| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Python3 compat | Armin Rigo | 2016-06-05 | 1 | -1/+1 |
| | | |||||
| * | Haaaack! Have 'lib.__class__' return &PyModule_Type. It makes | Armin Rigo | 2016-05-25 | 1 | -3/+7 |
| | | | | | | | | | | help(lib) behave and display a nice module-like view of your compiled lib. Note that this is seriously unexpected, but I hope nobody should get hit by it by looking at 'lib.__class__' directly... Might be reverted if I hear about someone :-) | ||||
| * | fix: Lib objects didn't have the cyclic GC enabled | Armin Rigo | 2016-05-25 | 1 | -3/+3 |
| | | |||||
| * | Support help(lib.foo) | Armin Rigo | 2016-04-19 | 1 | -36/+69 |
| | | |||||
| * | Another error case missing | Armin Rigo | 2015-12-11 | 1 | -0/+2 |
| | | |||||
| * | Add a __class__ attribute on lib objects | Armin Rigo | 2015-12-11 | 1 | -3/+11 |
| | | |||||
| * | update the code to use 'extern "Python"' | Armin Rigo | 2015-11-18 | 1 | -2/+2 |
| | | |||||
| * | in-progress | Armin Rigo | 2015-11-13 | 1 | -1/+1 |
| | | |||||
| * | tweak tweak tweak until we can at least read the function pointer out of | Armin Rigo | 2015-11-13 | 1 | -0/+10 |
| | | | | | the 'lib' object | ||||
| * | Explicitly protect against variables that end up resolving at address NULL, | Armin Rigo | 2015-07-06 | 1 | -2/+2 |
| | | | | | which can occur in random cases like test_macro_var_callback | ||||
| * | Refactor the way global variables are accessed. Now, every access to | Armin Rigo | 2015-07-03 | 1 | -13/+29 |
| | | | | | | "lib.var" calls a generated C function to get "&var". This fixes issue #212 and also make it work in more cases, like __thread variables. | ||||
| * | Test and fix: Python 3.5 (at least the current beta) is unhappy | Armin Rigo | 2015-06-26 | 1 | -0/+5 |
| | | | | | with module-like objects without a '__name__' | ||||
| * | Redo lib.__dict__, like in the in-line mode. Figured out that it's like | Armin Rigo | 2015-06-19 | 1 | -0/+31 |
| | | | | | | lib.__all__: a special attribute that can be shadowed if necessary because it's not essential. | ||||
| * | "from foo.lib import *" | Armin Rigo | 2015-06-15 | 1 | -18/+37 |
| | | |||||
| * | Some details | Armin Rigo | 2015-06-07 | 1 | -1/+1 |
| | | |||||
| * | Issue 200: bad interaction between "ffi.typeof(function_t)" and | Armin Rigo | 2015-05-30 | 1 | -12/+19 |
| | | | | | attribute access "lib.function" | ||||
| * | Better compatibility: detect if we have an old cffi running a new ABI | Armin Rigo | 2015-05-28 | 1 | -0/+3 |
| | | | | | | mode script that makes use of the new OP_CONSTANT feature --- now renamed to OP_DLOPEN_CONST. | ||||
| * | ABI mode: allow constants of any type, which are looked up from the lib | Armin Rigo | 2015-05-28 | 1 | -12/+19 |
| | | | | | on their first access. | ||||
| * | Issue #198 bis: fix for constants of unknown size | Armin Rigo | 2015-05-28 | 1 | -1/+4 |
| | | |||||
| * | Issue 198: corrupted constant of type 'struct' | Armin Rigo | 2015-05-28 | 1 | -1/+11 |
| | | | | | Thanks Lisandro! | ||||
| * | ffi.addressof(lib, "function") now returns a regular cdata function pointer | Armin Rigo | 2015-05-26 | 1 | -22/+39 |
| | | |||||
| * | Comment | Armin Rigo | 2015-05-26 | 1 | -0/+5 |
| | | |||||
| * | test and fix | Armin Rigo | 2015-05-17 | 1 | -0/+2 |
| | | |||||
| * | ffi.dlclose(). Global variables. | Armin Rigo | 2015-05-16 | 1 | -13/+15 |
| | | |||||
| * | Various tweaks until we get some more-or-less complete dlopen() support | Armin Rigo | 2015-05-16 | 1 | -24/+34 |
| | | |||||
| * | Python 3 compat | Armin Rigo | 2015-05-12 | 1 | -2/+2 |
| | | |||||
| * | the big Moving Files Around step | Armin Rigo | 2015-05-12 | 1 | -0/+495 |
 |
index : delta/python-packages/cffi.git | |
| foss.heptapod.net: pypy/cffi |
| summaryrefslogtreecommitdiff |