git-p4: improve import performance by avoiding reallocations.
[git/git-bigfiles.git] / symlinks.c
blobf262b7c44b387f4c60901124c30585b3c059fa73
1 #include "cache.h"
3 static struct cache_def {
4 char path[PATH_MAX + 1];
5 int len;
6 int flags;
7 int track_flags;
8 int prefix_len_stat_func;
9 } cache;
12 * Returns the length (on a path component basis) of the longest
13 * common prefix match of 'name' and the cached path string.
15 static inline int longest_match_lstat_cache(int len, const char *name,
16 int *previous_slash)
18 int max_len, match_len = 0, match_len_prev = 0, i = 0;
20 max_len = len < cache.len ? len : cache.len;
21 while (i < max_len && name[i] == cache.path[i]) {
22 if (name[i] == '/') {
23 match_len_prev = match_len;
24 match_len = i;
26 i++;
28 /* Is the cached path string a substring of 'name'? */
29 if (i == cache.len && cache.len < len && name[cache.len] == '/') {
30 match_len_prev = match_len;
31 match_len = cache.len;
32 /* Is 'name' a substring of the cached path string? */
33 } else if ((i == len && len < cache.len && cache.path[len] == '/') ||
34 (i == len && len == cache.len)) {
35 match_len_prev = match_len;
36 match_len = len;
38 *previous_slash = match_len_prev;
39 return match_len;
42 static inline void reset_lstat_cache(int track_flags, int prefix_len_stat_func)
44 cache.path[0] = '\0';
45 cache.len = 0;
46 cache.flags = 0;
47 cache.track_flags = track_flags;
48 cache.prefix_len_stat_func = prefix_len_stat_func;
51 #define FL_DIR (1 << 0)
52 #define FL_NOENT (1 << 1)
53 #define FL_SYMLINK (1 << 2)
54 #define FL_LSTATERR (1 << 3)
55 #define FL_ERR (1 << 4)
56 #define FL_FULLPATH (1 << 5)
59 * Check if name 'name' of length 'len' has a symlink leading
60 * component, or if the directory exists and is real, or not.
62 * To speed up the check, some information is allowed to be cached.
63 * This can be indicated by the 'track_flags' argument, which also can
64 * be used to indicate that we should check the full path.
66 * The 'prefix_len_stat_func' parameter can be used to set the length
67 * of the prefix, where the cache should use the stat() function
68 * instead of the lstat() function to test each path component.
70 static int lstat_cache(int len, const char *name,
71 int track_flags, int prefix_len_stat_func)
73 int match_len, last_slash, last_slash_dir, previous_slash;
74 int match_flags, ret_flags, save_flags, max_len, ret;
75 struct stat st;
77 if (cache.track_flags != track_flags ||
78 cache.prefix_len_stat_func != prefix_len_stat_func) {
80 * As a safeguard we clear the cache if the values of
81 * track_flags and/or prefix_len_stat_func does not
82 * match with the last supplied values.
84 reset_lstat_cache(track_flags, prefix_len_stat_func);
85 match_len = last_slash = 0;
86 } else {
88 * Check to see if we have a match from the cache for
89 * the 2 "excluding" path types.
91 match_len = last_slash =
92 longest_match_lstat_cache(len, name, &previous_slash);
93 match_flags = cache.flags & track_flags & (FL_NOENT|FL_SYMLINK);
94 if (match_flags && match_len == cache.len)
95 return match_flags;
97 * If we now have match_len > 0, we would know that
98 * the matched part will always be a directory.
100 * Also, if we are tracking directories and 'name' is
101 * a substring of the cache on a path component basis,
102 * we can return immediately.
104 match_flags = track_flags & FL_DIR;
105 if (match_flags && len == match_len)
106 return match_flags;
110 * Okay, no match from the cache so far, so now we have to
111 * check the rest of the path components.
113 ret_flags = FL_DIR;
114 last_slash_dir = last_slash;
115 max_len = len < PATH_MAX ? len : PATH_MAX;
116 while (match_len < max_len) {
117 do {
118 cache.path[match_len] = name[match_len];
119 match_len++;
120 } while (match_len < max_len && name[match_len] != '/');
121 if (match_len >= max_len && !(track_flags & FL_FULLPATH))
122 break;
123 last_slash = match_len;
124 cache.path[last_slash] = '\0';
126 if (last_slash <= prefix_len_stat_func)
127 ret = stat(cache.path, &st);
128 else
129 ret = lstat(cache.path, &st);
131 if (ret) {
132 ret_flags = FL_LSTATERR;
133 if (errno == ENOENT)
134 ret_flags |= FL_NOENT;
135 } else if (S_ISDIR(st.st_mode)) {
136 last_slash_dir = last_slash;
137 continue;
138 } else if (S_ISLNK(st.st_mode)) {
139 ret_flags = FL_SYMLINK;
140 } else {
141 ret_flags = FL_ERR;
143 break;
147 * At the end update the cache. Note that max 3 different
148 * path types, FL_NOENT, FL_SYMLINK and FL_DIR, can be cached
149 * for the moment!
151 save_flags = ret_flags & track_flags & (FL_NOENT|FL_SYMLINK);
152 if (save_flags && last_slash > 0 && last_slash <= PATH_MAX) {
153 cache.path[last_slash] = '\0';
154 cache.len = last_slash;
155 cache.flags = save_flags;
156 } else if (track_flags & FL_DIR &&
157 last_slash_dir > 0 && last_slash_dir <= PATH_MAX) {
159 * We have a separate test for the directory case,
160 * since it could be that we have found a symlink or a
161 * non-existing directory and the track_flags says
162 * that we cannot cache this fact, so the cache would
163 * then have been left empty in this case.
165 * But if we are allowed to track real directories, we
166 * can still cache the path components before the last
167 * one (the found symlink or non-existing component).
169 cache.path[last_slash_dir] = '\0';
170 cache.len = last_slash_dir;
171 cache.flags = FL_DIR;
172 } else {
173 reset_lstat_cache(track_flags, prefix_len_stat_func);
175 return ret_flags;
179 * Invalidate the given 'name' from the cache, if 'name' matches
180 * completely with the cache.
182 void invalidate_lstat_cache(int len, const char *name)
184 int match_len, previous_slash;
186 match_len = longest_match_lstat_cache(len, name, &previous_slash);
187 if (len == match_len) {
188 if ((cache.track_flags & FL_DIR) && previous_slash > 0) {
189 cache.path[previous_slash] = '\0';
190 cache.len = previous_slash;
191 cache.flags = FL_DIR;
192 } else
193 reset_lstat_cache(cache.track_flags,
194 cache.prefix_len_stat_func);
199 * Completely clear the contents of the cache
201 void clear_lstat_cache(void)
203 reset_lstat_cache(0, 0);
206 #define USE_ONLY_LSTAT 0
209 * Return non-zero if path 'name' has a leading symlink component
211 int has_symlink_leading_path(int len, const char *name)
213 return lstat_cache(len, name,
214 FL_SYMLINK|FL_DIR, USE_ONLY_LSTAT) &
215 FL_SYMLINK;
219 * Return non-zero if path 'name' has a leading symlink component or
220 * if some leading path component does not exists.
222 int has_symlink_or_noent_leading_path(int len, const char *name)
224 return lstat_cache(len, name,
225 FL_SYMLINK|FL_NOENT|FL_DIR, USE_ONLY_LSTAT) &
226 (FL_SYMLINK|FL_NOENT);
230 * Return non-zero if all path components of 'name' exists as a
231 * directory. If prefix_len > 0, we will test with the stat()
232 * function instead of the lstat() function for a prefix length of
233 * 'prefix_len', thus we then allow for symlinks in the prefix part as
234 * long as those points to real existing directories.
236 int has_dirs_only_path(int len, const char *name, int prefix_len)
238 return lstat_cache(len, name,
239 FL_DIR|FL_FULLPATH, prefix_len) &
240 FL_DIR;