Hi Edward
ah, so you are talking about the literate programming aspects of TeX.
Fair enough. I came to Leo via that side of things (as did you, I know) but
I can't remember what I ever knew about Knuth's tangle and weave etc.
FWIW, although TeX was originally written in Pascal (or SAIL, some sort of
derived language IIRC), I think nowadays there is a C version which pretty
much everyone uses. I would think that this impliest hat there are plenty
of unit tests for the implementation, presumably written after the fact.
But I don't know too much about that.
Regards, J^n
On Friday, October 15, 2021 at 1:12:28 PM UTC+1 Edward K. Ream wrote:
> On Fri, Oct 15, 2021 at 3:25 AM jkn <[email protected]> wrote:
>
> > I'm curious as to what you mean by this. Are you referring to the
> design of the TeX language?
>
> The short answer: no, but this is an interesting question!
>
> The most important things to know about TeX:
>
> 1. It's never going to change.
> 2. It does what it was intended to do.
> 3. It is an important de facto standard.
> 4. LaTex hides the horrendous details.
>
> However, I was referring to the web language, which imo is a horror show.
> I think I am entitled to this opinion since I have spent about 40 years
> trying to do better :-)
>
> > I have more than once wondered what would have happened if 'TeX' had
> been written in something like Python (which came much later, of
> course)...
>
> The question I have been asking myself lately is whether there are unit
> tests for TeX that would allow re-implementations in other languages. Unit
> testing as a concept did not exist when Knuth created TeX, but such tests
> should be completely feasible. After all the tests would check that, given
> a TeX input, the *text *output matches expected output in one of several
> forms: dvi, svg, pdf, postscript.
>
> These kinds of text-to-text checks are the easiest form of unit tests.
> However, it would likely take at least a year to create tests that fully
> cover TeX. But is that a big deal? I think not, given the importance of
> TeX in the world. There may be such tests, but I haven't (yet) found
> evidence of them.
>
> Absent such tests, we are stuck with precompiled versions of the original
> TeX (in pascal!) program. But I could be mistaken about that.
>
> vs-code has a textab extension that's written in rust! I am busy
> installing lyx (and tex live) so I can see in what language lyx is written.
> And then there is the whole mathjax tool chain, which I'm guessing is built
> on top of TeX/LaTex. There are several LaTeX extensions for vs-code.
>
> *Summary*
>
> There has been *tremendous* progress in all areas since ca 1980, so
> without more research it's too early to say how easy it would be to make
> sense of tex.web :-) And let me emphasize that everything I have said here
> could be mistaken, in small or large ways!
>
> Edward
>
--
You received this message because you are subscribed to the Google Groups
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/leo-editor/5d822c82-0686-45b2-a22f-ff11d9c2cbf7n%40googlegroups.com.
