[bakefile-devel] ANN: 0.2.4 pre-release

[Vaclav Slavik]

Hi,

I uploaded pre-release packages for 0.2.4:
http://sourceforge.net/project/showfiles.php?group_id=83016&package_id=85308&release_id=627653

The changelog is here:
http://www.bakefile.org/browser/bakefile/trunk/NEWS

As usual, the plan is to release 0.2.4 in a week or so if there are no
non-trivial problems with these packages.

I'd appreciate any testing you can give it and reports of any
regressions you encounter.

Thanks!
Vaclav

[Francesco Montorsi]

Hi,

Post by Vaclav Slavik
I'd appreciate any testing you can give it and reports of any
regressions you encounter.

I've updated to latest SVN and tried:

***@ubuntu:~/work/bakefile/tests$ bakefile_gen -f msvs2008prj

result:

[1/26] generating msvs2008prj from library/library.bkl
/home/frm/work/bakefile/tests/library/library.bkl:3: error: unknown tag 'lib'
[bakefile_gen] error: bakefile exited with error (1)

:/

I don't have time right now to look at this, but probably needs investigation
before a release...

Francesco

[Vaclav Slavik]

Hi,

Post by Francesco Montorsi
[1/26] generating msvs2008prj from library/library.bkl
/home/frm/work/bakefile/tests/library/library.bkl:3: error: unknown tag 'lib'
[bakefile_gen] error: bakefile exited with error (1)
:/
I don't have time right now to look at this, but probably needs investigation
before a release...

Hmm, no such thing here:

$ /tmp/bkl024/bin/bakefile_gen -f msvs2008prj -b library/library.bkl
[1/1] generating msvs2008prj from library/library.bkl
writing library/library_vc9_mylib.vcproj
writing library/library_vc9.sln
writing library/library_vc9_hello.vcproj
3 files modified
$

Do you have some local changes by a chance? If not, is anything
suspicious in the output with -V flag (e.g. picking files from installed
bakefile version instead of the SVN tree)?

Thanks,
Vaclav

[Francesco Montorsi]

Hi,

Post by Vaclav Slavik
Do you have some local changes by a chance?
If not, is anything
suspicious in the output with -V flag (e.g. picking files from installed
bakefile version instead of the SVN tree)?

I've found the root of the problem; msvs2008prj.bkl was not being installed in
/usr/local/share/bakefile/rules because I forgot to rerun "./configure"; the
result was that msvs2008prj.bkl was not loaded and bakefile wasn't aware of
<lib> targets.

I think the fact no warnings/errors about the missing .bkl were shown means that
a safer loadModule() is required ;)

Attached is a raw first-version of such safer loadModule(); it may require
various fixes as I'm not sure about the

# include module-specific makefiles:
global availableFiles
for f in availableFiles:
if m in f.modules:
f.modules.remove(m)
if len(f.modules) == 0:
processFile(f.file)
imported = True
availableFiles = [f for f in availableFiles if len(f.modules)>0]

part.

Francesco

[Vaclav Slavik]

Post by Francesco Montorsi
I think the fact no warnings/errors about the missing .bkl were shown
means that a safer loadModule() is required ;)

Done in r1260.

Thanks!
Vaclav

[Francesco Montorsi]

Post by Vaclav Slavik

Post by Francesco Montorsi
I think the fact no warnings/errors about the missing .bkl were shown
means that a safer loadModule() is required ;)

Done in r1260.

Thanks.

Except for ticket #227 I've found no problems in this release (I've tested it
for wxLua and some wxCode components).

Francesco

[Warren Young]

Post by Vaclav Slavik
As usual, the plan is to release 0.2.4 in a week or so if there are no
non-trivial problems with these packages.

It seems to be working fine for MySQL++, even after turning on VC++ 2008
support. Thanks for that!

I do get a new compiler complaint in VC++ 2008:

Command line warning D9030 : '/Gm' is incompatible with multiprocessing;
ignoring /MP switch

Personally, I'd rather have /MP than /Gm.

Most of my users build a new MySQL++ release only once, the install it
and move on. Overall, /Gm makes the build slower for these users.

That's true for me, too, most of the time: I do most of the ongoing
development on Linux and OS X, building directly on Windows only to
verify that it still works, and in those rare cases where a bug only
appears on Windows. Since Windows builds are so few and far between for
me, I end up rebuilding the whole thing almost every time, if only to be
sure I'm seeing what end users will see.

[Vaclav Slavik]

Hi,

Post by Warren Young
Command line warning D9030 : '/Gm' is incompatible with
multiprocessing; ignoring /MP switch

Ah, /Gm is causing problems again...

Post by Warren Young
Personally, I'd rather have /MP than /Gm.

I agree. It's not like disabling /Gm would mean everything is recompiled
from scratch, it only filters out some irrelevant change -- something
other compilers didn't even bother to implement. Fixed in r1258.

Thanks!
Vaclav

[Warren Young]

Post by Vaclav Slavik
Fixed in r1258.

Verified. Thank you!

[Scott Bailey]

We have not found defects.

SB

Post by Vaclav Slavik
Hi,
http://sourceforge.net/project/showfiles.php?group_id=83016&package_id=85308&release_id=627653
http://www.bakefile.org/browser/bakefile/trunk/NEWS
As usual, the plan is to release 0.2.4 in a week or so if there are no
non-trivial problems with these packages.
I'd appreciate any testing you can give it and reports of any
regressions you encounter.
Thanks!
Vaclav
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's
challenge
Build the coolest Linux based applications with Moblin SDK & win great
prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Bakefile-devel mailing list
https://lists.sourceforge.net/lists/listinfo/bakefile-devel