cw_overlapping_partition_by_regexp.sqlx

config { hasOutput: true }
/*
 * Copyright 2024 Google LLC
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

CREATE OR REPLACE FUNCTION ${self()}(firstRn INT64, haystack STRING, regex STRING)
  RETURNS ARRAY<INT64> 
  OPTIONS(description="""Returns overlapping matches for the regex with the sequence of rows encoded in
 custom format. The expected input haystack format is
 "row-1@row-number-1#row-2@row-number-2# ... #row-n@row-number-n#".
 If the regex matches, then it returns all rows matched at the prefix of the
 haystack,  e.g. if firstRn is 1, firstRn is "A@1#A@2#B@3#A@4#B@5#" and regex
 is "(?:A@\\d+#)+(?:B@\\d+#)" then it will return [1, 2, 3] as output since
 matched subsequence is "A@1#A@2#B@3#". If haystack is "B@3#A@4#B@5#" for the
 same regex it will return an empty array since regex does not match at the
 prefix of the haystack.

 By repeatedly calling cw_overlapping_partition_by_regexp with increasing values
 of firstRn and a substring of sequence starting at firstRn, this UDF
 effectively returns overlapping sets of row numbers that match the given regex.

 Continuing the above example, if we call this UDF with firstRn having
 increasing values, and haystack having substring starting with firstRn, then we
 get corresponding outputs as follows:
 firstRn    haystack                output
 1          "A@1#A@2#B@3#A@4#B@5#"  [1, 2, 3]
 2          "A@2#B@3#A@4#B@5#"      [2, 3]
 3          "B@3#A@4#B@5#"          []
 4          "A@4#B@5#"              [4, 5]
 5          "B@5#"                  []
""")
  AS (
   (WITH t AS (
      SELECT ARRAY_LENGTH(REGEXP_EXTRACT_ALL(REGEXP_EXTRACT(haystack, '^(' || regex || ')'), '@\\d+#')) AS n_rows
    )
    SELECT GENERATE_ARRAY(firstRn, firstRn + n_rows- 1) FROM t)
  );