TeX, LaTeX and Friends

David Carlisle

7:34 AM

@PauloCereda important news

Paulo Cereda

7:54 AM

@PabloGonzálezL oh my

Joseph Wright

@DavidCarlisle, @UlrikeFischer I'm enjoying GitHub Project support ;)

Paulo Cereda

@DavidCarlisle yay

Later

@JosephWright Hello. Can I talk to you about something irrelevant to TEX?

Joseph Wright

Of course

Later

@JosephWright Thanks. According to the SE rules, can a user change completely one of their questions having no answer?

Joseph Wright

8:03 AM

@Later It would be better to simply delete, although if there are no comments or answrs it's somewhat academic

Later

@JosephWright Ok. It has some comments but no answer. I want to know whether the user do an illegal thing by changing completely a question or not; is there any SE policy or rule preventing the user from that?

Joseph Wright

10:09 AM

@UlrikeFischer Expect loads of PRs ;)

Paulo Cereda

@JosephWright oh no

Ulrike Fischer

@JosephWright just came in (I had to go to the vampires and leave my blood ...) and see the rush of emails. This color stuff is extending at a tremendious speed ;-). But I see a bit the danger that the discussion starts to mix "package level stuff" with "kernel interfaces".

Joseph Wright

@UlrikeFischer I'm going to do everything as PR, so there is proper review: I only want real kernel functions

Ulrike Fischer

@JosephWright I'm not worried about the code, only about the discussion -- it is so easy to drift away in all sort of directions ;-)

Paulo Cereda

10:25 AM

@UlrikeFischer ooh das Vampir

Joseph Wright

@UlrikeFischer That's one reason for splitting up into a small project with multiple focussed issues :)

Paulo Cereda

@JosephWright cli.github.com

Joseph Wright

11:50 AM

@PauloCereda I'm usually OK with GitKraken and the website, but I can see some people want lots of power (all the keyboard shortcuts, etc.)

@PauloCereda You usingit?

Paulo Cereda

@JosephWright will give it a go :)

@JosephWright ooh power

Diaa

1:09 PM

Hello everyone,

Until I get a response to my comment on this answer of Joseph https://tex.stackexchange.com/a/548560/2288, how can I pass and escape the following special characters to the command \getunicode{}?
# , \ , {

David Carlisle

@Diaa looking at the code \# \\ and \{ should work don't they?

cis

@Skillmonlikestopanswers.xyz

I have an idea or a (generalizing) suggestion:

I would also consider letting the parser access on an auxiliary file.
The auxiliary file, say `kana.dat`, could look like this:

a, U+3041, LaTeXFontIdentifier, Idon'tKnow,  Idon'tKnow, ... | あ (explanation)
i, U+3043, LaTeXFontIdentifier, Idon'tKnow,  Idon'tKnow, ... | ぃ (explanation)
u, U+3043, LaTeXFontIdentifier, Idon'tKnow,  Idon'tKnow, ... | ぅ (explanation)
.........

(A similar auxiliary auxiliary file `kanji.dat` could also exist.)

Why the whole thing?

Well, with this level of generalization, users could create more help files over time.
For example `thai.dat`:

k,  U+...,  LaTeXFontIdentifier, Idon'tKnow,  Idon'tKnow, ...| ก (explanation)
kh, U+..., LaTeXFontIdentifier, Idon'tKnow,  Idon'tKnow, ... | ข (explanation)
ng, U+..., LaTeXFontIdentifier, Idon'tKnow,  Idon'tKnow, ... | ง (explanation)
.........

I think you can see where it's going:
The package could provide a general parser for various Asian languages (Japanese, Thai, Chinese, ...).

@Skillmonlikestopanswers.xyz

Diaa

@DavidCarlisle Unfortunately, they don't.

David Carlisle

@cis at least for mapping from the code points to the unicode names the Unicode data files have the information to easily make such mappings eg: w3c.github.io/xml-entities/…

@Diaa hmm it would for the core part: \cs_new:Npn \__diaa_getunicode_auxii:n #1 { \int_to_Hex:n { `#1 } }

Diaa

@DavidCarlisle I am sorry for not understanding what you mean. Should the code be edited to escape all the special characters? Or I have to input them different way!

David Carlisle

1:25 PM

@Diaa I mean this works, so it would just need a bit of adjustments on the initial tests so a command of that form is passed through:

\diaa_getunicode:n{a} ~

\__diaa_getunicode_auxii:n{\#} ~

\__diaa_getunicode_auxii:n{\{} ~

\__diaa_getunicode_auxii:n{\\}

@Diaa because for any character you can use \int_to_Hex:n {`\?} where you replace ? by that character

cis

@DavidCarlisle Unfortunately, I don't know enough about it. So this is how I would build a package (if I could ... which I cannot):

As general as possible!

The principle of @Skillmonlikestopanswers.xyz parser should not be significantly different with regard to various other Asian languages.

David Carlisle

@cis but what input or output do you want the above page (that I wrote in javascript) takes any input text and gives the unicode number (hex) the unicode name and any tex or html names for the characters, see the output from w3c.github.io/xml-entities/…

Diaa

@DavidCarlisle As a beginner user of latex, if my life depends on understanding what is going on, I would have been completely dead X|). I would be greatly thankful if you don't mind providing an enhanced version of the command \getunicode syntax that escapes all the special characters. Thanks for your consideration.

David Carlisle

@Diaa it's not clear what your use case is or what you want the input to be you could use the form with \ for all characters there is no need to special case anything. \__diaa_getunicode_auxii:n{\A} works fine.

Diaa

1:47 PM

Thanks for your patience, generally speaking, I just need to use, for example, \getunicode{\} to have the output of 5C or use \getunicode{#} to have the output 23.
So, what should be modified in the code to have what I need?

I sincerely apologize for not understanding what you are trying to explain to me.

David Carlisle

define \let\getunicode\__diaa_getunicode_auxii:n then you can use \getunicode{\\}

cis

@DavidCarlisle Ui, https://w3c.github.io/xml-entities/unicode-names.html?%E3%81%82%E3%81%83%E3%81%85

So as I said, I would try to include something general in the package.

@Skillmonlikestopanswers.xyz parser works in the form

In: `\command{<romanization (a, i, u, ..., ka, ki, ...)>}`
Out: Symbol-Output.

that should be transferable to other Asian languages (if necessary).
For example, other users could create appropriate help files.

That's an idea of mine that came to me last night. Of course, I don't know how difficult it is.

Joseph Wright

@DavidCarlisle The more complex definition is there just to allow for the various expansion cases ...

David Carlisle

@JosephWright yes but I only read the last line and told @Diaa that \# would work but it doesn't survive the tests (it could be fixed but if the only requirement is looking up a single character you only need that one line so I was asking...

Ulrike Fischer

@Diaa but for which end do you want to get the unicode? And what is the range of the input? Getting the unicode from something like "A" is easy but if you add commands like \beta to the lot then it gets quite complicated - in pdftex \beta maps to a glyph position that has no connection to the unicode value.

David Carlisle

1:57 PM

@Diaa you are not being clear enough on your use case for example \getunicode{\} it would be possible to make getunicode read its argument verbatim so that worked rather than use \getunicode{\\} as I suggested but then (a) like any verbatim command you would not be able to use \getunicode in the argument of another macro and (b) \getunicode{\beta} would not work as \beta would be read verbatim not expanded to a beta.

Diaa

@UlrikeFischer I didn't think about the range of the input. I thought about a lookup table where an input of any sort would return its respective or linked unicode number.

Ulrike Fischer

@Diaa but how should a lookup table work if a tex input can contain \blub and \zzz?

David Carlisle

@Diaa that isn't a usable specification:-) what does "any sort" mean? do you want \getunicode{abc} to do anything? or \getunicode{\textbf{x}} or ...

Diaa

I mean a table like this utf8-chartable.de/unicode-utf8-table.pl?names=2&utf8=0x

David Carlisle

@Diaa but you don't need a table tex is already treating characters as numbers.

Ulrike Fischer

2:02 PM

@Diaa but this table doesn't contain $\leq$ and \beta as in your question.

David Carlisle

@Diaa it sounds like the table that I just referenced for cis, w3c.github.io/xml-entities/…

Diaa

@UlrikeFischer this one does johndcook.com/blog/math_symbols

Ulrike Fischer

@Diaa but this contains only around twenty entries. So what is your input range?

David Carlisle

@Diaa yes but that just has a few simple cases that map to a single unicode character but what about tex comands that expand to ooalign constructs or multiple character or fractions or ...

Diaa

@UlrikeFischer Alphanumeric, opening/closing brackets, slashes, math symbols, greek letters as far as I remember

@DavidCarlisle didn't think about this

yo'

2:07 PM

@PauloCereda Hi, you helped us with some ABNT templates couple months ago. Please do you remember whether you put them in the Overleaf Gallery?

David Carlisle

@Diaa that isn't a usable spec in classic tex \rightarrow is a single character, \longrightarrow is a - and some negative spacing and then a rightarrow, what do you want to do if given \longraightarrow?? . If you restrict to luatex and xetex unicode-math then more characters use unicode slots but not all

Diaa

@DavidCarlisle You pointed out the potholes of usage I completely overlooked. I got what you are trying to warn about

David Carlisle

@Diaa what is your actual use case why are you trying to get these hex values?

Diaa

I use the Redacted Script font https://github.com/christiannaths/Redacted-Font to dummify my document. So, this font lack some text/math characters. Therefore, I need to find the unicode of those characters to put them in the range of this font like this

\setmainfont{redacted-script-regular.ttf}

\setmathfont{redacted-script-regular.ttf}%
[range = {up , it , bb , bbit , scr , cal , bfcal , frak , tt , sfup , sfit , bfup , bfit , bfscr , bffrak , bfsfup , bfsfit, "002B, "002D, "2796, "207B , "03B7 , "03BC , "03D5 , "03C1, "2264, "3D, "2212}]

Paulo Cereda

@yo' Was there an Overleaf Gallery? :)

yo'

2:19 PM

@PauloCereda Yeah, the Gallery's existed for years (one of the issues that the Gallery has :-) )

David Carlisle

@Diaa huh? why not just look up in fontforge to get the list of characters in the font? why do you need mappings from tex names?

Paulo Cereda

@yo' ooh :)

Diaa

@DavidCarlisle I don't know how to do what you say. I just tried what first came to my mind

David Carlisle

@Diaa but if I understand you correctly all you need is a list of ranges that the font supports to put in to the range= list, but how would your \getunicode question help with that, it appears to be going in the wrong direction.

Diaa

@DavidCarlisle For example, \leq is not supported or recognized by the Redacted font, so I need first to know its unicode to put it in the range list

cis

2:25 PM

Hey.... @github project (which I was invited)....

How to 'post' there a suggestion?

'Create new file'? No...

Rmano

@cis open an issue?

David Carlisle

@Diaa no that's backwards, just get the range of characters in the font from the font. You don't want to go through three thousand tex math commands one by one looking up their unicode number. Also are these for anonymising documents printed on paper (as it doesn't do much to hide the text in pdf)

Ulrike Fischer

@Diaa ? The list should contain chars the font support, not chars it doesn't have.

cis

@Rmano O.ke.ye! I found 'issues' --- thx.

Diaa

Yes, I need to completely anonymize the document.
In order to understand you properly, how to anonymize \leq by replacing it with the Redacted font?
Did I do it wrong knowing that without putting the unicode of \leq in the range, it will be printed?

cis

2:35 PM

@Skillmonlikestopanswers.xyz I found
`Skillmon has invited you to collaborate on the Skillmon/ltx_quack repository`
and posted an 'issue'. (I hope I did everything right.)

Ulrike Fischer

@Diaa well the pdf will not be anonym. Try to copy & paste.

Diaa

2:51 PM

@UlrikeFischer I already did before, and without "2264 in the range of \setmathfont{redacted-script-regular.ttf}, ≤ would be printed in the output PDF

Ulrike Fischer

@Diaa the problem is not the print. Did you try to copy & paste some of the text?

Joseph Wright

@UlrikeFischer I'd appreciate your thoughts on github.com/latex3/latex3/issues/742, in particular do we want to go fully-expandable?

Diaa

@UlrikeFischer I got your point :) For me, I use it for printing the exam using a public printer (I actually haven't yet :), and I hope this fake document doesn't come out with the real text

Paulo Cereda

3:07 PM

@UlrikeFischer blub updated. :) Nothing major, I just commented the code. :)

@barbarabeeton ooh wombats and roos twitter.com/dodo/status/1270068074549596164

Paulo Cereda

3:31 PM

GUYS

TOMORROW

2020-06-10: TeX - LaTeX Stack Exchange

/quacks in despair

We are doomed

DOOMED, I SAY

Joseph Wright

@PauloCereda Huh?

Paulo Cereda

348

Q: We're switching to CommonMark

I’m Ham and I’m a developer on the Teams team here at Stack Overflow. Over the past few months, I’ve been heads down working on the way we turn Markdown into HTML when writing and editing posts across the network. I’d love to share what I’ve come up with. In a nutshell: We’re planning to use Comm...