def _cache_file(self, src_path):
"""
Load the data from `self._xml_roots`
for `src_path`, if it hasn't been already.
"""
# If we have not yet loaded this source file
if src_path not in self._info_cache:
# We only want to keep violations that show up in each xml source.
# Thus, each time, we take the intersection. However, to do this
# we must treat the first time as a special case and just add all
# the violations from the first xml report.
violations = None
# A line is measured if it is measured in any of the reports, so
# we take set union each time and can just start with the empty set
measured = set()
# Loop through the files that contain the xml roots
for xml_document in self._xml_roots:
line_nodes = self._get_src_path_line_nodes(xml_document,
src_path)
if line_nodes is None:
continue
# First case, need to define violations initially
if violations is None:
violations = set(
Violation(int(line.get('number')), None)
for line in line_nodes
if int(line.get('hits', 0)) == 0)
# If we already have a violations set,
# take the intersection of the new
# violations set and its old self
else:
violations = violations & set(
Violation(int(line.get('number')), None)
for line in line_nodes
if int(line.get('hits', 0)) == 0
)
# Measured is the union of itself and the new measured
measured = measured | set(
int(line.get('number')) for line in line_nodes
)
# If we don't have any information about the source file,
# don't report any violations
if violations is None:
violations = set()
self._info_cache[src_path] = (violations, measured)
def parse_reports(self, reports):
"""
Args:
reports: list[str] - output from the report
Return:
A dict[Str:Violation]
Violation is a simple named tuple Defined above
"""
violations_dict = defaultdict(list)
for report in reports:
xml_document = etree.fromstring("".join(report))
bugs = xml_document.findall(".//BugInstance")
for bug in bugs:
category = bug.get('category')
short_message = bug.find('ShortMessage').text
line = bug.find('SourceLine')
if line.get('start') is None or line.get('end') is None:
continue
start = int(line.get('start'))
end = int(line.get('end'))
for line_number in range(start, end + 1):
error_str = "{}: {}".format(category, short_message)
violation = Violation(line_number, error_str)
filename = GitPathTool.relative_path(
line.get('sourcepath'))
violations_dict[filename].append(violation)
return violations_dict
def parse_reports(self, reports):
"""
Args:
reports: list[str] - output from the report
Return:
A dict[Str:Violation]
Violation is a simple named tuple Defined above
"""
violations_dict = defaultdict(list)
for report in reports:
output_lines = report.splitlines()
for line in output_lines:
match = self.cppcheck_expression.match(line)
# Ignore any line that isn't matched
# (for example, snippets from the source code)
if match is not None:
(cppcheck_src_path, line_number, message) = match.groups()
violation = Violation(int(line_number), message)
violations_dict[cppcheck_src_path].append(violation)
return violations_dict
def parse_reports(self, reports):
"""
Args:
reports: list[str] - output from the report
Return:
A dict[Str:Violation]
Violation is a simple named tuple Defined above
"""
violations_dict = defaultdict(list)
for report in reports:
output_lines = report.split("\n")
for output_line_number, line in enumerate(output_lines):
match = self.pylint_expression.match(line)
# Ignore any line that isn't matched
# (for example, snippets from the source code)
if match is not None:
(
pylint_src_path,
line_number,
pylint_code,
function_name,
message,
) = match.groups()
if pylint_code == self.dupe_code_violation:
files_involved = self._process_dupe_code_violation(
output_lines, output_line_number, message
)
else:
files_involved = [(pylint_src_path, line_number)]
for violation in files_involved:
pylint_src_path, line_number = violation
# If we're looking for a particular source file,
# ignore any other source files.
if function_name:
error_str = "{}: {}: {}".format(
pylint_code, function_name, message
)
else:
error_str = f"{pylint_code}: {message}"
violation = Violation(int(line_number), error_str)
violations_dict[pylint_src_path].append(violation)
return violations_dict
def violations(src_path: str) -> List[Violation]:
"""Return list of violations.
Given the path to a .sql file, analyze it and return a list of
violations (i.e. formatting or style issues).
"""
linter = Linter(config=FluffConfig.from_root())
linted_path = linter.lint_path(src_path,
ignore_non_existent_files=True)
result = []
for violation in linted_path.get_violations():
try:
# Normal SQLFluff warnings
message = f"{violation.rule_code()}: {violation.description}"
except AttributeError:
# Parse errors
message = str(violation)
result.append(Violation(violation.line_no, message))
return result
def parse_reports(self, reports):
"""
Args:
reports: list[str] - output from the report
Return:
A dict[Str:Violation]
Violation is a simple named tuple Defined above
"""
violations_dict = defaultdict(list)
for report in reports:
xml_document = etree.fromstring("".join(report))
files = xml_document.findall(".//file")
for file_tree in files:
for error in file_tree.findall('error'):
line_number = error.get('line')
error_str = "{}: {}".format(error.get('severity'),
error.get('message'))
violation = Violation(int(line_number), error_str)
filename = GitPathTool.relative_path(file_tree.get('name'))
violations_dict[filename].append(violation)
return violations_dict
def violations(src_path):
"""Return list of violations.
Given the path to a .sql file, analyze it and return a list of
violations (i.e. formatting or style issues).
:param src_path:
:return: list of Violation
"""
linter = Linter(config=FluffConfig())
linted_path = linter.lint_path(src_path,
ignore_non_existent_files=True)
result = []
for violation in linted_path.get_violations():
try:
# Normal SQLFluff warnings
message = violation.description
except AttributeError:
# Parse errors
message = str(violation)
result.append(Violation(violation.line_no(), message))
return result
def parse_reports(self, reports):
"""
Args:
reports: list[str] - output from the report
Return:
A dict[Str:Violation]
Violation is a simple named tuple Defined above
"""
violations_dict = defaultdict(list)
for report in reports:
xml_document = etree.fromstring("".join(report))
node_files = xml_document.findall(".//file")
for node_file in node_files:
for error in node_file.findall("violation"):
line_number = error.get("beginline")
error_str = "{}: {}".format(error.get("rule"), error.text.strip())
violation = Violation(int(line_number), error_str)
filename = GitPathTool.relative_path(node_file.get("name"))
filename = filename.replace(os.sep, "/")
violations_dict[filename].append(violation)
return violations_dict
def violations(self, src_path):
"""Return list of violations.
Given the path to a .sql file, analyze it and return a list of
violations (i.e. formatting or style issues).
:param src_path:
:return: list of Violation
"""
linter = get_linter(get_config())
linter.output_func = None
linted_path = linter.lint_path(src_path)
result = []
for violation in linted_path.files[0].violations:
try:
# Normal SQLFluff warnings
message = violation.description
except AttributeError:
# Parse errors
message = str(violation)
result.append(Violation(violation.line_no(), message))
return result
def _cache_file(self, src_path):
"""
Load the data from `self._xml_roots`
for `src_path`, if it hasn't been already.
"""
# If we have not yet loaded this source file
if src_path not in self._info_cache:
# We only want to keep violations that show up in each xml source.
# Thus, each time, we take the intersection. However, to do this
# we must treat the first time as a special case and just add all
# the violations from the first xml report.
violations = None
# A line is measured if it is measured in any of the reports, so
# we take set union each time and can just start with the empty set
measured = set()
# Loop through the files that contain the xml roots
for xml_document in self._xml_roots:
if xml_document.findall(".[@clover]"):
# see etc/schema/clover.xsd at https://bitbucket.org/atlassian/clover/src
line_nodes = self._get_src_path_line_nodes_clover(
xml_document, src_path)
_number = "num"
_hits = "count"
elif xml_document.findall(".[@name]"):
# https://github.com/jacoco/jacoco/blob/master/org.jacoco.report/src/org/jacoco/report/xml/report.dtd
line_nodes = self._get_src_path_line_nodes_jacoco(
xml_document, src_path)
_number = "nr"
_hits = "ci"
else:
# https://github.com/cobertura/web/blob/master/htdocs/xml/coverage-04.dtd
line_nodes = self._get_src_path_line_nodes_cobertura(
xml_document, src_path)
_number = "number"
_hits = "hits"
if line_nodes is None:
continue
# First case, need to define violations initially
if violations is None:
violations = {
Violation(int(line.get(_number)), None)
for line in line_nodes if int(line.get(_hits, 0)) == 0
}
# If we already have a violations set,
# take the intersection of the new
# violations set and its old self
else:
violations = violations & {
Violation(int(line.get(_number)), None)
for line in line_nodes if int(line.get(_hits, 0)) == 0
}
# Measured is the union of itself and the new measured
measured = measured | {
int(line.get(_number))
for line in line_nodes
}
# If we don't have any information about the source file,
# don't report any violations
if violations is None:
violations = set()
self._info_cache[src_path] = (violations, measured)