This might be a configure bug, although it doesn't find SQLite prerequisites it tries to compile anyway.
I'm trying to compile pike 7.7.25 from cvs.
Output of configure is:
SQLite.............. no (dependencies failed)
Output of make is:
Compilation command was: /usr/src/pike/bin/smartlink gcc -I. -I/usr/src/pike/src/post_modules/ SQLite -I/usr/src/pike/src -I/usr/src/pike/src -I/usr/src/pike/src/ bundles/include -I/usr/local/include -I/usr/X11R6/include - DHAVE_CONFIG_H -g -ggdb3 -mcpu=i686 -I/usr/src/pike/src/bundles/ include -I/usr/local/include -I/usr/X11R6/include -W -Wall -Wno- unused -Wcomment -Wformat -Wimplicit-function-declaration -Wmultichar -Wswitch -Wuninitialized -Wpointer-arith -Wchar-subscripts -Wno-long- long -c sqlite.c -o sqlite.o make[2]: *** [sqlite.o] Error 1 make[2]: Leaving directory `/var/src/pike/src/post_modules/SQLite' make[1]: *** [SQLite] Error 1
Best regards,
Marc Dirix
I don't knwo that it's a bug per se, because I think you'll end up with a SQLite module that doesn't actually do anything (much like the other database access modules). It's actually kind of frustrating, but I understand the rationale behind doing it that way (avoiding non-existent symbol errors during compilation, etc).
To the point of the error, there doesn't actually appear to be an error... did it get printed out somewhere outside of what you included in your mail?
Bill
On Wed, 19 Apr 2006, Marc Dirix wrote:
This might be a configure bug, although it doesn't find SQLite prerequisites it tries to compile anyway.
...
The start of the error was way up, I didn't copy the whole error message, because I didn't want to overload the message. I don't want to have a SQLite module, working or not, I just want to have a compiled version of Pike which SQLite is blocking, regardless of adding --without-SQLite to configure.
Op 19-apr-2006, om 22:18 heeft Bill Welliver het volgende geschreven:
I don't knwo that it's a bug per se, because I think you'll end up with a SQLite module that doesn't actually do anything (much like the other database access modules). It's actually kind of frustrating, but I understand the rationale behind doing it that way (avoiding non- existent symbol errors during compilation, etc).
To the point of the error, there doesn't actually appear to be an error... did it get printed out somewhere outside of what you included in your mail?
Bill
On Wed, 19 Apr 2006, Marc Dirix wrote:
This might be a configure bug, although it doesn't find SQLite prerequisites it tries to compile anyway.
...
You could just remove the SQLite directory temporarily. It [the error message] might be helpful to the original developers of that module (Nilsson was one of them, iirc).
Bill
On Wed, 19 Apr 2006, Marc Dirix wrote:
The start of the error was way up, I didn't copy the whole error message, because I didn't want to overload the message. I don't want to have a SQLite module, working or not, I just want to have a compiled version of Pike which SQLite is blocking, regardless of adding --without-SQLite to configure.
Yes, I already did that, but it does contain a bug, and reporting a bug is always good, right?
Except you still haven't reported the actual error you got, just the "make failed" part. If you send the actual error, fixing it will require less effort which helps improve the chances of a more immediate fix.
Just an idea.
Yes, I already did that, but it does contain a bug, and reporting a bug is always good, right?
Except you still haven't reported the actual error you got, just the "make failed" part. If you send the actual error, fixing it will require less effort which helps improve the chances of a more immediate fix.
Just an idea.
Yes, I already did that, but it does contain a bug, and reporting a bug is always good, right?
Except you still haven't reported the actual error you got, just the "make failed" part. If you send the actual error, fixing it will require less effort which helps improve the chances of a more immediate fix.
Just an idea.
Yes, I already did that, but it does contain a bug, and reporting a bug is always good, right?
Except you still haven't reported the actual error you got, just the "make failed" part. If you send the actual error, fixing it will require less effort which helps improve the chances of a more immediate fix.
Just an idea.
Because, I didn't get cvs to check the SQLite dir again, I hadn't had the chance to recompile.
Please find the error output of the make attached.
Hope it helps.
On Wed, Apr 19, 2006 at 10:10:01PM +0000, David Hedbor (Amazon.com) @ Pike (-) developers forum wrote:
Yes, I already did that, but it does contain a bug, and reporting a bug is always good, right?
Except you still haven't reported the actual error you got, just the "make failed" part. If you send the actual error, fixing it will require less effort which helps improve the chances of a more immediate fix.
Just an idea.
looks like the module could use some conditionals based on configure tests.
Bill
On Thu, 20 Apr 2006, Marc DIrix wrote:
Because, I didn't get cvs to check the SQLite dir again, I hadn't had the chance to recompile.
Please find the error output of the make attached.
Hope it helps.
On Wed, Apr 19, 2006 at 10:10:01PM +0000, David Hedbor (Amazon.com) @ Pike (-) developers forum wrote:
Yes, I already did that, but it does contain a bug, and reporting a bug is always good, right?
Except you still haven't reported the actual error you got, just the "make failed" part. If you send the actual error, fixing it will require less effort which helps improve the chances of a more immediate fix.
Just an idea.
pike-devel@lists.lysator.liu.se
-
Bill Welliver -
David Hedbor (Amazon.com) ＠ Pike (-) developers forum -
Marc Dirix