can config es id's constituent part

README.md

@@ -23,7 +23,7 @@ It uses `mysqldump` to fetch the origin data at first, then syncs data increment
 + binlog format must be **row**.
 + binlog row image must be **full** for MySQL, you may lost some field data if you update PK data in MySQL with minimal or noblob binlog row image. MariaDB only supports full row image.
 + Can not alter table format at runtime.
-+ MySQL table which will be synced must have a PK(primary key), multi columns PK is allowed now, e,g, if the PKs is (a, b), we will use "a:b" as the key. The PK data will be used as "id" in Elasticsearch.
++ MySQL table which will be synced should have a PK(primary key), multi columns PK is allowed now, e,g, if the PKs is (a, b), we will use "a:b" as the key. The PK data will be used as "id" in Elasticsearch. And you can also config the id's constituent part with other column.
 + You should create the associated mappings in Elasticsearch first, I don't think using the default mapping is a wise decision, you must know how to search accurately.
 + `mysqldump` must exist in the same node with go-mysql-elasticsearch, if not, go-mysql-elasticsearch will try to sync binlog only.
 + Don't change too many rows at same time in one SQL.

etc/river.toml

@@ -113,3 +113,23 @@ type = "tfilter"
 
 # Only sync following columns
 filter = ["id", "name"]
+
+# id rule
+#
+# desc tid_[0-9]{4};
+# +----------+--------------+------+-----+---------+-------+
+# | Field    | Type         | Null | Key | Default | Extra |
+# +----------+--------------+------+-----+---------+-------+
+# | id       | int(11)      | NO   | PRI | NULL    |       |
+# | tag      | varchar(256) | YES  |     | NULL    |       |
+# | desc     | varchar(256) | YES  |     | NULL    |       |
+# +----------+--------------+------+-----+---------+-------+
+#
+[[rule]]
+schema = "test"
+table = "tid_[0-9]{4}"
+index = "test"
+type = "t"
+# The es doc's id will be `id`:`tag`
+# It is useful for merge muliple table into one type while theses tables have same PK 
+id = ["id", "tag"]

river/river.go

@@ -212,6 +212,7 @@ func (r *River) prepareRule() error {
 					rr.Index = rule.Index
 					rr.Type = rule.Type
 					rr.Parent = rule.Parent
+					rr.ID = rule.ID
 					rr.FieldMapping = rule.FieldMapping
 				}
 			} else {

river/river_test.go

@@ -47,6 +47,7 @@ func (s *riverTestSuite) SetUpSuite(c *C) {
 
 	s.testExecute(c, "DROP TABLE IF EXISTS test_river")
 	s.testExecute(c, fmt.Sprintf(schema, "test_river"))
+	s.testExecute(c, fmt.Sprintf(schema, "test_for_id"))
 
 	for i := 0; i < 10; i++ {
 		table := fmt.Sprintf("test_river_%04d", i)
@@ -72,7 +73,7 @@ func (s *riverTestSuite) SetUpSuite(c *C) {
 
 	os.RemoveAll(cfg.DataDir)
 
-	cfg.Sources = []SourceConfig{SourceConfig{Schema: "test", Tables: []string{"test_river", "test_river_[0-9]{4}"}}}
+	cfg.Sources = []SourceConfig{SourceConfig{Schema: "test", Tables: []string{"test_river", "test_river_[0-9]{4}", "test_for_id"}}}
 
 	cfg.Rules = []*Rule{
 		&Rule{Schema: "test",
@@ -81,6 +82,14 @@ func (s *riverTestSuite) SetUpSuite(c *C) {
 			Type:         "river",
 			FieldMapping: map[string]string{"title": "es_title", "mylist": "es_mylist,list"},
 		},
+
+		&Rule{Schema: "test",
+			Table:        "test_for_id",
+			Index:        "river",
+			Type:         "river",
+			ID:           []string{"id", "title"},
+			FieldMapping: map[string]string{"title": "es_title", "mylist": "es_mylist,list"},
+		},
 
 		&Rule{Schema: "test",
 			Table:        "test_river_[0-9]{4}",
@@ -120,7 +129,7 @@ data_dir = "./var"
 [[source]]
 schema = "test"
 
-tables = ["test_river", "test_river_[0-9]{4}"]
+tables = ["test_river", "test_river_[0-9]{4}", "test_for_id"]
 
 [[rule]]
 schema = "test"
@@ -133,6 +142,19 @@ parent = "pid"
     title = "es_title"
     mylist = "es_mylist,list"
 
+
+[[rule]]
+schema = "test"
+table = "test_for_id"
+index = "river"
+type = "river"
+parent = "pid"
+id = ["id", "title"]
+    [rule.field]
+    title = "es_title"
+    mylist = "es_mylist,list"
+
+
 [[rule]]
 schema = "test"
 table = "test_river_[0-9]{4}"
@@ -148,8 +170,8 @@ type = "river"
 	cfg, err := NewConfig(str)
 	c.Assert(err, IsNil)
 	c.Assert(cfg.Sources, HasLen, 1)
-	c.Assert(cfg.Sources[0].Tables, HasLen, 2)
-	c.Assert(cfg.Rules, HasLen, 2)
+	c.Assert(cfg.Sources[0].Tables, HasLen, 3)
+	c.Assert(cfg.Rules, HasLen, 3)
 }
 
 func (s *riverTestSuite) testExecute(c *C, query string, args ...interface{}) {
@@ -162,7 +184,8 @@ func (s *riverTestSuite) testPrepareData(c *C) {
 	s.testExecute(c, "INSERT INTO test_river (id, title, content, tenum, tset) VALUES (?, ?, ?, ?, ?)", 2, "second", "hello mysql 2", "e2", "b,c")
 	s.testExecute(c, "INSERT INTO test_river (id, title, content, tenum, tset) VALUES (?, ?, ?, ?, ?)", 3, "third", "hello elaticsearch 3", "e3", "c")
 	s.testExecute(c, "INSERT INTO test_river (id, title, content, tenum, tset, tbit) VALUES (?, ?, ?, ?, ?, ?)", 4, "fouth", "hello go-mysql-elasticserach 4", "e1", "a,b,c", 0)
-
+	s.testExecute(c, "INSERT INTO test_for_id (id, title, content, tenum, tset) VALUES (?, ?, ?, ?, ?)", 1, "first", "hello go 1", "e1", "a,b")
+
 	for i := 0; i < 10; i++ {
 		table := fmt.Sprintf("test_river_%04d", i)
 		s.testExecute(c, fmt.Sprintf("INSERT INTO %s (id, title, content, tenum, tset) VALUES (?, ?, ?, ?, ?)", table), 5+i, "abc", "hello", "e1", "a,b,c")
@@ -208,7 +231,10 @@ func (s *riverTestSuite) TestRiver(c *C) {
 	c.Assert(r.Found, Equals, true)
 	c.Assert(r.Source["tenum"], Equals, "e1")
 	c.Assert(r.Source["tset"], Equals, "a,b")
-
+
+	r = s.testElasticGet(c, "1:first")
+	c.Assert(r.Found, Equals, true)
+
 	r = s.testElasticGet(c, "100")
 	c.Assert(r.Found, Equals, false)
 

river/rule.go

@@ -13,6 +13,7 @@ type Rule struct {
 	Index  string `toml:"index"`
 	Type   string `toml:"type"`
 	Parent string `toml:"parent"`
+	ID []string `toml:"id"`
 
 	// Default, a MySQL table field name is mapped to Elasticsearch field name.
 	// Sometimes, you want to use different name, e.g, the MySQL file name is title,

river/sync.go

@@ -384,20 +384,35 @@ func (r *River) makeUpdateReqData(req *elastic.BulkRequest, rule *Rule,
 	}
 }
 
-// Get primary keys in one row and format them into a string
-// PK must not be nil
+// If id in toml file is none, get primary keys in one row and format them into a string, and PK must not be nil
+// Else get the ID's column in one row and format them into a string
 func (r *River) getDocID(rule *Rule, row []interface{}) (string, error) {
-	pks, err := canal.GetPKValues(rule.TableInfo, row)
-	if err != nil {
-		return "", err
+	var (
+  		ids []interface{}
+  		err error 
+	)
+	if rule.ID == nil {
+		ids, err = canal.GetPKValues(rule.TableInfo, row)
+		if err != nil {
+			return "", err
+		}
+	} else {
+		ids = make([]interface{}, 0, len(rule.ID))
+		for _, column := range rule.ID {
+			value, err := canal.GetColumnValue(rule.TableInfo, column, row)
+			if err != nil {
+				return "", err
+			}
+			ids = append(ids, value)
+		}
 	}
 
 	var buf bytes.Buffer
 
 	sep := ""
-	for i, value := range pks {
+	for i, value := range ids {
 		if value == nil {
-			return "", errors.Errorf("The %ds PK value is nil", i)
+			return "", errors.Errorf("The %ds id or PK value is nil", i)
 		}
 
 		buf.WriteString(fmt.Sprintf("%s%v", sep, value))