1 #ifndef GIT_UTF8_H
2 #define GIT_UTF8_H
3
4 struct strbuf;
5
6 typedef unsigned int ucs_char_t; /* assuming 32bit int */
7
8 size_t display_mode_esc_sequence_len(const char *s);
9 int utf8_width(const char **start, size_t *remainder_p);
10 int utf8_strnwidth(const char *string, int len, int skip_ansi);
11 int utf8_strwidth(const char *string);
12 int is_utf8(const char *text);
13 int is_encoding_utf8(const char *name);
14 int same_encoding(const char *, const char *);
15 __attribute__((format (printf, 2, 3)))
16 int utf8_fprintf(FILE *, const char *, ...);
17
18 extern const char utf8_bom[];
19 int skip_utf8_bom(char **, size_t);
20
21 void strbuf_add_wrapped_text(struct strbuf *buf,
22 const char *text, int indent, int indent2, int width);
23 void strbuf_add_wrapped_bytes(struct strbuf *buf, const char *data, int len,
24 int indent, int indent2, int width);
25 void strbuf_utf8_replace(struct strbuf *sb, int pos, int width,
26 const char *subst);
27
28 #ifndef NO_ICONV
29 char *reencode_string_iconv(const char *in, size_t insz,
30 iconv_t conv, size_t bom_len, size_t *outsz);
31 char *reencode_string_len(const char *in, size_t insz,
32 const char *out_encoding,
33 const char *in_encoding,
34 size_t *outsz);
35 #else
reencode_string_len(const char * a,size_t b,const char * c,const char * d,size_t * e)36 static inline char *reencode_string_len(const char *a, size_t b,
37 const char *c, const char *d, size_t *e)
38 { if (e) *e = 0; return NULL; }
39 #endif
40
reencode_string(const char * in,const char * out_encoding,const char * in_encoding)41 static inline char *reencode_string(const char *in,
42 const char *out_encoding,
43 const char *in_encoding)
44 {
45 return reencode_string_len(in, strlen(in),
46 out_encoding, in_encoding,
47 NULL);
48 }
49
50 int mbs_chrlen(const char **text, size_t *remainder_p, const char *encoding);
51
52 /*
53 * Returns true if the path would match ".git" after HFS case-folding.
54 * The path should be NUL-terminated, but we will match variants of both ".git\0"
55 * and ".git/..." (but _not_ ".../.git"). This makes it suitable for both fsck
56 * and verify_path().
57 *
58 * Likewise, the is_hfs_dotgitfoo() variants look for ".gitfoo".
59 */
60 int is_hfs_dotgit(const char *path);
61 int is_hfs_dotgitmodules(const char *path);
62 int is_hfs_dotgitignore(const char *path);
63 int is_hfs_dotgitattributes(const char *path);
64 int is_hfs_dotmailmap(const char *path);
65
66 typedef enum {
67 ALIGN_LEFT,
68 ALIGN_MIDDLE,
69 ALIGN_RIGHT
70 } align_type;
71
72 /*
73 * Align the string given and store it into a strbuf as per the
74 * 'position' and 'width'. If the given string length is larger than
75 * 'width' than then the input string is not truncated and no
76 * alignment is done.
77 */
78 void strbuf_utf8_align(struct strbuf *buf, align_type position, unsigned int width,
79 const char *s);
80
81 /*
82 * If a data stream is declared as UTF-16BE or UTF-16LE, then a UTF-16
83 * BOM must not be used [1]. The same applies for the UTF-32 equivalents.
84 * The function returns true if this rule is violated.
85 *
86 * [1] http://unicode.org/faq/utf_bom.html#bom10
87 */
88 int has_prohibited_utf_bom(const char *enc, const char *data, size_t len);
89
90 /*
91 * If the endianness is not defined in the encoding name, then we
92 * require a BOM. The function returns true if a required BOM is missing.
93 *
94 * The Unicode standard instructs to assume big-endian if there in no
95 * BOM for UTF-16/32 [1][2]. However, the W3C/WHATWG encoding standard
96 * used in HTML5 recommends to assume little-endian to "deal with
97 * deployed content" [3].
98 *
99 * Therefore, strictly requiring a BOM seems to be the safest option for
100 * content in Git.
101 *
102 * [1] http://unicode.org/faq/utf_bom.html#gen6
103 * [2] http://www.unicode.org/versions/Unicode10.0.0/ch03.pdf
104 * Section 3.10, D98, page 132
105 * [3] https://encoding.spec.whatwg.org/#utf-16le
106 */
107 int is_missing_required_utf_bom(const char *enc, const char *data, size_t len);
108
109 #endif
110