def get_table_chunk_size()

in core/lib/payload/copy.py [0:0]

• 27 lines of code
• 5 McCabe index (conditional complexity)

    def get_table_chunk_size(self):
        """
        Calculate the number of rows for each table dump query table based on
        average row length and the chunks size we've specified
        """
        result = self.query(
            sql.table_avg_row_len,
            (
                self._current_db,
                self.table_name,
            ),
        )
        if result:
            tbl_avg_length = result[0]["AVG_ROW_LENGTH"]
            # avoid huge chunk row count
            if tbl_avg_length < 20:
                tbl_avg_length = 20
            self.select_chunk_size = self.chunk_size // tbl_avg_length
            # This means either the avg row size is huge, or user specified
            # a tiny select_chunk_size on CLI. Let's make it one row per
            # outfile to avoid zero division
            if not self.select_chunk_size:
                self.select_chunk_size = 1
            log.info(
                "TABLE contains {} rows, table_avg_row_len: {} bytes,"
                "chunk_size: {} bytes, ".format(
                    result[0]["TABLE_ROWS"], tbl_avg_length, self.chunk_size
                )
            )
            log.info("Outfile will contain {} rows each".format(self.select_chunk_size))
            self.eta_chunks = max(
                int(result[0]["TABLE_ROWS"] / self.select_chunk_size), 1
            )
        else:
            raise OSCError("FAIL_TO_GUESS_CHUNK_SIZE")