Skip to content

visitors

BindVisitor

Bases: BooleanExpressionVisitor[BooleanExpression]

Rewrites a boolean expression by replacing unbound references with references to fields in a struct schema.

Parameters:

Name Type Description Default
schema Schema

A schema to use when binding the expression.

 required
case_sensitive bool

Whether to consider case when binding a reference to a field in a schema, defaults to True.

 required

Raises:

Type Description
TypeError

In the case a predicate is already bound.
Source code in pyiceberg/expressions/visitors.py 
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
class BindVisitor(BooleanExpressionVisitor[BooleanExpression]):
    """Rewrites a boolean expression by replacing unbound references with references to fields in a struct schema.

    Args:
      schema (Schema): A schema to use when binding the expression.
      case_sensitive (bool): Whether to consider case when binding a reference to a field in a schema, defaults to True.

    Raises:
        TypeError: In the case a predicate is already bound.
    """

    schema: Schema
    case_sensitive: bool

    def __init__(self, schema: Schema, case_sensitive: bool) -> None:
        self.schema = schema
        self.case_sensitive = case_sensitive

    def visit_true(self) -> BooleanExpression:
        return AlwaysTrue()

    def visit_false(self) -> BooleanExpression:
        return AlwaysFalse()

    def visit_not(self, child_result: BooleanExpression) -> BooleanExpression:
        return Not(child=child_result)

    def visit_and(self, left_result: BooleanExpression, right_result: BooleanExpression) -> BooleanExpression:
        return And(left=left_result, right=right_result)

    def visit_or(self, left_result: BooleanExpression, right_result: BooleanExpression) -> BooleanExpression:
        return Or(left=left_result, right=right_result)

    def visit_unbound_predicate(self, predicate: UnboundPredicate[L]) -> BooleanExpression:
        return predicate.bind(self.schema, case_sensitive=self.case_sensitive)

    def visit_bound_predicate(self, predicate: BoundPredicate[L]) -> BooleanExpression:
        raise TypeError(f"Found already bound predicate: {predicate}")

BooleanExpressionVisitor

Bases: Generic[T], ABC

Source code in pyiceberg/expressions/visitors.py 
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
class BooleanExpressionVisitor(Generic[T], ABC):
    @abstractmethod
    def visit_true(self) -> T:
        """Visit method for an AlwaysTrue boolean expression.

        Note: This visit method has no arguments since AlwaysTrue instances have no context.
        """

    @abstractmethod
    def visit_false(self) -> T:
        """Visit method for an AlwaysFalse boolean expression.

        Note: This visit method has no arguments since AlwaysFalse instances have no context.
        """

    @abstractmethod
    def visit_not(self, child_result: T) -> T:
        """Visit method for a Not boolean expression.

        Args:
            child_result (T): The result of visiting the child of the Not boolean expression.
        """

    @abstractmethod
    def visit_and(self, left_result: T, right_result: T) -> T:
        """Visit method for an And boolean expression.

        Args:
            left_result (T): The result of visiting the left side of the expression.
            right_result (T): The result of visiting the right side of the expression.
        """

    @abstractmethod
    def visit_or(self, left_result: T, right_result: T) -> T:
        """Visit method for an Or boolean expression.

        Args:
            left_result (T): The result of visiting the left side of the expression.
            right_result (T): The result of visiting the right side of the expression.
        """

    @abstractmethod
    def visit_unbound_predicate(self, predicate: UnboundPredicate[L]) -> T:
        """Visit method for an unbound predicate in an expression tree.

        Args:
            predicate (UnboundPredicate[L): An instance of an UnboundPredicate.
        """

    @abstractmethod
    def visit_bound_predicate(self, predicate: BoundPredicate[L]) -> T:
        """Visit method for a bound predicate in an expression tree.

        Args:
            predicate (BoundPredicate[L]): An instance of a BoundPredicate.
        """

visit_and(left_result, right_result) abstractmethod

Visit method for an And boolean expression.

Parameters:

Name Type Description Default
left_result T

The result of visiting the left side of the expression.

 required
right_result T

The result of visiting the right side of the expression.

 required
Source code in pyiceberg/expressions/visitors.py 
105
106
107
108
109
110
111
112
@abstractmethod
def visit_and(self, left_result: T, right_result: T) -> T:
    """Visit method for an And boolean expression.

    Args:
        left_result (T): The result of visiting the left side of the expression.
        right_result (T): The result of visiting the right side of the expression.
    """

visit_bound_predicate(predicate) abstractmethod

Visit method for a bound predicate in an expression tree.

Parameters:

Name Type Description Default
predicate BoundPredicate[L]

An instance of a BoundPredicate.

 required
Source code in pyiceberg/expressions/visitors.py 
131
132
133
134
135
136
137
@abstractmethod
def visit_bound_predicate(self, predicate: BoundPredicate[L]) -> T:
    """Visit method for a bound predicate in an expression tree.

    Args:
        predicate (BoundPredicate[L]): An instance of a BoundPredicate.
    """

visit_false() abstractmethod

Visit method for an AlwaysFalse boolean expression.

Note: This visit method has no arguments since AlwaysFalse instances have no context.

Source code in pyiceberg/expressions/visitors.py 
90
91
92
93
94
95
@abstractmethod
def visit_false(self) -> T:
    """Visit method for an AlwaysFalse boolean expression.

    Note: This visit method has no arguments since AlwaysFalse instances have no context.
    """

visit_not(child_result) abstractmethod

Visit method for a Not boolean expression.

Parameters:

Name Type Description Default
child_result T

The result of visiting the child of the Not boolean expression.

 required
Source code in pyiceberg/expressions/visitors.py 
 97
 98
 99
100
101
102
103
@abstractmethod
def visit_not(self, child_result: T) -> T:
    """Visit method for a Not boolean expression.

    Args:
        child_result (T): The result of visiting the child of the Not boolean expression.
    """

visit_or(left_result, right_result) abstractmethod

Visit method for an Or boolean expression.

Parameters:

Name Type Description Default
left_result T

The result of visiting the left side of the expression.

 required
right_result T

The result of visiting the right side of the expression.

 required
Source code in pyiceberg/expressions/visitors.py 
114
115
116
117
118
119
120
121
@abstractmethod
def visit_or(self, left_result: T, right_result: T) -> T:
    """Visit method for an Or boolean expression.

    Args:
        left_result (T): The result of visiting the left side of the expression.
        right_result (T): The result of visiting the right side of the expression.
    """

visit_true() abstractmethod

Visit method for an AlwaysTrue boolean expression.

Note: This visit method has no arguments since AlwaysTrue instances have no context.

Source code in pyiceberg/expressions/visitors.py 
83
84
85
86
87
88
@abstractmethod
def visit_true(self) -> T:
    """Visit method for an AlwaysTrue boolean expression.

    Note: This visit method has no arguments since AlwaysTrue instances have no context.
    """

visit_unbound_predicate(predicate) abstractmethod

Visit method for an unbound predicate in an expression tree.

Parameters:

Name Type Description Default
predicate UnboundPredicate[L

An instance of an UnboundPredicate.

 required
Source code in pyiceberg/expressions/visitors.py 
123
124
125
126
127
128
129
@abstractmethod
def visit_unbound_predicate(self, predicate: UnboundPredicate[L]) -> T:
    """Visit method for an unbound predicate in an expression tree.

    Args:
        predicate (UnboundPredicate[L): An instance of an UnboundPredicate.
    """

BoundBooleanExpressionVisitor

Bases: BooleanExpressionVisitor[T], ABC

Source code in pyiceberg/expressions/visitors.py 
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
class BoundBooleanExpressionVisitor(BooleanExpressionVisitor[T], ABC):
    @abstractmethod
    def visit_in(self, term: BoundTerm[L], literals: Set[L]) -> T:
        """Visit a bound In predicate."""

    @abstractmethod
    def visit_not_in(self, term: BoundTerm[L], literals: Set[L]) -> T:
        """Visit a bound NotIn predicate."""

    @abstractmethod
    def visit_is_nan(self, term: BoundTerm[L]) -> T:
        """Visit a bound IsNan predicate."""

    @abstractmethod
    def visit_not_nan(self, term: BoundTerm[L]) -> T:
        """Visit a bound NotNan predicate."""

    @abstractmethod
    def visit_is_null(self, term: BoundTerm[L]) -> T:
        """Visit a bound IsNull predicate."""

    @abstractmethod
    def visit_not_null(self, term: BoundTerm[L]) -> T:
        """Visit a bound NotNull predicate."""

    @abstractmethod
    def visit_equal(self, term: BoundTerm[L], literal: Literal[L]) -> T:
        """Visit a bound Equal predicate."""

    @abstractmethod
    def visit_not_equal(self, term: BoundTerm[L], literal: Literal[L]) -> T:
        """Visit a bound NotEqual predicate."""

    @abstractmethod
    def visit_greater_than_or_equal(self, term: BoundTerm[L], literal: Literal[L]) -> T:
        """Visit a bound GreaterThanOrEqual predicate."""

    @abstractmethod
    def visit_greater_than(self, term: BoundTerm[L], literal: Literal[L]) -> T:
        """Visit a bound GreaterThan predicate."""

    @abstractmethod
    def visit_less_than(self, term: BoundTerm[L], literal: Literal[L]) -> T:
        """Visit a bound LessThan predicate."""

    @abstractmethod
    def visit_less_than_or_equal(self, term: BoundTerm[L], literal: Literal[L]) -> T:
        """Visit a bound LessThanOrEqual predicate."""

    @abstractmethod
    def visit_true(self) -> T:
        """Visit a bound True predicate."""

    @abstractmethod
    def visit_false(self) -> T:
        """Visit a bound False predicate."""

    @abstractmethod
    def visit_not(self, child_result: T) -> T:
        """Visit a bound Not predicate."""

    @abstractmethod
    def visit_and(self, left_result: T, right_result: T) -> T:
        """Visit a bound And predicate."""

    @abstractmethod
    def visit_or(self, left_result: T, right_result: T) -> T:
        """Visit a bound Or predicate."""

    @abstractmethod
    def visit_starts_with(self, term: BoundTerm[L], literal: Literal[L]) -> T:
        """Visit bound StartsWith predicate."""

    @abstractmethod
    def visit_not_starts_with(self, term: BoundTerm[L], literal: Literal[L]) -> T:
        """Visit bound NotStartsWith predicate."""

    def visit_unbound_predicate(self, predicate: UnboundPredicate[L]) -> T:
        """Visit an unbound predicate.

        Args:
            predicate (UnboundPredicate[L]): An unbound predicate.
        Raises:
            TypeError: This always raises since an unbound predicate is not expected in a bound boolean expression.
        """
        raise TypeError(f"Not a bound predicate: {predicate}")

    def visit_bound_predicate(self, predicate: BoundPredicate[L]) -> T:
        """Visit a bound predicate.

        Args:
            predicate (BoundPredicate[L]): A bound predicate.
        """
        return visit_bound_predicate(predicate, self)

visit_and(left_result, right_result) abstractmethod

Visit a bound And predicate.

Source code in pyiceberg/expressions/visitors.py 
318
319
320
@abstractmethod
def visit_and(self, left_result: T, right_result: T) -> T:
    """Visit a bound And predicate."""

visit_bound_predicate(predicate)

Visit a bound predicate.

Parameters:

Name Type Description Default
predicate BoundPredicate[L]

A bound predicate.

 required
Source code in pyiceberg/expressions/visitors.py 
344
345
346
347
348
349
350
def visit_bound_predicate(self, predicate: BoundPredicate[L]) -> T:
    """Visit a bound predicate.

    Args:
        predicate (BoundPredicate[L]): A bound predicate.
    """
    return visit_bound_predicate(predicate, self)

visit_equal(term, literal) abstractmethod

Visit a bound Equal predicate.

Source code in pyiceberg/expressions/visitors.py 
282
283
284
@abstractmethod
def visit_equal(self, term: BoundTerm[L], literal: Literal[L]) -> T:
    """Visit a bound Equal predicate."""

visit_false() abstractmethod

Visit a bound False predicate.

Source code in pyiceberg/expressions/visitors.py 
310
311
312
@abstractmethod
def visit_false(self) -> T:
    """Visit a bound False predicate."""

visit_greater_than(term, literal) abstractmethod

Visit a bound GreaterThan predicate.

Source code in pyiceberg/expressions/visitors.py 
294
295
296
@abstractmethod
def visit_greater_than(self, term: BoundTerm[L], literal: Literal[L]) -> T:
    """Visit a bound GreaterThan predicate."""

visit_greater_than_or_equal(term, literal) abstractmethod

Visit a bound GreaterThanOrEqual predicate.

Source code in pyiceberg/expressions/visitors.py 
290
291
292
@abstractmethod
def visit_greater_than_or_equal(self, term: BoundTerm[L], literal: Literal[L]) -> T:
    """Visit a bound GreaterThanOrEqual predicate."""

visit_in(term, literals) abstractmethod

Visit a bound In predicate.

Source code in pyiceberg/expressions/visitors.py 
258
259
260
@abstractmethod
def visit_in(self, term: BoundTerm[L], literals: Set[L]) -> T:
    """Visit a bound In predicate."""

visit_is_nan(term) abstractmethod

Visit a bound IsNan predicate.

Source code in pyiceberg/expressions/visitors.py 
266
267
268
@abstractmethod
def visit_is_nan(self, term: BoundTerm[L]) -> T:
    """Visit a bound IsNan predicate."""

visit_is_null(term) abstractmethod

Visit a bound IsNull predicate.

Source code in pyiceberg/expressions/visitors.py 
274
275
276
@abstractmethod
def visit_is_null(self, term: BoundTerm[L]) -> T:
    """Visit a bound IsNull predicate."""

visit_less_than(term, literal) abstractmethod

Visit a bound LessThan predicate.

Source code in pyiceberg/expressions/visitors.py 
298
299
300
@abstractmethod
def visit_less_than(self, term: BoundTerm[L], literal: Literal[L]) -> T:
    """Visit a bound LessThan predicate."""

visit_less_than_or_equal(term, literal) abstractmethod

Visit a bound LessThanOrEqual predicate.

Source code in pyiceberg/expressions/visitors.py 
302
303
304
@abstractmethod
def visit_less_than_or_equal(self, term: BoundTerm[L], literal: Literal[L]) -> T:
    """Visit a bound LessThanOrEqual predicate."""

visit_not(child_result) abstractmethod

Visit a bound Not predicate.

Source code in pyiceberg/expressions/visitors.py 
314
315
316
@abstractmethod
def visit_not(self, child_result: T) -> T:
    """Visit a bound Not predicate."""

visit_not_equal(term, literal) abstractmethod

Visit a bound NotEqual predicate.

Source code in pyiceberg/expressions/visitors.py 
286
287
288
@abstractmethod
def visit_not_equal(self, term: BoundTerm[L], literal: Literal[L]) -> T:
    """Visit a bound NotEqual predicate."""

visit_not_in(term, literals) abstractmethod

Visit a bound NotIn predicate.

Source code in pyiceberg/expressions/visitors.py 
262
263
264
@abstractmethod
def visit_not_in(self, term: BoundTerm[L], literals: Set[L]) -> T:
    """Visit a bound NotIn predicate."""

visit_not_nan(term) abstractmethod

Visit a bound NotNan predicate.

Source code in pyiceberg/expressions/visitors.py 
270
271
272
@abstractmethod
def visit_not_nan(self, term: BoundTerm[L]) -> T:
    """Visit a bound NotNan predicate."""

visit_not_null(term) abstractmethod

Visit a bound NotNull predicate.

Source code in pyiceberg/expressions/visitors.py 
278
279
280
@abstractmethod
def visit_not_null(self, term: BoundTerm[L]) -> T:
    """Visit a bound NotNull predicate."""

visit_not_starts_with(term, literal) abstractmethod

Visit bound NotStartsWith predicate.

Source code in pyiceberg/expressions/visitors.py 
330
331
332
@abstractmethod
def visit_not_starts_with(self, term: BoundTerm[L], literal: Literal[L]) -> T:
    """Visit bound NotStartsWith predicate."""

visit_or(left_result, right_result) abstractmethod

Visit a bound Or predicate.

Source code in pyiceberg/expressions/visitors.py 
322
323
324
@abstractmethod
def visit_or(self, left_result: T, right_result: T) -> T:
    """Visit a bound Or predicate."""

visit_starts_with(term, literal) abstractmethod

Visit bound StartsWith predicate.

Source code in pyiceberg/expressions/visitors.py 
326
327
328
@abstractmethod
def visit_starts_with(self, term: BoundTerm[L], literal: Literal[L]) -> T:
    """Visit bound StartsWith predicate."""

visit_true() abstractmethod

Visit a bound True predicate.

Source code in pyiceberg/expressions/visitors.py 
306
307
308
@abstractmethod
def visit_true(self) -> T:
    """Visit a bound True predicate."""

visit_unbound_predicate(predicate)

Visit an unbound predicate.

Parameters:

Name Type Description Default
predicate UnboundPredicate[L]

An unbound predicate.

 required

Raises: TypeError: This always raises since an unbound predicate is not expected in a bound boolean expression.

Source code in pyiceberg/expressions/visitors.py 
334
335
336
337
338
339
340
341
342
def visit_unbound_predicate(self, predicate: UnboundPredicate[L]) -> T:
    """Visit an unbound predicate.

    Args:
        predicate (UnboundPredicate[L]): An unbound predicate.
    Raises:
        TypeError: This always raises since an unbound predicate is not expected in a bound boolean expression.
    """
    raise TypeError(f"Not a bound predicate: {predicate}")

ResidualVisitor

Bases: BoundBooleanExpressionVisitor[BooleanExpression], ABC

Finds the residuals for an Expression the partitions in the given PartitionSpec.

A residual expression is made by partially evaluating an expression using partition values. For example, if a table is partitioned by day(utc_timestamp) and is read with a filter expression utc_timestamp > a and utc_timestamp < b, then there are 4 possible residuals expressions for the partition data, d:

  1. If d > day(a) and d < day(b), the residual is always true
  2. If d == day(a) and d != day(b), the residual is utc_timestamp > a
  3. if d == day(b) and d != day(a), the residual is utc_timestamp < b
  4. If d == day(a) == day(b), the residual is utc_timestamp > a and utc_timestamp < b Partition data is passed using StructLike. Residuals are returned by residualFor(StructLike).
Source code in pyiceberg/expressions/visitors.py 
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
class ResidualVisitor(BoundBooleanExpressionVisitor[BooleanExpression], ABC):
    """Finds the residuals for an Expression the partitions in the given PartitionSpec.

    A residual expression is made by partially evaluating an expression using partition values.
    For example, if a table is partitioned by day(utc_timestamp) and is read with a filter expression
    utc_timestamp > a and utc_timestamp < b, then there are 4 possible residuals expressions
    for the partition data, d:


    1. If d > day(a) and d &lt; day(b), the residual is always true
    2. If d == day(a) and d != day(b), the residual is utc_timestamp > a
    3. if d == day(b) and d != day(a), the residual is utc_timestamp < b
    4. If d == day(a) == day(b), the residual is utc_timestamp > a and utc_timestamp < b
    Partition data is passed using StructLike. Residuals are returned by residualFor(StructLike).
    """

    schema: Schema
    spec: PartitionSpec
    case_sensitive: bool
    expr: BooleanExpression

    def __init__(self, schema: Schema, spec: PartitionSpec, case_sensitive: bool, expr: BooleanExpression) -> None:
        self.schema = schema
        self.spec = spec
        self.case_sensitive = case_sensitive
        self.expr = expr

    def eval(self, partition_data: Record) -> BooleanExpression:
        self.struct = partition_data
        return visit(self.expr, visitor=self)

    def visit_true(self) -> BooleanExpression:
        return AlwaysTrue()

    def visit_false(self) -> BooleanExpression:
        return AlwaysFalse()

    def visit_not(self, child_result: BooleanExpression) -> BooleanExpression:
        return Not(child_result)

    def visit_and(self, left_result: BooleanExpression, right_result: BooleanExpression) -> BooleanExpression:
        return And(left_result, right_result)

    def visit_or(self, left_result: BooleanExpression, right_result: BooleanExpression) -> BooleanExpression:
        return Or(left_result, right_result)

    def visit_is_null(self, term: BoundTerm[L]) -> BooleanExpression:
        if term.eval(self.struct) is None:
            return AlwaysTrue()
        else:
            return AlwaysFalse()

    def visit_not_null(self, term: BoundTerm[L]) -> BooleanExpression:
        if term.eval(self.struct) is not None:
            return AlwaysTrue()
        else:
            return AlwaysFalse()

    def visit_is_nan(self, term: BoundTerm[L]) -> BooleanExpression:
        val = term.eval(self.struct)
        if isinstance(val, SupportsFloat) and math.isnan(val):
            return self.visit_true()
        else:
            return self.visit_false()

    def visit_not_nan(self, term: BoundTerm[L]) -> BooleanExpression:
        val = term.eval(self.struct)
        if isinstance(val, SupportsFloat) and not math.isnan(val):
            return self.visit_true()
        else:
            return self.visit_false()

    def visit_less_than(self, term: BoundTerm[L], literal: Literal[L]) -> BooleanExpression:
        if term.eval(self.struct) < literal.value:
            return self.visit_true()
        else:
            return self.visit_false()

    def visit_less_than_or_equal(self, term: BoundTerm[L], literal: Literal[L]) -> BooleanExpression:
        if term.eval(self.struct) <= literal.value:
            return self.visit_true()
        else:
            return self.visit_false()

    def visit_greater_than(self, term: BoundTerm[L], literal: Literal[L]) -> BooleanExpression:
        if term.eval(self.struct) > literal.value:
            return self.visit_true()
        else:
            return self.visit_false()

    def visit_greater_than_or_equal(self, term: BoundTerm[L], literal: Literal[L]) -> BooleanExpression:
        if term.eval(self.struct) >= literal.value:
            return self.visit_true()
        else:
            return self.visit_false()

    def visit_equal(self, term: BoundTerm[L], literal: Literal[L]) -> BooleanExpression:
        if term.eval(self.struct) == literal.value:
            return self.visit_true()
        else:
            return self.visit_false()

    def visit_not_equal(self, term: BoundTerm[L], literal: Literal[L]) -> BooleanExpression:
        if term.eval(self.struct) != literal.value:
            return self.visit_true()
        else:
            return self.visit_false()

    def visit_in(self, term: BoundTerm[L], literals: Set[L]) -> BooleanExpression:
        if term.eval(self.struct) in literals:
            return self.visit_true()
        else:
            return self.visit_false()

    def visit_not_in(self, term: BoundTerm[L], literals: Set[L]) -> BooleanExpression:
        if term.eval(self.struct) not in literals:
            return self.visit_true()
        else:
            return self.visit_false()

    def visit_starts_with(self, term: BoundTerm[L], literal: Literal[L]) -> BooleanExpression:
        eval_res = term.eval(self.struct)
        if eval_res is not None and str(eval_res).startswith(str(literal.value)):
            return AlwaysTrue()
        else:
            return AlwaysFalse()

    def visit_not_starts_with(self, term: BoundTerm[L], literal: Literal[L]) -> BooleanExpression:
        if not self.visit_starts_with(term, literal):
            return AlwaysTrue()
        else:
            return AlwaysFalse()

    def visit_bound_predicate(self, predicate: BoundPredicate[Any]) -> BooleanExpression:
        """
        If there is no strict projection or if it evaluates to false, then return the predicate.

        Get the strict projection and inclusive projection of this predicate in partition data,
        then use them to determine whether to return the original predicate. The strict projection
        returns true iff the original predicate would have returned true, so the predicate can be
        eliminated if the strict projection evaluates to true. Similarly the inclusive projection
        returns false iff the original predicate would have returned false, so the predicate can
        also be eliminated if the inclusive projection evaluates to false.

        """
        parts = self.spec.fields_by_source_id(predicate.term.ref().field.field_id)
        if parts == []:
            return predicate

        def struct_to_schema(struct: StructType) -> Schema:
            return Schema(*struct.fields)

        for part in parts:
            strict_projection = part.transform.strict_project(part.name, predicate)
            strict_result = None

            if strict_projection is not None:
                bound = strict_projection.bind(
                    struct_to_schema(self.spec.partition_type(self.schema)), case_sensitive=self.case_sensitive
                )
                if isinstance(bound, BoundPredicate):
                    strict_result = super().visit_bound_predicate(bound)
                else:
                    # if the result is not a predicate, then it must be a constant like alwaysTrue or alwaysFalse
                    strict_result = bound

            if isinstance(strict_result, AlwaysTrue):
                return AlwaysTrue()

            inclusive_projection = part.transform.project(part.name, predicate)
            inclusive_result = None
            if inclusive_projection is not None:
                bound_inclusive = inclusive_projection.bind(
                    struct_to_schema(self.spec.partition_type(self.schema)), case_sensitive=self.case_sensitive
                )
                if isinstance(bound_inclusive, BoundPredicate):
                    # using predicate method specific to inclusive
                    inclusive_result = super().visit_bound_predicate(bound_inclusive)
                else:
                    # if the result is not a predicate, then it must be a constant like alwaysTrue or
                    # alwaysFalse
                    inclusive_result = bound_inclusive
            if isinstance(inclusive_result, AlwaysFalse):
                return AlwaysFalse()

        return predicate

    def visit_unbound_predicate(self, predicate: UnboundPredicate[L]) -> BooleanExpression:
        bound = predicate.bind(self.schema, case_sensitive=self.case_sensitive)

        if isinstance(bound, BoundPredicate):
            bound_residual = self.visit_bound_predicate(predicate=bound)
            if not isinstance(bound_residual, (AlwaysFalse, AlwaysTrue)):
                # replace inclusive original unbound predicate
                return predicate

            # use the non-predicate residual (e.g. alwaysTrue)
            return bound_residual

        # if binding didn't result in a Predicate, return the expression
        return bound

visit_bound_predicate(predicate)

If there is no strict projection or if it evaluates to false, then return the predicate.

Get the strict projection and inclusive projection of this predicate in partition data, then use them to determine whether to return the original predicate. The strict projection returns true iff the original predicate would have returned true, so the predicate can be eliminated if the strict projection evaluates to true. Similarly the inclusive projection returns false iff the original predicate would have returned false, so the predicate can also be eliminated if the inclusive projection evaluates to false.

Source code in pyiceberg/expressions/visitors.py 
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
def visit_bound_predicate(self, predicate: BoundPredicate[Any]) -> BooleanExpression:
    """
    If there is no strict projection or if it evaluates to false, then return the predicate.

    Get the strict projection and inclusive projection of this predicate in partition data,
    then use them to determine whether to return the original predicate. The strict projection
    returns true iff the original predicate would have returned true, so the predicate can be
    eliminated if the strict projection evaluates to true. Similarly the inclusive projection
    returns false iff the original predicate would have returned false, so the predicate can
    also be eliminated if the inclusive projection evaluates to false.

    """
    parts = self.spec.fields_by_source_id(predicate.term.ref().field.field_id)
    if parts == []:
        return predicate

    def struct_to_schema(struct: StructType) -> Schema:
        return Schema(*struct.fields)

    for part in parts:
        strict_projection = part.transform.strict_project(part.name, predicate)
        strict_result = None

        if strict_projection is not None:
            bound = strict_projection.bind(
                struct_to_schema(self.spec.partition_type(self.schema)), case_sensitive=self.case_sensitive
            )
            if isinstance(bound, BoundPredicate):
                strict_result = super().visit_bound_predicate(bound)
            else:
                # if the result is not a predicate, then it must be a constant like alwaysTrue or alwaysFalse
                strict_result = bound

        if isinstance(strict_result, AlwaysTrue):
            return AlwaysTrue()

        inclusive_projection = part.transform.project(part.name, predicate)
        inclusive_result = None
        if inclusive_projection is not None:
            bound_inclusive = inclusive_projection.bind(
                struct_to_schema(self.spec.partition_type(self.schema)), case_sensitive=self.case_sensitive
            )
            if isinstance(bound_inclusive, BoundPredicate):
                # using predicate method specific to inclusive
                inclusive_result = super().visit_bound_predicate(bound_inclusive)
            else:
                # if the result is not a predicate, then it must be a constant like alwaysTrue or
                # alwaysFalse
                inclusive_result = bound_inclusive
        if isinstance(inclusive_result, AlwaysFalse):
            return AlwaysFalse()

    return predicate

_ColumnNameTranslator

Bases: BooleanExpressionVisitor[BooleanExpression]

Converts the column names with the ones in the actual file.

Parameters:

Name Type Description Default
file_schema Schema

The schema of the file.

 required
case_sensitive bool

Whether to consider case when binding a reference to a field in a schema, defaults to True.

 required

Raises:

Type Description
TypeError

In the case of an UnboundPredicate.
ValueError

When a column name cannot be found.
Source code in pyiceberg/expressions/visitors.py 
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
class _ColumnNameTranslator(BooleanExpressionVisitor[BooleanExpression]):
    """Converts the column names with the ones in the actual file.

    Args:
      file_schema (Schema): The schema of the file.
      case_sensitive (bool): Whether to consider case when binding a reference to a field in a schema, defaults to True.

    Raises:
        TypeError: In the case of an UnboundPredicate.
        ValueError: When a column name cannot be found.
    """

    file_schema: Schema
    case_sensitive: bool

    def __init__(self, file_schema: Schema, case_sensitive: bool) -> None:
        self.file_schema = file_schema
        self.case_sensitive = case_sensitive

    def visit_true(self) -> BooleanExpression:
        return AlwaysTrue()

    def visit_false(self) -> BooleanExpression:
        return AlwaysFalse()

    def visit_not(self, child_result: BooleanExpression) -> BooleanExpression:
        return Not(child=child_result)

    def visit_and(self, left_result: BooleanExpression, right_result: BooleanExpression) -> BooleanExpression:
        return And(left=left_result, right=right_result)

    def visit_or(self, left_result: BooleanExpression, right_result: BooleanExpression) -> BooleanExpression:
        return Or(left=left_result, right=right_result)

    def visit_unbound_predicate(self, predicate: UnboundPredicate[L]) -> BooleanExpression:
        raise TypeError(f"Expected Bound Predicate, got: {predicate.term}")

    def visit_bound_predicate(self, predicate: BoundPredicate[L]) -> BooleanExpression:
        file_column_name = self.file_schema.find_column_name(predicate.term.ref().field.field_id)

        if file_column_name is None:
            # In the case of schema evolution, the column might not be present
            # in the file schema when reading older data
            if isinstance(predicate, BoundIsNull):
                return AlwaysTrue()
            else:
                return AlwaysFalse()

        if isinstance(predicate, BoundUnaryPredicate):
            return predicate.as_unbound(file_column_name)
        elif isinstance(predicate, BoundLiteralPredicate):
            return predicate.as_unbound(file_column_name, predicate.literal)
        elif isinstance(predicate, BoundSetPredicate):
            return predicate.as_unbound(file_column_name, predicate.literals)
        else:
            raise ValueError(f"Unsupported predicate: {predicate}")

_ExpressionFieldIDs

Bases: BooleanExpressionVisitor[Set[int]]

Extracts the field IDs used in the BooleanExpression.

Source code in pyiceberg/expressions/visitors.py 
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
class _ExpressionFieldIDs(BooleanExpressionVisitor[Set[int]]):
    """Extracts the field IDs used in the BooleanExpression."""

    def visit_true(self) -> Set[int]:
        return set()

    def visit_false(self) -> Set[int]:
        return set()

    def visit_not(self, child_result: Set[int]) -> Set[int]:
        return child_result

    def visit_and(self, left_result: Set[int], right_result: Set[int]) -> Set[int]:
        return left_result.union(right_result)

    def visit_or(self, left_result: Set[int], right_result: Set[int]) -> Set[int]:
        return left_result.union(right_result)

    def visit_unbound_predicate(self, predicate: UnboundPredicate[L]) -> Set[int]:
        raise ValueError("Only works on bound records")

    def visit_bound_predicate(self, predicate: BoundPredicate[L]) -> Set[int]:
        return {predicate.term.ref().field.field_id}

_InclusiveMetricsEvaluator

Bases: _MetricsEvaluator

Source code in pyiceberg/expressions/visitors.py 
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
class _InclusiveMetricsEvaluator(_MetricsEvaluator):
    struct: StructType
    expr: BooleanExpression

    def __init__(
        self, schema: Schema, expr: BooleanExpression, case_sensitive: bool = True, include_empty_files: bool = False
    ) -> None:
        self.struct = schema.as_struct()
        self.include_empty_files = include_empty_files
        self.expr = bind(schema, rewrite_not(expr), case_sensitive)

    def eval(self, file: DataFile) -> bool:
        """Test whether the file may contain records that match the expression."""
        if not self.include_empty_files and file.record_count == 0:
            return ROWS_CANNOT_MATCH

        if file.record_count < 0:
            # Older version don't correctly implement record count from avro file and thus
            # set record count -1 when importing avro tables to iceberg tables. This should
            # be updated once we implemented and set correct record count.
            return ROWS_MIGHT_MATCH

        self.value_counts = file.value_counts or EMPTY_DICT
        self.null_counts = file.null_value_counts or EMPTY_DICT
        self.nan_counts = file.nan_value_counts or EMPTY_DICT
        self.lower_bounds = file.lower_bounds or EMPTY_DICT
        self.upper_bounds = file.upper_bounds or EMPTY_DICT

        return visit(self.expr, self)

    def _may_contain_null(self, field_id: int) -> bool:
        return self.null_counts is None or (field_id in self.null_counts and self.null_counts.get(field_id) is not None)

    def _contains_nans_only(self, field_id: int) -> bool:
        if (nan_count := self.nan_counts.get(field_id)) and (value_count := self.value_counts.get(field_id)):
            return nan_count == value_count
        return False

    def visit_is_null(self, term: BoundTerm[L]) -> bool:
        field_id = term.ref().field.field_id

        if self.null_counts.get(field_id) == 0:
            return ROWS_CANNOT_MATCH

        return ROWS_MIGHT_MATCH

    def visit_not_null(self, term: BoundTerm[L]) -> bool:
        # no need to check whether the field is required because binding evaluates that case
        # if the column has no non-null values, the expression cannot match
        field_id = term.ref().field.field_id

        if self._contains_nulls_only(field_id):
            return ROWS_CANNOT_MATCH

        return ROWS_MIGHT_MATCH

    def visit_is_nan(self, term: BoundTerm[L]) -> bool:
        field_id = term.ref().field.field_id

        if self.nan_counts.get(field_id) == 0:
            return ROWS_CANNOT_MATCH

        # when there's no nanCounts information, but we already know the column only contains null,
        # it's guaranteed that there's no NaN value
        if self._contains_nulls_only(field_id):
            return ROWS_CANNOT_MATCH

        return ROWS_MIGHT_MATCH

    def visit_not_nan(self, term: BoundTerm[L]) -> bool:
        field_id = term.ref().field.field_id

        if self._contains_nans_only(field_id):
            return ROWS_CANNOT_MATCH

        return ROWS_MIGHT_MATCH

    def visit_less_than(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        field = term.ref().field
        field_id = field.field_id

        if self._contains_nulls_only(field_id) or self._contains_nans_only(field_id):
            return ROWS_CANNOT_MATCH

        if not isinstance(field.field_type, PrimitiveType):
            raise ValueError(f"Expected PrimitiveType: {field.field_type}")

        if lower_bound_bytes := self.lower_bounds.get(field_id):
            lower_bound = from_bytes(field.field_type, lower_bound_bytes)

            if self._is_nan(lower_bound):
                # NaN indicates unreliable bounds. See the InclusiveMetricsEvaluator docs for more.
                return ROWS_MIGHT_MATCH

            if lower_bound >= literal.value:  # type: ignore[operator]
                return ROWS_CANNOT_MATCH

        return ROWS_MIGHT_MATCH

    def visit_less_than_or_equal(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        field = term.ref().field
        field_id = field.field_id

        if self._contains_nulls_only(field_id) or self._contains_nans_only(field_id):
            return ROWS_CANNOT_MATCH

        if not isinstance(field.field_type, PrimitiveType):
            raise ValueError(f"Expected PrimitiveType: {field.field_type}")

        if lower_bound_bytes := self.lower_bounds.get(field_id):
            lower_bound = from_bytes(field.field_type, lower_bound_bytes)
            if self._is_nan(lower_bound):
                # NaN indicates unreliable bounds. See the InclusiveMetricsEvaluator docs for more.
                return ROWS_MIGHT_MATCH

            if lower_bound > literal.value:  # type: ignore[operator]
                return ROWS_CANNOT_MATCH

        return ROWS_MIGHT_MATCH

    def visit_greater_than(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        field = term.ref().field
        field_id = field.field_id

        if self._contains_nulls_only(field_id) or self._contains_nans_only(field_id):
            return ROWS_CANNOT_MATCH

        if not isinstance(field.field_type, PrimitiveType):
            raise ValueError(f"Expected PrimitiveType: {field.field_type}")

        if upper_bound_bytes := self.upper_bounds.get(field_id):
            upper_bound = from_bytes(field.field_type, upper_bound_bytes)
            if upper_bound <= literal.value:  # type: ignore[operator]
                if self._is_nan(upper_bound):
                    # NaN indicates unreliable bounds. See the InclusiveMetricsEvaluator docs for more.
                    return ROWS_MIGHT_MATCH

                return ROWS_CANNOT_MATCH

        return ROWS_MIGHT_MATCH

    def visit_greater_than_or_equal(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        field = term.ref().field
        field_id = field.field_id

        if self._contains_nulls_only(field_id) or self._contains_nans_only(field_id):
            return ROWS_CANNOT_MATCH

        if not isinstance(field.field_type, PrimitiveType):
            raise ValueError(f"Expected PrimitiveType: {field.field_type}")

        if upper_bound_bytes := self.upper_bounds.get(field_id):
            upper_bound = from_bytes(field.field_type, upper_bound_bytes)
            if upper_bound < literal.value:  # type: ignore[operator]
                if self._is_nan(upper_bound):
                    # NaN indicates unreliable bounds. See the InclusiveMetricsEvaluator docs for more.
                    return ROWS_MIGHT_MATCH

                return ROWS_CANNOT_MATCH

        return ROWS_MIGHT_MATCH

    def visit_equal(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        field = term.ref().field
        field_id = field.field_id

        if self._contains_nulls_only(field_id) or self._contains_nans_only(field_id):
            return ROWS_CANNOT_MATCH

        if not isinstance(field.field_type, PrimitiveType):
            raise ValueError(f"Expected PrimitiveType: {field.field_type}")

        if lower_bound_bytes := self.lower_bounds.get(field_id):
            lower_bound = from_bytes(field.field_type, lower_bound_bytes)
            if self._is_nan(lower_bound):
                # NaN indicates unreliable bounds. See the InclusiveMetricsEvaluator docs for more.
                return ROWS_MIGHT_MATCH

            if lower_bound > literal.value:  # type: ignore[operator]
                return ROWS_CANNOT_MATCH

        if upper_bound_bytes := self.upper_bounds.get(field_id):
            upper_bound = from_bytes(field.field_type, upper_bound_bytes)
            if self._is_nan(upper_bound):
                # NaN indicates unreliable bounds. See the InclusiveMetricsEvaluator docs for more.
                return ROWS_MIGHT_MATCH

            if upper_bound < literal.value:  # type: ignore[operator]
                return ROWS_CANNOT_MATCH

        return ROWS_MIGHT_MATCH

    def visit_not_equal(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        return ROWS_MIGHT_MATCH

    def visit_in(self, term: BoundTerm[L], literals: Set[L]) -> bool:
        field = term.ref().field
        field_id = field.field_id

        if self._contains_nulls_only(field_id) or self._contains_nans_only(field_id):
            return ROWS_CANNOT_MATCH

        if len(literals) > IN_PREDICATE_LIMIT:
            # skip evaluating the predicate if the number of values is too big
            return ROWS_MIGHT_MATCH

        if not isinstance(field.field_type, PrimitiveType):
            raise ValueError(f"Expected PrimitiveType: {field.field_type}")

        if lower_bound_bytes := self.lower_bounds.get(field_id):
            lower_bound = from_bytes(field.field_type, lower_bound_bytes)
            if self._is_nan(lower_bound):
                # NaN indicates unreliable bounds. See the InclusiveMetricsEvaluator docs for more.
                return ROWS_MIGHT_MATCH

            literals = {lit for lit in literals if lower_bound <= lit}  # type: ignore[operator]
            if len(literals) == 0:
                return ROWS_CANNOT_MATCH

        if upper_bound_bytes := self.upper_bounds.get(field_id):
            upper_bound = from_bytes(field.field_type, upper_bound_bytes)
            # this is different from Java, here NaN is always larger
            if self._is_nan(upper_bound):
                return ROWS_MIGHT_MATCH

            literals = {lit for lit in literals if upper_bound >= lit}  # type: ignore[operator]
            if len(literals) == 0:
                return ROWS_CANNOT_MATCH

        return ROWS_MIGHT_MATCH

    def visit_not_in(self, term: BoundTerm[L], literals: Set[L]) -> bool:
        # because the bounds are not necessarily a min or max value, this cannot be answered using
        # them. notIn(col, {X, ...}) with (X, Y) doesn't guarantee that X is a value in col.
        return ROWS_MIGHT_MATCH

    def visit_starts_with(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        field = term.ref().field
        field_id: int = field.field_id

        if self._contains_nulls_only(field_id):
            return ROWS_CANNOT_MATCH

        if not isinstance(field.field_type, PrimitiveType):
            raise ValueError(f"Expected PrimitiveType: {field.field_type}")

        prefix = str(literal.value)
        len_prefix = len(prefix)

        if lower_bound_bytes := self.lower_bounds.get(field_id):
            lower_bound = str(from_bytes(field.field_type, lower_bound_bytes))

            # truncate lower bound so that its length is not greater than the length of prefix
            if lower_bound and lower_bound[:len_prefix] > prefix:
                return ROWS_CANNOT_MATCH

        if upper_bound_bytes := self.upper_bounds.get(field_id):
            upper_bound = str(from_bytes(field.field_type, upper_bound_bytes))

            # truncate upper bound so that its length is not greater than the length of prefix
            if upper_bound is not None and upper_bound[:len_prefix] < prefix:
                return ROWS_CANNOT_MATCH

        return ROWS_MIGHT_MATCH

    def visit_not_starts_with(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        field = term.ref().field
        field_id: int = field.field_id

        if self._may_contain_null(field_id):
            return ROWS_MIGHT_MATCH

        if not isinstance(field.field_type, PrimitiveType):
            raise ValueError(f"Expected PrimitiveType: {field.field_type}")

        prefix = str(literal.value)
        len_prefix = len(prefix)

        # not_starts_with will match unless all values must start with the prefix. This happens when
        # the lower and upper bounds both start with the prefix.
        if (lower_bound_bytes := self.lower_bounds.get(field_id)) and (upper_bound_bytes := self.upper_bounds.get(field_id)):
            lower_bound = str(from_bytes(field.field_type, lower_bound_bytes))
            upper_bound = str(from_bytes(field.field_type, upper_bound_bytes))

            # if lower is shorter than the prefix then lower doesn't start with the prefix
            if len(lower_bound) < len_prefix:
                return ROWS_MIGHT_MATCH

            if lower_bound[:len_prefix] == prefix:
                # if upper is shorter than the prefix then upper can't start with the prefix
                if len(upper_bound) < len_prefix:
                    return ROWS_MIGHT_MATCH

                if upper_bound[:len_prefix] == prefix:
                    return ROWS_CANNOT_MATCH

        return ROWS_MIGHT_MATCH

eval(file)

Test whether the file may contain records that match the expression.

Source code in pyiceberg/expressions/visitors.py 
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
def eval(self, file: DataFile) -> bool:
    """Test whether the file may contain records that match the expression."""
    if not self.include_empty_files and file.record_count == 0:
        return ROWS_CANNOT_MATCH

    if file.record_count < 0:
        # Older version don't correctly implement record count from avro file and thus
        # set record count -1 when importing avro tables to iceberg tables. This should
        # be updated once we implemented and set correct record count.
        return ROWS_MIGHT_MATCH

    self.value_counts = file.value_counts or EMPTY_DICT
    self.null_counts = file.null_value_counts or EMPTY_DICT
    self.nan_counts = file.nan_value_counts or EMPTY_DICT
    self.lower_bounds = file.lower_bounds or EMPTY_DICT
    self.upper_bounds = file.upper_bounds or EMPTY_DICT

    return visit(self.expr, self)

_RewriteNotVisitor

Bases: BooleanExpressionVisitor[BooleanExpression]

Inverts the negations.

Source code in pyiceberg/expressions/visitors.py 
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
class _RewriteNotVisitor(BooleanExpressionVisitor[BooleanExpression]):
    """Inverts the negations."""

    def visit_true(self) -> BooleanExpression:
        return AlwaysTrue()

    def visit_false(self) -> BooleanExpression:
        return AlwaysFalse()

    def visit_not(self, child_result: BooleanExpression) -> BooleanExpression:
        return ~child_result

    def visit_and(self, left_result: BooleanExpression, right_result: BooleanExpression) -> BooleanExpression:
        return And(left=left_result, right=right_result)

    def visit_or(self, left_result: BooleanExpression, right_result: BooleanExpression) -> BooleanExpression:
        return Or(left=left_result, right=right_result)

    def visit_unbound_predicate(self, predicate: UnboundPredicate[L]) -> BooleanExpression:
        return predicate

    def visit_bound_predicate(self, predicate: BoundPredicate[L]) -> BooleanExpression:
        return predicate

_StrictMetricsEvaluator

Bases: _MetricsEvaluator

Source code in pyiceberg/expressions/visitors.py 
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
class _StrictMetricsEvaluator(_MetricsEvaluator):
    struct: StructType
    expr: BooleanExpression

    def __init__(
        self, schema: Schema, expr: BooleanExpression, case_sensitive: bool = True, include_empty_files: bool = False
    ) -> None:
        self.struct = schema.as_struct()
        self.include_empty_files = include_empty_files
        self.expr = bind(schema, rewrite_not(expr), case_sensitive)

    def eval(self, file: DataFile) -> bool:
        """Test whether all records within the file match the expression.

        Args:
            file: A data file

        Returns: false if the file may contain any row that doesn't match
                    the expression, true otherwise.
        """
        if file.record_count <= 0:
            # Older version don't correctly implement record count from avro file and thus
            # set record count -1 when importing avro tables to iceberg tables. This should
            # be updated once we implemented and set correct record count.
            return ROWS_MUST_MATCH

        self.value_counts = file.value_counts or EMPTY_DICT
        self.null_counts = file.null_value_counts or EMPTY_DICT
        self.nan_counts = file.nan_value_counts or EMPTY_DICT
        self.lower_bounds = file.lower_bounds or EMPTY_DICT
        self.upper_bounds = file.upper_bounds or EMPTY_DICT

        return visit(self.expr, self)

    def visit_is_null(self, term: BoundTerm[L]) -> bool:
        # no need to check whether the field is required because binding evaluates that case
        # if the column has any non-null values, the expression does not match
        field_id = term.ref().field.field_id

        if self._contains_nulls_only(field_id):
            return ROWS_MUST_MATCH
        else:
            return ROWS_MIGHT_NOT_MATCH

    def visit_not_null(self, term: BoundTerm[L]) -> bool:
        # no need to check whether the field is required because binding evaluates that case
        # if the column has any non-null values, the expression does not match
        field_id = term.ref().field.field_id

        if (null_count := self.null_counts.get(field_id)) is not None and null_count == 0:
            return ROWS_MUST_MATCH
        else:
            return ROWS_MIGHT_NOT_MATCH

    def visit_is_nan(self, term: BoundTerm[L]) -> bool:
        field_id = term.ref().field.field_id

        if self._contains_nans_only(field_id):
            return ROWS_MUST_MATCH
        else:
            return ROWS_MIGHT_NOT_MATCH

    def visit_not_nan(self, term: BoundTerm[L]) -> bool:
        field_id = term.ref().field.field_id

        if (nan_count := self.nan_counts.get(field_id)) is not None and nan_count == 0:
            return ROWS_MUST_MATCH

        if self._contains_nulls_only(field_id):
            return ROWS_MUST_MATCH

        return ROWS_MIGHT_NOT_MATCH

    def visit_less_than(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        # Rows must match when: <----------Min----Max---X------->

        field_id = term.ref().field.field_id

        if self._can_contain_nulls(field_id) or self._can_contain_nans(field_id):
            return ROWS_MIGHT_NOT_MATCH

        if upper_bytes := self.upper_bounds.get(field_id):
            field = self._get_field(field_id)
            upper = _from_byte_buffer(field.field_type, upper_bytes)

            if upper < literal.value:
                return ROWS_MUST_MATCH

        return ROWS_MIGHT_NOT_MATCH

    def visit_less_than_or_equal(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        # Rows must match when: <----------Min----Max---X------->

        field_id = term.ref().field.field_id

        if self._can_contain_nulls(field_id) or self._can_contain_nans(field_id):
            return ROWS_MIGHT_NOT_MATCH

        if upper_bytes := self.upper_bounds.get(field_id):
            field = self._get_field(field_id)
            upper = _from_byte_buffer(field.field_type, upper_bytes)

            if upper <= literal.value:
                return ROWS_MUST_MATCH

        return ROWS_MIGHT_NOT_MATCH

    def visit_greater_than(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        # Rows must match when: <-------X---Min----Max---------->

        field_id = term.ref().field.field_id

        if self._can_contain_nulls(field_id) or self._can_contain_nans(field_id):
            return ROWS_MIGHT_NOT_MATCH

        if lower_bytes := self.lower_bounds.get(field_id):
            field = self._get_field(field_id)
            lower = _from_byte_buffer(field.field_type, lower_bytes)

            if self._is_nan(lower):
                # NaN indicates unreliable bounds.
                # See the _StrictMetricsEvaluator docs for more.
                return ROWS_MIGHT_NOT_MATCH

            if lower > literal.value:
                return ROWS_MUST_MATCH

        return ROWS_MIGHT_NOT_MATCH

    def visit_greater_than_or_equal(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        # Rows must match when: <-------X---Min----Max---------->
        field_id = term.ref().field.field_id

        if self._can_contain_nulls(field_id) or self._can_contain_nans(field_id):
            return ROWS_MIGHT_NOT_MATCH

        if lower_bytes := self.lower_bounds.get(field_id):
            field = self._get_field(field_id)
            lower = _from_byte_buffer(field.field_type, lower_bytes)

            if self._is_nan(lower):
                # NaN indicates unreliable bounds.
                # See the _StrictMetricsEvaluator docs for more.
                return ROWS_MIGHT_NOT_MATCH

            if lower >= literal.value:
                return ROWS_MUST_MATCH

        return ROWS_MIGHT_NOT_MATCH

    def visit_equal(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        # Rows must match when Min == X == Max
        field_id = term.ref().field.field_id

        if self._can_contain_nulls(field_id) or self._can_contain_nans(field_id):
            return ROWS_MIGHT_NOT_MATCH

        if (lower_bytes := self.lower_bounds.get(field_id)) and (upper_bytes := self.upper_bounds.get(field_id)):
            field = self._get_field(field_id)
            lower = _from_byte_buffer(field.field_type, lower_bytes)
            upper = _from_byte_buffer(field.field_type, upper_bytes)

            if lower != literal.value or upper != literal.value:
                return ROWS_MIGHT_NOT_MATCH
            else:
                return ROWS_MUST_MATCH

        return ROWS_MIGHT_NOT_MATCH

    def visit_not_equal(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        # Rows must match when X < Min or Max < X because it is not in the range
        field_id = term.ref().field.field_id

        if self._can_contain_nulls(field_id) or self._can_contain_nans(field_id):
            return ROWS_MUST_MATCH

        field = self._get_field(field_id)

        if lower_bytes := self.lower_bounds.get(field_id):
            lower = _from_byte_buffer(field.field_type, lower_bytes)

            if self._is_nan(lower):
                # NaN indicates unreliable bounds.
                # See the _StrictMetricsEvaluator docs for more.
                return ROWS_MIGHT_NOT_MATCH

            if lower > literal.value:
                return ROWS_MUST_MATCH

        if upper_bytes := self.upper_bounds.get(field_id):
            upper = _from_byte_buffer(field.field_type, upper_bytes)

            if upper < literal.value:
                return ROWS_MUST_MATCH

        return ROWS_MIGHT_NOT_MATCH

    def visit_in(self, term: BoundTerm[L], literals: Set[L]) -> bool:
        field_id = term.ref().field.field_id

        if self._can_contain_nulls(field_id) or self._can_contain_nans(field_id):
            return ROWS_MIGHT_NOT_MATCH

        field = self._get_field(field_id)

        if (lower_bytes := self.lower_bounds.get(field_id)) and (upper_bytes := self.upper_bounds.get(field_id)):
            # similar to the implementation in eq, first check if the lower bound is in the set
            lower = _from_byte_buffer(field.field_type, lower_bytes)
            if lower not in literals:
                return ROWS_MIGHT_NOT_MATCH

            # check if the upper bound is in the set
            upper = _from_byte_buffer(field.field_type, upper_bytes)
            if upper not in literals:
                return ROWS_MIGHT_NOT_MATCH

            # finally check if the lower bound and the upper bound are equal
            if lower != upper:
                return ROWS_MIGHT_NOT_MATCH

            # All values must be in the set if the lower bound and the upper bound are
            # in the set and are equal.
            return ROWS_MUST_MATCH

        return ROWS_MIGHT_NOT_MATCH

    def visit_not_in(self, term: BoundTerm[L], literals: Set[L]) -> bool:
        field_id = term.ref().field.field_id

        if self._can_contain_nulls(field_id) or self._can_contain_nans(field_id):
            return ROWS_MUST_MATCH

        field = self._get_field(field_id)

        if lower_bytes := self.lower_bounds.get(field_id):
            lower = _from_byte_buffer(field.field_type, lower_bytes)

            if self._is_nan(lower):
                # NaN indicates unreliable bounds.
                # See the StrictMetricsEvaluator docs for more.
                return ROWS_MIGHT_NOT_MATCH

            literals = {val for val in literals if lower <= val}
            if len(literals) == 0:
                return ROWS_MUST_MATCH

        if upper_bytes := self.upper_bounds.get(field_id):
            upper = _from_byte_buffer(field.field_type, upper_bytes)

            literals = {val for val in literals if upper >= val}

            if len(literals) == 0:
                return ROWS_MUST_MATCH

        return ROWS_MIGHT_NOT_MATCH

    def visit_starts_with(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        return ROWS_MIGHT_NOT_MATCH

    def visit_not_starts_with(self, term: BoundTerm[L], literal: Literal[L]) -> bool:
        return ROWS_MIGHT_NOT_MATCH

    def _get_field(self, field_id: int) -> NestedField:
        field = self.struct.field(field_id=field_id)
        if field is None:
            raise ValueError(f"Cannot find field, might be nested or missing: {field_id}")

        return field

    def _can_contain_nulls(self, field_id: int) -> bool:
        return (null_count := self.null_counts.get(field_id)) is not None and null_count > 0

    def _can_contain_nans(self, field_id: int) -> bool:
        return (nan_count := self.nan_counts.get(field_id)) is not None and nan_count > 0

eval(file)

Test whether all records within the file match the expression.

Parameters:

Name Type Description Default
file DataFile

A data file

 required

false if the file may contain any row that doesn't match

Type Description
bool

the expression, true otherwise.
Source code in pyiceberg/expressions/visitors.py 
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
def eval(self, file: DataFile) -> bool:
    """Test whether all records within the file match the expression.

    Args:
        file: A data file

    Returns: false if the file may contain any row that doesn't match
                the expression, true otherwise.
    """
    if file.record_count <= 0:
        # Older version don't correctly implement record count from avro file and thus
        # set record count -1 when importing avro tables to iceberg tables. This should
        # be updated once we implemented and set correct record count.
        return ROWS_MUST_MATCH

    self.value_counts = file.value_counts or EMPTY_DICT
    self.null_counts = file.null_value_counts or EMPTY_DICT
    self.nan_counts = file.nan_value_counts or EMPTY_DICT
    self.lower_bounds = file.lower_bounds or EMPTY_DICT
    self.upper_bounds = file.upper_bounds or EMPTY_DICT

    return visit(self.expr, self)

bind(schema, expression, case_sensitive)

Travers over an expression to bind the predicates to the schema.

Parameters:

Name Type Description Default
schema Schema

A schema to use when binding the expression.

 required
expression BooleanExpression

An expression containing UnboundPredicates that can be bound.

 required
case_sensitive bool

Whether to consider case when binding a reference to a field in a schema, defaults to True.

 required

Raises:

Type Description
TypeError

In the case a predicate is already bound.
Source code in pyiceberg/expressions/visitors.py 
203
204
205
206
207
208
209
210
211
212
213
214
def bind(schema: Schema, expression: BooleanExpression, case_sensitive: bool) -> BooleanExpression:
    """Travers over an expression to bind the predicates to the schema.

    Args:
      schema (Schema): A schema to use when binding the expression.
      expression (BooleanExpression): An expression containing UnboundPredicates that can be bound.
      case_sensitive (bool): Whether to consider case when binding a reference to a field in a schema, defaults to True.

    Raises:
        TypeError: In the case a predicate is already bound.
    """
    return visit(expression, BindVisitor(schema, case_sensitive))

expression_to_plain_format(expressions, cast_int_to_datetime=False)

Format a Disjunctive Normal Form expression.

These are the formats that the expression can be fed into:

  • https://arrow.apache.org/docs/python/generated/pyarrow.parquet.read_table.html
  • https://docs.dask.org/en/stable/generated/dask.dataframe.read_parquet.html

Contrary to normal DNF that may contain Not expressions, but here they should have been rewritten. This can be done using rewrite_not(...).

Keep in mind that this is only used for page skipping, and still needs to filter on a row level.

Parameters:

Name Type Description Default
expressions Tuple[BooleanExpression, ...]

Expression in Disjunctive Normal Form.

 required

Returns:

Type Description
List[List[Tuple[str, str, Any]]]

Formatter filter compatible with Dask and PyArrow.
Source code in pyiceberg/expressions/visitors.py 
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
def expression_to_plain_format(
    expressions: Tuple[BooleanExpression, ...], cast_int_to_datetime: bool = False
) -> List[List[Tuple[str, str, Any]]]:
    """Format a Disjunctive Normal Form expression.

    These are the formats that the expression can be fed into:

    - https://arrow.apache.org/docs/python/generated/pyarrow.parquet.read_table.html
    - https://docs.dask.org/en/stable/generated/dask.dataframe.read_parquet.html

    Contrary to normal DNF that may contain Not expressions, but here they should have
    been rewritten. This can be done using ``rewrite_not(...)``.

    Keep in mind that this is only used for page skipping, and still needs to filter
    on a row level.

    Args:
        expressions: Expression in Disjunctive Normal Form.

    Returns:
        Formatter filter compatible with Dask and PyArrow.
    """
    # In the form of expr1 ∨ expr2 ∨ ... ∨ exprN
    visitor = ExpressionToPlainFormat(cast_int_to_datetime)
    return [visit(expression, visitor) for expression in expressions]

visit(obj, visitor)

Apply a boolean expression visitor to any point within an expression.

The function traverses the expression in post-order fashion.

Parameters:

Name Type Description Default
obj BooleanExpression

An instance of a BooleanExpression.

 required
visitor BooleanExpressionVisitor[T]

An instance of an implementation of the generic BooleanExpressionVisitor base class.

 required

Raises:

Type Description
NotImplementedError

If attempting to visit an unsupported expression.
Source code in pyiceberg/expressions/visitors.py 
140
141
142
143
144
145
146
147
148
149
150
151
152
153
@singledispatch
def visit(obj: BooleanExpression, visitor: BooleanExpressionVisitor[T]) -> T:
    """Apply a boolean expression visitor to any point within an expression.

    The function traverses the expression in post-order fashion.

    Args:
        obj (BooleanExpression): An instance of a BooleanExpression.
        visitor (BooleanExpressionVisitor[T]): An instance of an implementation of the generic BooleanExpressionVisitor base class.

    Raises:
        NotImplementedError: If attempting to visit an unsupported expression.
    """
    raise NotImplementedError(f"Cannot visit unsupported expression: {obj}")