jacks oddities

Tue Jul 13 07:54:00 GMT 2004

> And I'm working on a fix.
> The issue is this, I have tclsh8.4 and tclsh on my darwin box. Tclsh is
> a sym link to tclsh8.4.
> Now in the jacks script there is an explicit call to tclsh8.3!!!!!
>> So every time I launched it with make check I only analyzed the exisiting
> gcj.log instead of running the suite from scratch.

Aaah! That's bad.
> Running the suite with 'tclsh jacks gcj' works fine ....
>> The idea is now to override the jacks call in jacks.exp like this:
>> - catch {exec ./jacks gcj} msg
> + set tcl_version [info tclversion]
> + catch {exec tclsh$tcl_version jacks gcj} msg

Correct me if I'm wrong (I don't know anything about Tcl/Tk),
but isn't "tclsh" supposed to be a symlink to the correct
version of the Tcl interpreter? So shouldn't the above be just:
 - catch {exec ./jacks gcj} msg
 + catch {exec tclsh jacks gcj} msg
Does it also make sense to check "info tclversion" and
complain if it is not at least 8.3?
However, I think a better fix would be to submit the 
proper fix to the Jacks guys. (And while you're at it,
perhaps an updated gcj.log as well for the current
mainline - or not, as it is always in flux. :-/)
> >Yes: I don't know whether to exclaim "D'oh!" or "Wow!". ;-)
> >
> >I have verified that your patch indeed gets us 58 XPASSes
> >(58 testcases out of 68 in 4.7.10) and no FAILs.
>> Yes, you're right, only 58.

Don't say "only" - any monotonic reduction in Jacks
XFAILs is highly welcome, at least as far as I am 
concerned.
If you meant w.r.t. your statement of "about 60 XPASSes",
then I was just confirming your findings and putting
in the exact figure. I am sorry if you thought I was pointing
out an inaccuracy.
Thanks,
Ranjit.
-- 
Ranjit Mathew Email: rmathew AT gmail DOT com
Bangalore, INDIA. Web: http://ranjitmathew.tripod.com/