From tkacvins at gmail.com  Sat Apr  1 16:09:44 2023
From: tkacvins at gmail.com (Tom Kacvinsky)
Date: Sat, 1 Apr 2023 10:09:44 -0400
Subject: MacTex 2023 installation failure
Message-ID: <CAG_eJLc3NntsGr8jLZcFgASt_SgLQz8nv+1VqoUXx8cGWsX+oQ@mail.gmail.com>

Hi,

I am getting an error while trying to install the latest version of
MacTeX.  There is no indication of what the error is.

I am running macOS Venture 13.3 on a mac mini with Apple M1 silicon.

Any and all help would be appreciated.

Thanks,

Tom
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230401/c123e5e1/attachment.html>

From koch at uoregon.edu  Sat Apr  1 18:00:02 2023
From: koch at uoregon.edu (Richard Koch)
Date: Sat, 1 Apr 2023 16:00:02 +0000
Subject: MacTex 2023 installation failure
Message-ID: <529C0A07-F2B9-4320-9F85-102C8CAA1230@uoregon.edu>

Tom Kacvinsky,

At the top of the MacTeX download page, it says

"While downloading, please skim read the rest of the page. Users who run into trouble often write us without noticing that the solution is on this very page."

But you don't mention any of the preliminary actions suggested there, leaving us to guess what kind of problem you might have run into.

Richard Koch


 

From tkacvins at gmail.com  Sat Apr  1 19:07:49 2023
From: tkacvins at gmail.com (Tom Kacvinsky)
Date: Sat, 1 Apr 2023 13:07:49 -0400
Subject: MacTex 2023 installation failure
In-Reply-To: <529C0A07-F2B9-4320-9F85-102C8CAA1230@uoregon.edu>
References: <529C0A07-F2B9-4320-9F85-102C8CAA1230@uoregon.edu>
Message-ID: <CAG_eJLd158qJtXXmGQ37Q28F1_Zm2adeaABcga_vaKtdvuzxGA@mail.gmail.com>

HI Richard,


On Sat, Apr 1, 2023 at 12:00?PM Richard Koch <koch at uoregon.edu> wrote:

> Tom Kacvinsky,
>
> At the top of the MacTeX download page, it says
>
> "While downloading, please skim read the rest of the page. Users who run
> into trouble often write us without noticing that the solution is on this
> very page."
>
> But you don't mention any of the preliminary actions suggested there,
> leaving us to guess what kind of problem you might have run into.


I read through the section you mentioned, and this is the specific error I
get

Sometimes at the end of installation, the installer will report "The
Installer encountered an error that caused the installation to fail.
Contact the software manufacturer for assistance." In almost all cases,
this error is caused by a faulty download. There is an easy way to check
this.

I checked the md5 sum and that matches with what is on the download page.

So I am filling out the MacTeX support form.

By the way, this is the second time today I haven't followed instructions.
*sigh*.

Tom
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230401/3b656977/attachment.html>

From hugo.parelho at gmail.com  Sun Apr  2 15:51:56 2023
From: hugo.parelho at gmail.com (Hugo Gomes)
Date: Sun, 2 Apr 2023 14:51:56 +0100
Subject: Misplaced files in pgf-PeriodicTable (v2.0.0)
Message-ID: <CANF6nuqMugOXJeYd0m4yW5bwLN2vEuuczs=-H6pTiRPnOa+4TQ@mail.gmail.com>

Hi, there are two misplaced files in texlive tree:
- the files ?pgfPT.input.library.tex? and ?pgfPTlibrary.colorschemes.tex?
are in
\texlive\2023\texmf-dist\doc\latex\pgf-periodictable
and must be in
\texlive\2023\texmf-dist\tex\latex\pgf-periodictable

Best regards,

Hugo Gomes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230402/d9df49a0/attachment.html>

From peter at silmaril.ie  Sun Apr  2 20:49:41 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Sun, 2 Apr 2023 19:49:41 +0100
Subject: Historic query
Message-ID: <865225d3-b417-979b-d368-6e90071e316d@silmaril.ie>

My Ubuntu texlive-full under Mint 20.1 (server) is (I think) 2019, so 
it's long out of date but long-term stability is all on this system, and 
Norbert's packaging is flawless.

However, I just noticed that typing 'texdoc setspace' in the terminal 
brings up TeXStudio with setspace.sty. I just wanted to raise a flag to 
see if current TL does include the PDF, in case it was an accidental 
omission.

Peter

From zdenek.wagner at gmail.com  Sun Apr  2 21:11:39 2023
From: zdenek.wagner at gmail.com (Zdenek Wagner)
Date: Sun, 2 Apr 2023 21:11:39 +0200
Subject: Historic query
In-Reply-To: <865225d3-b417-979b-d368-6e90071e316d@silmaril.ie>
References: <865225d3-b417-979b-d368-6e90071e316d@silmaril.ie>
Message-ID: <CAC1phyaFVceKr+_EXQRwSvnNVgCQf-aA8DMnktBvCv9bRQv=-w@mail.gmail.com>

Hi,

at my computer "texdoc setspace" opens setspace.sty up to TL 2021 and
the PDF document since TL 2022.

Zden?k Wagner
https://www.zdenek-wagner.eu/

ne 2. 4. 2023 v 21:05 odes?latel Peter Flynn <peter at silmaril.ie> napsal:
>
> My Ubuntu texlive-full under Mint 20.1 (server) is (I think) 2019, so
> it's long out of date but long-term stability is all on this system, and
> Norbert's packaging is flawless.
>
> However, I just noticed that typing 'texdoc setspace' in the terminal
> brings up TeXStudio with setspace.sty. I just wanted to raise a flag to
> see if current TL does include the PDF, in case it was an accidental
> omission.
>
> Peter


From peter at silmaril.ie  Sun Apr  2 23:04:46 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Sun, 2 Apr 2023 22:04:46 +0100
Subject: Historic query
In-Reply-To: <CAC1phyaFVceKr+_EXQRwSvnNVgCQf-aA8DMnktBvCv9bRQv=-w@mail.gmail.com>
References: <865225d3-b417-979b-d368-6e90071e316d@silmaril.ie>
 <CAC1phyaFVceKr+_EXQRwSvnNVgCQf-aA8DMnktBvCv9bRQv=-w@mail.gmail.com>
Message-ID: <9f5f6d09-ed83-af63-c8b3-5d83fe5c2b9e@silmaril.ie>

On 02/04/2023 20:11, Zdenek Wagner wrote:
> at my computer "texdoc setspace" opens setspace.sty up to TL 2021 and
> the PDF document since TL 2022.

Excellent, thank you very much.

Peter

From karl at freefriends.org  Sun Apr  2 23:35:18 2023
From: karl at freefriends.org (Karl Berry)
Date: Sun, 2 Apr 2023 15:35:18 -0600
Subject: Misplaced files in pgf-PeriodicTable (v2.0.0)
In-Reply-To: <CANF6nuqMugOXJeYd0m4yW5bwLN2vEuuczs=-H6pTiRPnOa+4TQ@mail.gmail.com>
Message-ID: <202304022135.332LZIXk008119@freefriends.org>

    - the files <<pgfPT.input.library.tex>> and <<pgfPTlibrary.colorschemes.tex>>
    are in
    \texlive\2023\texmf-dist\doc\latex\pgf-periodictable
    and must be in
    \texlive\2023\texmf-dist\tex\latex\pgf-periodictable

Thanks Hugo. Fixed for tonight's update. --best, karl.

From norbert at preining.info  Mon Apr  3 02:23:07 2023
From: norbert at preining.info (Norbert Preining)
Date: Mon, 3 Apr 2023 09:23:07 +0900
Subject: Historic query
In-Reply-To: <865225d3-b417-979b-d368-6e90071e316d@silmaril.ie>
References: <865225d3-b417-979b-d368-6e90071e316d@silmaril.ie>
Message-ID: <ZCocay21JO5aE9Gw@bulldog>

On Sun, 02 Apr 2023, Peter Flynn wrote:
> My Ubuntu texlive-full under Mint 20.1 (server) is (I think) 2019, so it's
> long out of date but long-term stability is all on this system, and
> Norbert's packaging is flawless.

Thanks for the flowers ;-)

> However, I just noticed that typing 'texdoc setspace' in the terminal brings
> up TeXStudio with setspace.sty. I just wanted to raise a flag to see if
> current TL does include the PDF, in case it was an accidental omission.

Already answered.

All the best

Norbert

--
PREINING Norbert                              https://www.preining.info
Mercari Inc.     +     IFMGA Guide     +     TU Wien     +     TeX Live
GPG: 0x860CDC13   fp: F7D8 A928 26E3 16A1 9FA0 ACF0 6CAC A448 860C DC13

From werner at suse.de  Mon Apr  3 09:34:31 2023
From: werner at suse.de (Dr. Werner Fink)
Date: Mon, 3 Apr 2023 09:34:31 +0200
Subject: TL 2023: package pgf-periodictable
Message-ID: <ZCqBh3nr7n8Vy7KG@boole.suse.de>

Hi,

just noted that pgf-PeriodicTable.sty from pgf-periodictable tries to
load pgfPT.input.library.tex but this is part of the doc package.
IMHO it should be part of the runfiles ;)

Werner

-- 
  "Having a smoking section in a restaurant is like having
          a peeing section in a swimming pool." -- Edward Burr
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 894 bytes
Desc: not available
URL: <https://tug.org/pipermail/tex-live/attachments/20230403/658d66b0/attachment.sig>

From cursus.publicus at sacrumromanumimperium.org  Mon Apr  3 09:39:59 2023
From: cursus.publicus at sacrumromanumimperium.org (Andreas Hirsch)
Date: Mon, 3 Apr 2023 09:39:59 +0200
Subject: TL 2023: package pgf-periodictable
In-Reply-To: <ZCqBh3nr7n8Vy7KG@boole.suse.de>
References: <ZCqBh3nr7n8Vy7KG@boole.suse.de>
Message-ID: <787081c3-192f-3856-2f5e-776ad1be4894@sacrumromanumimperium.org>

Hi Werner,

Dr. Werner Fink schrieb am 03.04.2023 um 09:34:
> Hi,
> 
> just noted that pgf-PeriodicTable.sty from pgf-periodictable tries to
> load pgfPT.input.library.tex but this is part of the doc package.
> IMHO it should be part of the runfiles ;)

This has already been mentioned and repaired:

> Karl Berry schrieb am 02.04.2023 um 23:35:
>>      - the files <<pgfPT.input.library.tex>> and <<pgfPTlibrary.colorschemes.tex>>
>>      are in
>>      \texlive\2023\texmf-dist\doc\latex\pgf-periodictable
>>      and must be in
>>      \texlive\2023\texmf-dist\tex\latex\pgf-periodictable
>> 
>> Thanks Hugo. Fixed for tonight's update. --best, karl.

best,
Andreas

-- 
'Es w?re dumm,
  sich ?ber die Welt zu ?rgern.
  Sie k?mmert sich nicht darum.'
	Marcus Aurelius (121-180)

From werner at suse.de  Tue Apr  4 11:30:52 2023
From: werner at suse.de (Dr. Werner Fink)
Date: Tue, 4 Apr 2023 09:30:52 +0000
Subject: [tex-live] Re: Running `context` fails with TeX Live 2023:
 mtxrun | unknown script 'mtx-context.lua' or 'mtx-mtx-context.lua'
In-Reply-To: <e32b38630b3e3f275a1ac8ee465c847261b79573.camel@telus.net>
References: <20230327110053.sairqyp22uhsti3e@inspiron>
 <e32b38630b3e3f275a1ac8ee465c847261b79573.camel@telus.net>
Message-ID: <20230404093052.GA25366@wotan.suse.de>

On 2023/03/27 16:58:33 -0600, Max Chernoff wrote:
> Hi Vit,
> 
> > Running `mtxrun --generate` produces the following output:
> 
> > resolvers       | resolving | variable 'SELFAUTOLOC' set to '/usr/bin'
> > resolvers       | resolving | variable 'SELFAUTODIR' set to '/usr'
> > resolvers       | resolving | variable 'SELFAUTOPARENT' set to '.'
> > resolvers       | resolving | variable 'TEXMFCNF' set to ''
> > resolvers       | resolving | variable 'TEXMF' set to ''
> > resolvers       | resolving | variable 'TEXOS' set to 'usr'
> 
> That's a problem. 
> 
> > This issue was originally reported at
> > https://gitlab.com/islandoftex/images/texlive/-/issues/30
> 
> This line in the source
> 
>    echo "tlpdbopt_sys_bin /usr/bin" >> install.profile && \
> 
>    (https://gitlab.com/islandoftex/images/texlive/-/blob/228ded/Dockerfile#L64)
>    
> is what's causing the issue. When you start LuaMetaTeX, it looks for a
> .lua file with the same name as the binary in the same folder. So
> running "/usr/bin/mtxrun" tries to run "/usr/bin/mtxrun.lua". Then
> mtxrun.lua tries to load "../../texmf-dist/web2c/texmfcnf.lua". That
> file exists if you run the default
> "$SELFAUTOPARENT/bin/<platform>/mtxrun", but it doesn't exist if you run
> "/usr/bin/mtxrun".
> 
> Two possible solutions:
> 
> 1. Don't use tlpdbopt_sys_bin; instead, set the path to contain
>    /usr/local/texlive/2023/bin/<platform>
>    
> 2. Set $TEXMF to the correct value and regenerate. You'll need to keep
>    this set any time that you run ConTeXt
> 
>       export TEXMF=/usr/local/texlive/2023/
>       mtxrun --generate
>       
> 3. Fix mtxrun.lua so that it looks in the proper location. I've tried to do?
>    this, but I couldn't get it to work properly.

I had seen similar problems for an installation of the binaries in /usr/bin
and the TeXLive tree below /usr/share/texmf ...

To solve that not only texmfcnf.lua (here below /etc/texmf/web2c) but also
resolvers.luacnfspec in texmf-dist/scripts/context/lua/mtxrun.lua
and texmf-dist/tex/context/base/mkiv/data-res.lua had to be changed to
enable mtxrun as well luametatex (called also with -ini for the formats
by mtxrun).

I'm using now

 resolvers.luacnfspec=concat ({
    "home:texmf/web2c",
    "/etc/texmf/web2c",
    "selfautodir:/share/texmf/web2c",
 }, ";")

works perfect.  At calling mtxrun --generate and/or --script fonts --reload
there whould be not set any different environment as then the databases have
a different hash.



Werner

-- 
  "Having a smoking section in a restaurant is like having
          a peeing section in a swimming pool." -- Edward Burr
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 880 bytes
Desc: not available
URL: <https://tug.org/pipermail/tex-live/attachments/20230404/87dbceff/attachment.sig>

From jfine2358 at gmail.com  Tue Apr  4 21:42:46 2023
From: jfine2358 at gmail.com (Jonathan Fine)
Date: Tue, 4 Apr 2023 20:42:46 +0100
Subject: Thu 6 Apr: TeX Hour: arXiv Access Research Report: 6:30 to 7:30 BST
In-Reply-To: <CALD=Yf_NuaF1LC=xRjGjZwqd4Lhp7c7EPS6aqAfmPe_CWcjVrg@mail.gmail.com>
References: <1d76208b-ba54-49be-81a5-2b9165120704n@googlegroups.com>
 <CALD=Yf_3EfRKYSO8EC9S7tDyNZUdJH1TunSrcy8smYEO-573aQ@mail.gmail.com>
 <CALD=Yf_NuaF1LC=xRjGjZwqd4Lhp7c7EPS6aqAfmPe_CWcjVrg@mail.gmail.com>
Message-ID: <CALD=Yf8uPugN3Xv8+gH4p3qfwMehrvq=CLpz2YXsDGhQPzj9Yw@mail.gmail.com>

Hi

The week after next, there'll be the most important and best-attended
meeting related to the future of TeX in STEM publishing for several years.
It's the very first arXiv Accessibility Forum, Monday 17 April, 1:00pm to
5:00pm Eastern Time US. It is an online meeting. The next two TeX Hour are
focussed on this forum

If you're interested in the arXiv forum, read about it and perhaps sign
right up now, while there are still spaces available. Here's the URL:
https://info.arxiv.org/about/accessibility_forum.html

And between now and 17 April, perhaps find some time to look at the arXiv
Accessibility Report:
https://info.arxiv.org/about/accessibility_research_report.html

If you've still got time and energy, the next two TeX Hours are related to
the arXiv Accessibility Forum. They are all on Thursdays, 6:30 to 7:30pm
British Summer Time. Here's the UK time now: https://time.is/UK.

The Zoom details:
https://us02web.zoom.us/j/78551255396?pwd=cHdJN0pTTXRlRCtSd1lCTHpuWmNIUT09.
Meeting ID: 785 5125 5396
Passcode: knuth

Thursday 6 April: The arXiv Accessibility Report
https://texhour.github.io/2023/04/06/arxiv-access-report/

Thursday 13 April: Unlatex: Results and Prospects
https://texhour.github.io/2023/04/13/unlatex-results-prospects/

The video of last week's meeting LaTeX and the arXiv Access Tree is
available here:
https://texhour.github.io/2023/03/30/latex-arxiv-access-tree/

wishing you accessible TeXing

Jonathan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230404/5e1dc723/attachment.html>

From Martin.Muench at Uni-Bonn.de  Wed Apr  5 11:23:13 2023
From: Martin.Muench at Uni-Bonn.de (=?UTF-8?Q?Martin_M=c3=bcnch?=)
Date: Wed, 5 Apr 2023 11:23:13 +0200
Subject: Nitpicking about texlive-de.pdf
In-Reply-To: <22c6d3c2-ee05-d504-1f43-7727682c1b2c@Uni-Bonn.de>
References: <55dd2b4b-78a4-c948-8a4e-d54fc3341f30@Uni-Bonn.de>
 <ZBOdn17WICA9VB7i@bulldog> <22c6d3c2-ee05-d504-1f43-7727682c1b2c@Uni-Bonn.de>
Message-ID: <d09bdb4b-f9ee-b4ee-35f1-3136c1a26f6c@Uni-Bonn.de>

In https://www.tug.org/texlive/doc/texlive-de/texlive-de.pdf there are 
six cases of "??" (i.e. broken links).

Additionally by chance I found:

Section 10.13, page 91: "Zus?tzlich wurden einige Plattformen von der 
DVD (einfach um Platz zu sparen), k?nnen aber ?ber das Internet 
installiert werden." Die Plattformen wurden >>entfernt<< (fehlt), oder?

Section 10.14, page 92: "afm2tfm makes" dt.: "macht"

Section 10.15, page 92f.: "basiertjetzt" -> "basiert jetzt"
page 93: "konsisten" -> "konsistent" (and please add a \linebreak two 
lines after the "konsisten")
"in in" -> "in"
"Modus(updmap" -> "Modus (updmap"
"; Neu" -> "; neu" (twice)

Section 10.16, p. 94: "vier contains four GUI Programme sind jetzt 
enthalten" remove "contains four "

Section 10.17, p. 95: "ist der der eingeschr?nkten Befehle, die 
standardm??ig verf?gbar sind"
-> "und wurde hinzugef?gt zur Liste eingeschr?nkter Befehle, die 
standardm??ig verf?gbar sind"

"Neue Primitiveexpanded" should have a space between "Primitive" and 
"expanded" (two occurrences of "Neue Primitiveexpanded"!) and there are 
two significantly overfull hboxes and one occurrence of the "??" is on 
that page.

Section 10.18, p. 96: "in LuaTEXm?glich" -> "in LuaTEX m?glich"

Section 10.18, p. 97: one overfull box ("auch"), missing samepage or 
goodbreak at "LuaTeX:"

10.21, p. 102: There are people on the web which call for help, because 
they do have a 32bit Windows release and need to install TeX Live. 
(Windows 8.1 was supported by Microsoft until this January, Windows 10 
will be supported until end of 2025.) And the Windows XP people need TeX 
Live 2017. (Well, in the first place they need a different OS, of 
course.) So, maybe point them to
ftp://tug.org/tex/texlive/historic/
to at least get some (old) distribution? Would also apply to page 7 and 
to the versions of texlive.pdf in other languages.

From ziegenhagen at gmail.com  Wed Apr  5 13:51:03 2023
From: ziegenhagen at gmail.com (Uwe Ziegenhagen)
Date: Wed, 5 Apr 2023 13:51:03 +0200
Subject: Nitpicking about texlive-de.pdf
In-Reply-To: <d09bdb4b-f9ee-b4ee-35f1-3136c1a26f6c@Uni-Bonn.de>
References: <d09bdb4b-f9ee-b4ee-35f1-3136c1a26f6c@Uni-Bonn.de>
Message-ID: <DA71755E-F791-4543-B3E2-E1862B1D46B4@gmail.com>

Will check it. Thank you

Von meinem iPhone gesendet

> Am 05.04.2023 um 13:35 schrieb Martin M?nch <Martin.Muench at uni-bonn.de>:
> 
> ?In https://www.tug.org/texlive/doc/texlive-de/texlive-de.pdf there are six cases of "??" (i.e. broken links).
> 
> Additionally by chance I found:
> 
> Section 10.13, page 91: "Zus?tzlich wurden einige Plattformen von der DVD (einfach um Platz zu sparen), k?nnen aber ?ber das Internet installiert werden." Die Plattformen wurden >>entfernt<< (fehlt), oder?
> 
> Section 10.14, page 92: "afm2tfm makes" dt.: "macht"
> 
> Section 10.15, page 92f.: "basiertjetzt" -> "basiert jetzt"
> page 93: "konsisten" -> "konsistent" (and please add a \linebreak two lines after the "konsisten")
> "in in" -> "in"
> "Modus(updmap" -> "Modus (updmap"
> "; Neu" -> "; neu" (twice)
> 
> Section 10.16, p. 94: "vier contains four GUI Programme sind jetzt enthalten" remove "contains four "
> 
> Section 10.17, p. 95: "ist der der eingeschr?nkten Befehle, die standardm??ig verf?gbar sind"
> -> "und wurde hinzugef?gt zur Liste eingeschr?nkter Befehle, die standardm??ig verf?gbar sind"
> 
> "Neue Primitiveexpanded" should have a space between "Primitive" and "expanded" (two occurrences of "Neue Primitiveexpanded"!) and there are two significantly overfull hboxes and one occurrence of the "??" is on that page.
> 
> Section 10.18, p. 96: "in LuaTEXm?glich" -> "in LuaTEX m?glich"
> 
> Section 10.18, p. 97: one overfull box ("auch"), missing samepage or goodbreak at "LuaTeX:"
> 
> 10.21, p. 102: There are people on the web which call for help, because they do have a 32bit Windows release and need to install TeX Live. (Windows 8.1 was supported by Microsoft until this January, Windows 10 will be supported until end of 2025.) And the Windows XP people need TeX Live 2017. (Well, in the first place they need a different OS, of course.) So, maybe point them to
> ftp://tug.org/tex/texlive/historic/
> to at least get some (old) distribution? Would also apply to page 7 and to the versions of texlive.pdf in other languages.


From i at estela.cn  Wed Apr  5 05:55:18 2023
From: i at estela.cn (Estela ad Astra)
Date: Wed, 5 Apr 2023 11:55:18 +0800
Subject: Biber on aarch64 is available
In-Reply-To: <4388907.LvFx2qVVIh@komascript.de>
References: <4388907.LvFx2qVVIh@komascript.de>
Message-ID: <3fcfa1b2-e0d2-7bed-9355-3449f78c0038@estela.cn>

Sorry for that I am not good at using mail list.
I reached https://tug.org/pipermail/tex-live/2022-April/047999.html
when looking for solve to my texlive problem of biber on Ubuntu arm64.
I followed https://github.com/plk/biber to build manually and it works.
So I reckon architecture is not a problem.
Hope to see biber present in texlive.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <https://tug.org/pipermail/tex-live/attachments/20230405/fb36709c/attachment.sig>

From siepo at bitmuis.nl  Wed Apr  5 21:36:09 2023
From: siepo at bitmuis.nl (Siep Kroonenberg)
Date: Wed, 5 Apr 2023 21:36:09 +0200
Subject: Glitch in Windows 11 GUI installer for TeX Live 2023
In-Reply-To: <CMM.0.95.0.1679764376.beebe@gamma.math.utah.edu>
References: <CMM.0.95.0.1679764376.beebe@gamma.math.utah.edu>
Message-ID: <ZC3NqZMGUP5loRdd@bitmuis.nl>

On Sat, Mar 25, 2023 at 11:12:56AM -0600, Nelson H. F. Beebe wrote:
> Earlier today, I downloaded the latest texcol2023.iso* files from
> 
> 	https://tug.org/~manfred/
> 
> to our mirror site at
> 
> 	https://www.math.utah.edu/pub/texlive-utah/Images
> 
> and updated the README.html file there.
> 
> I then moved to a Windows 11 laptop, downloaded those files to my
> Downloads directory, clicked on the installer in a file manager
> window, and selected the Advanced button.
> 
> That brings up a screen that has a menu item "GUI language".  For fun,
> I changed to other languages than the default English, and found that
> the change worked.  However, when I reverted to English without
> leaving the GUI installer, I found that the second line on the
> right-hand Options panel remained in the previously selected language;
> it is the line immediately under the "Default paper size" line.
> 
> This is most evident if you select Chinese or Japanese, but the
> problem remains for every other language that I tried.
> 
> The only way to repair it is to exit the GUI installer, then start it
> again.

This should be fixed in r66771.

-- 
Siep Kroonenberg

From karl at freefriends.org  Wed Apr  5 23:16:46 2023
From: karl at freefriends.org (Karl Berry)
Date: Wed, 5 Apr 2023 15:16:46 -0600
Subject: Biber on aarch64 is available
In-Reply-To: <3fcfa1b2-e0d2-7bed-9355-3449f78c0038@estela.cn>
Message-ID: <202304052116.335LGk64009415@freefriends.org>

    I reached https://tug.org/pipermail/tex-live/2022-April/047999.html
    ...
    Hope to see biber present in texlive.

As I explained in the message you cited above, someone (you, if you're
up for doing it) will need to contact Phil (maybe a submission at
https://github.com/plk/biber/issues is easest), offer the binary, and
arrange for it to get onto CTAN.  (I'm not sure of Phil's process for that.)

Then, and only then, will a new biber binary appear in TL.

Glad you were able to build it, in any case. --best, karl.

From fischer at unix-ag.uni-kl.de  Fri Apr  7 20:48:45 2023
From: fischer at unix-ag.uni-kl.de (Thomas Fischer)
Date: Fri, 7 Apr 2023 20:48:45 +0200
Subject: Feature request: package for font 'Gelasio', metrically compatible
 with 'Georgia'
Message-ID: <20230407183507.vl4fg76fn7xa7wnv@1fa4b100459ec294b9874642cf4714e3.l38.home.arpa>

Hello,

I would like to request the creation and inclusion of a package for 
font 'Gelasio' into TeX Live. Gelasio [1] is a font published under 
the OFL 1.1 and mimics Microsoft's Georgia font to which its metrics 
are compatible.
As such, Gelasio is a legal drop-in if documents have to be created 
on non-Microsoft systems where the requirement is to 'use Georgia'.

The package should ideally ship both OpenType font files for use 
with Lua(La)TeX and Xe(La)TeX as well as all the magical files to 
make it work with pdf(La)TeX. The user, irrespective of TeX flavour, 
would only need to issue
\usepackage{gelasio}
or equivalent.

Thank you for considering this request!

Greetings,
Thomas Fischer

[1] https://github.com/SorkinType/Gelasio

From reinhard.kotucha at gmx.de  Sat Apr  8 00:39:45 2023
From: reinhard.kotucha at gmx.de (Reinhard Kotucha)
Date: Sat, 8 Apr 2023 00:39:45 +0200
Subject: Feature request: package for font 'Gelasio', metrically compatible
 with 'Georgia'
In-Reply-To: <20230407183507.vl4fg76fn7xa7wnv@1fa4b100459ec294b9874642cf4714e3.l38.home.arpa>
References: <20230407183507.vl4fg76fn7xa7wnv@1fa4b100459ec294b9874642cf4714e3.l38.home.arpa>
Message-ID: <25648.39857.651305.870744@gargle.gargle.HOWL>

On 2023-04-07 at 20:48:45 +0200, Thomas Fischer wrote:

 > Hello,
 >
 > I would like to request the creation and inclusion of a package for
 > font 'Gelasio' into TeX Live. Gelasio [1] is a font published under
 > the OFL 1.1 and mimics Microsoft's Georgia font to which its metrics
 > are compatible.
 > As such, Gelasio is a legal drop-in if documents have to be created
 > on non-Microsoft systems where the requirement is to 'use Georgia'.
 >
 > The package should ideally ship both OpenType font files for use
 > with Lua(La)TeX and Xe(La)TeX as well as all the magical files to
 > make it work with pdf(La)TeX. The user, irrespective of TeX flavour,
 > would only need to issue
 > \usepackage{gelasio}
 > or equivalent.
 >
 > Thank you for considering this request!
 >
 > Greetings,
 > Thomas Fischer
 >
 > [1] https://github.com/SorkinType/Gelasio

Hi Thomas,
thank you very much for the link.  Georgia is one of the beautiful
fonts created by Matthew Carter which display well on paper and also
on screen.

TeX Live only supports packages which are on CTAN.  It would be nice
if you upload the fonts to CTAN:

   https://ctan.org/upload

It's probably not useful to upload the variable fonts and webfonts
(*.woff2) to CTAN as well because TeX can't do anything with them.

Regards,
  Reinhard

--
------------------------------------------------------------------
Reinhard Kotucha                            Phone: +49-511-3373112
Marschnerstr. 25
D-30167 Hannover                    mailto:reinhard.kotucha at gmx.de
------------------------------------------------------------------


From Herbert.Voss at fu-berlin.de  Sat Apr  8 07:43:35 2023
From: Herbert.Voss at fu-berlin.de (Herbert Voss)
Date: Sat, 8 Apr 2023 07:43:35 +0200
Subject: Feature request: package for font 'Gelasio', metrically
 compatible with 'Georgia'
In-Reply-To: <25648.39857.651305.870744@gargle.gargle.HOWL>
References: <20230407183507.vl4fg76fn7xa7wnv@1fa4b100459ec294b9874642cf4714e3.l38.home.arpa>
 <25648.39857.651305.870744@gargle.gargle.HOWL>
Message-ID: <2cc7c25e-2184-2a1a-d9a3-6d90899869f1@fu-berlin.de>

Am 08.04.23 um 00:39 schrieb Reinhard Kotucha:
> It's probably not useful to upload the variable fonts and webfonts

The variable fonts can be used by luatex/xetex and maybe useful too.

Herbert




> (*.woff2) to CTAN as well because TeX can't do anything with them.
>


From peter at silmaril.ie  Sun Apr  9 12:29:47 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Sun, 9 Apr 2023 11:29:47 +0100
Subject: TEXMFHOME on Windows
Message-ID: <b201b819-2ee0-107e-398c-be0f325aaca9@silmaril.ie>

A user has asked me why a private class file isn't being found. They are 
running TL under Windows (I have asked them which versions) and has put 
the file in C:/Users/username/texmf which is what TEXMFHOME gets 
expanded to.

My understanding is that TEXMFHOME does not need any ls-R database, but 
my understanding of Windows is poor; and I thought the canonical 
location was Computer\System\Users\username\texmf (avoiding the need for 
C:, D:, etc)

I have asked them to move the file to 
C:/Users/username/texmf/tex/latex/username/algoslides.cls and report back.

If C:/Users/username/texmf is created at install time, presumably no 
further action is needed by a user, but do they need to use the TDS 
structure within that directory?

Peter

From P.Taylor at Rhul.Ac.Uk  Sun Apr  9 12:45:25 2023
From: P.Taylor at Rhul.Ac.Uk (Philip Taylor (RHBNC))
Date: Sun, 9 Apr 2023 11:45:25 +0100
Subject: [EXT] TEXMFHOME on Windows
In-Reply-To: <b201b819-2ee0-107e-398c-be0f325aaca9@silmaril.ie>
References: <b201b819-2ee0-107e-398c-be0f325aaca9@silmaril.ie>
Message-ID: <75d9ce66-eb51-5f1f-4f31-a5a772952f71@Rhul.Ac.Uk>

On 09/04/2023 11:29, Peter Flynn wrote:
 I thought the canonical location was Computer\System\Users\username\texmf (avoiding the need for C:, D:, etc)

Highly unlikely, IMHO.  How would Windows know that "Computer" was to be replaced by something more meaningful, why "System", and why not %username% ?
I would suggest either

  *   %homedrive%%homepath%\texmf
     *   or
  *   %userprofile%\texmf

--
** Phil.

This email, its contents and any attachments are intended solely for the addressee and may contain confidential information. In certain circumstances, it may also be subject to legal privilege. Any unauthorised use, disclosure, or copying is not permitted. If you have received this email in error, please notify us and immediately and permanently delete it. Any views or opinions expressed in personal emails are solely those of the author and do not necessarily represent those of Royal Holloway, University of London. It is your responsibility to ensure that this email and any attachments are virus free.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230409/1bf4a70f/attachment.html>

From siepo at bitmuis.nl  Sun Apr  9 13:51:59 2023
From: siepo at bitmuis.nl (Siep Kroonenberg)
Date: Sun, 9 Apr 2023 13:51:59 +0200
Subject: TEXMFHOME on Windows
In-Reply-To: <b201b819-2ee0-107e-398c-be0f325aaca9@silmaril.ie>
References: <b201b819-2ee0-107e-398c-be0f325aaca9@silmaril.ie>
Message-ID: <ZDKm3z1AC+O7KKMi@bitmuis.nl>

On Sun, Apr 09, 2023 at 11:29:47AM +0100, Peter Flynn wrote:
> A user has asked me why a private class file isn't being found. They are
> running TL under Windows (I have asked them which versions) and has put the
> file in C:/Users/username/texmf which is what TEXMFHOME gets expanded to.
> 
> My understanding is that TEXMFHOME does not need any ls-R database, but my
> understanding of Windows is poor; and I thought the canonical location was
> Computer\System\Users\username\texmf (avoiding the need for C:, D:, etc)
> 
> I have asked them to move the file to
> C:/Users/username/texmf/tex/latex/username/algoslides.cls and report back.
> 
> If C:/Users/username/texmf is created at install time, presumably no further
> action is needed by a user, but do they need to use the TDS structure within
> that directory?
> 
> Peter

The windows version of $HOME is %userprofile%. TEXMFHOME is by
default set to %userprofile%/texmf, which is indeed usually
C:/Users/username/texmf.

The user has to create both this directory and its subdirectory
tex/latex, according to the TDS.

Your understanding is correct that this tree does not require an
ls-R.

Beware that Windows Explorer may display a localized name instead of
'Users', e.g. 'Gebruikers'. I am not sure whether the actual
directory name is ever localized.

-- 
Siep Kroonenberg

From news3 at nililand.de  Sun Apr  9 15:14:21 2023
From: news3 at nililand.de (Ulrike Fischer)
Date: Sun, 9 Apr 2023 15:14:21 +0200
Subject: TEXMFHOME on Windows
References: <b201b819-2ee0-107e-398c-be0f325aaca9@silmaril.ie>
Message-ID: <1qgpoytdw3ons.dlg@nililand.de>

Am Sun, 9 Apr 2023 11:29:47 +0100 schrieb Peter Flynn:

> If C:/Users/username/texmf is created at install time, presumably no 
> further action is needed by a user, but do they need to use the TDS 
> structure within that directory?

*All* texmf trees must follow the TDS structure. 

cls files belong below <whatevertexmf>/tex/latex and tfm below 
 <whatevertexmf>/fonts/tfm

-- 
Ulrike Fischer 
http://www.troubleshooting-tex.de/


From peter at silmaril.ie  Sun Apr  9 16:32:37 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Sun, 9 Apr 2023 15:32:37 +0100
Subject: [EXT] TEXMFHOME on Windows
In-Reply-To: <75d9ce66-eb51-5f1f-4f31-a5a772952f71@Rhul.Ac.Uk>
References: <b201b819-2ee0-107e-398c-be0f325aaca9@silmaril.ie>
 <75d9ce66-eb51-5f1f-4f31-a5a772952f71@Rhul.Ac.Uk>
Message-ID: <d70a9709-16a5-62dd-b57b-cdd870008be9@silmaril.ie>

On 09/04/2023 11:45, Philip Taylor (RHBNC) wrote:
> On 09/04/2023 11:29, Peter Flynn wrote:
>> I thought the canonical location was 
>> Computer\System\Users\username\texmf (avoiding the need for C:, D:,
>> etc)
> 
> Highly unlikely, IMHO.? How would Windows know that "Computer" was to be 
> replaced by something more meaningful, why "System", and why not 
> %username% ?

Sorry, my all-lowercase username was meant to be replaced by the user's 
actual username. But don't ask me why Computer and System :-) that's 
what a Windows user told me (in good faith) and I believe it is true, 
via some kind of alias or soft link. I touch not the Windows.

> I would suggest either
> 
>   * %homedrive%%homepath%\texmf
>       o /or/
>   * %userprofile%\texmf

That's useful, thanks.


On 09/04/2023 12:51, Siep Kroonenberg wrote:

 > The windows version of $HOME is %userprofile%. TEXMFHOME is by
 > default set to %userprofile%/texmf, which is indeed usually
 > C:/Users/username/texmf.

OK, thanks. I think I understand now.

I wonder what happened to Computer\System\Users\... or maybe that was a 
Windows 10 thing now dropped.

 > The user has to create both this directory and its subdirectory
 > tex/latex, according to the TDS.

Right. I recommend duplicating the distribution texmf tree, which is 
overkill but does at least ensure that all the expected subdirectories 
are present. But if ../tex/latex is sufficient to make it work, that's 
much easier.

 > Beware that Windows Explorer may display a localized name instead of
 > 'Users', e.g. 'Gebruikers'. I am not sure whether the actual
 > directory name is ever localized.

Good point, thanks.

I knew there were good reasons not to use Windows :-)

Peter


From P.Taylor at Rhul.Ac.Uk  Sun Apr  9 17:57:44 2023
From: P.Taylor at Rhul.Ac.Uk (Philip Taylor (RHBNC))
Date: Sun, 9 Apr 2023 16:57:44 +0100
Subject: [EXT] TEXMFHOME on Windows
In-Reply-To: <d70a9709-16a5-62dd-b57b-cdd870008be9@silmaril.ie>
References: <b201b819-2ee0-107e-398c-be0f325aaca9@silmaril.ie>
 <75d9ce66-eb51-5f1f-4f31-a5a772952f71@Rhul.Ac.Uk>
 <d70a9709-16a5-62dd-b57b-cdd870008be9@silmaril.ie>
Message-ID: <005e2fbe-9e22-f8a7-94b3-055ac34f248c@Rhul.Ac.Uk>

On 09/04/2023 15:32, Peter Flynn wrote:
On 09/04/2023 11:45, Philip Taylor (RHBNC) wrote:
On 09/04/2023 11:29, Peter Flynn wrote:
I thought the canonical location was Computer\System\Users\username\texmf (avoiding the need for C:, D:,
etc)

Highly unlikely, IMHO.  How would Windows know that "Computer" was to be replaced by something more meaningful, why "System", and why not %username% ?

Sorry, my all-lowercase username was meant to be replaced by the user's actual username. But don't ask me why Computer and System :-) that's what a Windows user told me (in good faith) and I believe it is true, via some kind of alias or soft link. I touch not the Windows.
I think that said user was perhaps a little na?ve ? he/she believes that all Windows users perform an untweaked install, such that %SystemDrive" is also the root of %UserProfile%, whereas more sophisticated users frequently intervene to ensure that all user files are on any device other than %SystemDrive%.
I wonder what happened to Computer\System\Users\... or maybe that was a Windows 10 thing now dropped.
Not impossible, but perhaps improbable ? I am using something even older than Windows 10 (Windows 7, for Very Good Reasons?), and it wots not of "Computer\System\Users".
> Beware that Windows Explorer may display a localized name instead of
> 'Users', e.g. 'Gebruikers'. I am not sure whether the actual > directory name is ever localized.

Good point, thanks.
Ah, so you are using Irish (Gaeilge) Windows ? I did wonder ...
I knew there were good reasons not to use Windows :-)
And there are even better reasons not to use anything else !

This email, its contents and any attachments are intended solely for the addressee and may contain confidential information. In certain circumstances, it may also be subject to legal privilege. Any unauthorised use, disclosure, or copying is not permitted. If you have received this email in error, please notify us and immediately and permanently delete it. Any views or opinions expressed in personal emails are solely those of the author and do not necessarily represent those of Royal Holloway, University of London. It is your responsibility to ensure that this email and any attachments are virus free.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230409/efc1ccab/attachment-0001.html>

From tex-live at opsec.eu  Sun Apr  9 19:27:59 2023
From: tex-live at opsec.eu (Kurt Jaeger)
Date: Sun, 9 Apr 2023 19:27:59 +0200
Subject: TeXlive FreeBSD Port, distfile split-ups and licensing
Message-ID: <20230409172759.GM35206@home.opsec.eu>

Hello,

there's a FreeBSD Port for TeX Live 2021 here:
https://www.freshports.org/print/texlive-full/
and an update to TeX Live 2023 is in preparation.
TeX Live is distributed in several sub-ports, among them:

- devel/tex-kpathsea
- print/texlive-tlmgr
- devel/tex-libtexlua
- devel/tex-libtexluajit
[...]

One problem of those ports is the *huge* distfile. Hosts that build
FreeBSD ports might need to download the full distfile and then,
as the ports are build in parallel, might unpack the whole distfile
multiple times in parallel, which has huuuuge requirements
when it comes to disk-IO and RAM usage.

This can kill even hosts with 128 GB of RAM or more and fast SSDs.

The author of the update to TeX Live 2023 wants to split the
distfile into several sub-distfiles to lessen the load on the
users that build TeX Live from the ports.

This has licensing implications and we need to find some
person that is able to comment on those implications.

Who might be able to help ?

-- 
pi at opsec.eu            +49 171 3101372                    Now what ?

From peter at silmaril.ie  Sun Apr  9 23:02:08 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Sun, 9 Apr 2023 22:02:08 +0100
Subject: [EXT] TEXMFHOME on Windows
In-Reply-To: <005e2fbe-9e22-f8a7-94b3-055ac34f248c@Rhul.Ac.Uk>
References: <b201b819-2ee0-107e-398c-be0f325aaca9@silmaril.ie>
 <75d9ce66-eb51-5f1f-4f31-a5a772952f71@Rhul.Ac.Uk>
 <d70a9709-16a5-62dd-b57b-cdd870008be9@silmaril.ie>
 <005e2fbe-9e22-f8a7-94b3-055ac34f248c@Rhul.Ac.Uk>
Message-ID: <07bd79d1-7820-a175-83a3-b59911202bef@silmaril.ie>

On 09/04/2023 16:57, Philip Taylor (RHBNC) wrote:
[...]
> I think that said user was perhaps a little na?ve ? he/she believes
> that all Windows users perform an untweaked install, 

In the case of users with their own laptop/desktop this is largely true, 
I think. Almost certainly not for corporate users, if they are even 
allowed to install non-corporate-approved software.

> such that %SystemDrive" is also the root of %UserProfile%, whereas
> more sophisticated users frequently intervene to ensure that all
> user files are on any device /other than/ %SystemDrive%.
Fortunately I am replying in the case of unsophisticated users.

> [...] I am using something even older than Windows 10 (Windows 7, for
> Very Good Reasons?), and it wots not of "Computer\System\Users".
I wouldn't expect it to. %HOME% was apparently the standard on DOS, and 
therefore got embedded into early Windows, certainly up as far as W95.

> Ah, so you are using Irish (Gaeilge) Windows ? I did wonder ...

Nope :-) I don't use Windows at all.

>> I knew there were good reasons not to use Windows :-)
> And there are even better reasons not to use anything else !

Precious few of them, but this is probably neither the time nor place to 
rehearse the articles of thy belief in that respect :-)

Peter



From reinhard.kotucha at gmx.de  Mon Apr 10 00:11:38 2023
From: reinhard.kotucha at gmx.de (Reinhard Kotucha)
Date: Mon, 10 Apr 2023 00:11:38 +0200
Subject: [EXT] TEXMFHOME on Windows
In-Reply-To: <d70a9709-16a5-62dd-b57b-cdd870008be9@silmaril.ie>
References: <b201b819-2ee0-107e-398c-be0f325aaca9@silmaril.ie>
 <75d9ce66-eb51-5f1f-4f31-a5a772952f71@Rhul.Ac.Uk>
 <d70a9709-16a5-62dd-b57b-cdd870008be9@silmaril.ie>
Message-ID: <25651.14362.694688.424115@gargle.gargle.HOWL>

On 2023-04-09 at 15:32:37 +0100, Peter Flynn wrote:

 > On 09/04/2023 11:45, Philip Taylor (RHBNC) wrote:
 > > On 09/04/2023 11:29, Peter Flynn wrote:
 > >> I thought the canonical location was
 > >> Computer\System\Users\username\texmf (avoiding the need for C:,
 > >> D:, etc)
 > > 
 > > Highly unlikely, IMHO.? How would Windows know that "Computer"
 > > was to be replaced by something more meaningful, why "System",
 > > and why not %username% ?
 > 
 > Sorry, my all-lowercase username was meant to be replaced by the
 > user's actual username. But don't ask me why Computer and System
 > :-) that's what a Windows user told me (in good faith) and I
 > believe it is true, via some kind of alias or soft link. I touch
 > not the Windows.
 > 
 > > I would suggest either
 > > 
 > >   * %homedrive%%homepath%\texmf
 > >       o /or/
 > >   * %userprofile%\texmf
 > 
 > That's useful, thanks.
 > 
 > 
 > On 09/04/2023 12:51, Siep Kroonenberg wrote:
 > 
 >  > The windows version of $HOME is %userprofile%. TEXMFHOME is by
 >  > default set to %userprofile%/texmf, which is indeed usually
 >  > C:/Users/username/texmf.
 > 
 > OK, thanks. I think I understand now.
 > 
 > I wonder what happened to Computer\System\Users\... or maybe that
 > was a Windows 10 thing now dropped.
 > 
 >  > The user has to create both this directory and its subdirectory
 >  > tex/latex, according to the TDS.
 > 
 > Right. I recommend duplicating the distribution texmf tree, which
 > is overkill but does at least ensure that all the expected
 > subdirectories are present. But if ../tex/latex is sufficient to
 > make it work, that's much easier.
 > 
 >  > Beware that Windows Explorer may display a localized name
 >  > instead of 'Users', e.g. 'Gebruikers'. I am not sure whether the
 >  > actual directory name is ever localized.
 > 
 > Good point, thanks.

Hello Peter,
sorry for the late response.  TeX Live is supposed to behave the same
way on all supported platforms.  Thus you get the definitive answer
with

  kpsewhich -var-value TEXMFHOME 

This returns the full path (including the drive letter on Windows).
It doesn't mean that this directory already exists but that kpathsea
looks there for files.  The installer cannot create this directory
because it isn't aware of users.

Phil mentioned

  %homedrive%%homepath%\texmf
  %userprofile%\texmf

But this assumes that the installer was invoked with the default
parameters -- what we strongly recommend unless users have special
requirements.  The root of TEXMFHOME is not hard coded and the only
reliable way to determine its location is to ask kpathsea.

If you support Windows users it's important to know that there is
nothing special about Windows as far as TeX Live is concerned.  On
Windows the same configuration files are read as on Unix.  All
programs are compiled from the same sources, etc.

The nasty differences between Windows and other systems are handled by
the installer and tlmgr internally.  

In order to support Windows users efficiently the only thing you have
to do is to convince them to use TeX Live.  What works for you should
then also work for them the same way, even if they are under Windows.

 > I knew there were good reasons not to use Windows :-)

I know this probably even better than you because I was forced to use
Windows in the past, though I did almost all of my work within a
virtual machine running Debian.  Windows is a pain. It was amazing to
see things running up to 12 times faster on Unix than on Windows, even
within a virtual machine running on Windows.  Windows is horribly
slow.  Nevertheless, TeX Live at least always provides the same
results on every system it supports.

Regards,
  Reinhard

-- 
------------------------------------------------------------------
Reinhard Kotucha                            Phone: +49-511-3373112
Marschnerstr. 25
D-30167 Hannover                    mailto:reinhard.kotucha at gmx.de
------------------------------------------------------------------


From norbert at preining.info  Mon Apr 10 01:27:50 2023
From: norbert at preining.info (Norbert Preining)
Date: Mon, 10 Apr 2023 08:27:50 +0900
Subject: TeXlive FreeBSD Port, distfile split-ups and licensing
In-Reply-To: <20230409172759.GM35206@home.opsec.eu>
References: <20230409172759.GM35206@home.opsec.eu>
Message-ID: <ZDNJ9uwt3DssH5ux@bulldog>

Hi Kurt,

> The author of the update to TeX Live 2023 wants to split the
> distfile into several sub-distfiles to lessen the load on the
> users that build TeX Live from the ports.

Sounds like a good plan.

> This has licensing implications and we need to find some
> person that is able to comment on those implications.

No need for licensing considerations, you can split as you like.

I recommend splitting at *collection* levels, out of several reasons:
- still a reasonable amount of packages (not too many)
- guaranteed to be disjoint (no file in two collections)

This is what I have done for Debian packaging about 15 years or so ago
;-)

Best

Norbert

--
PREINING Norbert                              https://www.preining.info
Mercari Inc.     +     IFMGA Guide     +     TU Wien     +     TeX Live
GPG: 0x860CDC13   fp: F7D8 A928 26E3 16A1 9FA0 ACF0 6CAC A448 860C DC13

From tex-live at opsec.eu  Mon Apr 10 09:56:16 2023
From: tex-live at opsec.eu (Kurt Jaeger)
Date: Mon, 10 Apr 2023 09:56:16 +0200
Subject: TeXlive FreeBSD Port, distfile split-ups and licensing
In-Reply-To: <ZDNJ9uwt3DssH5ux@bulldog>
References: <20230409172759.GM35206@home.opsec.eu> <ZDNJ9uwt3DssH5ux@bulldog>
Message-ID: <20230410075616.GO35206@home.opsec.eu>

Hi!

> > The author of the update to TeX Live 2023 wants to split the
> > distfile into several sub-distfiles to lessen the load on the
> > users that build TeX Live from the ports.
> 
> Sounds like a good plan.
> 
> > This has licensing implications and we need to find some
> > person that is able to comment on those implications.
> 
> No need for licensing considerations, you can split as you like.

To be more precise: We would split the distfile and put the parts
on distcache.freebsd.org so that the ports will reference those
distfiles, not the original one.

Is that still OK ?

> I recommend splitting at *collection* levels, out of several reasons:
> - still a reasonable amount of packages (not too many)
> - guaranteed to be disjoint (no file in two collections)
> 
> This is what I have done for Debian packaging about 15 years or so ago
> ;-)

Are those distfiles available somewhere ? Then maybe we can reference
those instead of doing our own splitting ?

-- 
pi at opsec.eu            +49 171 3101372                    Now what ?

From norbert at preining.info  Mon Apr 10 10:03:59 2023
From: norbert at preining.info (Norbert Preining)
Date: Mon, 10 Apr 2023 17:03:59 +0900
Subject: TeXlive FreeBSD Port, distfile split-ups and licensing
In-Reply-To: <20230410075616.GO35206@home.opsec.eu>
References: <20230409172759.GM35206@home.opsec.eu> <ZDNJ9uwt3DssH5ux@bulldog>
 <20230410075616.GO35206@home.opsec.eu>
Message-ID: <ZDPC7zounX3POaNL@bulldog>

Hi Kurt,

On Mon, 10 Apr 2023, Kurt Jaeger wrote:
> To be more precise: We would split the distfile and put the parts
> on distcache.freebsd.org so that the ports will reference those
> distfiles, not the original one.
> 
> Is that still OK ?

There are some cases where you should not split certain files from each
other, but since in the end all would be distributed together, that is
fine.

Generally, TeX Live follows the Debian guidelines, that means you can
practically do whatever you want.

> Are those distfiles available somewhere ? Then maybe we can reference
> those instead of doing our own splitting ?

No. I wrote a script that collected all the files for each collection
and packaged them up. 

In Debian we have in fact only 4 *source* packages:
- texlive-bin: practically the source directory of svn
- texlive-base: the base collections
- texlive-extra: all the extra collections (like fontsextra etc) and
  some more
- texlive-lang: files for all the "language packs" (collection-langXXX)

>From that we do build about 40 *binary* packages, where each binary
package more or less matches with the original TeX Live collections.

I don't know how FreeBSD works, but from what you wrote:
- you can host your own *source* files
- you can build arbitrary binary packages from it (assumption of mine)
it sounds that this should be possible for you, too.

Hope that helps.

Last but not least, I **strongly** suggest continuing this discussion
on
	tldistro at tug.org
since this is the *best* place to discuss this topic. This mailing list
is for "upstream" TeX Live.

All the best

Norbert

--
PREINING Norbert                              https://www.preining.info
Mercari Inc.     +     IFMGA Guide     +     TU Wien     +     TeX Live
GPG: 0x860CDC13   fp: F7D8 A928 26E3 16A1 9FA0 ACF0 6CAC A448 860C DC13

From cezary.wagner at gmail.com  Mon Apr 10 13:20:39 2023
From: cezary.wagner at gmail.com (Cezary Wagner)
Date: Mon, 10 Apr 2023 13:20:39 +0200
Subject: Bug in Windows Installer - unable to change installation path
Message-ID: <CAGAHrWGUArVCVz+ALU625XFhZ6ETVi0VfWmskXj63BGQQHLuPQ@mail.gmail.com>

Hi,

Problem is:
1. download installer
2. run
3. try to change path.

No problem in 2022.

[image: obraz.png]


[image: obraz.png]
error writing "file1afbe1d12a0": broken pipe
error writing "file1afbe1d12a0": broken pipe
    while executing
"chan puts $::inst "checkdir""
    (procedure "update_full_path" line 6)
    invoked from within
"update_full_path"
    (procedure "texdir_setup" line 88)
    invoked from within
"texdir_setup"
    invoked from within
".tdirb invoke "
    invoked from within
".tdirb instate !disabled { .tdirb invoke } "
    invoked from within
".tdirb instate pressed { .tdirb state !pressed; .tdirb instate !disabled {
.tdirb invoke } } "
    (command bound to event)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230410/d5a5d7fb/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: obraz.png
Type: image/png
Size: 12748 bytes
Desc: not available
URL: <https://tug.org/pipermail/tex-live/attachments/20230410/d5a5d7fb/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: obraz.png
Type: image/png
Size: 33054 bytes
Desc: not available
URL: <https://tug.org/pipermail/tex-live/attachments/20230410/d5a5d7fb/attachment-0003.png>

From norbert at preining.info  Mon Apr 10 16:00:56 2023
From: norbert at preining.info (Norbert Preining)
Date: Mon, 10 Apr 2023 23:00:56 +0900
Subject: TeX Live repo stats 2023
Message-ID: <ZDQWmC1YgylPMi8p@burischnitzel>

Hi all,

I have updated the TeX Live repo stats for 2022 (including data till
yesterday). Please see
	http://www.texlive.info/tlstats/2022/
for details.

Best

Norbert

--
PREINING Norbert                              https://www.preining.info
Mercari Inc.     +     IFMGA Guide     +     TU Wien     +     TeX Live
GPG: 0x860CDC13   fp: F7D8 A928 26E3 16A1 9FA0 ACF0 6CAC A448 860C DC13

From zdenek.wagner at gmail.com  Mon Apr 10 18:29:37 2023
From: zdenek.wagner at gmail.com (Zdenek Wagner)
Date: Mon, 10 Apr 2023 18:29:37 +0200
Subject: Bug in Windows Installer - unable to change installation path
In-Reply-To: <CAGAHrWGUArVCVz+ALU625XFhZ6ETVi0VfWmskXj63BGQQHLuPQ@mail.gmail.com>
References: <CAGAHrWGUArVCVz+ALU625XFhZ6ETVi0VfWmskXj63BGQQHLuPQ@mail.gmail.com>
Message-ID: <CAC1phyabcfDVcZ0eRS6+HoekDsBDWLnJW0n36vcbG4bskPeGcA@mail.gmail.com>

Is C:\texlive writable by the user who installs it?

Zden?k Wagner
https://www.zdenek-wagner.eu/


po 10. 4. 2023 v 14:17 odes?latel Cezary Wagner <cezary.wagner at gmail.com>
napsal:

> Hi,
>
> Problem is:
> 1. download installer
> 2. run
> 3. try to change path.
>
> No problem in 2022.
>
> [image: obraz.png]
>
>
> [image: obraz.png]
> error writing "file1afbe1d12a0": broken pipe
> error writing "file1afbe1d12a0": broken pipe
>     while executing
> "chan puts $::inst "checkdir""
>     (procedure "update_full_path" line 6)
>     invoked from within
> "update_full_path"
>     (procedure "texdir_setup" line 88)
>     invoked from within
> "texdir_setup"
>     invoked from within
> ".tdirb invoke "
>     invoked from within
> ".tdirb instate !disabled { .tdirb invoke } "
>     invoked from within
> ".tdirb instate pressed { .tdirb state !pressed; .tdirb instate !disabled
> { .tdirb invoke } } "
>     (command bound to event)
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230410/66d02d4b/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: obraz.png
Type: image/png
Size: 12748 bytes
Desc: not available
URL: <https://tug.org/pipermail/tex-live/attachments/20230410/66d02d4b/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: obraz.png
Type: image/png
Size: 33054 bytes
Desc: not available
URL: <https://tug.org/pipermail/tex-live/attachments/20230410/66d02d4b/attachment-0003.png>

From siepo at bitmuis.nl  Mon Apr 10 20:55:15 2023
From: siepo at bitmuis.nl (Siep Kroonenberg)
Date: Mon, 10 Apr 2023 20:55:15 +0200
Subject: Bug in Windows Installer - unable to change installation path
In-Reply-To: <CAC1phyabcfDVcZ0eRS6+HoekDsBDWLnJW0n36vcbG4bskPeGcA@mail.gmail.com>
References: <CAGAHrWGUArVCVz+ALU625XFhZ6ETVi0VfWmskXj63BGQQHLuPQ@mail.gmail.com>
 <CAC1phyabcfDVcZ0eRS6+HoekDsBDWLnJW0n36vcbG4bskPeGcA@mail.gmail.com>
Message-ID: <ZDRbk1xGab6fWQ+W@bitmuis.nl>

On Mon, Apr 10, 2023 at 06:29:37PM +0200, Zdenek Wagner wrote:
> Is C:\texlive writable by the user who installs it?

If not, that should merely lead to a warning 'Cannot be created or
cannot be written to', but not to a crash. Also, there is plenty of
disk space available.

-- 
Siep Kroonenberg

From usenet at siart.de  Tue Apr 11 10:29:23 2023
From: usenet at siart.de (Uwe Siart)
Date: Tue, 11 Apr 2023 10:29:23 +0200
Subject: [tlmgr]: mtxrun and context no longer executed
Message-ID: <86ttxmrfdo.fsf@siart.de>

After today's infra update the two post-update commands

mtxrun --generate
context --luatex --generate

are no longer executed. At least I did not see them in the messages.
Don't know if it's intentional or not.

-- 
Uwe


From norbert at preining.info  Tue Apr 11 10:35:34 2023
From: norbert at preining.info (Norbert Preining)
Date: Tue, 11 Apr 2023 17:35:34 +0900
Subject: [tlmgr]: mtxrun and context no longer executed
In-Reply-To: <86ttxmrfdo.fsf@siart.de>
References: <86ttxmrfdo.fsf@siart.de>
Message-ID: <ZDUb1j02PcmyalbE@bulldog>

> Don't know if it's intentional or not.

It is intentional ;-)
The cache will be automatically updated when you run context.

Best

Norbert

--
PREINING Norbert                              https://www.preining.info
Mercari Inc.     +     IFMGA Guide     +     TU Wien     +     TeX Live
GPG: 0x860CDC13   fp: F7D8 A928 26E3 16A1 9FA0 ACF0 6CAC A448 860C DC13

From mseven at telus.net  Tue Apr 11 12:00:36 2023
From: mseven at telus.net (Max Chernoff)
Date: Tue, 11 Apr 2023 04:00:36 -0600
Subject: [tlmgr]: mtxrun and context no longer executed
In-Reply-To: <86ttxmrfdo.fsf@siart.de>
Message-ID: <ccd42639d50bd5ca19d3535c2749104e9e813188.camel@telus.net>

Hi Uwe,

> After today's infra update the two post-update commands
> 
> mtxrun --generate
> context --luatex --generate
> 
> are no longer executed. At least I did not see them in the messages.
> Don't know if it's intentional or not.

As Norbert said, this is intentional. kpathsea-based TeX keeps per-tree
filename databases (1 ls-R file in each tree), but ConTeXt keeps per-
user filename databases (cache files in ~/.texlive2023/texmf-var/). 

If you had a system-wide installation of TL, then tlmgr would run
"mtxrun --generate" with the system account (root/Admin). This would
only update the cache for the system account which is both (1)
pointless, since no one should ever be typesetting as root, and (2)
ineffective, since all the per-user caches wouldn't be updated.

The new code has ConTeXt compare the modified date of its cache with the
modified date of the ls-R files every time that you compile a document.
If the ConTeXt date is older, then it regenerates the caches at startup.

So now running "mktexlsr" will also update the ConTeXt cache
(indirectly). The only downside is that the first time you run ConTeXt
after an update, it'll take about 2 seconds longer since it has to
rebuild the cache.

-- Max


From siepo at bitmuis.nl  Tue Apr 11 17:32:59 2023
From: siepo at bitmuis.nl (Siep Kroonenberg)
Date: Tue, 11 Apr 2023 17:32:59 +0200
Subject: Bug in Windows Installer - unable to change installation path
In-Reply-To: <CAGAHrWGUArVCVz+ALU625XFhZ6ETVi0VfWmskXj63BGQQHLuPQ@mail.gmail.com>
References: <CAGAHrWGUArVCVz+ALU625XFhZ6ETVi0VfWmskXj63BGQQHLuPQ@mail.gmail.com>
Message-ID: <ZDV9qzNucYBO+vUI@bitmuis.nl>

On Mon, Apr 10, 2023 at 01:20:39PM +0200, Cezary Wagner wrote:
> Hi,
> 
> Problem is:
> 1. download installer
> 2. run
> 3. try to change path.
> 
> No problem in 2022.
> 
> [image: obraz.png]
> 
> 
> [image: obraz.png]
> error writing "file1afbe1d12a0": broken pipe
> error writing "file1afbe1d12a0": broken pipe
>     while executing
> "chan puts $::inst "checkdir""
>     (procedure "update_full_path" line 6)
>     invoked from within
> "update_full_path"
>     (procedure "texdir_setup" line 88)
>     invoked from within
> "texdir_setup"
>     invoked from within
> ".tdirb invoke "
>     invoked from within
> ".tdirb instate !disabled { .tdirb invoke } "
>     invoked from within
> ".tdirb instate pressed { .tdirb state !pressed; .tdirb instate !disabled {
> .tdirb invoke } } "
>     (command bound to event)

'broken pipe' means that the back end, i.e. the perl script
install-tl, crashed. See whether you get better feedback when you
run the installer in text mode:

install-tl-windows.bat -no-gui

Also, the installer assumes that it is in step with the online
repository. So, if you downloaded the repository yesterday, then you
should not try to run it today but re-download it.

-- 
Siep Kroonenberg

From cezary.wagner at gmail.com  Tue Apr 11 18:34:16 2023
From: cezary.wagner at gmail.com (Cezary Wagner)
Date: Tue, 11 Apr 2023 18:34:16 +0200
Subject: Bug in Windows Installer - unable to change installation path
In-Reply-To: <ZDV9qzNucYBO+vUI@bitmuis.nl>
References: <CAGAHrWGUArVCVz+ALU625XFhZ6ETVi0VfWmskXj63BGQQHLuPQ@mail.gmail.com>
 <ZDV9qzNucYBO+vUI@bitmuis.nl>
Message-ID: <CAGAHrWG=tKjLLtUQhvHv+G-29JKpvDN_4EedsAFfq1K7Ow5QzQ@mail.gmail.com>

It works in text mode but not with gui.

wt., 11 kwi 2023 o 17:33 Siep Kroonenberg <siepo at bitmuis.nl> napisa?(a):

> On Mon, Apr 10, 2023 at 01:20:39PM +0200, Cezary Wagner wrote:
> > Hi,
> >
> > Problem is:
> > 1. download installer
> > 2. run
> > 3. try to change path.
> >
> > No problem in 2022.
> >
> > [image: obraz.png]
> >
> >
> > [image: obraz.png]
> > error writing "file1afbe1d12a0": broken pipe
> > error writing "file1afbe1d12a0": broken pipe
> >     while executing
> > "chan puts $::inst "checkdir""
> >     (procedure "update_full_path" line 6)
> >     invoked from within
> > "update_full_path"
> >     (procedure "texdir_setup" line 88)
> >     invoked from within
> > "texdir_setup"
> >     invoked from within
> > ".tdirb invoke "
> >     invoked from within
> > ".tdirb instate !disabled { .tdirb invoke } "
> >     invoked from within
> > ".tdirb instate pressed { .tdirb state !pressed; .tdirb instate
> !disabled {
> > .tdirb invoke } } "
> >     (command bound to event)
>
> 'broken pipe' means that the back end, i.e. the perl script
> install-tl, crashed. See whether you get better feedback when you
> run the installer in text mode:
>
> install-tl-windows.bat -no-gui
>
> Also, the installer assumes that it is in step with the online
> repository. So, if you downloaded the repository yesterday, then you
> should not try to run it today but re-download it.
>
> --
> Siep Kroonenberg
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230411/db2d2fa5/attachment.html>

From cezary.wagner at gmail.com  Tue Apr 11 18:36:07 2023
From: cezary.wagner at gmail.com (Cezary Wagner)
Date: Tue, 11 Apr 2023 18:36:07 +0200
Subject: Bug in Windows Installer - unable to change installation path
In-Reply-To: <CAC1phyabcfDVcZ0eRS6+HoekDsBDWLnJW0n36vcbG4bskPeGcA@mail.gmail.com>
References: <CAGAHrWGUArVCVz+ALU625XFhZ6ETVi0VfWmskXj63BGQQHLuPQ@mail.gmail.com>
 <CAC1phyabcfDVcZ0eRS6+HoekDsBDWLnJW0n36vcbG4bskPeGcA@mail.gmail.com>
Message-ID: <CAGAHrWH31MtCfUrSi7+FKQSq50n9GYfuPo_Zd+Oase_6AJrJ9w@mail.gmail.com>

Probably not since I do no want install here - but in c:\root\texlive.

I try to change destination.

pon., 10 kwi 2023 o 18:30 Zdenek Wagner <zdenek.wagner at gmail.com>
napisa?(a):

> Is C:\texlive writable by the user who installs it?
>
> Zden?k Wagner
> https://www.zdenek-wagner.eu/
>
>
> po 10. 4. 2023 v 14:17 odes?latel Cezary Wagner <cezary.wagner at gmail.com>
> napsal:
>
>> Hi,
>>
>> Problem is:
>> 1. download installer
>> 2. run
>> 3. try to change path.
>>
>> No problem in 2022.
>>
>> [image: obraz.png]
>>
>>
>> [image: obraz.png]
>> error writing "file1afbe1d12a0": broken pipe
>> error writing "file1afbe1d12a0": broken pipe
>>     while executing
>> "chan puts $::inst "checkdir""
>>     (procedure "update_full_path" line 6)
>>     invoked from within
>> "update_full_path"
>>     (procedure "texdir_setup" line 88)
>>     invoked from within
>> "texdir_setup"
>>     invoked from within
>> ".tdirb invoke "
>>     invoked from within
>> ".tdirb instate !disabled { .tdirb invoke } "
>>     invoked from within
>> ".tdirb instate pressed { .tdirb state !pressed; .tdirb instate !disabled
>> { .tdirb invoke } } "
>>     (command bound to event)
>>
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230411/0f58b11b/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: obraz.png
Type: image/png
Size: 12748 bytes
Desc: not available
URL: <https://tug.org/pipermail/tex-live/attachments/20230411/0f58b11b/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: obraz.png
Type: image/png
Size: 33054 bytes
Desc: not available
URL: <https://tug.org/pipermail/tex-live/attachments/20230411/0f58b11b/attachment-0003.png>

From cezary.wagner at gmail.com  Tue Apr 11 18:37:51 2023
From: cezary.wagner at gmail.com (Cezary Wagner)
Date: Tue, 11 Apr 2023 18:37:51 +0200
Subject: Bug in Windows Installer - unable to change installation path
In-Reply-To: <ZDRbk1xGab6fWQ+W@bitmuis.nl>
References: <CAGAHrWGUArVCVz+ALU625XFhZ6ETVi0VfWmskXj63BGQQHLuPQ@mail.gmail.com>
 <CAC1phyabcfDVcZ0eRS6+HoekDsBDWLnJW0n36vcbG4bskPeGcA@mail.gmail.com>
 <ZDRbk1xGab6fWQ+W@bitmuis.nl>
Message-ID: <CAGAHrWEpqU48bGTmceF-nKW9bJ=WYFA4A7BCEtVft++4qYOXjA@mail.gmail.com>

There is a lot of space and I do not want to install but change the
destination folder - it fails when I try to change the folder not when I
start to install.

pon., 10 kwi 2023 o 20:55 Siep Kroonenberg <siepo at bitmuis.nl> napisa?(a):

> On Mon, Apr 10, 2023 at 06:29:37PM +0200, Zdenek Wagner wrote:
> > Is C:\texlive writable by the user who installs it?
>
> If not, that should merely lead to a warning 'Cannot be created or
> cannot be written to', but not to a crash. Also, there is plenty of
> disk space available.
>
> --
> Siep Kroonenberg
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230411/5f9e7c32/attachment.html>

From siepo at bitmuis.nl  Tue Apr 11 20:49:24 2023
From: siepo at bitmuis.nl (Siep Kroonenberg)
Date: Tue, 11 Apr 2023 20:49:24 +0200
Subject: Bug in Windows Installer - unable to change installation path
In-Reply-To: <CAGAHrWEpqU48bGTmceF-nKW9bJ=WYFA4A7BCEtVft++4qYOXjA@mail.gmail.com>
References: <CAGAHrWGUArVCVz+ALU625XFhZ6ETVi0VfWmskXj63BGQQHLuPQ@mail.gmail.com>
 <CAC1phyabcfDVcZ0eRS6+HoekDsBDWLnJW0n36vcbG4bskPeGcA@mail.gmail.com>
 <ZDRbk1xGab6fWQ+W@bitmuis.nl>
 <CAGAHrWEpqU48bGTmceF-nKW9bJ=WYFA4A7BCEtVft++4qYOXjA@mail.gmail.com>
Message-ID: <ZDWrtHgMKWSnAc55@bitmuis.nl>

On Tue, Apr 11, 2023 at 06:37:51PM +0200, Cezary Wagner wrote:
> There is a lot of space and I do not want to install but change the
> destination folder - it fails when I try to change the folder not when I
> start to install.

I could reproduce your problem when running the install script from
an MSYS2 terminal window, and also from a regular command-prompt
with msys on the searchpath.

If this is indeed the problem: keep msys/cygwin from your path
when installing TL, or, if you have cygwin, try the unix/linux
installer, which supports cygwin binaries.

-- 
Siep Kroonenberg

From cezary.wagner at gmail.com  Tue Apr 11 23:36:35 2023
From: cezary.wagner at gmail.com (Cezary Wagner)
Date: Tue, 11 Apr 2023 23:36:35 +0200
Subject: Bug in Windows Installer - unable to change installation path
In-Reply-To: <ZDWrtHgMKWSnAc55@bitmuis.nl>
References: <CAGAHrWGUArVCVz+ALU625XFhZ6ETVi0VfWmskXj63BGQQHLuPQ@mail.gmail.com>
 <CAC1phyabcfDVcZ0eRS6+HoekDsBDWLnJW0n36vcbG4bskPeGcA@mail.gmail.com>
 <ZDRbk1xGab6fWQ+W@bitmuis.nl>
 <CAGAHrWEpqU48bGTmceF-nKW9bJ=WYFA4A7BCEtVft++4qYOXjA@mail.gmail.com>
 <ZDWrtHgMKWSnAc55@bitmuis.nl>
Message-ID: <CAGAHrWGbv1AW3s0bGUtCpZdw8V17C2chUEVmfTEcU68tSzv_mw@mail.gmail.com>

It is possible that cygwin has an impact since I am using it.
Maybe I will check soon.

wt., 11 kwi 2023 o 20:49 Siep Kroonenberg <siepo at bitmuis.nl> napisa?(a):

> On Tue, Apr 11, 2023 at 06:37:51PM +0200, Cezary Wagner wrote:
> > There is a lot of space and I do not want to install but change the
> > destination folder - it fails when I try to change the folder not when I
> > start to install.
>
> I could reproduce your problem when running the install script from
> an MSYS2 terminal window, and also from a regular command-prompt
> with msys on the searchpath.
>
> If this is indeed the problem: keep msys/cygwin from your path
> when installing TL, or, if you have cygwin, try the unix/linux
> installer, which supports cygwin binaries.
>
> --
> Siep Kroonenberg
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230411/ad0a2fe9/attachment.html>

From werner at suse.de  Wed Apr 12 12:46:56 2023
From: werner at suse.de (Dr. Werner Fink)
Date: Wed, 12 Apr 2023 12:46:56 +0200
Subject: Rare fails of bibtex8 and/or bibtexu test
Message-ID: <ZDaMIEZNchsfVgrl@boole.suse.de>

Hi,

I face rare and random fails of tests/bibtex8.test and/or tests/bibtexu.test
... e.g. this on an e.g. x86_64, the logs looks like (I've added an ERR trap
to catch this)

 [ 1028s] FAIL: tests/bibtex8
 [ 1028s] ===================
 [ 1028s] 
 [ 1028s] #! /bin/sh -vx
 [ 1028s] # $Id: bibtex8.test 63257 2022-05-08 13:25:05Z takuji $
 [ 1028s] # Copyright 2017 Karl Berry <tex-live at tug.org>
 [ 1028s] # Copyright 2010 Peter Breitenlohner <tex-live at tug.org>
 [ 1028s] # You may freely use, modify and/or distribute this file.
 [ 1028s] 
 [ 1028s] test -d tests || mkdir -p tests
 [ 1028s] + test -d tests
 [ 1028s] rm -f tests/xexampl.*
 [ 1028s] + rm -f 'tests/xexampl.*'
 [ 1028s] 
 [ 1028s] cp $srcdir/../web2c/tests/exampl.aux tests/xexampl.aux
 [ 1028s] + cp ../../../texk/bibtex-x/../web2c/tests/exampl.aux tests/xexampl.aux
 [ 1028s] 
 [ 1028s] TEXMFCNF=$srcdir/../kpathsea \
 [ 1028s]   BSTINPUTS=$srcdir/../web2c/tests \
 [ 1028s]   BIBINPUTS=$srcdir/../web2c/tests \
 [ 1028s]   ./bibtex8 -7 tests/xexampl || test $? = 1 || exit 1
 [ 1028s] + TEXMFCNF=../../../texk/bibtex-x/../kpathsea
 [ 1028s] + BSTINPUTS=../../../texk/bibtex-x/../web2c/tests
 [ 1028s] + BIBINPUTS=../../../texk/bibtex-x/../web2c/tests
 [ 1028s] + ./bibtex8 -7 tests/xexampl
 [...]
 [ 1028s] The top-level auxiliary file: tests/xexampl.aux
 [ 1028s] The style file: apalike.bst
 [ 1028s] Database file #1: xampl.bib
 [ 1028s] Warning--empty author in whole-journal
 [ 1028s] Warning--empty title in whole-journal
 [ 1028s] Warning--empty year in booklet-minimal
 [ 1028s] Warning--empty year in manual-minimal
 [ 1028s] Warning--empty year in misc-minimal
 [ 1028s] Warning--empty year in unpublished-minimal
 [ 1028s] Warning--empty year in random-note-crossref
 [ 1028s] (There were 7 warnings)
 [ 1028s] + test 1 = 1
 [ 1028s] diff $srcdir/tests/exampl.bbl tests/xexampl.bbl || exit 2
 [ 1028s] + diff ../../../texk/bibtex-x/tests/exampl.bbl tests/xexampl.bbl
 [ 1028s] 1,212d0
 [ 1028s] < \newcommand{\noopsort}[1]{} \newcommand{\printfirst}[2]{#1}
 [ 1028s] <   \newcommand{\singleletter}[1]{#1} \newcommand{\switchargs}[2]{#2#1}
 [ 1028s] < \begin{thebibliography}{}
 [ 1028s] < 
 [...]
 [ 1028s] < \end{thebibliography}
 [ 1028s] + exit 2
 [ 1028s] FAIL tests/bibtex8.test (exit status: 2)
 [ 1028s] 

No idea what the cause of the empty tests/xexampl.bbl is.  Should I increase
string values of bibtex

 % BibTeX only (max_strings also determines hash_size and hash_prime).
 ent_str_size = 500
 glob_str_size = 200000
 max_strings.bibtex  = 200000
 max_strings.bibtex8 = 200000
 max_strings.bibtexu = 200000
 max_strings.pbibtex = 200000
 max_strings.upbibtex= 200000

at least the stack size is set to unlimited for the full build process.

Werner

-- 
  "Having a smoking section in a restaurant is like having
          a peeing section in a swimming pool." -- Edward Burr
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 894 bytes
Desc: not available
URL: <https://tug.org/pipermail/tex-live/attachments/20230412/05fce149/attachment.sig>

From karl at freefriends.org  Wed Apr 12 19:59:33 2023
From: karl at freefriends.org (Karl Berry)
Date: Wed, 12 Apr 2023 11:59:33 -0600
Subject: Rare fails of bibtex8 and/or bibtexu test
In-Reply-To: <ZDaMIEZNchsfVgrl@boole.suse.de>
Message-ID: <202304121759.33CHxXGc025392@freefriends.org>

    I face rare and random fails of tests/bibtex8.test and/or
    tests/bibtexu.test

Maybe it's a timing/parallelism/caching/something problem.
Are you running with make -j?

I see bibtex8.test and bibtexu.test are using the same input and output
filenames. I just changed bibtexu.test to change that (r66832). Hope
that helps.

    Should I increase string values of bibtex

No. These test bibs are tiny. --thanks, karl.

From jfine2358 at gmail.com  Wed Apr 12 21:30:39 2023
From: jfine2358 at gmail.com (Jonathan Fine)
Date: Wed, 12 Apr 2023 20:30:39 +0100
Subject: Thu 13 Apr: TeX Hour: unlatex; Fermi estimate for reprocessing the
 arXiv: 6:30 to 7:30 BST
In-Reply-To: <CALD=Yf8TsJ77ifJjqTDhxHrpctr4K7uGp2GiRjZwmQULRG5ZGQ@mail.gmail.com>
References: <CALD=Yf8DdYSpjocehvkZfu1xyWK4fJaCSrZacwsUGg2C3o2F_A@mail.gmail.com>
 <CALD=Yf8TsJ77ifJjqTDhxHrpctr4K7uGp2GiRjZwmQULRG5ZGQ@mail.gmail.com>
Message-ID: <CALD=Yf_bfxi4Q0wrLBfX9zOOcBuv6EGtfVe6WPBvkd91o53WGA@mail.gmail.com>

Hi

Well, next week, on Monday 17 April 1:00-5:00pm Eastern Time, we have the
very first arXiv Access Forum. I'm looking forward to that. I'm most
grateful to the organisers, and I hope they're not overwhelmed. They have a
massive responsibility. As do the esteemed presenters and panelists, and
the many participants.

Tomorrow's TeX Hour (Thursday 6:30 to 7:30pm UK Summer Time) is about my
emerging unlatex tool for reprocessing TeX documents, to provide more
accessible outputs. I'm close to creating in Python an equivalent to TeX's
internal boxes. This involves an interesting parser + builder combination,
linked by a stream of control symbols, constructors and leaf nodes.

Here are some arXiv stats (in round numbers):

Total number of submissions: 2.25 million.
Downloads per month: 25 million.
Seconds in a month: 2.6 million.
Registered for arXiv Access Forum: 2,000 people.

Why seconds in a month? Well, it's approximately equal to the total number
of submissions. So we can make a Fermi estimate as to how long it will take
to reprocess the entire arXiv to get accessible outputs (assuming suitable
software).

Suppose we have a desktop PC with 12 cores, so 24 threads, so about 20
cores doing useful work. On such a machine, if not bottlenecked, we could
do the whole lot in a month provided each item takes only 20 seconds. The
download might take a while, and the electricity would be about ?150 (or
$150).

Harder is to make a Fermi estimate for creating suitable software, and yet
harder is writing and testing the software. Also very important is field
testing of its outputs for accessibility..

Here's the URL for Monday's arXiv forum:
https://accessibility2023.arxiv.org/
The TeX Hour zoom URL:
https://us02web.zoom.us/j/78551255396?pwd=cHdJN0pTTXRlRCtSd1lCTHpuWmNIUT09
The home page tomorrow's TeX Hour:
https://texhour.github.io/2023/04/13/unlatex-results-prospects/

wishing you happy arXiving

Jonathan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230412/96cb79ee/attachment.html>

From werner at suse.de  Thu Apr 13 08:18:02 2023
From: werner at suse.de (Dr. Werner Fink)
Date: Thu, 13 Apr 2023 08:18:02 +0200
Subject: Rare fails of bibtex8 and/or bibtexu test
In-Reply-To: <202304121759.33CHxXGc025392@freefriends.org>
References: <ZDaMIEZNchsfVgrl@boole.suse.de>
 <202304121759.33CHxXGc025392@freefriends.org>
Message-ID: <ZDeemi+PN35k57OS@boole.suse.de>

On 2023/04/12 11:59:33 -0600, Karl Berry wrote:
>     I face rare and random fails of tests/bibtex8.test and/or
>     tests/bibtexu.test
> 
> Maybe it's a timing/parallelism/caching/something problem.
> Are you running with make -j?

Yep our OBS aka autobuild system uses VMs with different numbers of
CPUs mostly 8 and therefore -j8 ... and this leads to random order
of the bibtex[u8] tests.

Meanwhile I've a patch

----------------------------------------------------------------------
--- texlive-20230311-source/texk/bibtex-x/Makefile.in
+++ texlive-20230311-source/texk/bibtex-x/Makefile.in   2023-04-12 13:47:24.619293127 +0000
@@ -13,7 +13,7 @@
 # PARTICULAR PURPOSE.
 
 @SET_MAKE@
-
+.NOTPARALLEL:
 
 VPATH = @srcdir@
 am__is_gnu_make = { \
----------------------------------------------------------------------

with this the order of the tests are repeatable for every run
and every architecture.

> 
> I see bibtex8.test and bibtexu.test are using the same input and output
> filenames. I just changed bibtexu.test to change that (r66832). Hope
> that helps.

That also avoid this, indeed.

> 
>     Should I increase string values of bibtex
> 
> No. These test bibs are tiny. --thanks, karl.

Thanks a lot

Werner

-- 
  "Having a smoking section in a restaurant is like having
          a peeing section in a swimming pool." -- Edward Burr
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 894 bytes
Desc: not available
URL: <https://tug.org/pipermail/tex-live/attachments/20230413/af5e08e4/attachment.sig>

From christophe.jorssen at gmail.com  Fri Apr 14 14:38:53 2023
From: christophe.jorssen at gmail.com (Christophe Jorssen)
Date: Fri, 14 Apr 2023 14:38:53 +0200
Subject: How should I deal with packages that require FFI external libraries?
Message-ID: <CAN1usO2ePAALUzU3h1ceXuNodv_ia3t_0rY7KZ42DMt2ojANzg@mail.gmail.com>

Hello all,

I hope this is the right place to post.

I'm currently writing a generic luatex package that relies on an
external library (https://github.com/CoolProp/CoolProp) called via the
FFI library in luatex (WIP:
https://github.com/cjorssen/luatex-coolprop). I'm not an expert at
compiling complex projects (I can read a `README` file and type
`make`), neither at binding C to lua, nor at the subtleties of OS and
architectures. So please forgive me if I don't use the correct
vocabulary and/or if I misunderstand something.

I work under linux but I'd like to make the package available for
windows and macos.

1) When the external library is loaded, I have something like:

lib = ffi.load('<path to the .so file>')

For now, the path is hardcoded. Where should I put the .so file so
that luatex finds it? What is good practice for CTAN/TeXLive? Does
luatex rely on kpsewhich for that kind of thing?

2) Can I distribute the .so file (or .dll) with the CTAN/TeXLive
package or should it be compiled on the user computer?

3) Does FFI work in luatex on all architectures supported by TeXlive?
By the way, I cannot find any reference to the ffi library in the
luatex manual: I missed something?

4) Can I compile the library for windows and macos on my linux
computer? (Sorry, I know it is out of scope.)

5) Regarding security, I don't think it is an issue since ffi est
shell-escape protected by default in TeXLive. But what do you think?

6) Regarding the licence: the library has an MIT licence. Is it ok for TeXLive?

Many thanks.

Best regards

-- 
Christophe

From norbert at preining.info  Fri Apr 14 17:37:09 2023
From: norbert at preining.info (Norbert Preining)
Date: Sat, 15 Apr 2023 00:37:09 +0900
Subject: How should I deal with packages that require FFI external
 libraries?
In-Reply-To: <CAN1usO2ePAALUzU3h1ceXuNodv_ia3t_0rY7KZ42DMt2ojANzg@mail.gmail.com>
References: <CAN1usO2ePAALUzU3h1ceXuNodv_ia3t_0rY7KZ42DMt2ojANzg@mail.gmail.com>
Message-ID: <ZDlzJZVaW-4bNKeL@burischnitzel>

Hi Christophe,

> I hope this is the right place to post.

I guess the luatex-dev mailing list is better for that, but I try to
answer what I can.

> I'm currently writing a generic luatex package that relies on an
> external library (https://github.com/CoolProp/CoolProp) called via the
> FFI library in luatex (WIP:

Good luck.

> https://github.com/cjorssen/luatex-coolprop). I'm not an expert at
> compiling complex projects (I can read a `README` file and type
> `make`), neither at binding C to lua, nor at the subtleties of OS and

Then maybe doing a ffi based library development might be rather on the
challenging side...

> 1) When the external library is loaded, I have something like:
> 
> lib = ffi.load('<path to the .so file>')
> 
> For now, the path is hardcoded. Where should I put the .so file so
> that luatex finds it? What is good practice for CTAN/TeXLive? Does
> luatex rely on kpsewhich for that kind of thing?

No idea. I don't think kpsewhich is used, but the normal ldconfig /
ld.so library path. But I might be wrong.

> 2) Can I distribute the .so file (or .dll) with the CTAN/TeXLive
> package or should it be compiled on the user computer?

Compilation on user computer is out of discussion, since you cannot know
whether a compiler is installed.

> 3) Does FFI work in luatex on all architectures supported by TeXlive?
> By the way, I cannot find any reference to the ffi library in the
> luatex manual: I missed something?

Please ask on the luatex-dev or luatex user lists.

> 4) Can I compile the library for windows and macos on my linux
> computer? (Sorry, I know it is out of scope.)

I guess there are some ways to cross compile, but that is rather tricky.

> 5) Regarding security, I don't think it is an issue since ffi est
> shell-escape protected by default in TeXLive. But what do you think?

Good point, I guess it is ok.

> 6) Regarding the licence: the library has an MIT licence. Is it ok for TeXLive?

Yes.

Best

Norbert

--
PREINING Norbert                              https://www.preining.info
Mercari Inc.     +     IFMGA Guide     +     TU Wien     +     TeX Live
GPG: 0x860CDC13   fp: F7D8 A928 26E3 16A1 9FA0 ACF0 6CAC A448 860C DC13

From karl at freefriends.org  Sat Apr 15 00:31:41 2023
From: karl at freefriends.org (Karl Berry)
Date: Fri, 14 Apr 2023 16:31:41 -0600
Subject: How should I deal with packages that require FFI external
 libraries?
In-Reply-To: <CAN1usO2ePAALUzU3h1ceXuNodv_ia3t_0rY7KZ42DMt2ojANzg@mail.gmail.com>
Message-ID: <202304142231.33EMVfZY019977@freefriends.org>

    1) When the external library is loaded, I have something like:
    lib = ffi.load('<path to the .so file>')

    For now, the path is hardcoded. Where should I put the .so file so
    that luatex finds it? What is good practice for CTAN/TeXLive? Does
    luatex rely on kpsewhich for that kind of thing?

Kpathsea has a path CLUAINPUTS, but I don't know if LuaTeX actually uses
it. When you ask on the luatex list, they can advise.

>From texmf.cnf:
CLUAINPUTS = $TEXMFDOTDIR;$SELFAUTOLOC/lib/{$progname,$engine,}/lua//

    2) Can I distribute the .so file (or .dll) with the CTAN/TeXLive
    package or should it be compiled on the user computer?

TeX Live does not distribute any user-created .so files, and I don't
foresee this changing. I can't speak for whether CTAN is willing to hold
them.

For the rest, I have nothing to add to Norbert's replies.
Sorry I can't be more help ... --best, karl.

From mseven at telus.net  Sat Apr 15 03:29:35 2023
From: mseven at telus.net (Max Chernoff)
Date: Fri, 14 Apr 2023 19:29:35 -0600
Subject: How should I deal with packages that require FFI external
 libraries?
In-Reply-To: <CAN1usO2ePAALUzU3h1ceXuNodv_ia3t_0rY7KZ42DMt2ojANzg@mail.gmail.com>
Message-ID: <9d16472af1514e0a507e6e9f1e6c0350cc11b259.camel@telus.net>



Hi Christophe,

> 

> 
> 1) When the external library is loaded, I have something like:
> 
> lib = ffi.load('<path to the .so file>')
> 
> For now, the path is hardcoded. Where should I put the .so file so
> that luatex finds it? What is good practice for CTAN/TeXLive? Does
> luatex rely on kpsewhich for that kind of thing?

To get the path to the file, the "proper" way is to use

   kpse.find_file("library-name", "clua")

which will look in

   $ kpsewhich --var-value=CLUAINPUTS
   .:/usr/local/texlive/2023/bin/x86_64-linux/lib/{kpsewhich,unsetengine,}/lua//

by default. If you're distributing your package in a .zip/.tar, that's
not the easiest path to add things to, so something like

   kpse.find_file("library-name.so") or kpse.find_file("library-name.dll") 

will likely be easier to distribute (although not technically
"correct").

> 
> 2) Can I distribute the .so file (or .dll) with the CTAN/TeXLive
> package

My full TL installation doesn't have any shared libraries installed

   $ find /usr/local/texlive/2023 -iname '*.so'

so you probably can't distribute it in TL. TLContrib and MiKTeX _might_
accept it, I have no idea. 

There aren't any .so's on CTAN, but there are a few .dll's. All the
.dll's are over 15 years old though, so I have no idea if they'll accept
any new ones. You'll have to ask them.

> 
> 3) Does FFI work in luatex on all architectures supported by TeXlive?

The documentation

   https://github.com/TeX-Live/luatex/tree/master/source/texk/web2c/luatexdir/luaffi

says that it only supports Linux and macOS, x86 and x86_64. There is a
msvc build file and a bunch of files with "win" in the name, so I'm
pretty sure that it supports Windows too. There are also support files
with "arm" in the name, but they're 5 years old so probably Linux only.

> By the way, I cannot find any reference to the ffi library in the
> luatex manual: I missed something?
> 

You are correct, there is no reference to it at all in the manual. I
didn't even know that LuaTeX included an FFI library until this email.

> 4) Can I compile the library for windows and macos on my linux
> computer? (Sorry, I know it is out of scope.)
> 

You can use mingw from Linux to compile for Windows, and there is
probably a clang-based cross-compiler for macOS. Cross-compiling can
range in difficulty from "replace gcc with x86_64-w64-mingw32-gcc and
compile as normal" to a really complicated and time consuming process.

It looks like CoolProp distributes binaries for Linux, macOS, and
Windows, so you could just use those.

> 5) Regarding security, I don't think it is an issue since ffi est
> shell-escape protected by default in TeXLive. But what do you think?
> 

Well there's the same security issue as always: if you enable shell
escape, then any package can do all sorts of bad things to your
computer. If you don't enable shell escape, then you can't use ffi. But
just adding the shared library to your TEXMF trees won't have any
security implications.

> 6) Regarding the licence: the library has an MIT licence. Is it ok for
> TeXLive?
> 

That should be fine, but (as mentioned above) it's unlikely that you'll
get any binaries into TL.

> I'm currently writing a generic luatex package that relies on an
> external library (https://github.com/CoolProp/CoolProp) called via the
> FFI library in luatex

A much simpler way to do all of this would be to use PythonTeX. CoolProp
distributes binary wheels, so you'd just need to tell your users to run 

   $ pip install CoolProp
   
Then, your package code would look something like

   \pyc{import CoolProp}
   \def\dowhatever#1{\py{CoolProp.do_whatever("#1")}
   
and a document would look like

   \usepackage{coolprop}
   ...
   The value of whatever at 55 is \dowhatever{55}.
   
Here are the docs:

   https://texdoc.org/serve/pythontex/2

> I'm not an expert at
> compiling complex projects (I can read a `README` file and type
> `make`), neither at binding C to lua,

You could also write a Lua module in C directly. This would require you
to write some C code, but it may end up being easier than using the ffi
library since (1) all the CoolProp example code is probably in C, and
(2) the compiler would give you proper warnings and errors so you'll
always be sure that you're calling the functions correctly.

The FFI approach should work, but I think that you'll have better luck
using either PythonTeX or a custom C module.

-- Max
> 


From christophe.jorssen at gmail.com  Sat Apr 15 13:24:58 2023
From: christophe.jorssen at gmail.com (Christophe Jorssen)
Date: Sat, 15 Apr 2023 13:24:58 +0200
Subject: How should I deal with packages that require FFI external
 libraries?
In-Reply-To: <9d16472af1514e0a507e6e9f1e6c0350cc11b259.camel@telus.net>
References: <CAN1usO2ePAALUzU3h1ceXuNodv_ia3t_0rY7KZ42DMt2ojANzg@mail.gmail.com>
 <9d16472af1514e0a507e6e9f1e6c0350cc11b259.camel@telus.net>
Message-ID: <CAN1usO2Ta2_feOmURVQHu0f2kPhOaMnKVWVBeJP99KJkvWXMeA@mail.gmail.com>

Thank you Norbert, Karl and Max for your detailed answers.

I continue the discussion here as the main issue is related to
distribution of TeX related materials.

Le sam. 15 avr. 2023 ? 03:29, Max Chernoff <mseven at telus.net> a ?crit :
> >
> > 1) When the external library is loaded, I have something like:
> >
> > lib = ffi.load('<path to the .so file>')
> >
> > For now, the path is hardcoded. Where should I put the .so file so
> > that luatex finds it? What is good practice for CTAN/TeXLive? Does
> > luatex rely on kpsewhich for that kind of thing?
>
> To get the path to the file, the "proper" way is to use
>
>    kpse.find_file("library-name", "clua")
>
> which will look in
>
>    $ kpsewhich --var-value=CLUAINPUTS
>    .:/usr/local/texlive/2023/bin/x86_64-linux/lib/{kpsewhich,unsetengine,}/lua//
>
> by default. If you're distributing your package in a .zip/.tar, that's
> not the easiest path to add things to, so something like
>
>    kpse.find_file("library-name.so") or kpse.find_file("library-name.dll")
>
> will likely be easier to distribute (although not technically
> "correct").
>

Thanks a lot for this suggestion.

> > 2) Can I distribute the .so file (or .dll) with the CTAN/TeXLive
> > package
>
> My full TL installation doesn't have any shared libraries installed
>
>    $ find /usr/local/texlive/2023 -iname '*.so'
>
> so you probably can't distribute it in TL. TLContrib and MiKTeX _might_
> accept it, I have no idea.
>
> There aren't any .so's on CTAN, but there are a few .dll's. All the
> .dll's are over 15 years old though, so I have no idea if they'll accept
> any new ones. You'll have to ask them.
>

I think I understand why a TeX distribution didn't need to distribute
.so or .dll in the past. But with luatex and its "modular" approach
(everything that can be done outside the engine must be done outside,
at format level, with lua or even with ffi), maybe that policy could
change?

I guess there are security reasons or storage reasons or...

As I said earlier, I am not an expert and as a TeXLive user I will
always abide by the rules stated by those who know better (Karl,
Norbert and all the others that are doing an amazing job with the
entire TeX ecosystem).

TLContrib is indeed a good idea (Norbert?) but, in my opinion, not for
the average TeX user I'll talk about below.

> > 3) Does FFI work in luatex on all architectures supported by TeXlive?
>
> The documentation
>
>    https://github.com/TeX-Live/luatex/tree/master/source/texk/web2c/luatexdir/luaffi
>
> says that it only supports Linux and macOS, x86 and x86_64. There is a
> msvc build file and a bunch of files with "win" in the name, so I'm
> pretty sure that it supports Windows too. There are also support files
> with "arm" in the name, but they're 5 years old so probably Linux only.
>

Ok. I'll have a look. But mac M{1,2} won't work easely. Got it.

> > By the way, I cannot find any reference to the ffi library in the
> > luatex manual: I missed something?
>
> You are correct, there is no reference to it at all in the manual. I
> didn't even know that LuaTeX included an FFI library until this email.
>

I got aware of FFI with a couple of questions on tex.stackexchange.com
(https://tex.stackexchange.com/questions/tagged/ffi) and an article in
tugboat written by Henri Menke:

https://tug.org/TUGboat/tb39-1/tb121menke-ffi.pdf

[...]

> > I'm currently writing a generic luatex package that relies on an
> > external library (https://github.com/CoolProp/CoolProp) called via the
> > FFI library in luatex
>
> A much simpler way to do all of this would be to use PythonTeX. CoolProp
> distributes binary wheels, so you'd just need to tell your users to run
>

That would be a nice idea but probably not for the average TeX user I
know. I teach physics and most of my colleagues just want to use LaTeX
out of the box. They often stick with what works for them (even some
really outdated bad practice). If I come to them and say: "Well, there
is a nice new package that can draw advanced thermodynamics diagrams.
All you have to do is update your TeXLive and run luatex with
--shell-escape enabled", maybe they'll give it a try. I know for sure
that if they have to install an entire python ecosystem, use pip
install, understand how pythontex works, mess with their system
python, etc., they just won't use the package.

>
> You could also write a Lua module in C directly. This would require you
> to write some C code, but it may end up being easier than using the ffi
> library since (1) all the CoolProp example code is probably in C, and
> (2) the compiler would give you proper warnings and errors so you'll
> always be sure that you're calling the functions correctly.
>

AFAIU, CoolProp is in C++ but maybe the difference is not relevant
here (I don't know what I'm talking about).

That would not solve the initial problem of distributing the compiled C code?

Thanks again for those detailed answers.

Best regards

-- 
Christophe


From mseven at telus.net  Sat Apr 15 23:32:23 2023
From: mseven at telus.net (Max Chernoff)
Date: Sat, 15 Apr 2023 15:32:23 -0600
Subject: How should I deal with packages that require FFI external
 libraries?
In-Reply-To: <CAN1usO2Ta2_feOmURVQHu0f2kPhOaMnKVWVBeJP99KJkvWXMeA@mail.gmail.com>
References: <CAN1usO2ePAALUzU3h1ceXuNodv_ia3t_0rY7KZ42DMt2ojANzg@mail.gmail.com>
 <9d16472af1514e0a507e6e9f1e6c0350cc11b259.camel@telus.net>
 <CAN1usO2Ta2_feOmURVQHu0f2kPhOaMnKVWVBeJP99KJkvWXMeA@mail.gmail.com>
Message-ID: <cc6445e46c4800e3775cafb1d314eaf7b34aedf8.camel@telus.net>

Hi Christophe,

(Karl and Norbert, a question for you at the bottom)

On Sat, 2023-04-15 at 13:24 +0200, Christophe Jorssen wrote:
> > 
> I think I understand why a TeX distribution didn't need to distribute
> .so or .dll in the past. But with luatex and its "modular" approach
> (everything that can be done outside the engine must be done outside,
> at format level, with lua or even with ffi), maybe that policy could
> change?

(I'm not on the TL team, so this may not be accurate)

There isn't any specific policy against including binary libraries, it's
just (1) TL builds for lots of obscure platforms and you're unlikely to
compile the library for all of them, (2) binaries are less auditable
than interpreted code, and (3) binaries make things more complicated for
downstream distributors. 

If you look at biber and luametatex, they just compile binaries for a
few common platforms then upload them to CTAN/TL. This works pretty
well, and there's no technical reason that you couldn't do the same.
>From a policy level though, I don't think that anyone is too happy about
the biber/luametatex situation and they'd rather not add another similar
package.

> > > 3) Does FFI work in luatex on all architectures supported by TeXlive?
> 
> Ok. I'll have a look. But mac M{1,2} won't work easely. Got it.

If the macOS arm ABI is close enough to the Linux arm ABI, it might just
work. If it doesn't work, you _might_ be able to convince one of the
MacTeX maintainers to add support, but that wouldn't be distributed
until TL2024 at the earliest.

> > 
> I got aware of FFI with a couple of questions on tex.stackexchange.com
> (https://tex.stackexchange.com/questions/tagged/ffi) and an article in
> tugboat written by Henri Menke:
> 
> https://tug.org/TUGboat/tb39-1/tb121menke-ffi.pdf

Interesting. Thanks!
> 

> > You could also write a Lua module in C directly. This would require you
> > to write some C code, but it may end up being easier than using the ffi
> > library since (1) all the CoolProp example code is probably in C, and
> > (2) the compiler would give you proper warnings and errors so you'll
> > always be sure that you're calling the functions correctly.
> 
> That would not solve the initial problem of distributing the compiled C code?

No, it wouldn't really solve any of the problems that you're having.
I've never used the ffi library before, but I've had a pretty good
experience writing Lua modules directly in C. If the ffi library is
working well for you though, then keep doing that. 

> A much simpler way to do all of this would be to use PythonTeX. CoolProp
> distributes binary wheels, so you'd just need to tell your users to run
> 
> > That would be a nice idea but probably not for the average TeX user

Ok, new idea.

If you load your package with no options, then it will try to load
libcoolprop.so from CLUAINPUTS, then from the standard system paths
(/usr/lib, /usr/local/lib, etc.). You can do this with something like

   local path = kpse.find_file("libcoolprop", "clua")
   if path then
       ffi.load(path)
   else
       ffi.load("coolprop")
   end
   
and if it doesn't find the library, it just throws an error and exists.

However, if you load the package with a "download" option, then it will
use luasocket to download "libcoolprop.so" to $TEXMFCACHE, then it will
load the library from there.

This makes it so that non-technical users can easily install the
binaries at runtime, and TL won't need to distribute any binaries.

Karl and Norbert,

Would doing this be okay in a TL package? The library that it's
downloading is open source, the package would only do this if the user
requests it via an option, and the download would be user-local. Doing
this seems fine to me, but I can also understand a policy against
downloading and running external code.

Thanks,
-- Max
   




From luecking at uark.edu  Mon Apr 17 20:50:32 2023
From: luecking at uark.edu (Daniel H. Luecking)
Date: Mon, 17 Apr 2023 18:50:32 +0000
Subject: Bug in gzip (package wintools.windows)
Message-ID: <SN6PR04MB4144743E851C34E275CD986EB99C9@SN6PR04MB4144.namprd04.prod.outlook.com>

For every self-update of tlmgr (so far) I have gotten the following error message from windows (Win 10):

  gzip.exe - Application Error
  The application was unable to start correctly [0xc000007b]
  Click OK to close the application.

I could reproduce the same error under Win 7.

To make sure it was not me at fault, I disabled each locally available version of gzip one by one (and waited for another self-update to arise). Finally, I thought to look for a version of gzip supplied by TeXLive and found gzip.exe in the bin/windows directory. Executing it directly caused the same error, as did gunzip.exe.

Replacing them with locally available versions made the error go away. I also tested the other programs in wintools.windows (just executing "<progname> --help") and found no other obvious problem.

Oddly, all of the self-updates seem to have succeeded without a working gzip.

Cheers,

Daniel H. Luecking [luecking at uark.edu]
Department of Mathematical Sciences
1 University of Arkansas
Fayetteville, AR, USA 72701-1201
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230417/2a3906f2/attachment.html>

From kakuto at jcom.zaq.ne.jp  Tue Apr 18 01:25:20 2023
From: kakuto at jcom.zaq.ne.jp (Akira Kakuto)
Date: Tue, 18 Apr 2023 08:25:20 +0900
Subject: Bug in gzip (package wintools.windows)
In-Reply-To: <SN6PR04MB4144743E851C34E275CD986EB99C9@SN6PR04MB4144.namprd04.prod.outlook.com>
References: <SN6PR04MB4144743E851C34E275CD986EB99C9@SN6PR04MB4144.namprd04.prod.outlook.com>
Message-ID: <97d617bb-c4ea-3412-fcb5-e35d65db8e17@jcom.zaq.ne.jp>

On 2023/04/18 3:50, Daniel H. Luecking wrote:
> For every self-update of tlmgr (so far) I have gotten the following error message from windows (Win 10):
> 
>    gzip.exe - Application Error
>    The application was unable to start correctly [0xc000007b]
>    Click OK to close the application.
> 
> I could reproduce the same error under Win 7.
>

Thanks for the report.
I cannot reproduce the error (win11). But I suspect that it is due to
some dll problem (libwinpthread-1.dll?).
The 64bit gzip.exe seems to depend on libwinpthread-1.dll, and the
32bit gzip.exe does not depend on the dll.
As a test, I replaced by the 32bit gzip.exe (r66874).

Akira

From siepo at bitmuis.nl  Tue Apr 18 10:33:21 2023
From: siepo at bitmuis.nl (Siep Kroonenberg)
Date: Tue, 18 Apr 2023 10:33:21 +0200
Subject: Bug in gzip (package wintools.windows)
In-Reply-To: <97d617bb-c4ea-3412-fcb5-e35d65db8e17@jcom.zaq.ne.jp>
References: <SN6PR04MB4144743E851C34E275CD986EB99C9@SN6PR04MB4144.namprd04.prod.outlook.com>
 <97d617bb-c4ea-3412-fcb5-e35d65db8e17@jcom.zaq.ne.jp>
Message-ID: <ZD5V0TNVW/9XwdsU@bitmuis.nl>

On Tue, Apr 18, 2023 at 08:25:20AM +0900, Akira Kakuto wrote:
> On 2023/04/18 3:50, Daniel H. Luecking wrote:
> > For every self-update of tlmgr (so far) I have gotten the following error message from windows (Win 10):
> > 
> >    gzip.exe - Application Error
> >    The application was unable to start correctly [0xc000007b]
> >    Click OK to close the application.
> > 
> > I could reproduce the same error under Win 7.
> > 
> 
> Thanks for the report.
> I cannot reproduce the error (win11). But I suspect that it is due to
> some dll problem (libwinpthread-1.dll?).
> The 64bit gzip.exe seems to depend on libwinpthread-1.dll, and the
> 32bit gzip.exe does not depend on the dll.
> As a test, I replaced by the 32bit gzip.exe (r66874).
> 
> Akira

I got an error message "The code execution cannot proceed because
libwinpthread-1.dll was not found."

But there is a libwinpthread-1.dll in tlpkg/tlperl/bin. I guess gzip
uses that one when running from tlmgr.

-- 
Siep Kroonenberg

From jfine2358 at gmail.com  Wed Apr 19 20:47:32 2023
From: jfine2358 at gmail.com (Jonathan Fine)
Date: Wed, 19 Apr 2023 19:47:32 +0100
Subject: Thu 20 Apr: TeX Hour: Why paper? Why print? Why PDF? Why HTML?: 6:30
 to 7:30 BST
In-Reply-To: <CALD=Yf8+qMwiHVMr9A8w4MfBDz3vTuOhk0SBAf2y7XXEp8wJLg@mail.gmail.com>
References: <76108cbf-c075-45b9-a31b-32dc88233351n@googlegroups.com>
 <CALD=Yf8+qMwiHVMr9A8w4MfBDz3vTuOhk0SBAf2y7XXEp8wJLg@mail.gmail.com>
Message-ID: <CALD=Yf9UzgotF1mjtWbwbvimPSVhtnfmWsekDcf6qAYRG849wA@mail.gmail.com>

Hi

Tomorrow's TeX Hour uses history to put into context the importance of the
first arXiv Access Forum, which was earlier this week. The arXiv intends to
offer research articles in both PDF and HTML. They retain the current
LaTeX-based production process, and add to another that provides HTML. This
is a major development.

TeX Hour zoom URL:
https://us02web.zoom.us/j/78551255396?pwd=cHdJN0pTTXRlRCtSd1lCTHpuWmNIUT09
Tomorrow's TeX Hour:
https://texhour.github.io/2023/04/20/why-paper-print-pdf-html/
Monday's arXiv forum: https://accessibility2023.arxiv.org/
The TeX Hour zoom URL:
https://us02web.zoom.us/j/78551255396?pwd=cHdJN0pTTXRlRCtSd1lCTHpuWmNIUT09

All are welcome to the TeX Hour. Especially if you bring some memories,
experience or conversation related to print, paper, PDF and HTML.

wishing you well

Jonathan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230419/ff318f41/attachment.html>

From bvoisin at icloud.com  Wed Apr 19 22:41:17 2023
From: bvoisin at icloud.com (Bruno Voisin)
Date: Wed, 19 Apr 2023 22:41:17 +0200
Subject: How should I deal with packages that require FFI external
 libraries?
Message-ID: <8D745C1D-E3BF-415F-BD10-46C2AEB22177@icloud.com>

> Christophe Jorssen wrote:
> 
> 4) Can I compile the library for windows and macos on my linux
> computer? (Sorry, I know it is out of scope.)

> Max Chernoff answered, speaking of FFI:
> 
> There are also support files
> with "arm" in the name, but they're 5 years old so probably Linux only.

> Christophe Jorssen answered:
> 
> Ok. I'll have a look. But mac M{1,2} won't work easely. Got it.

> Then Max Chernoff:
> 
> If the macOS arm ABI is close enough to the Linux arm ABI, it might just
> work. If it doesn't work, you _might_ be able to convince one of the
> MacTeX maintainers to add support, but that wouldn't be distributed
> until TL2024 at the earliest.

Coming in a bit late, and speaking only for the Mac (as one of the MacTeX folks):


- To compile for the Mac, you most likely need two Macs, one with an Intel chip and another with an ARM chip. You compile on both with Xcode (compiler + IDE) or the Command Line Developer Tools (compiler alone), then you combine the two binaries with lipo (part of the developer tools) to create a universal binary.


- Regarding the addition of external libraries into MacTeX, the answer will most certainly be a plain "no".

The future need, owing to the changes in Ghostscript 10.01, to provide for dvisvgm not only libgs (from Ghostscript) but also mutool (from MuPDF), led already to some debate. Finally it was decided to separate libgs from gs and provide it as a separate package with mutool, and see how the community reacts, whether it's worth continuing to provide these additions to basic gs or let the user install them.

The packages are at

https://pages.uoregon.edu/koch/

The concern here is size: mutool adds about 100 MB to the already gigantic MacTeX download. I have colleagues (in a fluid mechanics research department) who complained they did not understand why a TeX install was more than a few hundred magabytes, and asked whether there were alternatives to MacTeX/TeX Live.

Another problem is the complexity of compiling for the Mac, if you want to create a user-friendly install package not requiring the user to bypass os security at install time: you need to sign the software (with a Developer ID), make it a hardened runtime (by explicitly allowing the actions that need to get outside the sandbox, like calling a library signed by another developer or letting the software be affected by environment variables) and notarize it (ie, upload the binary to Apple servers which scan it automatically for potentially harmful actions).

Most likely, if the tendency for TeX packages to require external binaries or libraries persists and develops, the answer (at least on the Mac) will be to require users to install these additions themselves, either by compiling them or using a package manager like MacPorts or Homebrew.


Bruno Voisin

(I'm not on the tex-live list, so I'm more likely to see answers if they're cc'ed to me)

From harald.jele at uni-klu.ac.at  Thu Apr 20 09:32:00 2023
From: harald.jele at uni-klu.ac.at (Harald Jele)
Date: Thu, 20 Apr 2023 09:32:00 +0200
Subject: TeXLive 2023 and local fonts
Message-ID: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>

Dear all,

I deleted my local directory of TeXLive 2022 and installed 2023 as a portable installation in my homedir.
After this I had to recognize that TeXLive does not use my locally installed fonts anymore.
I could reproduce this on a second device, installing TeXLive 2023 the same way.
Is there a way telling TeXLive to search for the fonts again and put them into the fonts database of fontspec??
An example is available here.

https://wwwu.aau.at/hjele/texlive_vs_debian_testing/chapterheading_indent.zip

Thanks for help



Harald
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_0xA909F62C45FC7802.asc
Type: application/pgp-keys
Size: 3232 bytes
Desc: OpenPGP public key
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/3acf824f/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 495 bytes
Desc: OpenPGP digital signature
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/3acf824f/attachment.sig>

From romano at rgtti.com  Thu Apr 20 09:47:16 2023
From: romano at rgtti.com (Romano Giannetti)
Date: Thu, 20 Apr 2023 09:47:16 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
Message-ID: <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>

On 20/04/2023 09:32, Harald Jele wrote:
> Dear all,
> 
> I deleted my local directory of TeXLive 2022 and installed 2023 as a 
> portable installation in my homedir.

This is my configuration too. Where did you put your "locally installed 
fonts" before?

> Is there a way telling TeXLive to search for the fonts again and put 
> them into the fonts database of fontspec??

Basically, with a portable installation, you put all your local things 
into $HOME/texmf, from there they should be picked up automatically (at 
least for pdflatex and company). Lualatex and xelatex should see all the 
font your system see, so basically it depends on how you add a font to 
your system.

Romano

-- 
Romano Giannetti, Ph.D, Electronic Engineer, Professor
http://www.rgtti.com/


From harald.jele at uni-klu.ac.at  Thu Apr 20 11:19:47 2023
From: harald.jele at uni-klu.ac.at (Harald Jele)
Date: Thu, 20 Apr 2023 11:19:47 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
Message-ID: <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>

Dear Romano,

I did not change the directory where TeXLive 2022 already found them: $HOME./local/share/fonts
See the logfile within the ZIP. Lualatex loads but did not use them (Myriad Pro).


Thanks, Harald


Am 20.04.23 um 09:47 schrieb Romano Giannetti:
> On 20/04/2023 09:32, Harald Jele wrote:
>> Dear all,
>>
>> I deleted my local directory of TeXLive 2022 and installed 2023 as a portable installation in my homedir.
> 
> This is my configuration too. Where did you put your "locally installed fonts" before?
> 
>> Is there a way telling TeXLive to search for the fonts again and put them into the fonts database of fontspec??
> 
> Basically, with a portable installation, you put all your local things into $HOME/texmf, from there they should be picked up automatically (at least for pdflatex and company). Lualatex and xelatex should see all the font your system see, so basically it depends on how you add a font to your system.
> 
> Romano
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_0xA909F62C45FC7802.asc
Type: application/pgp-keys
Size: 3232 bytes
Desc: OpenPGP public key
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/546a883e/attachment-0001.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 495 bytes
Desc: OpenPGP digital signature
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/546a883e/attachment-0001.sig>

From romano at rgtti.com  Thu Apr 20 11:27:42 2023
From: romano at rgtti.com (Romano Giannetti)
Date: Thu, 20 Apr 2023 11:27:42 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
 <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>
Message-ID: <e3fb1a96-2c9c-38fc-567b-62ef01c9c487@rgtti.com>

Strange. I also have fonts there... (nad FiraGo and/or Iosevka makes the 
first lualatex run ultra slow, 15 minutes of cache building).

You should be able to see them using `fc-list`; otherwise, try to issue
`fc-cache -f -v`, it should output (among three millions things) 
something like:

/home/romano/.local/share/fonts: caching, new cache contents: 760 fonts, 
0 dirs

Romano

On 20/04/2023 11:19, Harald Jele wrote:
> Dear Romano,
> 
> I did not change the directory where TeXLive 2022 already found them: 
> $HOME./local/share/fonts
> See the logfile within the ZIP. Lualatex loads but did not use them 
> (Myriad Pro).
> 
> 
> Thanks, Harald
> 
> 
> Am 20.04.23 um 09:47 schrieb Romano Giannetti:
>> On 20/04/2023 09:32, Harald Jele wrote:
>>> Dear all,
>>>
>>> I deleted my local directory of TeXLive 2022 and installed 2023 as a 
>>> portable installation in my homedir.
>>
>> This is my configuration too. Where did you put your "locally 
>> installed fonts" before?
>>
>>> Is there a way telling TeXLive to search for the fonts again and put 
>>> them into the fonts database of fontspec??
>>
>> Basically, with a portable installation, you put all your local things 
>> into $HOME/texmf, from there they should be picked up automatically 
>> (at least for pdflatex and company). Lualatex and xelatex should see 
>> all the font your system see, so basically it depends on how you add a 
>> font to your system.
>>
>> Romano
>>

-- 
Romano Giannetti, Ph.D, Electronic Engineer, Professor
http://www.rgtti.com/


From harald.jele at uni-klu.ac.at  Thu Apr 20 11:45:33 2023
From: harald.jele at uni-klu.ac.at (Harald Jele)
Date: Thu, 20 Apr 2023 11:45:33 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <e3fb1a96-2c9c-38fc-567b-62ef01c9c487@rgtti.com>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
 <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>
 <e3fb1a96-2c9c-38fc-567b-62ef01c9c487@rgtti.com>
Message-ID: <5899d0eb-6523-ecca-4e98-2a7962c29d2f@uni-klu.ac.at>

Dear Romano,

yes, the system recognizes the local fonts.

fc-list | grep local
/home/admunix/.local/share/fonts/otf/MyriadPro-It.otf: Myriad Pro:style=Italic
/home/admunix/.local/share/fonts/truetype/NotoEmoji-Bold.ttf: Noto Emoji:style=Bold
/home/admunix/.local/share/fonts/otf/MyriadPro-Regular.otf: Myriad Pro:style=Regular
/home/admunix/.local/share/fonts/truetype/NotoEmoji-SemiBold.ttf: Noto Emoji,Noto Emoji SemiBold:style=SemiBold,Regular
/home/admunix/.local/share/fonts/otf/MyriadPro-Bold.otf: Myriad Pro:style=Bold
/home/admunix/.local/share/fonts/truetype/NotoEmoji-Medium.ttf: Noto Emoji,Noto Emoji Medium:style=Medium,Regular
/home/admunix/.local/share/fonts/otf/MyriadPro-BoldIt.otf: Myriad Pro:style=Bold Italic
/home/admunix/.local/share/fonts/truetype/NotoEmoji-Light.ttf: Noto Emoji,Noto Emoji Light:style=Light,Regular
/home/admunix/.local/share/fonts/truetype/NotoEmoji-Regular.ttf: Noto Emoji:style=Regular

In the log file (out of the given ZIP) lualatex says:
Package fontspec Info: Font family 'MyriadPro(0)' created for font 'Myriad
(fontspec)             Pro' with options [Ligatures=TeX,Ligatures=TeX].
(fontspec)
(fontspec)              This font family consists of the following NFSS
(fontspec)             series/shapes:
(fontspec)
(fontspec)             - 'normal' (m/n) with NFSS spec.:
(fontspec)             <->"MyriadPro:mode=node;script=latn;language=dflt;+tlig;
"
(fontspec)             - 'small caps'  (m/sc) with NFSS spec.:
(fontspec)             - 'bold' (b/n) with NFSS spec.:
(fontspec)             <->"MyriadPro/B:mode=node;script=latn;language=dflt;+tli
g;"
(fontspec)             - 'bold small caps'  (b/sc) with NFSS spec.:
(fontspec)             - 'italic' (m/it) with NFSS spec.:
(fontspec)             <->"MyriadPro/I:mode=node;script=latn;language=dflt;+tli
g;"
(fontspec)             - 'italic small caps'  (m/scit) with NFSS spec.:
(fontspec)             - 'bold italic' (b/it) with NFSS spec.:
(fontspec)             <->"MyriadPro/BI:mode=node;script=latn;language=dflt;+tl
ig;"
(fontspec)             - 'bold italic small caps'  (b/scit) with NFSS spec.:

So I am wondering, why lualatex does not use it since there is
\setsansfont[Ligatures=TeX]{Myriad Pro}
in this example.


Thanks, Harald


Am 20.04.23 um 11:27 schrieb Romano Giannetti:
> Strange. I also have fonts there... (nad FiraGo and/or Iosevka makes the first lualatex run ultra slow, 15 minutes of cache building).
> 
> You should be able to see them using `fc-list`; otherwise, try to issue
> `fc-cache -f -v`, it should output (among three millions things) something like:
> 
> /home/romano/.local/share/fonts: caching, new cache contents: 760 fonts, 0 dirs
> 
> Romano
> 
> On 20/04/2023 11:19, Harald Jele wrote:
>> Dear Romano,
>>
>> I did not change the directory where TeXLive 2022 already found them: $HOME./local/share/fonts
>> See the logfile within the ZIP. Lualatex loads but did not use them (Myriad Pro).
>>
>>
>> Thanks, Harald
>>
>>
>> Am 20.04.23 um 09:47 schrieb Romano Giannetti:
>>> On 20/04/2023 09:32, Harald Jele wrote:
>>>> Dear all,
>>>>
>>>> I deleted my local directory of TeXLive 2022 and installed 2023 as a portable installation in my homedir.
>>>
>>> This is my configuration too. Where did you put your "locally installed fonts" before?
>>>
>>>> Is there a way telling TeXLive to search for the fonts again and put them into the fonts database of fontspec??
>>>
>>> Basically, with a portable installation, you put all your local things into $HOME/texmf, from there they should be picked up automatically (at least for pdflatex and company). Lualatex and xelatex should see all the font your system see, so basically it depends on how you add a font to your system.
>>>
>>> Romano
>>>
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_0xA909F62C45FC7802.asc
Type: application/pgp-keys
Size: 3232 bytes
Desc: OpenPGP public key
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/9de54067/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 491 bytes
Desc: OpenPGP digital signature
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/9de54067/attachment.sig>

From Herbert.Voss at fu-berlin.de  Thu Apr 20 11:52:52 2023
From: Herbert.Voss at fu-berlin.de (Herbert Voss)
Date: Thu, 20 Apr 2023 11:52:52 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <5899d0eb-6523-ecca-4e98-2a7962c29d2f@uni-klu.ac.at>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
 <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>
 <e3fb1a96-2c9c-38fc-567b-62ef01c9c487@rgtti.com>
 <5899d0eb-6523-ecca-4e98-2a7962c29d2f@uni-klu.ac.at>
Message-ID: <4c3c28a5-94dd-a093-0140-fb63d1fa6d34@fu-berlin.de>



Am 20.04.23 um 11:45 schrieb Harald Jele:
>
> yes, the system recognizes the local fonts.


what is the output of

luafindfont myriad

running in a terminal.

Herbert



From harald.jele at uni-klu.ac.at  Thu Apr 20 11:57:49 2023
From: harald.jele at uni-klu.ac.at (Harald Jele)
Date: Thu, 20 Apr 2023 11:57:49 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <4c3c28a5-94dd-a093-0140-fb63d1fa6d34@fu-berlin.de>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
 <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>
 <e3fb1a96-2c9c-38fc-567b-62ef01c9c487@rgtti.com>
 <5899d0eb-6523-ecca-4e98-2a7962c29d2f@uni-klu.ac.at>
 <4c3c28a5-94dd-a093-0140-fb63d1fa6d34@fu-berlin.de>
Message-ID: <61ab2c5e-c1f1-b9de-41dc-f3f2e9712474@uni-klu.ac.at>

Dear Herbert,

luafindfont myriad
   No.              Filename Symbolic name                                   Path
    1.    MyriadPro-Bold.otf myriadpro  /home/admunix/.local/share/fonts/otf/
    2.  MyriadPro-BoldIt.otf myriadpro  /home/admunix/.local/share/fonts/otf/
    3.      MyriadPro-It.otf myriadpro  /home/admunix/.local/share/fonts/otf/
    4. MyriadPro-Regular.otf myriadpro  /home/admunix/.local/share/fonts/otf/

Is there something wrong with
\setsansfont[Ligatures=TeX]{Myriad Pro}
??


Harald

Am 20.04.23 um 11:52 schrieb Herbert Voss:
> 
> 
> Am 20.04.23 um 11:45 schrieb Harald Jele:
>>
>> yes, the system recognizes the local fonts.
> 
> 
> what is the output of
> 
> luafindfont myriad
> 
> running in a terminal.
> 
> Herbert
> 
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_0xA909F62C45FC7802.asc
Type: application/pgp-keys
Size: 3232 bytes
Desc: OpenPGP public key
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/f638cc3f/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 495 bytes
Desc: OpenPGP digital signature
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/f638cc3f/attachment.sig>

From zdenek.wagner at gmail.com  Thu Apr 20 12:03:48 2023
From: zdenek.wagner at gmail.com (Zdenek Wagner)
Date: Thu, 20 Apr 2023 12:03:48 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <5899d0eb-6523-ecca-4e98-2a7962c29d2f@uni-klu.ac.at>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
 <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>
 <e3fb1a96-2c9c-38fc-567b-62ef01c9c487@rgtti.com>
 <5899d0eb-6523-ecca-4e98-2a7962c29d2f@uni-klu.ac.at>
Message-ID: <CAC1phyb8-xVg1zRfsrPnacS9e5xFKh484Bh4u6WKiihjALAT6g@mail.gmail.com>

Hi,

I think that this is the conflict between the fontenc and fontspec
packages. Myriad Pro is certainy a Unicode font, not a font with the
T1 encoding. I do not have the font thus I cannot try it but if you
remove \usepackage[T1]{fontenc}, it will most probably work. You can
see in the log that te font was found by lualatex but not used.

Zden?k Wagner
https://www.zdenek-wagner.eu/

?t 20. 4. 2023 v 11:45 odes?latel Harald Jele
<harald.jele at uni-klu.ac.at> napsal:
>
> Dear Romano,
>
> yes, the system recognizes the local fonts.
>
> fc-list | grep local
> /home/admunix/.local/share/fonts/otf/MyriadPro-It.otf: Myriad Pro:style=Italic
> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Bold.ttf: Noto Emoji:style=Bold
> /home/admunix/.local/share/fonts/otf/MyriadPro-Regular.otf: Myriad Pro:style=Regular
> /home/admunix/.local/share/fonts/truetype/NotoEmoji-SemiBold.ttf: Noto Emoji,Noto Emoji SemiBold:style=SemiBold,Regular
> /home/admunix/.local/share/fonts/otf/MyriadPro-Bold.otf: Myriad Pro:style=Bold
> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Medium.ttf: Noto Emoji,Noto Emoji Medium:style=Medium,Regular
> /home/admunix/.local/share/fonts/otf/MyriadPro-BoldIt.otf: Myriad Pro:style=Bold Italic
> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Light.ttf: Noto Emoji,Noto Emoji Light:style=Light,Regular
> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Regular.ttf: Noto Emoji:style=Regular
>
> In the log file (out of the given ZIP) lualatex says:
> Package fontspec Info: Font family 'MyriadPro(0)' created for font 'Myriad
> (fontspec)             Pro' with options [Ligatures=TeX,Ligatures=TeX].
> (fontspec)
> (fontspec)              This font family consists of the following NFSS
> (fontspec)             series/shapes:
> (fontspec)
> (fontspec)             - 'normal' (m/n) with NFSS spec.:
> (fontspec)             <->"MyriadPro:mode=node;script=latn;language=dflt;+tlig;
> "
> (fontspec)             - 'small caps'  (m/sc) with NFSS spec.:
> (fontspec)             - 'bold' (b/n) with NFSS spec.:
> (fontspec)             <->"MyriadPro/B:mode=node;script=latn;language=dflt;+tli
> g;"
> (fontspec)             - 'bold small caps'  (b/sc) with NFSS spec.:
> (fontspec)             - 'italic' (m/it) with NFSS spec.:
> (fontspec)             <->"MyriadPro/I:mode=node;script=latn;language=dflt;+tli
> g;"
> (fontspec)             - 'italic small caps'  (m/scit) with NFSS spec.:
> (fontspec)             - 'bold italic' (b/it) with NFSS spec.:
> (fontspec)             <->"MyriadPro/BI:mode=node;script=latn;language=dflt;+tl
> ig;"
> (fontspec)             - 'bold italic small caps'  (b/scit) with NFSS spec.:
>
> So I am wondering, why lualatex does not use it since there is
> \setsansfont[Ligatures=TeX]{Myriad Pro}
> in this example.
>
>
> Thanks, Harald
>
>
> Am 20.04.23 um 11:27 schrieb Romano Giannetti:
> > Strange. I also have fonts there... (nad FiraGo and/or Iosevka makes the first lualatex run ultra slow, 15 minutes of cache building).
> >
> > You should be able to see them using `fc-list`; otherwise, try to issue
> > `fc-cache -f -v`, it should output (among three millions things) something like:
> >
> > /home/romano/.local/share/fonts: caching, new cache contents: 760 fonts, 0 dirs
> >
> > Romano
> >
> > On 20/04/2023 11:19, Harald Jele wrote:
> >> Dear Romano,
> >>
> >> I did not change the directory where TeXLive 2022 already found them: $HOME./local/share/fonts
> >> See the logfile within the ZIP. Lualatex loads but did not use them (Myriad Pro).
> >>
> >>
> >> Thanks, Harald
> >>
> >>
> >> Am 20.04.23 um 09:47 schrieb Romano Giannetti:
> >>> On 20/04/2023 09:32, Harald Jele wrote:
> >>>> Dear all,
> >>>>
> >>>> I deleted my local directory of TeXLive 2022 and installed 2023 as a portable installation in my homedir.
> >>>
> >>> This is my configuration too. Where did you put your "locally installed fonts" before?
> >>>
> >>>> Is there a way telling TeXLive to search for the fonts again and put them into the fonts database of fontspec??
> >>>
> >>> Basically, with a portable installation, you put all your local things into $HOME/texmf, from there they should be picked up automatically (at least for pdflatex and company). Lualatex and xelatex should see all the font your system see, so basically it depends on how you add a font to your system.
> >>>
> >>> Romano
> >>>
> >


From harald.jele at uni-klu.ac.at  Thu Apr 20 12:10:22 2023
From: harald.jele at uni-klu.ac.at (Harald Jele)
Date: Thu, 20 Apr 2023 12:10:22 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <CAC1phyb8-xVg1zRfsrPnacS9e5xFKh484Bh4u6WKiihjALAT6g@mail.gmail.com>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
 <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>
 <e3fb1a96-2c9c-38fc-567b-62ef01c9c487@rgtti.com>
 <5899d0eb-6523-ecca-4e98-2a7962c29d2f@uni-klu.ac.at>
 <CAC1phyb8-xVg1zRfsrPnacS9e5xFKh484Bh4u6WKiihjALAT6g@mail.gmail.com>
Message-ID: <f3c574ff-fa8f-1565-024b-375de5fbc89d@uni-klu.ac.at>

Dear Zden?k,

sorry, makes no difference.
In TeXLive 2022 I did not recognize this Problem, so I am a bit surprised.


Harald

Am 20.04.23 um 12:03 schrieb Zdenek Wagner:
> Hi,
> 
> I think that this is the conflict between the fontenc and fontspec
> packages. Myriad Pro is certainy a Unicode font, not a font with the
> T1 encoding. I do not have the font thus I cannot try it but if you
> remove \usepackage[T1]{fontenc}, it will most probably work. You can
> see in the log that te font was found by lualatex but not used.
> 
> Zden?k Wagner
> https://www.zdenek-wagner.eu/
> 
> ?t 20. 4. 2023 v 11:45 odes?latel Harald Jele
> <harald.jele at uni-klu.ac.at> napsal:
>>
>> Dear Romano,
>>
>> yes, the system recognizes the local fonts.
>>
>> fc-list | grep local
>> /home/admunix/.local/share/fonts/otf/MyriadPro-It.otf: Myriad Pro:style=Italic
>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Bold.ttf: Noto Emoji:style=Bold
>> /home/admunix/.local/share/fonts/otf/MyriadPro-Regular.otf: Myriad Pro:style=Regular
>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-SemiBold.ttf: Noto Emoji,Noto Emoji SemiBold:style=SemiBold,Regular
>> /home/admunix/.local/share/fonts/otf/MyriadPro-Bold.otf: Myriad Pro:style=Bold
>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Medium.ttf: Noto Emoji,Noto Emoji Medium:style=Medium,Regular
>> /home/admunix/.local/share/fonts/otf/MyriadPro-BoldIt.otf: Myriad Pro:style=Bold Italic
>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Light.ttf: Noto Emoji,Noto Emoji Light:style=Light,Regular
>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Regular.ttf: Noto Emoji:style=Regular
>>
>> In the log file (out of the given ZIP) lualatex says:
>> Package fontspec Info: Font family 'MyriadPro(0)' created for font 'Myriad
>> (fontspec)             Pro' with options [Ligatures=TeX,Ligatures=TeX].
>> (fontspec)
>> (fontspec)              This font family consists of the following NFSS
>> (fontspec)             series/shapes:
>> (fontspec)
>> (fontspec)             - 'normal' (m/n) with NFSS spec.:
>> (fontspec)             <->"MyriadPro:mode=node;script=latn;language=dflt;+tlig;
>> "
>> (fontspec)             - 'small caps'  (m/sc) with NFSS spec.:
>> (fontspec)             - 'bold' (b/n) with NFSS spec.:
>> (fontspec)             <->"MyriadPro/B:mode=node;script=latn;language=dflt;+tli
>> g;"
>> (fontspec)             - 'bold small caps'  (b/sc) with NFSS spec.:
>> (fontspec)             - 'italic' (m/it) with NFSS spec.:
>> (fontspec)             <->"MyriadPro/I:mode=node;script=latn;language=dflt;+tli
>> g;"
>> (fontspec)             - 'italic small caps'  (m/scit) with NFSS spec.:
>> (fontspec)             - 'bold italic' (b/it) with NFSS spec.:
>> (fontspec)             <->"MyriadPro/BI:mode=node;script=latn;language=dflt;+tl
>> ig;"
>> (fontspec)             - 'bold italic small caps'  (b/scit) with NFSS spec.:
>>
>> So I am wondering, why lualatex does not use it since there is
>> \setsansfont[Ligatures=TeX]{Myriad Pro}
>> in this example.
>>
>>
>> Thanks, Harald
>>
>>
>> Am 20.04.23 um 11:27 schrieb Romano Giannetti:
>>> Strange. I also have fonts there... (nad FiraGo and/or Iosevka makes the first lualatex run ultra slow, 15 minutes of cache building).
>>>
>>> You should be able to see them using `fc-list`; otherwise, try to issue
>>> `fc-cache -f -v`, it should output (among three millions things) something like:
>>>
>>> /home/romano/.local/share/fonts: caching, new cache contents: 760 fonts, 0 dirs
>>>
>>> Romano
>>>
>>> On 20/04/2023 11:19, Harald Jele wrote:
>>>> Dear Romano,
>>>>
>>>> I did not change the directory where TeXLive 2022 already found them: $HOME./local/share/fonts
>>>> See the logfile within the ZIP. Lualatex loads but did not use them (Myriad Pro).
>>>>
>>>>
>>>> Thanks, Harald
>>>>
>>>>
>>>> Am 20.04.23 um 09:47 schrieb Romano Giannetti:
>>>>> On 20/04/2023 09:32, Harald Jele wrote:
>>>>>> Dear all,
>>>>>>
>>>>>> I deleted my local directory of TeXLive 2022 and installed 2023 as a portable installation in my homedir.
>>>>>
>>>>> This is my configuration too. Where did you put your "locally installed fonts" before?
>>>>>
>>>>>> Is there a way telling TeXLive to search for the fonts again and put them into the fonts database of fontspec??
>>>>>
>>>>> Basically, with a portable installation, you put all your local things into $HOME/texmf, from there they should be picked up automatically (at least for pdflatex and company). Lualatex and xelatex should see all the font your system see, so basically it depends on how you add a font to your system.
>>>>>
>>>>> Romano
>>>>>
>>>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_0xA909F62C45FC7802.asc
Type: application/pgp-keys
Size: 3232 bytes
Desc: OpenPGP public key
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/68e50a25/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 495 bytes
Desc: OpenPGP digital signature
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/68e50a25/attachment.sig>

From zdenek.wagner at gmail.com  Thu Apr 20 12:12:29 2023
From: zdenek.wagner at gmail.com (Zdenek Wagner)
Date: Thu, 20 Apr 2023 12:12:29 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <f3c574ff-fa8f-1565-024b-375de5fbc89d@uni-klu.ac.at>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
 <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>
 <e3fb1a96-2c9c-38fc-567b-62ef01c9c487@rgtti.com>
 <5899d0eb-6523-ecca-4e98-2a7962c29d2f@uni-klu.ac.at>
 <CAC1phyb8-xVg1zRfsrPnacS9e5xFKh484Bh4u6WKiihjALAT6g@mail.gmail.com>
 <f3c574ff-fa8f-1565-024b-375de5fbc89d@uni-klu.ac.at>
Message-ID: <CAC1phya-NFUkLj6MH81eVAnacFUFNkPBzXu2P3_AqROXiDPE3w@mail.gmail.com>

Can you, please, send the new .tex and .log so that I can see the
difference? It's now lunch time here, I will be back at the computer
after 1 hour.

Zden?k Wagner
https://www.zdenek-wagner.eu/

?t 20. 4. 2023 v 12:10 odes?latel Harald Jele
<harald.jele at uni-klu.ac.at> napsal:
>
> Dear Zden?k,
>
> sorry, makes no difference.
> In TeXLive 2022 I did not recognize this Problem, so I am a bit surprised.
>
>
> Harald
>
> Am 20.04.23 um 12:03 schrieb Zdenek Wagner:
> > Hi,
> >
> > I think that this is the conflict between the fontenc and fontspec
> > packages. Myriad Pro is certainy a Unicode font, not a font with the
> > T1 encoding. I do not have the font thus I cannot try it but if you
> > remove \usepackage[T1]{fontenc}, it will most probably work. You can
> > see in the log that te font was found by lualatex but not used.
> >
> > Zden?k Wagner
> > https://www.zdenek-wagner.eu/
> >
> > ?t 20. 4. 2023 v 11:45 odes?latel Harald Jele
> > <harald.jele at uni-klu.ac.at> napsal:
> >>
> >> Dear Romano,
> >>
> >> yes, the system recognizes the local fonts.
> >>
> >> fc-list | grep local
> >> /home/admunix/.local/share/fonts/otf/MyriadPro-It.otf: Myriad Pro:style=Italic
> >> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Bold.ttf: Noto Emoji:style=Bold
> >> /home/admunix/.local/share/fonts/otf/MyriadPro-Regular.otf: Myriad Pro:style=Regular
> >> /home/admunix/.local/share/fonts/truetype/NotoEmoji-SemiBold.ttf: Noto Emoji,Noto Emoji SemiBold:style=SemiBold,Regular
> >> /home/admunix/.local/share/fonts/otf/MyriadPro-Bold.otf: Myriad Pro:style=Bold
> >> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Medium.ttf: Noto Emoji,Noto Emoji Medium:style=Medium,Regular
> >> /home/admunix/.local/share/fonts/otf/MyriadPro-BoldIt.otf: Myriad Pro:style=Bold Italic
> >> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Light.ttf: Noto Emoji,Noto Emoji Light:style=Light,Regular
> >> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Regular.ttf: Noto Emoji:style=Regular
> >>
> >> In the log file (out of the given ZIP) lualatex says:
> >> Package fontspec Info: Font family 'MyriadPro(0)' created for font 'Myriad
> >> (fontspec)             Pro' with options [Ligatures=TeX,Ligatures=TeX].
> >> (fontspec)
> >> (fontspec)              This font family consists of the following NFSS
> >> (fontspec)             series/shapes:
> >> (fontspec)
> >> (fontspec)             - 'normal' (m/n) with NFSS spec.:
> >> (fontspec)             <->"MyriadPro:mode=node;script=latn;language=dflt;+tlig;
> >> "
> >> (fontspec)             - 'small caps'  (m/sc) with NFSS spec.:
> >> (fontspec)             - 'bold' (b/n) with NFSS spec.:
> >> (fontspec)             <->"MyriadPro/B:mode=node;script=latn;language=dflt;+tli
> >> g;"
> >> (fontspec)             - 'bold small caps'  (b/sc) with NFSS spec.:
> >> (fontspec)             - 'italic' (m/it) with NFSS spec.:
> >> (fontspec)             <->"MyriadPro/I:mode=node;script=latn;language=dflt;+tli
> >> g;"
> >> (fontspec)             - 'italic small caps'  (m/scit) with NFSS spec.:
> >> (fontspec)             - 'bold italic' (b/it) with NFSS spec.:
> >> (fontspec)             <->"MyriadPro/BI:mode=node;script=latn;language=dflt;+tl
> >> ig;"
> >> (fontspec)             - 'bold italic small caps'  (b/scit) with NFSS spec.:
> >>
> >> So I am wondering, why lualatex does not use it since there is
> >> \setsansfont[Ligatures=TeX]{Myriad Pro}
> >> in this example.
> >>
> >>
> >> Thanks, Harald
> >>
> >>
> >> Am 20.04.23 um 11:27 schrieb Romano Giannetti:
> >>> Strange. I also have fonts there... (nad FiraGo and/or Iosevka makes the first lualatex run ultra slow, 15 minutes of cache building).
> >>>
> >>> You should be able to see them using `fc-list`; otherwise, try to issue
> >>> `fc-cache -f -v`, it should output (among three millions things) something like:
> >>>
> >>> /home/romano/.local/share/fonts: caching, new cache contents: 760 fonts, 0 dirs
> >>>
> >>> Romano
> >>>
> >>> On 20/04/2023 11:19, Harald Jele wrote:
> >>>> Dear Romano,
> >>>>
> >>>> I did not change the directory where TeXLive 2022 already found them: $HOME./local/share/fonts
> >>>> See the logfile within the ZIP. Lualatex loads but did not use them (Myriad Pro).
> >>>>
> >>>>
> >>>> Thanks, Harald
> >>>>
> >>>>
> >>>> Am 20.04.23 um 09:47 schrieb Romano Giannetti:
> >>>>> On 20/04/2023 09:32, Harald Jele wrote:
> >>>>>> Dear all,
> >>>>>>
> >>>>>> I deleted my local directory of TeXLive 2022 and installed 2023 as a portable installation in my homedir.
> >>>>>
> >>>>> This is my configuration too. Where did you put your "locally installed fonts" before?
> >>>>>
> >>>>>> Is there a way telling TeXLive to search for the fonts again and put them into the fonts database of fontspec??
> >>>>>
> >>>>> Basically, with a portable installation, you put all your local things into $HOME/texmf, from there they should be picked up automatically (at least for pdflatex and company). Lualatex and xelatex should see all the font your system see, so basically it depends on how you add a font to your system.
> >>>>>
> >>>>> Romano
> >>>>>
> >>>


From harald.jele at uni-klu.ac.at  Thu Apr 20 12:21:38 2023
From: harald.jele at uni-klu.ac.at (Harald Jele)
Date: Thu, 20 Apr 2023 12:21:38 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <CAC1phya-NFUkLj6MH81eVAnacFUFNkPBzXu2P3_AqROXiDPE3w@mail.gmail.com>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
 <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>
 <e3fb1a96-2c9c-38fc-567b-62ef01c9c487@rgtti.com>
 <5899d0eb-6523-ecca-4e98-2a7962c29d2f@uni-klu.ac.at>
 <CAC1phyb8-xVg1zRfsrPnacS9e5xFKh484Bh4u6WKiihjALAT6g@mail.gmail.com>
 <f3c574ff-fa8f-1565-024b-375de5fbc89d@uni-klu.ac.at>
 <CAC1phya-NFUkLj6MH81eVAnacFUFNkPBzXu2P3_AqROXiDPE3w@mail.gmail.com>
Message-ID: <eaefee5f-bf46-8407-b533-d0b84718678a@uni-klu.ac.at>

yes, of course:

https://wwwu.aau.at/hjele/texlive_vs_debian_testing/chapterheading_indent.log


Harald

Am 20.04.23 um 12:12 schrieb Zdenek Wagner:
> Can you, please, send the new .tex and .log so that I can see the
> difference? It's now lunch time here, I will be back at the computer
> after 1 hour.
> 
> Zden?k Wagner
> https://www.zdenek-wagner.eu/
> 
> ?t 20. 4. 2023 v 12:10 odes?latel Harald Jele
> <harald.jele at uni-klu.ac.at> napsal:
>>
>> Dear Zden?k,
>>
>> sorry, makes no difference.
>> In TeXLive 2022 I did not recognize this Problem, so I am a bit surprised.
>>
>>
>> Harald
>>
>> Am 20.04.23 um 12:03 schrieb Zdenek Wagner:
>>> Hi,
>>>
>>> I think that this is the conflict between the fontenc and fontspec
>>> packages. Myriad Pro is certainy a Unicode font, not a font with the
>>> T1 encoding. I do not have the font thus I cannot try it but if you
>>> remove \usepackage[T1]{fontenc}, it will most probably work. You can
>>> see in the log that te font was found by lualatex but not used.
>>>
>>> Zden?k Wagner
>>> https://www.zdenek-wagner.eu/
>>>
>>> ?t 20. 4. 2023 v 11:45 odes?latel Harald Jele
>>> <harald.jele at uni-klu.ac.at> napsal:
>>>>
>>>> Dear Romano,
>>>>
>>>> yes, the system recognizes the local fonts.
>>>>
>>>> fc-list | grep local
>>>> /home/admunix/.local/share/fonts/otf/MyriadPro-It.otf: Myriad Pro:style=Italic
>>>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Bold.ttf: Noto Emoji:style=Bold
>>>> /home/admunix/.local/share/fonts/otf/MyriadPro-Regular.otf: Myriad Pro:style=Regular
>>>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-SemiBold.ttf: Noto Emoji,Noto Emoji SemiBold:style=SemiBold,Regular
>>>> /home/admunix/.local/share/fonts/otf/MyriadPro-Bold.otf: Myriad Pro:style=Bold
>>>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Medium.ttf: Noto Emoji,Noto Emoji Medium:style=Medium,Regular
>>>> /home/admunix/.local/share/fonts/otf/MyriadPro-BoldIt.otf: Myriad Pro:style=Bold Italic
>>>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Light.ttf: Noto Emoji,Noto Emoji Light:style=Light,Regular
>>>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Regular.ttf: Noto Emoji:style=Regular
>>>>
>>>> In the log file (out of the given ZIP) lualatex says:
>>>> Package fontspec Info: Font family 'MyriadPro(0)' created for font 'Myriad
>>>> (fontspec)             Pro' with options [Ligatures=TeX,Ligatures=TeX].
>>>> (fontspec)
>>>> (fontspec)              This font family consists of the following NFSS
>>>> (fontspec)             series/shapes:
>>>> (fontspec)
>>>> (fontspec)             - 'normal' (m/n) with NFSS spec.:
>>>> (fontspec)             <->"MyriadPro:mode=node;script=latn;language=dflt;+tlig;
>>>> "
>>>> (fontspec)             - 'small caps'  (m/sc) with NFSS spec.:
>>>> (fontspec)             - 'bold' (b/n) with NFSS spec.:
>>>> (fontspec)             <->"MyriadPro/B:mode=node;script=latn;language=dflt;+tli
>>>> g;"
>>>> (fontspec)             - 'bold small caps'  (b/sc) with NFSS spec.:
>>>> (fontspec)             - 'italic' (m/it) with NFSS spec.:
>>>> (fontspec)             <->"MyriadPro/I:mode=node;script=latn;language=dflt;+tli
>>>> g;"
>>>> (fontspec)             - 'italic small caps'  (m/scit) with NFSS spec.:
>>>> (fontspec)             - 'bold italic' (b/it) with NFSS spec.:
>>>> (fontspec)             <->"MyriadPro/BI:mode=node;script=latn;language=dflt;+tl
>>>> ig;"
>>>> (fontspec)             - 'bold italic small caps'  (b/scit) with NFSS spec.:
>>>>
>>>> So I am wondering, why lualatex does not use it since there is
>>>> \setsansfont[Ligatures=TeX]{Myriad Pro}
>>>> in this example.
>>>>
>>>>
>>>> Thanks, Harald
>>>>
>>>>
>>>> Am 20.04.23 um 11:27 schrieb Romano Giannetti:
>>>>> Strange. I also have fonts there... (nad FiraGo and/or Iosevka makes the first lualatex run ultra slow, 15 minutes of cache building).
>>>>>
>>>>> You should be able to see them using `fc-list`; otherwise, try to issue
>>>>> `fc-cache -f -v`, it should output (among three millions things) something like:
>>>>>
>>>>> /home/romano/.local/share/fonts: caching, new cache contents: 760 fonts, 0 dirs
>>>>>
>>>>> Romano
>>>>>
>>>>> On 20/04/2023 11:19, Harald Jele wrote:
>>>>>> Dear Romano,
>>>>>>
>>>>>> I did not change the directory where TeXLive 2022 already found them: $HOME./local/share/fonts
>>>>>> See the logfile within the ZIP. Lualatex loads but did not use them (Myriad Pro).
>>>>>>
>>>>>>
>>>>>> Thanks, Harald
>>>>>>
>>>>>>
>>>>>> Am 20.04.23 um 09:47 schrieb Romano Giannetti:
>>>>>>> On 20/04/2023 09:32, Harald Jele wrote:
>>>>>>>> Dear all,
>>>>>>>>
>>>>>>>> I deleted my local directory of TeXLive 2022 and installed 2023 as a portable installation in my homedir.
>>>>>>>
>>>>>>> This is my configuration too. Where did you put your "locally installed fonts" before?
>>>>>>>
>>>>>>>> Is there a way telling TeXLive to search for the fonts again and put them into the fonts database of fontspec??
>>>>>>>
>>>>>>> Basically, with a portable installation, you put all your local things into $HOME/texmf, from there they should be picked up automatically (at least for pdflatex and company). Lualatex and xelatex should see all the font your system see, so basically it depends on how you add a font to your system.
>>>>>>>
>>>>>>> Romano
>>>>>>>
>>>>>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_0xA909F62C45FC7802.asc
Type: application/pgp-keys
Size: 3232 bytes
Desc: OpenPGP public key
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/af69a78c/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 495 bytes
Desc: OpenPGP digital signature
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/af69a78c/attachment.sig>

From Herbert.Voss at fu-berlin.de  Thu Apr 20 12:57:04 2023
From: Herbert.Voss at fu-berlin.de (Herbert Voss)
Date: Thu, 20 Apr 2023 12:57:04 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <eaefee5f-bf46-8407-b533-d0b84718678a@uni-klu.ac.at>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
 <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>
 <e3fb1a96-2c9c-38fc-567b-62ef01c9c487@rgtti.com>
 <5899d0eb-6523-ecca-4e98-2a7962c29d2f@uni-klu.ac.at>
 <CAC1phyb8-xVg1zRfsrPnacS9e5xFKh484Bh4u6WKiihjALAT6g@mail.gmail.com>
 <f3c574ff-fa8f-1565-024b-375de5fbc89d@uni-klu.ac.at>
 <CAC1phya-NFUkLj6MH81eVAnacFUFNkPBzXu2P3_AqROXiDPE3w@mail.gmail.com>
 <eaefee5f-bf46-8407-b533-d0b84718678a@uni-klu.ac.at>
Message-ID: <186fb5ff-f9b1-0e74-77cb-d221a2ef8d77@fu-berlin.de>



Am 20.04.23 um 12:21 schrieb Harald Jele:
> yes, of course:
>
> https://wwwu.aau.at/hjele/texlive_vs_debian_testing/chapterheading_indent.log 
>

it looks like that you do not used any sans serif text:

</home/admunix/texlive/texmf-dist/fonts/truetype/public/gentium-tug/GentiumPlus-
Regular.ttf></home/admunix/texlive/texmf-dist/fonts/truetype/public/gentium-tug/
GentiumPlus-Bold.ttf>
Output written on chapterheading_indent.pdf (1 page, 65563 bytes).


What happens with

\documentclass{article}
\usepackage{fontspec}
\setsansfont{Myriad Pro}

\begin{document}

foo and \textsf{Myriad Pro}

\end{document}

and running lualatex

Herbert

From zdenek.wagner at gmail.com  Thu Apr 20 13:21:57 2023
From: zdenek.wagner at gmail.com (Zdenek Wagner)
Date: Thu, 20 Apr 2023 13:21:57 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <eaefee5f-bf46-8407-b533-d0b84718678a@uni-klu.ac.at>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
 <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>
 <e3fb1a96-2c9c-38fc-567b-62ef01c9c487@rgtti.com>
 <5899d0eb-6523-ecca-4e98-2a7962c29d2f@uni-klu.ac.at>
 <CAC1phyb8-xVg1zRfsrPnacS9e5xFKh484Bh4u6WKiihjALAT6g@mail.gmail.com>
 <f3c574ff-fa8f-1565-024b-375de5fbc89d@uni-klu.ac.at>
 <CAC1phya-NFUkLj6MH81eVAnacFUFNkPBzXu2P3_AqROXiDPE3w@mail.gmail.com>
 <eaefee5f-bf46-8407-b533-d0b84718678a@uni-klu.ac.at>
Message-ID: <CAC1phyaBgOAsKoV5OVnb8-LAMZtTmMKoj_WE5BvkGdn9QykZ3Q@mail.gmail.com>

Hi,

the log shows that Myriad Pro is found and will be used eve in math
but I looked again into your file and the problem is on this line:


\large\textsf\textbf{EvenNotMyriadPro}

\textsf is a macro which requires one parameter, the parameter is
either a token or balanced text enclosed by {}. It is followed by a
token \textbf, thus it executes \textsf{\textbf}. I am not sure where
\textbf gets its parameter but you can see, that it was consumed by
\textsf and the text is not even bold. You have to add outer braces:


\large\textsf{\textbf{EvenNotMyriadPro}}

Zden?k Wagner
https://www.zdenek-wagner.eu/

?t 20. 4. 2023 v 12:21 odes?latel Harald Jele
<harald.jele at uni-klu.ac.at> napsal:
>
> yes, of course:
>
> https://wwwu.aau.at/hjele/texlive_vs_debian_testing/chapterheading_indent.log
>
>
> Harald
>
> Am 20.04.23 um 12:12 schrieb Zdenek Wagner:
> > Can you, please, send the new .tex and .log so that I can see the
> > difference? It's now lunch time here, I will be back at the computer
> > after 1 hour.
> >
> > Zden?k Wagner
> > https://www.zdenek-wagner.eu/
> >
> > ?t 20. 4. 2023 v 12:10 odes?latel Harald Jele
> > <harald.jele at uni-klu.ac.at> napsal:
> >>
> >> Dear Zden?k,
> >>
> >> sorry, makes no difference.
> >> In TeXLive 2022 I did not recognize this Problem, so I am a bit surprised.
> >>
> >>
> >> Harald
> >>
> >> Am 20.04.23 um 12:03 schrieb Zdenek Wagner:
> >>> Hi,
> >>>
> >>> I think that this is the conflict between the fontenc and fontspec
> >>> packages. Myriad Pro is certainy a Unicode font, not a font with the
> >>> T1 encoding. I do not have the font thus I cannot try it but if you
> >>> remove \usepackage[T1]{fontenc}, it will most probably work. You can
> >>> see in the log that te font was found by lualatex but not used.
> >>>
> >>> Zden?k Wagner
> >>> https://www.zdenek-wagner.eu/
> >>>
> >>> ?t 20. 4. 2023 v 11:45 odes?latel Harald Jele
> >>> <harald.jele at uni-klu.ac.at> napsal:
> >>>>
> >>>> Dear Romano,
> >>>>
> >>>> yes, the system recognizes the local fonts.
> >>>>
> >>>> fc-list | grep local
> >>>> /home/admunix/.local/share/fonts/otf/MyriadPro-It.otf: Myriad Pro:style=Italic
> >>>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Bold.ttf: Noto Emoji:style=Bold
> >>>> /home/admunix/.local/share/fonts/otf/MyriadPro-Regular.otf: Myriad Pro:style=Regular
> >>>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-SemiBold.ttf: Noto Emoji,Noto Emoji SemiBold:style=SemiBold,Regular
> >>>> /home/admunix/.local/share/fonts/otf/MyriadPro-Bold.otf: Myriad Pro:style=Bold
> >>>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Medium.ttf: Noto Emoji,Noto Emoji Medium:style=Medium,Regular
> >>>> /home/admunix/.local/share/fonts/otf/MyriadPro-BoldIt.otf: Myriad Pro:style=Bold Italic
> >>>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Light.ttf: Noto Emoji,Noto Emoji Light:style=Light,Regular
> >>>> /home/admunix/.local/share/fonts/truetype/NotoEmoji-Regular.ttf: Noto Emoji:style=Regular
> >>>>
> >>>> In the log file (out of the given ZIP) lualatex says:
> >>>> Package fontspec Info: Font family 'MyriadPro(0)' created for font 'Myriad
> >>>> (fontspec)             Pro' with options [Ligatures=TeX,Ligatures=TeX].
> >>>> (fontspec)
> >>>> (fontspec)              This font family consists of the following NFSS
> >>>> (fontspec)             series/shapes:
> >>>> (fontspec)
> >>>> (fontspec)             - 'normal' (m/n) with NFSS spec.:
> >>>> (fontspec)             <->"MyriadPro:mode=node;script=latn;language=dflt;+tlig;
> >>>> "
> >>>> (fontspec)             - 'small caps'  (m/sc) with NFSS spec.:
> >>>> (fontspec)             - 'bold' (b/n) with NFSS spec.:
> >>>> (fontspec)             <->"MyriadPro/B:mode=node;script=latn;language=dflt;+tli
> >>>> g;"
> >>>> (fontspec)             - 'bold small caps'  (b/sc) with NFSS spec.:
> >>>> (fontspec)             - 'italic' (m/it) with NFSS spec.:
> >>>> (fontspec)             <->"MyriadPro/I:mode=node;script=latn;language=dflt;+tli
> >>>> g;"
> >>>> (fontspec)             - 'italic small caps'  (m/scit) with NFSS spec.:
> >>>> (fontspec)             - 'bold italic' (b/it) with NFSS spec.:
> >>>> (fontspec)             <->"MyriadPro/BI:mode=node;script=latn;language=dflt;+tl
> >>>> ig;"
> >>>> (fontspec)             - 'bold italic small caps'  (b/scit) with NFSS spec.:
> >>>>
> >>>> So I am wondering, why lualatex does not use it since there is
> >>>> \setsansfont[Ligatures=TeX]{Myriad Pro}
> >>>> in this example.
> >>>>
> >>>>
> >>>> Thanks, Harald
> >>>>
> >>>>
> >>>> Am 20.04.23 um 11:27 schrieb Romano Giannetti:
> >>>>> Strange. I also have fonts there... (nad FiraGo and/or Iosevka makes the first lualatex run ultra slow, 15 minutes of cache building).
> >>>>>
> >>>>> You should be able to see them using `fc-list`; otherwise, try to issue
> >>>>> `fc-cache -f -v`, it should output (among three millions things) something like:
> >>>>>
> >>>>> /home/romano/.local/share/fonts: caching, new cache contents: 760 fonts, 0 dirs
> >>>>>
> >>>>> Romano
> >>>>>
> >>>>> On 20/04/2023 11:19, Harald Jele wrote:
> >>>>>> Dear Romano,
> >>>>>>
> >>>>>> I did not change the directory where TeXLive 2022 already found them: $HOME./local/share/fonts
> >>>>>> See the logfile within the ZIP. Lualatex loads but did not use them (Myriad Pro).
> >>>>>>
> >>>>>>
> >>>>>> Thanks, Harald
> >>>>>>
> >>>>>>
> >>>>>> Am 20.04.23 um 09:47 schrieb Romano Giannetti:
> >>>>>>> On 20/04/2023 09:32, Harald Jele wrote:
> >>>>>>>> Dear all,
> >>>>>>>>
> >>>>>>>> I deleted my local directory of TeXLive 2022 and installed 2023 as a portable installation in my homedir.
> >>>>>>>
> >>>>>>> This is my configuration too. Where did you put your "locally installed fonts" before?
> >>>>>>>
> >>>>>>>> Is there a way telling TeXLive to search for the fonts again and put them into the fonts database of fontspec??
> >>>>>>>
> >>>>>>> Basically, with a portable installation, you put all your local things into $HOME/texmf, from there they should be picked up automatically (at least for pdflatex and company). Lualatex and xelatex should see all the font your system see, so basically it depends on how you add a font to your system.
> >>>>>>>
> >>>>>>> Romano
> >>>>>>>
> >>>>>


From harald.jele at uni-klu.ac.at  Thu Apr 20 14:02:05 2023
From: harald.jele at uni-klu.ac.at (Harald Jele)
Date: Thu, 20 Apr 2023 14:02:05 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <186fb5ff-f9b1-0e74-77cb-d221a2ef8d77@fu-berlin.de>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <91198cb1-2e89-91e3-e494-ffebb6c3bb2e@rgtti.com>
 <aa4acab9-99a0-81fd-bb02-57e166a94119@uni-klu.ac.at>
 <e3fb1a96-2c9c-38fc-567b-62ef01c9c487@rgtti.com>
 <5899d0eb-6523-ecca-4e98-2a7962c29d2f@uni-klu.ac.at>
 <CAC1phyb8-xVg1zRfsrPnacS9e5xFKh484Bh4u6WKiihjALAT6g@mail.gmail.com>
 <f3c574ff-fa8f-1565-024b-375de5fbc89d@uni-klu.ac.at>
 <CAC1phya-NFUkLj6MH81eVAnacFUFNkPBzXu2P3_AqROXiDPE3w@mail.gmail.com>
 <eaefee5f-bf46-8407-b533-d0b84718678a@uni-klu.ac.at>
 <186fb5ff-f9b1-0e74-77cb-d221a2ef8d77@fu-berlin.de>
Message-ID: <ba84a6a1-beba-f0d0-af27-b5634f30a8d9@uni-klu.ac.at>

Thanks Herbert,

so there must have been a change in Koma Script and headings (of chapters, sections and subsections) are not printed in a Sans Serif Font per default anymore.
Sorry for the circumstances


Harald

Am 20.04.23 um 12:57 schrieb Herbert Voss:
> 
> 
> Am 20.04.23 um 12:21 schrieb Harald Jele:
>> yes, of course:
>>
>> https://wwwu.aau.at/hjele/texlive_vs_debian_testing/chapterheading_indent.log
> 
> it looks like that you do not used any sans serif text:
> 
> </home/admunix/texlive/texmf-dist/fonts/truetype/public/gentium-tug/GentiumPlus-
> Regular.ttf></home/admunix/texlive/texmf-dist/fonts/truetype/public/gentium-tug/
> GentiumPlus-Bold.ttf>
> Output written on chapterheading_indent.pdf (1 page, 65563 bytes).
> 
> 
> What happens with
> 
> \documentclass{article}
> \usepackage{fontspec}
> \setsansfont{Myriad Pro}
> 
> \begin{document}
> 
> foo and \textsf{Myriad Pro}
> 
> \end{document}
> 
> and running lualatex
> 
> Herbert
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_0xA909F62C45FC7802.asc
Type: application/pgp-keys
Size: 3232 bytes
Desc: OpenPGP public key
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/af478424/attachment-0001.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 495 bytes
Desc: OpenPGP digital signature
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/af478424/attachment-0001.sig>

From wl at gnu.org  Thu Apr 20 14:07:30 2023
From: wl at gnu.org (Werner LEMBERG)
Date: Thu, 20 Apr 2023 12:07:30 +0000 (UTC)
Subject: TeXLive 2023 and local fonts
In-Reply-To: <ba84a6a1-beba-f0d0-af27-b5634f30a8d9@uni-klu.ac.at>
References: <eaefee5f-bf46-8407-b533-d0b84718678a@uni-klu.ac.at>
 <186fb5ff-f9b1-0e74-77cb-d221a2ef8d77@fu-berlin.de>
 <ba84a6a1-beba-f0d0-af27-b5634f30a8d9@uni-klu.ac.at>
Message-ID: <20230420.140730.993528444807433017.wl@gnu.org>


> so there must have been a change in Koma Script and headings (of
> chapters, sections and subsections) are not printed in a Sans Serif
> Font per default anymore.

I've noted this, too.  Either this is a new bug, or it wasn't properly
announced.


    Werner

From komascript at gmx.info  Thu Apr 20 14:31:17 2023
From: komascript at gmx.info (Markus Kohm)
Date: Thu, 20 Apr 2023 14:31:17 +0200
Subject: TeXLive 2023 and local fonts
In-Reply-To: <ba84a6a1-beba-f0d0-af27-b5634f30a8d9@uni-klu.ac.at>
References: <365f2d69-11de-9298-1a45-eee96da653dd@uni-klu.ac.at>
 <186fb5ff-f9b1-0e74-77cb-d221a2ef8d77@fu-berlin.de>
 <ba84a6a1-beba-f0d0-af27-b5634f30a8d9@uni-klu.ac.at>
Message-ID: <2677035.mvXUDI8C0e@komascript.de>

Am Donnerstag, 20. April 2023, 14:02:05 CEST schrieb Harald Jele:
> so there must have been a change in Koma Script and headings (of chapters,
> sections and subsections) are not printed in a Sans Serif Font per default
> anymore.

? https://sourceforge.net/p/koma-script/tickets/48/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: This is a digitally signed message part.
URL: <https://tug.org/pipermail/tex-live/attachments/20230420/72ede2e5/attachment.sig>

From karl at freefriends.org  Fri Apr 21 23:43:58 2023
From: karl at freefriends.org (Karl Berry)
Date: Fri, 21 Apr 2023 15:43:58 -0600
Subject: How should I deal with packages that require FFI external
 libraries?
In-Reply-To: <8D745C1D-E3BF-415F-BD10-46C2AEB22177@icloud.com>
Message-ID: <202304212143.33LLhwJW013599@freefriends.org>

Hi Bruno - you know this, but I can't replying anyway:

    they did not understand why a TeX
    install was more than a few hundred magabytes

You can tell them that if all they want to do is typeset their math
papers, they can use BasicTeX == scheme-small ~ 550MB. Admittedly still
not tiny, but plausible nowadays even on fairly small machines.

The most minimal scheme including LaTeX is scheme-basic, some 265MB.
Perfectly capable of typesetting straightforward documents, but of
course lacks almost all add-on packages, say, revtex, which is included
in scheme-small.

The answer to why full TL is so big (currently 7.5+GB) is that it
includes all the fonts, support for all languages including CJK, all
macros, etc. Support to typeset any document in (more or less) any
script anywhere in the world takes space ...

Feel free to forward this to anyone who questions why TL contains what
it does. --hope this helps somehow, karl.

From peter at silmaril.ie  Sat Apr 22 01:09:26 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Sat, 22 Apr 2023 00:09:26 +0100
Subject: How should I deal with packages that require FFI external
 libraries?
In-Reply-To: <202304212143.33LLhwJW013599@freefriends.org>
References: <202304212143.33LLhwJW013599@freefriends.org>
Message-ID: <0a905443-718a-b9ea-528d-583fee57135c@silmaril.ie>

I'd like to include this in Formatting information, if I may.

Peter

On 21/04/2023 22:43, Karl Berry wrote:
> Hi Bruno - you know this, but I can't replying anyway:
> 
>      they did not understand why a TeX
>      install was more than a few hundred magabytes
> 
> You can tell them that if all they want to do is typeset their math
> papers, they can use BasicTeX == scheme-small ~ 550MB. Admittedly still
> not tiny, but plausible nowadays even on fairly small machines.
> 
> The most minimal scheme including LaTeX is scheme-basic, some 265MB.
> Perfectly capable of typesetting straightforward documents, but of
> course lacks almost all add-on packages, say, revtex, which is included
> in scheme-small.
> 
> The answer to why full TL is so big (currently 7.5+GB) is that it
> includes all the fonts, support for all languages including CJK, all
> macros, etc. Support to typeset any document in (more or less) any
> script anywhere in the world takes space ...
> 
> Feel free to forward this to anyone who questions why TL contains what
> it does. --hope this helps somehow, karl.

From karl at freefriends.org  Sat Apr 22 23:18:03 2023
From: karl at freefriends.org (Karl Berry)
Date: Sat, 22 Apr 2023 15:18:03 -0600
Subject: How should I deal with packages that require FFI external
 libraries?
In-Reply-To: <0a905443-718a-b9ea-528d-583fee57135c@silmaril.ie>
Message-ID: <202304222118.33MLI322025795@freefriends.org>

    I'd like to include this in Formatting information, if I may.

Hi Peter (and anyone else who might find this) - surely. I'm honored :).
Feel free to use/edit anything I post in public email however you like.
No need for attribution as far as I'm concerned.

One tidbit I perhaps should have mentioned is that you can select the
scheme on the command line invoking install-tl:
  install-tl -s small
  install-tl -s basic
etc. Can be more convenient than messing with the menus.

Thanks,
Karl

From peter at silmaril.ie  Mon Apr 24 15:52:58 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Mon, 24 Apr 2023 14:52:58 +0100
Subject: Installation sizes (was: Re: How should I deal with packages that
 require FFI external libraries?
In-Reply-To: <202304212143.33LLhwJW013599@freefriends.org>
References: <202304212143.33LLhwJW013599@freefriends.org>
Message-ID: <8e3be43a-c778-2081-d985-6d5d0cfb09ed@silmaril.ie>

On 21/04/2023 22:43, Karl Berry wrote:
> The answer to why full TL is so big (currently 7.5+GB) is that it 
> includes all the fonts, support for all languages including CJK, all 
> macros, etc. Support to typeset any document in (more or less) any 
> script anywhere in the world takes space ...

If I do

cd /usr/share/texlive/texmf-dist/fonts
ls -l */*|grep '^d'| awk '{print $NF}'|sort|uniq| wc -l

I get 499 typefaces...well, in theory. Is "TeX comes with 500 fonts" 
anywhere near acceptable (modulo the abuse of the word "fonts")?

Peter

From borisv at lk.net  Mon Apr 24 18:15:49 2023
From: borisv at lk.net (Boris Veytsman)
Date: Mon, 24 Apr 2023 09:15:49 -0700
Subject: Installation sizes (was: Re: How should I deal with packages that
 require FFI external libraries?
In-Reply-To: <8e3be43a-c778-2081-d985-6d5d0cfb09ed@silmaril.ie> (message from
 Peter Flynn on Mon, 24 Apr 2023 14:52:58 +0100)
References: <202304212143.33LLhwJW013599@freefriends.org>
 <8e3be43a-c778-2081-d985-6d5d0cfb09ed@silmaril.ie>
Message-ID: <202304241615.33OGFn5d029795@bilbo.localnet>

PF> Date: Mon, 24 Apr 2023 14:52:58 +0100
PF> From: Peter Flynn <peter at silmaril.ie>


PF> If I do

PF> cd /usr/share/texlive/texmf-dist/fonts
PF> ls -l */*|grep '^d'| awk '{print $NF}'|sort|uniq| wc -l

PF> I get 499 typefaces...well, in theory. Is "TeX comes with 500 fonts" 
PF> anywhere near acceptable (modulo the abuse of the word "fonts")?

On my system

cd /usr/local/texlive/2023/texmf-dist/fonts
ls -l */*|grep '^d'| awk '{print $NF}'|sort|uniq| wc -l

gives 573...


-- 
Good luck

-Boris

From reinhard.kotucha at gmx.de  Mon Apr 24 20:35:01 2023
From: reinhard.kotucha at gmx.de (Reinhard Kotucha)
Date: Mon, 24 Apr 2023 20:35:01 +0200
Subject: Installation sizes (was: Re: How should I deal with packages that
 require FFI external libraries?
In-Reply-To: <202304241615.33OGFn5d029795@bilbo.localnet>
References: <202304212143.33LLhwJW013599@freefriends.org>
 <8e3be43a-c778-2081-d985-6d5d0cfb09ed@silmaril.ie>
 <202304241615.33OGFn5d029795@bilbo.localnet>
Message-ID: <25670.52181.480975.363144@gargle.gargle.HOWL>

On 2023-04-24 at 09:15:49 -0700, Boris Veytsman wrote:

 > PF> Date: Mon, 24 Apr 2023 14:52:58 +0100
 > PF> From: Peter Flynn <peter at silmaril.ie>
 >
 >
 > PF> If I do
 >
 > PF> cd /usr/share/texlive/texmf-dist/fonts
 > PF> ls -l */*|grep '^d'| awk '{print $NF}'|sort|uniq| wc -l
 >
 > PF> I get 499 typefaces...well, in theory. Is "TeX comes with 500 fonts"
 > PF> anywhere near acceptable (modulo the abuse of the word "fonts")?
 >
 > On my system
 >
 > cd /usr/local/texlive/2023/texmf-dist/fonts
 > ls -l */*|grep '^d'| awk '{print $NF}'|sort|uniq| wc -l
 >
 > gives 573...

... I get 572.  However I responded already, accidentally off-list.
Here is the content:

Hi Peter,
there are much, much more.  With your command you only see the tip
of the iceberg.

If you omit the wc -l you'll see vntex for instance.  vntex is not a
font or typeface.  The package contains

  arevvn
  chartervn
  cmbrightvn
  concretevn
  grotesqvn
  txttvn
  urwvn
  vnrother
  vnrtext
  vntopia

where urwvn contains all the lw35 fonts extended for Vietnamese.
There are certainly many such packages in /texmf-dist/fonts/*/* and I
have no idea how to estimate the number of typefaces without looking
manually into each directory.

  $ find . -name '*otf'|wc -l
    1872

  $ find . -name '*ttf'|wc -l
    419

  $ find . -name '*pfb'|wc -l
    7417

It's better to say "TeX comes with several thousand fonts".

Regards,
  Reinhard

--
------------------------------------------------------------------
Reinhard Kotucha                            Phone: +49-511-3373112
Marschnerstr. 25
D-30167 Hannover                    mailto:reinhard.kotucha at gmx.de
------------------------------------------------------------------


From karl at freefriends.org  Mon Apr 24 22:48:18 2023
From: karl at freefriends.org (Karl Berry)
Date: Mon, 24 Apr 2023 22:48:18 +0200
Subject: Installation sizes (was: Re: How should I deal with packages that
 require FFI external libraries?
In-Reply-To: <25670.52181.480975.363144@gargle.gargle.HOWL>
Message-ID: <202304242048.33OKmITo011018@tug.org>

Hi Peter and all,

    rk> It's better to say "TeX comes with several thousand fonts".

Several thousand font files, yes, but I think Peter wants to count
"typeface families" rather than individual pfb/ttf/otf files. 

Let's refine the command to skip map/enc/etc. which don't include any
real fonts, and to print all directories, not just those at a given level:
  find opentype source truetype tfm -type d | sed 's,.*/,,' | sort -u >/tmp/e
(The sed command reduces each path to its basename.)

I get 597 names, including arevvn and the like. Clearly still imperfect
(some fonts are missed, some non-fonts are included, some directories
have multiple designs, etc.), but "more than 500 fonts" seems like a
reasonable statement to me, given that "font" means "typeface family".

Best,
Karl

From glhoffman at gmail.com  Mon Apr 24 23:22:40 2023
From: glhoffman at gmail.com (Gary Hoffman)
Date: Mon, 24 Apr 2023 16:22:40 -0500
Subject: Problem with expl3.sty
Message-ID: <CAMGsiTGkyJh86RGbwMSSjA6gOLr0PB-i8P-O1d4s3XTkMtU_gQ@mail.gmail.com>

I am attempting to create a document in MLA format. However, I received the
following error message:

Command Line:   pdflatex.exe --interaction=errorstopmode --synctex=-1
"main.tex"
Startup Folder: Z:\Documents\MLA

This is pdfTeX, Version 3.141592653-2.6-1.40.25 (TeX Live 2023) (preloaded
format=pdflatex)
 restricted \write18 enabled.
entering extended mode
(./main.tex
LaTeX2e <2022-11-01> patch level 1
L3 programming layer <2023-03-30>
(c:/texlive/2023/texmf-dist/tex/latex/base/article.cls
Document Class: article 2022/07/02 v1.4n Standard LaTeX document class
(c:/texlive/2023/texmf-dist/tex/latex/base/size12.clo))
(c:/texlive/2023/texmf-dist/tex/generic/babel/babel.sty
(c:/texlive/2023/texmf-dist/tex/generic/babel/txtbabel.def)
(c:/texlive/2023/texmf-dist/tex/generic/babel-english/american.ldf
(c:/texlive/2023/texmf-dist/tex/generic/babel-english/english.ldf)))
(c:/texlive/2023/texmf-dist/tex/generic/babel/locale/en/babel-american.tex)
(c:/texlive/2023/texmf-dist/tex/latex/csquotes/csquotes.sty
(c:/texlive/2023/texmf-dist/tex/latex/etoolbox/etoolbox.sty)
(c:/texlive/2023/texmf-dist/tex/latex/graphics/keyval.sty)
(c:/texlive/2023/texmf-dist/tex/latex/csquotes/csquotes.def)
(c:/texlive/2023/texmf-dist/tex/latex/csquotes/csquotes.cfg))
(c:/texlive/2023/texmf-dist/tex/latex/biblatex/biblatex.sty
(c:/texlive/2023/texmf-dist/tex/generic/pdftexcmds/pdftexcmds.sty
(c:/texlive/2023/texmf-dist/tex/generic/infwarerr/infwarerr.sty)
(c:/texlive/2023/texmf-dist/tex/generic/iftex/iftex.sty)
(c:/texlive/2023/texmf-dist/tex/generic/ltxcmds/ltxcmds.sty))
(c:/texlive/2023/texmf-dist/tex/latex/kvoptions/kvoptions.sty
(c:/texlive/2023/texmf-dist/tex/latex/kvsetkeys/kvsetkeys.sty))
(c:/texlive/2023/texmf-dist/tex/latex/logreq/logreq.sty
(c:/texlive/2023/texmf-dist/tex/latex/logreq/logreq.def))
(c:/texlive/2023/texmf-dist/tex/latex/base/ifthen.sty)
(c:/texlive/2023/texmf-dist/tex/latex/url/url.sty)
(c:/texlive/2023/texmf-dist/tex/latex/biblatex/blx-dm.def)
(c:/texlive/2023/texmf-dist/tex/latex/biblatex-mla/mla.dbx)
(c:/texlive/2023/texmf-dist/tex/latex/biblatex/blx-compat.def)
(c:/texlive/2023/texmf-dist/tex/latex/biblatex/biblatex.def)
(c:/texlive/2023/texmf-dist/tex/latex/biblatex-mla/mla.bbx
(c:/texlive/2023/texmf-dist/tex/latex/biblatex-mla/mla-strict.bbx
(c:/texlive/2023/texmf-dist/tex/latex/biblatex/bbx/standard.bbx)))
(c:/texlive/2023/texmf-dist/tex/latex/biblatex-mla/mla.cbx
(c:/texlive/2023/texmf-dist/tex/latex/biblatex-mla/mla-footnotes.cbx))
(c:/texlive/2023/texmf-dist/tex/latex/biblatex/biblatex.cfg)
(c:/texlive/2023/texmf-dist/tex/latex/l3kernel/expl3.sty

! LaTeX Error: Mismatched LaTeX support files detected.
(LaTeX)        Loading 'expl3.sty' aborted!
(LaTeX)
(LaTeX)        The L3 programming layer in the LaTeX format
(LaTeX)        is dated 2023-03-30, but in your TeX tree the files require
(LaTeX)        at least 2023-04-20.

For immediate help type H <return>.
 ...

l.77      \ExplLoaderFileDate{expl3.sty}}
                                         %
?

For what it's worth, I use WinEdt 11 on Windows 11. I update Tex Live on a
nearly daily basis. So I suspect that this error arises from some change
that occurred over the last few days.

Thank you,

Gary Hoffman
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230424/f82ab293/attachment.html>

From phe.h.o1 at gmail.com  Mon Apr 24 23:32:41 2023
From: phe.h.o1 at gmail.com (Phelype Oleinik)
Date: Mon, 24 Apr 2023 18:32:41 -0300
Subject: Problem with expl3.sty
In-Reply-To: <CAMGsiTGkyJh86RGbwMSSjA6gOLr0PB-i8P-O1d4s3XTkMtU_gQ@mail.gmail.com>
References: <CAMGsiTGkyJh86RGbwMSSjA6gOLr0PB-i8P-O1d4s3XTkMtU_gQ@mail.gmail.com>
Message-ID: <CANfjC08Ka9aqfBQZkWfTLRqpY0vPDGXuV+fiNxhjC8vFgGEznA@mail.gmail.com>

Hi Gary,

On Mon, Apr 24, 2023 at 6:25?PM Gary Hoffman <glhoffman at gmail.com> wrote:
>
> ! LaTeX Error: Mismatched LaTeX support files detected.
> (LaTeX)        Loading 'expl3.sty' aborted!
> (LaTeX)
> (LaTeX)        The L3 programming layer in the LaTeX format
> (LaTeX)        is dated 2023-03-30, but in your TeX tree the files require
> (LaTeX)        at least 2023-04-20.

This error means exactly that:  your LaTeX format has expl3 2023-03-30
preloaded, but since you recently updated TeX Live, your installed expl3
is from 2023-04-20.

This error usually happens because TeX Live sometimes creates local
format files when updating, and "forgets" to regenerate them when
updating again.  The solution is usually to remove those local format
files.  See: https://tex.stackexchange.com/a/598609/134574

If you type "H <return>" at the error prompt, LaTeX will try to tell
you where that extra format file is, if it has some shell access.
Otherwise, you can easily locate the format file manually.

Hope it helps,
Phelype


From peter at silmaril.ie  Tue Apr 25 00:30:59 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Mon, 24 Apr 2023 23:30:59 +0100
Subject: Installation sizes (was: Re: How should I deal with packages that
 require FFI external libraries?
In-Reply-To: <202304242048.33OKmITo011018@tug.org>
References: <202304242048.33OKmITo011018@tug.org>
Message-ID: <33dd1c18-5e3d-7240-a3b9-a818e856ea75@silmaril.ie>

On 24/04/2023 21:48, Karl Berry wrote:
[snip]
> [...] "more than 500 fonts" seems like a reasonable statement to me,
> given that "font" means "typeface family".

Thank you all for your contributions, I'll go with that.

Of course, someone is bound to come back and say "well, what are they 
and where is the font sampler?" :-)

Peter

From reinhard.kotucha at gmx.de  Tue Apr 25 02:34:04 2023
From: reinhard.kotucha at gmx.de (Reinhard Kotucha)
Date: Tue, 25 Apr 2023 02:34:04 +0200
Subject: Installation sizes (was: Re: How should I deal with packages that
 require FFI external libraries?
In-Reply-To: <33dd1c18-5e3d-7240-a3b9-a818e856ea75@silmaril.ie>
References: <202304242048.33OKmITo011018@tug.org>
 <33dd1c18-5e3d-7240-a3b9-a818e856ea75@silmaril.ie>
Message-ID: <25671.8188.38886.165637@gargle.gargle.HOWL>

On 2023-04-24 at 23:30:59 +0100, Peter Flynn wrote:

 > On 24/04/2023 21:48, Karl Berry wrote:
 > [snip]
 > > [...] "more than 500 fonts" seems like a reasonable statement to me,
 > > given that "font" means "typeface family".
 >
 > Thank you all for your contributions, I'll go with that.
 >
 > Of course, someone is bound to come back and say "well, what are
 > they and where is the font sampler?" :-)

The definite answer is

   https://tug.org/FontCatalogue/

If you select a particular font you get a code-snippet which you can
paste into the preamble of your LaTeX document.

Regards,
  Reinhard

--
------------------------------------------------------------------
Reinhard Kotucha                            Phone: +49-511-3373112
Marschnerstr. 25
D-30167 Hannover                    mailto:reinhard.kotucha at gmx.de
------------------------------------------------------------------


From peter at silmaril.ie  Tue Apr 25 11:33:00 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Tue, 25 Apr 2023 10:33:00 +0100
Subject: Installation sizes
In-Reply-To: <25671.8188.38886.165637@gargle.gargle.HOWL>
References: <202304242048.33OKmITo011018@tug.org>
 <33dd1c18-5e3d-7240-a3b9-a818e856ea75@silmaril.ie>
 <25671.8188.38886.165637@gargle.gargle.HOWL>
Message-ID: <31760c72-dc60-0a4a-0b42-402ce04402cc@silmaril.ie>


On 25/04/2023 01:34, Reinhard Kotucha wrote:
> On 2023-04-24 at 23:30:59 +0100, Peter Flynn wrote:
>> Of course, someone is bound to come back and say "well, what are 
>> they and where is the font sampler?" :-)
> 
> The definite answer is
> 
>     https://tug.org/FontCatalogue/

Yes, it's excellent, but I was thinking of something with rather more 
examples (different sizes, and a full alphabet). Maybe I should start 
work :-)

Peter


From zdenek.wagner at gmail.com  Tue Apr 25 11:58:16 2023
From: zdenek.wagner at gmail.com (Zdenek Wagner)
Date: Tue, 25 Apr 2023 11:58:16 +0200
Subject: Installation sizes
In-Reply-To: <31760c72-dc60-0a4a-0b42-402ce04402cc@silmaril.ie>
References: <202304242048.33OKmITo011018@tug.org>
 <33dd1c18-5e3d-7240-a3b9-a818e856ea75@silmaril.ie>
 <25671.8188.38886.165637@gargle.gargle.HOWL>
 <31760c72-dc60-0a4a-0b42-402ce04402cc@silmaril.ie>
Message-ID: <CAC1phyYnmUpssyRXQXec8eNb8DLtq_ydpeRwrwpQsnMiPHoFRw@mail.gmail.com>

?t 25. 4. 2023 v 11:33 odes?latel Peter Flynn <peter at silmaril.ie> napsal:
>
>
> On 25/04/2023 01:34, Reinhard Kotucha wrote:
> > On 2023-04-24 at 23:30:59 +0100, Peter Flynn wrote:
> >> Of course, someone is bound to come back and say "well, what are
> >> they and where is the font sampler?" :-)
> >
> > The definite answer is
> >
> >     https://tug.org/FontCatalogue/
>
> Yes, it's excellent, but I was thinking of something with rather more
> examples (different sizes, and a full alphabet). Maybe I should start
> work :-)
>
It is not easy. Imagine a font such as FreeSerif which covers almost
the whole Unicode so that it contains 50 thousand characters. It even
contains variants, Devan?gar? in Sanskrit uses conjuncts not used in
nowadays Hindi, therefore in LaTeX you have to specify not only the
script but also the language. Modern web browsers also honour the
language variants. The web sites offering fonts usually contain such
small snipets and then you can clich on the font and you will see
details, you can select italics, boldface etc (whatever is available).
You can change the size and you can even type your own text which is
displayed in the selected font. It is useful for examination of
support of a particular language. For instance, in Czech we want to
see the word "lo?ka", in Slovak "ko?ko" and "st?p". We want to see
whether AV and ?V have the same kerning. In case of an Arabic font we
want to see whether it connects properly only Arabich characters or
also ? ? ? ? ? used in Urdu. Thus instead of showing the whole
alphabet it should preferably offer dislay of a particular unicode
page (selectable by the user) and the display of a text entered by the
user with the possibility of selection a language.

> Peter
>

Zden?k Wagner
https://www.zdenek-wagner.eu/


From frank.mittelbach at latex-project.org  Tue Apr 25 12:01:53 2023
From: frank.mittelbach at latex-project.org (Frank Mittelbach)
Date: Tue, 25 Apr 2023 12:01:53 +0200
Subject: Installation sizes
In-Reply-To: <31760c72-dc60-0a4a-0b42-402ce04402cc@silmaril.ie>
References: <202304242048.33OKmITo011018@tug.org>
 <33dd1c18-5e3d-7240-a3b9-a818e856ea75@silmaril.ie>
 <25671.8188.38886.165637@gargle.gargle.HOWL>
 <31760c72-dc60-0a4a-0b42-402ce04402cc@silmaril.ie>
Message-ID: <d15655bd-605f-29d7-b859-53f9081f4c58@latex-project.org>

Am 25.04.23 um 11:33 schrieb Peter Flynn:
> 
> On 25/04/2023 01:34, Reinhard Kotucha wrote:
>> On 2023-04-24 at 23:30:59 +0100, Peter Flynn wrote:
>>> Of course, someone is bound to come back and say "well, what are they 
>>> and where is the font sampler?" :-)
>>
>> The definite answer is
>>
>> ??? https://tug.org/FontCatalogue/
> 
> Yes, it's excellent, but I was thinking of something with rather more 
> examples (different sizes, and a full alphabet). Maybe I should start 
> work :-)

shameless ad:  chapter 10 of TLC3

https://www.latex-project.org/news/2023/03/17/TLC3/ pages 47-50 in the 
sample

does not contain everything that is in TL but a lot. Got a lot of help 
from Adam Twardoch to identify the really good ones and present them in 
a hopefully useful way. It classifies and shows more than 100 font 
families on 120 pages (where family is meta family if there are 
different faces or styles so in total more like 300 in Karl's counting). 
A lot of heart-blood went into that chapter to make it a bit of a useful 
coffeetable chapter --- but no, it is not free.

frank


From peter at silmaril.ie  Tue Apr 25 15:39:28 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Tue, 25 Apr 2023 14:39:28 +0100
Subject: Installation sizes
In-Reply-To: <d15655bd-605f-29d7-b859-53f9081f4c58@latex-project.org>
References: <202304242048.33OKmITo011018@tug.org>
 <33dd1c18-5e3d-7240-a3b9-a818e856ea75@silmaril.ie>
 <25671.8188.38886.165637@gargle.gargle.HOWL>
 <31760c72-dc60-0a4a-0b42-402ce04402cc@silmaril.ie>
 <d15655bd-605f-29d7-b859-53f9081f4c58@latex-project.org>
Message-ID: <345a76d7-2483-37d8-20ac-6f90c59dbfc5@silmaril.ie>


On 25/04/2023 11:01, Frank Mittelbach wrote:
> Am 25.04.23 um 11:33 schrieb Peter Flynn:
>> [...] I was thinking of something with rather more examples
>> (different sizes, and a full alphabet). Maybe I should start work
> 
> shameless ad:? chapter 10 of TLC3
> https://www.latex-project.org/news/2023/03/17/TLC3/ pages 47-50 

This is much more detailed than the simple ones at 
http://latex/formattinginformation/fontuse.html#ottt-packages (aimed at 
a less analytic beginner population).

> does not contain everything that is in TL but a lot. Got a lot of
> help from Adam Twardoch to identify the really good ones and present
> them in a hopefully useful way.

It is very useful, especially because it shows bold, italic, and small 
caps as well as the roman.

> It classifies 

As a sideline, I am working on a simple recognition/identity system for 
faces based on the appearance of some letterforms ? for example the 
double-loop or fish-hook lowercase "g" ? designed to let beginners 
identify the typeface they want. More later.

> and shows more than 100 font families on 120 pages

It's really impressive

> A lot of heart-blood went into that chapter to make it a bit of a
> useful coffeetable chapter --- but no, it is not free.
That's also fine. The work involved is very significant, and we are 
indebted to you and Ulrike and your contributors. It would make a nice 
booklet for the coffee table too, one that people would be happy to pay for.

Peter


From peter at silmaril.ie  Tue Apr 25 15:48:24 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Tue, 25 Apr 2023 14:48:24 +0100
Subject: Typeface identification
Message-ID: <fc94fcbe-b55c-cc2a-1f14-eb0c9c0d9ad6@silmaril.ie>


> As a sideline, I am working on a simple recognition/identity system
> for faces based on the appearance of some letterforms ? for example
> the double-loop or fish-hook lowercase "g" ? designed to let
> beginners identify the typeface they want.
Beginners usually lack the technical vocabulary to describe their 
difficulties, especially when English is not their native language, and 
they often use the only words they can think of, which sometimes leads 
to confusion.

Unlike other classifications (Vox, especially) I am not trying to group 
the faces into categories, but to provide some features which can be 
used to locate a specific face.

(It was triggered by requests like "What's that typeface where the bar 
on the e is slanted and the capitals are lower than the ascenders and 
the f is so narrow you hardly need ff fl fi ffl ffi?")

As a test, I'm slowly working through the faces listed in the LaTeX Font 
Catalogue, building on the list of features as I go. So far:

a-form	hooked/round
e-form	horixontal/sloped
g-form	fish-hook/double-loop
Ascenders	above/cap-height/below
Serifs	bracketed/square/hairline
R-form	straight/convex/concave/s-shape
c-form	bulb/stroke

I am not trying to capture all design features, just those that are 
immediately explainable and visible to a beginner.

If anyone else has done this before, please scream now :-)


-- 
Peter Flynn
Cork ?? Ireland ??


From jfine2358 at gmail.com  Tue Apr 25 21:27:07 2023
From: jfine2358 at gmail.com (Jonathan Fine)
Date: Tue, 25 Apr 2023 20:27:07 +0100
Subject: Thu 27 Apr: TeX Hour: Documents, warnings and errors at scale: 6:30
 to 7:30 BST
In-Reply-To: <CALD=Yf9OcmLToaQtoEio2P-nKqHSE054TvQNaQ5YFG_Sdmuxdw@mail.gmail.com>
References: <76108cbf-c075-45b9-a31b-32dc88233351n@googlegroups.com>
 <CALD=Yf8+qMwiHVMr9A8w4MfBDz3vTuOhk0SBAf2y7XXEp8wJLg@mail.gmail.com>
 <CALD=Yf9saEBGfh4oQsPjGFN4F68hGzPrHYU-_qeZhPNR3FSHcg@mail.gmail.com>
 <CALD=Yf9OcmLToaQtoEio2P-nKqHSE054TvQNaQ5YFG_Sdmuxdw@mail.gmail.com>
Message-ID: <CALD=Yf_GqfqF0AfjAXo-ab-M73BmVG87UNE4sM3ZVuO3qDPcDQ@mail.gmail.com>

Hi

The arXiv contains over 2.5 million STEM articles, most written in LaTeX.
This week's TeX Hour is a small contribution to improving the rendering of
archived articles to HTML. The focus is on the backlog. You can help if you
want to. (Few special skills beyond knowledge of TeX required.)

TeX Hour: Thursday 27 April, 6:30 to 7:30pm
More information:
https://texhour.github.io/2023/04/27/docs-warns-errs-scale/
Zoom URL:
https://us02web.zoom.us/j/78551255396?pwd=cHdJN0pTTXRlRCtSd1lCTHpuWmNIUT09

wishing you well

Jonathan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230425/0bdbd704/attachment.html>

From zarniwhoop at ntlworld.com  Tue Apr 25 21:50:58 2023
From: zarniwhoop at ntlworld.com (Ken Moffat)
Date: Tue, 25 Apr 2023 20:50:58 +0100
Subject: Typeface identification
In-Reply-To: <fc94fcbe-b55c-cc2a-1f14-eb0c9c0d9ad6@silmaril.ie>
References: <fc94fcbe-b55c-cc2a-1f14-eb0c9c0d9ad6@silmaril.ie>
Message-ID: <ZEgvIopjhsxqlPpE@llamedos.localdomain>

On Tue, Apr 25, 2023 at 02:48:24PM +0100, Peter Flynn wrote:
> 
> > As a sideline, I am working on a simple recognition/identity system
> > for faces based on the appearance of some letterforms ? for example
> > the double-loop or fish-hook lowercase "g" ? designed to let
> > beginners identify the typeface they want.
> Beginners usually lack the technical vocabulary to describe their
> difficulties, especially when English is not their native language, and they
> often use the only words they can think of, which sometimes leads to
> confusion.
> 

As a native English speaker, most of the details for how to describe
the variations in fonts are not common knowledge.  In the past I was
interested in trying to use TTF and OTF fonts to cover the main
languages I was[1] likely to find on the web (and a few less-common
variants).  In the end I went with the descriptions from wikipedia
when I could find them (and it is possible I misclassified some
fotns).

Those results are at http://zarniwhoop.uk/ttf-otf-notes.html
(now old, but very little now changes in unicode for the
subset of current languages found on the web).

Please note it is plain http://, also (if anyone accesses it) the
'Please send any comments to' should be disregarded - that address
no-longer gets to me.

Anyway - the TeX connection is that I used xelatex to produce the
PDFs of Glyphs and Languages.  For me, part of the interest was
"does it reliably cover what I encounter", part was "does it look
nice? (very subjective, my ideas of niceness are a million miles
away from Knuth's).  Oh, and for general creation of (non-TeX)
documents it helps to not have very many fonts (scrolling down
though several hundred when editing a paragraph to use a new font is
not fun).

> Unlike other classifications (Vox, especially) I am not trying to group the
> faces into categories, but to provide some features which can be used to
> locate a specific face.
> 
> (It was triggered by requests like "What's that typeface where the bar on
> the e is slanted and the capitals are lower than the ascenders and the f is
> so narrow you hardly need ff fl fi ffl ffi?")
> 
> As a test, I'm slowly working through the faces listed in the LaTeX Font
> Catalogue, building on the list of features as I go. So far:
> 
> a-form	hooked/round
> e-form	horixontal/sloped
> g-form	fish-hook/double-loop
> Ascenders	above/cap-height/below
> Serifs	bracketed/square/hairline
> R-form	straight/convex/concave/s-shape
> c-form	bulb/stroke
> 
> I am not trying to capture all design features, just those that are
> immediately explainable and visible to a beginner.
> 
> If anyone else has done this before, please scream now :-)
> 
> 
> -- 
> Peter Flynn
> Cork ?? Ireland ??

Sounds interesting.

?en

1. In those days it was easy to use google news to look at reports
from different countries.

-- 
git gets easier once you get the basic idea that branches are
homeomorphic endofunctors mapping submanifolds of a Hilbert space.
                -- Isaac Wolkerstorfer

From karl at freefriends.org  Tue Apr 25 22:30:45 2023
From: karl at freefriends.org (Karl Berry)
Date: Tue, 25 Apr 2023 14:30:45 -0600
Subject: Problem with expl3.sty
In-Reply-To: <CANfjC08Ka9aqfBQZkWfTLRqpY0vPDGXuV+fiNxhjC8vFgGEznA@mail.gmail.com>
Message-ID: <202304252030.33PKUjHZ005882@freefriends.org>

    This error usually happens because TeX Live sometimes creates local
    format files when updating, and "forgets" to regenerate them 
    ...
    https://tex.stackexchange.com/a/598609/134574

As far as I can think of, install-tl/tlmgr never creates "local" format
files, despite the statements in that tex.sx post. Users have to
knowingly create them themselves by running fmtutil-user. Am I wrong?
Anyway, it would be wrong for tlmgr to update them. The advice in that
post to remove them is usually what is desired. --best, karl.

From joseph.wright at morningstar2.co.uk  Tue Apr 25 22:33:56 2023
From: joseph.wright at morningstar2.co.uk (Joseph Wright)
Date: Tue, 25 Apr 2023 21:33:56 +0100
Subject: Problem with expl3.sty
In-Reply-To: <202304252030.33PKUjHZ005882@freefriends.org>
References: <202304252030.33PKUjHZ005882@freefriends.org>
Message-ID: <85db59cc-ce07-8249-8710-ac75968e61e0@morningstar2.co.uk>

On 25/04/2023 21:30, Karl Berry wrote:
>      This error usually happens because TeX Live sometimes creates local
>      format files when updating, and "forgets" to regenerate them
>      ...
>      https://tex.stackexchange.com/a/598609/134574
> 
> As far as I can think of, install-tl/tlmgr never creates "local" format
> files, despite the statements in that tex.sx post. Users have to
> knowingly create them themselves by running fmtutil-user. Am I wrong?
> Anyway, it would be wrong for tlmgr to update them. The advice in that
> post to remove them is usually what is desired. --best, karl.

We've seen this repeatedly, with some users who definitely know what 
they are doing ...

Joseph

From peter at silmaril.ie  Tue Apr 25 22:58:53 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Tue, 25 Apr 2023 21:58:53 +0100
Subject: Typeface identification
In-Reply-To: <ZEgvIopjhsxqlPpE@llamedos.localdomain>
References: <fc94fcbe-b55c-cc2a-1f14-eb0c9c0d9ad6@silmaril.ie>
 <ZEgvIopjhsxqlPpE@llamedos.localdomain>
Message-ID: <7abf7608-90e0-73e1-c56e-86eb93d9579a@silmaril.ie>


On 25/04/2023 20:50, Ken Moffat via tex-live wrote:
[...]
> As a native English speaker, most of the details for how to describe
> the variations in fonts are not common knowledge.  

Yep. Talk of bowls and bracketed serifs is baffling for the font beginner.

> [...] Those results are at http://zarniwhoop.uk/ttf-otf-notes.html 
> (now old, but very little now changes in unicode for the subset of
> current languages found on the web).

That is a very useful page. I'd like to refer people to it when they 
want to find out about language support.

> Please note it is plain http://, also (if anyone accesses it) the 
> 'Please send any comments to' should be disregarded - that address 
> no longer gets to me.

Noted.

> [...] (very subjective, my ideas of niceness are a million miles away
> from Knuth's).

American and European typographic aesthetics have some significant 
differences too :-)

> Oh, and for general creation of (non-TeX) documents it helps to not
> have very many fonts (scrolling down though several hundred when
> editing a paragraph to use a new font is not fun).

I can't imagine what it must be like to be stuck with a graphical font 
selection tool that insists on using each font to render each entry in 
the drop-down menu, when you have several thousand fonts to pick from.

Peter

From jacobgyntelberg at mac.com  Wed Apr 26 00:10:42 2023
From: jacobgyntelberg at mac.com (Jacob Gyntelberg)
Date: Wed, 26 Apr 2023 00:10:42 +0200
Subject: Problem with expl3.sty
In-Reply-To: <85db59cc-ce07-8249-8710-ac75968e61e0@morningstar2.co.uk>
References: <85db59cc-ce07-8249-8710-ac75968e61e0@morningstar2.co.uk>
Message-ID: <2A2F9A9D-6CFA-448F-B5F2-0F00DC90D747@mac.com>

?

Sent from my iPhone

> On 25 Apr 2023, at 22:34, Joseph Wright <joseph.wright at morningstar2.co.uk> wrote:
> 
> ?On 25/04/2023 21:30, Karl Berry wrote:
>>     This error usually happens because TeX Live sometimes creates local
>>     format files when updating, and "forgets" to regenerate them
>>     ...
>>     https://tex.stackexchange.com/a/598609/134574
>> As far as I can think of, install-tl/tlmgr never creates "local" format
>> files, despite the statements in that tex.sx post. Users have to
>> knowingly create them themselves by running fmtutil-user. Am I wrong?
>> Anyway, it would be wrong for tlmgr to update them. The advice in that
>> post to remove them is usually what is desired. --best, karl.
> 
> We've seen this repeatedly, with some users who definitely know what they are doing ...
> 
> Joseph


From andreas_tex at freenet.de  Wed Apr 26 08:48:01 2023
From: andreas_tex at freenet.de (Andreas Scherer)
Date: Wed, 26 Apr 2023 08:48:01 +0200
Subject: Typeface identification
In-Reply-To: <fc94fcbe-b55c-cc2a-1f14-eb0c9c0d9ad6@silmaril.ie>
References: <fc94fcbe-b55c-cc2a-1f14-eb0c9c0d9ad6@silmaril.ie>
Message-ID: <ce69b1ed-72b4-4dd3-fbe8-01791fff7af7@freenet.de>

Peter,

 > I am not trying to capture all design features, just those that are
 > immediately explainable and visible to a beginner.
 >
 > If anyone else has done this before, please scream now :-)

In 1991, I bought "Rookledge's International Typefinder". Sarema Press, 
1990, curiously marked as "not for sale in the UK". It lists over 700 
typeface specimens in several categories. As a second method for 
identifying a font is the use of a selection order of "earmarks" 
(Capital letters, lower case letters, figures).

A more recent acquisition is Erik Spiekermann's "Sheep book" ("Find our 
how type works", 4th ed., The Other Collection, 2022) with tons of type 
examples; it has a three-page "typeface index".

Later,
Andreas


From andreas_tex at freenet.de  Wed Apr 26 08:48:18 2023
From: andreas_tex at freenet.de (Andreas Scherer)
Date: Wed, 26 Apr 2023 08:48:18 +0200
Subject: Typeface identification
In-Reply-To: <7abf7608-90e0-73e1-c56e-86eb93d9579a@silmaril.ie>
References: <7abf7608-90e0-73e1-c56e-86eb93d9579a@silmaril.ie>
Message-ID: <c5e3eec6-dc1d-a18d-3f8b-26c5784cce52@freenet.de>

Peter,

 > I can't imagine what it must be like to be stuck with a graphical font
 > selection tool that insists on using each font to render each entry in
 > the drop-down menu, when you have several thousand fonts to pick from.

People are complaining: 
https://forum.affinity.serif.com/index.php?/topic/177870-we-need-flexible-font-categories

(At least on macOS, Affinity users can resort to font categories that 
they can define in the system font management tool. Inkscape 1.3 seems 
to bring a similar approach on all platforms.)

Best,
Andreas

From peter at silmaril.ie  Wed Apr 26 11:10:43 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Wed, 26 Apr 2023 10:10:43 +0100
Subject: Typeface identification
In-Reply-To: <c5e3eec6-dc1d-a18d-3f8b-26c5784cce52@freenet.de>
References: <7abf7608-90e0-73e1-c56e-86eb93d9579a@silmaril.ie>
 <c5e3eec6-dc1d-a18d-3f8b-26c5784cce52@freenet.de>
Message-ID: <054f0d4e-0406-cec3-8be3-b2fd81448b8c@silmaril.ie>

On 26/04/2023 07:48, Andreas Scherer wrote:
> Peter,
> 
>  > I can't imagine what it must be like to be stuck with a graphical font
>  > selection tool that insists on using each font to render each entry in
>  > the drop-down menu, when you have several thousand fonts to pick from.
> 
> People are complaining: 
> https://forum.affinity.serif.com/index.php?/topic/177870-we-need-flexible-font-categories

Vox and Vox-based systems do provide these categories, but until 
recently, the Vox ecosystem was used (and managed?) by Atypi. I don't 
know what the position is since they stopped using it. The problem isn't 
a shortage of candidate ontologies but which one to choose from, because 
switching to another one afterwards is probably expensive.

> (At least on macOS, Affinity users can resort to font categories that 
> they can define in the system font management tool. Inkscape 1.3 seems 
> to bring a similar approach on all platforms.)

The implementation of collections is a step in the right direction. The 
big problem is that the categorisation is a human job because the 
metadata provided by the font-file creators is often very poor (and 
there isn't much provision to store it). For example, how would you go 
about listing all the sans-serif fonts you have installed, having access 
only to the font files themselves? AFAIK the only indication *might* be 
in the font name or the file name; but Raleway (for example) has no 
indication that I can find that it is a sans-serif typeface.

Peter

From peter at silmaril.ie  Wed Apr 26 11:23:10 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Wed, 26 Apr 2023 10:23:10 +0100
Subject: Typeface identification
In-Reply-To: <ce69b1ed-72b4-4dd3-fbe8-01791fff7af7@freenet.de>
References: <fc94fcbe-b55c-cc2a-1f14-eb0c9c0d9ad6@silmaril.ie>
 <ce69b1ed-72b4-4dd3-fbe8-01791fff7af7@freenet.de>
Message-ID: <777a2917-dd94-8fba-2a6d-830d285afe22@silmaril.ie>

On 26/04/2023 07:48, Andreas Scherer wrote:
> Peter,
> 
>  > I am not trying to capture all design features, just those that are
>  > immediately explainable and visible to a beginner.
>  >
>  > If anyone else has done this before, please scream now :-)
> 
> In 1991, I bought "Rookledge's International Typefinder". Sarema Press, 
> 1990, curiously marked as "not for sale in the UK". 

\begin{rant}
That was at a time when the private cartel of publishers was still in 
operation, which stitched up the world into zones in which they agreed 
not to sell their editions in each others territories. That excuse "For 
copyright reasons this edition is not for sale in..." was just that: an 
excuse to shaft the book-buying public.  Publishers are still at it 
today: the zones used by DVDs and ebooks are a modern implementation of 
the same racket.
\end{rant}

> It lists over 700 typeface specimens in several categories. As a
> second method for identifying a font is the use of a selection order
> of "earmarks" (Capital letters, lower case letters, figures).

It's a wonderful collection.

> A more recent acquisition is Erik Spiekermann's "Sheep book" ("Find
> our how type works", 4th ed., The Other Collection, 2022) with tons
> of type examples; it has a three-page "typeface index".

That too, I love it.

Peter

From jfkthame at gmail.com  Wed Apr 26 12:38:03 2023
From: jfkthame at gmail.com (Jonathan Kew)
Date: Wed, 26 Apr 2023 11:38:03 +0100
Subject: Typeface identification
In-Reply-To: <054f0d4e-0406-cec3-8be3-b2fd81448b8c@silmaril.ie>
References: <7abf7608-90e0-73e1-c56e-86eb93d9579a@silmaril.ie>
 <c5e3eec6-dc1d-a18d-3f8b-26c5784cce52@freenet.de>
 <054f0d4e-0406-cec3-8be3-b2fd81448b8c@silmaril.ie>
Message-ID: <a2d350a1-8b70-b7fd-b76d-97820f89b2f2@gmail.com>

On 26/04/2023 10:10, Peter Flynn wrote:
> The implementation of collections is a step in the right direction. The 
> big problem is that the categorisation is a human job because the 
> metadata provided by the font-file creators is often very poor (and 
> there isn't much provision to store it). For example, how would you go 
> about listing all the sans-serif fonts you have installed, having access 
> only to the font files themselves? AFAIK the only indication *might* be 
> in the font name or the file name; but Raleway (for example) has no 
> indication that I can find that it is a sans-serif typeface.

In theory, for OpenType fonts you should be able to determine this from 
the panose classification values in the OS/2 table.

And looking at Raleway, it does indeed identify itself as sans-serif:

     <panose>
       <bFamilyType value="2"/>        Latin Text
       <bSerifStyle value="11"/>       Normal Sans
       <bWeight value="5"/>            Book
       <bProportion value="3"/>        Modern
       <bContrast value="3"/>          Very Low
       <bStrokeVariation value="1"/>   No Fit
       <bArmStyle value="1"/>          No Fit
       <bLetterForm value="6"/>        Normal/Rounded
       <bMidline value="0"/>           Any
       <bXHeight value="3"/>           Constant/Standard
     </panose>

However, I wouldn't trust the panose data to be accurate/meaningful in 
many of the fonts out there in the world... maybe some heroic person 
would care to do (or has done) a survey to see how often these values 
are set reasonably by designers/vendors.

JK


From peter at silmaril.ie  Wed Apr 26 16:42:15 2023
From: peter at silmaril.ie (Peter Flynn)
Date: Wed, 26 Apr 2023 15:42:15 +0100
Subject: Typeface identification
In-Reply-To: <a2d350a1-8b70-b7fd-b76d-97820f89b2f2@gmail.com>
References: <7abf7608-90e0-73e1-c56e-86eb93d9579a@silmaril.ie>
 <c5e3eec6-dc1d-a18d-3f8b-26c5784cce52@freenet.de>
 <054f0d4e-0406-cec3-8be3-b2fd81448b8c@silmaril.ie>
 <a2d350a1-8b70-b7fd-b76d-97820f89b2f2@gmail.com>
Message-ID: <bd291b6e-e8a9-5eb3-13d6-37074ec4bea0@silmaril.ie>

On 26/04/2023 11:38, Jonathan Kew wrote:
> On 26/04/2023 10:10, Peter Flynn wrote:
>> [...] but Raleway (for example) has no indication that I can find
>> that it is a sans-serif typeface.
> 
> In theory, for OpenType fonts you should be able to determine this from 
> the panose classification values in the OS/2 table.

Ooooh. I love being proved wrong! Thank you. I'm only just beginning to 
use OT fonts, so I hadn't come across this.
> 
> And looking at Raleway, it does indeed identify itself as sans-serif:

Is there a command-line tool that extracts panose data? I can see 
several tools to *edit* it, and tools like panosifier that you have to 
embed in a Python script, but not a simple extracter. neither fc-list 
nor otfinfo make any mention of panose.

>  ??? <panose>
>  ????? <bFamilyType value="2"/>??????? Latin Text
>  ????? <bSerifStyle value="11"/>?????? Normal Sans
>  ????? <bWeight value="5"/>??????????? Book
>  ????? <bProportion value="3"/>??????? Modern
>  ????? <bContrast value="3"/>????????? Very Low
>  ????? <bStrokeVariation value="1"/>?? No Fit
>  ????? <bArmStyle value="1"/>????????? No Fit
>  ????? <bLetterForm value="6"/>??????? Normal/Rounded
>  ????? <bMidline value="0"/>?????????? Any
>  ????? <bXHeight value="3"/>?????????? Constant/Standard
>  ??? </panose>
> 
> However, I wouldn't trust the panose data to be accurate/meaningful 

That's another whole problem. Metadata isn't a concern for most makers.

> many of the fonts out there in the world... maybe some heroic person 
> would care to do (or has done) a survey to see how often these values 
> are set reasonably by designers/vendors.

I can certainly do it for the fonts I'm dealing with from the LaTeX Font 
Catalog if I can find a tool to extract the data in a manipulable form 
(XML would be perfect).

Peter


From P.Taylor at Rhul.Ac.Uk  Wed Apr 26 18:05:49 2023
From: P.Taylor at Rhul.Ac.Uk (Philip Taylor (RHBNC))
Date: Wed, 26 Apr 2023 17:05:49 +0100
Subject: [EXT] Re: Typeface identification
In-Reply-To: <054f0d4e-0406-cec3-8be3-b2fd81448b8c@silmaril.ie>
References: <7abf7608-90e0-73e1-c56e-86eb93d9579a@silmaril.ie>
 <c5e3eec6-dc1d-a18d-3f8b-26c5784cce52@freenet.de>
 <054f0d4e-0406-cec3-8be3-b2fd81448b8c@silmaril.ie>
Message-ID: <aa488dcc-beb2-64f4-d6ac-96d0e892b1ec@Rhul.Ac.Uk>

On 26/04/2023 10:10, Peter Flynn wrote:
Raleway (for example) has no indication that I can find that it is a sans-serif typeface.

If I interpret https://fonts.adobe.com/fonts/raleway#details-section+raleway-thin correctly, Adobe classify it as being sans-serif, light in weight, regular width, high x-height, low contrast, both cases, and with old-style numerals.

--
Philip Taylor

This email, its contents and any attachments are intended solely for the addressee and may contain confidential information. In certain circumstances, it may also be subject to legal privilege. Any unauthorised use, disclosure, or copying is not permitted. If you have received this email in error, please notify us and immediately and permanently delete it. Any views or opinions expressed in personal emails are solely those of the author and do not necessarily represent those of Royal Holloway, University of London. It is your responsibility to ensure that this email and any attachments are virus free.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230426/ca6bb493/attachment.html>

From jfkthame at gmail.com  Wed Apr 26 18:24:50 2023
From: jfkthame at gmail.com (Jonathan Kew)
Date: Wed, 26 Apr 2023 18:24:50 +0200
Subject: Typeface identification
In-Reply-To: <bd291b6e-e8a9-5eb3-13d6-37074ec4bea0@silmaril.ie>
References: <7abf7608-90e0-73e1-c56e-86eb93d9579a@silmaril.ie>
 <c5e3eec6-dc1d-a18d-3f8b-26c5784cce52@freenet.de>
 <054f0d4e-0406-cec3-8be3-b2fd81448b8c@silmaril.ie>
 <a2d350a1-8b70-b7fd-b76d-97820f89b2f2@gmail.com>
 <bd291b6e-e8a9-5eb3-13d6-37074ec4bea0@silmaril.ie>
Message-ID: <932f8970-857c-45a6-cf80-ea0a9c7224cf@gmail.com>

On 26/04/2023 15:42, Peter Flynn wrote:
> On 26/04/2023 11:38, Jonathan Kew wrote:
>> On 26/04/2023 10:10, Peter Flynn wrote:
>>> [...] but Raleway (for example) has no indication that I can find
>>> that it is a sans-serif typeface.
>>
>> In theory, for OpenType fonts you should be able to determine this 
>> from the panose classification values in the OS/2 table.
> 
> Ooooh. I love being proved wrong! Thank you. I'm only just beginning to 
> use OT fonts, so I hadn't come across this.
>>
>> And looking at Raleway, it does indeed identify itself as sans-serif:
> 
> Is there a command-line tool that extracts panose data?

The example I showed was extracted using TTX, from the python FontTools 
package,[1] Dump the OS/2 table with a command like:

     ttx -t 'OS/2' Raleway-Regular.otf

to get an XML file "Raleway-Regular.ttx". Then I looked through the 
Monotype github pages[2] to see what the numbers mean, and added those 
glosses.

- JK


[1] https://github.com/fonttools/fonttools
[2] https://monotype.github.io/panose/pan2.htm

> I can see 
> several tools to *edit* it, and tools like panosifier that you have to 
> embed in a Python script, but not a simple extracter. neither fc-list 
> nor otfinfo make any mention of panose.
> 
>> ???? <panose>
>> ?????? <bFamilyType value="2"/>??????? Latin Text
>> ?????? <bSerifStyle value="11"/>?????? Normal Sans
>> ?????? <bWeight value="5"/>??????????? Book
>> ?????? <bProportion value="3"/>??????? Modern
>> ?????? <bContrast value="3"/>????????? Very Low
>> ?????? <bStrokeVariation value="1"/>?? No Fit
>> ?????? <bArmStyle value="1"/>????????? No Fit
>> ?????? <bLetterForm value="6"/>??????? Normal/Rounded
>> ?????? <bMidline value="0"/>?????????? Any
>> ?????? <bXHeight value="3"/>?????????? Constant/Standard
>> ???? </panose>
>>
>> However, I wouldn't trust the panose data to be accurate/meaningful 
> 
> That's another whole problem. Metadata isn't a concern for most makers.
> 
>> many of the fonts out there in the world... maybe some heroic person 
>> would care to do (or has done) a survey to see how often these values 
>> are set reasonably by designers/vendors.
> 
> I can certainly do it for the fonts I'm dealing with from the LaTeX Font 
> Catalog if I can find a tool to extract the data in a manipulable form 
> (XML would be perfect).
> 
> Peter


From karl at freefriends.org  Wed Apr 26 19:34:47 2023
From: karl at freefriends.org (Karl Berry)
Date: Wed, 26 Apr 2023 11:34:47 -0600
Subject: Problem with expl3.sty
In-Reply-To: <85db59cc-ce07-8249-8710-ac75968e61e0@morningstar2.co.uk>
Message-ID: <202304261734.33QHYlD1006472@freefriends.org>

    We've seen this repeatedly,

I don't doubt your observation.  Please provide a recipe to reproduce. 

The only way I see tlmgr invoking fmtutil-user is if --usermode is
explicitly specified, which I doubt these expert users are doing. (I
also presume they are not explicitly invoking fmtutil-user.)
Maybe I'm missing something.

    with some users who definitely know what they are doing ...

Or maybe that's the problem :). -k

From karl at freefriends.org  Wed Apr 26 19:34:42 2023
From: karl at freefriends.org (Karl Berry)
Date: Wed, 26 Apr 2023 11:34:42 -0600
Subject: Typeface identification
In-Reply-To: <ce69b1ed-72b4-4dd3-fbe8-01791fff7af7@freenet.de>
Message-ID: <202304261734.33QHYg2p006399@freefriends.org>

    A more recent acquisition is Erik Spiekermann's "Sheep book" ("Find our 
    how type works", 4th ed., The Other Collection, 2022) with tons of type 
    examples; it has a three-page "typeface index".

Let me mention that this book is available at no charge online, thanks
to Google, and was reviewed for TUGboat.
https://tug.org/books/#stealingsheep


From mseven at telus.net  Thu Apr 27 11:26:32 2023
From: mseven at telus.net (Max Chernoff)
Date: Thu, 27 Apr 2023 03:26:32 -0600
Subject: Problem with expl3.sty
In-Reply-To: <202304261734.33QHYlD1006472@freefriends.org>
Message-ID: <6dec8fd29f975da5cc6d70c3c1af93eeb28a62a7.camel@telus.net>

Hi Karl,

> Please provide a recipe to reproduce.
> 
> The only way I see tlmgr invoking fmtutil-user is if --usermode is
> explicitly specified, which I doubt these expert users are doing. (I
> also presume they are not explicitly invoking fmtutil-user.)
> Maybe I'm missing something.
> 
>     with some users who definitely know what they are doing ...

I've managed to accidentally make a user format before by deleting the
system format file:

   $ sudo rm $(kpsewhich --engine=pdftex pdftex.fmt)
   $ pdftex '\end'
   This is pdfTeX, Version 3.141592653-2.6-1.40.25 (TeX Live 2023) (preloaded format=pdftex)
    restricted \write18 enabled.
   
   kpathsea: Running mktexfmt pdftex.fmt
   [...]
   mktexfmt [INFO]: /home/max/.texlive2023/texmf-var/web2c/pdftex/pdftex.fmt installed.
   mktexfmt [INFO]: disabled formats: 17
   mktexfmt [INFO]: successfully rebuilt formats: 1
   mktexfmt [INFO]: not selected formats: 40
   mktexfmt [INFO]: total formats: 58
   mktexfmt [INFO]: 
   *************************************************************
   *                                                           *
   * WARNING: you are switching to fmtutil's per-user formats. *
   *         Please read the following warnings!               *
   *                                                           *
   *************************************************************
   [...]
   mktexfmt [INFO]: exiting with status 0
   entering extended mode
   No pages of output.
   Transcript written on texput.log.
   
It would be pretty hard to do this accidentally. I guess it could happen
if fmtutil-sys crashes on the initial install or if an antivirus
randomly deletes the file. Pretty unlikely though.

(I've also found that LuaTeX will build per-user format files if you
replace the one in the TL binary directory with a newer one built from
source, but there's no way that an ordinary user would ever do that.)

I doubt that any of this is very helpful, but maybe it will give someone
some clues.

Thanks,
-- Max


From karl at freefriends.org  Thu Apr 27 23:26:32 2023
From: karl at freefriends.org (Karl Berry)
Date: Thu, 27 Apr 2023 15:26:32 -0600
Subject: Problem with expl3.sty
In-Reply-To: <6dec8fd29f975da5cc6d70c3c1af93eeb28a62a7.camel@telus.net>
Message-ID: <202304272126.33RLQWU5020165@freefriends.org>

Hi Max,

    I've managed to accidentally make a user format before by deleting the
    system format file:

True, if dynamic mktexfmt runs, it will create user formats.
I don't think that could or should be changed.

But maybe we should disable mktexfmt by default. Norbert, wdyt?

    (I've also found that LuaTeX will build per-user format files if you
    replace the one in the TL binary directory with a newer one 

I don't understand ... -k

From mseven at telus.net  Fri Apr 28 06:53:18 2023
From: mseven at telus.net (Max Chernoff)
Date: Thu, 27 Apr 2023 22:53:18 -0600
Subject: Problem with expl3.sty
In-Reply-To: <202304272126.33RLQWU5020165@freefriends.org>
References: <202304272126.33RLQWU5020165@freefriends.org>
Message-ID: <c0a6ba7be00d8647aef0b185589431eb43f9df6f.camel@telus.net>

Hi Karl,

>     (I've also found that LuaTeX will build per-user format files if you
>     replace the one in the TL binary directory with a newer one 
> 
> I don't understand

I've been testing some changes to LuaTeX by doing something like this:

   $ git clone https://github.com/TeX-Live/luatex/
   $ cd luatex
   [make whatever changes to the source]
   $ ./build.sh --make --parallel
   $ sudo cp build/texk/web2c/luatex /usr/local/texlive/2023/bin/x86_64-linux/luatex
   
Then, when I run:

   $ luatex '\end'
   
I've had all of the following happen:

   1. LuaTeX compiles the file as normal
      
   2. "Fatal format file error; I?m stymied"
      
   3. LuaTeX detects that the format files are old, so it rebuilds the
      format files just like it does if you've deleted them.

No regular user would ever do this, so this unpredictable behaviour is
perfectly fine with me.

Also, if I symlink "luahbtex" to the freshly-built "luatex", then
running "lualatex" will rebuild its format every time that you run it.
Doing this is obviously completely unsupported, so this behaviour isn't
a problem either.

Thanks,
-- Max



From denis.bitouze at univ-littoral.fr  Fri Apr 28 09:55:35 2023
From: denis.bitouze at univ-littoral.fr (=?utf-8?Q?Denis_Bitouz=C3=A9?=)
Date: Fri, 28 Apr 2023 09:55:35 +0200
Subject: `notocondensed` package: why not provided by TL?
Message-ID: <87ildge8zs.fsf@example.com>

Hi,

the `notocondensed` package is provided by TeX ?Live Contrib and MiKTeX
and not by TeX ?Live, though the licenses (SIL Open Font License
and LaTeX Project Public License) seems to be compatible.

Why?

Thanks.
-- 
Denis


From frank.mittelbach at latex-project.org  Fri Apr 28 10:07:35 2023
From: frank.mittelbach at latex-project.org (Frank Mittelbach)
Date: Fri, 28 Apr 2023 10:07:35 +0200
Subject: `notocondensed` package: why not provided by TL?
In-Reply-To: <87ildge8zs.fsf@example.com>
References: <87ildge8zs.fsf@example.com>
Message-ID: <864fa0d1-e731-f54c-0a1d-5fa1d1b50d07@latex-project.org>

Am 28.04.23 um 09:55 schrieb Denis Bitouz?:
> Hi,
> 
> the `notocondensed` package is provided by TeX ?Live Contrib and MiKTeX
> and not by TeX ?Live, though the licenses (SIL Open Font License
> and LaTeX Project Public License) seems to be compatible.
> 
> Why?

too big (caught me by surprise too a while back).

TLC3 therefore says on that one:

The condensed faces have their own support packages notocondensed (for 
Serif and Sans) and notocondensed-mono.
Because of their size, the condensed fonts are not automatically 
included in the TeX Live distribution but need a simple manual 
installation. Follow the instructions at installation 
https://contrib.texlive.info/ for that or download the package from CTAN.

frank

From denis.bitouze at univ-littoral.fr  Fri Apr 28 12:30:28 2023
From: denis.bitouze at univ-littoral.fr (=?utf-8?Q?Denis_Bitouz=C3=A9?=)
Date: Fri, 28 Apr 2023 12:30:28 +0200
Subject: `notocondensed` package: why not provided by TL?
In-Reply-To: <864fa0d1-e731-f54c-0a1d-5fa1d1b50d07@latex-project.org> (Frank
 Mittelbach's message of "Fri, 28 Apr 2023 10:07:35 +0200")
References: <87ildge8zs.fsf@example.com>
 <864fa0d1-e731-f54c-0a1d-5fa1d1b50d07@latex-project.org>
Message-ID: <87edo4e1tn.fsf@example.com>

Le 28/04/23 ? 10h07, Frank Mittelbach a ?crit :

> too big (caught me by surprise too a while back).
>
> TLC3 therefore says on that one:
>
> The condensed faces have their own support packages notocondensed (for Serif and
> Sans) and notocondensed-mono.
> Because of their size, the condensed fonts are not automatically included in the
> TeX Live distribution but need a simple manual installation. Follow the
> instructions at installation https://contrib.texlive.info/ for that or download
> the package from CTAN.

OK. Thanks! 
-- 
Denis


From sergio.callegari at unibo.it  Sun Apr 30 15:10:54 2023
From: sergio.callegari at unibo.it (Sergio Callegari)
Date: Sun, 30 Apr 2023 15:10:54 +0200
Subject: Explanation about update to luatex received via tlmgr
Message-ID: <6b8b544c-1b14-35e0-f3f3-2733f4410673@unibo.it>

Hi,

Today I have received some updates to my texlive 2023 system via `tlmgr 
update --all`.

Among them, I got:

`[2/3, 00:02/00:17] update: luatex [1916k] (66784 -> 66967) ... done`

It is my understanding that the `66967` corresponds to an svn commit in 
the texlive trunk branch. So I went looking for it and I found:

`texlive[66967] trunk: LuaTeX 1.17.0` 
(https://tug.org/pipermail/tex-live-commits/2023-April/025590.html).

However, I have not received luatex 1.17.0.? As a matter of fact, my 
luatex appears to be exactly the same it was before the update.

So, what was this update about? What am I missing about how to 
understand what goes into an updated package?

Incidentally, I would like to also ask why 
http://tug.org/svn/texlive?view=revision&revision=66967 is not browsable.

Thanks,

Sergio



From P.Taylor at Rhul.Ac.Uk  Sun Apr 30 15:30:42 2023
From: P.Taylor at Rhul.Ac.Uk (Philip Taylor (RHBNC))
Date: Sun, 30 Apr 2023 14:30:42 +0100
Subject: [EXT] Explanation about update to luatex received via tlmgr
In-Reply-To: <6b8b544c-1b14-35e0-f3f3-2733f4410673@unibo.it>
References: <6b8b544c-1b14-35e0-f3f3-2733f4410673@unibo.it>
Message-ID: <553a448a-25ca-ad2a-a814-6a568606969d@Rhul.Ac.Uk>

On 30/04/2023 14:10, Sergio Callegari wrote:
I would like to also ask why http://tug.org/svn/texlive?view=revision&revision=66967 is not browsable.

For me at least, nothing from https://tug.org/svn/ downwards is accessible.

--
Philip Taylor

This email, its contents and any attachments are intended solely for the addressee and may contain confidential information. In certain circumstances, it may also be subject to legal privilege. Any unauthorised use, disclosure, or copying is not permitted. If you have received this email in error, please notify us and immediately and permanently delete it. Any views or opinions expressed in personal emails are solely those of the author and do not necessarily represent those of Royal Holloway, University of London. It is your responsibility to ensure that this email and any attachments are virus free.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://tug.org/pipermail/tex-live/attachments/20230430/086c2af0/attachment.html>

From sergio.callegari at unibo.it  Sun Apr 30 15:56:01 2023
From: sergio.callegari at unibo.it (Sergio Callegari)
Date: Sun, 30 Apr 2023 15:56:01 +0200
Subject: [EXT] Explanation about update to luatex received via tlmgr
In-Reply-To: <553a448a-25ca-ad2a-a814-6a568606969d@Rhul.Ac.Uk>
References: <6b8b544c-1b14-35e0-f3f3-2733f4410673@unibo.it>
 <553a448a-25ca-ad2a-a814-6a568606969d@Rhul.Ac.Uk>
Message-ID: <7c8cfb8f-440c-1a9a-8694-87a8ceb27168@unibo.it>

On 30/04/2023 15:30, Philip Taylor (RHBNC) wrote:
> On 30/04/2023 14:10, Sergio Callegari wrote:
>> I would like to also ask why 
>> http://tug.org/svn/texlive?view=revision&revision=66967 is not 
>> browsable. 
>
> For me at least, nothing from https://tug.org/svn/ downwards is 
> accessible.
>
Indeed, however on https://www.tug.org/texlive/svn/ you can read:

For anonymous read access

For public (read-only, anonymous) access, there are a variety of methods:

You can browse the repository via the web; (pointing to 
https://www.tug.org/svn/texlive/trunk/)

Best,

Sergio


From mojca.miklavec.lists at gmail.com  Sun Apr 30 19:55:34 2023
From: mojca.miklavec.lists at gmail.com (Mojca Miklavec)
Date: Sun, 30 Apr 2023 19:55:34 +0200
Subject: [EXT] Explanation about update to luatex received via tlmgr
In-Reply-To: <7c8cfb8f-440c-1a9a-8694-87a8ceb27168@unibo.it>
References: <6b8b544c-1b14-35e0-f3f3-2733f4410673@unibo.it>
 <553a448a-25ca-ad2a-a814-6a568606969d@Rhul.Ac.Uk>
 <7c8cfb8f-440c-1a9a-8694-87a8ceb27168@unibo.it>
Message-ID: <CALBOmsYJgoaCPt9x9wq7c25ks_r7sZpSjW6bQZ3cZFVrNW-NPg@mail.gmail.com>

On Sun, 30 Apr 2023 at 16:01, Sergio Callegari wrote:
> On 30/04/2023 15:30, Philip Taylor (RHBNC) wrote:
> > On 30/04/2023 14:10, Sergio Callegari wrote:
> >> I would like to also ask why
> >> http://tug.org/svn/texlive?view=revision&revision=66967 is not
> >> browsable.
> >
> > For me at least, nothing from https://tug.org/svn/ downwards is
> > accessible.
> >
> Indeed, however on https://www.tug.org/texlive/svn/ you can read:
>
> For anonymous read access
>
> For public (read-only, anonymous) access, there are a variety of methods:
>
> You can browse the repository via the web; (pointing to
> https://www.tug.org/svn/texlive/trunk/)

That one doesn't seem to work for me either.

The regular subversion cloning / svn log etc. seems to be unaffected,
and of course the git mirror does a good enough job:
    https://git.texlive.info/texlive/
to see what has changed.

I'm just somewhat confused that the documentation for LuaTeX has now
been updated (with files like
Master/texmf-dist/doc/luatex/base/luatex-lua.tex, which also triggered
a "package update" with "fake version update"), but the binary itself
wasn't changed?

Mojca

From karl at freefriends.org  Sun Apr 30 22:19:32 2023
From: karl at freefriends.org (Karl Berry)
Date: Sun, 30 Apr 2023 14:19:32 -0600
Subject: [EXT] Explanation about update to luatex received via tlmgr
In-Reply-To: <CALBOmsYJgoaCPt9x9wq7c25ks_r7sZpSjW6bQZ3cZFVrNW-NPg@mail.gmail.com>
Message-ID: <202304302019.33UKJWEA026263@freefriends.org>

    https://www.tug.org/svn/texlive/trunk/

Online repo browsing should be back now.

    mm> the documentation for LuaTeX has now
    been updated (with files like
    Master/texmf-dist/doc/luatex/base/luatex-lua.tex, which also triggered
    a "package update" with "fake version update"), 
    
Right.

    but the binary itself wasn't changed?

Luigi commits doc updates when he commits source updates, independent of
binary updates.  --best, karl.