The data structure used in unaccent is a trie, not suffix tree.

Fix the term used in variable and struct names, and comments.

Alexander Korotkov
This commit is contained in:
Heikki Linnakangas 2013-05-08 20:57:42 +03:00
parent 2ffa66f497
commit 4b06c1820a
1 changed files with 28 additions and 28 deletions

View File

@ -23,30 +23,29 @@
PG_MODULE_MAGIC;
/*
* Unaccent dictionary uses uncompressed suffix tree to find a
* character to replace. Each node of tree is an array of
* SuffixChar struct with length = 256 (n-th element of array
* Unaccent dictionary uses a trie to find a character to replace. Each node of
* the trie is an array of 256 TrieChar structs (n-th element of array
* corresponds to byte)
*/
typedef struct SuffixChar
typedef struct TrieChar
{
struct SuffixChar *nextChar;
struct TrieChar *nextChar;
char *replaceTo;
int replacelen;
} SuffixChar;
} TrieChar;
/*
* placeChar - put str into tree's structure, byte by byte.
* placeChar - put str into trie's structure, byte by byte.
*/
static SuffixChar *
placeChar(SuffixChar *node, unsigned char *str, int lenstr, char *replaceTo, int replacelen)
static TrieChar *
placeChar(TrieChar *node, unsigned char *str, int lenstr, char *replaceTo, int replacelen)
{
SuffixChar *curnode;
TrieChar *curnode;
if (!node)
{
node = palloc(sizeof(SuffixChar) * 256);
memset(node, 0, sizeof(SuffixChar) * 256);
node = palloc(sizeof(TrieChar) * 256);
memset(node, 0, sizeof(TrieChar) * 256);
}
curnode = node + *str;
@ -71,13 +70,14 @@ placeChar(SuffixChar *node, unsigned char *str, int lenstr, char *replaceTo, int
}
/*
* initSuffixTree - create suffix tree from file. Function converts
* UTF8-encoded file into current encoding.
* initTrie - create trie from file.
*
* Function converts UTF8-encoded file into current encoding.
*/
static SuffixChar *
initSuffixTree(char *filename)
static TrieChar *
initTrie(char *filename)
{
SuffixChar *volatile rootSuffixTree = NULL;
TrieChar *volatile rootTrie = NULL;
MemoryContext ccxt = CurrentMemoryContext;
tsearch_readline_state trst;
volatile bool skip;
@ -161,7 +161,7 @@ initSuffixTree(char *filename)
}
if (state >= 3)
rootSuffixTree = placeChar(rootSuffixTree,
rootTrie = placeChar(rootTrie,
(unsigned char *) src, srclen,
trg, trglen);
@ -192,14 +192,14 @@ initSuffixTree(char *filename)
tsearch_readline_end(&trst);
return rootSuffixTree;
return rootTrie;
}
/*
* findReplaceTo - find multibyte character in tree
* findReplaceTo - find multibyte character in trie
*/
static SuffixChar *
findReplaceTo(SuffixChar *node, unsigned char *src, int srclen)
static TrieChar *
findReplaceTo(TrieChar *node, unsigned char *src, int srclen)
{
while (node)
{
@ -221,7 +221,7 @@ Datum
unaccent_init(PG_FUNCTION_ARGS)
{
List *dictoptions = (List *) PG_GETARG_POINTER(0);
SuffixChar *rootSuffixTree = NULL;
TrieChar *rootTrie = NULL;
bool fileloaded = false;
ListCell *l;
@ -235,7 +235,7 @@ unaccent_init(PG_FUNCTION_ARGS)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("multiple Rules parameters")));
rootSuffixTree = initSuffixTree(defGetString(defel));
rootTrie = initTrie(defGetString(defel));
fileloaded = true;
}
else
@ -254,7 +254,7 @@ unaccent_init(PG_FUNCTION_ARGS)
errmsg("missing Rules parameter")));
}
PG_RETURN_POINTER(rootSuffixTree);
PG_RETURN_POINTER(rootTrie);
}
PG_FUNCTION_INFO_V1(unaccent_lexize);
@ -262,21 +262,21 @@ Datum unaccent_lexize(PG_FUNCTION_ARGS);
Datum
unaccent_lexize(PG_FUNCTION_ARGS)
{
SuffixChar *rootSuffixTree = (SuffixChar *) PG_GETARG_POINTER(0);
TrieChar *rootTrie = (TrieChar *) PG_GETARG_POINTER(0);
char *srcchar = (char *) PG_GETARG_POINTER(1);
int32 len = PG_GETARG_INT32(2);
char *srcstart,
*trgchar = NULL;
int charlen;
TSLexeme *res = NULL;
SuffixChar *node;
TrieChar *node;
srcstart = srcchar;
while (srcchar - srcstart < len)
{
charlen = pg_mblen(srcchar);
node = findReplaceTo(rootSuffixTree, (unsigned char *) srcchar, charlen);
node = findReplaceTo(rootTrie, (unsigned char *) srcchar, charlen);
if (node && node->replaceTo)
{
if (!res)