« first day (4256 days earlier)      last day (46 days later) » 

 
1 hour later…
6:53 AM
@DavidCarlisle Yes, I'm aware of that. The topic was solely about how to write my log message.
 
 
2 hours later…
8:30 AM
@FaheemMitha yes, so as I say I'd just call them macros, no need to explain tex's processing model in your change log.
@CarLaTeX you should try both and tell us which you prefer
 
9:14 AM
@DavidCarlisle italian have a special function: @CarLaTeX
user image
2
 
9:26 AM
@UlrikeFischer @DavidCarlisle I think @UlrikeFischer's function is the most suitable
 
9:43 AM
@DavidCarlisle :D That makes it much easier to find the example in the doc - nobody can miss them :)
 
@samcarter once I move \rowcolors from xcolor to colortbl I'm sure there will be a whole new era of publications using alternating primary coloured row backgrounds.
 
@DavidCarlisle Can you patch the \rowcolors macro to automatically add a note to the previous page saying "time to put your sun glasses on"?
 
10:34 AM
@samcarter ooh
 
10:47 AM
@samcarter oh goodness. we're going to go back to the first days of webpages. #seisurewarning 😂
 
@Plergux I used the web for some years before I had a colour screen
 
@DavidCarlisle oh
 
@PauloCereda that's amateur stuff. there is no garish, badly tiled background nor scrolling rainbow coloured text. 😝
 
@Plergux ooh
 
10:55 AM
@DavidCarlisle I was probably not born 😝
@PauloCereda apparently intentionally ugly web design is a thing now clearph.com/brutalist-web-design
 
@Plergux :P maybe some animation could be added, blinking borders for the table or something
 
@samcarter oh yes, that would be amazing. very shabby chic. 😂
 
@Plergux :)
 
11:26 AM
@Plergux You can always wait for Halloween and open CTAN...
 
@JosephWright I can't remember does l3color have a supported way of doing an urestored color change? this example isn't that clear but it's stacking up a lot of aftergroup'd restores....
 
 
2 hours later…
1:18 PM
user image
4
 
@AlanMunn what else can you expect from an author with the name Cook?
 
 
1 hour later…
@DavidCarlisle I saw it ;-)
 
 
1 hour later…
3:27 PM
I have a fontspec question.
I want to use "FONT ONE" as my main font for my document. But "FONT ONE" doesn't include bold and italics scripts. So to type text in bold and italics, I want to use "FONT TWO". How can I do that? I just scanned the fontspec manual and haven't such a use case.
 
@tush scan the manual again and look for keys like BoldFont.
 
@UlrikeFischer Thanks.\newfontfamily\englishfont{FONT ONE}[Script=Latin,Language=English,BoldFont=FONT TWO,ItalicFont=FONT THREE] works
 
 
1 hour later…
4:59 PM
@JosephWright vv
\label is supposed to be safe, as is \ref , \eqref could be made safe with \tl_put_right:Nn \l_text_case_exclude_arg_tl { \eqref } but caption package is not expecting this somewhere and things going wrong. looking...... — David Carlisle 35 secs ago
 
5:38 PM
@JosephWright I thought we had handled \label inside \MakeUppercase, but it errors for me:
\documentclass{article}
\begin{document}
abc \MakeUppercase{x\label{blub}x}
\end{document}
 
6:14 PM
@UlrikeFischer yes I'd got to similar point but had to stop and eat a duck
@UlrikeFischer it's safe in the l3 code but dies in the legacy \protected@edef
abc \MakeUppercase{x\protect\label{blub}x}
 
@DavidCarlisle yes. But it would be a nuisance if people have to \protect labels.
 
@UlrikeFischer why is label neither \protect nor \protected robust....
 
@DavidCarlisle no idea. I wondered about that too.
 
@UlrikeFischer team list ?
 
@DavidCarlisle yes I guess.
 
6:21 PM
@UlrikeFischer example has blub so you send first...
 
@DavidCarlisle ok. But wasn't it a good idea that I made \ref protected in the new latex?
 
@UlrikeFischer yes because it means we can blame you for every other command
 
6:51 PM
@DavidCarlisle I just found the \caption problem too ;-(. And naturally hyperref/nameref needs an adaption.
 
 
2 hours later…
9:04 PM
@DavidCarlisle Indeed: 'it's not my fault'
 
@JosephWright hmm, does code I just sent to the list look OK? although arguably \label, \ref etc should always make themselves safe in a protected@edef irrspective of use in makeuppercase
@JosephWright PL4 ?
 
@DavidCarlisle I guess
@DavidCarlisle Yup code looks good, but really ... \protected
 
@JosephWright most pure format uses are fine but the "textcase is obsolete" story is rather broken currently, and as I made textcase do nothing, can't really fix there
@JosephWright we could make \label \protected (probably) but wouldn't you still need to protect its argument from premature expansion (or possibly not...)
 
9:23 PM
@DavidCarlisle but it is imho not trivial. We would have to hunt down first which packages (beside hyperref/nameref) patch \label, caption is perhaps not the only one.
 
@UlrikeFischer yes making them safe (again) in MakeUppercase is a bit infficient but avoids issues with package redefinitions
 
9:41 PM
@DavidCarlisle and probably not only packages. I bet lots of documents patch \label.
 
@yo' bigl bigr not big (just got called out in a comment here:-) ... overleaf.com/learn/latex/…
2
 
@UlrikeFischer @DavidCarlisle's plan looks fine to me: it's essentially what the expl3 code is already doing inside \text_<thing>case:n
 
@JosephWright yes, looks fine to me too.
 
@JosephWright yes of course that's arguably what's wrong with the plan, doing it all twice, but avoiding that would require more extensive changes I suspect
 
@DavidCarlisle That's though really the fact that \MakeUppercase is non-expandable in the first place and/or the need to deal with non-UTF8 actives ...
 
10:01 PM
@JosephWright yes, but even if we only supported utf-8 makeuppercase has always edef expanded before uppercasing, would be nice if MakeUppercase was just a :e variant of the expl3 case changer, but not sure if we can get there
 
@DavidCarlisle Not sure an e-type variant makes a lot of difference here: the code already applies \text_expand:n, after all ...
 
@JosephWright ah so it does, Ok so if we dropped the \protected@edef you just lose the LICR handling/
 
@DavidCarlisle Yes exactly
@DavidCarlisle I suppose I might be able to find a way to do that, but it looks like pretty hard work
 
@JosephWright yes I guess I knew that once
 
@DavidCarlisle will you also add \eqref support?
 
10:13 PM
@UlrikeFischer may as well I think
 

« first day (4256 days earlier)      last day (46 days later) »