TeX, LaTeX and Friends

@MarcelKrüger did you see tex.stackexchange.com/questions/511230/… (i get the same, but the example is a bit long so I'm not sure if it is user error or not...) on the second pass I get

ABD: EveryShipout initializing macros...xmf-dist/tex/luatex/luaotfload/fontload
er-basics-gen.lua:148: attempt to call a nil value (field 'suffix').
<to be read again>
relax
l.126 \section*{Editor Comment}

?
! Font \TU/LatinModernRoman(0)/bx/n/10=LatinModernRoman/B:mode=node;script=latn
;language=DFLT;+tlig;-liga; at 10pt not loadable: metric data not found or bad.

<to be read again>
relax
l.126 \section*{Editor Comment}

?

We could blame @UlrikeFischer as she is not here.

Marcel Krüger

12:39 PM

@DavidCarlisle It is a user error, but we might still want to make luaotfload more resistant: You shall not overwrite the global file... I'll write an answer.

daleif

@DavidCarlisle David, shouldn't the graphics bundle test for LaTeX version and complain if incompatible? ref question om tl list. Just to catch those who forget to update everything.

Ulrike Fischer

@DavidCarlisle Don't rely on it - in the modern world you can watch even from Syracuse.

David Carlisle

@daleif possibly but graphics is part of "required" so it is the core release, we can't make every file check the version of every other. Of course by the time it is filtered through ctan and the tex distribution graphics being part of required is lost and it looks like any other contrib package so these things happen....

@daleif actually it is _already checking \NeedsTeXFormat{LaTeX2e}[1995/12/01] which is now pointless, I should update that date....

Marcel Krüger

@UlrikeFischer We can blame you anyway: The faulty code comes from your old answer :P

David Carlisle

12:57 PM

@MarcelKrüger thanks

Ulrike Fischer

@MarcelKrüger et tu Brute?

David Carlisle

@UlrikeFischer He was just following team policy

@UlrikeFischer I guess I can't put off updating oberdiek monster for long:-) just ran test suite on the updated format:

./build/test/alphalph-test2.pdftex.diff
./build/test/alphalph-test3.pdftex.diff
./build/test/attachfile2.luatex.pdf.diff
./build/test/attachfile2.xetex.pdf.diff
./build/test/bigintcalc-test2.luatex.diff
./build/test/bigintcalc-test2.pdftex.diff
./build/test/bigintcalc-test2.xetex.diff
./build/test/catchfile-test2.luatex.diff
./build/test/catchfile-test2.pdftex.diff
./build/test/catchfile-test2.xetex.diff
./build/test/hologo-test-spacefactor.luatex.diff
./build/test/hologo-test-spacefactor.pdftex.diff

Paulo Cereda

1:57 PM

@PhelypeOleinik since you were the first to reply to the TL mailing list, you are the one to be blamed.

David Carlisle

@PauloCereda important point is It's not my fault

Phelype Oleinik

@PauloCereda I don't like your reasoning :-)

Paulo Cereda

@DavidCarlisle ooh

@PhelypeOleinik :)

@PhelypeOleinik From a Futurama quote: "When you do things right, people won't be sure you've done anything at all."

Phelype Oleinik

@PauloCereda Futurama is genius! I hope it gets re-un-cancelled some day :-)

Paulo Cereda

@PhelypeOleinik it is, one of my favourite series of all times!

barbara beeton

2:10 PM

@egreg -- The item in the photograph looks like a branding iron. The function of such an object is to be distinctly different and recognizable, or, more importantly, to be able to render an image that is distinct and quickly recognizable.

@egreg -- On another topic, are you preparing another "cherry-picking" entry for the upcoming TUGboat issue?

@CarLaTeX -- See my comment to @egreg, above.

StrongBad

2:26 PM

Some of you may be aware that there has been a [mini-revolt going on across the network](https://meta.stackexchange.com/questions/333965/firing-mods-and-forced-relicensing-is-stack-exchange-still-interested-in-cooper). There is a [timeline of events](https://meta.stackexchange.com/questions/334399/summing-up-the-main-issues-the-story-so-far) but it is a deep rabbit hole.

Basically, it was sparked by the firing of a moderator, but I personally think it goes deeper than that and is related to the disconnect between SE and users that was seen during the site redesign fiasco. There is an [open…

egreg

@barbarabeeton Not sure I'm able to.

barbara beeton

@egreg -- Okay, thanks for the ack. A possibility for next issue, maybe?

Alan Munn

2:41 PM

@DavidCarlisle You seem to be a good programmer. Could you please have a look at this and tell me why the macro jednotky{22} prints the correct value BUT when given in \hg\hgunits an error occurs? — user2925716 2 hours ago

@DavidCarlisle You seem to be a good programmer. :)

David Carlisle

@AlanMunn I was offended by the seem

@AlanMunn almost as good a programmer as I am a linguist (applying very similar techniques it has to be said)

Alan Munn

3:03 PM

@DavidCarlisle So you have a transcompiler to TeX? That would be a feat!

David Carlisle

@AlanMunn no but I have infallible oracles to consult for each discipline.

Paulo Cereda

@DavidCarlisle latex.ltx on an emacs buffer

@AlanMunn ooh

David Carlisle

3:18 PM

@PauloCereda well obviously that as well

Ulrike Fischer

3:44 PM

@PauloCereda In Siracusa they care about ducks!!

Paulo Cereda

@UlrikeFischer ooh

@UlrikeFischer saw your keynote in the GuIT meeting. :)

Ulrike Fischer

@PauloCereda ? Is there more than the title somewhere already?

Paulo Cereda

@UlrikeFischer only the title.

Ulrike Fischer

@DavidCarlisle buh. Why does it break?

SBM

Hello

Alan Munn

4:01 PM

@SBM Hi

SBM

4:16 PM

Hope everyone has a happy time

David Carlisle

4:33 PM

@UlrikeFischer no idea yet, mostly because the packages poke into low level latex definitions that ~~Frank~~ we changed.

@UlrikeFischer or it could just be a l3build normalisation issue, I haven't had time to look at the diffs at all,

Ulrike Fischer

@DavidCarlisle I'm just checking. A few are harmless (someone changed a message from "Try loading font" to "Try to load font", some are due to hyperref changes, in one case a version change in listings which is not suppress. All the askinclude probably because of the filecontents change but need some checking.

@DavidCarlisle but I don't get a failure for ./build/test/attachfile2.luatex.pdf.diff and ./build/test/attachfile2.xetex.pdf.diff so you should perhaps check these.

David Carlisle

@UlrikeFischer I wish I'd jumped in on that PR and said no:-)

meanwhile I guess we need a patch level 1

Ulrike Fischer

@DavidCarlisle you mean the "try to load"? It is rather harmless, the filecontents stuff will need more work.

Paulo Cereda

@DavidCarlisle ooh

Ulrike Fischer

@DavidCarlisle for the graphics stuff? looks so. probably we need also a few more testfiles for curious files names ;-).

David Carlisle

4:49 PM

@UlrikeFischer \extracolsep

@UlrikeFischer changing the messages is looks harmless but breaks any scripts or test files (or printed documentation) that was assuming the old message

Ulrike Fischer

@DavidCarlisle a yes, that too. ;-) I will remake the harmless failures and push and then look at the other ones if there is real problem.

@DavidCarlisle Regarding documentation: the newest latex news isn't on the website

Faheem Mitha

5:12 PM

It seems that PyX is affiliated with TeX in some way. The project page says:

> The authors kindly acknowledge financial support from the German Speaking TeX User Group DANTE e.V.

I thought it was just a drawing package. Python, if you're not familiar with it.

David Carlisle

5:27 PM

@UlrikeFischer aren't you on holiday?

Ulrike Fischer

5:40 PM

@DavidCarlisle ;-). This evening I have some time - hearing the sea below our balcony ...

David Carlisle

@UlrikeFischer I think I want to start with @Skillmon {a.b.c}.png issue and then work up from there to grffile before looking at the other failures

Ulrike Fischer

@DavidCarlisle the first askinclude diff got reduced to "Overwriting file" <->"Writing or overwriting file" after I changed the filecontents commands ...

Skillmon

6:16 PM

@DavidCarlisle to be fair, it's not my bug, that's your fault!

Ulrike Fischer

@Skillmon then it is not a bug but a feature ;-)

Skillmon

6:37 PM

@UlrikeFischer yes, no, maybe, I don't know...

Faheem Mitha

7:18 PM

@UlrikeFischer I think that's Microsoft's line.

Ulrike Fischer

8:11 PM

@DavidCarlisle can we blame Frank?

David Carlisle

@UlrikeFischer also

1

A: The error occurred after the last update

If you put \makeatletter \def\nobreak{\penalty\@M} \makeatother at the start of the file you will restore the behaviour, I've not traced enough to say yet where the correct fix should be.

Ulrike Fischer

@DavidCarlisle yes, that's what I meant.

David Carlisle

@UlrikeFischer ah I thought you meant for the grffile ones:-)

Ulrike Fischer

@DavidCarlisle ;-). But hyperref at least is not involved in this one.

Sebastiano

Good evening to all users into chat.

egreg

8:22 PM

@DavidCarlisle Might it be because \Writetofile does \let\protect\string?

Sebastiano

@egreg Hi, dearest Prof. I'm searching exactly you :-) I hope you're okay.

egreg

@Sebastiano Ciao!

@DavidCarlisle The error disappears if I add x before \Writetofile.

David Carlisle

@egreg I saw that in the \tracingall but got bored working back as to which package I was in (as I don't really know any of those or what they are trying to do)

Ulrike Fischer

@egreg the problem is the \@nobreaktrue from the chapter.

@DavidCarlisle it is only the answers package.

David Carlisle

@UlrikeFischer I don't know any answers:-)

egreg

8:25 PM

@DavidCarlisle We can blame @JosephWright

David Carlisle

@egreg so, get the package changed, or adjust the format again?

Ulrike Fischer

@DavidCarlisle smaller:

\documentclass[]{book}
\usepackage{answers}
\makeatletter
%\def\nobreak{\penalty10000}
\begin{document}
\Opensolutionfile{answer}[this-Answers]
\@nobreaktrue
\tracingmacros=1
\Writetofile{answer}{\@esphack}
\end{document}

Sebastiano

@egreg Sometimes come and visit me at math.stackexchange where there are some my questions. Please in English the word "maggiorazione" in the mathematical sense how does it translate? Excuse me for my intrusion for your discussion. math.stackexchange.com/users/705338/sebastiano

egreg

@Sebastiano “upper bound”

Sebastiano

@egreg Thank you very very much. I have written augment or ("increase") :-(

David Carlisle

8:32 PM

@egreg @JosephWright is going to say we should have used \protected\def instead of \protect

egreg

@DavidCarlisle I agree

David Carlisle

@egreg you agree with @JosephWright or you agree with my prediction? (or both:-)

Ulrike Fischer

@egreg Imho you are right, that is the problem.

egreg

@DavidCarlisle Both

Joseph Wright

@DavidCarlisle Yup

David Carlisle

8:40 PM

@egreg (@JosephWright) although arguably the package is wrong to use \protect=\string at that point if unknown latex-protected commands can appear there?

Joseph Wright

8:54 PM

@DavidCarlisle Er, I think this has come up before: It's not my fault

David Carlisle

@JosephWright when @egreg blamed you I assumed it was just on general principles, I didn't realise your name was in the file:-)

Ulrike Fischer

9:43 PM

@DavidCarlisle I will push in a few minutes the last test changes I made. There are a number of open "askinclude"-failures: The new compilation shows an error that I don't understand, but it is too late now for them.

David Carlisle

10:04 PM

@UlrikeFischer thanks a lot:-)

Alan Munn

10:40 PM

Fun with nonsense graphs: twitter.com/JamesCleverly/status/1181158343571918849/photo/1

David Carlisle

@AlanMunn if he'd done the graph with picture mode it would have been more authoritative

Alan Munn

@DavidCarlisle Undoubtedly.

Transcript for

$TeX - LaTeX$ TeX, LaTeX and Friends