#1056 closed defect (fixed)
ipp-2.4-cr0: DBD::mysql ulong undeclared on Mac OS X
| Reported by: | Michael Wood-Vasey | Owned by: | jhoblitt |
|---|---|---|---|
| Priority: | high | Milestone: | |
| Component: | misc | Version: | 2.4 |
| Severity: | minor | Keywords: | |
| Cc: | jester@… |
Description
Compiling DBD::mysql on Mac OS X (10.5.2) finds 'ulong' undeclared.
Can make work by forcing the cflags:
perl Makefile.PL PREFIX=/Volumes/data/PS1/code/ipp-2.4.darwin LIB=/Volumes/data/PS1/code/ipp-2.4.darwin/lib --cflags="-D'ulong=unsigned long'"
in the DBD-mysql-4.006 directory
Here's the original error
cc -c -I/System/Library/Perl/Extras/5.8.8/darwin-thread-multi-2level/auto/DBI -D'ulong=usigned long' -DDBD_MYSQL_INSERT_ID_IS_GOOD -g -arch i386 -arch ppc -g -pipe -fno-common -DPERL_DARWIN -no-cpp-precomp -fno-strict-aliasing -Wdeclaration-after-statement -I/usr/local/include -O3 -DVERSION=\"4.006\" -DXS_VERSION=\"4.006\" "-I/System/Library/Perl/5.8.8/darwin-thread-multi-2level/CORE" dbdimp.c
dbdimp.c: In function âmysql_dr_connectâ:
dbdimp.c:1538: error: âusignedâ undeclared (first use in this function)dbdimp.c: In function âmysql_dr_connectâ:
dbdimp.c:1538: error: âusignedâ undeclared (first use in this function)
dbdimp.c:1538: error: (Each undeclared identifier is reported only once
dbdimp.c:1538: error: for each function it appears in.)
dbdimp.c:1538: error: syntax error before âlongâ
dbdimp.c:1538: error: (Each undeclared identifier is reported only once
dbdimp.c:1540: error: syntax error before âlongâ
dbdimp.c:1538: error: for each function it appears in.)
dbdimp.c:1538: error: syntax error before âlongâ
dbdimp.c:1540: error: syntax error before âlongâ
dbdimp.c:1654: error: syntax error before âlongâdbdimp.c:1654: error: syntax error before âlongâ
[...]
Change History (17)
comment:1 by , 18 years ago
comment:3 by , 18 years ago
This appears to be a mysql header problem that doesn't effect linux.
http://bugs.mysql.com/bug.php?id=22227
"Appears to be fixed in 5.0.33. It is very confusing when in-between releases disappear,and difficult to maintain backwards compatibility with older releases if we are unable to download them. The free bitkeeper UI also makes it difficult to track which version a particular changeset is included in."
What version of MySQL do you have installed?
comment:4 by , 18 years ago
MySQL 5.0.27 from the extlibs tarball.
[wwoodvas@serenity psLib] /Volumes/data/PS1/code/ipp-2.4.darwin/bin/mysql -V
/Volumes/data/PS1/code/ipp-2.4.darwin/bin/mysql Ver 14.12 Distrib 5.0.27, for apple-darwin9.2.0 (powerpc) using EditLine wrapper
I'll try a more recent version and see.
comment:5 by , 18 years ago
Please give 5.0.51a a try. If that fixes it we'll upgrade the IPP supplied version.
comment:6 by , 18 years ago
I succesfully downloaded and installed 5.0.45
http://downloads.mysql.com/archives/mysql-5.0/mysql-5.0.45.tar.gz
and it didn't work. I'll look for 5.0.51a and try that.
comment:7 by , 18 years ago
5.0.45 seems like the latest stable version and I'm not going to login and setup an account at dev.mysql.org just for this. I'm happy with the workaround for now, although I understand how this is an issue that you need Mac OS X folks to help you resolve.
comment:8 by , 18 years ago
| Cc: | added |
|---|---|
| Summary: | DBD::mysql ulong undeclared on Mac OS X → ipp-2.4-cr0: DBD::mysql ulong undeclared on Mac OS X |
Didn't have this problem with mysql-5.1.19-beta. Could you IPP folks add a minimum version requirement to sql on Mac, or is that difficult?
comment:9 by , 18 years ago
It'd be simpler (especially for compile time issues) if we just ship the same version across all platforms. I'd also really like to find a version of 5.0.x that works on OSX as that's what we using in "production" here. Has anyone tried 5.0.51 ?
comment:10 by , 18 years ago
Just checked http://dev.mysql.com/downloads/ and they link to 5.0.51a as latest "stable" production release. However, I'm fuzzy as to whether that version still is subject to some apparently bad bugs that you can find by googling 'mysql 5.0.51' - see http://dev.mysql.com/doc/refman/5.0/en/releasenotes-cs-5-0-51a.html
But given that they list it as "working" for all platforms, that version should be ok.
Can't we just release IPP as it is now, and do a small update later once we've worked out the mySQL problem? I don't feel like installing 5.0.51 since I have an (apparently) working version that's more recent...
Alternatively: Michael, do you want to give it a try? Why do you think 5.0.45 is the most stable, not 5.0.51a?
Cheers,
Sebastian
comment:11 by , 18 years ago
| Version: | unspecified → 2.4 |
|---|
comment:12 by , 18 years ago
I haven't looked into this in detail. Quick Google searches delivered 5.0.45 as easily downloadable, whereas 5.0.51a took me to the devs.mysql.com site where I had to register, which I declined to do. If you have to go to devs to get it, it doesn't suggest to me that it's the public stable release. I'm really not up-to-date on what the verified maintenance release number of the latest version of MySQL and don't mean to claim any particular knowledge.
comment:13 by , 18 years ago
I don't get the registration comment. As of NOW, I can go to
http://dev.mysql.com/downloads/mysql/5.0.html#macosx-dmg
and download a .dmg or .tar for 5.0.51a without registering or anything. This is where last week's quick google search for the most recent stable $$$-free mySQL led me.
Does that settle that we can use 5.0.51a?
comment:14 by , 18 years ago
Oh, I see. Underneath the registration stuff is
"""
» No thanks, just take me to the downloads!
"""
I hate this sort of stuff. I'll refrain for going into a rant, but this is one of the things about the whole MySQL experience that bugs me. At least as of v5.0 it's a real relational database unlike earlier versions.
Sure, include 5.0.51a in the default dist.
comment:15 by , 18 years ago
I started the ipp-2.5 config files yesterday with MySQL 5.0.51a in them. Can I consider this issue resolved?
comment:16 by , 18 years ago
| Resolution: | → fixed |
|---|---|
| Status: | new → closed |
Sure. If it crops up again we'll reopen this thread.

That really needs to be fixed upstream and I don't have an OSX machine to test on. Could you send basically the same bug report to perl@… and Cc: me?