diff options
author | Petri Lehtinen <petri@digip.org> | 2018-02-09 09:33:41 +0200 |
---|---|---|
committer | Petri Lehtinen <petri@digip.org> | 2018-02-09 09:35:53 +0200 |
commit | a6138a07b6f302567254c1f51193d9017dcb8d3e (patch) | |
tree | a39f982e663e9a2d908a39bb17c988b6f4f4ead4 /doc/apiref.rst | |
parent | 9e5af7c3b77f81a7d1720073dca600e51fe9962d (diff) | |
download | jansson-doc-utf-8.zip jansson-doc-utf-8.tar.gz jansson-doc-utf-8.tar.bz2 |
Document encoding requirements for callbacksdoc-utf-8
Original patch by @phst.
Fixes #369.
Diffstat (limited to 'doc/apiref.rst')
-rw-r--r-- | doc/apiref.rst | 18 |
1 files changed, 15 insertions, 3 deletions
diff --git a/doc/apiref.rst b/doc/apiref.rst index a8cf8d6..f5493c6 100644 --- a/doc/apiref.rst +++ b/doc/apiref.rst @@ -1119,6 +1119,10 @@ These functions output UTF-8: the length of the buffer, and *data* is the corresponding :func:`json_dump_callback()` argument passed through. + *buffer* is guaranteed to be a valid UTF-8 string (i.e. multi-byte + code unit sequences are preserved). *buffer* never contains + embedded null bytes. + On error, the function should return -1 to stop the encoding process. On success, it should return 0. @@ -1314,11 +1318,19 @@ If no error or position information is needed, you can pass *NULL*. *buffer* points to a buffer of *buflen* bytes, and *data* is the corresponding :func:`json_load_callback()` argument passed through. - On success, the function should return the number of bytes read; a - returned value of 0 indicates that no data was read and that the - end of file has been reached. On error, the function should return + On success, the function should write at most *buflen* bytes to + *buffer*, and return the number of bytes written; a returned value + of 0 indicates that no data was produced and that the end of file + has been reached. On error, the function should return ``(size_t)-1`` to abort the decoding process. + In UTF-8, some code points are encoded as multi-byte sequences. The + callback function doesn't need to worry about this, as Jansson + handles it at a higher level. For example, you can safely read a + fixed number of bytes from a network connection without having to + care about code unit sequences broken apart by the chunk + boundaries. + .. versionadded:: 2.4 .. function:: json_t *json_load_callback(json_load_callback_t callback, void *data, size_t flags, json_error_t *error) |