/* hash - hashing table processing.
- Copyright (C) 1998-2004, 2006-2007, 2009-2011 Free Software Foundation, Inc.
+ Copyright (C) 1998-2004, 2006-2007, 2009-2013 Free Software Foundation, Inc.
Written by Jim Meyering, 1992.
#include "hash.h"
#include "bitrotate.h"
-#include "xalloc.h"
+#include "xalloc-oversized.h"
#include <stdint.h>
#include <stdio.h>
# endif
#endif
-/* The attribute __const__ was added in gcc 2.95. */
-#undef _GL_ATTRIBUTE_CONST
-#if __GNUC__ > 2 || (__GNUC__ == 2 && __GNUC_MINOR__ >= 95)
-# define _GL_ATTRIBUTE_CONST __attribute__ ((__const__))
-#else
-# define _GL_ATTRIBUTE_CONST /* empty */
-#endif
-
struct hash_entry
{
void *data;
/* Tuning arguments, kept in a physically separate structure. */
const Hash_tuning *tuning;
- /* Three functions are given to `hash_initialize', see the documentation
+ /* Three functions are given to 'hash_initialize', see the documentation
block for this function. In a word, HASHER randomizes a user entry
into a number up from 0 up to some maximum minus 1; COMPARATOR returns
true if two user entries compare equally; and DATA_FREER is the cleanup
some user-provided data (also called a user entry). An entry indistinctly
refers to both the internal entry and its associated user entry. A user
entry contents may be hashed by a randomization function (the hashing
- function, or just `hasher' for short) into a number (or `slot') between 0
+ function, or just "hasher" for short) into a number (or "slot") between 0
and the current table size. At each slot position in the hash table,
starts a linked chain of entries for which the user data all hash to this
slot. A bucket is the collection of all entries hashing to the same slot.
- A good `hasher' function will distribute entries rather evenly in buckets.
+ A good "hasher" function will distribute entries rather evenly in buckets.
In the ideal case, the length of each bucket is roughly the number of
entries divided by the table size. Finding the slot for a data is usually
- done in constant time by the `hasher', and the later finding of a precise
+ done in constant time by the "hasher", and the later finding of a precise
entry is linear in time with the size of the bucket. Consequently, a
larger hash table size (that is, a larger number of buckets) is prone to
- yielding shorter chains, *given* the `hasher' function behaves properly.
+ yielding shorter chains, *given* the "hasher" function behaves properly.
Long buckets slow down the lookup algorithm. One might use big hash table
sizes in hope to reduce the average length of buckets, but this might
become inordinate, as unused slots in the hash table take some space. The
- best bet is to make sure you are using a good `hasher' function (beware
+ best bet is to make sure you are using a good "hasher" function (beware
that those are not that easy to write! :-), and to use a table size
larger than the actual number of entries. */
1.0). The growth threshold defaults to 0.8, and the growth factor
defaults to 1.414, meaning that the table will have doubled its size
every second time 80% of the buckets get used. */
-#define DEFAULT_GROWTH_THRESHOLD 0.8
-#define DEFAULT_GROWTH_FACTOR 1.414
+#define DEFAULT_GROWTH_THRESHOLD 0.8f
+#define DEFAULT_GROWTH_FACTOR 1.414f
/* If a deletion empties a bucket and causes the ratio of used buckets to
table size to become smaller than the shrink threshold (a number between
number greater than the shrink threshold but smaller than 1.0). The shrink
threshold and factor default to 0.0 and 1.0, meaning that the table never
shrinks. */
-#define DEFAULT_SHRINK_THRESHOLD 0.0
-#define DEFAULT_SHRINK_FACTOR 1.0
+#define DEFAULT_SHRINK_THRESHOLD 0.0f
+#define DEFAULT_SHRINK_FACTOR 1.0f
/* Use this to initialize or reset a TUNING structure to
some sensible values. */
number of buckets (used plus unused), or the maximum number of slots, are
the same quantity. */
-size_t _GL_ATTRIBUTE_PURE
+size_t
hash_get_n_buckets (const Hash_table *table)
{
return table->n_buckets;
/* Return the number of slots in use (non-empty buckets). */
-size_t _GL_ATTRIBUTE_PURE
+size_t
hash_get_n_buckets_used (const Hash_table *table)
{
return table->n_buckets_used;
/* Return the number of active entries. */
-size_t _GL_ATTRIBUTE_PURE
+size_t
hash_get_n_entries (const Hash_table *table)
{
return table->n_entries;
/* Return the length of the longest chain (bucket). */
-size_t _GL_ATTRIBUTE_PURE
+size_t
hash_get_max_bucket_length (const Hash_table *table)
{
struct hash_entry const *bucket;
/* Do a mild validation of a hash table, by traversing it and checking two
statistics. */
-bool _GL_ATTRIBUTE_PURE
+bool
hash_table_ok (const Hash_table *table)
{
struct hash_entry const *bucket;
/* Return the first data in the table, or NULL if the table is empty. */
-void * _GL_ATTRIBUTE_PURE
+void *
hash_get_first (const Hash_table *table)
{
struct hash_entry const *bucket;
}
/* Return the user data for the entry following ENTRY, where ENTRY has been
- returned by a previous call to either `hash_get_first' or `hash_get_next'.
+ returned by a previous call to either 'hash_get_first' or 'hash_get_next'.
Return NULL if there are no more entries. */
void *
algorithms tend to be domain-specific, so what's good for [diffutils'] io.c
may not be good for your application." */
-size_t _GL_ATTRIBUTE_PURE
+size_t
hash_string (const char *string, size_t n_buckets)
{
# define HASH_ONE_CHAR(Value, Byte) \
#else /* not USE_DIFF_HASH */
-/* This one comes from `recode', and performs a bit better than the above as
+/* This one comes from 'recode', and performs a bit better than the above as
per a few experiments. It is inspired from a hashing routine found in the
- very old Cyber `snoop', itself written in typical Greg Mansfield style.
+ very old Cyber 'snoop', itself written in typical Greg Mansfield style.
(By the way, what happened to this excellent man? Is he still alive?) */
size_t
The user-supplied DATA_FREER function, when not NULL, may be later called
with the user data as an argument, just before the entry containing the
- data gets freed. This happens from within `hash_free' or `hash_clear'.
+ data gets freed. This happens from within 'hash_free' or 'hash_clear'.
You should specify this function only if you want these functions to free
- all of your `data' data. This is typically the case when your data is
+ all of your 'data' data. This is typically the case when your data is
simply an auxiliary struct that you have malloc'd to aggregate several
values. */
return false;
}
-/* Return -1 upon memory allocation failure.
+/* Insert ENTRY into hash TABLE if there is not already a matching entry.
+
+ Return -1 upon memory allocation failure.
Return 1 if insertion succeeded.
Return 0 if there is already a matching entry in the table,
and in that case, if MATCHED_ENT is non-NULL, set *MATCHED_ENT
hash_insert, the only way to distinguish those cases is to compare
the return value and ENTRY. That works only when you can have two
different ENTRY values that point to data that compares "equal". Thus,
- when the ENTRY value is a simple scalar, you must use hash_insert0.
- ENTRY must not be NULL. */
+ when the ENTRY value is a simple scalar, you must use
+ hash_insert_if_absent. ENTRY must not be NULL. */
int
-hash_insert0 (Hash_table *table, void const *entry, void const **matched_ent)
+hash_insert_if_absent (Hash_table *table, void const *entry,
+ void const **matched_ent)
{
void *data;
struct hash_entry *bucket;
return 1;
}
+/* hash_insert0 is the deprecated name for hash_insert_if_absent.
+ . */
+int
+hash_insert0 (Hash_table *table, void const *entry, void const **matched_ent)
+{
+ return hash_insert_if_absent (table, entry, matched_ent);
+}
+
/* If ENTRY matches an entry already in the hash table, return the pointer
to the entry from the table. Otherwise, insert ENTRY and return ENTRY.
Return NULL if the storage required for insertion cannot be allocated.
hash_insert (Hash_table *table, void const *entry)
{
void const *matched_ent;
- int err = hash_insert0 (table, entry, &matched_ent);
+ int err = hash_insert_if_absent (table, entry, &matched_ent);
return (err == -1
? NULL
: (void *) (err == 0 ? matched_ent : entry));
projects / gnulib.git / blobdiff