summaryrefslogtreecommitdiff
path: root/doc/source/ref.rst
Commit message (Collapse)AuthorAgeFilesLines
* Write explicitly that byte strings passed to `char *` arguments are not ↵Armin Rigo2021-12-031-5/+20
| | | | meant to live as long as the byte string is alive, but only for the duration of the call
* doc: Turn ffi.new() references into links in Overview pageRasmus Bondesson2021-07-171-0/+2
|
* Remove trailing whitespace in doc/*Rasmus Bondesson2021-07-171-1/+1
|
* 'bitbucket => heptapod' for most linksArmin Rigo2020-08-081-2/+2
|
* The point of this example is that it doesn't work with a variable, only with aArmin Rigo2020-01-311-1/+1
| | | | constant number. So don't use `n`.
* Simplify the implementation, cut down the explanations, and allowArmin Rigo2019-06-031-9/+5
| | | | | | accessing multiple items instead of just one in from_buffer("type *"), trading off security for the realisation that it makes sense to access multiple items in some cases (unlike after ffi.new()).
* Document the new ffi.from_buffer("type *")Armin Rigo2019-06-031-4/+27
|
* typoArmin Rigo2019-02-221-1/+1
|
* Issue #402Armin Rigo2019-02-221-14/+22
| | | | More rewording
* Issue #402Armin Rigo2019-02-211-3/+4
| | | | Fix bug in documentation
* tweak docsArmin Rigo2019-01-271-2/+3
|
* Implement a limited form of from_buffer-release() on pypyArmin Rigo2019-01-081-1/+3
|
* Implement ffi.from_buffer("foo[]", x)Armin Rigo2019-01-071-27/+61
| | | | | Also contains some improvements to the documentation about other recent additions
* ffi.release()Armin Rigo2019-01-051-1/+46
|
* Issue #394Armin Rigo2018-12-161-1/+14
| | | | Implement ffi.from_buffer(x, require_writable=True)
* Document for issue 340Armin Rigo2017-10-131-0/+5
|
* Write down an explicit example of what not to doArmin Rigo2017-08-311-0/+3
|
* Expand the docs about wchar_t/char16_t/char32_tArmin Rigo2017-08-311-14/+65
|
* Document ffi.gc(..., size)Armin Rigo2017-08-021-12/+32
|
* Didn't figure out how to cleanly define charN_t inside _cffi_include.h.Armin Rigo2017-06-041-4/+14
| | | | So for now, don't.
* Document the signedness of charN_tArmin Rigo2017-06-021-2/+9
|
* Document char16_t and char32_tArmin Rigo2017-06-021-12/+14
|
* DocumentationArmin Rigo2017-05-301-13/+25
|
* tweaksArmin Rigo2017-05-291-2/+2
|
* Document the missing add_memory_pressure in PyPy <= 5.6Armin Rigo2017-03-101-17/+25
|
* typoArmin Rigo2017-03-071-1/+1
|
* Document more explicitly the difference between CompiledFFI and FFI.Armin Rigo2017-02-281-0/+8
|
* issue #255: comparing primitive cdatasArmin Rigo2017-02-191-7/+7
|
* Make 'ffi.buffer' be the type of minibuffers returned by cffi, andArmin Rigo2017-02-041-0/+7
| | | | 'ffi.buffer()' now calls the constructor.
* issue300: return _Bool as Python booleans; related fixesArmin Rigo2017-01-231-5/+15
|
* Tweaks the docsArmin Rigo2017-01-041-13/+13
|
* merge strbuf-as-bufferRichard Plangger2017-01-041-4/+22
|\
| * Document explicit calls to lib.malloc()Armin Rigo2016-12-221-2/+20
| |
| * typosArmin Rigo2016-12-121-2/+2
| |
* | kill invalid_input_buffer_type, update docs to state that now only unicode ↵Richard Plangger2017-01-031-3/+2
|/ | | | objects are forbidden
* add warningArmin Rigo2016-12-091-1/+1
|
* Document 'FILE *'Armin Rigo2016-12-091-0/+32
|
* Document should_clear_after_alloc a bit moreArmin Rigo2016-11-281-1/+15
|
* Add a noteArmin Rigo2016-10-291-1/+3
|
* Decided to fix ffi.sizeof() too. Update the documentation.Armin Rigo2016-10-271-1/+10
|
* Docs: remove "char *" from the special casing in the referenceArmin Rigo2016-09-151-3/+3
|
* Update the docs according to issue #281Armin Rigo2016-08-221-1/+1
|
* Update docsArmin Rigo2016-08-061-3/+3
|
* The null_byte_after_str branch of PyPy makes ffi.from_buffer(str)Armin Rigo2016-08-021-1/+4
| | | | possible.
* english attemptsArmin Rigo2016-06-061-2/+2
|
* Rewrite the code more clearly, and fix the wrong commentArmin Rigo2016-06-061-4/+6
|
* Write down a typical ffi.new_handle() usage exampleArmin Rigo2016-06-061-0/+29
|
* Rename 'ffi' to 'ffibuilder' in the docs and in a few demos, whenArmin Rigo2016-06-051-1/+1
| | | | | | it is used in out-of-line builders. I think it makes things clearer, particularly in examples where the two 'ffi' are close together (and even sometimes used in the same sentence...)
* Add a warningArmin Rigo2016-06-031-0/+3
|
* Document 61e03368485c and 4d6671385e08. Remove the "new in version 0.9"Armin Rigo2016-06-031-4/+11
| | | | as deprecated; keep only "new in version 1.x".