rugged 0.19.0 → 0.28.0

data/vendor/libgit2/src/iterator.h

@@ -8,9 +8,11 @@
 #define INCLUDE_iterator_h__
 
 #include "common.h"
+
 #include "git2/index.h"
 #include "vector.h"
 #include "buffer.h"
+#include "ignore.h"
 
 typedef struct git_iterator git_iterator;
 
@@ -24,40 +26,85 @@ typedef enum {
 
 typedef enum {
 	/** ignore case for entry sort order */
-	GIT_ITERATOR_IGNORE_CASE = (1 << 0),
+	GIT_ITERATOR_IGNORE_CASE = (1u << 0),
 	/** force case sensitivity for entry sort order */
-	GIT_ITERATOR_DONT_IGNORE_CASE = (1 << 1),
+	GIT_ITERATOR_DONT_IGNORE_CASE = (1u << 1),
 	/** return tree items in addition to blob items */
-	GIT_ITERATOR_INCLUDE_TREES    = (1 << 2),
+	GIT_ITERATOR_INCLUDE_TREES    = (1u << 2),
 	/** don't flatten trees, requiring advance_into (implies INCLUDE_TREES) */
-	GIT_ITERATOR_DONT_AUTOEXPAND  = (1 << 3),
+	GIT_ITERATOR_DONT_AUTOEXPAND  = (1u << 3),
+	/** convert precomposed unicode to decomposed unicode */
+	GIT_ITERATOR_PRECOMPOSE_UNICODE = (1u << 4),
+	/** never convert precomposed unicode to decomposed unicode */
+	GIT_ITERATOR_DONT_PRECOMPOSE_UNICODE = (1u << 5),
+	/** include conflicts */
+	GIT_ITERATOR_INCLUDE_CONFLICTS = (1u << 6),
+	/** descend into symlinked directories */
+	GIT_ITERATOR_DESCEND_SYMLINKS = (1u << 7),
+	/** hash files in workdir or filesystem iterators */
+	GIT_ITERATOR_INCLUDE_HASH = (1u << 8),
 } git_iterator_flag_t;
 
+typedef enum {
+	GIT_ITERATOR_STATUS_NORMAL = 0,
+	GIT_ITERATOR_STATUS_IGNORED = 1,
+	GIT_ITERATOR_STATUS_EMPTY = 2,
+	GIT_ITERATOR_STATUS_FILTERED = 3
+} git_iterator_status_t;
+
+typedef struct {
+	const char *start;
+	const char *end;
+
+	/* paths to include in the iterator (literal).  if set, any paths not
+	 * listed here will be excluded from iteration.
+	 */
+	git_strarray pathlist;
+
+	/* flags, from above */
+	unsigned int flags;
+} git_iterator_options;
+
+#define GIT_ITERATOR_OPTIONS_INIT {0}
+
 typedef struct {
 	int (*current)(const git_index_entry **, git_iterator *);
 	int (*advance)(const git_index_entry **, git_iterator *);
 	int (*advance_into)(const git_index_entry **, git_iterator *);
-	int (*seek)(git_iterator *, const char *prefix);
-	int (*reset)(git_iterator *, const char *start, const char *end);
-	int (*at_end)(git_iterator *);
+	int (*advance_over)(
+		const git_index_entry **, git_iterator_status_t *, git_iterator *);
+	int (*reset)(git_iterator *);
 	void (*free)(git_iterator *);
 } git_iterator_callbacks;
 
 struct git_iterator {
 	git_iterator_type_t type;
 	git_iterator_callbacks *cb;
+
 	git_repository *repo;
+	git_index *index;
+
 	char *start;
+	size_t start_len;
+
 	char *end;
+	size_t end_len;
+
+	bool started;
+	bool ended;
+	git_vector pathlist;
+	size_t pathlist_walk_idx;
+	int (*strcomp)(const char *a, const char *b);
+	int (*strncomp)(const char *a, const char *b, size_t n);
 	int (*prefixcomp)(const char *str, const char *prefix);
+	int (*entry_srch)(const void *key, const void *array_member);
+	size_t stat_calls;
 	unsigned int flags;
 };
 
 extern int git_iterator_for_nothing(
 	git_iterator **out,
-	git_iterator_flag_t flags,
-	const char *start,
-	const char *end);
+	git_iterator_options *options);
 
 /* tree iterators will match the ignore_case value from the index of the
  * repository, unless you override with a non-zero flag value
@@ -65,27 +112,24 @@ extern int git_iterator_for_nothing(
 extern int git_iterator_for_tree(
 	git_iterator **out,
 	git_tree *tree,
-	git_iterator_flag_t flags,
-	const char *start,
-	const char *end);
+	git_iterator_options *options);
 
 /* index iterators will take the ignore_case value from the index; the
  * ignore_case flags are not used
  */
 extern int git_iterator_for_index(
 	git_iterator **out,
+	git_repository *repo,
 	git_index *index,
-	git_iterator_flag_t flags,
-	const char *start,
-	const char *end);
+	git_iterator_options *options);
 
 extern int git_iterator_for_workdir_ext(
 	git_iterator **out,
 	git_repository *repo,
 	const char *repo_workdir,
-	git_iterator_flag_t flags,
-	const char *start,
-	const char *end);
+	git_index *index,
+	git_tree *tree,
+	git_iterator_options *options);
 
 /* workdir iterators will match the ignore_case value from the index of the
  * repository, unless you override with a non-zero flag value
@@ -93,11 +137,11 @@ extern int git_iterator_for_workdir_ext(
 GIT_INLINE(int) git_iterator_for_workdir(
 	git_iterator **out,
 	git_repository *repo,
-	git_iterator_flag_t flags,
-	const char *start,
-	const char *end)
+	git_index *index,
+	git_tree *tree,
+	git_iterator_options *options)
 {
-	return git_iterator_for_workdir_ext(out, repo, NULL, flags, start, end);
+	return git_iterator_for_workdir_ext(out, repo, NULL, index, tree, options);
 }
 
 /* for filesystem iterators, you have to explicitly pass in the ignore_case
@@ -106,9 +150,7 @@ GIT_INLINE(int) git_iterator_for_workdir(
 extern int git_iterator_for_filesystem(
 	git_iterator **out,
 	const char *root,
-	git_iterator_flag_t flags,
-	const char *start,
-	const char *end);
+	git_iterator_options *options);
 
 extern void git_iterator_free(git_iterator *iter);
 
@@ -163,54 +205,38 @@ GIT_INLINE(int) git_iterator_advance_into(
 	return iter->cb->advance_into(entry, iter);
 }
 
-/**
- * Advance into a tree or skip over it if it is empty.
+/* Advance over a directory and check if it contains no files or just
+ * ignored files.
  *
- * Because `git_iterator_advance_into` may return GIT_ENOTFOUND if the
- * directory is empty (only with filesystem and working directory
- * iterators) and a common response is to just call `git_iterator_advance`
- * when that happens, this bundles the two into a single simple call.
+ * In a tree or the index, all directories will contain files, but in the
+ * working directory it is possible to have an empty directory tree or a
+ * tree that only contains ignored files.  Many Git operations treat these
+ * cases specially.  This advances over a directory (presumably an
+ * untracked directory) but checks during the scan if there are any files
+ * and any non-ignored files.
  */
-GIT_INLINE(int) git_iterator_advance_into_or_over(
-	const git_index_entry **entry, git_iterator *iter)
-{
-	int error = iter->cb->advance_into(entry, iter);
-	if (error == GIT_ENOTFOUND) {
-		giterr_clear();
-		error = iter->cb->advance(entry, iter);
-	}
-	return error;
-}
-
-/* Seek is currently unimplemented */
-GIT_INLINE(int) git_iterator_seek(
-	git_iterator *iter, const char *prefix)
+GIT_INLINE(int) git_iterator_advance_over(
+	const git_index_entry **entry,
+	git_iterator_status_t *status,
+	git_iterator *iter)
 {
-	return iter->cb->seek(iter, prefix);
+	return iter->cb->advance_over(entry, status, iter);
 }
 
 /**
  * Go back to the start of the iteration.
- *
- * This resets the iterator to the start of the iteration.  It also allows
- * you to reset the `start` and `end` pathname boundaries of the iteration
- * when doing so.
  */
-GIT_INLINE(int) git_iterator_reset(
-	git_iterator *iter, const char *start, const char *end)
+GIT_INLINE(int) git_iterator_reset(git_iterator *iter)
 {
-	return iter->cb->reset(iter, start, end);
+	return iter->cb->reset(iter);
 }
 
 /**
- * Check if the iterator is at the end
- *
- * @return 0 if not at end, >0 if at end
+ * Go back to the start of the iteration after updating the `start` and
+ * `end` pathname boundaries of the iteration.
  */
-GIT_INLINE(int) git_iterator_at_end(git_iterator *iter)
-{
-	return iter->cb->at_end(iter);
-}
+extern int git_iterator_reset_range(
+	git_iterator *iter, const char *start, const char *end);
 
 GIT_INLINE(git_iterator_type_t) git_iterator_type(git_iterator *iter)
 {
@@ -222,6 +248,11 @@ GIT_INLINE(git_repository *) git_iterator_owner(git_iterator *iter)
 	return iter->repo;
 }
 
+GIT_INLINE(git_index *) git_iterator_index(git_iterator *iter)
+{
+	return iter->index;
+}
+
 GIT_INLINE(git_iterator_flag_t) git_iterator_flags(git_iterator *iter)
 {
 	return iter->flags;
@@ -232,18 +263,18 @@ GIT_INLINE(bool) git_iterator_ignore_case(git_iterator *iter)
 	return ((iter->flags & GIT_ITERATOR_IGNORE_CASE) != 0);
 }
 
-extern int git_iterator_set_ignore_case(git_iterator *iter, bool ignore_case);
+extern void git_iterator_set_ignore_case(
+	git_iterator *iter, bool ignore_case);
 
 extern int git_iterator_current_tree_entry(
 	const git_tree_entry **entry_out, git_iterator *iter);
 
 extern int git_iterator_current_parent_tree(
-	const git_tree **tree_out, git_iterator *iter, const char *parent_path);
+	const git_tree **tree_out, git_iterator *iter, size_t depth);
 
 extern bool git_iterator_current_is_ignored(git_iterator *iter);
 
-extern int git_iterator_cmp(
-	git_iterator *iter, const char *path_prefix);
+extern bool git_iterator_current_tree_is_ignored(git_iterator *iter);
 
 /**
  * Get full path of the current item from a workdir iterator.  This will
@@ -253,7 +284,39 @@ extern int git_iterator_cmp(
 extern int git_iterator_current_workdir_path(
 	git_buf **path, git_iterator *iter);
 
-/* Return index pointer if index iterator, else NULL */
-extern git_index *git_iterator_get_index(git_iterator *iter);
+/**
+ * Retrieve the index stored in the iterator.
+ *
+ * Only implemented for the workdir and index iterators.
+ */
+extern git_index *git_iterator_index(git_iterator *iter);
+
+typedef int (*git_iterator_foreach_cb)(
+	const git_index_entry *entry,
+	void *data);
+
+/**
+ * Walk the given iterator and invoke the callback for each path
+ * contained in the iterator.
+ */
+extern int git_iterator_foreach(
+	git_iterator *iterator,
+	git_iterator_foreach_cb cb,
+	void *data);
+
+typedef int (*git_iterator_walk_cb)(
+	const git_index_entry **entries,
+	void *data);
+
+/**
+ * Walk the given iterators in lock-step.  The given callback will be
+ * called for each unique path, with the index entry in each iterator
+ * (or NULL if the given iterator does not contain that path).
+ */
+extern int git_iterator_walk(
+	git_iterator **iterators,
+	size_t cnt,
+	git_iterator_walk_cb cb,
+	void *data);
 
 #endif

data/vendor/libgit2/src/khash.h

@@ -46,6 +46,19 @@ int main() {
 */
 
 /*
+  2013-05-02 (0.2.8):
+
+	* Use quadratic probing. When the capacity is power of 2, stepping function
+	  i*(i+1)/2 guarantees to traverse each bucket. It is better than double
+	  hashing on cache performance and is more robust than linear probing.
+
+	  In theory, double hashing should be more robust than quadratic probing.
+	  However, my implementation is probably not for large hash tables, because
+	  the second hash function is closely tied to the first hash function,
+	  which reduce the effectiveness of double hashing.
+
+	Reference: http://research.cs.vt.edu/AVresearch/hashing/quadratic.php
+
   2011-12-29 (0.2.7):
 
     * Minor code clean up; no actual effect.
@@ -110,13 +123,13 @@ int main() {
   Generic hash table library.
  */
 
-#define AC_VERSION_KHASH_H "0.2.6"
+#define AC_VERSION_KHASH_H "0.2.8"
 
 #include <stdlib.h>
 #include <string.h>
 #include <limits.h>
 
-/* compipler specific configuration */
+/* compiler specific configuration */
 
 #if UINT_MAX == 0xffffffffu
 typedef unsigned int khint32_t;
@@ -130,11 +143,15 @@ typedef unsigned long khint64_t;
 typedef unsigned long long khint64_t;
 #endif
 
+#ifndef kh_inline
 #ifdef _MSC_VER
 #define kh_inline __inline
+#elif defined(__GNUC__)
+#define kh_inline __inline__
 #else
-#define kh_inline inline
+#define kh_inline
 #endif
+#endif /* kh_inline */
 
 typedef khint32_t khint_t;
 typedef khint_t khiter_t;
@@ -147,12 +164,6 @@ typedef khint_t khiter_t;
 #define __ac_set_isboth_false(flag, i) (flag[i>>4]&=~(3ul<<((i&0xfU)<<1)))
 #define __ac_set_isdel_true(flag, i) (flag[i>>4]|=1ul<<((i&0xfU)<<1))
 
-#ifdef KHASH_LINEAR
-#define __ac_inc(k, m) 1
-#else
-#define __ac_inc(k, m) (((k)>>3 ^ (k)<<3) | 1) & (m)
-#endif
-
 #define __ac_fsize(m) ((m) < 16? 1 : (m)>>4)
 
 #ifndef kroundup32
@@ -168,6 +179,9 @@ typedef khint_t khiter_t;
 #ifndef krealloc
 #define krealloc(P,Z) realloc(P,Z)
 #endif
+#ifndef kreallocarray
+#define kreallocarray(P,N,Z) ((SIZE_MAX - N < Z) ? NULL : krealloc(P, (N*Z)))
+#endif
 #ifndef kfree
 #define kfree(P) free(P)
 #endif
@@ -175,7 +189,7 @@ typedef khint_t khiter_t;
 static const double __ac_HASH_UPPER = 0.77;
 
 #define __KHASH_TYPE(name, khkey_t, khval_t) \
-	typedef struct { \
+	typedef struct kh_##name##_s { \
 		khint_t n_buckets, size, n_occupied, upper_bound; \
 		khint32_t *flags; \
 		khkey_t *keys; \
@@ -213,19 +227,19 @@ static const double __ac_HASH_UPPER = 0.77;
 	SCOPE khint_t kh_get_##name(const kh_##name##_t *h, khkey_t key) 	\
 	{																	\
 		if (h->n_buckets) {												\
-			khint_t inc, k, i, last, mask;								\
+			khint_t k, i, last, mask, step = 0; \
 			mask = h->n_buckets - 1;									\
 			k = __hash_func(key); i = k & mask;							\
-			inc = __ac_inc(k, mask); last = i; /* inc==1 for linear probing */ \
+			last = i; \
 			while (!__ac_isempty(h->flags, i) && (__ac_isdel(h->flags, i) || !__hash_equal(h->keys[i], key))) { \
-				i = (i + inc) & mask; 									\
+				i = (i + (++step)) & mask; \
 				if (i == last) return h->n_buckets;						\
 			}															\
 			return __ac_iseither(h->flags, i)? h->n_buckets : i;		\
 		} else return 0;												\
 	}																	\
 	SCOPE int kh_resize_##name(kh_##name##_t *h, khint_t new_n_buckets) \
-	{ /* This function uses 0.25*n_bucktes bytes of working space instead of [sizeof(key_t+val_t)+.25]*n_buckets. */ \
+	{ /* This function uses 0.25*n_buckets bytes of working space instead of [sizeof(key_t+val_t)+.25]*n_buckets. */ \
 		khint32_t *new_flags = 0;										\
 		khint_t j = 1;													\
 		{																\
@@ -233,16 +247,16 @@ static const double __ac_HASH_UPPER = 0.77;
 			if (new_n_buckets < 4) new_n_buckets = 4;					\
 			if (h->size >= (khint_t)(new_n_buckets * __ac_HASH_UPPER + 0.5)) j = 0;	/* requested size is too small */ \
 			else { /* hash table size to be changed (shrink or expand); rehash */ \
-				new_flags = (khint32_t*)kmalloc(__ac_fsize(new_n_buckets) * sizeof(khint32_t));	\
+				new_flags = (khint32_t*)kreallocarray(NULL, __ac_fsize(new_n_buckets), sizeof(khint32_t)); \
 				if (!new_flags) return -1;								\
 				memset(new_flags, 0xaa, __ac_fsize(new_n_buckets) * sizeof(khint32_t)); \
 				if (h->n_buckets < new_n_buckets) {	/* expand */		\
-					khkey_t *new_keys = (khkey_t*)krealloc((void *)h->keys, new_n_buckets * sizeof(khkey_t)); \
-					if (!new_keys) return -1;							\
+					khkey_t *new_keys = (khkey_t*)kreallocarray((void *)h->keys, new_n_buckets, sizeof(khkey_t)); \
+					if (!new_keys) { kfree(new_flags); return -1; }		\
 					h->keys = new_keys;									\
 					if (kh_is_map) {									\
-						khval_t *new_vals = (khval_t*)krealloc((void *)h->vals, new_n_buckets * sizeof(khval_t)); \
-						if (!new_vals) return -1;						\
+						khval_t *new_vals = (khval_t*)kreallocarray((void *)h->vals, new_n_buckets, sizeof(khval_t)); \
+						if (!new_vals) { kfree(new_flags); return -1; }	\
 						h->vals = new_vals;								\
 					}													\
 				} /* otherwise shrink */								\
@@ -258,11 +272,10 @@ static const double __ac_HASH_UPPER = 0.77;
 					if (kh_is_map) val = h->vals[j];					\
 					__ac_set_isdel_true(h->flags, j);					\
 					while (1) { /* kick-out process; sort of like in Cuckoo hashing */ \
-						khint_t inc, k, i;								\
+						khint_t k, i, step = 0; \
 						k = __hash_func(key);							\
 						i = k & new_mask;								\
-						inc = __ac_inc(k, new_mask);					\
-						while (!__ac_isempty(new_flags, i)) i = (i + inc) & new_mask; \
+						while (!__ac_isempty(new_flags, i)) i = (i + (++step)) & new_mask; \
 						__ac_set_isempty_false(new_flags, i);			\
 						if (i < h->n_buckets && __ac_iseither(h->flags, i) == 0) { /* kick out the existing element */ \
 							{ khkey_t tmp = h->keys[i]; h->keys[i] = key; key = tmp; } \
@@ -277,8 +290,8 @@ static const double __ac_HASH_UPPER = 0.77;
 				}														\
 			}															\
 			if (h->n_buckets > new_n_buckets) { /* shrink the hash table */ \
-				h->keys = (khkey_t*)krealloc((void *)h->keys, new_n_buckets * sizeof(khkey_t)); \
-				if (kh_is_map) h->vals = (khval_t*)krealloc((void *)h->vals, new_n_buckets * sizeof(khval_t)); \
+				h->keys = (khkey_t*)kreallocarray((void *)h->keys, new_n_buckets, sizeof(khkey_t)); \
+				if (kh_is_map) h->vals = (khval_t*)kreallocarray((void *)h->vals, new_n_buckets, sizeof(khval_t)); \
 			}															\
 			kfree(h->flags); /* free the working space */				\
 			h->flags = new_flags;										\
@@ -301,14 +314,14 @@ static const double __ac_HASH_UPPER = 0.77;
 			}															\
 		} /* TODO: to implement automatically shrinking; resize() already support shrinking */ \
 		{																\
-			khint_t inc, k, i, site, last, mask = h->n_buckets - 1;		\
+			khint_t k, i, site, last, mask = h->n_buckets - 1, step = 0; \
 			x = site = h->n_buckets; k = __hash_func(key); i = k & mask; \
 			if (__ac_isempty(h->flags, i)) x = i; /* for speed up */	\
 			else {														\
-				inc = __ac_inc(k, mask); last = i;						\
+				last = i; \
 				while (!__ac_isempty(h->flags, i) && (__ac_isdel(h->flags, i) || !__hash_equal(h->keys[i], key))) { \
 					if (__ac_isdel(h->flags, i)) site = i;				\
-					i = (i + inc) & mask; 								\
+					i = (i + (++step)) & mask; \
 					if (i == last) { x = site; break; }					\
 				}														\
 				if (x == h->n_buckets) {								\
@@ -449,7 +462,8 @@ static kh_inline khint_t __ac_Wang_hash(khint_t key)
   @param  name  Name of the hash table [symbol]
   @param  h     Pointer to the hash table [khash_t(name)*]
   @param  k     Key [type of keys]
-  @param  r     Extra return code: 0 if the key is present in the hash table;
+  @param  r     Extra return code: -1 if the operation failed;
+                0 if the key is present in the hash table;
                 1 if the bucket is empty (never used); 2 if the element in
 				the bucket has been deleted [int*]
   @return       Iterator to the inserted element [khint_t]
@@ -461,7 +475,7 @@ static kh_inline khint_t __ac_Wang_hash(khint_t key)
   @param  name  Name of the hash table [symbol]
   @param  h     Pointer to the hash table [khash_t(name)*]
   @param  k     Key [type of keys]
-  @return       Iterator to the found element, or kh_end(h) is the element is absent [khint_t]
+  @return       Iterator to the found element, or kh_end(h) if the element is absent [khint_t]
  */
 #define kh_get(name, h, k) kh_get_##name(h, k)