summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Experimental serialisation of cffi.model types into Python code that ↵split-verifyJeremy Thurgood2013-10-072-5/+99
| | | | instantiates them.
* Bootstrapping documentationStefano Rivera2013-10-071-0/+15
|
* Port setuptools tests to FFIBuilderStefano Rivera2013-10-0712-77/+102
|
* Port distutils_package_2 to FFIBuilderStefano Rivera2013-10-075-26/+35
|
* We'd want to ship ffibuilder in the sdistStefano Rivera2013-10-071-0/+1
|
* Support ext_packageStefano Rivera2013-10-071-6/+9
|
* Port test_distutils_package_1 to FFIBuilderStefano Rivera2013-10-076-27/+38
|
* Allow specifying a parent packageStefano Rivera2013-10-071-15/+23
|
* More whitespaceStefano Rivera2013-10-061-1/+1
|
* Port distutils_module to FFIBuilderStefano Rivera2013-10-064-24/+32
|
* Create lib_dir if it doesn't existStefano Rivera2013-10-061-0/+2
|
* Avoid setting force_generic_engine twiceStefano Rivera2013-10-061-2/+2
|
* Add verifier_distutils_module, so that we can transition all the other ↵Stefano Rivera2013-10-063-0/+31
| | | | zintegration tests to FFIBuilder
* PEP-8 order the test_zintegration importsStefano Rivera2013-10-061-1/+6
|
* I think this is the most horrible thing I have written in months, but... ↵Jeremy Thurgood2013-10-061-3/+6
| | | | pickle all the _declarations.
* Fix (and test) ffi proxy functions.Jeremy Thurgood2013-10-062-70/+89
|
* Add cffi.packaging module which has everything you need for your setup.pyStefano Rivera2013-10-061-0/+21
|
* Better API for generated module.Jeremy Thurgood2013-10-062-32/+127
|
* Move FFIBuilder into its own module.Jeremy Thurgood2013-10-063-73/+75
|
* Collect built filepaths.Jeremy Thurgood2013-10-062-4/+27
|
* Clean up a few things, fix python3 compat.Jeremy Thurgood2013-10-062-19/+15
|
* FFIBuilder implementation.Jeremy Thurgood2013-10-054-1/+146
|
* (arigo, fijal, jerith) More complicated (and better) juggling of compiler ↵Jeremy Thurgood2013-10-051-7/+10
| | | | args for gcc/clang in verify tests.
* A redo of pull request 19: dynamically determine if the C compilerArmin Rigo2013-09-253-6/+15
| | | | | supports "__thread" or not on this platform. Done slightly more simply in setup.py.
* Two tests from the mailing list, and fixes: bogus "const" detection,Armin Rigo2013-09-144-11/+41
| | | | and global array variables should never be "const".
* issue 102: allow ffi.typeof() to work on functions taking a structArmin Rigo2013-09-113-2/+18
| | | | | argument (with vengine_gen). Note that it still fails on functions taking an auto-completed struct. Unsure if and how to fix.
* Tweak: like this, pypy-shadowstack crashes systematicallyArmin Rigo2013-09-031-0/+1
|
* Fix bitfields on big-endian machinesArmin Rigo2013-09-012-8/+61
|
* In this case, we must call backend.load_library() with a pathArmin Rigo2013-08-071-1/+3
| | | | that contains a '/'.
* Update the version number to 0.7.2Armin Rigo2013-08-025-5/+6
|
* Bah, fixArmin Rigo2013-08-021-1/+2
|
* Remove this temporary warningArmin Rigo2013-08-011-2/+0
|
* Update the versionArmin Rigo2013-08-015-7/+9
|
* Fix for win64Armin Rigo2013-08-011-1/+6
|
* Fix the docArmin Rigo2013-08-011-1/+2
|
* Add a warningArmin Rigo2013-08-011-1/+3
|
* The Win64 problem with the .asm was fixed by checking in the .obj file.Armin Rigo2013-08-011-5/+6
|
* Removed tag release-0.1Armin Rigo2013-08-010-0/+0
|
* Removed tag release-0.2Armin Rigo2013-08-010-0/+0
|
* Removed tag release-0.2.1Armin Rigo2013-08-010-0/+0
|
* Removed tag release-0.3Armin Rigo2013-08-010-0/+0
|
* Removed tag release-0.4Armin Rigo2013-08-010-0/+0
|
* Removed tag release-0.4.1Armin Rigo2013-08-010-0/+0
|
* Removed tag release-0.4.2Armin Rigo2013-08-010-0/+0
|
* Remove the hg tags. We're using hg branches 'release-*' nowadays.Armin Rigo2013-08-010-0/+0
|
* Don't care too much about the ctypes backendArmin Rigo2013-08-011-0/+2
|
* Write "status of Win64".Armin Rigo2013-08-011-0/+3
|
* Py3k fixArmin Rigo2013-08-011-2/+2
|
* hg merge win64Armin Rigo2013-08-013-5/+15
|\
| * Close branch, ready to mergewin64Armin Rigo2013-08-010-0/+0
| |