| 1 | /* classes: h_files */ |
| 2 | |
| 3 | #ifndef SCM_STRINGS_H |
| 4 | #define SCM_STRINGS_H |
| 5 | |
| 6 | /* Copyright (C) 1995,1996,1997,1998,2000,2001, 2004 Free Software Foundation, Inc. |
| 7 | * |
| 8 | * This library is free software; you can redistribute it and/or |
| 9 | * modify it under the terms of the GNU Lesser General Public |
| 10 | * License as published by the Free Software Foundation; either |
| 11 | * version 2.1 of the License, or (at your option) any later version. |
| 12 | * |
| 13 | * This library is distributed in the hope that it will be useful, |
| 14 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 15 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 16 | * Lesser General Public License for more details. |
| 17 | * |
| 18 | * You should have received a copy of the GNU Lesser General Public |
| 19 | * License along with this library; if not, write to the Free Software |
| 20 | * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA |
| 21 | */ |
| 22 | |
| 23 | \f |
| 24 | |
| 25 | #include "libguile/__scm.h" |
| 26 | |
| 27 | \f |
| 28 | |
| 29 | /* String representation. |
| 30 | |
| 31 | A string is a piece of a stringbuf. A stringbuf can be used by |
| 32 | more than one string. When a string is written to and the |
| 33 | stringbuf of that string is used by more than one string, a new |
| 34 | stringbuf is created. That is, strings are copy-on-write. This |
| 35 | behavior can be used to make the substring operation quite |
| 36 | efficient. |
| 37 | |
| 38 | The implementation is tuned so that mutating a string is costly, |
| 39 | but just reading it is cheap and lock-free. |
| 40 | |
| 41 | There are also mutation-sharing strings. They refer to a part of |
| 42 | an ordinary string. Writing to a mutation-sharing string just |
| 43 | writes to the ordinary string. |
| 44 | |
| 45 | |
| 46 | Internal, low level interface to the character arrays |
| 47 | |
| 48 | - Use scm_i_string_chars to get a pointer to the byte array of a |
| 49 | string for reading. Use scm_i_string_length to get the number of |
| 50 | bytes in that array. The array is not null-terminated. |
| 51 | |
| 52 | - The array is valid as long as the corresponding SCM object is |
| 53 | protected but only until the next SCM_TICK. During such a 'safe |
| 54 | point', strings might change their representation. |
| 55 | |
| 56 | - Use scm_i_string_writable_chars to get the same pointer as with |
| 57 | scm_i_string_chars, but for reading and writing. This is a |
| 58 | potentially costly operation since it implements the |
| 59 | copy-on-write behavior. When done with the writing, call |
| 60 | scm_i_string_stop_writing. You must do this before the next |
| 61 | SCM_TICK. (This means, before calling almost any other scm_ |
| 62 | function and you can't allow throws, of course.) |
| 63 | |
| 64 | - New strings can be created with scm_i_make_string. This gives |
| 65 | access to a writable pointer that remains valid as long as nobody |
| 66 | else makes a copy-on-write substring of the string. Do not call |
| 67 | scm_i_string_stop_writing for this pointer. |
| 68 | |
| 69 | Legacy interface |
| 70 | |
| 71 | - SCM_STRINGP is just scm_is_string. |
| 72 | |
| 73 | - SCM_STRING_CHARS uses scm_i_string_writable_chars and immediately |
| 74 | calls scm_i_stop_writing, hoping for the best. SCM_STRING_LENGTH |
| 75 | is the same as scm_i_string_length. SCM_STRING_CHARS will throw |
| 76 | an error for for strings that are not null-terminated. |
| 77 | */ |
| 78 | |
| 79 | SCM_API SCM scm_string_p (SCM x); |
| 80 | SCM_API SCM scm_string (SCM chrs); |
| 81 | SCM_API SCM scm_make_string (SCM k, SCM chr); |
| 82 | SCM_API SCM scm_string_length (SCM str); |
| 83 | SCM_API SCM scm_string_ref (SCM str, SCM k); |
| 84 | SCM_API SCM scm_string_set_x (SCM str, SCM k, SCM chr); |
| 85 | SCM_API SCM scm_substring (SCM str, SCM start, SCM end); |
| 86 | SCM_API SCM scm_substring_read_only (SCM str, SCM start, SCM end); |
| 87 | SCM_API SCM scm_substring_shared (SCM str, SCM start, SCM end); |
| 88 | SCM_API SCM scm_substring_copy (SCM str, SCM start, SCM end); |
| 89 | SCM_API SCM scm_string_append (SCM args); |
| 90 | |
| 91 | SCM_API SCM scm_c_make_string (size_t len, SCM chr); |
| 92 | SCM_API size_t scm_c_string_length (SCM str); |
| 93 | SCM_API SCM scm_c_string_ref (SCM str, size_t pos); |
| 94 | SCM_API void scm_c_string_set_x (SCM str, size_t pos, SCM chr); |
| 95 | SCM_API SCM scm_c_substring (SCM str, size_t start, size_t end); |
| 96 | SCM_API SCM scm_c_substring_read_only (SCM str, size_t start, size_t end); |
| 97 | SCM_API SCM scm_c_substring_shared (SCM str, size_t start, size_t end); |
| 98 | SCM_API SCM scm_c_substring_copy (SCM str, size_t start, size_t end); |
| 99 | |
| 100 | SCM_API int scm_is_string (SCM x); |
| 101 | SCM_API SCM scm_from_locale_string (const char *str); |
| 102 | SCM_API SCM scm_from_locale_stringn (const char *str, size_t len); |
| 103 | SCM_API SCM scm_take_locale_string (char *str); |
| 104 | SCM_API SCM scm_take_locale_stringn (char *str, size_t len); |
| 105 | SCM_API char *scm_to_locale_string (SCM str); |
| 106 | SCM_API char *scm_to_locale_stringn (SCM str, size_t *lenp); |
| 107 | SCM_API size_t scm_to_locale_stringbuf (SCM str, char *buf, size_t max_len); |
| 108 | |
| 109 | SCM_API SCM scm_makfromstrs (int argc, char **argv); |
| 110 | |
| 111 | /* internal accessor functions. Arguments must be valid. */ |
| 112 | |
| 113 | SCM_API SCM scm_i_make_string (size_t len, char **datap); |
| 114 | SCM_API SCM scm_i_substring (SCM str, size_t start, size_t end); |
| 115 | SCM_API SCM scm_i_substring_read_only (SCM str, size_t start, size_t end); |
| 116 | SCM_API SCM scm_i_substring_shared (SCM str, size_t start, size_t end); |
| 117 | SCM_API SCM scm_i_substring_copy (SCM str, size_t start, size_t end); |
| 118 | SCM_API size_t scm_i_string_length (SCM str); |
| 119 | SCM_API const char *scm_i_string_chars (SCM str); |
| 120 | SCM_API char *scm_i_string_writable_chars (SCM str); |
| 121 | SCM_API void scm_i_string_stop_writing (void); |
| 122 | |
| 123 | /* internal functions related to symbols. */ |
| 124 | |
| 125 | SCM_API SCM scm_i_make_symbol (SCM name, scm_t_bits flags, |
| 126 | unsigned long hash, SCM props); |
| 127 | SCM_API const char *scm_i_symbol_chars (SCM sym); |
| 128 | SCM_API size_t scm_i_symbol_length (SCM sym); |
| 129 | SCM_API SCM scm_i_symbol_substring (SCM sym, size_t start, size_t end); |
| 130 | |
| 131 | /* internal GC functions. */ |
| 132 | |
| 133 | SCM_API SCM scm_i_string_mark (SCM str); |
| 134 | SCM_API SCM scm_i_stringbuf_mark (SCM buf); |
| 135 | SCM_API SCM scm_i_symbol_mark (SCM buf); |
| 136 | SCM_API void scm_i_string_free (SCM str); |
| 137 | SCM_API void scm_i_stringbuf_free (SCM buf); |
| 138 | SCM_API void scm_i_symbol_free (SCM sym); |
| 139 | |
| 140 | /* internal utility functions. */ |
| 141 | |
| 142 | SCM_API char **scm_i_allocate_string_pointers (SCM list); |
| 143 | SCM_API void scm_i_free_string_pointers (char **pointers); |
| 144 | SCM_API void scm_i_get_substring_spec (size_t len, |
| 145 | SCM start, size_t *cstart, |
| 146 | SCM end, size_t *cend); |
| 147 | |
| 148 | /* deprecated stuff */ |
| 149 | |
| 150 | #if SCM_ENABLE_DEPRECATED |
| 151 | |
| 152 | SCM_API int scm_i_deprecated_stringp (SCM obj); |
| 153 | SCM_API char *scm_i_deprecated_string_chars (SCM str); |
| 154 | SCM_API size_t scm_i_deprecated_string_length (SCM str); |
| 155 | |
| 156 | #define SCM_STRINGP(x) scm_i_deprecated_stringp(x) |
| 157 | #define SCM_STRING_CHARS(x) scm_i_deprecated_string_chars(x) |
| 158 | #define SCM_STRING_LENGTH(x) scm_i_deprecated_string_length(x) |
| 159 | #define SCM_STRING_UCHARS ((unsigned char *)SCM_STRING_CHARS (str)) |
| 160 | |
| 161 | #endif |
| 162 | |
| 163 | SCM_API void scm_init_strings (void); |
| 164 | |
| 165 | #endif /* SCM_STRINGS_H */ |
| 166 | |
| 167 | /* |
| 168 | Local Variables: |
| 169 | c-file-style: "gnu" |
| 170 | End: |
| 171 | */ |