utf8.h

   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, size_t 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
  36 static inline char *reencode_string_len(const char *a UNUSED, size_t b UNUSED,
  37                                         const char *c UNUSED,
  38                                         const char *d UNUSED, size_t *e)
  39 { if (e) *e = 0; return NULL; }
  40 #endif
  41
  42 static inline char *reencode_string(const char *in,
  43                                     const char *out_encoding,
  44                                     const char *in_encoding)
  45 {
  46         return reencode_string_len(in, strlen(in),
  47                                    out_encoding, in_encoding,
  48                                    NULL);
  49 }
  50
  51 int mbs_chrlen(const char **text, size_t *remainder_p, const char *encoding);
  52
  53 /*
  54  * Returns true if the path would match ".git" after HFS case-folding.
  55  * The path should be NUL-terminated, but we will match variants of both ".git\0"
  56  * and ".git/..." (but _not_ ".../.git"). This makes it suitable for both fsck
  57  * and verify_path().
  58  *
  59  * Likewise, the is_hfs_dotgitfoo() variants look for ".gitfoo".
  60  */
  61 int is_hfs_dotgit(const char *path);
  62 int is_hfs_dotgitmodules(const char *path);
  63 int is_hfs_dotgitignore(const char *path);
  64 int is_hfs_dotgitattributes(const char *path);
  65 int is_hfs_dotmailmap(const char *path);
  66
  67 typedef enum {
  68         ALIGN_LEFT,
  69         ALIGN_MIDDLE,
  70         ALIGN_RIGHT
  71 } align_type;
  72
  73 /*
  74  * Align the string given and store it into a strbuf as per the
  75  * 'position' and 'width'. If the given string length is larger than
  76  * 'width' than then the input string is not truncated and no
  77  * alignment is done.
  78  */
  79 void strbuf_utf8_align(struct strbuf *buf, align_type position, unsigned int width,
  80                        const char *s);
  81
  82 /*
  83  * If a data stream is declared as UTF-16BE or UTF-16LE, then a UTF-16
  84  * BOM must not be used [1]. The same applies for the UTF-32 equivalents.
  85  * The function returns true if this rule is violated.
  86  *
  87  * [1] https://unicode.org/faq/utf_bom.html#bom10
  88  */
  89 int has_prohibited_utf_bom(const char *enc, const char *data, size_t len);
  90
  91 /*
  92  * If the endianness is not defined in the encoding name, then we
  93  * require a BOM. The function returns true if a required BOM is missing.
  94  *
  95  * The Unicode standard instructs to assume big-endian if there in no
  96  * BOM for UTF-16/32 [1][2]. However, the W3C/WHATWG encoding standard
  97  * used in HTML5 recommends to assume little-endian to "deal with
  98  * deployed content" [3].
  99  *
 100  * Therefore, strictly requiring a BOM seems to be the safest option for
 101  * content in Git.
 102  *
 103  * [1] https://unicode.org/faq/utf_bom.html#gen6
 104  * [2] https://www.unicode.org/versions/Unicode10.0.0/ch03.pdf
 105  *     Section 3.10, D98, page 132
 106  * [3] https://encoding.spec.whatwg.org/#utf-16le
 107  */
 108 int is_missing_required_utf_bom(const char *enc, const char *data, size_t len);
 109
 110 #endif