6:25 AM
I was just looking something up in the multicol source code and came across its license header. There it says:
%%   The producer of a proprietary and commercially market product
%%   that involves typesetting using multicol is asked to determine
%%   the value of a license fee for using multicol if
%%
%%   - the product is a document and the producer has decided to
%%     include multicol to typeset (parts of) the document or has
%%     directed the author of the document to include multicol (for
%%     example, by providing a class file to be used by the author)
%%
%%   - the product is a LaTeX class or package that includes multicol
Isn't this a violation of the LPPL?

@HenriMenke It's a 'moral' obligation

@JosephWright But then this whole (sort of threatening) legal mumbo-jumbo can be replaced by “Please donate if you find this useful”
@JosephWright On another note, I have built a script to extract all the codeexample environment from the PGF documentation, so that each example can be passed to l3build individually. github.com/pgf-tikz/pgf/blob/master/doc/generic/pgf/extract.lua

6:46 AM
@HenriMenke most of that wording is older than LPPL and originally was a non-optional part of the licence under which multicol was distributed.

@HenriMenke Oooh, wowo
@HenriMenke I'll probably look at the l3build stuff at the weekend ... I'm still trying to push PDF stuff at the moment

7:04 AM
hopefully…

7:58 AM

@UlrikeFischer I think I'm going to 'pick off' a few easy PDF things: info, catalog, etc. I wonder about \pdfinfo: should this be handled with individual key-value pairs or a simple list
@UlrikeFischer Actually, it's clear it has to be: dvips needs the keyvals with some pre-processing

'ello

8:14 AM
@JosephWright imho everything that at the end is a dictionary is at best handled with properties/key-val unless we find a strong argument not to do it. I have a simple \driver_pdf_info:n in the dtx, which I used as \driver_pdf_info:n { /Author(\@pdfauthor) /Title(\@pdftitle)...} in the driver. But only because I wasn't sure if I can fill the fields without changing hyperref.sty (and disturbing other packages like pdfx and hyperxmp). I can look at it again in the afternoon.

@UlrikeFischer I've just had a look and have some code in hand; wait for check-in, probably over my lunch break :)
@UlrikeFischer At the driver level, I don't think we need a prop: that's probably something for the \pdf_... module, as it's common to all drivers
@UlrikeFischer BTW, I think we can ignore \pdfinfoomitdate as you get the same effect setting the two dates to a blank string
@UlrikeFischer I'll need a little while for dvips, as I'll probably have to generalise a bit the \pdfmark code I've borrowed ... at present, it's entirely focussed on making links

@JosephWright yes, one can do something like \driver_pdf_info:x{<expand some prop>}.
@JosephWright I think we should open more issues in the repo to keep track of such topics.

@UlrikeFischer Exactly
@UlrikeFischer OK, go for it
@UlrikeFischer I'll log that one with an example in a bit

8:32 AM
@JosephWright should I label them?

8:56 AM
@UlrikeFischer Go for it

9:20 AM
@DavidCarlisle The modern trend is for documentation to be online /only/ – at Read the Docs for example. We hates it!

9:42 AM
Hi, there was a trick to prevent lists or objects like this to start directly after an environment.
It happens as the disalignement in the case of itemize.
The command was something like escape vertical space...
OK, an appropriate search re-revealed the solution :)
It was \leavevmode

Hi, I want to polish up my question tex.stackexchange.com/questions/487529/… Do you have any suggestions?

10:32 AM
@marmot Here is how I envision this to be used:
\documentclass[a4paper]{article}
\usepackage{pgfkeys}

\pgfkeys{%
/dlf/.is family,%
/dlf,%
A/.initial=,
B/.initial=,
C/.initial=,
set default/.code={
\pgfkeys{/dlf,
A=#1,
B=#1,
C=#1,
}
},
set default/.default=\textbf,
set default,
my value/.initial=,
}

\newcommand\apply[2]{\pgfkeysvalueof{/dlf/#1}{\pgfkeysvalueof{/dlf/#2}}}

\pgfkeys{/dlf/.cd,
my value=test,
B=\textit,
C=\MakeUppercase,
% set default=\textsf,
}

\begin{document}

\apply{A}{my value}

\apply{B}{my value}

\apply{C}{my value}
It works, but the nesting of \pgfkeys just seems strange to me.
I'm playing with a beamer theme, where a lot needs to be in uppercase, and I want this is be individually configurable (frame titles in upper case takes too much space in most cases), just seems easier to use a method like this to set a lot of default values.

11:16 AM
@marmot ahh, you are absolutely right, .forward to seems the way to go. It is just not obvious from the manual, that you can specify this several times on the same parent key. Source: tex.stackexchange.com/q/368736/3929

11:45 AM
Some pgfplots expert currently here? @marmot? I am looking for creating a "stacked" plot: 3 plots, one on top of the other which share the same abscissa (time axis, in my case), to be shown only on the lowermost plot. Any pointers where to look for an example in the pgfplots manual?

@AlexG groupplots?

@JosephWright Thank you! Going to look for it.

@AlexG I'll send you a demo

@JosephWright oh no
@JosephWright 👿

@PauloCereda demo, not d[a]emon I hope.

11:49 AM
@AlexG phew :)

@AlexG Example on it's way

@JosephWright Thank you so much!

@DavidCarlisle 𝓷𝓸𝓽 𝓶𝔂 𝓯𝓪𝓾𝓵𝓽
Unicode is amazing

@PauloCereda we can blame @barbarabeeton for those

@DavidCarlisle ooh

11:56 AM
@JosephWright Perfect example! Exactly what I need, even the font fits! (I need it for my boss, to be pasted into ppt.)

@AlexG oh no

@AlexG Good: its from a paper a couple of years ago

12:11 PM
@egreg thief

@DavidCarlisle oh no

@PauloCereda tempting people with flashy _: tricks and personally named modules

@DavidCarlisle where is this world going
In other news, will any of you boys and girls watch Endgame?

Is there a way to put caption like a), b) of this type in the figures (using caption command)
3

Use positioning library. right = of horizontally aligns nodes by their center: \documentclass{standalone} \usepackage{tikz} \usetikzlibrary{positioning} \begin{document} \begin{tikzpicture}[every node/.style={inner sep=0pt, outer sep=0pt}] \node (img1) {\includegraphics[height=2cm]{example-...

@Shamina I simply use minipages. :)

12:20 PM
@PauloCereda I agree with you. But some how I have made a collage along with the coordinate axes, however, now I need a caption below each fig

12:31 PM
@JosephWright Hello Joseph! Can you please turn this answer into a comment?

12:58 PM
@daleif You do not need to nest the \pgfkeys, the following seems to yield the same result:
\documentclass[a4paper]{article}
\usepackage{pgfkeys}

\pgfkeys{%
/dlf/.is family,%
/dlf/.cd,%
A/.initial=,
B/.initial=,
C/.initial=,
set default/.style={/dlf/.cd,
A=#1,
B=#1,
C=#1,
},
set default/.default=\textbf,
set default,
my value/.initial=,
}

\newcommand\apply[2]{\pgfkeysvalueof{/dlf/#1}{\pgfkeysvalueof{/dlf/#2}}}

\pgfkeys{/dlf/.cd,
my value=test,
B=\textit,
C=\MakeUppercase,
% set default=\textsf,
However, I am not sure I understand the purpose.

1:24 PM
@marmot your hint with the forward to, also works.

@daleif Yes, but I am wondering if you are overall just looking for /.append code.

As for the purpose, I have a beamer template, that is modeled after a power point template. It has \MakeTextUppercase all over the place. Instead of hard coding these., I'd like to make them adjustable. So in this case A would be a formatter for the name of the presenter in the footer, B would be the formatter for the frame title.

> Daniel Craig is back in a sequel from the director of True Detective
bah
He does not look English. :)

Some of these formatters I'm going to leave blank by default (fx the frame title), but if the user wanted to fully comply to the PP formatting, they could issue use all caps as a configuration key for this beamer theme, and it would set all the appropriate formatters to run uppercase.

@daleif Couldn't you just have one key, say text formatter/.code={\textbf} to which you append locally text formatter/.append code={\MakeTextUppercase}? (Not that I tried this... ;-)

1:31 PM
Pip pip tally ho cheerio ta ta
Happy Penguin day!

@DavidCarlisle Which theft are you referring to? There have been a few, lately. :-P

@egreg the width one, replacing my elegant use of measuring the width of an alignment with some dull expl3 loop:-)

@DavidCarlisle Mathematicians are fond of generalizations. :-)

I got my tick back.

@marmot how do you propagate this to say 10 different keys? that can also be set on their own.

1:46 PM
2
weeeeeee

@daleif Are these variations in their local groups? If so, you can append these extra options locally. If not, you may have to reset these things.

2:07 PM
@UlrikeFischer Be careful. Ticks carry all kinds of scary diseases.

@HaraldHanche-Olsen Which ones? minor or major? x ticks or y ticks?

@marmot depends on what you mean by local group? The keys are meant to be set once (in the preamble) (I'm experimenting with providing all the configuration for using this beamer theme as a key-val interface, so no confusing \author[short]{long} etc

@HaraldHanche-Olsen I'm quite safe, I assume that @egreg will generalize his answer and take it back ;-)

@daleif These A, B and C are then just styles that you set?

2:19 PM
@marmot they are just values, here macros that can take up to one argument, hence my example with \textbf. (I'm having a bit of trouble with the concept of style here)

2:57 PM
@daleif What I want to say is that it seems to me that what you are after is conceptually the same as wanting a node to have the styles, say, circle,draw,inner sep=3pt. You want some text or whatever have the properties \MakeUpperCase and \bfseries, say, so I am wondering if you might be better off if you just tried to copy the concept from the nodes and define styles bold,upper case and so on that you apply to these texts.

3:16 PM
@marmot might be. I think I'll keep working on the features and then settle on the interface later on. Somepeople might be getting upset that I'm ditching \author, \title \institution etc for a key-val based interface. So we'll see whether the interface in the end will be better than the standard interface.

@daleif Better than powerpoint? I bet it will be. ;-)

@marmot Well we have some freedom here (I've been putting off for year to actually make this theme, and in the mean time they changed to a much much simpler PP theme). I'm more thinking about beamer users, how they are going to feel about \author and \title perhaps not doing anything.

@daleif I do not know the details of it at all, so I cannot say much.

@marmot some of it comes from the confusing of \title[short]{long}. Because shortis normally in the footer, this can easily something completely different than long. So why not split these completely and give them key names that make more sense to the users. At least that is the end goal.
long could be the title of your talk at a conference, and short be the name of the conference to go into the footer.

1 hour later…
4:50 PM
Hello geektexs,
I'm in trouble with this code:
   \begin{verbatim}
length=2*3.14*r
print(length)
#this is a comment
\end{verbatim}
In a beamer documentclass
Is there anything wrong?

@santimirandarp fragile option?

Do you mean to insert fragile option? I'm not sure what's that @JosephWright

@santimirandarp Yes: \begin{frame}[fragile]

fine, it works. I'll check whats that @JosephWright
Last question, is it ok if \setlength{\itemsep}{neededsep} is set on the preamble?

@santimirandarp it won't work. itemsep is reset at the begin of lists. Try \setlength\itemsep{3cm}.

5:05 PM
Yes, it doesn't...

@PauloCereda -- Yup. The creators of the Brown Corpus were among my professors, and I actually keypunched part of it. It was one of the very first applications of computers to linguistics, and is still recognized as an important source. (Henry Kucera, one of the creators, taught the only formal programming course I ever took; the language was PL/1, now nearly unknown, but very interesting in terms of its design principles.)

@barbarabeeton How nice! I know the Brown corpus and (surprisingly) PL/1 as well! :)

5:21 PM
@AlexG -- You need to check out tug.org/TUGboat/tb36-3/tb114cartoon.pdf
@DavidCarlisle -- The only problem is, I can't see anything but little empty square boxes.

5:44 PM
@UlrikeFischer Thinking about \pdfdest: we have to do some processing for drivers other than pdfTeX, so have to work out how to handle the optional zoom ...

6:04 PM
@UlrikeFischer Or do we just rule that out: I can only see support in pdfTeX, not dvipdfmx or dvips

@UlrikeFischer Hmm, so we do need the zoom ...

@JosephWright dvips and dvipdmx don't allow to set it?
@JosephWright in pdfmark.def I can find a \@pdfviewparams which perhaps (!) means that it supports the zoom, in hxetex is doesn't looks as if it is there. If the drivers have different capabilities we could perhaps collect them in (driver) dependant tables.

6:28 PM
@barbarabeeton Too bad. Those are letters from the MATHEMATICAL BOLD SCRIPT SMALL set of letters. The declared fonts are, in order, Verdana, Arial, sans-serif. But I am beginning to suspect that my browser is actually reaching for the STIX Two Math font to render those glyphs.

@barbarabeeton you have been using TeX too long, that is the tfm view of a font.

@UlrikeFischer Probably I'll go for an interface that supports zoom then deal with lack-of-support at the low level

@JosephWright sounds good.

6:43 PM
@UlrikeFischer Go for xyz for 'no zoom', xzy<number> for scaling?
@UlrikeFischer Or xyz [zoom]? Not sure I like having an optional value at the driver level. I thought about xyz:<zoom>

@JosephWright do you think xzy in pdfpages is by purpose? I thought it is a typo.

@UlrikeFischer Hmm, yes, could well be a typo ....
@UlrikeFischer So xyz<zoom> or xyz:<zoom>?
@UlrikeFischer Or do we just have an extra argument that's only used for xyz?

@JosephWright or simply beside fit etc a number which is then used for xyz.

@UlrikeFischer Yes, that did occur to me too
@UlrikeFischer Out of 100? So 100 = xyz

7:04 PM
@JosephWright well 100 as 100% would be actually quite good, it took me some time to figure out that one has to use 1000 in pdfpages to get 100 ;-)

@UlrikeFischer Yes, don't plan to use 1000

@DavidCarlisle -- Not exactly. The only "reliable" device I currently have access to is one without a tex installation, and an out-of-date operating system. It's a long story.

@DavidCarlisle -- If I could figure out how to cut and paste what's there on my screen, you would also see little empty boxes.

@barbarabeeton oh something like this I assume 🪓🪓🪓🪓🪓🪓🪓🪓

7:27 PM
@DavidCarlisle -- Not sure whether how this will show here, but it's what a copy/paste results in: @DavidCarlisle \303\260\303\260\303\260 \303\260\303\260 \303\260\303\260\303\260\303\260\303\260 How's that? (Not very helpful.)

@barbarabeeton octal? very retro:-)

@DavidCarlisle -- I told you this machine was out of date. Got nothing to do with either tfm or tex.

7:41 PM
@barbarabeeton Hi! Got any response regarding the theorem format?

\cs_new_protected:Npn \driver_pdf_destination:nn #1#2
{
\cs_if_exist:NTF \tex_pdfextension:D
{ \exp_not:N \tex_pdfextension:D dest ~ }
{ \exp_not:N \tex_pdfdest:D }
name {#1}
\exp_not:N \str_case:nnF {#2}
{
{ xyz }   { xyz }
{ fit }   { fit }
{ fitb }  { fitb }
{ fitbh } { fitbh }
{ fitbv } { fitbv }
{ fith }  { fith }
{ fitv }  { fitv }
}
{ xyz ~ zoom \fp_eval:n { #2 * 10 } }
@UlrikeFischer ^^^
@UlrikeFischer Look OK>

@mickep -- I've had no feedback from my former colleagues. Please remind me of the particular question, and I'll try again.

@JosephWright yes, clever.

@UlrikeFischer Need it for dvipdfmx, picking up from hyperref
@UlrikeFischer Need to work out how the zoom works in XeTeX ...

@barbarabeeton It was about the style of theorems, that has changed during the years. Previously they were indented and set in small caps, now (in the style guide) the head is set bold, and not indented.

@JosephWright if I go the docu right, simply the array is passed, so perhaps some argument [@dest /XYZ @xpos @ypos 1.0]

@UlrikeFischer Yes, I'm just testing
@UlrikeFischer I think something like this!
@UlrikeFischer The form hyperref uses is [@thispage /XYZ @xpos @ypos null], so I just need to work out by hand
@UlrikeFischer Where did you find docs ...

@JosephWright I looked in dvipdfm.pdf

@mickep -- I've sent another inquiry. I do doubt that there is anything useful recorded, so it will depend on the memory of someone else who was in the editorial department at the time. Will report if I hear anything.

@barbarabeeton Much appreciated.

8:04 PM
@UlrikeFischer Yup, it's a scale in dvipdfmx. I wonder why hyperref doesn't support it ... at least that I can see

@AlanMunn -- Hmmm. It seems I'm strengthening these statistics when I "English" some submissions to TUGboat. (This often consists of adding "the" or "a"; articles are not required in many languages, and picking up "standard" English usage is apparently not exactly automatic.)

@JosephWright I didn't find anything too. Do one need the keywork null is there is no zoom, or does 0 works too?

@UlrikeFischer Good question: I suspect not, as null means 'do nothing' whereas 0 is a fixed zoom

@JosephWright I think 0 = null (looking at pdfmark spec) (which isn't directly involved but i assume has same meaning)

@DavidCarlisle Yeah, seems to be
@DavidCarlisle Don't need to worry, I think ....
@UlrikeFischer, @DavidCarlisle I think one can do zoom in dvips

8:15 PM
@JosephWright can't you in principle write any pdfmark for dvips?

@DavidCarlisle Yes, looks that way; I think the hyperref dvips driver like the dvipdfmx one just skips the idea

@JosephWright the pdf reference says "A zoom value of 0 has the same meaning as a null val".

@UlrikeFischer Ah, right
@UlrikeFischer, @DavidCarlisle I have pdfTeX/LuaTeX/XeTeX sorted, just dvips to do ...
@UlrikeFischer Not sure it's that important ...

@JosephWright running history in reverse

@DavidCarlisle The PostScript stuff is by far the most complex to sort out; I have to get it all working. With pdfTeX, I just need the primitive (OK, without that much documentation!)

8:19 PM
@JosephWright I think last time I thought about this it was with discussions with Sebastian and probably predated pdftex:-)

@DavidCarlisle Yes, I suspect so

@DavidCarlisle Kind David when you have a little of time look this: cdn2.scuolabook.it/Uploaded/loescher_0376e_preview/…

@Sebastiano You don't have permission to access /Uploaded/loescher_0376e_preview/loescher_0376e_preview.pdf on this server.
@UlrikeFischer seen the greek luatex ligatures question?

@DavidCarlisle ok I just wanted to show you a pdf that is a preview of a publishing house available online.

@Sebastiano as I showed, I have no permission to access it.

8:35 PM
@DavidCarlisle yes, will look at it now.

@Sebastiano but how can you tell from a pdf that it was made with latex and indesign with figures made in illustrator? especially pdfs that are not readable, or do you have password access for that?
@Sebastiano if it is the same as this one (found by google) scuolabook.it/Uploaded/loescher_0376e_preview/… that is made with QuarkExpress
@Sebastiano also the question was about Pearson and Wiley and I asked how you could tell how their books were made using tex, I don't see how linking to a book made by a different publisher using QuarkExpress really answers that?

@UlrikeFischer that was quick:-)

8:51 PM
@UlrikeFischer yes I got that far (seeing the link didn't work) and then gave up and asked you google suggested some other places to download it but most of those font sites i don't trust:-)

@DavidCarlisle I just showed you that Loescher's book is out of production written in LaTeX and modified with other programs.
@DavidCarlisle 2nd question Fondamenti di fisica di David Halliday, Robert Resnick, Jearl Walker before are of the company Zanichelli.

@Sebastiano no you didn't you just pointed to a non public URL that was unreadable, I found (the same?0 pdf vis google, and can tell it's made with QuarkExpress, how can you tell it started as latex, do you know the authors or can you tell from the pdf, that was the question i originally asked.
@Sebastiano a link to what? to loescher_0376e_preview.pdf ? as I say I found that via google already.

@DavidCarlisle Directly from the site I have copiated this: Le basi di partenza per i nostri elaborati con formule sono il TeX, ideato da Knuth, e LaTeX, di Lamport, affiancati dai più potenti programmi per l'impaginazione tradizionale. You can see that there is also Pearson.
@DavidCarlisle Do you now believe me? The book LinX Pearson Walker of Physics that I adopt is made with QuarkXpress. I have the email from the publisher.

@Sebastiano yep I just translated it at the link you gave, but I's say it doesn't really answer the question on the main site. a specialist math typesetter did enough to put the pearson logo on their site, that could mean anything, perhaps they worked on one book, or 1000 or ... You stated that pearson and wiley produced there mathematical books starting with tex and then touching up in indesign and quarkexpress, and using illustrator for the figures. it seems that was a guess.
@Sebastiano as I thought, your original statement is just a guess and you don't have inside knowledge of a tex->indesign->illustrator workflow inside pearson or wiley.

9:07 PM
@DavidCarlisle And I confirm. The most famous publishing houses use other smaller publishing houses to create the texts.

@Sebastiano I give up.

@AlanMunn Thank you!

@PauloCereda Hi, Paolo

@Sebastiano Do you know that? Have you worked there? If so, you are probably breaking an NDA
@Sebastiano These are big places creating 1000s of books: does just one count? 10% of their output? ... It's far to open a question

@JosephWright I didn't work on it. On the covers of the school books there are the references of the publishing house. I'm not violating anything. If you want, I'll take a picture of you and show you that the graphic processing is done that way.

9:16 PM
@Sebastiano hi.

@Sebastiano your comment under the question is at best misleading and probably wrong, you should delete it.

@Sebastiano All they tell you normally is 'Typeset by Company X of Place Y'

@PauloCereda I got today a very serious looking poster with a footnote saying Förderung: Bundesamt für magische Wesen ;-)

@JosephWright it also says who is in charge of the graphic processing of the company.

@UlrikeFischer ooh :)

9:19 PM
@Sebastiano Not the academic books I'm familiar with (OUP, largely, but also other people)

@DavidCarlisle Can you tell me exactly what I'm supposed to eliminate?

@Sebastiano do you have any knowledge of a pearson book that was started as tex, modified in indesign and with figures generated in illustrator? if you have not delete the comment saying that is how pearson produce mathematical books.

@JosephWright I think this page answers the question that Wiley does not use LaTeX: The template will not be used for the presentation of the final page layout.

@DavidCarlisle I've deleted the comments. See if they are correct. I personally have never worked in a publishing house but I have had some clarifications via email also from school textbooks. Many Physics schoolbooks, for example, such as Walker or Halliday have been translated into Italian and reset by publishers with the same characteristics as academic books.

@StrongBad Could be, or could be they have a strange workflow. For example, I had to submit a chapter using Word, but the proofs were from LaTeX. They could be doing LaTeX/Word -> XML - > TeX ouptut for all we know

9:29 PM
I could simply forge the PDF metadata and typeset with a potato? :)

@JosephWright Ah, I thought "typed" meant "keyed" and not outputted to the printed page such that TeX had to be the first stage of the process.

@Sebastiano the question has been closed anyway but you should also delete your comment starting "The answer is simple: Adobe InDesign, Illustrato..." as it is unrelated to the question about Pearson and Wiley so misleading in that context.

@JosephWright This is LuaTeX, Version 1.10.0 (TeX Live 2019/W32TeX) (format=lualatex-expl3 2019.4.25).
         ^^^^^^^^

@StrongBad I just wanted to make my knowledge available. But am I wrong by some reason?

@UlrikeFischer does anything work? :-)

9:31 PM
@DavidCarlisle ooh

@UlrikeFischer Hi
@DavidCarlisle I have deleted all.

@Sebastiano thanks

@DavidCarlisle a test document that loads fontspec and siunitx didn't explode.

@UlrikeFischer :-)

@UlrikeFischer drat, since @JosephWright is a chemist, it should explode. :)

9:32 PM
@UlrikeFischer who cares about those packages? does picture mode work?
3

@Sebastiano well according to Wiley's webpage, if you submit your book with their LaTeX template, they still do things to it so it is not possible to submit CRC in LaTeX. As @JosephWright points out, it is possible TeX plays a role someplace in their process.

@DavidCarlisle if you give me an example I will try.

78

It was suggested in chat http://chat.stackexchange.com/transcript/message/9482087#9482087 That picture mode would be the ideal tool for the job here: \documentclass{article} \begin{document} \begin{picture}(200,100) \put(30,40){\line(1,0){150}} \put(30,40){\line(0,1){60}} \put(30,100){\line(...

@StrongBad @JosephWright is what I was going to say :-(. They then change everything but the initial setting can also be done with LaTeX. A few years ago I loaded modern computer fonts and with adobe illustrator, for example, with images you can do what you want.

@Sebastiano the question was not about what is possible or what you have done, it was about what Pearson and Wiley do, that's why there was surprise at your definite statements, as most people have no knowledge of their internal production processes.

9:38 PM
@DavidCarlisle It compiled fine - both with pdflatex-expl3 and lualatex-expl3 - I'm very relieved.

@Sebastiano My point is that unless you are actually involved in the work, it's hard to be sure of the route(s) used. So the question on the main site is probably OT anyway: we can't answer.

@UlrikeFischer l3draw version next I suppose....

@DavidCarlisle @JosephWright In any case, I have removed all comments.

@barbarabeeton Yes, the English article system is deceptively difficult for speakers of languages without articles.

@JosephWright What do you mean OT?

9:40 PM
@Sebastiano Off Topic
@UlrikeFischer, @DavidCarlisle As with other stuff, it seems dvips is by far the most awkward for \pdfdest: there seems to be some need to measure stuff in PostScript again ...

Greetings, everyone.

@JosephWright simple just reverse engineer pstricks as well as pgf:-)

@DavidCarlisle ooh

10:05 PM
@JosephWright what do you need to measure for pdfdest?

@DavidCarlisle Trying to work that out now ...
@DavidCarlisle I think it's to allow you to know where the destination actually is
@DavidCarlisle Ah, was reading the wrong part of the log

10:23 PM
@DavidCarlisle What I'm not getting is why in dvips we seem to be measuring something either side of some input (text), whereas for the other routes it's a single point. It would be a lot easier just to measure everything in one shot. I suspect that should be doable. Need to understand the PostScript.

@JosephWright yes a single point was what I was expecting, hence the question about what you were measuring, although I should probably look at the code....

10:37 PM
@DavidCarlisle I think I get the basic idea: pretty sure I don't need two points
@DavidCarlisle I'd appreciate thoughts on H.S and H.A in hdvips.def. They are adding/subtracting 72 converted to points to things ... but shouldn't it be 72.27? Have I found a bug in hyperref here?