On Wednesday 02 August 2006 11:58, david reid wrote: > First pass at importing some apreq functions. Not yet had time to wirte > a test! Rushing out of door, so just a quick mail :-)
fairy nuff1 > +/** > + * @fn apr_size_t apr_uri_query_string_encode(char *dest, const char *src, > + const apr_size_t slen) > + * @brief URI encode a query string > + * @param dest the destination pointer > + * @param src the source pointer > + * @param slen length of source string > + */ > +APU_DECLARE(apr_size_t) apr_uri_query_string_encode(char *, const char *, > + const apr_size_t); No pool, so this presumably requires the destination to be already allocated. Which in turn presumes a size computation (documentation?), and the user has done most of the work! Or am I missing something? Also returning a size is not really APR-style, and is conspicuously at variance with ... > + * @fn apr_status_t apr_uri_query_string_decode(char *d, apr_size_t *dlen, > + const char *s, > + apr_size_t slen) > + * @brief URI Decode a query string > + * @param d destination pointer > + * @param dlen length of created string > + * @param s source pointer > + * @param slen length of source string > + */ > +APU_DECLARE(apr_status_t) apr_uri_query_string_decode(char *, apr_size_t > *, + const char *, + > apr_size_t); + Again no pool, though at least there's an easy size estimate this time:-) No comments on the implementation - that requires more time&effort, and since it's existing code it's presumably tried&tested. -- Nick Kew
