Do not broke words with numbers and letters, divided by hyphen

Author: Arthur Zakirov

Diff for: README.md

@@ -3,9 +3,27 @@
 ## Introduction
 
 The **pg_tsparser** module is the modified default text search parser from
-PostgreSQL 9.6.
-The difference between **tsparser** and **default** parsers is that **tsparser**
-gives also unbroken words by underscore character.
+PostgreSQL 9.6. The differences are:
+* **tsparser** gives unbroken words by underscore character
+* **tsparser** gives unbroken words with numbers and letters by hyphen character
+
+For example:
+
+```sql
+SELECT to_tsvector('english', 'pg_trgm') as def_parser,
+       to_tsvector('english_ts', 'pg_trgm')  as new_parser;
+   def_parser    |         new_parser
+-----------------+-----------------------------
+ 'pg':1 'trgm':2 | 'pg':2 'pg_trgm':1 'trgm':3
+(1 row)
+
+SELECT to_tsvector('english', '123-abc') as def_parser,
+       to_tsvector('english_ts', '123-abc')  as new_parser;
+   def_parser    |         new_parser
+-----------------+-----------------------------
+ '123':1 'abc':2 | '123':2 '123-abc':1 'abc':3
+(1 row)
+```
 
 ## License
 
@@ -40,21 +58,3 @@ ALTER TEXT SEARCH CONFIGURATION english_ts
     word, hword, hword_part
     WITH english_stem;
 ```
-
-## Examples
-
-Example of difference between **tsparser** and **default**:
-
-```sql
-SELECT to_tsvector('english_ts', 'pg_trgm');
-         to_tsvector
------------------------------
- 'pg':2 'pg_trgm':1 'trgm':3
-(1 row)
-
-SELECT to_tsvector('english', 'pg_trgm');
-   to_tsvector
------------------
- 'pg':1 'trgm':2
-(1 row)
-```

Diff for: expected/pg_tsparser.out

@@ -193,3 +193,15 @@ SELECT to_tsvector('english_ts', 'pg_trgm');
  'pg':2 'pg_trgm':1 'trgm':3
 (1 row)
 
+SELECT to_tsvector('english_ts', '12_abc');
+        to_tsvector        
+---------------------------
+ '12':2 '12_abc':1 'abc':3
+(1 row)
+
+SELECT to_tsvector('english_ts', '12-abc');
+        to_tsvector        
+---------------------------
+ '12':2 '12-abc':1 'abc':3
+(1 row)
+

Diff for: sql/pg_tsparser.sql

@@ -22,3 +22,5 @@ ALTER TEXT SEARCH CONFIGURATION english_ts
     WITH english_stem;
 
 SELECT to_tsvector('english_ts', 'pg_trgm');
+SELECT to_tsvector('english_ts', '12_abc');
+SELECT to_tsvector('english_ts', '12-abc');

Diff for: tsparser.c

@@ -1132,6 +1132,8 @@ static const TParserStateActionItem actionTPS_InUnsignedInt[] = {
 	{p_iseqC, '-', A_PUSH, TPS_InHostFirstAN, 0, NULL},
 	{p_iseqC, '_', A_PUSH, TPS_InHostFirstAN, 0, NULL},
 	{p_iseqC, '@', A_PUSH, TPS_InEmail, 0, NULL},
+	{p_iseqC, '-', A_PUSH, TPS_InHyphenNumWordFirst, 0, NULL},
+	{p_iseqC, '_', A_PUSH, TPS_InHyphenNumWordFirst, 0, NULL},
 	{p_isasclet, 0, A_PUSH, TPS_InHost, 0, NULL},
 	{p_isalpha, 0, A_NEXT, TPS_InNumWord, 0, NULL},
 	{p_isspecial, 0, A_NEXT, TPS_InNumWord, 0, NULL},
@@ -1658,7 +1660,7 @@ static const TParserStateActionItem actionTPS_InParseHyphen[] = {
 	{p_isEOF, 0, A_RERUN, TPS_Base, 0, NULL},
 	{p_isasclet, 0, A_NEXT, TPS_InHyphenAsciiWordPart, 0, NULL},
 	{p_isalpha, 0, A_NEXT, TPS_InHyphenWordPart, 0, NULL},
-	{p_isdigit, 0, A_PUSH, TPS_InHyphenUnsignedInt, 0, NULL},
+	{p_isdigit, 0, A_PUSH, TPS_InHyphenNumWordPart, 0, NULL},
 	{p_iseqC, '-', A_PUSH, TPS_InParseHyphenHyphen, 0, NULL},
 	{p_iseqC, '_', A_PUSH, TPS_InParseHyphenHyphen, 0, NULL},
 	{NULL, 0, A_RERUN, TPS_Base, 0, NULL}