On Dec 23, 2007 6:23 PM, Nagy Gabor <ngaba@bibl.u-szeged.hu> wrote:
Nagy Gabor wrote:
Updated patch attached without the memory leaks and adding return of removal success.
Cheers, Allan
Is newcachedir freed?;-) Bye
No... but it wasn't in alpm_option_add_cachedir() either and I was trying for coding style consistency. :) Both are fixed in attached patch. No. You mustn't free it in add_cachedir, because alpm_list_add just adds a POINTER to the (handle->cachedirs) list. (So you can always think alpm_list as a list of pointers.)
Ok, enough of this junk. Sorry to rain down on the holiday spirit, but can we please not be cryptic on this mailing list with cute little remarks and smiley faces?
Something like this would have saved us all 5 minutes of our time. "I noticed that newcachedir was not freed in remove_cachedir- as far as I can tell, it needs to be freed because we aren't adding it to a list and it isn't needed outside of the function."
-Dan
1. If you refer to "Is newcachedir freed?;-)", then you should have answered to my previous mail, because we had an _other_ bug here. 2. These "junks" were addressed to Allen, who fortunately understood them. He is new to the code (as he said), and I thought he need other information than you/us (alpm_list_t explanation). But if this is forbidden here, next time I will write a direct e-mail to him or be silent. Merry Christmas!