Hello @all,

On Thu, 9 Jan 2003, Stenberg, Daniel wrote:
 
> Just be aware of that I don't know if anyone ever has done
> this on Windows before. I can't say I know any reason why
> it would work differently, but you never know... :-)

Never ever... =]

> So, do you get a file output that you can browse manually at
> least?

Yep!

> --- testfile start---
> MEM url.c:241 malloc(67260) = 0x8104304
> MEM url.c:250 malloc(256) = 0x81149d4
> --- testfile end ---
> $ memanalyze.pl < testfile
> Leak detected: memory still allocated: 67516 bytes
> At 8104304, there's 67260 bytes.
> allocated by url.c:241
> At 81149d4, there's 256 bytes.
> allocated by url.c:250

> If this isn't shown, [...]

Nope, "memanalyze.pl < testfile" spits out an empty line...

> [...] can you post your memdebug dump here and let me try it
> and see if I can figure out why memanalyze fails?

Sure, thank you. Please find attached the memdump of a
small testapp that misses the curl-cleanup; "memanalyze.pl"
shows nothing.
BTW: You can feed "memanalyze" with arbitrary files on Win32,
nothing happens...
(Sorry to say that i don't know much about Perl, i've just
installed "ActiveState ActivePerl 5.8" for Win32, the examples
seem to run...)

> Daniel Stenberg -- curl, cURL, Curl, CURL. Groks URLs. <<
> File: Markus Henke.vcf >>

Many thanx & Ciao, Markus

P.S.: WRT Win32 related probs i dare to point you to a
posting of mine that bothers with the "you can't pass
FILE* between DLLs"-thing, that (IMHO) is somehow
mistakable...
http://curl.haxx.se/mail/lib-2003-01/0056.html

-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com