[tex-live] Is TeXk from TeXlive 2003 TRIP certified?

[Petr Olsak]

On Tue, 16 Mar 2004, Fabrice Popineau wrote:

>
> > isn't it? I thought the point of the trip test was for the developer
> > to check their work, not the end user.
>
> Definitely.
>
> Who would want to run TeX with those defaults ???? Not to speak about
> the rather limited memory size (!), but why would my terminal be stuck to 72
> chars, error lines to 64 and so on ? Please ...

OK.

> If you lost 2 pages out of 16, it is simply because the trip test
> expects 72 char lines.

NO, if I have set error_line = 64, half_error_line = 32, max_print_line = 72
I stil get only 14 pages instedad 16. I don't know how.

See the "Emergency stop" message below.

diff trip.log trip-ori.log
1,3c1
< This is TeXk, Version 3.141592 (Web2C 7.5.2) (format=trip 2004.3.16)  16
MAR 2004 16:08
<  %&-line parsing enabled.
<  (/usr/local/texlive/texmf/web2c/cp8bit.tcx)
---
> This is TeX, Version 3.14159 (preloaded format=trip 95.3.8)  8 MAR 1995
14:24
5c3
< (./trip.tex ##
---
> (trip.tex ##
61,64d58
< \openout0 = `8terminal.tex'.
<
< \openout10 = `tripos.tex'.
<
... etc
2024c2018
< \vbox(16383.99998+1.0)x100.0, glue set 16238.99998fil
---
> \vbox(16383.99998+1.0)x100.0, glue set 16239.0fil
... etc
5907,5910c5900
< ! Emergency stop.
< l.408   \global\let\endt=&\endt
<                                &$&&&.}
< (interwoven alignment preambles are not allowed)
---
> {end of alignment template}
5912,5921c5902,7304
<
< Here is how much of TeX's memory you used:
<  31 strings out of 98658
<  187 string characters out of 1226551
<  3556 words of memory out of 1500001
<  361 multiletter control sequences out of 10000+50000
<  3041 words of font info for 4 fonts, out of 1000000 for 2000
<  1 hyphenation exception out of 1000
<  5i,7n,9p,113b,38s stack positions out of
5000i,500n,6000p,200000b,40000s
< Output written on trip.dvi (14 pages, 2568 bytes).
---
> \l #1->
> #1<-$
> {end of alignment template}
... etc

The "glue set 16238.99998fil" instead "glue set 16239.0fil" is not fatal
problem and it is acceptable by TRIP. But, it is very interesting for me:
why the different web2c versions behave differently on the same machine
(hardware+OS)? I don't know how it is possible. Who did a change in TeX
source and why?

Best regards

Petr Olsak