I sketched a a kind of code analyzer for (La)TeX documents. It can do things like pop up a little documentation or suggest completions:

It's a Lua script, so it could become a plugin for any text editor. If anybody wants to take a look, it's here: github.com/astoff/digestif

@HaraldHanche-Olsen ooh I like it

Moo

@PauloCereda steak dinner!

@DavidCarlisle oh no

@PauloCereda I did wonder if your "I like it" comment above was referring to the screenshot of an editor just above.

@DavidCarlisle heavens no, it's the edtior no one should use. :)

@MarcelKrüger do you have some time this week to discuss some luaotfload stuff?

Hi

@UlrikeFischer Certainly, I have some changes to discuss anyway.

@DavidCarlisle I'm looking at the floatrow package now. It hasn't been updated in about 9 years. Do you figure it would still be adaequate to get a row of figures with different aspect ratios adjusted to the same height and captions or should I be looking elsewhere?

@MarcelKrüger good. Perhaps better per mail? Or do you prefer on github?

@thymaro I know I updated it but I never really looked at its capabilities sorry, it was just an "emergency fix" for a change in the latex insert allocation.

@UlrikeFischer I think mail is the best solution, especially because my GitHub notifications seem to be broken.

@DavidCarlisle Oh, @h<tab> would do it, I suppose.

@MarcelKrüger mail sent.

@UlrikeFischer Just received the most beautiful gift wrapping paper in the world <3<3<3 I'll use it for all my Xmas gifts!

@UlrikeFischer Thank you so much!

@CarLaTeX finally ;-) I was getting worried - the time started to look brasilien like ;-). Glad that you like it ...

@UlrikeFischer I love it

@UlrikeFischer oh

@JosephWright @egreg wow not even checked out the files locally yet and already got an issue posted, how do people find these things:-)

@DavidCarlisle ooh

@DavidCarlisle ;-). And you are mad to invite me to something that needs the skill not to confuse left and right ;-).

@DavidCarlisle I took a heart and moved the testfiles I made to hyperref, merged the code for two issues and will now push the whole lot ...

@UlrikeFischer thanks, as we agreed on the call not to do the 2e changes that means I should have time to package them both for a ctan release this week.

@DavidCarlisle you could perhaps check if the testfiles run for you - this space business is still lurking ...

@UlrikeFischer hmm so far:

Running checks on
  25-sec-non-ascii (1/19)
          --> failed

  38-pdfminorversion (2/19)

oh they are all failing, I'll look at the diffs...

@DavidCarlisle ;-((

@UlrikeFischer PDF is very different, the first few chunks in the first diff:

*** ./build/test/25-sec-non-ascii.luatex.tpf	2018-11-27 13:56:25.182479600 +0000
--- ./build/test/25-sec-non-ascii.luatex.pdf	2018-11-27 13:56:30.162854200 +0000
***************
*** 1,54 ****
  %PDF-1.5
! %ÐÔÅØ
  1 0 obj
! <</S/GoTo/D (section.1)>>
  endobj
  4 0 obj
  (\376\377\000\045\000\341)
  endobj
  5 0 obj
! <</S/GoTo/D[6 0 R /Fit]>>
  endobj
  8 0 obj
! <</Length 128       >>
  stream
  BT
.....
--- 1,56 ----
  %PDF-1.5
! %ÌÕÁÔÅØÐÄÆ
  1 0 obj
! << /S /GoTo /D (section.1) >>
  endobj
  4 0 obj

@DavidCarlisle looks like new luatex?

@UlrikeFischer oops yes I'll revert again (I was testing @KhaledHosny's callback issue:-)

@DavidCarlisle I saw this ! %ÌÕÁÔÅØÐÄÆ already in another test and wondered about it. And the spaces in the pdf are different in the new version. That will give us some fun.

@UlrikeFischer first one passes:-)

@UlrikeFischer do you recognise this where do the pagebox=cropbox come from in my version

*** ./build/test/test0.luatex.tlg	2018-11-27 14:07:44.168474500 +0000
--- ./build/test/test0.luatex.log	2018-11-27 14:07:50.569562600 +0000
***************
*** 14,41 ****
  (hyperref)                than one, level fixed on input line ....
  Package hyperref Warning: Difference (2) between bookmark levels is greater
  (hyperref)                than one, level fixed on input line ....
! <hog.pdf, id=81, 251.94125pt x 213.79875pt>
  File: hog.pdf Graphic file (type pdf)
  <use hog.pdf>
! Package luatex.def Info: hog.pdf  used on input line ....

@UlrikeFischer ignore that, I found it

@UlrikeFischer "reproducible pdf glitch" ! /CreationDate (D:20160520110000+02'00')/ModDate (D:20160520110000+02'00')/Trapped/False/PTEX.FullBanner (This is LuaTeX, Version 1.07.0 (TeX Live 2018/W32TeX))>>

@UlrikeFischer I don't get /W32TeX although I suppose @JosephWright could normalize that....

@DavidCarlisle I forgot the config for the new xetex test in build.lua, and I found a forgotten temporary pdf in the testfiles. So will push in a few minutes again.

@DavidCarlisle the +02'00 is a problem too or not?

@UlrikeFischer yes

@UlrikeFischer and then of course to make it hard to normalize away you get a difference in the xref table ! 9236 although I suppose we could always delete the xref table

  Check failed with difference files
  - ./build/test/test8.luatex.pdf.diff
  - ./build/test/test8.pdftex.pdf.diff

@UlrikeFischer phew rtl issue closed.

@DavidCarlisle the fullbanner should be easy to suppress with \pdfvariable suppressoptionalinfo ..., I only wonder why it doesn't happen yet. @JosephWright

@DavidCarlisle and what is the failure?

@UlrikeFischer the full banner (pdftex and luatex both show differences)

@UlrikeFischer the full pdftex diff is:

*** ./build/test/test8.tpf	2018-11-27 14:13:20.357022100 +0000
--- ./build/test/test8.pdftex.pdf	2018-11-27 14:13:23.257608400 +0000
***************
*** 33,39 ****
  /CreationDate (D:20160520090000Z)
  /ModDate (D:20160520090000Z)
  /Trapped /False
! /PTEX.Fullbanner (This is pdfTeX, Version 3.14159265-2.6-1.40.19 (TeX Live 2018/W32TeX) kpathsea version 6.3.0)
  >>
  endobj
  2 0 obj
--- 33,39 ----
  /CreationDate (D:20160520090000Z)
  /ModDate (D:20160520090000Z)
  /Trapped /False
! /PTEX.Fullbanner (This is pdfTeX, Version 3.14159265-2.6-1.40.19 (TeX Live 2018) kpathsea version 6.3.0)

@DavidCarlisle blush. regression-test is missing ;-)

@DavidCarlisle I pushed the change.

  All checks passed

@UlrikeFischer ^^

@DavidCarlisle ;-). I didn't know which versions numbers need to be changed. I only adapted the one in hyperref.dtx.

@UlrikeFischer I'll check but that's probably all (it's not like the oberdiek bundle where if you change one you need to change another hundred or it complains) then I need to make sure the date in readme is at least the date of the last change (or I get told off by Petra:-)

@DavidCarlisle I wondered if Petra will complain if one updates with a new hyperref e.g. nameref.dtx with an unchanged version number.

@UlrikeFischer I don't think so: Ive been updating tools without changing the version of longtable for decades:-)

@UlrikeFischer, @DavidCarlisle Phew :)

@JosephWright you missed all the fun ;-)

@UlrikeFischer Grant proposal ... has to be done today

@UlrikeFischer I did do the team minutes

@JosephWright I saw it (and read it ...).

@DavidCarlisle tex.stackexchange.com/questions/462021/…

@JosephWright yes. @AlanMunn before you answer that, check the current situation (which is not as it was this morning) tex.stackexchange.com/questions/462021/…

@DavidCarlisle I think the rant is inappropriate. I'll add an answer and delete it immediately. If you and @JosephWright think it's ok, I can undelete.

@AlanMunn the timing may be a coincidence but some combination of "us" adopted bidi on github this morning... github.com/bidi-tex (at Vafa's request)

@DavidCarlisle So you can restore François to his rightful place in the code history. :)

@DavidCarlisle Do you want me to add something to that effect to the answer?

@AlanMunn well it's a bit premature (I haven't even checked it out locally yet) I've just imported the git repo from sourceforge and set up some basic structure, ctan are primed to expect an email asking for the listed maintainer to change in the next few days.. However since that question has been posted... an answer probably shoudl say that maintenance has moved to github (not sure who we are going to say the maintainers are, named individuals or latex-team or "bidi-tex github organisation"

@DavidCarlisle Ok. Feel free to edit the answer then. I've put LaTeX Team for the moment. But the answer is still deleted.

@AlanMunn no it's fine really. Once things are set up I may post another answer (although since the question is just a rant really I'm not sure I will)

@DavidCarlisle ooh the Oberdiek™ bundle, part 2

@PauloCereda A bit that way, but circumstances are different

@JosephWright Oh I can imagine

@PauloCereda See WhatsApp

@JosephWright ack

@DavidCarlisle Weird that I got a downvote on the answer. Maybe a "don't encourage rants" downvote?

@DavidCarlisle ^^

@PauloCereda Howzat!

@PauloCereda too many ducks :D

@boycott.se-yo' ooh :)

@AlanMunn ooh cricket references

And how do you all do, by the way?

@AlanMunn who knows. By the way if you want the back story about today's change, drop me an email.

@AlanMunn I have absolutely no idea about the bidi package, let alone its history. But I can see why someone might think that it might be appropriate to mention the initial author of the package somewhere. Of course, not mentioning him is not a copyright violation, as you say. (And of course I did not downvote your answer.)

@marmot I agree. But it's a case of manners more than anything. But ranting about it publicly isn't really appropriate.

@AlanMunn Sure, I agree with that. (Seems like CTAN and this site work a bit differently from academia. So I would not know how where to complain about such things, but certainly our main site is not the right place. Needless to say that the tone of his "question" isn't appropriate either.)

@DavidCarlisle ^^ interesting entry on my feed. :)

@PauloCereda a couple more since then

@PauloCereda Cool

@DavidCarlisle Thanks for all the work on that

@JosephWright grr I went to the trouble of cloning the soureforge repo and splitting xepersian and bidi now i see there is no history anyway, just an initial commit:-)

@DavidCarlisle Hah!

@DavidCarlisle Lucky you ;)

@DavidCarlisle Glad you decided to take this job :)

@JosephWright Well I'm hoping that we can just park the files really but having large chunks of contrib going unmaintained doesn't seem that good. If Ctan will just update the catalogue I'm hoping not to touch the files at all.

@PauloCereda Oh, a duckification of your avatar :) looks just like the original :)

@marmot @AlanMunn I deleted that "question".

@DavidCarlisle @JosephWright I got today the newest texnische Komödie, and it says that on the context meeting they decided to concentrate after texlive 2019 on a stripped down luatex called MetaTeX, without dvi and bitmap fonts, and which only cares about context.

Is the “upload…” button dimmed also to others?

Oh, now it no longer is dimmed, curious.

@Sebastiano Hi! As friendly moderators we already took action several times regarding inappropriate activities of others, regarding bullying towards you in various ways. I take it seriously, so I escalated to the community team of all StackExchange sites.

It may take days, since they are responsible for more than a hundred sites, so let's see. I insist on solving that since it happened since months.

@StefanKottwitz Thanks! (it is certainly off-topic, so one does not even need to appeal to the "ranting" part to close it, I think).