Hi Everyone,

I know a few people want to see a stable release of Padre.  That's fine,
so do I.

However there are still a few niggles with the code base and Padre doing
the right thing.

By this I mean, I haven't seen any major crashes or anyone complain of
loss of work due to stability issues.  This isn't the problem as I see
it.

What I do see for now is a few stand out matters, and most are still
around the matter of the big threading change merged in just after the
0.56 release.

Adam has made headway with the database and threading issue.

The matter under "heavy debate" is if turning off the optimisations in
Startup.pm: http://padre.perlide.org/trac/changeset/11138 is enough to
mark a Stable release of Padre.

Myself, I found some issues, which with updating to this revision seems
addressed, however I now see this:

Thread 2 terminated abnormally: lock can only be used on shared values
at /home/pete/Programming/Perl/Padre/trunk/Padre/lib/Padre/SlaveDriver.pm line 
210.
Thread 3 terminated abnormally: lock can only be used on shared values
at /home/pete/Programming/Perl/Padre/trunk/Padre/lib/Padre/SlaveDriver.pm line 
210.
Thread 6 terminated abnormally: lock can only be used on shared values
at /home/pete/Programming/Perl/Padre/trunk/Padre/lib/Padre/Wx/Ack.pm
line 535, <$__ANONIO__> line 107.
Thread 7 terminated abnormally: lock can only be used on shared values
at /home/pete/Programming/Perl/Padre/trunk/Padre/lib/Padre/Wx/Ack.pm
line 535, <$__ANONIO__> line 107.
Thread 8 terminated abnormally: lock can only be used on shared values
at /home/pete/Programming/Perl/Padre/trunk/Padre/lib/Padre/Wx/Ack.pm
line 535, <$__ANONIO__> line 47.

this is over a number of searches.

What's worse is that ack doesn't find the method name refresh.. and I
know it's in there some where!

So something isn't quite right yet.


I'm quite sure Padre is stable enough in general use, but when you have
these little doubts it's a risk to release it as stable giving the
impression for all and sundry to take this next version, package it up
for other platforms ( such as OSX, ActiveState PPM, Windows etc ) only
to have the user base increase and then leave a little disappointed.

Reputation is hard to win, easy to lose.  In fact, that's paraphrasing
one of my favourite sayings to the kids, trust is hard to earn, however
it's easy to lose.

I'd hate to see Padre "stable" rushed out only to undo all the good work
to date with "silly" little issues like these.

Granted in truth they are likely to be difficult ones to resolve.

It's my understanding we have two people really clued up on threads,
that's Steffen and Andrew.  Both of whom are busy with their lives at
the moment.

It may just mean we have another unstable release with the requisite
warnings until either of them are available or someone else cracks the
issue.  

Or we release as stable understanding what we risk if Padre isn't as
good as it could be.

Also, there are 4 tickets set as "blocker" currently:

http://padre.perlide.org/trac/ticket/460

Seriously??? Artwork is a blocker???



http://padre.perlide.org/trac/ticket/836

Which I think should be down graded.

and two that relate to the new threading code:

http://padre.perlide.org/trac/ticket/859
http://padre.perlide.org/trac/ticket/866


Just my thoughts on the matter... happy to accept the broader
community's view here.

Pete.
_______________________________________________
Padre-dev mailing list
[email protected]
http://mail.perlide.org/mailman/listinfo/padre-dev

Reply via email to