this will fix the problem: ? diff.txt Index: nsStringBundle.cpp =================================================================== RCS file: /cvsroot/mozilla/intl/strres/src/nsStringBundle.cpp,v retrieving revision 1.60 diff -p -r1.60 nsStringBundle.cpp *** nsStringBundle.cpp 2000/04/25 05:29:58 1.60 --- nsStringBundle.cpp 2000/04/25 15:10:54 *************** nsStringBundle::GetStringFromName(const *** 145,151 **** return ret; } ! NS_IMPL_ISUPPORTS1(nsStringBundle, nsIStringBundle) /* void GetStringFromID (in long aID, out wstring aResult); */ NS_IMETHODIMP --- 145,151 ---- return ret; } ! NS_IMPL_THREADSAFE_ISUPPORTS1(nsStringBundle, nsIStringBundle) /* void GetStringFromID (in long aID, out wstring aResult); */ NS_IMETHODIMP
yikes, that's odd.
Status: NEW → ASSIGNED
Whiteboard: fix in hand
Hi, Alec: Did you ask Warren to review the threadsafe portion of the constructor? He was not in favor of using the threadsfafe version of the MACRO to warrant the thread safety of the stringbunlde class.
I figured out what's going on here. Basically I have extended the lifetime of a stringbundle now, because it's cached. What happens is that thread 1 asks for a stringbundle, and thread 2 later asks for the same string bundle. Since the stringbundle was created on thread 1, it asserts. This means I have to 1) make sure the stringbundle itself is threadsafe (should be trivial) 2) fix the macro as seth suggested.
you'd probably only see the assertion if you run with the PSM. I've noticed that when I use the PSM (for secure imap, http or news) I get tons of these assertions, when I don't with out the PSM. this is one of those assertions.
ah, good to know. thanks
Ok I've fixed both of these in my tree. if this ends up being a blocker I can check in.
fixed this yesterday.
Status: ASSIGNED → RESOLVED
Last Resolved: 19 years ago
Resolution: --- → FIXED
Sorry for the spam. New QA Contact for Browser General. Thanks for your help Joseph (good luck with the new job) and welcome aboard Doron Rosenberg
QA Contact: jelwell → doronr
Alec Flett can you update the QA contact on this bug. Thanks
to what? I don't know who the QA contact should be...tao?
You need to log in before you can comment on or make changes to this bug.