« first day (3524 days earlier)      last day (1406 days later) » 

12:09 AM
Does anyone know what \BibitemShut {NoStop} does?
 
12:55 AM
@DavidCarlisle I spent all day on it, and after several hours, I got the 48 lines in the preamble of the .bbl, down to only 5 !! Finally, I posted the question here after so much hesitation about it not being a great question (I'm not very active on this site, so don't really know the culture):
0
Q: arXiv is not word-wrapping long bibliography entries the way Overleaf and my local installation do, for precisely the same MWE

user1271772After extensive discussion with Barbara Beeton in the chat room, and later with David Carlisle in the same place, I was recommended by both to write my question here, it has just taken some hours to strip everything down to as minimal of a working example as possible, and still get the problem re...

 
 
1 hour later…
2:04 AM
@user1271772 -- Your question is just fine -- it shows that you've done your research, and the example is compilable and error-free. That said, I've just tried it out and get yet a different result (with an up-to-date installation); the authors all fit on one line and the journal name on the next, no need for any breaks. What's different, I can't say, and have no way of checking. But you're not hallucinating. (My laptop is misbehaving; I can't post on the main site.)
 
2:19 AM
@barbarabeeton The result that you got, is the same as what I got on my laptop and in Overleaf (see the first screenshot in my question, all authors are in one line, just like in your case). It's arXiv that's giving something anomalous!
 
@user1271772 -- I believe there's a slight difference: in your image, the name of the last author is hyphenated -- "Fil-ippi" -- but the spacing is certainly much better than what arXiv is producing. I hope that someone out there can figure this out. It's horrible!
@user1271772 -- I apologize! I looked again at your question on the main site, and the result (the first image) is the same. I was looking at the images at the bottom.
 
@barbarabeeton Yea the images at the bottom are the ones I first showed you here in chat. Those are the ones I originally had when I submitted my paper. The ones at the top of my question were generated using the MWE (hence why your result was the same!). The MWE result is probably a bit different because I removed 43 lines from the .bbl file. That took some time: basically I commented one line at a time and re-compiled every time in between, until I got the 5 remaining lines that were absolutely
necessary to compile without errors :)
 
2:42 AM
@user1271772 -- I've honestly never seen a .bbl file with coding like that. What does the .bib file look like for that entry, and what \bibliographystyle is used? (I think it was there in your original question, but I can't find it now.) The only package I recognize that I know is related to a biblio is natbib, and I know natbib doesn't produce any .bbl code like that. I suspect the .bst file.
 
Let me try David's recommendation for how to get multi-line code in the chat (in order to paste the .bib entry):
    @article{Daday2012,
    abstract = {We employ the recently developed full configuration interaction quantum Monte Carlo (FCIQMC) method to compute the $\pi$ → $\pi$* vertical excitation energies of ethene and all-trans butadiene. These excitations have been the subject of extensive theoretical studies, and their location with respect to the corresponding absorption band maximum is the source of a long lingering debate. Here, we reliably estimate the vertical excitations of ethene and butadiene by performing FCIQMC calculations for spaces as large as 1018 and 1029 Slater determinants, re
This is just exported directly from the Journal of Chemical Theory and Computation, via Mendeley.
\bibliographystyle{apsrev4-1}
APS is not as good as AMS ;)
 
3:22 AM
@user1271772 -- Ignoring the "abstract" and "file" elements, this looks pretty routine. A little digging in the bibtex/bst area of the tex-live directory finds no apsrev4 files, but a couple of revtex4 files that are about double the size of the base and amscls offerings. The thing that surprises me is that they were written by Patrick Daly. They were generated from sources (.mbs files) about which I know nothing, whereas the "base" and "ams" .bst files were created "directly".
I'll do some more poking around, but tomorrow -- it's heading toward midnight, so time to tuck in.
 
@barbarabeeton 11:30pm here for me too :)
 
Somebody's in a bad mood tonight.
 
@AlanMunn -- Yes, I noticed that. (Since my laptop is being obnoxious, I can't post on the main site, or even look at comments that aren't already shown. So I can't see what other "damage" might have been done. But I can demonstrate that it wasn't me! Even if I felt like it!) How's life locked down in Michigan?
 
3:38 AM
Wasn't me either
 
@barbarabeeton You were definitely at the top of my suspects list. :) Life goes on. I go shopping once every three weeks and we go to the barn and that's about it. I'm not optimistic about the Fall semester. Already 14 people at a local bar popular with students tested positive. I can't imagine what it will be like when more students come back.
@barbarabeeton My mum just had her 90th birthday yesterday, but she's in lockdown in Toronto. So that was a bit sad. But she's doing fine, thankfully.
 
@AlanMunn I'm in Toronto (well, Waterloo.. close enough).
 
@user1271772 I went to Waterloo my first year of university. It's been a long time though... :)
 
@AlanMunn What about the rest of the years?!
Switched to McGill? What was the reason?
 
@AlanMunn -- Oy! But students are immortal, didn't you know? (At least they think they are.) Oh, happy birthday to your mum! I hope she stays safe and has lots to keep herself interested.
 
3:45 AM
@user1271772 I dropped out of CS/Pure math, moved to Montreal, worked for a few years as a programmer and then did linguistics at McGill. A linguist convinced me that human natural science problems were more interesting than computational problems. :)
 
I think we talked here a few months ago about Chomsky and Knuth
 
@barbarabeeton She has plenty to read especially after she raided the 7 boxes of books I shipped to her house in 1992 that I never retrieved. :)
 
@AlanMunn -- Gad! Comp Sci didn't even exist when I was an undergrad. The Appl Math department had a computer (it used biquinary arithmetic and programs had to take into account the rotational delay of the memory drum0
 
@user1271772 Ah so we did. You really ought to get a more memorable user name. :)
 
(cont'd) but the university president said "There are already three math departments. NO we can't have a fourth." (The third was History of math.) So I ended up in linguistics too, but largely because a pure math professor didn't like applied mathematicians, and liked female mathematicians even less. But it turned out well, I think. Now I get to play in the TeX sandbox.
 
3:56 AM
@barbarabeeton One of my former profs at McGill ended up as a linguist for similar reasons: the math dept at UPenn wouldn't accept women.
 
@AlanMunn -- The crazy thing is, before WWII, Brown awarded more PhDs to women in Appl Math than all but one other school in the U.S. (I think it might have been Chicago or Berkeley.) And if I had managed to graduate, I would have been the first ScB. (The department chair was terribly sad when I told him I was decamping.)
 
@barbarabeeton So things changed a lot after the war? Was this because of the GI bill?
 
@AlanMunn -- That's an interesting question. It might have had some effect, but Pembroke, the former women's college at Brown, wasn't really affected, and most of the "technical" departments (engineering, physics, geology) didn't discourage women, as far as I know. My guess is that women just didn't apply as readily, once the men came home.
 
4:12 AM
@barbarabeeton I see.
 
 
5 hours later…
cis
8:54 AM
Is there a font, which contains nice "stealth" arrows, known from TikZ?
 
9:33 AM
@UlrikeFischer @PauloCereda Obrigado pelo jantar assado
 
@DavidCarlisle ;-) the first arrived !
 
@DavidCarlisle oh no
@UlrikeFischer woo
 
9:49 AM
@PauloCereda but can we say it reached a safe haven?
 
@UlrikeFischer oopsie
 
10:03 AM
@UlrikeFischer Color model export now working: the branches should be ready for you to review properly. I'll start back on separations tomorrow.
@UlrikeFischer BTW, did you see the ICC question on the site?
 
@JosephWright is the master merged? I had to switch the branches yesterday again as lualatex-dev errored. Or is there a way to install only the color code for testing?
 
@UlrikeFischer I'll sort that
 
@PauloCereda Thanks you so much for the lovely duck! (chat.stackoverflow.com/transcript/message/49735365#49735365)
 
@JosephWright yes, but I hadn't yet the time to comment.
 
@UlrikeFischer I've merged master into color-export and color-multi-models: I'd recommend looking at the latter as it also covers color-export
@UlrikeFischer :)
@UlrikeFischer 'Come back next week once Joseph has sorted this'? ;)
 
10:09 AM
@samcarter_is_at_topanswers.xyz ooh more ducks
@JosephWright ooh
 
@PauloCereda And such a cute one nonetheless!
 
@JosephWright well it hasn't so much to do with colors. Mostly you have to add a stream and perhaps add values to the outputintents and similar. So it is more "come back when I have sorted the pdfresource stuff" (where I'm currently trying to sort it so that it can be loaded and then activated optionally - the main problem is where to put the bdc-commands: they need the full resource management currently, I don't see how to write variants without it.).
@PauloCereda we should start with chalk strokes to count ...
 
cis
That is the beginning of an implementation as font. Only the arrowheads are not as nice as at TikZ.
 
10:28 AM
@UlrikeFischer ooh we could have a counter
 
@PauloCereda will it have the macro \duckstepcount?
 
@samcarter_is_at_topanswers.xyz ooh a new macro
 
10:55 AM
@UlrikeFischer Seen the dvipdfmx email?
 
@JosephWright I just wanted to ask you the same ;-) I have to make lunch now, will answer after it.
 
@UlrikeFischer Cool
 
@JosephWright at first glance it looks right. It should be easy to retrieve the color obj reference.
 
11:18 AM
@UlrikeFischer Looks good to me too
 
@barbarabeeton It sounds as if other browsers experience similar problems as you: meta.stackoverflow.com/questions/398671/…
 
 
1 hour later…
12:49 PM
Here is a strange one with memoir vs biblaex vs \appendix vs hyperref
% \documentclass[a4paper]{book} % works
\documentclass[a4paper]{memoir} % the citation on page 3 points to
% page 1
\begin{filecontents}[overwrite]{\jobname-refs.bib}
@article{test,
author = "Test Testson",
title = "Test",
journal = "Some journal",
year = 2020,
}
\end{filecontents}
\begin{filecontents}[overwrite]{\jobname-a.tex}
\chapter{Test}
\cite{test}
\printbibliography[heading = subbibliography]
\end{filecontents}
\begin{filecontents}[overwrite]{\jobname-b.tex}
\chapter{Test}
\cite{test}
\printbibliography[heading = subbibliography]
If book is used instead of memoir then the \cite in filename-b.tex points to it self, in memoir it points to page 1. Why?
Hmm, I'd better post it as a proper question instead
 
1:02 PM
Observation: memoir added \anappendixtrue to \appendix this is enough to tilt stuff, but is the tilt in biblatex or hyperref?
 
@daleif I can look but only in few minutes, just finishing something else.
 
@UlrikeFischer thanks. For the project at hand I can probably remove \anappendixtrue from \appendix and it seems to work. Just wanted to know where on earth this issue comes from
 
1:28 PM
I've added it here
1
Q: biblatex vs hyperref vs \appendix vs memoir

daleifHere is an interesting one. In the example below, if the book class is used then the \cite in the second included document points to the bibliography generated for that page. But when memoir is loaded it points to the previous file. What gives? Update: It seems to be the definition of \appendix H...

 
1:59 PM
@daleif I added an answer.
 
2:22 PM
@UlrikeFischer thanks, I added an issue on the biblatex project with a suggested solution.
 
2:48 PM
@DavidCarlisle since you're the best linguist I know of, I have a question for you. In German there is a verb for calculating $\ln x$ called "Logarithmieren", what is the correct verb for the inverse $\mathrm{e}^x$ (or if you so want $e^x$)?
 
@Skillmonlikestopanswers.xyz "taking the log(arithm) of"
@Skillmonlikestopanswers.xyz English has this useful thing called spaces so we don't need a single word for each concept
 
@DavidCarlisle you misunderstood my question. I don't want to know how to say "taking the logarithm of", but I want to say "e to the power of" within a single word in German.
@DavidCarlisle that seems very inefficient to me
 
@Skillmonlikestopanswers.xyz oh sorry you want the German: exponentiationeren
 
@DavidCarlisle that is actually pretty close to what I came up with... :)
@DavidCarlisle I had "exponentiieren", but I doubt that any German fellow would understand what I mean. I guess there is no word in German for this up to now.
 
3:04 PM
@Skillmonlikestopanswers.xyz is "exponential growth" a common term in German (for bank interest rates, or virus infection rates)?
 
@DavidCarlisle See Skype
 
@JosephWright dare I follow those links:-)
 
@DavidCarlisle Hic sunt leones?
@DavidCarlisle They've been sorted by Karl ;)
 
@JosephWright more Hic idem I think
 
@DavidCarlisle "Exponentielles Wachstum" (the direct translation is pretty common, yes)
 
3:17 PM
@Skillmonlikestopanswers.xyz exponenzieren ?
 
@samcarter_is_at_topanswers.xyz most likely the correct variant of because of the "ti" :)
@samcarter_is_at_topanswers.xyz I'm just a bit confused because the Duden doesn't know it.
 
@Skillmonlikestopanswers.xyz Not really surprising, a lot of technical terms are missing.
 
3:56 PM
I have a tikzcd diagram that is a bit too dense, MWE: pastebin.com/ZtUvsiZs
How can I remove the clash in the middle? I tried adjusting some numeric values in \&[...pt], but it didn't make much difference. I'd like to retain the same font size.
 
4:08 PM
I upgraded my ghostscript from 9.27 to 9.52. Now the opacity does not work.
 
@TooFatManNoNeck Yup
@TooFatManNoNeck There's been lots of discussion of this
 
Oh my ghost. Thank you!
I roll back.
 
4:26 PM
@JosephWright About this, the specific problem seems to be that neither \filename@parse nor \file_parse_full_name:nNNN is expandable... It would make sense to have an expandable \file_parse_full_name:n that returns {path}{name}{ext}, wouldn't it?
 
@PhelypeOleinik Yes, I guess so: comes up often enough to be useful
 
@JosephWright Expect a PR in a bit ;-)
 
@UlrikeFischer Now I test colorspace a bit, to find out what really happens with mixing, and get weird errors :)
So, everyone signed to up TUG2020?
 
@JosephWright ?
 
@JosephWright What's the policy regarding leading spaces in file names? I'd like to use \__kernel_file_name_sanitize:n, but it trims spaces, so the results aren't identical to \file_parse_full_name:nNNN
@JosephWright Yes :-)
 
4:41 PM
@PhelypeOleinik No spaces ;)
@UlrikeFischer I've made a branch that now creates new Separation color models (no backend part yet, so I have to sort that); next I need to make sure they mix correctly. So I'm finding out what happens in colorspace
@PhelypeOleinik Hmm, i must have missed that: spaces Should be stripped off names at the ends (they are not 'safe' in all use cases)
@PhelypeOleinik Looks like a good programme: I'm working on my 'talk' (i.e. what to say on a recording)
 
@JosephWright Great, that makes things a lot easier. I asked because there's a test in m3file001 that explicitly tests leading spaces:
\test:n { test~file.tex }
\test:n { ~test~file.tex }
@JosephWright And the results differ in the test log
 
@PhelypeOleinik Perhaps I missed something somewhere ...
@PhelypeOleinik We should fix that: spaces at the start and end of names are not reliable
 
@JosephWright OK, I'll make \file_parse_full_name:nNNN use my expandable implementation underneath, so everything will (hopefully) do the same
 
@PhelypeOleinik Great: same as we've done for other stuff
\documentclass{article}
\usepackage{xcolor,colorspace}
\usepackage{l3pdf}
\ExplSyntaxOn
\pdf_uncompress:
\ExplSyntaxOff
\definespotcolor{textcmyk}{BarTone CMYK}{.8,.2,.5,.3}
\begin{document}
\textcolor{textcmyk!50}{Some text}
\textcolor{textcmyk!50!black}{Some text}
%\textcolor{textcmyk!50!cyan}{Some text}
%\textcolor{textcmyk!50!blue}{Some text}
\end{document}
@UlrikeFischer ^^^ Guess the outcome
@UlrikeFischer ^^ Then uncomment the line with cyan ...
 
@JosephWright error with black and a loop with cyan?
 
4:50 PM
@UlrikeFischer Yup :)
@UlrikeFischer Good guess
@UlrikeFischer I'm probably going to start with the backend part: that's much more trackable
 
@JosephWright ;-). I cheated ;-)
@JosephWright but basically this is documentated: mixing works only under conditions in colorspace. We should try to get either better errors or fallback with warning.
 
@UlrikeFischer Certainly better errors!
@UlrikeFischer I'm going to sort mixing with white then the backend, then more general mixing, then DeviceN
@UlrikeFischer github.com/latex3/latex3/tree/color-models is where i'm working on this
 
 
2 hours later…
7:17 PM
@Skillmonlikestopanswers.xyz (and perhaps @DavidCarlisle): Isn't "potenzieren" as in "e potenzieren mit ..." the verb you are looking for?
 
@AlexG no, "potenzieren" means x^(something), I want (something)^x, the other way around.
 
@Skillmonlikestopanswers.xyz But couldn't it be used in either context? "etwas potenzieren mit x"
 
@AlexG I'd not use it that way. Used in one word it means something different, you need more to describe what you're meaning, that's not what I want :)
 
@Skillmonlikestopanswers.xyz ok, was just an idea...
 
@AlexG appreciated :)
 
7:28 PM
@Skillmonlikestopanswers.xyz Wikipedia suggests: anitlogarithmieren, exponenzieren (not in Duden, as @samcarter_is_at_topanswers.xyz already noted): de.wikipedia.org/wiki/Antilogarithmus
 
@AlexG thanks for a more reliable source (wikipedia) :)
 
8:09 PM
@DavidCarlisle regarding the pdftexcmds issue: wouldn't it make sense to set \let\pdfelapsedtime\elapsedtime and similar for \pdfresettimer in xelatex?
 
@UlrikeFischer well generally it just defines \pdf@zzz names doesn't it? for the actual issue we could use \ifcsname rather than the grouping hack but I suppose the rest are all in that style
 
@DavidCarlisle ah, I thought it wouldn't define the command for xelatex, but it already does. Can we use ifcsname in style that nominally should work in plain too?
Regarding plain: try this with luatex:
\input pdftexcmds.sty

blub

\bye
 
@UlrikeFischer who needs catcode tables:-)
 
8:27 PM
@DavidCarlisle ;-). where is the number from?
 
@UlrikeFischer ltluatex.tex
 
@DavidCarlisle I wished there were less expandafter. Where is a place for "luatex +plain" to load the file? Or should we simply say, it is not for plain?
 
9:00 PM
@UlrikeFischer I'll look in a bit
@UlrikeFischer the sensible thing would be for plain luatex to be more like lualatex and preload some definitions and luaotfload etc, but we were not brave enough to do that and it may be too late now
 
@DavidCarlisle I think I will simply add after the \ifluatex \ifcsname catcodetable@string\endcsname\else\input{ltluatex}\fi. Does that sound sensible?
 
@UlrikeFischer yes certainly better than an undefined csname error
 
@DavidCarlisle ok added.
 

« first day (3524 days earlier)      last day (1406 days later) »