summaryrefslogtreecommitdiff
path: root/cffi/api.py
Commit message (Expand)AuthorAgeFilesLines
* Add locking. Not really tested, apart from the absence of double locking.thread-safeArmin Rigo2013-11-091-37/+68
* Clean up. (May also fix an obscure bug...)Armin Rigo2013-11-091-7/+12
* Docstring fixArmin Rigo2013-10-211-1/+1
* issue 102: allow ffi.typeof() to work on functions taking a structArmin Rigo2013-09-111-0/+3
* Bah, fixArmin Rigo2013-08-021-1/+2
* Propagate the original OSError, which contains information returned byArmin Rigo2013-06-041-2/+2
* issue 87: first stabArmin Rigo2013-05-221-3/+9
* Python 3 compatArmin Rigo2013-05-101-1/+1
* Try to preserve the exact error messageArmin Rigo2013-05-101-2/+2
* fix handling w/ py3 hasattr, which only swallows AttributeErrorsPhilip Jenvey2013-04-051-1/+4
* Fix the docstring.Armin Rigo2013-04-041-1/+1
* Test and fix for pull request #11: anonymous enums.Armin Rigo2013-03-291-1/+2
* Fix (also for Python 3).Armin Rigo2013-03-291-1/+1
* A bit hackish, but solves exactly issue #71: ffi.typeof(builtin_function).Armin Rigo2013-03-281-1/+20
* Fix for a potential race condition in multithreaded programs (hard to test).Armin Rigo2013-03-271-2/+2
* Issue #67Armin Rigo2013-03-271-0/+12
* Check that the backend's version matches the frontend'sArmin Rigo2013-03-081-0/+2
* (lazka, arigo) Change dlopen() to accept either a full path or a libraryArmin Rigo2013-02-221-7/+7
* ImplementationArmin Rigo2013-02-101-1/+0
* Python 3 compatArmin Rigo2013-02-071-0/+6
* issue #50: fix, at least as much as I could find tests for.Armin Rigo2013-01-021-11/+23
* Fix Python 3.1 compat (callable)Simon Sapin2012-12-271-0/+8
* Untested (how..?): record dependencies across ffi.include()Armin Rigo2012-12-041-0/+3
* ffi.include().Armin Rigo2012-11-301-0/+11
* Issue #42: Test and fixArmin Rigo2012-11-281-6/+5
* Remove the default backend's fallback to ctypes, with comments.Armin Rigo2012-11-231-8/+7
* Simplification.Armin Rigo2012-11-171-3/+1
* Minor performance improvement for PyPy.Armin Rigo2012-11-171-3/+11
* Fix issue #32: turn types like "size_t" into primitive types from theArmin Rigo2012-11-121-12/+1
* Kill again ffi.inspecttype(), and replace it with read-only attributesArmin Rigo2012-11-111-5/+0
* Issue #22: Add ffi.inspecttype(). See the doc.Armin Rigo2012-11-101-0/+5
* Fix the documentation.Armin Rigo2012-10-291-2/+7
* Fix docstring to work around xemacs syntax highlighting issues.Armin Rigo2012-10-261-1/+1
* FILE limited supportArmin Rigo2012-10-191-1/+1
* Keep the loaded libraries alive as long as the 'ffi' object is kept alive.Armin Rigo2012-10-091-1/+5
* Change the default dlopen() flags from RTLD_LAZY to RTLD_NOW.Armin Rigo2012-09-241-5/+10
* Kill _cffi_backend.offsetof(), using only _cffi_backend.typeoffsetof().Armin Rigo2012-09-191-1/+1
* ffi.addressof(struct, field).Armin Rigo2012-09-181-3/+6
* Remove a special case from _cffi_backend, and replace it with a generalArmin Rigo2012-09-181-0/+12
* Make ffi.callback work both in normal mode and in "decorator mode".Armin Rigo2012-09-181-8/+16
* Still trying to fix up the ordering of building stuff. One test passes,Armin Rigo2012-09-151-3/+4
* Add ffi.CData and ffi.CType.Armin Rigo2012-09-041-0/+1
* Change the default directory in which verify() builds the C module. NowArmin Rigo2012-09-041-3/+4
* hg merge default, and try to resolve all conflicts.Armin Rigo2012-08-121-32/+64
|\
| * Finally found out the "right" way to implement ffi.gc(), in just aArmin Rigo2012-08-091-0/+12
| * update docArmin Rigo2012-08-031-1/+1
| * As discussed on the mailing list: str() -> ffi.string()Armin Rigo2012-08-031-0/+18
| * Speed up.Armin Rigo2012-07-301-32/+34
* | An attempt to port cffi to python3.Amaury Forgeot d'Arc2012-07-291-12/+12
|/
* Fix: the code incorrectly accepted e.g. 'ffi.new(ffi.new("int*"))',Armin Rigo2012-07-271-22/+29