Edit on GitHub

sqlglot.dialects.prql

  1from __future__ import annotations
  2
  3import typing as t
  4
  5from sqlglot import exp, generator, parser, tokens
  6from sqlglot.dialects.dialect import Dialect
  7from sqlglot.tokens import TokenType
  8
  9if t.TYPE_CHECKING:
 10    pass
 11
 12
 13class PRQL(Dialect):
 14    class Tokenizer(tokens.Tokenizer):
 15        IDENTIFIERS = ["`"]
 16        QUOTES = ["'", '"']
 17
 18        SINGLE_TOKENS = {
 19            **tokens.Tokenizer.SINGLE_TOKENS,
 20            "=": TokenType.ALIAS,
 21            "'": TokenType.QUOTE,
 22            '"': TokenType.QUOTE,
 23            "`": TokenType.IDENTIFIER,
 24            "#": TokenType.COMMENT,
 25        }
 26
 27        KEYWORDS = {
 28            **tokens.Tokenizer.KEYWORDS,
 29        }
 30
 31    class Parser(parser.Parser):
 32        TRANSFORM_PARSERS = {
 33            "DERIVE": lambda self, query: self._parse_selection(query),
 34            "SELECT": lambda self, query: self._parse_selection(query, append=False),
 35        }
 36
 37        def _parse_statement(self) -> t.Optional[exp.Expression]:
 38            expression = self._parse_expression()
 39            expression = expression if expression else self._parse_query()
 40            return expression
 41
 42        def _parse_query(
 43            self,
 44        ) -> t.Optional[exp.Query]:
 45            from_ = self._parse_from()
 46
 47            if not from_:
 48                return None
 49
 50            query = exp.select("*").from_(from_, copy=False)
 51
 52            while self._match_texts(self.TRANSFORM_PARSERS):
 53                query = self.TRANSFORM_PARSERS[self._prev.text.upper()](self, query)
 54
 55            return query
 56
 57        def _parse_selection(self, query: exp.Query, append: bool = True) -> exp.Query:
 58            if self._match(TokenType.L_BRACE):
 59                selects = self._parse_csv(self._parse_expression)
 60
 61                if not self._match(TokenType.R_BRACE, expression=query):
 62                    self.raise_error("Expecting ]")
 63            else:
 64                expression = self._parse_expression()
 65                selects = [expression] if expression else []
 66
 67            projections = {
 68                select.alias_or_name: select.this if isinstance(select, exp.Alias) else select
 69                for select in query.selects
 70            }
 71
 72            selects = [
 73                select.transform(
 74                    lambda s: (projections[s.name].copy() if s.name in projections else s)
 75                    if isinstance(s, exp.Column)
 76                    else s,
 77                    copy=False,
 78                )
 79                for select in selects
 80            ]
 81
 82            return query.select(*selects, append=append, copy=False)
 83
 84        def _parse_expression(self) -> t.Optional[exp.Expression]:
 85            if self._next and self._next.token_type == TokenType.ALIAS:
 86                alias = self._parse_id_var(True)
 87                self._match(TokenType.ALIAS)
 88                return self.expression(exp.Alias, this=self._parse_conjunction(), alias=alias)
 89            return self._parse_conjunction()
 90
 91        def _parse_table(
 92            self,
 93            schema: bool = False,
 94            joins: bool = False,
 95            alias_tokens: t.Optional[t.Collection[TokenType]] = None,
 96            parse_bracket: bool = False,
 97            is_db_reference: bool = False,
 98        ) -> t.Optional[exp.Expression]:
 99            return self._parse_table_parts()
100
101        def _parse_from(
102            self, joins: bool = False, skip_from_token: bool = False
103        ) -> t.Optional[exp.From]:
104            if not skip_from_token and not self._match(TokenType.FROM):
105                return None
106
107            return self.expression(
108                exp.From, comments=self._prev_comments, this=self._parse_table(joins=joins)
109            )
110
111    class Generator(generator.Generator):
112        pass
class PRQL(sqlglot.dialects.dialect.Dialect):
 14class PRQL(Dialect):
 15    class Tokenizer(tokens.Tokenizer):
 16        IDENTIFIERS = ["`"]
 17        QUOTES = ["'", '"']
 18
 19        SINGLE_TOKENS = {
 20            **tokens.Tokenizer.SINGLE_TOKENS,
 21            "=": TokenType.ALIAS,
 22            "'": TokenType.QUOTE,
 23            '"': TokenType.QUOTE,
 24            "`": TokenType.IDENTIFIER,
 25            "#": TokenType.COMMENT,
 26        }
 27
 28        KEYWORDS = {
 29            **tokens.Tokenizer.KEYWORDS,
 30        }
 31
 32    class Parser(parser.Parser):
 33        TRANSFORM_PARSERS = {
 34            "DERIVE": lambda self, query: self._parse_selection(query),
 35            "SELECT": lambda self, query: self._parse_selection(query, append=False),
 36        }
 37
 38        def _parse_statement(self) -> t.Optional[exp.Expression]:
 39            expression = self._parse_expression()
 40            expression = expression if expression else self._parse_query()
 41            return expression
 42
 43        def _parse_query(
 44            self,
 45        ) -> t.Optional[exp.Query]:
 46            from_ = self._parse_from()
 47
 48            if not from_:
 49                return None
 50
 51            query = exp.select("*").from_(from_, copy=False)
 52
 53            while self._match_texts(self.TRANSFORM_PARSERS):
 54                query = self.TRANSFORM_PARSERS[self._prev.text.upper()](self, query)
 55
 56            return query
 57
 58        def _parse_selection(self, query: exp.Query, append: bool = True) -> exp.Query:
 59            if self._match(TokenType.L_BRACE):
 60                selects = self._parse_csv(self._parse_expression)
 61
 62                if not self._match(TokenType.R_BRACE, expression=query):
 63                    self.raise_error("Expecting ]")
 64            else:
 65                expression = self._parse_expression()
 66                selects = [expression] if expression else []
 67
 68            projections = {
 69                select.alias_or_name: select.this if isinstance(select, exp.Alias) else select
 70                for select in query.selects
 71            }
 72
 73            selects = [
 74                select.transform(
 75                    lambda s: (projections[s.name].copy() if s.name in projections else s)
 76                    if isinstance(s, exp.Column)
 77                    else s,
 78                    copy=False,
 79                )
 80                for select in selects
 81            ]
 82
 83            return query.select(*selects, append=append, copy=False)
 84
 85        def _parse_expression(self) -> t.Optional[exp.Expression]:
 86            if self._next and self._next.token_type == TokenType.ALIAS:
 87                alias = self._parse_id_var(True)
 88                self._match(TokenType.ALIAS)
 89                return self.expression(exp.Alias, this=self._parse_conjunction(), alias=alias)
 90            return self._parse_conjunction()
 91
 92        def _parse_table(
 93            self,
 94            schema: bool = False,
 95            joins: bool = False,
 96            alias_tokens: t.Optional[t.Collection[TokenType]] = None,
 97            parse_bracket: bool = False,
 98            is_db_reference: bool = False,
 99        ) -> t.Optional[exp.Expression]:
100            return self._parse_table_parts()
101
102        def _parse_from(
103            self, joins: bool = False, skip_from_token: bool = False
104        ) -> t.Optional[exp.From]:
105            if not skip_from_token and not self._match(TokenType.FROM):
106                return None
107
108            return self.expression(
109                exp.From, comments=self._prev_comments, this=self._parse_table(joins=joins)
110            )
111
112    class Generator(generator.Generator):
113        pass
tokenizer_class = <class 'PRQL.Tokenizer'>
parser_class = <class 'PRQL.Parser'>
generator_class = <class 'PRQL.Generator'>
TIME_TRIE: Dict = {}
FORMAT_TRIE: Dict = {}
INVERSE_TIME_MAPPING: Dict[str, str] = {}
INVERSE_TIME_TRIE: Dict = {}
INVERSE_ESCAPE_SEQUENCES: Dict[str, str] = {}
QUOTE_START = "'"
QUOTE_END = "'"
IDENTIFIER_START = '`'
IDENTIFIER_END = '`'
BIT_START: Optional[str] = None
BIT_END: Optional[str] = None
HEX_START: Optional[str] = None
HEX_END: Optional[str] = None
BYTE_START: Optional[str] = None
BYTE_END: Optional[str] = None
UNICODE_START: Optional[str] = None
UNICODE_END: Optional[str] = None
class PRQL.Tokenizer(sqlglot.tokens.Tokenizer):
15    class Tokenizer(tokens.Tokenizer):
16        IDENTIFIERS = ["`"]
17        QUOTES = ["'", '"']
18
19        SINGLE_TOKENS = {
20            **tokens.Tokenizer.SINGLE_TOKENS,
21            "=": TokenType.ALIAS,
22            "'": TokenType.QUOTE,
23            '"': TokenType.QUOTE,
24            "`": TokenType.IDENTIFIER,
25            "#": TokenType.COMMENT,
26        }
27
28        KEYWORDS = {
29            **tokens.Tokenizer.KEYWORDS,
30        }
IDENTIFIERS = ['`']
QUOTES = ["'", '"']
SINGLE_TOKENS = {'(': <TokenType.L_PAREN: 'L_PAREN'>, ')': <TokenType.R_PAREN: 'R_PAREN'>, '[': <TokenType.L_BRACKET: 'L_BRACKET'>, ']': <TokenType.R_BRACKET: 'R_BRACKET'>, '{': <TokenType.L_BRACE: 'L_BRACE'>, '}': <TokenType.R_BRACE: 'R_BRACE'>, '&': <TokenType.AMP: 'AMP'>, '^': <TokenType.CARET: 'CARET'>, ':': <TokenType.COLON: 'COLON'>, ',': <TokenType.COMMA: 'COMMA'>, '.': <TokenType.DOT: 'DOT'>, '-': <TokenType.DASH: 'DASH'>, '=': <TokenType.ALIAS: 'ALIAS'>, '>': <TokenType.GT: 'GT'>, '<': <TokenType.LT: 'LT'>, '%': <TokenType.MOD: 'MOD'>, '!': <TokenType.NOT: 'NOT'>, '|': <TokenType.PIPE: 'PIPE'>, '+': <TokenType.PLUS: 'PLUS'>, ';': <TokenType.SEMICOLON: 'SEMICOLON'>, '/': <TokenType.SLASH: 'SLASH'>, '\\': <TokenType.BACKSLASH: 'BACKSLASH'>, '*': <TokenType.STAR: 'STAR'>, '~': <TokenType.TILDA: 'TILDA'>, '?': <TokenType.PLACEHOLDER: 'PLACEHOLDER'>, '@': <TokenType.PARAMETER: 'PARAMETER'>, "'": <TokenType.QUOTE: 'QUOTE'>, '`': <TokenType.IDENTIFIER: 'IDENTIFIER'>, '"': <TokenType.QUOTE: 'QUOTE'>, '#': <TokenType.COMMENT: 'COMMENT'>}
KEYWORDS = {'{%': <TokenType.BLOCK_START: 'BLOCK_START'>, '{%+': <TokenType.BLOCK_START: 'BLOCK_START'>, '{%-': <TokenType.BLOCK_START: 'BLOCK_START'>, '%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '+%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '-%}': <TokenType.BLOCK_END: 'BLOCK_END'>, '{{+': <TokenType.BLOCK_START: 'BLOCK_START'>, '{{-': <TokenType.BLOCK_START: 'BLOCK_START'>, '+}}': <TokenType.BLOCK_END: 'BLOCK_END'>, '-}}': <TokenType.BLOCK_END: 'BLOCK_END'>, '/*+': <TokenType.HINT: 'HINT'>, '==': <TokenType.EQ: 'EQ'>, '::': <TokenType.DCOLON: 'DCOLON'>, '||': <TokenType.DPIPE: 'DPIPE'>, '>=': <TokenType.GTE: 'GTE'>, '<=': <TokenType.LTE: 'LTE'>, '<>': <TokenType.NEQ: 'NEQ'>, '!=': <TokenType.NEQ: 'NEQ'>, ':=': <TokenType.COLON_EQ: 'COLON_EQ'>, '<=>': <TokenType.NULLSAFE_EQ: 'NULLSAFE_EQ'>, '->': <TokenType.ARROW: 'ARROW'>, '->>': <TokenType.DARROW: 'DARROW'>, '=>': <TokenType.FARROW: 'FARROW'>, '#>': <TokenType.HASH_ARROW: 'HASH_ARROW'>, '#>>': <TokenType.DHASH_ARROW: 'DHASH_ARROW'>, '<->': <TokenType.LR_ARROW: 'LR_ARROW'>, '&&': <TokenType.DAMP: 'DAMP'>, '??': <TokenType.DQMARK: 'DQMARK'>, 'ALL': <TokenType.ALL: 'ALL'>, 'ALWAYS': <TokenType.ALWAYS: 'ALWAYS'>, 'AND': <TokenType.AND: 'AND'>, 'ANTI': <TokenType.ANTI: 'ANTI'>, 'ANY': <TokenType.ANY: 'ANY'>, 'ASC': <TokenType.ASC: 'ASC'>, 'AS': <TokenType.ALIAS: 'ALIAS'>, 'ASOF': <TokenType.ASOF: 'ASOF'>, 'AUTOINCREMENT': <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, 'AUTO_INCREMENT': <TokenType.AUTO_INCREMENT: 'AUTO_INCREMENT'>, 'BEGIN': <TokenType.BEGIN: 'BEGIN'>, 'BETWEEN': <TokenType.BETWEEN: 'BETWEEN'>, 'CACHE': <TokenType.CACHE: 'CACHE'>, 'UNCACHE': <TokenType.UNCACHE: 'UNCACHE'>, 'CASE': <TokenType.CASE: 'CASE'>, 'CHARACTER SET': <TokenType.CHARACTER_SET: 'CHARACTER_SET'>, 'CLUSTER BY': <TokenType.CLUSTER_BY: 'CLUSTER_BY'>, 'COLLATE': <TokenType.COLLATE: 'COLLATE'>, 'COLUMN': <TokenType.COLUMN: 'COLUMN'>, 'COMMIT': <TokenType.COMMIT: 'COMMIT'>, 'CONNECT BY': <TokenType.CONNECT_BY: 'CONNECT_BY'>, 'CONSTRAINT': <TokenType.CONSTRAINT: 'CONSTRAINT'>, 'CREATE': <TokenType.CREATE: 'CREATE'>, 'CROSS': <TokenType.CROSS: 'CROSS'>, 'CUBE': <TokenType.CUBE: 'CUBE'>, 'CURRENT_DATE': <TokenType.CURRENT_DATE: 'CURRENT_DATE'>, 'CURRENT_TIME': <TokenType.CURRENT_TIME: 'CURRENT_TIME'>, 'CURRENT_TIMESTAMP': <TokenType.CURRENT_TIMESTAMP: 'CURRENT_TIMESTAMP'>, 'CURRENT_USER': <TokenType.CURRENT_USER: 'CURRENT_USER'>, 'DATABASE': <TokenType.DATABASE: 'DATABASE'>, 'DEFAULT': <TokenType.DEFAULT: 'DEFAULT'>, 'DELETE': <TokenType.DELETE: 'DELETE'>, 'DESC': <TokenType.DESC: 'DESC'>, 'DESCRIBE': <TokenType.DESCRIBE: 'DESCRIBE'>, 'DISTINCT': <TokenType.DISTINCT: 'DISTINCT'>, 'DISTRIBUTE BY': <TokenType.DISTRIBUTE_BY: 'DISTRIBUTE_BY'>, 'DIV': <TokenType.DIV: 'DIV'>, 'DROP': <TokenType.DROP: 'DROP'>, 'ELSE': <TokenType.ELSE: 'ELSE'>, 'END': <TokenType.END: 'END'>, 'ENUM': <TokenType.ENUM: 'ENUM'>, 'ESCAPE': <TokenType.ESCAPE: 'ESCAPE'>, 'EXCEPT': <TokenType.EXCEPT: 'EXCEPT'>, 'EXECUTE': <TokenType.EXECUTE: 'EXECUTE'>, 'EXISTS': <TokenType.EXISTS: 'EXISTS'>, 'FALSE': <TokenType.FALSE: 'FALSE'>, 'FETCH': <TokenType.FETCH: 'FETCH'>, 'FILTER': <TokenType.FILTER: 'FILTER'>, 'FIRST': <TokenType.FIRST: 'FIRST'>, 'FULL': <TokenType.FULL: 'FULL'>, 'FUNCTION': <TokenType.FUNCTION: 'FUNCTION'>, 'FOR': <TokenType.FOR: 'FOR'>, 'FOREIGN KEY': <TokenType.FOREIGN_KEY: 'FOREIGN_KEY'>, 'FORMAT': <TokenType.FORMAT: 'FORMAT'>, 'FROM': <TokenType.FROM: 'FROM'>, 'GEOGRAPHY': <TokenType.GEOGRAPHY: 'GEOGRAPHY'>, 'GEOMETRY': <TokenType.GEOMETRY: 'GEOMETRY'>, 'GLOB': <TokenType.GLOB: 'GLOB'>, 'GROUP BY': <TokenType.GROUP_BY: 'GROUP_BY'>, 'GROUPING SETS': <TokenType.GROUPING_SETS: 'GROUPING_SETS'>, 'HAVING': <TokenType.HAVING: 'HAVING'>, 'ILIKE': <TokenType.ILIKE: 'ILIKE'>, 'IN': <TokenType.IN: 'IN'>, 'INDEX': <TokenType.INDEX: 'INDEX'>, 'INET': <TokenType.INET: 'INET'>, 'INNER': <TokenType.INNER: 'INNER'>, 'INSERT': <TokenType.INSERT: 'INSERT'>, 'INTERVAL': <TokenType.INTERVAL: 'INTERVAL'>, 'INTERSECT': <TokenType.INTERSECT: 'INTERSECT'>, 'INTO': <TokenType.INTO: 'INTO'>, 'IS': <TokenType.IS: 'IS'>, 'ISNULL': <TokenType.ISNULL: 'ISNULL'>, 'JOIN': <TokenType.JOIN: 'JOIN'>, 'KEEP': <TokenType.KEEP: 'KEEP'>, 'KILL': <TokenType.KILL: 'KILL'>, 'LATERAL': <TokenType.LATERAL: 'LATERAL'>, 'LEFT': <TokenType.LEFT: 'LEFT'>, 'LIKE': <TokenType.LIKE: 'LIKE'>, 'LIMIT': <TokenType.LIMIT: 'LIMIT'>, 'LOAD': <TokenType.LOAD: 'LOAD'>, 'LOCK': <TokenType.LOCK: 'LOCK'>, 'MERGE': <TokenType.MERGE: 'MERGE'>, 'NATURAL': <TokenType.NATURAL: 'NATURAL'>, 'NEXT': <TokenType.NEXT: 'NEXT'>, 'NOT': <TokenType.NOT: 'NOT'>, 'NOTNULL': <TokenType.NOTNULL: 'NOTNULL'>, 'NULL': <TokenType.NULL: 'NULL'>, 'OBJECT': <TokenType.OBJECT: 'OBJECT'>, 'OFFSET': <TokenType.OFFSET: 'OFFSET'>, 'ON': <TokenType.ON: 'ON'>, 'OR': <TokenType.OR: 'OR'>, 'XOR': <TokenType.XOR: 'XOR'>, 'ORDER BY': <TokenType.ORDER_BY: 'ORDER_BY'>, 'ORDINALITY': <TokenType.ORDINALITY: 'ORDINALITY'>, 'OUTER': <TokenType.OUTER: 'OUTER'>, 'OVER': <TokenType.OVER: 'OVER'>, 'OVERLAPS': <TokenType.OVERLAPS: 'OVERLAPS'>, 'OVERWRITE': <TokenType.OVERWRITE: 'OVERWRITE'>, 'PARTITION': <TokenType.PARTITION: 'PARTITION'>, 'PARTITION BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PARTITIONED BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PARTITIONED_BY': <TokenType.PARTITION_BY: 'PARTITION_BY'>, 'PERCENT': <TokenType.PERCENT: 'PERCENT'>, 'PIVOT': <TokenType.PIVOT: 'PIVOT'>, 'PRAGMA': <TokenType.PRAGMA: 'PRAGMA'>, 'PRIMARY KEY': <TokenType.PRIMARY_KEY: 'PRIMARY_KEY'>, 'PROCEDURE': <TokenType.PROCEDURE: 'PROCEDURE'>, 'QUALIFY': <TokenType.QUALIFY: 'QUALIFY'>, 'RANGE': <TokenType.RANGE: 'RANGE'>, 'RECURSIVE': <TokenType.RECURSIVE: 'RECURSIVE'>, 'REGEXP': <TokenType.RLIKE: 'RLIKE'>, 'REPLACE': <TokenType.REPLACE: 'REPLACE'>, 'RETURNING': <TokenType.RETURNING: 'RETURNING'>, 'REFERENCES': <TokenType.REFERENCES: 'REFERENCES'>, 'RIGHT': <TokenType.RIGHT: 'RIGHT'>, 'RLIKE': <TokenType.RLIKE: 'RLIKE'>, 'ROLLBACK': <TokenType.ROLLBACK: 'ROLLBACK'>, 'ROLLUP': <TokenType.ROLLUP: 'ROLLUP'>, 'ROW': <TokenType.ROW: 'ROW'>, 'ROWS': <TokenType.ROWS: 'ROWS'>, 'SCHEMA': <TokenType.SCHEMA: 'SCHEMA'>, 'SELECT': <TokenType.SELECT: 'SELECT'>, 'SEMI': <TokenType.SEMI: 'SEMI'>, 'SET': <TokenType.SET: 'SET'>, 'SETTINGS': <TokenType.SETTINGS: 'SETTINGS'>, 'SHOW': <TokenType.SHOW: 'SHOW'>, 'SIMILAR TO': <TokenType.SIMILAR_TO: 'SIMILAR_TO'>, 'SOME': <TokenType.SOME: 'SOME'>, 'SORT BY': <TokenType.SORT_BY: 'SORT_BY'>, 'START WITH': <TokenType.START_WITH: 'START_WITH'>, 'TABLE': <TokenType.TABLE: 'TABLE'>, 'TABLESAMPLE': <TokenType.TABLE_SAMPLE: 'TABLE_SAMPLE'>, 'TEMP': <TokenType.TEMPORARY: 'TEMPORARY'>, 'TEMPORARY': <TokenType.TEMPORARY: 'TEMPORARY'>, 'THEN': <TokenType.THEN: 'THEN'>, 'TRUE': <TokenType.TRUE: 'TRUE'>, 'TRUNCATE': <TokenType.TRUNCATE: 'TRUNCATE'>, 'UNION': <TokenType.UNION: 'UNION'>, 'UNKNOWN': <TokenType.UNKNOWN: 'UNKNOWN'>, 'UNNEST': <TokenType.UNNEST: 'UNNEST'>, 'UNPIVOT': <TokenType.UNPIVOT: 'UNPIVOT'>, 'UPDATE': <TokenType.UPDATE: 'UPDATE'>, 'USE': <TokenType.USE: 'USE'>, 'USING': <TokenType.USING: 'USING'>, 'UUID': <TokenType.UUID: 'UUID'>, 'VALUES': <TokenType.VALUES: 'VALUES'>, 'VIEW': <TokenType.VIEW: 'VIEW'>, 'VOLATILE': <TokenType.VOLATILE: 'VOLATILE'>, 'WHEN': <TokenType.WHEN: 'WHEN'>, 'WHERE': <TokenType.WHERE: 'WHERE'>, 'WINDOW': <TokenType.WINDOW: 'WINDOW'>, 'WITH': <TokenType.WITH: 'WITH'>, 'APPLY': <TokenType.APPLY: 'APPLY'>, 'ARRAY': <TokenType.ARRAY: 'ARRAY'>, 'BIT': <TokenType.BIT: 'BIT'>, 'BOOL': <TokenType.BOOLEAN: 'BOOLEAN'>, 'BOOLEAN': <TokenType.BOOLEAN: 'BOOLEAN'>, 'BYTE': <TokenType.TINYINT: 'TINYINT'>, 'MEDIUMINT': <TokenType.MEDIUMINT: 'MEDIUMINT'>, 'INT1': <TokenType.TINYINT: 'TINYINT'>, 'TINYINT': <TokenType.TINYINT: 'TINYINT'>, 'INT16': <TokenType.SMALLINT: 'SMALLINT'>, 'SHORT': <TokenType.SMALLINT: 'SMALLINT'>, 'SMALLINT': <TokenType.SMALLINT: 'SMALLINT'>, 'INT128': <TokenType.INT128: 'INT128'>, 'HUGEINT': <TokenType.INT128: 'INT128'>, 'INT2': <TokenType.SMALLINT: 'SMALLINT'>, 'INTEGER': <TokenType.INT: 'INT'>, 'INT': <TokenType.INT: 'INT'>, 'INT4': <TokenType.INT: 'INT'>, 'INT32': <TokenType.INT: 'INT'>, 'INT64': <TokenType.BIGINT: 'BIGINT'>, 'LONG': <TokenType.BIGINT: 'BIGINT'>, 'BIGINT': <TokenType.BIGINT: 'BIGINT'>, 'INT8': <TokenType.TINYINT: 'TINYINT'>, 'DEC': <TokenType.DECIMAL: 'DECIMAL'>, 'DECIMAL': <TokenType.DECIMAL: 'DECIMAL'>, 'BIGDECIMAL': <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, 'BIGNUMERIC': <TokenType.BIGDECIMAL: 'BIGDECIMAL'>, 'MAP': <TokenType.MAP: 'MAP'>, 'NULLABLE': <TokenType.NULLABLE: 'NULLABLE'>, 'NUMBER': <TokenType.DECIMAL: 'DECIMAL'>, 'NUMERIC': <TokenType.DECIMAL: 'DECIMAL'>, 'FIXED': <TokenType.DECIMAL: 'DECIMAL'>, 'REAL': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT4': <TokenType.FLOAT: 'FLOAT'>, 'FLOAT8': <TokenType.DOUBLE: 'DOUBLE'>, 'DOUBLE': <TokenType.DOUBLE: 'DOUBLE'>, 'DOUBLE PRECISION': <TokenType.DOUBLE: 'DOUBLE'>, 'JSON': <TokenType.JSON: 'JSON'>, 'CHAR': <TokenType.CHAR: 'CHAR'>, 'CHARACTER': <TokenType.CHAR: 'CHAR'>, 'NCHAR': <TokenType.NCHAR: 'NCHAR'>, 'VARCHAR': <TokenType.VARCHAR: 'VARCHAR'>, 'VARCHAR2': <TokenType.VARCHAR: 'VARCHAR'>, 'NVARCHAR': <TokenType.NVARCHAR: 'NVARCHAR'>, 'NVARCHAR2': <TokenType.NVARCHAR: 'NVARCHAR'>, 'BPCHAR': <TokenType.BPCHAR: 'BPCHAR'>, 'STR': <TokenType.TEXT: 'TEXT'>, 'STRING': <TokenType.TEXT: 'TEXT'>, 'TEXT': <TokenType.TEXT: 'TEXT'>, 'LONGTEXT': <TokenType.LONGTEXT: 'LONGTEXT'>, 'MEDIUMTEXT': <TokenType.MEDIUMTEXT: 'MEDIUMTEXT'>, 'TINYTEXT': <TokenType.TINYTEXT: 'TINYTEXT'>, 'CLOB': <TokenType.TEXT: 'TEXT'>, 'LONGVARCHAR': <TokenType.TEXT: 'TEXT'>, 'BINARY': <TokenType.BINARY: 'BINARY'>, 'BLOB': <TokenType.VARBINARY: 'VARBINARY'>, 'LONGBLOB': <TokenType.LONGBLOB: 'LONGBLOB'>, 'MEDIUMBLOB': <TokenType.MEDIUMBLOB: 'MEDIUMBLOB'>, 'TINYBLOB': <TokenType.TINYBLOB: 'TINYBLOB'>, 'BYTEA': <TokenType.VARBINARY: 'VARBINARY'>, 'VARBINARY': <TokenType.VARBINARY: 'VARBINARY'>, 'TIME': <TokenType.TIME: 'TIME'>, 'TIMETZ': <TokenType.TIMETZ: 'TIMETZ'>, 'TIMESTAMP': <TokenType.TIMESTAMP: 'TIMESTAMP'>, 'TIMESTAMPTZ': <TokenType.TIMESTAMPTZ: 'TIMESTAMPTZ'>, 'TIMESTAMPLTZ': <TokenType.TIMESTAMPLTZ: 'TIMESTAMPLTZ'>, 'DATE': <TokenType.DATE: 'DATE'>, 'DATETIME': <TokenType.DATETIME: 'DATETIME'>, 'INT4RANGE': <TokenType.INT4RANGE: 'INT4RANGE'>, 'INT4MULTIRANGE': <TokenType.INT4MULTIRANGE: 'INT4MULTIRANGE'>, 'INT8RANGE': <TokenType.INT8RANGE: 'INT8RANGE'>, 'INT8MULTIRANGE': <TokenType.INT8MULTIRANGE: 'INT8MULTIRANGE'>, 'NUMRANGE': <TokenType.NUMRANGE: 'NUMRANGE'>, 'NUMMULTIRANGE': <TokenType.NUMMULTIRANGE: 'NUMMULTIRANGE'>, 'TSRANGE': <TokenType.TSRANGE: 'TSRANGE'>, 'TSMULTIRANGE': <TokenType.TSMULTIRANGE: 'TSMULTIRANGE'>, 'TSTZRANGE': <TokenType.TSTZRANGE: 'TSTZRANGE'>, 'TSTZMULTIRANGE': <TokenType.TSTZMULTIRANGE: 'TSTZMULTIRANGE'>, 'DATERANGE': <TokenType.DATERANGE: 'DATERANGE'>, 'DATEMULTIRANGE': <TokenType.DATEMULTIRANGE: 'DATEMULTIRANGE'>, 'UNIQUE': <TokenType.UNIQUE: 'UNIQUE'>, 'STRUCT': <TokenType.STRUCT: 'STRUCT'>, 'SEQUENCE': <TokenType.SEQUENCE: 'SEQUENCE'>, 'VARIANT': <TokenType.VARIANT: 'VARIANT'>, 'ALTER': <TokenType.ALTER: 'ALTER'>, 'ANALYZE': <TokenType.COMMAND: 'COMMAND'>, 'CALL': <TokenType.COMMAND: 'COMMAND'>, 'COMMENT': <TokenType.COMMENT: 'COMMENT'>, 'COPY': <TokenType.COMMAND: 'COMMAND'>, 'EXPLAIN': <TokenType.COMMAND: 'COMMAND'>, 'GRANT': <TokenType.COMMAND: 'COMMAND'>, 'OPTIMIZE': <TokenType.COMMAND: 'COMMAND'>, 'PREPARE': <TokenType.COMMAND: 'COMMAND'>, 'VACUUM': <TokenType.COMMAND: 'COMMAND'>, 'USER-DEFINED': <TokenType.USERDEFINED: 'USERDEFINED'>, 'FOR VERSION': <TokenType.VERSION_SNAPSHOT: 'VERSION_SNAPSHOT'>, 'FOR TIMESTAMP': <TokenType.TIMESTAMP_SNAPSHOT: 'TIMESTAMP_SNAPSHOT'>}
class PRQL.Parser(sqlglot.parser.Parser):
 32    class Parser(parser.Parser):
 33        TRANSFORM_PARSERS = {
 34            "DERIVE": lambda self, query: self._parse_selection(query),
 35            "SELECT": lambda self, query: self._parse_selection(query, append=False),
 36        }
 37
 38        def _parse_statement(self) -> t.Optional[exp.Expression]:
 39            expression = self._parse_expression()
 40            expression = expression if expression else self._parse_query()
 41            return expression
 42
 43        def _parse_query(
 44            self,
 45        ) -> t.Optional[exp.Query]:
 46            from_ = self._parse_from()
 47
 48            if not from_:
 49                return None
 50
 51            query = exp.select("*").from_(from_, copy=False)
 52
 53            while self._match_texts(self.TRANSFORM_PARSERS):
 54                query = self.TRANSFORM_PARSERS[self._prev.text.upper()](self, query)
 55
 56            return query
 57
 58        def _parse_selection(self, query: exp.Query, append: bool = True) -> exp.Query:
 59            if self._match(TokenType.L_BRACE):
 60                selects = self._parse_csv(self._parse_expression)
 61
 62                if not self._match(TokenType.R_BRACE, expression=query):
 63                    self.raise_error("Expecting ]")
 64            else:
 65                expression = self._parse_expression()
 66                selects = [expression] if expression else []
 67
 68            projections = {
 69                select.alias_or_name: select.this if isinstance(select, exp.Alias) else select
 70                for select in query.selects
 71            }
 72
 73            selects = [
 74                select.transform(
 75                    lambda s: (projections[s.name].copy() if s.name in projections else s)
 76                    if isinstance(s, exp.Column)
 77                    else s,
 78                    copy=False,
 79                )
 80                for select in selects
 81            ]
 82
 83            return query.select(*selects, append=append, copy=False)
 84
 85        def _parse_expression(self) -> t.Optional[exp.Expression]:
 86            if self._next and self._next.token_type == TokenType.ALIAS:
 87                alias = self._parse_id_var(True)
 88                self._match(TokenType.ALIAS)
 89                return self.expression(exp.Alias, this=self._parse_conjunction(), alias=alias)
 90            return self._parse_conjunction()
 91
 92        def _parse_table(
 93            self,
 94            schema: bool = False,
 95            joins: bool = False,
 96            alias_tokens: t.Optional[t.Collection[TokenType]] = None,
 97            parse_bracket: bool = False,
 98            is_db_reference: bool = False,
 99        ) -> t.Optional[exp.Expression]:
100            return self._parse_table_parts()
101
102        def _parse_from(
103            self, joins: bool = False, skip_from_token: bool = False
104        ) -> t.Optional[exp.From]:
105            if not skip_from_token and not self._match(TokenType.FROM):
106                return None
107
108            return self.expression(
109                exp.From, comments=self._prev_comments, this=self._parse_table(joins=joins)
110            )

Parser consumes a list of tokens produced by the Tokenizer and produces a parsed syntax tree.

Arguments:
  • error_level: The desired error level. Default: ErrorLevel.IMMEDIATE
  • error_message_context: The amount of context to capture from a query string when displaying the error message (in number of characters). Default: 100
  • max_errors: Maximum number of error messages to include in a raised ParseError. This is only relevant if error_level is ErrorLevel.RAISE. Default: 3
TRANSFORM_PARSERS = {'DERIVE': <function PRQL.Parser.<lambda>>, 'SELECT': <function PRQL.Parser.<lambda>>}
SHOW_TRIE: Dict = {}
SET_TRIE: Dict = {'GLOBAL': {0: True}, 'LOCAL': {0: True}, 'SESSION': {0: True}, 'TRANSACTION': {0: True}}
Inherited Members
sqlglot.parser.Parser
Parser
FUNCTIONS
NO_PAREN_FUNCTIONS
STRUCT_TYPE_TOKENS
NESTED_TYPE_TOKENS
ENUM_TYPE_TOKENS
AGGREGATE_TYPE_TOKENS
TYPE_TOKENS
SIGNED_TO_UNSIGNED_TYPE_TOKEN
SUBQUERY_PREDICATES
RESERVED_TOKENS
DB_CREATABLES
CREATABLES
ID_VAR_TOKENS
INTERVAL_VARS
TABLE_ALIAS_TOKENS
COMMENT_TABLE_ALIAS_TOKENS
UPDATE_ALIAS_TOKENS
TRIM_TYPES
FUNC_TOKENS
CONJUNCTION
EQUALITY
COMPARISON
BITWISE
TERM
FACTOR
EXPONENT
TIMES
TIMESTAMPS
SET_OPERATIONS
JOIN_METHODS
JOIN_SIDES
JOIN_KINDS
JOIN_HINTS
LAMBDAS
COLUMN_OPERATORS
EXPRESSION_PARSERS
STATEMENT_PARSERS
UNARY_PARSERS
STRING_PARSERS
NUMERIC_PARSERS
PRIMARY_PARSERS
PLACEHOLDER_PARSERS
RANGE_PARSERS
PROPERTY_PARSERS
CONSTRAINT_PARSERS
ALTER_PARSERS
SCHEMA_UNNAMED_CONSTRAINTS
NO_PAREN_FUNCTION_PARSERS
INVALID_FUNC_NAME_TOKENS
FUNCTIONS_WITH_ALIASED_ARGS
KEY_VALUE_DEFINITIONS
FUNCTION_PARSERS
QUERY_MODIFIER_PARSERS
SET_PARSERS
SHOW_PARSERS
TYPE_LITERAL_PARSERS
DDL_SELECT_TOKENS
PRE_VOLATILE_TOKENS
TRANSACTION_KIND
TRANSACTION_CHARACTERISTICS
CONFLICT_ACTIONS
CREATE_SEQUENCE
USABLES
INSERT_ALTERNATIVES
CLONE_KEYWORDS
HISTORICAL_DATA_KIND
OPCLASS_FOLLOW_KEYWORDS
OPTYPE_FOLLOW_TOKENS
TABLE_INDEX_HINT_TOKENS
WINDOW_ALIAS_TOKENS
WINDOW_BEFORE_PAREN_TOKENS
WINDOW_SIDES
JSON_KEY_VALUE_SEPARATOR_TOKENS
FETCH_TOKENS
ADD_CONSTRAINT_TOKENS
DISTINCT_TOKENS
NULL_TOKENS
UNNEST_OFFSET_ALIAS_TOKENS
STRICT_CAST
PREFIXED_PIVOT_COLUMNS
IDENTIFY_PIVOT_STRINGS
LOG_DEFAULTS_TO_LN
ALTER_TABLE_ADD_REQUIRED_FOR_EACH_COLUMN
TABLESAMPLE_CSV
SET_REQUIRES_ASSIGNMENT_DELIMITER
TRIM_PATTERN_FIRST
STRING_ALIASES
MODIFIERS_ATTACHED_TO_UNION
UNION_MODIFIERS
NO_PAREN_IF_COMMANDS
JSON_ARROWS_REQUIRE_JSON_TYPE
VALUES_FOLLOWED_BY_PAREN
SUPPORTS_IMPLICIT_UNNEST
error_level
error_message_context
max_errors
dialect
reset
parse
parse_into
check_errors
raise_error
expression
validate_expression
errors
sql
class PRQL.Generator(sqlglot.generator.Generator):
112    class Generator(generator.Generator):
113        pass

Generator converts a given syntax tree to the corresponding SQL string.

Arguments:
  • pretty: Whether to format the produced SQL string. Default: False.
  • identify: Determines when an identifier should be quoted. Possible values are: False (default): Never quote, except in cases where it's mandatory by the dialect. True or 'always': Always quote. 'safe': Only quote identifiers that are case insensitive.
  • normalize: Whether to normalize identifiers to lowercase. Default: False.
  • pad: The pad size in a formatted string. Default: 2.
  • indent: The indentation size in a formatted string. Default: 2.
  • normalize_functions: How to normalize function names. Possible values are: "upper" or True (default): Convert names to uppercase. "lower": Convert names to lowercase. False: Disables function name normalization.
  • unsupported_level: Determines the generator's behavior when it encounters unsupported expressions. Default ErrorLevel.WARN.
  • max_unsupported: Maximum number of unsupported messages to include in a raised UnsupportedError. This is only relevant if unsupported_level is ErrorLevel.RAISE. Default: 3
  • leading_comma: Whether the comma is leading or trailing in select expressions. This is only relevant when generating in pretty mode. Default: False
  • max_text_width: The max number of characters in a segment before creating new lines in pretty mode. The default is on the smaller end because the length only represents a segment and not the true line length. Default: 80
  • comments: Whether to preserve comments in the output SQL code. Default: True
SELECT_KINDS: Tuple[str, ...] = ()
AFTER_HAVING_MODIFIER_TRANSFORMS = {'qualify': <function Generator.<lambda>>, 'windows': <function Generator.<lambda>>}
Inherited Members
sqlglot.generator.Generator
Generator
TRANSFORMS
NULL_ORDERING_SUPPORTED
IGNORE_NULLS_IN_FUNC
LOCKING_READS_SUPPORTED
EXPLICIT_UNION
WRAP_DERIVED_VALUES
CREATE_FUNCTION_RETURN_AS
MATCHED_BY_SOURCE
SINGLE_STRING_INTERVAL
INTERVAL_ALLOWS_PLURAL_FORM
LIMIT_FETCH
LIMIT_ONLY_LITERALS
RENAME_TABLE_WITH_DB
GROUPINGS_SEP
INDEX_ON
JOIN_HINTS
TABLE_HINTS
QUERY_HINTS
QUERY_HINT_SEP
IS_BOOL_ALLOWED
DUPLICATE_KEY_UPDATE_WITH_SET
LIMIT_IS_TOP
RETURNING_END
COLUMN_JOIN_MARKS_SUPPORTED
EXTRACT_ALLOWS_QUOTES
TZ_TO_WITH_TIME_ZONE
NVL2_SUPPORTED
VALUES_AS_TABLE
ALTER_TABLE_INCLUDE_COLUMN_KEYWORD
UNNEST_WITH_ORDINALITY
AGGREGATE_FILTER_SUPPORTED
SEMI_ANTI_JOIN_WITH_SIDE
COMPUTED_COLUMN_WITH_TYPE
SUPPORTS_TABLE_COPY
TABLESAMPLE_REQUIRES_PARENS
TABLESAMPLE_SIZE_IS_ROWS
TABLESAMPLE_KEYWORDS
TABLESAMPLE_WITH_METHOD
TABLESAMPLE_SEED_KEYWORD
COLLATE_IS_FUNC
DATA_TYPE_SPECIFIERS_ALLOWED
ENSURE_BOOLS
CTE_RECURSIVE_KEYWORD_REQUIRED
SUPPORTS_SINGLE_ARG_CONCAT
LAST_DAY_SUPPORTS_DATE_PART
SUPPORTS_TABLE_ALIAS_COLUMNS
UNPIVOT_ALIASES_ARE_IDENTIFIERS
JSON_KEY_VALUE_PAIR_SEP
INSERT_OVERWRITE
SUPPORTS_SELECT_INTO
SUPPORTS_UNLOGGED_TABLES
SUPPORTS_CREATE_TABLE_LIKE
LIKE_PROPERTY_INSIDE_SCHEMA
MULTI_ARG_DISTINCT
JSON_TYPE_REQUIRED_FOR_EXTRACTION
JSON_PATH_BRACKETED_KEY_SUPPORTED
JSON_PATH_SINGLE_QUOTE_ESCAPE
SUPPORTED_JSON_PATH_PARTS
CAN_IMPLEMENT_ARRAY_ANY
TYPE_MAPPING
STAR_MAPPING
TIME_PART_SINGULARS
TOKEN_MAPPING
STRUCT_DELIMITER
PARAMETER_TOKEN
NAMED_PLACEHOLDER_TOKEN
PROPERTIES_LOCATION
RESERVED_KEYWORDS
WITH_SEPARATED_COMMENTS
EXCLUDE_COMMENTS
UNWRAPPED_INTERVAL_VALUES
PARAMETERIZABLE_TEXT_TYPES
EXPRESSIONS_WITHOUT_NESTED_CTES
SENTINEL_LINE_BREAK
pretty
identify
normalize
pad
unsupported_level
max_unsupported
leading_comma
max_text_width
comments
dialect
normalize_functions
unsupported_messages
generate
preprocess
unsupported
sep
seg
pad_comment
maybe_comment
wrap
no_identify
normalize_func
indent
sql
uncache_sql
cache_sql
characterset_sql
column_sql
columnposition_sql
columndef_sql
columnconstraint_sql
computedcolumnconstraint_sql
autoincrementcolumnconstraint_sql
compresscolumnconstraint_sql
generatedasidentitycolumnconstraint_sql
generatedasrowcolumnconstraint_sql
periodforsystemtimeconstraint_sql
notnullcolumnconstraint_sql
transformcolumnconstraint_sql
primarykeycolumnconstraint_sql
uniquecolumnconstraint_sql
createable_sql
create_sql
sequenceproperties_sql
clone_sql
describe_sql
heredoc_sql
prepend_ctes
with_sql
cte_sql
tablealias_sql
bitstring_sql
hexstring_sql
bytestring_sql
unicodestring_sql
rawstring_sql
datatypeparam_sql
datatype_sql
directory_sql
delete_sql
drop_sql
except_sql
except_op
fetch_sql
filter_sql
hint_sql
index_sql
identifier_sql
inputoutputformat_sql
national_sql
partition_sql
properties_sql
root_properties
properties
with_properties
locate_properties
property_name
property_sql
likeproperty_sql
fallbackproperty_sql
journalproperty_sql
freespaceproperty_sql
checksumproperty_sql
mergeblockratioproperty_sql
datablocksizeproperty_sql
blockcompressionproperty_sql
isolatedloadingproperty_sql
partitionboundspec_sql
partitionedofproperty_sql
lockingproperty_sql
withdataproperty_sql
withsystemversioningproperty_sql
insert_sql
intersect_sql
intersect_op
introducer_sql
kill_sql
pseudotype_sql
objectidentifier_sql
onconflict_sql
returning_sql
rowformatdelimitedproperty_sql
withtablehint_sql
indextablehint_sql
historicaldata_sql
table_parts
table_sql
tablesample_sql
pivot_sql
version_sql
tuple_sql
update_sql
values_sql
var_sql
into_sql
from_sql
group_sql
having_sql
connect_sql
prior_sql
join_sql
lambda_sql
lateral_op
lateral_sql
limit_sql
offset_sql
setitem_sql
set_sql
pragma_sql
lock_sql
literal_sql
escape_str
loaddata_sql
null_sql
boolean_sql
order_sql
withfill_sql
cluster_sql
distribute_sql
sort_sql
ordered_sql
matchrecognize_sql
query_modifiers
queryoption_sql
offset_limit_modifiers
after_limit_modifiers
select_sql
schema_sql
schema_columns_sql
star_sql
parameter_sql
sessionparameter_sql
placeholder_sql
subquery_sql
qualify_sql
set_operations
union_sql
union_op
unnest_sql
prewhere_sql
where_sql
window_sql
partition_by_sql
windowspec_sql
withingroup_sql
between_sql
bracket_sql
all_sql
any_sql
exists_sql
case_sql
constraint_sql
nextvaluefor_sql
extract_sql
trim_sql
convert_concat_args
concat_sql
concatws_sql
check_sql
foreignkey_sql
primarykey_sql
if_sql
matchagainst_sql
jsonkeyvalue_sql
jsonpath_sql
json_path_part
formatjson_sql
jsonobject_sql
jsonobjectagg_sql
jsonarray_sql
jsonarrayagg_sql
jsoncolumndef_sql
jsonschema_sql
jsontable_sql
openjsoncolumndef_sql
openjson_sql
in_sql
in_unnest_op
interval_sql
return_sql
reference_sql
anonymous_sql
paren_sql
neg_sql
not_sql
alias_sql
pivotalias_sql
aliases_sql
atindex_sql
attimezone_sql
fromtimezone_sql
add_sql
and_sql
xor_sql
connector_sql
bitwiseand_sql
bitwiseleftshift_sql
bitwisenot_sql
bitwiseor_sql
bitwiserightshift_sql
bitwisexor_sql
cast_sql
currentdate_sql
currenttimestamp_sql
collate_sql
command_sql
comment_sql
mergetreettlaction_sql
mergetreettl_sql
transaction_sql
commit_sql
rollback_sql
altercolumn_sql
renametable_sql
renamecolumn_sql
altertable_sql
add_column_sql
droppartition_sql
addconstraint_sql
distinct_sql
ignorenulls_sql
respectnulls_sql
havingmax_sql
intdiv_sql
dpipe_sql
div_sql
overlaps_sql
distance_sql
dot_sql
eq_sql
propertyeq_sql
escape_sql
glob_sql
gt_sql
gte_sql
ilike_sql
ilikeany_sql
is_sql
like_sql
likeany_sql
similarto_sql
lt_sql
lte_sql
mod_sql
mul_sql
neq_sql
nullsafeeq_sql
nullsafeneq_sql
or_sql
slice_sql
sub_sql
trycast_sql
log_sql
use_sql
binary
function_fallback_sql
func
format_args
text_width
format_time
expressions
op_expressions
naked_property
tag_sql
token_sql
userdefinedfunction_sql
joinhint_sql
kwarg_sql
when_sql
merge_sql
tochar_sql
dictproperty_sql
dictrange_sql
dictsubproperty_sql
oncluster_sql
clusteredbyproperty_sql
anyvalue_sql
querytransform_sql
indexconstraintoption_sql
checkcolumnconstraint_sql
indexcolumnconstraint_sql
nvl2_sql
comprehension_sql
columnprefix_sql
opclass_sql
predict_sql
forin_sql
refresh_sql
operator_sql
toarray_sql
tsordstotime_sql
tsordstodate_sql
unixdate_sql
lastday_sql
arrayany_sql
generateseries_sql
struct_sql
partitionrange_sql
truncatetable_sql
convert_sql