summaryrefslogtreecommitdiff
path: root/testing
Commit message (Collapse)AuthorAgeFilesLines
* Skip a couple of new tests on Windows, because this method doesn't seemArmin Rigo2015-12-071-0/+3
| | | | able to capture the C-level fprintf(stderr)
* py3k compatArmin Rigo2015-12-074-12/+26
|
* hg merge defaultArmin Rigo2015-11-233-1/+72
|\
| * Copy the PyPy behavior on CPython too: two calls to new_handle(x),Armin Rigo2015-11-231-0/+5
| | | | | | | | even with the same x, now return cdatas that compare as different.
| * issue #233: ffi.init_once()Armin Rigo2015-11-232-0/+66
| |
| * fixArmin Rigo2015-11-171-1/+1
| |
* | Change the @ffi.def_extern() decorator to not automatically replace theArmin Rigo2015-11-201-14/+16
| | | | | | | | | | | | function with the cdata. You need to get the cdata from the lib explicitly. This should make it clearer that there is only one cdata, even if you apply the decorator again.
* | Write one error message directly to stderr instead of sys.stderr. ThisArmin Rigo2015-11-183-37/+44
| | | | | | | | | | lets us avoid taking the GIL, which might crash in case the Python interpreter is not initialized at all.
* | update the code to use 'extern "Python"'Armin Rigo2015-11-181-62/+64
| |
* | Fix cparserArmin Rigo2015-11-181-8/+22
| |
* | hg merge defaultArmin Rigo2015-11-161-2/+5
|\ \ | |/
| * Skip this test when run inside pypy, where there is no ../c/commontypes.cArmin Rigo2015-11-161-2/+5
| |
* | More tests for what I thought would work because it uses the same codeArmin Rigo2015-11-151-1/+42
| | | | | | | | as ffi.callback(). Alas, it didn't
* | tests tests testsArmin Rigo2015-11-131-8/+61
| |
* | more testsArmin Rigo2015-11-131-7/+38
| |
* | More tests, make the name optional in ffi.call_python()Armin Rigo2015-11-131-2/+42
| |
* | ffi.call_python()Armin Rigo2015-11-131-0/+10
| |
* | passing testArmin Rigo2015-11-131-2/+11
| |
* | in-progressArmin Rigo2015-11-131-0/+1
| |
* | tweak tweak tweak until we can at least read the function pointer out ofArmin Rigo2015-11-131-0/+1
| | | | | | | | the 'lib' object
* | 'long double' is the only primitive type that doesn't necessarily fitArmin Rigo2015-11-131-0/+10
| | | | | | | | inside uint64_t, for now
* | Pass and return structsArmin Rigo2015-11-131-0/+12
| |
* | Finally found a way to arrange the generated codeArmin Rigo2015-11-131-0/+11
| |
* | Parsing CFFI_CALL_PYTHON in the cdefArmin Rigo2015-11-121-0/+14
|/
* Give a warning instead of crashing; in this way we can call the nextArmin Rigo2015-11-061-0/+3
| | | | | release 1.3.1 as planned and give some advance warning to users relying on this.
* Complain more readily about opaque enums, by refusing to guess whichArmin Rigo2015-11-051-1/+10
| | | | integer type (unsigned/signed, int/long) it is meant to be.
* One of the Windows common types was lost somehow. Add a testArmin Rigo2015-11-031-2/+18
| | | | that at least dependencies are there.
* Show <ctype 'FILE'> instead of <ctype 'struct _IO_FILE'>Armin Rigo2015-11-031-1/+1
|
* Allow different formattings of the underlying "%p"Armin Rigo2015-11-031-1/+2
|
* Issue #228: do the same for "FILE". Only remaining case now is theArmin Rigo2015-11-032-0/+29
| | | | | | | very obscure Windows type UNICODE_STRING. I think it is unlikely that someone hits the case of their cdef not mentioning UNICODE_STRING but their runtime code still wants to use it, so I'll close this issue.
* Issue #228: "bool" not working in out-of-line FFI objects. Same problemArmin Rigo2015-11-025-38/+39
| | | | | | | with all Windows common types. Mostly fixed by moving the list of common types to C code. The only remaining corner case I can think of is "FILE", which works only if it was mentioned in the cdef while building the out-of-line module.
* Found a pycparser issue. Work around it by adding some parentheses inArmin Rigo2015-11-022-0/+40
| | | | the csource passed to it, to avoid the buggy cases.
* Propagate "volatile" in addition to "const" and "restrict"Armin Rigo2015-11-021-3/+10
|
* Fix the error we get for 'int f(unknown_type);'Armin Rigo2015-11-021-0/+7
|
* wchar_t can be signed or not, apparently, even on the same platform (armArmin Rigo2015-10-222-6/+8
| | | | linux)
* Backed out changeset fbe55ed7e5e2Armin Rigo2015-10-211-7/+0
| | | | It's probably better written in the docs instead.
* Issue #225: Ignore basic SAL annotations on Windows.Armin Rigo2015-10-201-0/+7
|
* fix corner case of writing "typdef char INT8, *PINT8;" on WindowsArmin Rigo2015-10-181-0/+6
|
* py3 and no-g++ fixesArmin Rigo2015-10-161-2/+2
|
* win32 fixArmin Rigo2015-10-161-1/+2
|
* win32 fixArmin Rigo2015-10-161-1/+4
|
* win64 fixesArmin Rigo2015-10-161-5/+5
|
* win64 fixArmin Rigo2015-10-161-1/+1
|
* win64 fixArmin Rigo2015-10-161-2/+2
|
* win64 fixArmin Rigo2015-10-161-1/+1
|
* win64 fixArmin Rigo2015-10-161-7/+13
|
* Comment out the prints (not py3)Armin Rigo2015-10-082-9/+9
|
* ffi.memmove()Armin Rigo2015-10-082-0/+106
|
* remove the debug printfsArmin Rigo2015-10-062-24/+24
|
* Parse __cdecl and __stdcall in the built-in parserArmin Rigo2015-10-063-20/+41
|
View Lorry Controller Status