Optimize. Improve ispell support for compound words. This work was sponsored by ABC Startsiden AS.

contrib/tsearch2/dict_ispell.c

@@ -27,7 +27,7 @@ Datum		spell_lexize(PG_FUNCTION_ARGS);
 static void
 freeDictISpell(DictISpell * d)
 {
-	FreeIspell(&(d->obj));
+	NIFree(&(d->obj));
 	freestoplist(&(d->stoplist));
 	free(d);
 }
@@ -71,7 +71,7 @@ spell_init(PG_FUNCTION_ARGS)
 					  (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
 					   errmsg("dictionary already loaded")));
 			}
-			if (ImportDictionary(&(d->obj), pcfg->value))
+			if (NIImportDictionary(&(d->obj), pcfg->value))
 			{
 				freeDictISpell(d);
 				ereport(ERROR,
@@ -90,7 +90,7 @@ spell_init(PG_FUNCTION_ARGS)
 					  (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
 					   errmsg("affixes already loaded")));
 			}
-			if (ImportAffixes(&(d->obj), pcfg->value))
+			if (NIImportAffixes(&(d->obj), pcfg->value))
 			{
 				freeDictISpell(d);
 				ereport(ERROR,
@@ -132,8 +132,8 @@ spell_init(PG_FUNCTION_ARGS)
 
 	if (affloaded && dictloaded)
 	{
-		SortDictionary(&(d->obj));
-		SortAffixes(&(d->obj));
+		NISortDictionary(&(d->obj));
+		NISortAffixes(&(d->obj));
 	}
 	else if (!affloaded)
 	{
@@ -168,7 +168,7 @@ spell_lexize(PG_FUNCTION_ARGS)
 
 	res = palloc(sizeof(char *) * 2);
 	txt = pnstrdup(in, PG_GETARG_INT32(2));
-	res = NormalizeWord(&(d->obj), txt);
+	res = NINormalizeWord(&(d->obj), txt);
 	pfree(txt);
 
 	if (res == NULL)

contrib/tsearch2/ispell/spell.h

@@ -4,15 +4,43 @@
 #include <sys/types.h>
 #include <regex.h>
 
+
+struct SPNode;
+
+
+typedef struct {
+	u_int32_t 
+		val:8,
+		isword:1,
+		compoundallow:1,
+		affix:22;
+	struct SPNode *node; 
+} SPNodeData;
+
+typedef struct SPNode {
+	u_int32_t 	length;
+	SPNodeData	data[1];	
+} SPNode;
+
+#define SPNHRDSZ	(sizeof(u_int32_t))
+
+
 typedef struct spell_struct
 {
 	char	   *word;
-	char		flag[10];
+	union {
+		char		flag[16];
+		struct {
+			int		affix;
+			int 		len;
+		} d;
+	} p;
 }	SPELL;
 
 typedef struct aff_struct
 {
 	char		flag;
+	char		flagflags;
 	char		type;
 	char		mask[33];
 	char		find[16];
@@ -22,35 +50,66 @@ typedef struct aff_struct
 	char		compile;
 }	AFFIX;
 
+#define FF_CROSSPRODUCT 	0x01
+#define FF_COMPOUNDWORD 	0x02
+#define FF_COMPOUNDONLYAFX      0x04
+
+struct AffixNode;
+
+typedef struct {
+	u_int32_t
+		val:8,
+		naff:24;
+	AFFIX   **aff;
+	struct AffixNode *node;
+} AffixNodeData;
+
+typedef struct AffixNode {
+	u_int32_t length;
+	AffixNodeData	data[1];
+} AffixNode;
+
+#define ANHRDSZ        (sizeof(u_int32_t))
+
 typedef struct Tree_struct
 {
 	int			Left[256],
 				Right[256];
 }	Tree_struct;
 
+typedef struct {
+	char *affix;
+	int len;
+} CMPDAffix;
+
 typedef struct
 {
 	int			maffixes;
 	int			naffixes;
 	AFFIX	   *Affix;
+	char			compoundcontrol;
 
 	int			nspell;
 	int			mspell;
 	SPELL	   *Spell;
-	Tree_struct SpellTree;
-	Tree_struct PrefixTree;
-	Tree_struct SuffixTree;
+
+	AffixNode	*Suffix;
+	AffixNode	*Prefix;
+
+	SPNode	*Dictionary;
+	char	**AffixData;
+	CMPDAffix    *CompoundAffix;
 
 }	IspellDict;
 
-char	  **NormalizeWord(IspellDict * Conf, char *word);
-int			ImportAffixes(IspellDict * Conf, const char *filename);
-int			ImportDictionary(IspellDict * Conf, const char *filename);
+char	  **NINormalizeWord(IspellDict * Conf, char *word);
+int			NIImportAffixes(IspellDict * Conf, const char *filename);
+int			NIImportDictionary(IspellDict * Conf, const char *filename);
 
-int			AddSpell(IspellDict * Conf, const char *word, const char *flag);
-int			AddAffix(IspellDict * Conf, int flag, const char *mask, const char *find, const char *repl, int type);
-void		SortDictionary(IspellDict * Conf);
-void		SortAffixes(IspellDict * Conf);
-void		FreeIspell(IspellDict * Conf);
+int			NIAddSpell(IspellDict * Conf, const char *word, const char *flag);
+int			NIAddAffix(IspellDict * Conf, int flag, char flagflags, const char *mask, const char *find, const char *repl, int type);
+void		NISortDictionary(IspellDict * Conf);
+void		NISortAffixes(IspellDict * Conf);
+void		NIFree(IspellDict * Conf);
 
 #endif