« first day (3253 days earlier)      last day (1688 days later) » 

1:32 AM
@JosephWright yep google just warned me:-) (I think I had it listed from when we set it up)
 
 
4 hours later…
5:12 AM
@DavidCarlisle Princess Beatrice marries an Italian because she wants to keep the EU citizenship! (P.S. = pizza, even the good one, is not a plate served at wedding parties).
 
5:27 AM
@DavidCarlisle course, dish not plate, sorry for my English
 
 
2 hours later…
7:07 AM
@CarLaTeX The English was OK, more worrying was your astounding lack of knowledge about Italian food!
 
7:40 AM
Looking at the FAQ page, I wonder if @texfaq should be on Twitter: 'Hint of the day' or something (@PauloCereda ;)
@DavidCarlisle I was thinking about the FAQ: I wonder if we should just be honest and say 'managed by the LaTeX team and friends'
 
@JosephWright why not, certainly it makes no sense to say it's managed by uktug
 
8:02 AM
@JosephWright does that mean that I will have to learn how twitter works? ;-)
 
@DavidCarlisle Yes, exactly
@UlrikeFischer No, just that I'll need to work out how to script something to post things
Just a passing thought, when I was looking the hit numbers from Google
@DavidCarlisle Nothing from HO
 
@JosephWright but I would have to check what you post.
 
@DavidCarlisle It doesn't: I said something like 'a group of curators'
 
@JosephWright What is a "Google Webmaster list"?
 
Was also thinking that the design makes my blog look a bit ropey ... I should find someone to do it properly
@UlrikeFischer You can get info on what search terms they find for your site: you show that you actually own/control it, then you get to see search hits, page issues, etc.
Zoomed up:
 
8:16 AM
@JosephWright what does "Impressions" mean?
 
@UlrikeFischer 'Appears in the search results Google provide'
 
@JosephWright got it.
 
@UlrikeFischer Not that we really care, as we are not trying to make money or whatever
@UlrikeFischer Mainly interesting to see top search terms:
Same thing fro my blog:
 
@JosephWright looks as if appendix and underscore and outer par mode are the most pressing problems.
 
@UlrikeFischer Yes
 
8:41 AM
@JosephWright a bit surprising. Not what I would have guessed ;-)
 
@JosephWright yes but there are still a gazillion references to the uktug faq on the internet, not to mention ctan.org/tex-archive/usergrps/uktug/faq?lang=en
@JosephWright notice that it doesn't like the design of our top page, fails its mobile use test as links too close
@JosephWright not exactly a surprise, I have a few bits and pieces that I want to check in, I should have time at the weekend to get it into a reasonable state then move it across. (@UlrikeFischer)
 
@DavidCarlisle I've asked for removal
@DavidCarlisle I'd seen: I guess we should fix
@DavidCarlisle Moving is easy: it all redirects at GitHub
 
@JosephWright I think it's just the vertical spaces in the toc, so a line or two of css should be enough.
 
@DavidCarlisle Cool
@DavidCarlisle Like I said, I wonder about looking for a little design help ...
 
@JosephWright yes I know, but I wanted to move it in some reasonable state (and adjust the manual and copyright lines etc at the same time)
 
8:57 AM
@DavidCarlisle OK, doesn't have to happen before a move ...
@DavidCarlisle I'll leave it to you then :)
 
@DavidCarlisle I wanted to look at issue 100 and 102 later today. But beside this it is fine for me.
 
@JosephWright no but I think it makes the history look clearer if the edits to the status in the manual are more or less contemporary with the actual move
 
@JosephWright regarding redirects. Shouldn't we archiv the older luaotfload and lualibs repos at github.com/lualatex?
 
@UlrikeFischer I am at the company AGM all day (in the hotel lounge on the free wifi waiting fro it to start:-) so can't do anything before the weekend.
 
@UlrikeFischer Yes
 
9:01 AM
@DavidCarlisle I had to look up what "AGM" is ;-)
 
@DavidCarlisle Sure
 
@DavidCarlisle :P
 
@UlrikeFischer Ideally they'd have been moved, but we missed that
@DavidCarlisle I've cleaned up the FAQ copy on GitHub (the one that was under the uktug account)
@UlrikeFischer Done
 
@JosephWright ? You mean when I move them to my github? Yes, but at the time I didn't dare to do something so drastic. I wasn't sure if it would really work out in the end.
 
@UlrikeFischer Yeah, something like then, but it was missed and we don't want to loose data. So archiving makes sense
 
9:41 AM
@egreg This is probably for you tex.stackexchange.com/q/509975/3929
\detokenize seems to work
 
 
4 hours later…
1:23 PM
Can anyone give me details about how PDFTeX writes on PDFs? Using TikZ, for example. And how does this relate to the PDF standard? Is it a separate layer? I know that when PDFTeX writes on the PDF, it does not interfere with what is already there. And can apparently be easily removed by other tools, though I have not tried to do so.
Also, I notice that PDFTeX can write on JPEGs. Is that the same mechanism, or a different one?
 
1:56 PM
@FaheemMitha Huh?
@FaheemMitha pdfTeX creates the entire PDF from scratch: I'm not sure what you mean about 'already there'
 
@JosephWright and it's all hamburger? :)
 
Does anyone know what the dimension that specifies a \framebox(,){} rule thickness is? It appears NOT to be \fboxrule.
 
@StevenB.Segletes \framebox seems to rely on \frame, so it should be \@wholewidth
 
@PhelypeOleinik Thank you so much!!
 
@StevenB.Segletes You're welcome :-)
 
2:16 PM
user image
2
 
2:40 PM
@UlrikeFischer Sehr spassig!
@UlrikeFischer What leaks from my grease gun? Tool ooze!
 
@StevenB.Segletes the resident linguist should be pleased!
 
2:56 PM
@JosephWright I'm talking about writing on an existing PDF file.
In case that wasn't clear.
 
@FaheemMitha The PDF is always completely re-written
 
@JosephWright Oh. But the original PDF and the new stuff is separate, regardless.
 
@JosephWright in case you mean writing over a graphic input with e.g. \includegraphics: that is not different to writing over text or something else.
 
@UlrikeFischer I think you meant to address that to me.
 
@FaheemMitha yes, sorry.
 
3:07 PM
It appears me (perhaps incorrectly) tha the original PDF (rewritten or not) and whatever PDFTeX annotates on it are separate in some sense. But perhaps that is not true.
Also, why does PDFTeX rewrite the PDF?
 
3:21 PM
@FaheemMitha could you please try to bit a more precise? E.g. show some code, or screenshots or whatever that makes clear what you are meaning?
2
 
4:18 PM
@FaheemMitha it is completely unclear what you are talking about what do you mean by "write on" ?
 
 
3 hours later…
6:51 PM
@DavidCarlisle I'm talking about what \includepdf from the pdfpages package does.
It includes a PDF, but you can also write on it, say with TikZ, like so:
\node [font=\bfseries, align=center] at (12.4,-17.7){...}
So, overall something like:
 
@FaheemMitha it's a wrapper around \includegraphics, it seems.
 
\includepdf[pagecommand=
{\begin{tikzpicture}[remember picture, overlay]
    % \draw (-2,-24) to[grid with coordinates, major help lines/.append style=LawnGreen, minor help lines/.append style=LawnGreen] (16,0);
 \end{tikzpicture}}
,pages=3]{\filename}
@PauloCereda What is?
 
@FaheemMitha \includepdf
 
@PauloCereda Oh, is it?
 
$ texdef -t latex -p pdfpages includepdf

\includepdf:
macro:->\@protected@testopt \includepdf \\includepdf {}


\\includepdf:
macro:[#1]#2->\begingroup \let \AM@threadname \relax \def \AM@tmp {pdfpages} \expandafter \AM@split@options \expandafter \AM@tmp \expandafter {\the \AM@global@opts ,#1} \edef \AM@temp {{pdfpages}{\the \@temptokena }}\expandafter \setkeys \AM@temp \ifthenelse {\boolean {AM@pkg@draft} \and \boolean {AM@survey}}{\let \AM@currentdocname \relax \renewcommand \includegraphics [2][]{Survey in draft-mode}\def \AM@pagecount {0}}{\AM@findfile {#2}\if \AM@threadname \relax \de
 
6:55 PM
@PauloCereda T@@ m@ny @s :-)
 
@PhelypeOleinik @@h
 
@PauloCereda That looks like aah. Or maybe AAh.
@PauloCereda Huh. And all those \foos are part of the definition?
 
@FaheemMitha 🆈🅴🆂
 
Yes, I'm seeing the same thing here. \foo? Really? How unimaginative.
 
@FaheemMitha @n '@' c@n r@pl@c@ m@ny l@tt@rs. Th@ m@ss@g@ r@m@ins cl@@r @s d@y
 
7:00 PM
@PhelypeOleinik Yikes.
 
@PhelypeOleinik good old MSN times. :)
 
@PhelypeOleinik So, how is it going with being a TeX developer?
 
@PauloCereda I didn't use that much. At home we only had dial up internet, so my fun was mostly MS paint :-)
 
@PhelypeOleinik :)
 
@FaheemMitha I'm not developing much currently. The deadline to my dissertation is approaching so I have to focus on that a bit more.
@FaheemMitha But I like it :-)
 
7:09 PM
@PhelypeOleinik I see.
 
@FaheemMitha (probably because I wasn't blamed for anything so far)
 
@PhelypeOleinik Ok. Exciting times?
@PhelypeOleinik It's only a matter of time. :-)
 
@FaheemMitha the "writing with tikz" and the including of the pdf are unrelated
 
@FaheemMitha The Secret Plan not much :-)
@FaheemMitha I know
 
@DavidCarlisle They're unrelated? Ok.
@PhelypeOleinik Keeping secrets must be stressful. If you want to tell us, we'll quite understand. :-)
 
7:12 PM
@FaheemMitha why would you think they are any more related that if you typeset an x an dthen do a tikz picture or typeset two tiks pictures?
@PhelypeOleinik would it be helpful if we nudged you every so often to keep you on track as we did for @PauloCereda chat.stackexchange.com/…
 
@FaheemMitha It's quite fun, actually. One day everything will be revealed
 
@DavidCarlisle Well, both are done using one command. But in any case, my question was whether the included PDF and the stuff written (which then becomes part of the PDF) are separated in some way, from the internal PDF structure POV.
 
@DavidCarlisle I'm afraid to answer that question...
 
This question is vague because I don't really know anything about a PDF's internal structure.
 
@FaheemMitha no it was vague as you hadn't mentioned includepdf or tikz:-)
 
7:15 PM
@PhelypeOleinik I expect that shortly David will be telling people to blame you instead of Ulrike. Or possibly as well as.
The more the better, after all.
 
@PhelypeOleinik if in doubt, blame @FaheemMitha
 
@DavidCarlisle Oh dear.
 
@FaheemMitha I allow @DavidCarlisle to blame me once I have my dissertation finished
 
@PhelypeOleinik That's very generous.
 
@PhelypeOleinik ooh
 
7:17 PM
@FaheemMitha That does not remove your blame, though :-)
 
@DavidCarlisle Well, I think it's still a vague question. But opinions may vary.
@PhelypeOleinik Bugger.
 
@PhelypeOleinik that's why @PauloCereda is never blamed for anything
 
@DavidCarlisle yay
oh wait
 
@FaheemMitha well the image is an xobject: /XObject << /Im4 7 0 R >> and referenced in the page stream as /Im4 Do. The tikz code consists of lots of drawing instructions in the page stream (but you could store it in an xobject too). But what will you do with the info?
 
@UlrikeFischer Nothing. I just wondered how it worked.
Now looking up xobject.
@UlrikeFischer You could store it as an xobject, meaning, the TikZ code could?
 
7:28 PM
@FaheemMitha you could use the xsavebox package.
 
@UlrikeFischer To write on a PDF?
 
@FaheemMitha that is not relevant to write over something else. Every drawing instruction can do it.
 
@UlrikeFischer Can do what?
I wonder if I can get such a tree-like view without using some Adobe product.
Well, or other proprietary software.
 
@DavidCarlisle The hyperref manual claims that showkeys only works if hyperref is loaded implicit=false but I don't think that this still true.
 
xsavebox appears to be relatively new.
2016-02-25
* initial release (v0.1) to CTAN
 
8:01 PM
@UlrikeFischer hmm I don't think I changed showkeys...
 
@DavidCarlisle then perhaps hyperref, I found the line % Support for package |showkeys|.
 
@UlrikeFischer @UlrikeFischer yes \@ifpackageloaded{showkeys}{% looks like it's doing something sensible. Just remove the comment in the manual?
 
cis
Hello my friends in the USA and the United Kingdom!

I have a question:
Can you confirm that this is the right song text for the following song?

https://www.youtube.com/watch?v=8PkXIO3kkl0
Like the wind,

You came here runnin'

Take the consequence forever.

There's no space,

There's no tomorrow,

There's no safe communication.



Check it in, check it out,

And the sun will never shine.

They're a long way away,

In the subways of your mind.



Like the wind,

You're gonna suffer/somewhere,

Let us drown in your own failure.



There's no place,

And there's no sorrow,

In the young and restless dreamer.



Check it in, turn it down,

And the sun will never shine.

They're a long way away,
 
8:17 PM
@DavidCarlisle yes, I think so. I'm only wondering if one should remove the whole item. As hyperref redefines \ref etc in any case (and with AtBeginDocument), loading it with implicit=false will probably not help.
 
cis
It could be that he sings 'Time to win' instead of 'Like the wind' .....
I am sorry, I had to post the link again, so that you can see the nice picture (question above) ;)
 
@cis google suggests lots of people have asked about that:-)
 
cis
8:32 PM
@DavidCarlisle I know, I just wanted to ask the educational elite. ;)
 
 
2 hours later…
Bob
10:02 PM
There must be something fundamentally incorrect with this \newcommand{\varone}[1]{\gdef\@varone{#1}} but I cannot figure out why.
 
@Bob Seems good to me, why?
 
Bob
The error happens when you try to do it again.
 
@Bob When you use \varone again or when you use \newcommand{\varone} again?
 
Bob
@PhelypeOleinik Say you have in your preamble:

\newcommand{\varone}[1]{\gdef\@varone{#1}}
\newcommand{\vartwo}[1]{\gdef\@vartwo{#1}}

and in your document body

\varone{a variable}

Here is my \@variable it works fine.
 
@Bob Also, use the 'reply' feature of the chat so it's easier to follow the conversations.
@Bob Make sure you did \makeatletter before using commands with @ in their name...
 
Bob
10:12 PM
But if you do :

and in your document body

\varone{a variable}
\vartwo{another variable}

The code breaks
 
@Bob I see nothing wrong with that. Can you make a compilable example that shows the error?
 
Bob
@PhelypeOleinik I thought makeatletter makeatother was only for when referencing a @ that was inside a class or package?
 
@Bob No. TeX only allows you to type \<stuff> if all the characters in <stuff> are "letter" tokens. By default (in a document) the @ character is an "other" token, so \my@variable is the same as \my<space>@variable. \makeatletter stands for make the at character a letter token.
 
Bob
@PhelypeOleinik Here is my example to produce the error.

\documentclass[12pt]{report}
\usepackage[utf8]{inputenc}


\newcommand{\command}[1]{\gdef\@command{#1}}%
\newcommand{\anothercommand}[1]{\gdef\@anothercommand{#1}}%

\begin{document}

\command{some text}

\anothercommand{some more text}

\@command works.

\@anothercommand does not work.

\end{document}
 
@Bob \makeatletter will solve the problem here :)
 
Bob
10:21 PM
Ok I will see if I can figure out how to use them
 
@Bob The idea of needing \makeatletter and \makeatother is to make it less easy for the user to mess with LaTeX's internal macros, so they are "hidden" with the @ character. In theory you should not use @-commands in the middle of the document. You should do something like \newcommand\ValueOfVariable[1]{\csname @#1\endcsname} and then use \ValueOfVariable{command} instead of \@command.
@Bob Just put \makeatletter before the first \newcommand
 
Bob
@PhelypeOleinik Yes but doesnt that mean that the \makeatletter is active for the whole document? Isnt that dangerous?
 
@Bob \makeatletter is "active" until you do \makeatother: one reverses the effect of the other. Not dangerous, but not as it should be.
\documentclass[12pt]{report}
\usepackage[utf8]{inputenc}

\makeatletter
\newcommand{\command}[1]{\gdef\mypackage@command{#1}}%
\newcommand{\anothercommand}[1]{\gdef\mypackage@anothercommand{#1}}%
\newcommand{\MypackageValue}[1]{%
  \@ifundefined{mypackage@#1}%
    {\PackageError{mypackage}{Variable '#1' undefined.}{}}%
    {\csname mypackage@#1\endcsname}}
\makeatother

\begin{document}

\command{some text}

\anothercommand{some more text}

\MypackageValue{command} works.

\MypackageValue{anothercommand} works too.
@Bob ^^^
 
Bob
@PhelypeOleinik I see you wrote sort of a universal get command.
@PhelypeOleinik Is it necessary to have the words "mypackage" before the @? Or was that a stylistic choice to make things safer.
 
@Bob Not universal, just mypackage-versal :-)
@Bob Exactly. Imagine if all the hundreds of LaTeX packages decided to use \@macro to store something. Everything would be incompatible. The idea to use a prefix is to avoid these clashes.
 
Bob
10:35 PM
@PhelypeOleinik Ok great I understand now.
 
@Bob In expl3 we call this prefix the 'module' name, so that all code that belongs to a package has the same exclusive prefix and other packages are "forbidden" to use that (of course in TeX there's no way to enforce that, but we try :-).
 
Bob
@PhelypeOleinik Yes I remember a few years ago when I asked for help on something @egreg would repeatedly change all my variable names in his replies.
 
@Bob The main reason changes to the LaTeX kernel are so minimal is that a large number of packages, in one way or another, redefine some LaTeX internal command, and changing that breaks things. The expl3 approach of allowing only public interfaces to be used across modules making implementation changes much easier.
@Bob That's probably it :-)
 
Bob
@PhelypeOleinik In all honesty I asked this question in order to correct an error in an answer to a question found here. https://tex.stackexchange.com/questions/510046/input-variable-yields-undefined-control-sequence/510064#510064

I think you should post your answer.
 
@Bob There's no need. Just fix your answer and we're good ;-) (besides, I'm not at the computer right now)
 
Bob
10:50 PM
@PhelypeOleinik Ok I just wanted your permission first. Thanks!
 
@Bob Sure :D
 
11:17 PM
@PhelypeOleinik expl3 enforces using only the public interface?
 
@FaheemMitha As far as it's possible in TeX, it does, but nothing stops someone from shooting themselves on the foot ;-)
@FaheemMitha If you use l3doc's @@ feature, then it will complain if you're writing, for example, \__mypackage: code using \__otherpackage: functions.
@FaheemMitha But again, it just complains. Some people prefer to ignore warnings
\documentclass{l3doc}
\begin{document}
    \begin{macrocode}
%<@@=mypackage>
\cs_new:Npn \@@_somefunction:
  { \use:n { \__otherpackage_funcion: } }
%    \end{macrocode}
\end{document}
Package l3doc Warning: A control sequence of the form
(l3doc)                '...__otherpackage_funcion:' was used. It should only
(l3doc)                be used in the module 'otherpackage', not in
(l3doc)                'mypackage'.
@FaheemMitha ^^^
 
Bob
@FaheemMitha From what I understood and was able to drag out of egreg without him actually admitting it, there is no such thing as a local variable.
 
@Bob If you mean in TeX, I don't think TeX has variables. Period.
 
@Bob Not in the sense as other languages do, as in "local to a function". I think I saw the word "monolithic" being used to describe TeX.
 
@PhelypeOleinik I see. So it doesn't enforce it, in the sense of giving an error?
Honestly, I tend to ignore warnings too.
 
11:30 PM
@FaheemMitha No, that's as far as it goes (for now). Sometimes rules need to be bent, but the warning is there to make sure you think twice before doing so.
 
@PhelypeOleinik Ok.
 
@FaheemMitha My brain does not allow my logs to have warnings. It's annoying sometimes :P
 
@PhelypeOleinik Does that include overfull/underfull warnings?
 
@FaheemMitha My dissertation project doesn't have a single one :-)
 
@PhelypeOleinik Actually, I can relate. My OCD tends to kick in a lot, though much of the time I can keep it at bay.
@PhelypeOleinik I'm impressed.
 
Bob
11:32 PM
How do you even go about fixing overfull and underfull warnings?
 
@FaheemMitha More time spent on polishing text than actually writing it (I believe @PauloCereda can relate)
 
Bob
I have always just ignored them entirely
 
@Bob It depends. When it's not due to some stupid mistake then I rephrase the text, adjust page breaks and so on
@Bob It's not the most fun task, really
 
Bob
@PhelypeOleinik So you mean to say that overfull/underfull can be caused by box and glue issues because the words in your sentence cannot be put in place satisfactorily?
 
@Bob Usually, that's the case, yes. Mostly when you have long words near the first line break of a paragraph
 
Bob
11:36 PM
@PhelypeOleinik Huh. Back when I was just learning I could never figure out what caused them. Pretty quickly I just always ignored them.
 
@Bob Overfull boxes can be easily spotted by setting \overfullrule=5pt. Underfull boxes are a tad harder to find, but TeX does give you some clues
@Bob Oh, I just found this: tex.stackexchange.com/a/4146/134574. Pretty neat
 
@Bob you can usually tell just by looking at the output or increase \showboxdepth and \showboxwidth then tex will show the box contents in the log.
@FaheemMitha only by documenting that you should do that, you can never enforce anything in a macro language every document has full access to the internal implementation of every command, just if you use that power, things will break
111
A: Do I have to care about bad boxes?

David CarlisleIt is worth noting that TeX doesn't make the value judgements here. The user, or more likely, the class file on behalf of the user, has set constraints on the amount by which boxes may overflow, the amount of stretching allowed on short pages etc. TeX only warns if these user-set constraints are...

 
Bob
@PhelypeOleinik Oh cool
 

« first day (3253 days earlier)      last day (1688 days later) »