Building formats…

Welcome to TeX Live!


See /usr/local/texlive/2019/index.html for links to documentation.
The TeX Live web site (tug.org/texlive) contains any updates and
corrections. TeX Live is a joint project of the TeX user groups around the
world; please consider supporting it by joining the group best for you. The
list of groups is available on the web at tug.org/usergroups.html.

Tomorrow the fun starts!

Hello everyone!!

Suppose we have a diagram in a (32.9 x 48.3)cm surface (not the a4 paper). Is it possible to put that big diagram into a a4 paper using LaTeX?

@manooooh Yes.

but without deforming the size of the figures or arrows or text; reduce everything automatically

@manooooh Yes,

@PauloCereda oh, do we need a package for that?

And please, no bold, if possible.

@manooooh It depends. You were vague in your answer, I am being vague in my reply. :)

You can scale a tikzpicture with scale and transform shape, and there's also \scalebox.

@PauloCereda my idea is to create a diagram using LaTeX, but in a4 format the diagram is very tight. For that, my idea is to create it in a bigger size so I have more space. However, if you want to print the sheet with a standard printer, it does not support very large sheet sizes, so you would have to reduce the whole diagram. Is this a good idea?

Originally the diagram was designed for an a4 format, but because of the number of arrows and so on it is very complicated to create it in a4 because the font size should be very small, and the idea is to have the printed diagram

@PauloCereda oh, using scale is a good idea, thanks!

@manooooh suit yourself, it's really up to you. But if the diagram is not reasonably visible with A4, writing it in an arbitrary size and reducing it back won't make it clearer...

If you had to choose thinking that it is better to have printed the document, what would you prefer to create a diagram, not very understandable but that fits the sheet size, or an understandable diagram but with the sheet size greater than desired?

I think it must be a recurring dilemma

@UlrikeFischer what was the package and command to make a box with the possibilty of adding text inside it?

Hm, is a nice palindrome of number of comment worth to be shown here?

@PauloCereda @manooooh We marmots are considering to sue the manufacturers of tablets because their products do not respond to claws. Do ducks encounter similar problems?

@PhelypeOleinik Please let me know if tex.stackexchange.com/q/477206/121799 gets closed as a duplicate, I will be happy to reopen it right away. (Anyone tempted to close it please read the questions before.)

@marmot the Argentine ducks have not experienced that kind of problem, but we are sure that if it happened to us we would also sue the companies that created the tablets. So we accompany you in your demand

@manooooh Very good! (Thanks for the bounty ... ;-)

@marmot no problem! We will be accompanying you in judgment; we will be the people who appear behind the accused and the plaintiff

Hahaha

@manooooh TeX?

@UlrikeFischer I do not think so. You suggested to a user a package (or document class I do not remember) where you can add text to a text box in the PDF document output

@manooooh your question was so vage that I couldn't resist. I still don't know what you have in mind - boxes are everywhere-, perhaps pdfcomment.

@UlrikeFischer no, I should find the comment in the chat. My apologies

@UlrikeFischer spot anything? ^^^^

@DavidCarlisle :)

@egreg, @DavidCarlisle, @UlrikeFischer Which server did you use? I left it going last night, but after 4 hours my PC went onto standby ...

I found it!!

@UlrikeFischer ^^^^

@JosephWright I left it installing a gazillian files last night and this morning it doesn't seem quite right...

The"TextField" is also called "TextBox" by another programs

@JosephWright depend opt_location:ftp.cstug.cz/pub/tex/local/tlpretest

@JosephWright ftp.cstug.cz/pub/tex/local/tlpretest, but I remember from last year, that the download times varied a lot.

@DavidCarlisle looks okay here: This is TeX, Version 3.14159265 (TeX Live 2019/W32TeX).

@UlrikeFischer it may be just that the cygwin binaries not built yet, but I wish it had mentioned that before starting the download:-) I just asked on tl list....

@UlrikeFischer Cool

@JosephWright I had a bit problems at the end: after saying that it does some package postprocessing the tlmgr windows hanged and cmd-window flickered around. At the end I killed the perl process. But it seems to work fine. I will check later this day, what happens on the other PC.

@JosephWright and now like every year I need to remember how the gpg stuff needs to be done ;-(.

@user0 Shall we continue our discussion here? Please see the attached links in my previous comment. You could see why I made that remark.

Also, I am not accusing you :)

There is nothing to accuse you there. It is just a major constructive remark.

There is nothing to accuse you there. It is just a major constructive remark.

@Raaja There is no discussion. I made my point. You might have not wished to hurt me, but you did.

@user0 I no way in inteded to hurt you. But the fact is you made the same edits -- that possess no new information -- to the same question, in a matter of few days. Hence, to improve the community's user-experience, I mentioned it to you. And, I dont see any mistake in doing so. After all, we are all here to have fun, learn and help others right. It also involves constructive feedbacks, I believe.

@user0 More importantly, that is how this community is buit-up and thats the best part about this community.

@user0 Therefore I kindly propose to you to look into these two edits that you made: tex.stackexchange.com/review/reopen/166473 here and tex.stackexchange.com/review/reopen/165953 here.

@Raaja The latest edits are there to make the text grammatically (produces->produced) or typographically ({xe}|{lua}latex -> {xe|lua}latex) better or clearer (100%). I am not an ideal person and it was, perhaps, my fault for not getting the OP right at once. If you have questions concerning other edits, please feel free to ask.

@Raaja If you don't like the question to be reopened, you can vote for closing, and you have a full right to do so. I would object technically, but not procedurally: it's up to the community (in particular, to you) to vote for closing/reopening.

@Raaja What I don't think is the right thing to do is to accuse me of editing for the pupose of reopening. You might have asked first before writing this.

@Raaja just for the purpose of reopening, better to say.

@UlrikeFischer This is TeX, Version 3.14159265 (TeX Live 2019) (preloaded format=tex)

@UlrikeFischer @JosephWright And a bunch of updates on the first day!

@egreg yes got them too.

@JosephWright cstug; less than 45 minutes

@egreg Lucky you!

@egreg Same server here is looking like three hours

I wonder whether some people around to here would be able to say something to this post on Meta MathOverflow: MO links chopped in gmail pdf viewer.

@JosephWright That's an effect of Brexit. ;-)

I'd guess that Meta.MO is not really the right site where to ask - if this is really a problem, it is either bug in gmail pdf viewer or a bug in how pdflatex creates links. But I thought it might be worth mentioning it here - since this people in this room are likely to be experts on this and might be able to say where the problem actually is. (Or at least make a reasonable recommendation where to ask instead, for example, whether this could be changed into a suitable question on TeX.SE.)

@MartinSleziak no idea with just that information, if you made a real example and posted as a question here someone could look into the generated pdf and se what the link annotation says.

Would you (or somebody else around here) be willing to post a comment on that question what the OP should do if he wants to make it into TeX.SE question? Although I have made a few posts on this site, I do not have really enough experience to give instructions how a suitable question on TeX.SE look like.

Or should I simply suggest TeX.SE and add link to this chat with a brief comment that the question would have to contain some kind of minimal example and the pdf which causes problem would have to be uploaded somewhere online?

Would anybody resist to remark that “erronious is erroneous”? :-D

@MartinSleziak that can't work. If you put simply URL: https://mathoverflow.net/q/313198 in a document, then it depends on the pdf viewer how the link is interpreted. The OP needs something like URL: \url{https://mathoverflow.net/q/313198} or should use biblatex and the url field. (and the document should load hyperref.)

Time used for installing the packages: 173:53

@DavidCarlisle, @egreg, @UlrikeFischer Right, TL'19 installed, I'll start checking over all of the test suite. Plan to create a TL19 branch for test file changes so we can acutally track it (now we've gone all modern and Git-based)

@JosephWright since we can no longer update for tl2018 anyway shouldn't we just move the master branch tests to tl2019 (and get travis to pick that up)

@DavidCarlisle Er, I'd rather not: it means having to mess with the build script at least twice a year to 'swap'

@UlrikeFischer I suspected that but couldn't remember if amsrefs was magically linking URL: notes

@DavidCarlisle So far, the only changes are expected ones due to new primitives

@DavidCarlisle I didn't think about this possibility. But if the link break then probably because it is not there ;-)

@JosephWright yes sure but that's always a pain as you can't do l3build ctan if that fails

@DavidCarlisle But we (probably) shouldn't be: releases are meant to be against release engines ...

@JosephWright isn't @UlrikeFischer going to break every test output that includes a dash?

@JosephWright I thought there was a plan to get some updates in before tl2019 released?

tools has a dependency on base so won't be buildable in master with a 2019 engine

@DavidCarlisle don't remember me of dashes ;-( -- just trying to convince Hans to apply Marcel's patch.

@UlrikeFischer So does it mean that the bibtex entry generated by the cite button on Stack Exchange sites in incorrect? Should it have something like eprint={\url{https://mathoverflow.net/q/313198}} instead of eprint={https://mathoverflow.net/q/313198}?

@DavidCarlisle Sure, but what I meant is that the documentation PDFs should be build (I think) with release versions of pdfTeX, third-party packages, etc.

@DavidCarlisle I wonder if we can 'auto-detect' that TL pretesting is ongoing, and have the script switch

@MartinSleziak depends on if you use biblatex or not. With biblatex the first is wrong and the second okay, without biblatex it depends on the bib-style.

So what is actually the part which is incorrect in the OP's example? Is it this part: note={URL: https://mathoverflow.net/q/313198 (version: 2018-10-19)}?

@DavidCarlisle oh, Marcel found a very nice example of the "ligature" bug. Guess the output of

\documentclass[11pt]{article}

\begin{document}
  ff\discretionary{l}{}{l}
\end{document}

@marmot Sure thing :)

@marmot Here's the (almost) complete picture ^^ Just some adjustments left :)

@MartinSleziak yes there is nothing there to make a hyperlink, it is just text.

@DavidCarlisle So note={URL: \url{https://mathoverflow.net/q/313198} (version: 2018-10-19)} would be better?

@MartinSleziak without context impossible to say, that will cause latex to generate latex using \url which is undefined by default but if it is used with a document using hyperref then it will make a link, so having \url is probably what you want but if you are writing a generic tool to make bibliographies that work with any latex document not using hyperref might be what you have to do, and as Ulrike says if the bibliography is being used with biblatex the details are different again.

@MartinSleziak That is why we always complain about fragments of tex code:-) The details always depend on the full context...

Thanks for the response! I will mention what I was told here to the poster of that question - so the rest is up to him.

@DavidCarlisle See TeX Live list: found something!

@JosephWright nothing from you yet...

@DavidCarlisle On its way

@JosephWright they probably messed up the command table adding your \expanded patch, they'll probably blame you, I would:-)

@DavidCarlisle Was my feeling too, but it's not an integer setting: I wonder about the other new primitive they've added

@DavidCarlisle \readpapersizespecial ?

@JosephWright 完全に文書化されていますか？

@DavidCarlisle Sounds about right ;)

@JosephWright what should that do?

@UlrikeFischer see my comment!

@UlrikeFischer Something about paper sizes I guess :)

@DavidCarlisle 彼はそれが何をするのか言っていない！

@UlrikeFischer I assume it is like the support of \pdfpageheight in xetex that gives some access to the values on the tex side even though really they are passed to xdvipdfmx via a \special

@UlrikeFischer we don't read documentation, that's your job:-)

@JosephWright good call, it seems,

Hm I get \readpapersizespecial=undefined. when I run uptex \show\readpapersizespecial. Is it really a new primitive?

 This is e-upTeX, Version 3.14159265-p3.8.2-u1.24-190131-2.6 (utf8.uptex) (TeX Live 2019/W32TeX) (preloaded format=uplatex)
 restricted \write18 enabled.
entering extended mode
(./test.tex
pLaTeX2e <2018-12-01u02> (based on LaTeX2e <2018-12-01>)
(c:/texlive/2019/texmf-dist/tex/latex/base/article.cls
Document Class: article 2018/09/03 v1.4i Standard LaTeX document class
(c:/texlive/2019/texmf-dist/tex/latex/base/size10.clo)) (./test.aux)
> \readpapersizespecial=\readpapersizespecial.
l.3 \show\readpapersizespecial

@UlrikeFischer @^^^

@JosephWright ah, extended version ;-)

@UlrikeFischer Yes: pTeX/upTeX not really useful ...

@DavidCarlisle, @UlrikeFischer No issues with LuaTeX yet ... nearly finished with LaTeX3, moving to LaTeX2e

@PhelypeOleinik Looks nice!

@Sigur I always thought you need to use \tikzcdmatrixname

@marmot :)

o/ hi everybody

Is there any "canonical" post regarding (simple) tweaks to citation styles in biblatex?

Concretely, what I want to do is to change the minimum number of authors for which the "alphabetic" style starts using the [Bla+99] style from 4 to 5

@marmot I guess, you think I said something stupid in tex.meta.stackexchange.com/a/8125/165772 (I don't wish to rule out the possibility of this fact, since I was pretty tired when answering, but yes, I read the question.) So, I'm open for criticism.

@marmot, It worked. Thanks a lot.

Omg. What is going wrong? I can count up to 7 (!) just-do-it-for-me questions today (on a single day). It is the first time I can see such a large number of questions.

@JouleV It's Friday, and certain folks might dream of going home early :-).

@JouleV You can always vote things down if you don't think they're good questions!

@Danu Indeed that is what I was doing some hours ago. But now if I continue doing that, I will downvote more than upvote today, and I don't like that.

What happens with the user that give 500 rep to questions that were answered?

Because of Friday? :P

@DavidCarlisle On the TeX Live business, we could use an env variable to point to the server

@JosephWright yes I guess so, so you'd just need to fiddle with travis.yml to switch?

@DavidCarlisle Exactly: probably set up so we have to actively choose the pre-test

@JosephWright somewhere I set up somthing to extend the [ci skip] check in log thing so you could go [pretest] or some such (just worked by some basic shell script lookup of the commit message then setting the environment. I recall setting it up, not sure if I ever used it though..

@JosephWright found it:-)

case "$TRAVIS_COMMIT_MESSAGE" in
\[ci\ final\]*)
    DRAFT=no
    TARGET=om20-2017-12-01
esac

@DavidCarlisle Would we want it in a commit message? Isn't it more a 'state setting'?

@DavidCarlisle Key thing for me at present is we actually have basically no engine issues this year

@JosephWright possibly, I couldn't actually recall what I had when I started typing the above, there travis is building a document and normally it includes editors notes and other stuff but of you commit it with [ci final] it builds a version without such notes

@JosephWright did the --- thing not show up in any of our existing tests?

@DavidCarlisle No, or at least it probably is simply silently accepted in a LuaTeX-specific .tlg

@JosephWright I'll run it at the weekend I suppose

Uff I tagged my first link object ;-)

@UlrikeFischer the pdf reader crashed as a consequence?

@UlrikeFischer show off:-)

@DavidCarlisle ^^^ But it took me days starring on the reference until I figure out where one has to put all the numbers.

@UlrikeFischer reading manuals, again

@DavidCarlisle I realize I should better have asked you what was wrong with my pdf ;-)

I miss this with git commit ids, you don't get the same sense of progress as svn Committed revision 184814

Is there a canonical place to put sty files inside a directory heirarchy? Say for example you have a VC repository, and you want those sty files to be part of the repository. But you also don't want to duplicate them.

@DavidCarlisle Mercurial has sequential numbering.

I wonder if a texmf directory at the root (top level) of the repository would be reasonable. Thoughts?

Would using soft links to a sty file still work?

I found

Is it ok to use this?

@FaheemMitha Normally one would set TEXINPUTS as required

@JosephWright The environment variable?

@FaheemMitha Yup

@JosephWright So pointing to those files location in a checkout of that repository?

@FaheemMitha I using such small texmf trees a lot, but your question is much too vage to say if it is sensible in your case.

I mean, the texmf directory within the repository.

@FaheemMitha It's what I'd do (indeed what I do for beamer)

@UlrikeFischer Say I have a repository foo. I want a central place, say foo/texmf to keep style files etc.

But TeX files elsewhere in foo need to know to look in foo/texmf.

So the question is what a good way to manage that is.

Joseph recommends assigning to TEXINPUTS.

Not sure how TEXMFAUXTREES fits in there.

@FaheemMitha well as I said, I'm using this a lot. Quite a number of my repositories have also a texmf tree, and I'm activating/deactiving them with scripts.

I could make this a question if it would help. The important thing is that the sty files (and cls files), should live in the same repository as the tex files using them.

@UlrikeFischer so you don't use TEXINPUTS?

@FaheemMitha No.

@UlrikeFischer Ok.

I'm tempted to write a question asking what TEXMFAUXTREES is used for. Because I'm having difficulty finding documentation on it.

@FaheemMitha It depends in part on the detail of your set up; if you want to add/remove trees a lot, TEXMFAUXTREES is right. But if it's a 'main local' area, TEXINPUTS is the right choice

@JosephWright Still not sure how TEXMFAUXTREES works.

My main current repository has a standard location in my home directory. Hasn't changed in years.

I suppose TEXINPUTS doesn't care if it's in my home directory?

@FaheemMitha possibly, I don't remember but mercurial lost, git won, so the choices are svn or git, really.

@DavidCarlisle Not possibly. It's a fact.

@FaheemMitha yes I wasn't doubting you:-)

@FaheemMitha if you just have tex files .sty. .cls etc you may as well just use TEXINPUTS, but if you are @UlrikeFischer and have lua and fonts and tex files and stuff and you want local consistent copies of all of them, then it's simpler to set up a TDS structure and just have to switch in one tree rather than set TEXINPUTS and LUAINPUTS and TEXFONTS and ...

@FaheemMitha export TEXINPUTS=~//: would make tex find all files at any level below your home directory before looking in the standard places (if you have millions of files there you might want something more focussed but...

@DavidCarlisle I was going with export TEXINPUTS=~/repos//:.

@FaheemMitha no . at the end, that would be bad:-)

But I still don't know what TEXMFAUXTREES is for.

@DavidCarlisle That's the full stop for the sentence. :-)

@FaheemMitha well if you have some private font in your home directory for example, setting TEXINPUTS would not do anything useful because tex doesn't look along TEXINPUTS for fonts (or bibtex files or makeindex, so if you need to set up private search paths for all thouse things texmfauxtrees is better than TEXINPUTS

@FaheemMitha ah.

@FaheemMitha by default texlive has ~/texmf in its paths so if you put stuff in there instead of in ~/repos you would not have to do anything

@DavidCarlisle Or ~/Library/texmf on a Mac.

@HaraldHanche-Olsen but everyone uses windows:-)

@DavidCarlisle Sure, but I need those files to live in the repos.

@DavidCarlisle So it's more general?

@DavidCarlisle I know, that's why I am so amazed that someone maintains MacTeX just for me, the only Mac user in the world!

@FaheemMitha the default TEXINPUTS is like TEXMFHOME/tex// the default BIBINPUTS is TEXMFHOME/bibtex/bib// etc so dding a TEXMFAUXTREE essentially just adds a top levl TDS directory to TEXMFHOME so adds to all the search paths which is convenient as long as your local files are in a TDS structure

@HaraldHanche-Olsen I sometimes think that (possibly with more justification) about cygwin texlive

@DavidCarlisle Where is this documented? I could not find anything.

@FaheemMitha oh, somewhere the auxtrees thing is new it was discussed on tl list but it's only really setting some enviornment settings that have always been there since before texlive

@DavidCarlisle Then someone should ask a question about it.

@FaheemMitha tug.org/texlive/doc/tlmgr.html#conf

@UlrikeFischer Not the most understandable documentation.

@FaheemMitha get egreg to do it

@FaheemMitha Not more docu needed - they added the option for me and I knew what it should be doing, I was pestering them for year to add it.

@UlrikeFischer ooh a technique

@DavidCarlisle I don't think he knows where the question button is.

@UlrikeFischer Ah, the In Crowd.

Are there any gotchas I'm unaware of with relative paths for \includepdf? Because it's not working for me.

@UlrikeFischer -- If the doc isn't clear, I may be able to do something about it. This year, i will have to install tex live for myself (it will be on a linux system). So if I can't understand it, or think it is confusing, I will volunteer to clarify it. (At the very least, I can make sure everything is spelled correctly, and all the commas are in the right places.)

@barbarabeeton Commas are important.

@FaheemMitha -- Copyeditors probably love you.

@barbarabeeton I don't think I know any.

@PauloCereda David doesn't believe in Oxford commas.

Even though he lives in Oxfordshire.

@FaheemMitha He's from Manchester, understandable.

@PauloCereda Is he?

@FaheemMitha I have partial knowledge of a lot of things, so probably. :)

I am 100% right 50% of the time.

@AlanMunn perfect timing. :)

@barbarabeeton colour

@PauloCereda not originally

@DavidCarlisle oh

@FaheemMitha ;-) But it is really not complicated. You already have quite a lot texmf trees (texmf-dist, texmf-var, texmf-local, texmf-config, the texmf in your home directory,) and the option allows you add more.

@DavidCarlisle -- I'm willing to bend on that spelling when it's used consistently, except if it's the name of a package or the defined name os a command.

@DavidCarlisle -- Admittedly, i'm more skeptical about wikipedia articles if i happen to have firsthand knowledge. I've tried a few times to propose corrections, but since the "old guard" doesn't know or care who I might be, my proposals have always been rejected. So I've basically given up there; I'll stick to editing TUGboat, Unicode Tech Report 25, and similar items where my knowledge is appreciated.

@barbarabeeton do you have to propose corrections can't you just make them? or do you mean they were rolled back?

@DavidCarlisle -- If you're not "known", the edited text you submit is perused by individuals who have been admitted to the fraternity, and they decide. I have no idea who those individuals are, nor am I inclined to try to find out.

@barbarabeeton oh I forgot that (I'm not sure it used to be true) anyway it doesn't happen if you get an account and log in. Which isn't entirely unreasonable, edits from unregistered readers they really have no traceable information of who did the edit, just an IP address which is probably an internet proxy somewhere...

@barbarabeeton speaking of accounts, if you have a github one by now you might comment on this issue if you have any thoughts:

@barbarabeeton github.com/mathml-refresh/mathml/issues/61

@DavidCarlisle I don't think I've had any of my edits to wikipedia pages rejected, but my contributions tend to erode over time, as new edits occur. Some times for the better, other times for the worse.

@DavidCarlisle -- I think I did register, but in any event, whatever I did has been lost because all the records were stored in my AMS account, now terminated with uncertain prospects that the backup I took can be reconstructed.

@DavidCarlisle -- I'm more inclined toward having a github account now than I was at AMS; I'll see if I may already have registered, and if not, do so.

@barbarabeeton But all you'll need is your wikipedia username and password, and your contributions, even if rejected, should still be there in the history?

I did upload several portraits of Abel Prize winners to wikipedia, though, and at some point they all got removed because of copyright issues. All were my own, but it appears I would have to sign a bunch of papers and wait half a year for someone to accept them. That's a different issue, though, but it bugs me.

@barbarabeeton Speaking of accounts, I have noticed that your username here is still all lower case.

@DavidCarlisle -- I did register with github, and I'm trying to log in, but it won't accept the password I recorded, and it claims it can't find the email address I give it to send me a replacment, even though when I type in the ID, it feeds back both the AMS and the TUG addresses. Maybe it's the fact that I'm using an "outdated" browser. Until I get a stable new system set up here, I'm not going to be very useful.

@barbarabeeton no rush, but we may push Murray to see if we can get something done about script/calligraphic and I'll keep you informed even if you are not logged in to gh.

@DavidCarlisle -- I do have something to say on that issue. I provided to Murray some published examples where the same letters in both calligraphic and script were used for contrasting purposes, and I've asked him several times to please submit these to the UTC. I am pretty sure that they will be encoded using the variation selector, but that is something I have no control over. Both calligraphic and script are needed. I'll nag him again. Sigh.

@HaraldHanche-Olsen -- Yes, and I wish it to remain that way. (There are other reasons for that, but it's a rather long story. If we meet someday, face to face, ask me and I'll be happy to tall you.)

It's clear the underscore package is breaking this, but I'm not clear exactly why.

gist.github.com/b6c529e6128c7d03ceaf7f9698adba72

Is it worth asking as a question?

@FaheemMitha well underscore change underscores, and your file name has an underscore.

@UlrikeFischer What does it do to underscores, exactly? And the error message is really rather uninformative. It just says that a file of exactly the right name, is not found.

@FaheemMitha {\detokenize{blank_blank.pdf}} should work.

@UlrikeFischer Hmm. Not sure what that does, but ok.

Worth asking as a question?

Would it break any filename reference?

@FaheemMitha well you should at first read the documentation. It says on page one You must avoid “_” in file names.

@UlrikeFischer Yes, I see it. But suppose you need to?

@FaheemMitha as I said: \detokenize should work. But I would simply not use such packages.

@UlrikeFischer Not a fan of underscore?

@FaheemMitha not a fan of making to many ascii chars active.

@UlrikeFischer Ok. I don't actually know what underscore does, so...

@FaheemMitha read the docu. It has only three pages.

Does anyone know something about the "supposed" command \operartorfont?

@UlrikeFischer Can you show a minimal example?

@manooooh Never heard about it. You mean \operator@font, maybe.

@egreg oh, I do not know, a mathematician told me about that command. Can \operator@font be used to write for example "d" in upright or not, or change the font of a math symbol?

@egreg An arbitrary document that loads regression-test and hyperref should do it.

@egreg I don't think that will be very difficult to repair, it is only a nuisance to have to resave all the test. - and all the luatex tests too as luatex changed the spaces in the pdf.

@manooooh That's an internal command. If you don't know about it, forget it.

@egreg ok. I am going to ask him what he meant by that, since he told me that the correct way to write the vertical "d" is by using the command \operartorfont