diff options
author | Jan Christoph Ebersbach <jceb@e-jc.de> | 2015-05-22 16:06:57 +0200 |
---|---|---|
committer | Roberto E. Vargas Caballero <k0ga@shike2.com> | 2015-05-25 08:35:32 +0200 |
commit | caa97cc781ccf29f28c3d9e6683a66eb3f70e2bd (patch) | |
tree | 98a9dcd8e25fee71514f2a786fdaa785a1aea4bd | |
parent | c03548750b2527a6ddb5edfd945c5799066a6224 (diff) | |
download | st-patched-caa97cc781ccf29f28c3d9e6683a66eb3f70e2bd.tar.bz2 st-patched-caa97cc781ccf29f28c3d9e6683a66eb3f70e2bd.tar.xz st-patched-caa97cc781ccf29f28c3d9e6683a66eb3f70e2bd.zip |
Support UTF-8 characters as word delimiters
For a higher usefulness of the utf8strchr function, the index of the
UTF-8 character could be returned in addition with a Rune instead of a
char*. Since utf8strchr is currently only used by ISDELIM I didn't
bother to increase the complexity.
-rw-r--r-- | st.c | 18 |
1 files changed, 17 insertions, 1 deletions
@@ -71,7 +71,7 @@ char *argv0; | |||
71 | #define ISCONTROLC0(c) (BETWEEN(c, 0, 0x1f) || (c) == '\177') | 71 | #define ISCONTROLC0(c) (BETWEEN(c, 0, 0x1f) || (c) == '\177') |
72 | #define ISCONTROLC1(c) (BETWEEN(c, 0x80, 0x9f)) | 72 | #define ISCONTROLC1(c) (BETWEEN(c, 0x80, 0x9f)) |
73 | #define ISCONTROL(c) (ISCONTROLC0(c) || ISCONTROLC1(c)) | 73 | #define ISCONTROL(c) (ISCONTROLC0(c) || ISCONTROLC1(c)) |
74 | #define ISDELIM(u) (BETWEEN(u, 0, 127) && strchr(worddelimiters, u) != NULL) | 74 | #define ISDELIM(u) (utf8strchr(worddelimiters, u) != NULL) |
75 | #define LIMIT(x, a, b) (x) = (x) < (a) ? (a) : (x) > (b) ? (b) : (x) | 75 | #define LIMIT(x, a, b) (x) = (x) < (a) ? (a) : (x) > (b) ? (b) : (x) |
76 | #define ATTRCMP(a, b) ((a).mode != (b).mode || (a).fg != (b).fg || (a).bg != (b).bg) | 76 | #define ATTRCMP(a, b) ((a).mode != (b).mode || (a).fg != (b).fg || (a).bg != (b).bg) |
77 | #define IS_SET(flag) ((term.mode & (flag)) != 0) | 77 | #define IS_SET(flag) ((term.mode & (flag)) != 0) |
@@ -473,6 +473,7 @@ static size_t utf8decode(char *, Rune *, size_t); | |||
473 | static Rune utf8decodebyte(char, size_t *); | 473 | static Rune utf8decodebyte(char, size_t *); |
474 | static size_t utf8encode(Rune, char *); | 474 | static size_t utf8encode(Rune, char *); |
475 | static char utf8encodebyte(Rune, size_t); | 475 | static char utf8encodebyte(Rune, size_t); |
476 | static char *utf8strchr(char *s, Rune u); | ||
476 | static size_t utf8validate(Rune *, size_t); | 477 | static size_t utf8validate(Rune *, size_t); |
477 | 478 | ||
478 | static ssize_t xwrite(int, const char *, size_t); | 479 | static ssize_t xwrite(int, const char *, size_t); |
@@ -640,6 +641,21 @@ utf8encodebyte(Rune u, size_t i) { | |||
640 | return utfbyte[i] | (u & ~utfmask[i]); | 641 | return utfbyte[i] | (u & ~utfmask[i]); |
641 | } | 642 | } |
642 | 643 | ||
644 | char * | ||
645 | utf8strchr(char *s, Rune u) { | ||
646 | Rune r; | ||
647 | size_t i, j, len; | ||
648 | |||
649 | len = strlen(s); | ||
650 | for(i = 0, j = 0; i < len; i += j) { | ||
651 | if(!(j = utf8decode(&s[i], &r, len - i))) | ||
652 | break; | ||
653 | if(r == u) | ||
654 | return &(s[i]); | ||
655 | } | ||
656 | return NULL; | ||
657 | } | ||
658 | |||
643 | size_t | 659 | size_t |
644 | utf8validate(Rune *u, size_t i) { | 660 | utf8validate(Rune *u, size_t i) { |
645 | if(!BETWEEN(*u, utfmin[i], utfmax[i]) || BETWEEN(*u, 0xD800, 0xDFFF)) | 661 | if(!BETWEEN(*u, utfmin[i], utfmax[i]) || BETWEEN(*u, 0xD800, 0xDFFF)) |