buf->buf[newlen] = '/';
}
-static const char *common_list[] = {
- "/branches", "/hooks", "/info", "!/logs", "/lost-found",
- "/objects", "/refs", "/remotes", "/worktrees", "/rr-cache", "/svn",
- "config", "!gc.pid", "packed-refs", "shallow",
- NULL
+struct common_dir {
+ /* Not considered garbage for report_linked_checkout_garbage */
+ unsigned ignore_garbage:1;
+ unsigned is_dir:1;
+ /* Not common even though its parent is */
+ unsigned exclude:1;
+ const char *dirname;
};
-static void update_common_dir(struct strbuf *buf, int git_dir_len)
+static struct common_dir common_list[] = {
+ { 0, 1, 0, "branches" },
+ { 0, 1, 0, "hooks" },
+ { 0, 1, 0, "info" },
+ { 0, 0, 1, "info/sparse-checkout" },
+ { 1, 1, 0, "logs" },
+ { 1, 1, 1, "logs/HEAD" },
+ { 0, 1, 1, "logs/refs/bisect" },
+ { 0, 1, 0, "lost-found" },
+ { 0, 1, 0, "objects" },
+ { 0, 1, 0, "refs" },
+ { 0, 1, 1, "refs/bisect" },
+ { 0, 1, 0, "remotes" },
+ { 0, 1, 0, "worktrees" },
+ { 0, 1, 0, "rr-cache" },
+ { 0, 1, 0, "svn" },
+ { 0, 0, 0, "config" },
+ { 1, 0, 0, "gc.pid" },
+ { 0, 0, 0, "packed-refs" },
+ { 0, 0, 0, "shallow" },
+ { 0, 0, 0, NULL }
+};
+
+/*
+ * A compressed trie. A trie node consists of zero or more characters that
+ * are common to all elements with this prefix, optionally followed by some
+ * children. If value is not NULL, the trie node is a terminal node.
+ *
+ * For example, consider the following set of strings:
+ * abc
+ * def
+ * definite
+ * definition
+ *
+ * The trie would look look like:
+ * root: len = 0, children a and d non-NULL, value = NULL.
+ * a: len = 2, contents = bc, value = (data for "abc")
+ * d: len = 2, contents = ef, children i non-NULL, value = (data for "def")
+ * i: len = 3, contents = nit, children e and i non-NULL, value = NULL
+ * e: len = 0, children all NULL, value = (data for "definite")
+ * i: len = 2, contents = on, children all NULL,
+ * value = (data for "definition")
+ */
+struct trie {
+ struct trie *children[256];
+ int len;
+ char *contents;
+ void *value;
+};
+
+static struct trie *make_trie_node(const char *key, void *value)
{
- char *base = buf->buf + git_dir_len;
- const char **p;
-
- if (is_dir_file(base, "logs", "HEAD") ||
- is_dir_file(base, "info", "sparse-checkout"))
- return; /* keep this in $GIT_DIR */
- for (p = common_list; *p; p++) {
- const char *path = *p;
- int is_dir = 0;
- if (*path == '!')
- path++;
- if (*path == '/') {
- path++;
- is_dir = 1;
+ struct trie *new_node = xcalloc(1, sizeof(*new_node));
+ new_node->len = strlen(key);
+ if (new_node->len) {
+ new_node->contents = xmalloc(new_node->len);
+ memcpy(new_node->contents, key, new_node->len);
+ }
+ new_node->value = value;
+ return new_node;
+}
+
+/*
+ * Add a key/value pair to a trie. The key is assumed to be \0-terminated.
+ * If there was an existing value for this key, return it.
+ */
+static void *add_to_trie(struct trie *root, const char *key, void *value)
+{
+ struct trie *child;
+ void *old;
+ int i;
+
+ if (!*key) {
+ /* we have reached the end of the key */
+ old = root->value;
+ root->value = value;
+ return old;
+ }
+
+ for (i = 0; i < root->len; i++) {
+ if (root->contents[i] == key[i])
+ continue;
+
+ /*
+ * Split this node: child will contain this node's
+ * existing children.
+ */
+ child = malloc(sizeof(*child));
+ memcpy(child->children, root->children, sizeof(root->children));
+
+ child->len = root->len - i - 1;
+ if (child->len) {
+ child->contents = xstrndup(root->contents + i + 1,
+ child->len);
}
- if (is_dir && dir_prefix(base, path)) {
- replace_dir(buf, git_dir_len, get_git_common_dir());
- return;
+ child->value = root->value;
+ root->value = NULL;
+ root->len = i;
+
+ memset(root->children, 0, sizeof(root->children));
+ root->children[(unsigned char)root->contents[i]] = child;
+
+ /* This is the newly-added child. */
+ root->children[(unsigned char)key[i]] =
+ make_trie_node(key + i + 1, value);
+ return NULL;
+ }
+
+ /* We have matched the entire compressed section */
+ if (key[i]) {
+ child = root->children[(unsigned char)key[root->len]];
+ if (child) {
+ return add_to_trie(child, key + root->len + 1, value);
+ } else {
+ child = make_trie_node(key + root->len + 1, value);
+ root->children[(unsigned char)key[root->len]] = child;
+ return NULL;
}
- if (!is_dir && !strcmp(base, path)) {
- replace_dir(buf, git_dir_len, get_git_common_dir());
- return;
+ }
+
+ old = root->value;
+ root->value = value;
+ return old;
+}
+
+typedef int (*match_fn)(const char *unmatched, void *data, void *baton);
+
+/*
+ * Search a trie for some key. Find the longest /-or-\0-terminated
+ * prefix of the key for which the trie contains a value. Call fn
+ * with the unmatched portion of the key and the found value, and
+ * return its return value. If there is no such prefix, return -1.
+ *
+ * The key is partially normalized: consecutive slashes are skipped.
+ *
+ * For example, consider the trie containing only [refs,
+ * refs/worktree] (both with values).
+ *
+ * | key | unmatched | val from node | return value |
+ * |-----------------|------------|---------------|--------------|
+ * | a | not called | n/a | -1 |
+ * | refs | \0 | refs | as per fn |
+ * | refs/ | / | refs | as per fn |
+ * | refs/w | /w | refs | as per fn |
+ * | refs/worktree | \0 | refs/worktree | as per fn |
+ * | refs/worktree/ | / | refs/worktree | as per fn |
+ * | refs/worktree/a | /a | refs/worktree | as per fn |
+ * |-----------------|------------|---------------|--------------|
+ *
+ */
+static int trie_find(struct trie *root, const char *key, match_fn fn,
+ void *baton)
+{
+ int i;
+ int result;
+ struct trie *child;
+
+ if (!*key) {
+ /* we have reached the end of the key */
+ if (root->value && !root->len)
+ return fn(key, root->value, baton);
+ else
+ return -1;
+ }
+
+ for (i = 0; i < root->len; i++) {
+ /* Partial path normalization: skip consecutive slashes. */
+ if (key[i] == '/' && key[i+1] == '/') {
+ key++;
+ continue;
}
+ if (root->contents[i] != key[i])
+ return -1;
}
+
+ /* Matched the entire compressed section */
+ key += i;
+ if (!*key)
+ /* End of key */
+ return fn(key, root->value, baton);
+
+ /* Partial path normalization: skip consecutive slashes */
+ while (key[0] == '/' && key[1] == '/')
+ key++;
+
+ child = root->children[(unsigned char)*key];
+ if (child)
+ result = trie_find(child, key + 1, fn, baton);
+ else
+ result = -1;
+
+ if (result >= 0 || (*key != '/' && *key != 0))
+ return result;
+ if (root->value)
+ return fn(key, root->value, baton);
+ else
+ return -1;
+}
+
+static struct trie common_trie;
+static int common_trie_done_setup;
+
+static void init_common_trie(void)
+{
+ struct common_dir *p;
+
+ if (common_trie_done_setup)
+ return;
+
+ for (p = common_list; p->dirname; p++)
+ add_to_trie(&common_trie, p->dirname, p);
+
+ common_trie_done_setup = 1;
+}
+
+/*
+ * Helper function for update_common_dir: returns 1 if the dir
+ * prefix is common.
+ */
+static int check_common(const char *unmatched, void *value, void *baton)
+{
+ struct common_dir *dir = value;
+
+ if (!dir)
+ return 0;
+
+ if (dir->is_dir && (unmatched[0] == 0 || unmatched[0] == '/'))
+ return !dir->exclude;
+
+ if (!dir->is_dir && unmatched[0] == 0)
+ return !dir->exclude;
+
+ return 0;
+}
+
+static void update_common_dir(struct strbuf *buf, int git_dir_len)
+{
+ char *base = buf->buf + git_dir_len;
+ init_common_trie();
+ if (trie_find(&common_trie, base, check_common, NULL) > 0)
+ replace_dir(buf, git_dir_len, get_git_common_dir());
}
void report_linked_checkout_garbage(void)
{
struct strbuf sb = STRBUF_INIT;
- const char **p;
+ const struct common_dir *p;
int len;
if (!git_common_dir_env)
return;
strbuf_addf(&sb, "%s/", get_git_dir());
len = sb.len;
- for (p = common_list; *p; p++) {
- const char *path = *p;
- if (*path == '!')
+ for (p = common_list; p->dirname; p++) {
+ const char *path = p->dirname;
+ if (p->ignore_garbage)
continue;
strbuf_setlen(&sb, len);
strbuf_addstr(&sb, path);