Great work grubba! I see the following blocker bugs. Please add your own.
[ ] Fix dmalloc leak. [ ] Fix Windows java problem. [ ] Fix Windows dynamic load of the Image module (?) [ ] Verify the whole path from (make doc;make export) to make install. [ ] Get Windows packaging to work for 7.5. [ ] Write change log/release notes
Now is really the time for last requests from Roxen, Caudium and Aido developers.
Maybe revert to the old buggy dumper that dumps all nested programs wherever they are referenced from, so that there are less tricky reference cycle that the decoder can't handle?
/ Martin Stjernholm, Roxen IS
Previous text:
2004-03-22 02:59: Subject: 7.6
Great work grubba! I see the following blocker bugs. Please add your own.
[ ] Fix dmalloc leak. [ ] Fix Windows java problem. [ ] Fix Windows dynamic load of the Image module (?) [ ] Verify the whole path from (make doc;make export) to make install. [ ] Get Windows packaging to work for 7.5. [ ] Write change log/release notes
Now is really the time for last requests from Roxen, Caudium and Aido developers.
/ Martin Nilsson (provokatör)
It is not possible to have the dumper give up more easily instead of trying and fail?
/ Martin Nilsson (provokatör)
Previous text:
2004-03-22 03:34: Subject: 7.6
Maybe revert to the old buggy dumper that dumps all nested programs wherever they are referenced from, so that there are less tricky reference cycle that the decoder can't handle?
/ Martin Stjernholm, Roxen IS
What do you mean? I don't think the wasted time trying to do bogus dumps is important.
/ Martin Stjernholm, Roxen IS
Previous text:
2004-03-22 03:41: Subject: 7.6
It is not possible to have the dumper give up more easily instead of trying and fail?
/ Martin Nilsson (provokatör)
It already checks that it works to read the dump back afterwards. The problem is when several dumps are decoded due to cyclic dependencies and the whole resolver system turns over on its back. One can implement a check for that by loading each dump in a pike process that have essentially nothing else loaded aldready.
The problem is to decide from that which dump is the "faulty" one. Since the problem doesn't really lie in any single dump, that amounts toanalyzing the load dependencies in all the failed cycles and deduce a minimal set of dump(s) to remove. You could see it as a nice optimization problem if you like.
Also, I'm not positively sure that compiling a certain module from source will resolve the problematic situation in all cases, so the system should also tentatively remove those dumps and rerun the test to verify that the problem has gone away.
/ Martin Stjernholm, Roxen IS
Previous text:
2004-03-23 00:53: Subject: 7.6
I mean if it is possible for it to realize the dump is bogus and abort it.
/ Martin Nilsson (provokatör)
Regardless of decision, it would be good to have a summary of the pros and cons of the old and new dumper for release notes and other documentation.
/ Martin Nilsson (provokatör)
Previous text:
2004-03-22 03:34: Subject: 7.6
Maybe revert to the old buggy dumper that dumps all nested programs wherever they are referenced from, so that there are less tricky reference cycle that the decoder can't handle?
/ Martin Stjernholm, Roxen IS
If the new dumper is reverted then it's no use comparing it to something else.
If it stays there should probably be a bit of documentation on how it causes problems, but that actually amounts to documenting the ways in which the whole resolver system is flawed.
/ Martin Stjernholm, Roxen IS
Previous text:
2004-03-22 03:43: Subject: 7.6
Regardless of decision, it would be good to have a summary of the pros and cons of the old and new dumper for release notes and other documentation.
/ Martin Nilsson (provokatör)
In the last episode (Mar 22), Martin Nilsson (provokatr) @ Pike (-) developers forum said:
Now is really the time for last requests from Roxen, Caudium and Aido developers.
I'd like support for the realpath() function; a patch enabling it is in Crunch bug 3317. Pike already has resolvepath, but that's Solaris-only.
I would like to see this as a patch against the resolvepath code rather than a new function.
/ Martin Nilsson (provokatör)
Previous text:
2004-03-22 06:02: Subject: Re: 7.6
In the last episode (Mar 22), Martin Nilsson (provokatr) @ Pike (-) developers forum said:
Now is really the time for last requests from Roxen, Caudium and Aido developers.
I'd like support for the realpath() function; a patch enabling it is in Crunch bug 3317. Pike already has resolvepath, but that's Solaris-only.
-- Dan Nelson dnelson@allantgroup.com
/ Brevbäraren
In the last episode (Mar 22), Martin Nilsson (provokatr) @ Pike (-) developers forum said:
I would like to see this as a patch against the resolvepath code rather than a new function.
You mean just a rename of resolvepath to realpath, or make pike's resolvepath call the OS'es realpath? I would prefer to have a pike function called "realpath", since I would expect more people to be looking for that function as opposed to "resolvepath", but I'll be happy as long as it's available at all :)
And don't pay any attention to the Solaris realpath manpage's claim that it may return a relative path in some cases. It's a bug in the manpage:
http://sunsolve.sun.com/private-cgi/retrieve.pl?type=0&doc=bug/manpage/s...
I would like to avoid having two functions to avoid this kind of code:
#if constant(realpath) x = realpath(y); #elif constant(resolvepath) x = resolvepath(y); #endif
/ Martin Nilsson (provokatör)
Previous text:
2004-03-22 18:06: Subject: Re: 7.6
In the last episode (Mar 22), Martin Nilsson (provokatr) @ Pike (-) developers forum said:
I would like to see this as a patch against the resolvepath code rather than a new function.
You mean just a rename of resolvepath to realpath, or make pike's resolvepath call the OS'es realpath? I would prefer to have a pike function called "realpath", since I would expect more people to be looking for that function as opposed to "resolvepath", but I'll be happy as long as it's available at all :)
And don't pay any attention to the Solaris realpath manpage's claim that it may return a relative path in some cases. It's a bug in the manpage:
http://sunsolve.sun.com/private-cgi/retrieve.pl?type=0&doc=bug/manpage/s...
-- Dan Nelson dnelson@allantgroup.com
/ Brevbäraren
In the last episode (Mar 22), Martin Nilsson (provokatr) @ Pike (-) developers forum said:
I would like to avoid having two functions to avoid this kind of code:
#if constant(realpath) x = realpath(y); #elif constant(resolvepath) x = resolvepath(y); #endif
Luckily, the only place resolvepath is available is Solaris, and Solaris also also realpath (back as far as SunOS 4.1.3 at least). I don't know of any pike code that uses resolvepath.
[ ] Fix dmalloc leak. [ ] Fix Windows java problem. [ ] Fix Windows dynamic load of the Image module (?) [ ] Verify the whole path from (make doc;make export) to make install. [ ] Get Windows packaging to work for 7.5. [ ] Write change log/release notes +[ ] Mirar should commit the latest bittorrent changes
/ Mirar
Previous text:
2004-03-22 02:59: Subject: 7.6
Great work grubba! I see the following blocker bugs. Please add your own.
[ ] Fix dmalloc leak. [ ] Fix Windows java problem. [ ] Fix Windows dynamic load of the Image module (?) [ ] Verify the whole path from (make doc;make export) to make install. [ ] Get Windows packaging to work for 7.5. [ ] Write change log/release notes
Now is really the time for last requests from Roxen, Caudium and Aido developers.
/ Martin Nilsson (provokatör)
Le 22 mars 04, à 03:00, Martin Nilsson (provokatör) @ Pike (-) developers forum a écrit :
Great work grubba! I see the following blocker bugs. Please add your own.
[ ] Fix dmalloc leak. [ ] Fix Windows java problem.
I should add : [ ] Fix FreeBSD Java detection and utilization.
The native JDK for FreeBSD should be supported, but I cannot figure why it doesn't work with pike... And since it work natively with Mozilla and apache tomcat, I dunno why it should not on freebsd.
By the way I can give access to a freebsd 5.2.1 box for that... Contact me in private for that.
[ ] Fix Windows dynamic load of the Image module (?) [ ] Verify the whole path from (make doc;make export) to make install. [ ] Get Windows packaging to work for 7.5. [ ] Write change log/release notes
Now is really the time for last requests from Roxen, Caudium and Aido developers.
Caudium-devel mailing list caudium-devel@caudium.net http://lists.oav.net/wws/info/caudium-devel Archives are : http://lists.oav.net/wws/arc/caudium-devel Caudium Website : http://caudium.net/
-- Xavier Beaudouin - Unix System Administrator & Projects Leader. President of Kazar Organization : http://www.kazar.net/ Please visit http://caudium.net/, home of Caudium & Camas projects
[ ] Fix Windows java problem.
I should add : [ ] Fix FreeBSD Java detection and utilization.
I don't think this is a blocker bug. BSDs hasn't been well tested so a stable release for FreeBSD could take while as various issues appears. I think that could wait until a later 7.6 release. Especially since there isn't much of any active FreeBSD development going on.
Note that I wrote "problem" and not "detection and utilization". It is ok for me if Java support isn't compiled at all on Windows for the time being (given that it has always been broken).
/ Martin Nilsson (provokatör)
Previous text:
2004-03-22 11:26: Subject: Re: [caudium-devel] 7.6
Le 22 mars 04, à 03:00, Martin Nilsson (provokatör) @ Pike (-) developers forum a écrit :
Great work grubba! I see the following blocker bugs. Please add your own.
[ ] Fix dmalloc leak. [ ] Fix Windows java problem.
I should add : [ ] Fix FreeBSD Java detection and utilization.
The native JDK for FreeBSD should be supported, but I cannot figure why it doesn't work with pike... And since it work natively with Mozilla and apache tomcat, I dunno why it should not on freebsd.
By the way I can give access to a freebsd 5.2.1 box for that... Contact me in private for that.
[ ] Fix Windows dynamic load of the Image module (?) [ ] Verify the whole path from (make doc;make export) to make install. [ ] Get Windows packaging to work for 7.5. [ ] Write change log/release notes
Now is really the time for last requests from Roxen, Caudium and Aido developers.
Caudium-devel mailing list caudium-devel@caudium.net http://lists.oav.net/wws/info/caudium-devel Archives are : http://lists.oav.net/wws/arc/caudium-devel Caudium Website : http://caudium.net/
-- Xavier Beaudouin - Unix System Administrator & Projects Leader. President of Kazar Organization : http://www.kazar.net/ Please visit http://caudium.net/, home of Caudium & Camas projects
/ Brevbäraren
Hello,
[ ] Fix dmalloc leak. [ ] Fix Windows java problem. [ ] Fix Windows dynamic load of the Image module (?) [ ] Verify the whole path from (make doc;make export) to make install. [ ] Get Windows packaging to work for 7.5. [ ] Write change log/release notes
Now is really the time for last requests from Roxen, Caudium and Aido developers.
For me a rock stable Pike without any leaks is fine.
/ David Gourdelier
May I suggest
[ ] fix decoding of two-complementary binary integers (sscanf %+c or %C)
I think it would be good to have *before* the split.
/ Mirar
Previous text:
2004-03-22 02:59: Subject: 7.6
Great work grubba! I see the following blocker bugs. Please add your own.
[ ] Fix dmalloc leak. [ ] Fix Windows java problem. [ ] Fix Windows dynamic load of the Image module (?) [ ] Verify the whole path from (make doc;make export) to make install. [ ] Get Windows packaging to work for 7.5. [ ] Write change log/release notes
Now is really the time for last requests from Roxen, Caudium and Aido developers.
/ Martin Nilsson (provokatör)
The bug here is really that sscanf %+c doesn't work as sscanf %+1c.
array_sscanf("\xff","%+1c")[0];
(6) Result: -1
/ Martin Nilsson (provokatör)
Previous text:
2004-03-23 15:04: Subject: 7.6
May I suggest
[ ] fix decoding of two-complementary binary integers (sscanf %+c or %C)
I think it would be good to have *before* the split.
/ Mirar
Looks like there are more issues...
array_sscanf("\xff\xff","%+1c")[0];
(15) Result: -1
array_sscanf("\xff\xfe","%+1c")[0];
(16) Result: -1
array_sscanf("\xff\xfe","%+-1c")[0];
(17) Result: -1
array_sscanf("\xff\xfe","%-+1c")[0];
(18) Result: -1
/ Martin Nilsson (provokatör)
Previous text:
2004-03-23 15:42: Subject: 7.6
(Hmm... hidden comments.)
Well, I didn't say it wouldn't be easy to fix. :)
/ Mirar
array_sscanf("\xff\xff","%+1c")[0];
(15) Result: -1
array_sscanf("\xff\xfe","%+1c")[0];
(16) Result: -1
array_sscanf("\xff\xfe","%+-1c")[0];
(17) Result: -1
array_sscanf("\xff\xfe","%-+1c")[0];
(18) Result: -1
Looks correct to me. What behaviour did you expect?
Did you perhaps mean the following?
array_sscanf("\xff\xff","%+2c")[0];
(1) Result: -1
array_sscanf("\xff\xfe","%+2c")[0];
(2) Result: -2
array_sscanf("\xff\xfe","%+-2c")[0];
(3) Result: -257
array_sscanf("\xff\xfe","%-+2c")[0];
(4) Result: -257
/ Henrik Grubbström (Lysator)
Previous text:
2004-03-23 15:48: Subject: 7.6
Looks like there are more issues...
array_sscanf("\xff\xff","%+1c")[0];
(15) Result: -1
array_sscanf("\xff\xfe","%+1c")[0];
(16) Result: -1
array_sscanf("\xff\xfe","%+-1c")[0];
(17) Result: -1
array_sscanf("\xff\xfe","%-+1c")[0];
(18) Result: -1
/ Martin Nilsson (provokatör)
How silly of me. I need blood suger.
/ Martin Nilsson (provokatör)
Previous text:
2004-03-23 15:54: Subject: 7.6
array_sscanf("\xff\xff","%+1c")[0];
(15) Result: -1
array_sscanf("\xff\xfe","%+1c")[0];
(16) Result: -1
array_sscanf("\xff\xfe","%+-1c")[0];
(17) Result: -1
array_sscanf("\xff\xfe","%-+1c")[0];
(18) Result: -1
Looks correct to me. What behaviour did you expect?
Did you perhaps mean the following?
array_sscanf("\xff\xff","%+2c")[0];
(1) Result: -1
array_sscanf("\xff\xfe","%+2c")[0];
(2) Result: -2
array_sscanf("\xff\xfe","%+-2c")[0];
(3) Result: -257
array_sscanf("\xff\xfe","%-+2c")[0];
(4) Result: -257
/ Henrik Grubbström (Lysator)
Frankly I don't see what "%+c" would mean. "%c" means a character, not an integer. The string "ÿ" should always give the character 'ÿ' when scanned with "%c".
/ Marcus Comstedt (ACROSS) (Hail Ilpalazzo!)
Previous text:
2004-03-23 15:36: Subject: 7.6
The bug here is really that sscanf %+c doesn't work as sscanf %+1c.
array_sscanf("\xff","%+1c")[0];
(6) Result: -1
/ Martin Nilsson (provokatör)
pike-devel@lists.lysator.liu.se
-
Dan Nelson -
David Gourdelier -
Henrik Grubbstr�m (Lysator) ＠ Pike (-) developers forum -
Marcus Comstedt (ACROSS) (Hail Ilpalazzo!) ＠ Pike (-) developers forum
-
Martin Nilsson (provokat�r) ＠ Pike (-) developers forum
-
Martin Stjernholm, Roxen IS ＠ Pike developers forum
-
Mirar ＠ Pike developers forum
-
Xavier Beaudouin