« first day (3297 days earlier)      last day (1621 days later) » 

12:40 AM
@UlrikeFischer — so Renderer=HarfBuzz doesn't work still in fontspec? Sorry I'm behind on this, I thought it had been fixed up :(
@UlrikeFischer I mean Renderer=Harfbuzz
@UlrikeFischer Wait! It SHOULD be HarfBuzz but is currently Harfbuzz...
 
hi @WillRobertson bit late for Ulrike I guess:-)
 
 
6 hours later…
7:08 AM
@WillRobertson I'm not quite sure to what you are referring to, but Marcel said in a luaotfload issue that there could be a problem., But I'm out today most of the time (chess match) and can't look.
@WillRobertson that confused me already, I expected HarfBuzz ;-).
 
 
2 hours later…
8:42 AM
quack
 
 
1 hour later…
9:58 AM
hmmmmm
 
 
2 hours later…
11:53 AM
PL3 building...
 
12:24 PM
... gone to ctan
 
 
1 hour later…
1:27 PM
@JosephWright Do you have a moment (or several) for a few questions about l3build?
 
@moewe Sure
 
@JosephWright Great, thank you.
@JosephWright As you may remember I'm trying to test my biblatex styles with l3build.
There is very little expandable stuff that I could write to the log file so I think I need something like \showoutput to send the typesetting info to the log
 
@moewe Yes, that or a box to show
 
so far so good.
The files get quite long if I test a lot of output, so testing takes a while, but I guess I can live with that.
But then I realised that this setup doesn't catch errors that appear when the style is loaded.
 
@moewe You could just check around the loading line(s)
@moewe For example, for xfp we have
\documentclass{minimal}
\input{regression-test}
\RequirePackage[check-declarations]{expl3}
\ExplSyntaxOn
\debug_on:n { deprecation }
\ExplSyntaxOff

\START
\AUTHOR{Bruno Le Floch}
\RequirePackage{xfp}
\END
 
1:33 PM
which gives me a lot of stuff I don't need to see: github.com/moewew/biblatex-ext/blob/travis/testfiles/…
I was hoping there might be a clever way to only get warnings and errors for the style/package loading...
 
@moewe Load all the dependencies first, and if necessary redefine \e@alloc
@moewe I'd use a separate test file for the load test and the runtime stuff
 
@JosephWright I was hoping to be able to use only one file per style, but it might be better to separate those tests, yes.
 
1:51 PM
@JosephWright The idea of loading all dependencies first and messing with \e@alloc looks very nice, though.
Since I'm already messing with all kinds of internals of biblatex, I might as well mess with LaTeX...
 
@moewe It's more-or-less what we do for contrib tests
@moewe l3build working for you generally? It's good to know we are getting more users: I'd like to see it used for most LaTeX dev work
 
@JosephWright So far quite well. Mind you, I'm only testing my own biblatex style at the moment and have yet to build a larger test suite. So I'm still quite far away from moving looking into moving biblatex tests to l3build.
 
 
2 hours later…
4:06 PM
@moewe you can usually use \showbox rather than \showoutput, also if you don;t need to see it all, just catch errors etc you can set \showboxdepth and \showboxbreadth to something less than \maxdimen
 
@DavidCarlisle Currently I'm \showboxing stuff (github.com/moewew/biblatex-ext/blob/travis/testfiles/support/…), but I'm wondering whether I should be using \showoutput for the bibliography since that is a list...
I'm having a few troubles wit excludefiles, if I understand correctly the setup at github.com/moewew/biblatex-ext/blob/travis/build.lua should make sure that TODO.md does not end up in the CTAN .zip, but it somehow still does ...
@JosephWright ^^^^
 
@moewe excludefiles is mainly about Emacs backups: I'd just use a fixed textfiles list
@moewe I'd have to check back where excludefiles is applied: it was added for Frank ...
 
cis
4:22 PM
Hallo my friends in the U.S.!

Question: By the emotion... Is ''you" more "thou" (example: Pauolo, you....)
or
more "you" (example: Mr. Carlisle, you...)?
 
@cis Huh?
 
cis
@JosephWright :()
 
@JosephWright I think he wanted to know if it is considered formal or personal, as in German there is "Du" for informal/personal conversation and "Sie" for formal situations
 
cis
@Skillmon Yes, Mr. Skillmon
 
@cis Er, well, technically thou is singular and you is (or was) plural ..
That's like 500 years ago, of course
 
4:29 PM
@cis but since "thou" isn't used in modern English there is no distinction between the formal and informal use in modern English (but it might be worth asking people, who know things, @AlanMunn)
 
cis
@JosephWright Mmmhh... So "you" is like formal speech
 
@cis it was, in the not-so-near past. (but I'm no Englishman, so what do I know...)
 
cis
@Skillmon But todays usage is singular and plural. Mhh...
 
@cis As @Skillmon says, today in normal speech we don't use 'thou' at all: it's in say the King James bible, so it 'sounds' religous
@cis We just don't have the distinction in modern English
 
@moewe it depends how much you want to show, if you just want to show that the list is basically the same as expected, then turning down \showboxdepth might owrk or simply using \pdfsavepos and logging where it ends and not showing box output at all or ..
 
cis
4:33 PM
@JosephWright Ah ok. Yes, I know 'thou' is not used today. I wonder, what is the emotion, when I say 'you'...
 
@cis it depends on the context, it is fairly neutral as to formality. emotion etc.
 
cis
@DavidCarlisle Mmhh, OK.
 
@cis It's so archaic, it's hard to say: it sounds 'olde-worlde' more than anything
 
@cis this is English, either there are no rules, or the rules can be ignored, and you can make things up as you go along (where here you is not referring to anyone in particular, I noticed after writing it)
3
 
@DavidCarlisle 'One'
 
4:36 PM
@JosephWright only for posh people like you
 
@DavidCarlisle Tar
 
@JosephWright :)
 
cis
4:49 PM
Is it posible to add "Python- or Sage(math)-thing" here stackexchange.com/sites# ?
 
@cis you can ask Python questions on stackoverflow.
@cis and even if not, this chat is the wrong place to ask :)
 
cis
@Skillmon Ah. Or do you know: is there an official Sage-forum from the developers?
 
cis
@Skillmon Ah Ok.
 
@DavidCarlisle So I can just make up words like "multicriterionoptimizationtechnique"?
 
@Skillmon only if you put more spaces in it - english people love spaces ;-)
 
@UlrikeFischer yeah, but he said I could ignore that rule of spaces :(
 
@Skillmon yes, if you get more than a dozen people to use it in social media for a couple of years you will probably find that it gets added to an Oxford English Dictionary Supplement at some point.
@Skillmon there really is no deciding authority, if a string of letters is used as a word, then it is a word.
 
5:15 PM
@DavidCarlisle so, from this time on, I hereby enact that this chat never uses the combination of words "multi-criterion optimization technique", but substitutes it with the now correct "multicriterionoptimizationtechnique".
 
@Skillmon it's probably already true that a majority of the uses in this forum use the multicriterionoptimizationtechnique form
 
@DavidCarlisle most likely, yes. But that comes naturally, as multicriterionoptimizationtechnique is such a lovely easy to type and memorable word.
 
@Skillmon quite so, almost as good as duckpizza
 
@DavidCarlisle and, the best about English, imho, is the punctuation, with commas.
@DavidCarlisle sounds tasty. When where with whom?
@DavidCarlisle (actually I've shortened it, the original word would have been "multicriterionoptimizationimagereconstructiontechnique", but that would have been too easy to type)
 
@DavidCarlisle I think hyperref is broken (or iftex).
 
5:25 PM
@UlrikeFischer Ooopy
 
@UlrikeFischer just looking
@UlrikeFischer you get the error?
 
@DavidCarlisle yes, but only with latex. But I'm not sure if I have the newest hyperref, let me check.
 
@UlrikeFischer I cleaned out my local copies and ran with tl2019 install and it worked for me
@UlrikeFischer scrap that tl is at 7.0a
 
@DavidCarlisle what is the hyperref version?
 
@UlrikeFischer That's odd, I fixed that before looks like I reverted something by mistake, let me see...
 
5:49 PM
@cis As others have said, the thou forms were historically singular and you forms plural. During the Middle English period (1100-1500), there was a gradual adoption of using the plural you forms for formal address and the singular thou forms for familiar address, presumably influenced by French. By the 18th C, the thou forms are pretty much entirely gone except as used in religious texts, although regional dialects of English preserved them much longer.
 
@UlrikeFischer need to go for a bit but I pushed a fix... (I hope)
 
@cis Today they really can't be used at all, except in a somewhat jokey way. And typically people don't know the inflections and so will add –eth (3rd person singular) to the verb instead of –est.
 
6:12 PM
@DavidCarlisle I can't run l3build install to install ;-(
 
@AlanMunn -- Is it possible that "thou" might still be used by older members of some religious groups? (I'm thinking of Amish or Quakers, and probably only among other members of the same group.) It would certainly be very rare.
 
6:28 PM
@UlrikeFischer do this -- bundlectan= {error('Use ./mkctan for now')}
@JosephWright ^^ how do I disable l3build ctan without disabling install ? (@UlrikeFischer)
 
@DavidCarlisle Add it to the right table ;)
 
@JosephWright er yes, bundlectan sounded a good name and worked for what I wanted the other day (allowing check and disallowing ctan:-)
 
@DavidCarlisle target_list = target_list or { } target_list.ctan = {function = noop() }
 
@JosephWright If I get some time without patch releases or the day job I'd get rid of mkctan (should be easier in hyperref than oberdiek) But I just wanted to make sure we/I didn't send something to ctan from an unconfigured l3build (again:-)
 
cis
@barbarabeeton @AlanMunn @barbarabeeton And I thought, I could be most poshly if I say thou, you and so on... ;)
 
6:32 PM
@DavidCarlisle Got it
 
@cis -- If I heard that, I'd assume that you were either a member of a relict religious group, or rehearsing for a period play.
 
@DavidCarlisle did you steal the visible spaces for the file names? ;-)
 
6:55 PM
@MarcelKrüger I think the scripts works okay now. I will document it and then try to make an upload.
 
7:20 PM
@UlrikeFischer I just noticed that build.lua has `checkengines = {"pdftex","luatex"}` I guess we should check `latex` as well....
`
 
Bob
Can someone help me with tikz? I cannot figure out how to handle libraries.
 
@DavidCarlisle yes, good plan ;-). Perhaps we also need some tests which loads the other dvi drivers.
 
@UlrikeFischer checkengines = {"latex","pdftex","luatex","xetex","uptex"} ?
 
Bob
By which primarily I think I mean, how do I determine which libraries to use?
 
@DavidCarlisle you could try - if some test fails we will perhaps have to separate them.
 
Bob
7:22 PM
My problem is that, most of the examples in the tikzpgf manual do not work, and I cannot figure out why, I think its libraries but im not sure.
 
@UlrikeFischer I'm going to try that first with the fix I just made backed out, hopefully it will fail somewhere
@Bob they have been working on annotating the examples in the manual with which libraries are needed but I don't know if that is in teh released version yet
 
Bob
Ok thats good, there doesnt seem to be a list of libraries that I can find. Is there one?
 
@Bob I know nothing about tikz:-)
 
@Bob the index has a long list.
 
@UlrikeFischer oh I was going to suggest running find in the search path, looking in the manual honestly didn't occur to me....
 
7:27 PM
@Bob @DavidCarlisle ^^^^
 
@UlrikeFischer but can any of them do this ?
 
Bob
@UlrikeFischer ok thank you!
 
@UlrikeFischer Right, the bugs should be fixed. I was trying to figure out why the tests still failed, but I now found that the old code was buggy: Instead of "Kashmiri", it selected "Kölsch" as language for some arabic numerals.
 
@DavidCarlisle create a new repo for \usetikzlibrary{davids-art-gallery}?
@MarcelKrüger Kölsch?? some karneval joke?
 
@UlrikeFischer I was thinking \usetikzlibrary{picture-mode} where it replaces all the basic drawing commands with picture mode constructs
 
7:32 PM
@MarcelKrüger you mean the failure with amiri? I was just wondering about this too. Did you already pushed the fix?
 
@UlrikeFischer Just pushed it, Travis is testing it right now.
 
7:45 PM
@UlrikeFischer can you do pvt based pdf testing against latex+(dvipdfm or dvips+ps2pdf or anything) or would we need a lvt-only config directory?
 
@barbarabeeton I asked a Quaker friend of mine who said that some conservative (in the religious sense) Quakers do use it but only with family.
 
@DavidCarlisle I'm not sure what the state is here, I asked @JosephWright about dvips tests. The questions is how to start the backends.
 
@AlanMunn -- Thanks for confirmation. That's pretty much what I expected. (@cis -- see above)
 
@barbarabeeton Oh, and they only use 'thee' not 'thou' (independent of case).
 
@AlanMunn -- So "thou" is probably extinct.
 
7:54 PM
@barbarabeeton Although I would think they would still use 'thy' for the possessive.
 
@AlanMunn -- Makes sense. "Thine" is almost certainly extinct then.
 
@barbarabeeton Among standard English speakers, yes. I think there are still dialects in which it is used, Shetland English being one I'm aware of, but there may be more.
@barbarabeeton No idea. What's interesting about the 'mine' 'thine' forms, is that the distinction between 'my' and 'mine' used to be purely phonological: 'n' before vowels, just like the 'a' 'an' distinction, but then the 'my' form won, and the 'mine' forms remain only in what is traditionally called the noun form.
 
@UlrikeFischer for now (not pushed) I have separated the lvt and pvt tests into separate directories, then can do basic "not completely broken because I missed a vtex guard" tests without having to worry about pdf comparisons as I keep teh existing checkengines for the pvt ones
 
8:58 PM
@UlrikeFischer know anyone who can read pdf diffs:-)
My failures now down to
build/test-config-pvt/test0-pdf.luatex.pdf.diff
build/test-config-pvt/test0-pdf.pdftex.pdf.diff
build/test-config-pvt/test4-pdf.luatex.pdf.diff
build/test-config-pvt/test4-pdf.pdftex.pdf.diff
but not sure what to make of the diffs, the first chunk in the first pdftex failure is
*** ./build/test-config-pvt/test0-pdf.tpf	2019-11-10 20:44:26.993734700 +0000
--- ./build/test-config-pvt/test0-pdf.pdftex.pdf	2019-11-10 20:44:46.531911800 +0000
***************
*** 73,79 ****
  0 g 0 G
   [-903(E4)]TJ
  0 g 0 G
! /F17 9.9626 Tf -69.739 -21.918 Td [(6)-925(T)96(esting)-384(External)-383(Jumps)-19626(E5)]TJ/F16 14.3462 Tf 0 -32.946 Td [(Abstract)]TJ 0 -27.798 Td [(1)-1125(A)-375(\014rst,)-375(simple,)-375(section)-375(heading)]TJ/F8 9.9626 Tf 0 -21.821 Td [(And)-333(some)-334(text)]TJ/F16 14.3462 Tf 0 -32.946 Td [(2)-1125(A)-375(funn)31(y)]TJ/F18 14.3462 Tf 87.281 0 Td [(AN
 
@DavidCarlisle hm, the difference seems to be in the text: (E5) versus (E4).
 
@UlrikeFischer yes
 
@DavidCarlisle I need to check the test.
 
I'm re-checking my vtex guards, I guess I need to try to get this to ctan today really even if some tests not passing, can't really leave it broken in dvi latex
@UlrikeFischer the lvt/tlg tests with latex looked OK apart from obvious \pdfdest->\special differences
 
@DavidCarlisle do you have enough runs? This seems to be a page number in the toc.
@DavidCarlisle did you already pushed the new test setup?
 
9:13 PM
@UlrikeFischer config-pvt has checkruns=2 which was what build.lua had, but I'm not convinced they were passing before
@UlrikeFischer yes
 
@DavidCarlisle I think the test is missing some support files, I see an error as hog is not found.
 
@UlrikeFischer I did a second push that added eps versions of hog and picture for teh etex based tests, did you get those (should have hog.eps in testfiles/support
 
@DavidCarlisle the files are there, but l3build don't copy them into the build folder. Something is wrong with the support dir setting.
 
9:28 PM
@UlrikeFischer sorry you are right, the config-pvt file has testsuppdir = testfiledir .. "../testfiles/support" ah no / befoe the ..
@UlrikeFischer trying with testsuppdir = testfiledir .. "/../testfiles/support"
 
@DavidCarlisle just trying too
@DavidCarlisle hog is now in the build folder.
 
@UlrikeFischer that might help:-)
 
@DavidCarlisle the test passed.
 
yeh, OK thanks, I'll run the full test suite again and if that looks good I'll check in the remaining changes and push to ctan (they are going to be so cross with us)
 
9:50 PM
@UlrikeFischer all tests passed. Thanks, sometimes an extra pair of eyes helps:-)
 
10:38 PM
@egreg Hi, look here please: the code works correctly. tex.stackexchange.com/questions/515892/…
\documentclass[12pt]{report}
\usepackage[T1]{fontenc}
\usepackage[utf8]{inputenc}
\usepackage{graphicx}
\usepackage{amsmath,amssymb,amsfonts}
\usepackage{txfonts}

%\usepackage[polish]{babel}
\setlength{\parindent}{0pt}

\renewcommand{\chaptername}{Rozdział}
\renewcommand{\contentsname}{Spis treści}
\renewcommand{\figurename}{Rys.}
\renewcommand{\tablename}{Tab.}
\renewcommand{\listfigurename}{Spis rysunków}
\renewcommand{\listtablename}{Spis tabel}
\renewcommand{\bibname}{Bibliografia}

\pagestyle{headings}
 
@Sebastiano yes that is what egreg said.
 
@DavidCarlisle Thank you very much for your comment. I wanted to be of help.
 
10:55 PM
@Sebastiano the question is what did the user input to get the image shown, getting the output from the code posted is easy:-)
 
11:27 PM
@egreg oi I was still typing:-)
 
yo'
11:40 PM
@JosephWright Trump--Robertson (Champ of Champs final, 19 frames): Trump needed a snooker in frame 17, made it and went to 9:8. Then Robertson needed a snooker to equalize in frame, made it, potted the second black, and went to 9:9. And the final frame was finished with a hundred. What a match! This was worth waiting until late after midnight!
 
11:54 PM
@yo' I was confused at first I got some way into the sentence thinking it was about POTUS :-)
 
@DavidCarlisle -- This one's for you: tex.stackexchange.com/q/515900
 
@barbarabeeton I'm sure it's a dup but I might answer anyway
@barbarabeeton voted as dup of
23
A: Why LaTeX2e has this name?

David CarlisleAs Frank says in his interview Although LaTeX2e already addressed most, if not all, of the deficiencies identified in the first decade of LaTeX 2.09 use, it was originally thought that LaTeX2e would only be an intermediate step towards a LaTeX 3 version So that really is the name 2-e...

@barbarabeeton I suppose the non mathematicians on the site might need some words to explain why \epsilon denotes a small step...
 

« first day (3297 days earlier)      last day (1621 days later) »