memory leak in floatobject.c

[84401114-8e59-4056-83cb-632106c0b648]

BPO	3369
Nosy	@brettcannon, @mdickinson, @kristjanvalur
Files	tmp4.patch: patch for issue

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields: ```python assignee = 'https://github.com/mdickinson' closed_at = created_at = labels = ['interpreter-core', 'easy', 'performance'] title = 'memory leak in floatobject.c' updated_at = user = 'https://github.com/kristjanvalur' ``` bugs.python.org fields: ```python activity = actor = 'mark.dickinson' assignee = 'mark.dickinson' closed = True closed_date = closer = 'mark.dickinson' components = ['Interpreter Core'] creation = creator = 'kristjan.jonsson' dependencies = [] files = ['10901'] hgrepos = [] issue_num = 3369 keywords = ['patch', 'easy'] message_count = 4.0 messages = ['69721', '69725', '69830', '70126'] nosy_count = 3.0 nosy_names = ['brett.cannon', 'mark.dickinson', 'kristjan.jonsson'] pr_nums = [] priority = 'normal' resolution = 'fixed' stage = None status = 'closed' superseder = None type = 'resource usage' url = 'https://bugs.python.org/issue3369' versions = ['Python 3.0'] ```

[84401114-8e59-4056-83cb-632106c0b648]

When a nan or inf is generated in PyFloat_FromString(), a temporary buffer may be left behind. Patch attached. Found using Purify

[mdickinson]

Does this also apply to 2.6?

[mdickinson]

Does this also apply to 2.6?

Looks like it doesn't. :-).

The patch needs to touch the newly arrived comparison with "infinity", as well as the "inf" and "nan" comparisons.

[mdickinson]

Fixed in r65179. I modified the patch slightly to include the "infinity" check, and to change "if(s_buffer)" to the slightly more explicit "if (s_buffer != NULL)".

Thanks!