« first day (5225 days earlier)    last day (1 day later) » 

cfr
cfr
02:00
@DavidCarlisle is it really normal to use marks? is there any actual advantage to doing so?
 
4 hours later…
cfr
cfr
05:56
@JosephWright is there some reason l3build doesn't normalise line numbers generally, but only on line? I'm wondering if there is some reason changes shouldn't be ignored?
06:15
@cfr We try to get all those we can without zapping anything that might not be 'ignorable' - it's tricky - what example are you thinking of?
06:55
@DavidCarlisle The textcase package documentation, footnote 1 alises instead of aliases. Quite romantic, though :D
07:17
So, after having studied `lthooks`, `ltproperties` and parts of `ltoutput`, I am still not quite at a point where I understand what could be the reason for the following issue

If I put `\property_record:nn` in `\AtEndDocument`, it won't put anything to the `.aux`. Now, `lthooks` says that `\AddToHook{enddocument/afterlastpage}` should be used to add to the `.aux` file as this is called after all remaining floats or whatsnot have been processed. But ... in my case this still does not work. Only `\AddToHook{shipout/lastpage}` ... I tried to come up with a MWE, but it will still take some tim
It works now somehow, but I wonder what could be the reason for this. Maybe someone has an idea? If not, I'll continue dissecting the code today ...
07:35
Could a two-column layout be the culprit?
07:49
Which document class are you using? In the past, I had a nothing to aux problem in koma, but not in standard.
Gooooood morning,
@MaestroGlanz Good morning =) I use report, so should not be an issue
I have a thing here. You might know Edwardian Scrtipt, which is nice and curly and has a kerning, that hardly needs correction. Unfortunately, it is not free, only free of charge. I stumbled across Monte Carlo, which is really nice and has a lot of decoration features. I would be willing to create a wrapper here and submit the result to ctan, but.....
what is the policy here? Do we have enough fonts, is there a need, a standard?
The thing is, that some features must be accessed by weird unicode, where a cs would be helpful.
:67272286 You though, you had something and you didn't...?
@MaestroGlanz You can upload anything that you are entitled to to CTAN (so e.g. TeX support for commercial fonts) - but TeX Live won't include stuff that is not license-free - both the code and any dependencies
I got it! I use \afterpage{\enlargethispage{-10mm}} (in twocolumn). This somehow inferferes with \property_record:nn.
It's OFL. So it's good. I only redistribute OFL. Because, you know, license mess...
07:58
And I think it is because afterpage redefines \enddocument. So stuff is output after \AddToHook{enddocument/afterlastpage}. And this is the answer. Which leads me to teh question: shouldn't afterpage rather use a hook to output things?
Then I might create a curated fonts package soon. Requirements will be: Unique design but usable, free license, good kerning, wrapped symbols = compatibility with standard keyboard under babel, compatibility with an example document for example text. Because creating only one package per font, will create a mess imo.
@JasperHabicht We are creating a carpet here. One thread interwoven with the other one.
@MaestroGlanz That's not unusual for the chat ... =)
I just want to gather a bit of input for ideas here before starting and creating a chaos. But start slow than thrice.
@JasperHabicht One for @DavidCarlisle, but yes to me it looks like this needs updating
cfr
cfr
@MaestroGlanz you mean monte carlo is ofl? edwardian script is not, as you said. to be included in tl, the fonts have to be acceptable to linux distros such as debian. you can create a ctan package which includes several smaller packages. e.g. I have a package which is basically a bunch of tiny packages providing macros to access decorative initials, which were already on ctan. but if the fonts are substantial in size, consider separate packages so on-the-fly installing isn't too painful ;).
@JosephWright right now, I'm just adding empty lines to the test files to maintain the numbering, but e.g.
Overfull \hbox (5.64543pt too wide) in paragraph at lines 27--27
Overfull \hbox (22.33553pt too wide) in paragraph at lines 30--30
Overfull \hbox (33.63544pt too wide) in paragraph at lines 33--33
Overfull \hbox (41.11562pt too wide) in paragraph at lines 36--36
Overfull \hbox (13.74544pt too wide) in paragraph at lines 39--39
08:12
@JasperHabicht no not really. The hook is executed (add \show\x to it to see that), the problem is that the write command is lost. The problem is basically the same as this one which doesn't write the label either:
cfr
cfr
if the overfull box is precisely the same, I can't see a reason to care.
\documentclass{article}
\begin{document}
\label{where}
\write1000{Hallo}

\begin{table}
blub
\end{table}
\end{document}
cfr
cfr
@UlrikeFischer does latexdef work for you on miktex without specifying the format?
@JasperHabicht so your problem with afterpage is because the label is issued on a page that is then thrown away again and so never shipped out. It will reappear if you do e.g. \afterpage{x\enlargethispage{-3cm}}.
\documentclass{article}
\usepackage{afterpage}
\begin{document}
\AddToHook{enddocument}{\show\x\label{blub}}
xxx\afterpage{\enlargethispage{3cm}}

\end{document}
@UlrikeFischer Ah you are right! I tried with \iow_now:Nn and this wrote to the .aux.
@UlrikeFischer Right. It happens only if the right column remains empty.
08:21
@UlrikeFischer the problem with afterpage is using afterpage
@DavidCarlisle Well I only use it to shift up the lower margin of both columns on the first page. This can probably be achieved in a way without using afterpage ...
@CarLaTeX ah I'll fix in the sources thanks
@DavidCarlisle You're welcome!
@JasperHabicht but it doesn't support twocolumn at all?
@DavidCarlisle Well, that's what I learned today =3 I thought that \enlargethispage{-10mm}\afterpage{\enlargethispage{-10mm}} could work (and it does, except for the problem now)
08:24
@CarLaTeX oh that footnote (which answers a question @UlrikeFischer asked me the other day, she clearly never reads my documentation:-)
Mar 26, 2012 at 19:37, by David Carlisle
@Canageek moral of the story: never read the documentation, bad things happen
@JasperHabicht it's "don't read footnote's in David's documentation day"
\footnote{This is really a pre-release, to see whether
% people like the idea of a command like this. This implementation is
% \emph{not} particularly robust. This implementation does not work in
% two column mode, and can get `confused' by \LaTeX's floating
% environments.}
@JasperHabicht if anything goes wrong when you load afterpage, it's a documented feature
@DavidCarlisle Well, I remember that you warned me back when I tried to implement this in the first place ...
@DavidCarlisle one of my answers lately in the german group said: "Dazu hätte ich in die Doku statt in den Code schauen müssen, und David sagt immer, dann tut man nicht ;-)"
@UlrikeFischer very good advice
08:32
@cfr Yes. Which package are you referring to? I might have a look into that.
@cfr no, it works on texlive but not in miktex. miktex has Version 1.9 -- 2020/09/27 while texlive has already Version 1.10 -- 2025/02/17. You will have to wait there a few weeks. miktex generally lags behind.
@cfr well all the standard classes are set up to use marks for section headings, and setting the page head mid document is like accessing the page number, it's mostly unsupported although you can do it with care at a forced page break so if you only change the page had at \chapter you could directly change the page head, but if the head changes at section headings without a forced break then marks are the only possibility
@JasperHabicht if you have a small one column afterpage example that's losing data from the hooks could you make an issue in the main latex2e repo, thanks
@DavidCarlisle I will do.
And then I need to think about how to either hook into the start of the second column (to insert \enlargethispage there ... or how to restore the geometry on the next page ... mabe using a pagestyle ...
Other thing: I'm working on another package (again). It is for creating booklets for events. In German it is Festfolge, Ablauf, Liedheft. Like these small booklets for church weddings, where they print the songs and prayers. My question: What do you call these? I haven't found any English term yet, that would please me.

All that stuff, which I'm doing right now, is somehow connected. Including the font stuff.
@MaestroGlanz "order of service" ?
08:45
@DavidCarlisle iff there is actually a problem in a onecolumn layout at all ... that is
(which I should doubt as you are the creator of the package ... but I have a MWE now ... so will file an issue)
@JasperHabicht If "someone" added hooks to latex's enddocument some decades after I wrote afterpage and broke it, then that is possible, but clearly not my fault:-)
@DavidCarlisle I don't blame you! =)
@DavidCarlisle Would be strongly restricted to church events. I also think, it sound a but weird as a package name. Right now I favor the German term "festfolge" and the anglo saxonian world has to live with it. Yes, I pretend, that a lot of people will use it. It keeps me motivated and gives me a deceptive sense of importance.
@MaestroGlanz yes weddings and funerals, you wouldn't normally use that term for more secular events
09:46
@MaestroGlanz Would be a 'programme' for a secular event
Hello! With a document where no page has headers I just saw the warning "Package scrlayer-scrpage Warning: \headheight to low. ..." Beneath other things I think an o is missing.
I tried to report the missing o at sourceforge.net/p/koma-script/tickets . But there I only get the sourceforge logo and a message from Cloudflare that it is checked whether I am a human. (Apparently it is difficult to consider me as a human being, because otherwise at some point something else would be displayed...)
@MaestroGlanz "[...] gives me a deceptive sense of importance" :D You mean to say all our packages don't have many millions of users?
@JosephWright I was going to suggest that but then I thought all those m would confuse @barbarabeeton
@DavidCarlisle :)
10:06
@DavidCarlisle that's most likely because there's a mandatory argument?
10:21
Ah, I think, I just wait for build/column/before being implemented as hook in about ... two weeks?
@JasperHabicht I need to check with FMi - this week I hope in dev
This would probably be exactly what I need ... if I understand correctly, it hooks into the start of the (next) column, right?
@JasperHabicht yes although unlike afterpage it's an output routine level thing so you can write out information r affect the column layout but not (I think) insert new material into the main document flow, as afterpage does.
10:36
@DavidCarlisle Yes, I understand this
11:13
@DavidCarlisle But it should do what I want to do
\documentclass{article}
\usepackage{lipsum}

\makeatletter
\NewHook{build/column/before}
\let\orig@makecol\@makecol

\def\@makecol{%
    \UseHook{build/column/before}%
    \orig@makecol%
}
\makeatother

\begin{document}
\twocolumn
\enlargethispage{-20mm}
\AddToHookNext{build/column/before}{\enlargethispage{-20mm}}

\lipsum
\lipsum

\end{document}
@JasperHabicht er I'd have to think about it but using enlargethispage there looks weird, it's adding an \insert so you'd need to re-read that bit of the texbook 3 times to work out what happens if you add an insert node at that point in the output routine
Well, I have not found any better way to set the layout on the first page without needing \clearpage
@JasperHabicht what are you trying to do? you can't really change teh size of a column, as it needs to be the same size as th eother column, are you just adding some space at the bottom?
Yes, I want to increase the margin at the bottom on the first page without inserting a page break.
I tried with page styles and geometry ... this was the best I could come up with
And: I need to be able to change this on the page I am currently on.
So, this makes things tricky
11:55
I don't think that "just" is a good way to describe what I want to do ... what could work (but It feels strange) is \begin{figure}[b]\vspace*{20mm}\end{figure}.
No, can't work due to twocolumn ... sigh I'll just ask on the main site. Maybe someone has a good ides
12:25
@JasperHabicht I tried that (with nidanfloat and dblfloatfix) but they didn't work (in different ways) You could probably use \twocolumn[\vspace{20mm}] (which puts the space at the top then move the text bolck in one of the new hooks, biut I'm not sure I really want to say that is better
@DavidCarlisle Maybe I can just change \textheight?
(But I use geometry ... it is probably not good to interfere here)
@DavidCarlisle But \AddToHookNext{build/column/before}{\vspace*{-20mm}} would increase the size of the column again, because at this time, the column box has not been boxified. (Which is why I thought that inserting \enlargepage would not be totally wrong ...)
In the doc of ltoutput it even says later in the code at \UseSocket {build/column/outputbox}: "Then we deal with any \enlargethispage or run the normal code to build a column." ... But maybe something else it meant here ... e_e
 
2 hours later…
cfr
cfr
14:17
@DavidCarlisle maybe I just write odd documents. generally, I only want the content to change at a page break and marks always seemed too complicated.
@UlrikeFischer thanks, but that isn't what I meant. there's a question from a miktex user with updated perl. that gets the expected warnings, just as on linux, but it also fails to find a latex defn if latexdef is used without --format given explicitly. I just wondered if it was expected to work.
@MaestroGlanz ctan.org/pkg/cfr-initials there is not much to look at.
@cfr no, as I wrote: with miktex it doesn't work. latexdef section gives "undefined"
cfr
cfr
@UlrikeFischer oh.
 
1 hour later…
15:33
@cfr Even better. Then I will be finished soon. It's only about the concept.
What's an fd-file? I guess font definition. Can you tell me something about it?
16:04
@MaestroGlanz texdoc fntguide
@MaestroGlanz there are a load of them in latex/base directory
16:50
Whoever thought that surrounding hyperref links in coloured boxes was a good idea?
17:01
@AlanMunn dont you want your documents to scream "I was made by hyperref with default settings" ?
@DavidCarlisle :)
@AlanMunn But I suppose the answer to your question is "Adobe" in that it is more or less the default pdf link rendering from acrobat, the boxes are not added by tex,
@DavidCarlisle Oh really? I had no idea. It's been so long since I used any actual Adobe products.
@AlanMunn well all pdf readers do the same, but the boxes come from the pdf spec (and the earliest acrobat versions) they are not drawn by TeX rules. You need to set the border properties on the link annotation to 0 to stop the pdf reader drawing them,
@DavidCarlisle Yes, I know how to get rid of them (this image is from a paper I'm reading, not one of mine). I've never really thought about their actual source.
17:17
@AlanMunn This was pre-html (or at least before it was mainstream) and people were not really used to the idea of hypertext and inline links, so having a default display that is somewhat painfully obvious possibly made more sense then than it does now.
@DavidCarlisle I suppose so. An interesting history lesson.
 
1 hour later…
cfr
cfr
18:24
@MaestroGlanz something like this:
%% Based on a file created using fontinst v1.928

%% THIS FILE SHOULD BE PUT IN A TEX INPUTS DIRECTORY

\ProvidesFileSVN{$Id: ebgaramond-maths.dtx 10819 2025-02-19 05:45:52Z cfrees $}[v1.3 \revinfo][OMLEBGaramond-Maths.fd: font definitions for OML/EBGaramond-Maths.]

\DeclareFontFamily{OML}{EBGaramond-Maths}{}

\DeclareFontShape{OML}{EBGaramond-Maths}{m}{it}{
<-> EBGaramond-Italic--oml-ebgaramond
}{}
\DeclareFontShape{OML}{EBGaramond-Maths}{sb}{it}{
<-> EBGaramond-SemiBoldItalic--oml-ebgaramond
@MaestroGlanz but if you are looking to support type1 fonts for e.g. pdftex, you probably want to use one or more of the various fools available to help generate the support files. depending on which tool you use, you may or may not need to write your fd files by hand.
 
2 hours later…
20:07
@cfr you are referring to @egreg ?
2
 
1 hour later…
cfr
cfr
21:28
@DavidCarlisle is @egreg a delicious summer dessert made from fruit and cream?
cfr
cfr
22:14
 ....\mathon
 ....\OML/cmm/m/it/12 a
-....\OT1/cmr/m/n/12 (
+....\OT1/cmr/m/n/12 (-20) (
 ....\OML/cmm/m/it/12 b
 ....\glue(\medmuskip) 2.66663 plus 1.33331 minus 2.66663
-....\OT1/cmr/m/n/12 +
+....\OT1/cmr/m/n/12 (-20) +
 ....\penalty 700
 ....\glue(\medmuskip) 2.66663 plus 1.33331 minus 2.66663
 ....\OML/cmm/m/it/12 c
-....\OT1/cmr/m/n/12 )
+....\OT1/cmr/m/n/12 (-20) )
 ....\glue(\thickmuskip) 3.33328 plus 3.33328
-....\OT1/cmr/m/n/12 =
+....\OT1/cmr/m/n/12 (-20) =
 ....\penalty 500
-....\glue(\rightskip) 0.0
@DavidCarlisle why are the penalties different in places?
22:49
@cfr binoppenalty is 700, relpenalty is 500, is that what you mean?

« first day (5225 days earlier)    last day (1 day later) »