« first day (1761 days earlier)      last day (3178 days later) » 

12:51 AM
@DavidCarlisle Conditionally I would say 'yes'. :)
 
 
7 hours later…
8:06 AM
@DavidCarlisle Subjunctive?
 
8:33 AM
-2
Q: Can anyone provide me with patgen.c?

Arpit JaiswalI am unable to convert patgent.web to patgen.c due to lack of resources . SO if anyone has it , I would like to have it. Thanks in advance .

Is there more to say than I've already commented?
 
@egreg "sounds right" is the main classification system I use
 
8:44 AM
@DavidCarlisle :-)
 
9:03 AM
@JosephWright reading up on lua varargs/... stuff: something odd happening in the user callback code, the define font callback is being called with two arguments but as far as i can tell second one is never passed and comes through as nil....
 
@DavidCarlisle We've just copied that code from luatexbase haven't we?
 
@JosephWright it looks that way, except original works.... need to do some day job work, but am annotating versions to see where it diverges.
 
@DavidCarlisle I'll see if I can have a look too
 
in ltjfont-lua there is
      texio.write_nl("XXXXXX fn: " .. type(fn))
      fmtable = luatexbase.call_callback("luatexja.define_jfont", fmtable, fn)
@JosephWright where fn is a number but in
luatexbase.add_to_callback(
   "luatexja.define_jfont",
   function (fmtable, fnum)
      local vadv = {}; fmtable.v_advance = vadv
      local vorg = {}; fmtable.v_origin = vorg
      local ft = font_getfont(fnum)
fnum is nil so the font lookup sets ft to nil and eventually tex stops with
! LuaTeX error ...exlive/2015/texmf-dist/tex/luatex/luatexja/ltj-jfont.lua:843:
attempt to index local 'ft' (a nil value)
 
@DavidCarlisle I see that the version of call_callback we've got isn't the same at the one in luatexbase
  if not l then
    f = user_callbacks_defaults[name]
    if l == false then
	   return nil
	 end
  else
versus
if not l then
    f = lua_callbacks_defaults[name]
else
 
9:15 AM
@JosephWright yes there was a comment in the original that the default could have been made optional and that's what we did I think, but anyway in this case the luatexja.define_jfont is being called, unless I am totally confused
 
 
1 hour later…
yo'
10:15 AM
@PauloCereda Hey how!
 
@yo' yo hey yo!
<3
 
yo'
@PauloCereda :)
What iis the right choice for French? french or frenchb?
 
@yo' The former
 
yo'
@egreg ok, thanks
And please, is \emph{\foreignlanguage{}{...}} better than the other way around, or it doesn't matter?
 
@yo' Check the italic correction in either case.
 
yo'
10:28 AM
@egreg ok, thanks
 
@yo' It's the same.
 
yo'
@egreg thanks a lot
 
With \emph{\foreignlanguage{italian}{l}} b I get
...\hbox(6.8872+0.0)x345.0, glue set 317.52528fil
....\hbox(0.0+0.0)x15.0
....\setlanguage51 (hyphenmin 2,2)
....\T1/cmr/m/it/10 l
....\kern 1.03307
....\glue 3.33252 plus 1.66626 minus 1.11084
....\setlanguage0 (hyphenmin 2,3)
....\T1/cmr/m/n/10 b
With the reversed order I get
....\hbox(0.0+0.0)x15.0
....\T1/cmr/m/it/10 l
....\kern 1.03307
....\glue 3.33252 plus 1.66626 minus 1.11084
....\setlanguage0 (hyphenmin 2,3)
....\T1/cmr/m/n/10 b
 
@JosephWright @egreg instead of a private message, it got public: latex-community.org/forum/…
 
yo'
@egreg looks identical, since the position of \setlanguage doesn't matter here
 
10:34 AM
@egreg: Buongiorno! :) Haven't heard from the GuIT folks yet. :(
 
yo'
@PauloCereda maybe they don't speak any Native American language ... :-/
 
@yo' Not really: I can't see the language switch in the latter case.
 
yo'
@egreg ah right, neither I can. Let's test it, I've got some hyphenation of Nadání going on here :)
 
@yo' Oh no!
 
yo'
@egreg I get the correct hyphenation: Na-dání, which is forbidden in english, so it seems to work; just the \setlanguage is not shown for some reason,... sorry wrong test
 
10:38 AM
@yo' Long strikethrough line is long. :)
 
yo'
@PauloCereda but still I was correct, the hyphentaion works as expected...
A---------------------------------------------------------------------------------------------------------A
 
@yo' oooooooh <3
 
@PauloCereda you're not a duck on RG :(
 
@DavidCarlisle I tried to be, but I got a warning from my peers. :(
 
@yo' No hyphenation in \foreignlanguage! Curious.
 
yo'
10:43 AM
@egreg I seem to be fine...
 
@egreg In general? I got a sample document with it.
 
@PauloCereda Ive taken to marking all mail from RG as spam, but just got one that got through and noticed your picture:-)
 
@DavidCarlisle Oh my! :) I haven't marked them as spam, but I created a filter to silently move them directly to the trash. Those things are very annoying.
@David: I'm yet to try that OrcID thingy @Joseph mentioned.
 
@yo' Yes, I was testing the wrong thing. But there's some problem anyway.
 
yo'
@PauloCereda ah now I finally know what RG you speak about!
 
10:45 AM
@yo' ResearchGate. :)
 
@yo' It's a secret society. Best not to join it.
 
The first is \foreignlanguage{latin}{Totus...}
 
yo'
@PauloCereda yep, I think they keep sending mails on behalf of me without my permission. However, everybody knows that they're ejaculating spam all over the place, so it's probably fine.
 
@yo' The problem is with the apostrophe, for instance: in Italian it should participate to hyphenation, but it doesn't when in \foreignlanguage. :(
 
@yo' clearly unreliable as they just sent me an email claiming that @PauloCereda's picture was human and he'd written something that wasn't about ducks, parrots or church songs.
 
10:47 AM
@David: oooh I think I know why one mail got through! I accidentally endorsed you in LaTeX!
@DavidCarlisle <3
 
yo'
@egreg isn't there a catcode problem? Have you tried \begin{otherlanguage}{italian}...\end{otherlanguage}?
 
@yo' more likely lccode problem, for hyphenation
 
@DavidCarlisle Yes.
 
yo'
@DavidCarlisle well, anyway, an early tokenization problem :-)
 
@yo' no doesn't matter about tokenization for lccodes, what matters is the lccode at end of paragraph
 
yo'
10:51 AM
@DavidCarlisle ah that's bad...
 
@yo' blame @FrankMittelbach for not getting DEK to do TeX3 right:-)
 
@DavidCarlisle Well, if I ask for \showthe\lccode`' inside \foreignlanguage{italian} I get 39. But, of course, the lccodes are looked at at \par, so it can't work.
 
yo'
@DavidCarlisle well, couldn't at this moment the fact that I use LuaLaTeX matter?
 
@yo' you could almost certainly fix this in luatex yes
 
yo'
@DavidCarlisle well, I need not to I think :) For 2 French and 2 Czech half-sentences...
 
10:54 AM
@yo' adding a couple of manual \- might be easier than writing a lua hyphenation callback, yes.
 
yo'
@DavidCarlisle well, I don't need to consider ' for hyphenation, so \foreignlanguage works just fine :-)
 
New church songsheet! ^^
 
yo'
@PauloCereda have you tried \raggedright? (Only asking, I don't know if it's gonna be better)
 
@yo' I tried it, but I got some mixed feelings. :) I did a little experiment with my group and they found a little difficult to follow the lyrics (maybe because they are used to full justification).
 
yo'
@PauloCereda ah ok, good then
 
11:01 AM
@yo' <3
 
yo'
I would just set sp of / to the same as of . so that the slash has equally long spaces on either side :)
Ah well, it's more difficult than this, because ! has a different value of sp for your language...
 
@yo' ooh
 
yo'
maybe something like \def\divverse{\unskip\spacefactor2000\ /\spacefactor2000\ \ignorespaces}
 
@yo' thanks, I will try something later on. :)
@egreg: Yesterday I had Coca-Cola in a wine glass. :)
 
yo'
@PauloCereda sorry a small mistake of course:
\def\divverse{\unskip\spacefactor3000{} /\spacefactor3000{} \ignorespaces}
 
11:11 AM
@PauloCereda Horror! See you later!
 
@egreg :) Have a great day!
 
yo'
@egreg later
 
11:54 AM
@JosephWright When and where?
 
kan
Hello everyone!
 
yo'
@kan long time no see!
 
kan
Yuppie!
I was writing my masters thesis. :-)
Now, I am back with much more time!
 
yo'
@kan I'm writing down my (true) thesis :-) But I hope you're doing well!
 
kan
I should be able to post more regularly!
@yo' Hah! Good luck with that! Yes, I am! :-)
For Duck and German enthusiasts: Old Macdonald in Deutsch: youtube.com/watch?v=U4LHBQK870A :-)
 
yo'
12:01 PM
@kan thanks
 
yo'
12:11 PM
Can I declare a ligauture in lualatex/fontspec?
 
12:28 PM
@yo' well for tfm fonts you can change the ligature table from lua, possibly can do same for otf as well, not sure, see how fontspec implements "ligatures=TeX" I guess or of course you can leave the fonts as they are and re-arrange the node list after the fact replacing adjacent runs of character nodes by whatever ligature node you have
 
yo'
@DavidCarlisle the problem is that --- works, but not inside \textsc ... :-/ I solved it by using Unicode...
 
@yo' ping Will? tex ligatures are supposed to be on by default these days I think.
 
yo'
@DavidCarlisle even SmallCapsFeatures = { Ligatures = TeX } doesn't help. I'll make a MnWE
oh no! In a minimal example it works!
 
@yo' typical minimal example behavior:-)
 
yo'
Ah it's letterspace that causes it!
 
12:34 PM
@yo' well that seems reasonable
 
yo'
@DavidCarlisle Ummm should letterspace have priority over all ligatures?
@WillRobertson Is there a way to distinguish letter and non-letter ligatures so that LetterSpace does not prevent --- from being made into an em-dash? MWE is here:
%!TeX program = lualatex

\documentclass{article}

\usepackage{fontspec}

\setmainfont{texgyrepagella-regular.otf}[
	ItalicFont = texgyrepagella-italic.otf ,
	BoldFont = texgyrepagella-bold.otf ,
	BoldItalicFont = texgyrepagella-bolditalic.otf ,
	SmallCapsFeatures = { Ligatures = TeX, LetterSpace = 5.0 } ,
	ItalicFeatures = { Ligatures = TeX, LetterSpace = 20.0 } ,
	Ligatures = TeX ,
	]

\begin{document}

\rmfamily

Works --- but {\scshape Doesn't work --- Why?}

\textit{LetterSpace --- causes it}
 
@yo' well seems to me if you are inserting space between each letter you have no adjacent letters to ligature. of course --- is really a fake ligature abusing the ligature mechanism to get a variant markup so that isn't the behaviour you want there but ...
 
yo'
@DavidCarlisle yeah I see, let's wait what Will thinks about it. But I suppose that this is "a feature"... I think I can either live with it or make LuaTeX's mechanisms do the replacement.
 
@yo' that's what I'd say:)
 
yo'
@DavidCarlisle You and your features....
 
12:37 PM
@yo' but this one is I think really by design.
 
yo'
@DavidCarlisle Yeah I know :) Well, thanks, now I go carry on writing
 
cfr
1:29 PM
Could somebody who knows at least something about MikTeX take a look at this:
http://tex.stackexchange.com/questions/263150/fontawesome-wont-render-correctly-while-working-on-a-moderncv
 
Guys, stylistic question: how do you guys type an all caps acronym, for example, ASCII? Literal ASCII?
 
@PauloCereda if I remember correctly tugboat has a medium caps font setup for this, but in practice I use all caps small caps looks hmm small
 
@DavidCarlisle ooh good plan! :) Let me check TUGboat's \acro.
[paulo@cambridge ~] $ texdef -t latex -c ltugboat acro

\acro:
\long macro:#1->\textSMC {#1}\@
 
@PauloCereda stop messing around with texdef -t and just put tugboat.cls into emacs like a man duck
 
1:51 PM
@DavidCarlisle oh
@DavidCarlisle It worked, it worked!
 
@PauloCereda using emacs? of course it worked.
 
@DavidCarlisle boo :)
 
yo'
2:33 PM
60 references in a thesis, that starts to be a good number...
(including one auto-citation)
 
@DavidCarlisle Good spot :-)
 
@JosephWright half a day to spot a comma and three full stops were not there:(
@JosephWright thanks for the help though, spotted it within minutes of sending mail to you:-)
 
@DavidCarlisle So is luatexja working?
 
@JosephWright well I get no errors from an empty document, does that classify as working?
@JosephWright certainly working a lot better than yesterday
 
2:48 PM
@DavidCarlisle For a set up where we can't read the documentation, code comments or even bits of the code, I think so :-)
@DavidCarlisle So the big question is when do we release
 
@JosephWright been down that list of use of \luatex... primitives want to add a couple more used in packages (including latexja) that load luatexbase. then that just leave the directional primitives which i don't plan to add, only found them in eledmac and I can contact the maintainer for that and then the lua escape thing in polyglossia which can use expl3
@JosephWright 2015/11/01 it says on the tin?
@JosephWright advertise on latex-l we plan to drop the prefixes then aim for 2015/10/01 release (ie a months time)?
 
3:32 PM
@DavidCarlisle OK
@DavidCarlisle All sounds good to me: I'll write something later today unless you want to take it
 
@JosephWright you're team spokesperson:-)
 
3:52 PM
Even this naming scheme is making me furious. Medium Length Professional CV Who makes up a name like that?
And why is the downloaded zip file called cv_4? My mind cannot process this madness.
 
4:03 PM
@DavidCarlisle Sometimes it seems that way: I always think that's Frank :-)
 
4:39 PM
@clemens Do you know the syntax?
@clemens Each command you think very important, you should put as \command. The less often used as \command#*
@clemens commands which should not be autocompleted but shown as "correct syntax" in TeXStudio (quite handy highlighting feature of TS) should be written as \command#S
@clemens You can improve the experience for TS users of your package quite a lot by taking those decisions. Thanks.
@clemens feel free to put that file in your VCS. Just delete my vain "LaRiFaRi" signature
 
5:00 PM
@LaRiFaRi actually i don't: I've never used TeXStudio myself…
 
5:14 PM
@PauloCereda interesting – I would have expected the opposite…
 
yo'
@clemens It is easier to follow the lines when they are justified, because at least something is fixed: each verse is a block of text. I realized this with 2-column bibliography: raggedright is the working solution, but you can't really "read" the bibliography (well, you don't "read" it anyway, do you...) That's why I only suggested trying raggedright
 
@clemens Maybe it's something cultural. It's very difficult to read a text in here that isn't raggedright.
 
yo'
@PauloCereda "is"?
 
@yo' For my band I and my band members have written loads and loads of leadsheets – always ragged right: that way the text lines follow the song structure
 
yo'
@clemens ah you speak about something else: replace slashes by newlines. Yes, that's possible, but in general it takes a lot of extra space. With a songbook like here --- 700 songs, you don't want things to take extra space.
Actually, our songbooks are printed on bible paper so that they are not too heavy.
 
5:23 PM
@yo' Oh is. Sorry. :)
 
@yo' I know, I noticed them and I know that notation as well. But my perspective is different – I don't sing but I accompany the singers
@yo' sounds a bit like Das Ding. I think it only has about 400 songs in it, though.
 
yo'
@clemens Looks similar to Scottish (British?) Common Ground or Czech Svítá (Sunrise). These also have about 4-5 hundred. Also, they are much havier than our basic church songbook :)
 
@yo' it's been years sonce I had a church songbook in my hands… one main difference between church songbooks and songbooks like Das Ding is of course that the latter always have the guitar chords above the lines of the first verse and chorus, often even with chord charts – this alone needs more space
 
yo'
@clemens yep, I know
 
@barbarabeeton I'm confused I just added \listoffigures to the code in my answer and the short lh appeared in the list not the long form? (sorry it's argue-with-barbara-day today:-)
 
 
1 hour later…
6:39 PM
@clemens Wow, four volumes? I knew there was a second, but four?
 
@Johannes_B I was surprised, too.
 
@clemens My name is removed form the template site :-)
 
@Johannes_B cool
 
@clemens There was one sentence i couldn't quite understand: Look, basically the problem is, you write good code at the cost of usability, I work on usability at the cost of good code.
 
@Johannes_B if that statement were true both variants would be bad. He should really aim for good code and usability. (I believe he has a different idea of what usability means than you…)
 
6:49 PM
@clemens I know one thing for sure Don't use CC-licenses for code.
 
@Johannes_B why?
 
@clemens The class file is CC3.0 and the doc of the template encourages the user to change stuff in the class if there is any need.
 
@Johannes_B but doesn't the license allow to change files? (IIRC you have to indicate the changes and give credits if you distribute the changed files, though)
@Johannes_B IMHO the class should provide options for user settings so a user doesn't have to touch the class file itself (would also be easier for reverting to the default settings)
 
@clemens Oh, you are right, changes have to be indicated creativecommons.org/licenses/by-nc-sa/3.0
@clemens but to be honest, would a starter check out the license terms, especially if the doc says: Change it, gehet hin und vermehret euch.
 
@Johannes_B No. I probably wouldn't…
 
7:01 PM
@clemens geometry is loaded in the class file, fancyhdr is loaded afterwards. Changing the margins in the document will lead to messed headers/footers.
That's eat it or leave it! vs. The fingerfood approach.
 
@Johannes_B we could upload a repaired version to CTAN and tell people to use to use that class instead…
 
@clemens Adding another template? People will find the template site first.
 
@Johannes_B I know. Just thinking aloud
 
I watched an episode of South Park in German. I found my new favourite thing.
 
@Johannes_B nice :)
@PauloCereda oh! Which one?
Today I re-discoved Jeeves and Wooster (by searching for recordings of Minnie the Moocher)
 
7:44 PM
@DavidCarlisle -- ugh. i think that's probably something that is "correct" in our in-house versions of the cls files, but not yet updated in the "public" versions. (this has been driving me out of my poor feeble mind for years.) styling of the toc and other lists was deferred for a long time (by a manager who is no longer here) because "there were no official written specs for implementing such lists", and everything was fixed in-house by hand until the fixers got fed up and insisted.
 
@clemens I heard their Dreidel song in German. Amazing. :)
 
8:18 PM
I just got my first downvote since 2013. That's a long time compared to other sites lol
 
yo'
@Alenanno happens even here. The downvote rate is about 1% :)
 
8:53 PM
@PauloCereda :) I didn't know that one!
 
9:22 PM
@DavidCarlisle Drat, you just beat me to it with the \output question
@DavidCarlisle Waiting now for the complaints on LaTeX-L
 
(The latter will be supported by David Carlisle and me rather than formally by the team as a whole.)
@DavidCarlisle ^^
 
@JosephWright message looks good thanks, we'll see what comments come in...
@PauloCereda you think I didn't know?
 
@DavidCarlisle :)
 
9:43 PM
@JosephWright just saw the announcement about the next latex release which reminded me: Could someone add \DeclareUnicodeCharacter{00A0}{~} to t1enc.dfu? The non-breaking space is getting in quite a number of documents by copy and paste.
 
@UlrikeFischer there's really a whole load of characters we ought to add, space being less of a premium these days. I'll see what the others say....
 
@DavidCarlisle But this is the one which frequently leads to errors. I had various other examples (hyphen, soft hyphen etc), but only isolated cases. 00A0 is always my first bet.
 
10:02 PM
@UlrikeFischer yes, you've got mail:-)
 
@DavidCarlisle Good ;-) I can look up tomorrow which declarations I added to various documents.
 
@UlrikeFischer I agree.
 
11:10 PM
@clemens: southpark.de All episodes in German!
 
@UlrikeFischer @Joseph @david Agreed. biblatex/biber is adding some hyphen or endash, can't remember right now. I was meaning to ask you to add it as well. Gotta find that thing.
 
@yo' Short answer: no :)
 

« first day (1761 days earlier)      last day (3178 days later) »