summaryrefslogtreecommitdiff
path: root/c
Commit message (Expand)AuthorAgeFilesLines
* Issue #213: in case we'd give the error message "initializer for ctypeArmin Rigo2015-07-092-6/+42
* Explicitly protect against variables that end up resolving at address NULL,Armin Rigo2015-07-062-5/+21
* Port the new_allocator() work from pypyArmin Rigo2015-07-062-32/+162
* Found a simpler and more efficient way to implement any ffi.gc().Armin Rigo2015-07-064-144/+113
* Found a simpler way to implement the in-line ffi.gc() on top of theArmin Rigo2015-07-062-17/+0
* Minor clean upArmin Rigo2015-07-051-24/+24
* Generalize the test for pypyArmin Rigo2015-07-041-2/+2
* 'onerror' can also return a custom value to returnArmin Rigo2015-07-042-7/+43
* New argument "onerror" on ffi.callback()Armin Rigo2015-07-043-12/+81
* Use the logic in cgc.c to implement ffi.gc() also for the pure PythonArmin Rigo2015-07-042-0/+17
* Bump the version number to 1.2.0Armin Rigo2015-07-032-2/+2
* Refactor the way global variables are accessed. Now, every access toArmin Rigo2015-07-032-17/+58
* Issue #209: check for dereferencing NULL pointersArmin Rigo2015-06-302-7/+25
* Test and fix: Python 3.5 (at least the current beta) is unhappyArmin Rigo2015-06-261-0/+5
* Redo lib.__dict__, like in the in-line mode. Figured out that it's likeArmin Rigo2015-06-191-0/+31
* "from foo.lib import *"Armin Rigo2015-06-151-18/+37
* Allow "from _cffimodule.lib import a, b, c". (thanks DanielHolth on irc)Armin Rigo2015-06-151-2/+14
* Bump version to 1.1.2Armin Rigo2015-06-092-2/+2
* Some detailsArmin Rigo2015-06-072-2/+2
* Bump version number to 1.1.1Armin Rigo2015-06-072-2/+2
* Fix the logic to handle ffi.gc(x) being called several times with equalArmin Rigo2015-06-062-44/+88
* In the type parser, escape error messages and don't display the inputArmin Rigo2015-06-021-9/+35
* Issue #203: add keyword arguments to ffi.string(), ffi.buffer(),Armin Rigo2015-06-013-16/+22
* Leave with Py_FatalError() if we detect that the object is reallyArmin Rigo2015-05-311-0/+6
* The next release will be 1.1.0.Armin Rigo2015-05-302-2/+2
* Tweaks and minor fixesArmin Rigo2015-05-301-5/+10
* Issue 200: bad interaction between "ffi.typeof(function_t)" andArmin Rigo2015-05-302-12/+23
* Better compatibility: detect if we have an old cffi running a new ABIArmin Rigo2015-05-281-0/+3
* ABI mode: allow constants of any type, which are looked up from the libArmin Rigo2015-05-281-12/+19
* more about issue 198: Test and fixArmin Rigo2015-05-281-1/+4
* Issue #198 bis: fix for constants of unknown sizeArmin Rigo2015-05-281-1/+4
* Issue 198: corrupted constant of type 'struct'Armin Rigo2015-05-281-1/+11
* Bump version numberArmin Rigo2015-05-262-2/+2
* ffi.addressof(lib, "function") now returns a regular cdata function pointerArmin Rigo2015-05-262-23/+40
* CommentArmin Rigo2015-05-261-0/+5
* Bump the version number to 1.0.3Armin Rigo2015-05-252-2/+2
* Bump the version number to 1.0.2Armin Rigo2015-05-222-2/+2
* clarifyArmin Rigo2015-05-211-1/+1
* Bump the version to 1.0.1Armin Rigo2015-05-212-2/+2
* Windows: pass along the result (success/error) of FreeLibraryArmin Rigo2015-05-201-2/+1
* Forgot about the version checking and specifying in the ABI-mode modulesArmin Rigo2015-05-182-4/+13
* test and fixArmin Rigo2015-05-173-2/+4
* Add RTLD_xxx flags to the CompiledFFI typeArmin Rigo2015-05-162-18/+40
* ffi.dlclose(). Global variables.Armin Rigo2015-05-162-18/+24
* Bump the internal version numberArmin Rigo2015-05-161-2/+2
* Test and fix for struct names from included ffisArmin Rigo2015-05-161-1/+1
* Includes of dlopen-style ffis. Negative constants in "static const int = ...".Armin Rigo2015-05-161-4/+18
* Various tweaks until we get some more-or-less complete dlopen() supportArmin Rigo2015-05-164-25/+111
* StructsArmin Rigo2015-05-161-1/+6
* Windows fixArmin Rigo2015-05-131-1/+2
View Lorry Controller Status