Add json(b)_to_tsvector function

Jsonb has a complex nature so there isn't best-for-everything way to convert it
to tsvector for full text search. Current to_tsvector(json(b)) suggests to
convert only string values, but it's possible to index keys, numerics and even
booleans value. To solve that json(b)_to_tsvector has a second required
argument contained a list of desired types of json fields. Second argument is
a jsonb scalar or array right now with possibility to add new options in a
future.

Bump catalog version

Author: Dmitry Dolgov with some editorization by me
Reviewed by: Teodor Sigaev
Discussion: https://www.postgresql.org/message-id/CA+q6zcXJQbS1b4kJ_HeAOoOc=unfnOrUEL=KGgE32QKDww7d8g@mail.gmail.com

Author: feodor

doc/src/sgml/func.sgml

@@ -9727,6 +9727,26 @@ CREATE TYPE rainbow AS ENUM ('red', 'orange', 'yellow', 'green', 'blue', 'purple
         <entry><literal>to_tsvector('english', '{"a": "The Fat Rats"}'::json)</literal></entry>
         <entry><literal>'fat':2 'rat':3</literal></entry>
        </row>
+       <row>
+        <entry>
+         <literal><function>json(b)_to_tsvector(<optional> <replaceable class="parameter">config</replaceable> <type>regconfig</type>,
+         </optional> <replaceable class="parameter">document</replaceable> <type>json(b)</type>,
+         <replaceable class="parameter">filter</replaceable> <type>json(b)</type>)</function></literal>
+        </entry>
+        <entry><type>tsvector</type></entry>
+        <entry>
+          reduce each value in the document, specified by <replaceable class="parameter">filter</replaceable> to a <type>tsvector</type>,
+          and then concatenate those in document order to produce a single <type>tsvector</type>.
+          <replaceable class="parameter">filter</replaceable> is a jsonb array, that enumerates what kind of elements need to be included
+          into the resulting <type>tsvector</type>. Possible values for <replaceable class="parameter">filter</replaceable> are
+          <literal>"string"</literal> (to include all string values), <literal>"numeric"</literal> (to include all numeric values in the string format),
+          <literal>"boolean"</literal> (to include all boolean values in the string format "true"/"false"),
+          <literal>"key"</literal> (to include all keys) or <literal>"all"</literal> (to include all above). These values
+          can be combined together to include, e.g. all string and numeric values.
+        </entry>
+        <entry><literal>json_to_tsvector('english', '{"a": "The Fat Rats", "b": 123}'::json, '["string", "numeric"]')</literal></entry>
+        <entry><literal>'123':5 'fat':2 'rat':3</literal></entry>
+       </row>
        <row>
         <entry>
          <indexterm>

src/backend/tsearch/to_tsany.c

@@ -267,12 +267,12 @@ to_tsvector(PG_FUNCTION_ARGS)
 										PointerGetDatum(in)));
 }
 
-Datum
-jsonb_to_tsvector_byid(PG_FUNCTION_ARGS)
+/*
+ * Worker function for jsonb(_string)_to_tsvector(_byid)
+ */
+static TSVector
+jsonb_to_tsvector_worker(Oid cfgId, Jsonb *jb, uint32 flags)
 {
-	Oid			cfgId = PG_GETARG_OID(0);
-	Jsonb	   *jb = PG_GETARG_JSONB_P(1);
-	TSVector	result;
 	TSVectorBuildState state;
 	ParsedText	prs;
 
@@ -281,33 +281,77 @@ jsonb_to_tsvector_byid(PG_FUNCTION_ARGS)
 	state.prs = &prs;
 	state.cfgId = cfgId;
 
-	iterate_jsonb_string_values(jb, &state, add_to_tsvector);
+	iterate_jsonb_values(jb, flags, &state, add_to_tsvector);
 
-	PG_FREE_IF_COPY(jb, 1);
+	return make_tsvector(&prs);
+}
+
+Datum
+jsonb_string_to_tsvector_byid(PG_FUNCTION_ARGS)
+{
+	Oid			cfgId = PG_GETARG_OID(0);
+	Jsonb	   *jb = PG_GETARG_JSONB_P(1);
+	TSVector	result;
 
-	result = make_tsvector(&prs);
+	result = jsonb_to_tsvector_worker(cfgId, jb, jtiString);
+	PG_FREE_IF_COPY(jb, 1);
 
 	PG_RETURN_TSVECTOR(result);
 }
 
 Datum
-jsonb_to_tsvector(PG_FUNCTION_ARGS)
+jsonb_string_to_tsvector(PG_FUNCTION_ARGS)
 {
 	Jsonb	   *jb = PG_GETARG_JSONB_P(0);
 	Oid			cfgId;
+	TSVector	result;
 
 	cfgId = getTSCurrentConfig(true);
-	PG_RETURN_DATUM(DirectFunctionCall2(jsonb_to_tsvector_byid,
-										ObjectIdGetDatum(cfgId),
-										JsonbPGetDatum(jb)));
+	result = jsonb_to_tsvector_worker(cfgId, jb, jtiString);
+	PG_FREE_IF_COPY(jb, 0);
+
+	PG_RETURN_TSVECTOR(result);
 }
 
 Datum
-json_to_tsvector_byid(PG_FUNCTION_ARGS)
+jsonb_to_tsvector_byid(PG_FUNCTION_ARGS)
 {
 	Oid			cfgId = PG_GETARG_OID(0);
-	text	   *json = PG_GETARG_TEXT_P(1);
+	Jsonb	   *jb = PG_GETARG_JSONB_P(1);
+	Jsonb	   *jbFlags = PG_GETARG_JSONB_P(2);
+	TSVector	result;
+	uint32		flags = parse_jsonb_index_flags(jbFlags);
+
+	result = jsonb_to_tsvector_worker(cfgId, jb, flags);
+	PG_FREE_IF_COPY(jb, 1);
+	PG_FREE_IF_COPY(jbFlags, 2);
+
+	PG_RETURN_TSVECTOR(result);
+}
+
+Datum
+jsonb_to_tsvector(PG_FUNCTION_ARGS)
+{
+	Jsonb	   *jb = PG_GETARG_JSONB_P(0);
+	Jsonb	   *jbFlags = PG_GETARG_JSONB_P(1);
+	Oid			cfgId;
 	TSVector	result;
+	uint32		flags = parse_jsonb_index_flags(jbFlags);
+
+	cfgId = getTSCurrentConfig(true);
+	result = jsonb_to_tsvector_worker(cfgId, jb, flags);
+	PG_FREE_IF_COPY(jb, 0);
+	PG_FREE_IF_COPY(jbFlags, 1);
+
+	PG_RETURN_TSVECTOR(result);
+}
+
+/*
+ * Worker function for json(_string)_to_tsvector(_byid)
+ */
+static TSVector
+json_to_tsvector_worker(Oid cfgId, text *json, uint32 flags)
+{
 	TSVectorBuildState state;
 	ParsedText	prs;
 
@@ -316,11 +360,50 @@ json_to_tsvector_byid(PG_FUNCTION_ARGS)
 	state.prs = &prs;
 	state.cfgId = cfgId;
 
-	iterate_json_string_values(json, &state, add_to_tsvector);
+	iterate_json_values(json, flags, &state, add_to_tsvector);
+
+	return make_tsvector(&prs);
+}
+
+Datum
+json_string_to_tsvector_byid(PG_FUNCTION_ARGS)
+{
+	Oid			cfgId = PG_GETARG_OID(0);
+	text	   *json = PG_GETARG_TEXT_P(1);
+	TSVector	result;
 
+	result = json_to_tsvector_worker(cfgId, json, jtiString);
 	PG_FREE_IF_COPY(json, 1);
 
-	result = make_tsvector(&prs);
+	PG_RETURN_TSVECTOR(result);
+}
+
+Datum
+json_string_to_tsvector(PG_FUNCTION_ARGS)
+{
+	text	   *json = PG_GETARG_TEXT_P(0);
+	Oid			cfgId;
+	TSVector	result;
+
+	cfgId = getTSCurrentConfig(true);
+	result = json_to_tsvector_worker(cfgId, json, jtiString);
+	PG_FREE_IF_COPY(json, 0);
+
+	PG_RETURN_TSVECTOR(result);
+}
+
+Datum
+json_to_tsvector_byid(PG_FUNCTION_ARGS)
+{
+	Oid			cfgId = PG_GETARG_OID(0);
+	text	   *json = PG_GETARG_TEXT_P(1);
+	Jsonb	   *jbFlags = PG_GETARG_JSONB_P(2);
+	TSVector	result;
+	uint32		flags = parse_jsonb_index_flags(jbFlags);
+
+	result = json_to_tsvector_worker(cfgId, json, flags);
+	PG_FREE_IF_COPY(json, 1);
+	PG_FREE_IF_COPY(jbFlags, 2);
 
 	PG_RETURN_TSVECTOR(result);
 }
@@ -329,12 +412,17 @@ Datum
 json_to_tsvector(PG_FUNCTION_ARGS)
 {
 	text	   *json = PG_GETARG_TEXT_P(0);
+	Jsonb	   *jbFlags = PG_GETARG_JSONB_P(1);
 	Oid			cfgId;
+	TSVector	result;
+	uint32		flags = parse_jsonb_index_flags(jbFlags);
 
 	cfgId = getTSCurrentConfig(true);
-	PG_RETURN_DATUM(DirectFunctionCall2(json_to_tsvector_byid,
-										ObjectIdGetDatum(cfgId),
-										PointerGetDatum(json)));
+	result = json_to_tsvector_worker(cfgId, json, flags);
+	PG_FREE_IF_COPY(json, 0);
+	PG_FREE_IF_COPY(jbFlags, 1);
+
+	PG_RETURN_TSVECTOR(result);
 }
 
 /*
@@ -353,7 +441,7 @@ add_to_tsvector(void *_state, char *elem_value, int elem_len)
 		 * First time through: initialize words array to a reasonable size.
 		 * (parsetext() will realloc it bigger as needed.)
 		 */
-		prs->lenwords = Max(elem_len / 6, 64);
+		prs->lenwords = 16;
 		prs->words = (ParsedWord *) palloc(sizeof(ParsedWord) * prs->lenwords);
 		prs->curwords = 0;
 		prs->pos = 0;