3006 - [PP] layout/html/tests fails to link on Solaris

Status: RESOLVED WORKSFORME

(Core Graveyard :: Viewer App, defect, P2)

Description

[Chris McAfee]

layout/html/tests fails to link on Solaris:
Undefined                       first referenced
 symbol                             in file
fstream::fstream(int, char const *, int,
int)../../../dist/./bin/libraptorhtmlpars.so
fstreambase::close(void)            ../../../dist/./bin/libraptorhtmlpars.so
fstream::~fstream(void)             ../../../dist/./bin/libraptorhtmlpars.so

These are in libstdc++.a, I'm confused.  Something not pulling
this in?  This links Ok on linux, but linux also has a libstdc++.so.

I was building on mocha, GNU stuff is local in /usr/local.

Comment 1

[Brian Ostrom]

This is probably a case where you have to explicitly include
-lstdc++ on the link line.  What does the link line look like
for you (you didn't include it in this bug report)?

Comment 2

[Chris McAfee]

I tried explicitly adding -lstdc++ to the link line, no dice.

Comment 3

[Paul MacQuiddy]

per leger, assigning QA contacts to all open bugs without QA contacts according
to list at http://bugzilla.mozilla.org/describecomponents.cgi?product=Browser

Comment 4

[leger]

As long as this works on Linux, not a Dogfood blocker.  Setting QA Assigned to
petersen.  How does this look for Linux with latest build...Ok?  willmove to M4
once we know for sure.

Comment 5

[leger]

Moving to M5, not a Dogfood blocker.

Comment 6

[Brian Ostrom]

I've turned off tests for now (unless ENABLE_TESTS is defined).
Too many platforms are unable to build, link, or run many of our
existing "tests".  We can worry about these "non-essential" bits
of code later, I think.

Comment 7

[Chris Petersen]

Verifying as a Reminder.

Comment 8

[Christopher Hoess (gone)]

LATER is deprecated per bug 35839.

Comment 9

[Christopher Hoess (gone)]

->Bernd

Comment 10

[Bernd]

I don't have solaris. I cant test it, neither do I want to fix this. There are
several old layout test routines which have worked at some time but aren't
working now. They breakage is mostly due to string or xpcom foo, which is
definetely not in my area of interest.

Comment 11

[Bernd]

 I did not cause this bug and I have no intention to fix it.

Comment 12

[Chris McAfee]

> I did not cause this bug and I have no intention to fix it.

Nothing would ever get done if we put comments like this in bugs.
A much better comment would be, "this bug needs a new owner, please help".

Adding helpwanted keyword.

Comment 13

[Bernd]

Sorry for my rude comment, Chris your comment is correct. No intention ==> I
can't fix it. I asked on #mozilla what should I do with this bug and did not get
help, so I ended up with a bug which is in my oppinion a wontfix for me. And I
did not know to whom pass this hot (dead) potato. Before this bug  converts into
a zombie bug on my buglist because I would fix every other bug on that list
before I would touch this, I thought it was a good idea to state clear what will
happen with this bug. Obviously it wasn't a very bright idea. 

Comment 14

[Chris McAfee]

this works for me on worms, marking so.