def _coalesce_fix_list(
context: RuleContext,
coalesce_arg_1: BaseSegment,
coalesce_arg_2: BaseSegment,
preceding_not: bool = False,
) -> List[LintFix]:
"""Generate list of fixes to convert CASE statement to COALESCE function."""
# Add coalesce and opening parenthesis.
edits = [
KeywordSegment("coalesce"),
SymbolSegment("(", type="start_bracket"),
coalesce_arg_1,
SymbolSegment(",", type="comma"),
WhitespaceSegment(),
coalesce_arg_2,
SymbolSegment(")", type="end_bracket"),
]
if preceding_not:
not_edits: List[BaseSegment] = [
KeywordSegment("not"),
WhitespaceSegment(),
]
edits = not_edits + edits
fixes = [LintFix.replace(
context.segment,
edits,
)]
return fixes
def _handle_semicolon_newline(
self, context: RuleContext,
info: SegmentMoveContext) -> Optional[LintResult]:
# Adjust before_segment and anchor_segment for preceding inline
# comments. Inline comments can contain noqa logic so we need to add the
# newline after the inline comment.
(
before_segment,
anchor_segment,
) = self._handle_preceding_inline_comments(info.before_segment,
info.anchor_segment)
if (len(before_segment) == 1) and all(
s.is_type("newline") for s in before_segment):
return None
# If preceding segment is not a single newline then delete the old
# semi-colon/preceding whitespace and then insert the
# semi-colon in the correct location.
# This handles an edge case in which an inline comment comes after
# the semi-colon.
anchor_segment = self._handle_trailing_inline_comments(
context, anchor_segment)
fixes = []
if anchor_segment is context.segment:
fixes.append(
LintFix.replace(
anchor_segment,
[
NewlineSegment(),
SymbolSegment(raw=";", type="symbol",
name="semicolon"),
],
))
else:
fixes.extend([
LintFix.replace(
anchor_segment,
[
anchor_segment,
NewlineSegment(),
SymbolSegment(raw=";", type="symbol",
name="semicolon"),
],
),
LintFix.delete(context.segment, ),
])
fixes.extend(LintFix.delete(d) for d in info.whitespace_deletions)
return LintResult(
anchor=anchor_segment,
fixes=fixes,
)
def _handle_semicolon_same_line(
self,
target_segment: RawSegment,
parent_segment: BaseSegment,
info: SegmentMoveContext,
) -> Optional[LintResult]:
if not info.before_segment:
return None
# If preceding segments are found then delete the old
# semi-colon and its preceding whitespace and then insert
# the semi-colon in the correct location.
fixes = self._create_semicolon_and_delete_whitespace(
target_segment,
parent_segment,
info.anchor_segment,
info.whitespace_deletions,
[
SymbolSegment(raw=";", type="statement_terminator"),
],
)
return LintResult(
anchor=info.anchor_segment,
fixes=fixes,
)
def _eval(self, segment, parent_stack, **kwargs):
"""Trailing commas within select clause."""
if segment.is_type("select_clause"):
# Iterate content to find last element
last_content = None
for seg in segment.segments:
if seg.is_code:
last_content = seg
# What mode are we in?
if self.select_clause_trailing_comma == "forbid":
# Is it a comma?
if last_content.is_type("comma"):
return LintResult(
anchor=last_content,
fixes=[LintFix("delete", last_content)],
description="Trailing comma in select statement forbidden",
)
elif self.select_clause_trailing_comma == "require":
if not last_content.is_type("comma"):
new_comma = SymbolSegment(",", name="comma", type="comma")
return LintResult(
anchor=last_content,
fixes=[
LintFix("edit", last_content, [last_content, new_comma])
],
description="Trailing comma in select statement required",
)
return None
def _eval(self, context: RuleContext) -> Optional[LintResult]:
"""Trailing commas within select clause."""
# Config type hints
self.select_clause_trailing_comma: str
segment = FunctionalContext(context).segment
children = segment.children()
# Iterate content to find last element
last_content: BaseSegment = children.last(sp.is_code())[0]
# What mode are we in?
if self.select_clause_trailing_comma == "forbid":
# Is it a comma?
if last_content.is_type("comma"):
return LintResult(
anchor=last_content,
fixes=[LintFix.delete(last_content)],
description="Trailing comma in select statement forbidden",
)
elif self.select_clause_trailing_comma == "require":
if not last_content.is_type("comma"):
new_comma = SymbolSegment(",", type="comma")
return LintResult(
anchor=last_content,
fixes=[
LintFix.replace(last_content,
[last_content, new_comma])
],
description="Trailing comma in select statement required",
)
return None
def make_symbol(cls, raw, seg_type, name=None):
"""Make a symbol segment."""
# For the name of the segment, we force the string to lowercase.
symbol_seg = SymbolSegment.make(raw.lower(),
name=name or seg_type,
type=seg_type)
# At the moment we let the rule dictate *case* here.
return symbol_seg(raw=raw, pos_marker=None)
def _handle_semicolon_same_line(
context: RuleContext,
info: SegmentMoveContext) -> Optional[LintResult]:
if not info.before_segment:
return None
# If preceding segments are found then delete the old
# semi-colon and its preceding whitespace and then insert
# the semi-colon in the correct location.
fixes = [
LintFix.replace(
info.anchor_segment,
[
info.anchor_segment,
SymbolSegment(raw=";", type="symbol", name="semicolon"),
],
),
LintFix.delete(context.segment, ),
]
fixes.extend(LintFix.delete(d) for d in info.whitespace_deletions)
return LintResult(
anchor=info.anchor_segment,
fixes=fixes,
)
def _eval(self, segment, **kwargs):
"""Find rule violations and provide fixes.
0. Look for a case expression
1. Find the first expression and "then"
2. Determine if the "then" is followed by a boolean
3. If so, determine if the first then-bool is followed by an else-bool
4. If so, delete everything but the first expression
5a. If then-true-else-false
* return deletions
* wrap with coalesce
5b. If then-false-else-true
* return deletions
* add a not condition
* wrap with parenthesis and coalesce
"""
# Look for a case expression
if segment.is_type("case_expression") and segment.segments[0].name == "case":
# Find the first expression and "then"
idx = 0
while segment.segments[idx].name != "then":
if segment.segments[idx].is_type("expression"):
expression_idx = idx
idx += 1
# Determine if "then" is followed by a boolean
then_bool_type = None
while segment.segments[idx].name not in ["when", "else", "end"]:
if segment.segments[idx].raw_upper in ["TRUE", "FALSE"]:
then_bool_type = segment.segments[idx].raw_upper
idx += 1
if then_bool_type:
# Determine if the first then-bool is followed by an else-bool
while segment.segments[idx].name != "else":
# If the first then-bool is followed by a "WHEN" or "END", exit
if segment.segments[idx].name in ["when", "end"]:
return None
idx += 1 # pragma: no cover
# Determine if "else" is followed by a boolean
else_bool_type = None
while segment.segments[idx].name != "end":
if segment.segments[idx].raw_upper in ["TRUE", "FALSE"]:
else_bool_type = segment.segments[idx].raw_upper
idx += 1
# If then-bool-else-bool, return fixes
if (
then_bool_type is not None
and else_bool_type is not None
and then_bool_type != else_bool_type
):
# Generate list of segments to delete -- everything but the
# first expression.
delete_segments = []
for s in segment.segments:
if s != segment.segments[expression_idx]:
delete_segments.append(s)
# If then-false, add "not" and space
edits = []
if then_bool_type == "FALSE":
edits.extend(
[
KeywordSegment("not"),
WhitespaceSegment(),
]
)
# Add coalesce and parenthesis
edits.extend(
[
KeywordSegment("coalesce"),
SymbolSegment("(", name="start_bracket", type="start_bracket"),
]
)
edit_coalesce_target = segment.segments[0]
fixes = []
fixes.append(
LintFix(
"edit",
edit_coalesce_target,
edits,
)
)
# Add comma, bool, closing parenthesis
expression = segment.segments[expression_idx + 1]
closing_parenthesis = [
SymbolSegment(",", name="comma", type="comma"),
WhitespaceSegment(),
KeywordSegment("false"),
SymbolSegment(")", name="end_bracket", type="end_bracket"),
]
fixes.append(
LintFix(
"edit",
expression,
closing_parenthesis,
)
)
# Generate a "delete" action for each segment in
# delete_segments EXCEPT the one being edited to become a call
# to "coalesce(". Deleting and editing the same segment has
# unpredictable behavior.
fixes += [
LintFix("delete", s)
for s in delete_segments
if s is not edit_coalesce_target
]
return LintResult(
anchor=segment.segments[expression_idx],
fixes=fixes,
description="Case when returns booleans.",
)
exasol_fs_dialect.add(
FunctionScriptTerminatorSegment=NamedSegment.make(
"function_script_terminator", type="statement_terminator"
),
WalrusOperatorSegment=NamedSegment.make(
"walrus_operator", type="assignment_operator"
),
VariableNameSegment=ReSegment.make(
r"[A-Z][A-Z0-9_]*",
name="function_variable",
type="variable",
),
)
exasol_fs_dialect.replace(
SemicolonSegment=SymbolSegment.make(";", name="semicolon", type="semicolon"),
)
@exasol_fs_dialect.segment(replace=True)
class StatementSegment(BaseSegment):
"""A generic segment, to any of its child subsegments."""
type = "statement"
match_grammar = GreedyUntil(Ref("FunctionScriptTerminatorSegment"))
parse_grammar = OneOf(
Ref("CreateFunctionStatementSegment"),
Ref("CreateScriptingLuaScriptStatementSegment"),
Ref("CreateUDFScriptStatementSegment"),
Ref("CreateAdapterScriptStatementSegment"),
def _ensure_final_semicolon(self,
context: RuleContext) -> Optional[LintResult]:
# Locate the end of the file.
if not self.is_final_segment(context):
return None
# Include current segment for complete stack.
complete_stack: List[BaseSegment] = list(context.raw_stack)
complete_stack.append(context.segment)
# Iterate backwards over complete stack to find
# if the final semi-colon is already present.
anchor_segment = context.segment
semi_colon_exist_flag = False
is_one_line = False
before_segment = []
for segment in complete_stack[::-1]:
if segment.name == "semicolon":
semi_colon_exist_flag = True
elif segment.is_code:
is_one_line = self._is_one_line_statement(context, segment)
break
elif not segment.is_meta:
before_segment.append(segment)
anchor_segment = segment
semicolon_newline = self.multiline_newline if not is_one_line else False
if not semi_colon_exist_flag:
# Create the final semi-colon if it does not yet exist.
# Semi-colon on same line.
if not semicolon_newline:
fixes = [
LintFix.replace(
anchor_segment,
[
anchor_segment,
SymbolSegment(
raw=";", type="symbol", name="semicolon"),
],
)
]
# Semi-colon on new line.
else:
# Adjust before_segment and anchor_segment for inline
# comments.
(
before_segment,
anchor_segment,
) = self._handle_preceding_inline_comments(
before_segment, anchor_segment)
fixes = [
LintFix.replace(
anchor_segment,
[
anchor_segment,
NewlineSegment(),
SymbolSegment(
raw=";", type="symbol", name="semicolon"),
],
)
]
return LintResult(
anchor=anchor_segment,
fixes=fixes,
)
return None
snowflake_dialect.sets("reserved_keywords").update([
"CLONE",
"MASKING",
"NOTIFICATION",
"PIVOT",
"SAMPLE",
"TABLESAMPLE",
"UNPIVOT",
])
snowflake_dialect.add(
# In snowflake, these are case sensitive even though they're not quoted
# so they need a different `name` and `type` so they're not picked up
# by other rules.
ParameterAssignerSegment=SymbolSegment.make("=>",
name="parameter_assigner",
type="parameter_assigner"),
NakedSemiStructuredElementSegment=ReSegment.make(
r"[A-Z0-9_]*",
name="naked_semi_structured_element",
type="semi_structured_element",
),
QuotedSemiStructuredElementSegment=NamedSegment.make(
"double_quote",
name="quoted_semi_structured_element",
type="semi_structured_element",
),
ColumnIndexIdentifierSegment=ReSegment.make(
r"\$[0-9]+", name="column_index_identifier_segment",
type="identifier"),
)
def _ensure_final_semicolon(
self, parent_segment: BaseSegment) -> Optional[LintResult]:
# Iterate backwards over complete stack to find
# if the final semi-colon is already present.
anchor_segment = parent_segment.segments[-1]
trigger_segment = parent_segment.segments[-1]
semi_colon_exist_flag = False
is_one_line = False
before_segment = []
for segment in parent_segment.segments[::-1]:
anchor_segment = segment
if segment.is_type("statement_terminator"):
semi_colon_exist_flag = True
elif segment.is_code:
is_one_line = self._is_one_line_statement(
parent_segment, segment)
break
elif not segment.is_meta:
before_segment.append(segment)
trigger_segment = segment
self.logger.debug("Trigger on: %s", trigger_segment)
self.logger.debug("Anchoring on: %s", anchor_segment)
semicolon_newline = self.multiline_newline if not is_one_line else False
if not semi_colon_exist_flag:
# Create the final semi-colon if it does not yet exist.
# Semi-colon on same line.
if not semicolon_newline:
fixes = [
LintFix.create_after(
self._choose_anchor_segment(
parent_segment,
"create_after",
anchor_segment,
filter_meta=True,
),
[
SymbolSegment(raw=";",
type="statement_terminator"),
],
)
]
# Semi-colon on new line.
else:
# Adjust before_segment and anchor_segment for inline
# comments.
(
before_segment,
anchor_segment,
) = self._handle_preceding_inline_comments(
before_segment, anchor_segment)
self.logger.debug("Revised anchor on: %s", anchor_segment)
fixes = [
LintFix.create_after(
self._choose_anchor_segment(
parent_segment,
"create_after",
anchor_segment,
filter_meta=True,
),
[
NewlineSegment(),
SymbolSegment(raw=";",
type="statement_terminator"),
],
)
]
return LintResult(
anchor=trigger_segment,
fixes=fixes,
)
return None
),
] + exasol_fs_dialect.get_lexer_struct())
exasol_fs_dialect.add(
FunctionScriptTerminatorSegment=NamedSegment.make(
"function_script_terminator", type="statement_terminator"),
WalrusOperatorSegment=NamedSegment.make("walrus_operator",
type="assignment_operator"),
VariableNameSegment=ReSegment.make(
r"[A-Z][A-Z0-9_]*",
name="function_variable",
type="variable",
),
)
exasol_fs_dialect.replace(SemicolonSegment=SymbolSegment.make(
";", name="semicolon", type="semicolon"), )
@exasol_fs_dialect.segment(replace=True)
class StatementSegment(BaseSegment):
"""A generic segment, to any of its child subsegments."""
type = "statement"
match_grammar = GreedyUntil(Ref("FunctionScriptTerminatorSegment"))
parse_grammar = OneOf(
Ref("CreateFunctionStatementSegment"),
Ref("CreateScriptingLuaScriptStatementSegment"),
Ref("CreateUDFScriptStatementSegment"),
Ref("CreateAdapterScriptStatementSegment"),
)
