usr.bin/perl/sv.c

116  *   lib/utf8.t lib/Unicode/Collate/t/index.t
3382 cases, or if the input is already flagged as being in utf8.
3407      * compiled and individual nodes will remain non-utf8 even if the
3433             /* utf8 conversion not needed because all are invariants.  Mark
3441          * the string should be converted to utf8.  Everything from 's' to
3656         /* it is actually just a matter of turning the utf8 flag on, but
3657          * we want to make sure everything inside is valid utf8 first.
5584            bytes *and* utf8, which would indicate a bug elsewhere. */
7682 /* Create and update the UTF8 magic offset cache, with the proffered utf8/
7687    The proffered utf8/byte length pairing isn't used if the cache already has
7707                            const STRLEN utf8, const STRLEN blen)
7732            a pointer.  Note that we no longer cache utf8 offsets on refer-
7737         assert_uft8_cache_coherent("utf8_mg_pos_cache_update", utf8, realutf8,
7747         cache[0] = utf8;
7754             cache[0] = utf8;
7757             cache[2] = utf8;
7783             cache[0] = utf8;
7801                     cache[2] = utf8;
7805                     cache[0] = utf8;
7814                 cache[2] = utf8;
8611            be treated as utf-8 if the handle is utf8.
9715                Andreas would like keys he put in as utf8 to come back as utf8
12109 They assume that C<pat> has the same utf8-ness as C<sv>.  It's the caller's
12135     bool has_utf8 = DO_UTF8(sv);    /* has the result utf8? */
12136     const bool pat_utf8 = has_utf8; /* the pattern is in utf8? */
12242         bool is_utf8     = FALSE;     /* is this item utf8?   */
12262                 /* upgrade and copy the bytes of fmtstart..q-1 to utf8 on
12709              * %d%lu%4p  (UTF8f) A utf8 string. Consumes 3 args:
12710              *                       (cBOOL(utf8), len, string_buf).
12849                 /* if utf8 length is larger than 0x7ffff..., then it might
13315              * Note also things like ps_AF.utf8 which has
13330                     /* floating-point formats only get utf8 if the radix point
13331                      * is utf8. All other characters in the string are < 128
13332                      * and so can be safely appended to both a non-utf8 and utf8
13334                      * Note that this will convert the output to utf8 even if
13573                 /* XXX if sv was originally non-utf8 with a char in the
16546             /* clear pos and any utf8 cache */