I'm looking for a BNF grammar of the TeX language, does it exist?
EDIT
For those of us who are not computer scientists, a BNF grammar is one kind of formal description of a CFG: Backus Naur Form.
For those who don't know their Chomsky, a CFG is a context-free grammar, which means (very roughly...
@DavidCarlisle well, the problem with any "simplification" of LaTeX is missing finak output control via artificial tildes, looseness, manual figure placement etc. To this end, it acutally probably makes more sense to use something like LyX and do the final corrections in teh TeX code
@egreg I think it's this but haven't had time to look yet % In order to relieve the \textsf{save stack} we assign the % replacement texts for =\@halignto= globally.
@StefanKottwitz I temporarilz lost inet Connection. Over the Weekend was a user on LC )etp94= whom I still need to answer. Page number in the Header with memoir. Can zou take a look at it please?
I came across an odd problem when writing an interpreter that (should) hooks to external programs/functions: functions in 'C' and 'C++' can't hook variadic functions, e.g. I can't make a function that calls 'printf' with the exact same arguments that it got, and instead has to call an alternate v...
You should look into turing tarpits. They're a fascinating type of programming language that intentionally has as few operations available as possible while still being turing complete. Most of them lack basic data types, functions, even simple things like declaring variables. Personally, I think coding in them is great fun, since it forces you out of your comfort zone to try something needlessly difficult. — DJMcMayhemyesterday
@JosephWright A class is doing \addtocounter{cost}{1000 * \real{#2} * \real{#3}} (with the calc package) with much confidence that eventually dividing by 1000 with fp will return a good result (several additions like that are performed).
@MarkoLustro not a tool, it is a font format, but for tex fonts look at the tex gyre project or in general people use tools like fontforge
@MarkoLustro these people have done more unicode fonts for tex than anyone else, and have some papers describing methods used gust.org.pl/projects/e-foundry/tex-gyre
@AaronHall TeX fonts are 8-bit, so each has 256 slots and anything that requires more space has to use multiple fonts. For Unicode engines, you can use OTF fonts directly so one font file covers everything.
@AaronHall With an 8-bit font, mapping glyph-to-Unicode (for example for copy-paste from PDFs) is 'interesting'
@AaronHall TeX 8-bit fonts are a walled garden: they are only use by TeX
@AaronHall See also my earlier comment about hyphenation
@PauloCereda that's fine and not quite a problem, I'll get a large breakfast in tez morning :) (getting up at 5:55 as I teach at 8:00)
@egreg Ad your comment to the invoice thingy: Isn't it preferable to explicitly call expl3 if you're using it? You call it implicitly through xparse and siunitx.
@AaronHall as long as you use fonts available for pdflatex and write in one or two languages that use latin alphabet? Not at all... (btw, if you don't use the "reply to message" function of the chat, the people you want to reach very likely miss your message)
@AaronHall well the cjk-package managed to make all the chinese characters usable with pdflatex. So it can be done but it is not easy - a lot of fonts are involved.
@AaronHall what do you mean by an alphabet here? you can not use a font with more than 256 characters but you can use hundreds of fonts. (that's how the cjk packages for (pdf)tex work)
@DavidCarlisle Can I define \file_input:nT somehow so this doesn't throw an error? I'm not sure how to do this for a signature involving T when \file_input:n etc. exist.
@egreg @DavidCarlisle @anybody Or some other workaround? I just wanted it to load l3doc.cfg....
@cfr with pdflatex '\expandafter\def\csname file_input:nT\endcsname{\csname file_if_exist_input:nT\endcsname}\input{l3doc.dtx}' I get the work done, apart from an error about an "Illegal use of \verb" error on line 285
@cfr The error on line 285 is due to a stray | character; after removing it the compilation runs smoothly. But in the end you might as well fix the other piece of text.
@cfr here's the diff
diff l3doc.dtx /usr/local/texlive/2016/texmf-dist/source/latex/l3kernel/l3doc.dtx
285c285
< % same (\enquote{\texttt{\string\foo}}). In general, \cs{cs} is more robust since
---
> % same (\enquote{\texttt{\string\foo|}}). In general, \cs{cs} is more robust since
1076c1076
< \file_if_exist_input:nT { l3doc.cfg }
---
> \file_if_exist:nT { l3doc.cfg }
1078,1079c1078,1082
< \cs_if_exist:cF { ExplMakeTitle }
< { \msg_info:nn { l3doc } { input-cfg } }
---
> \file_input:nT { l3doc.cfg }
@egreg Thank you! That's very nice. Though I'm starting to think l3doc can't be used very much and probably I shouldn't be trying to use it either. (I thought it would be the L3 equivalent of ltxdoc, but clearly not!)
I'm looking for a BNF grammar of the TeX language, does it exist?
EDIT
For those of us who are not computer scientists, a BNF grammar is one kind of formal description of a CFG: Backus Naur Form.
For those who don't know their Chomsky, a CFG is a context-free grammar, which means (very roughly...
