coalib.bearlib.abstractions.Lint._process_corrected() - Code Metrics - Inspection of "Remove CorrectionBasedBear" - coala-analyzer/coala - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Pull Request — master (#1404)

by Abdeali

created 2016-02-11 13:56 UTC

_process_corrected() A

↳ Parent: coalib.bearlib.abstractions.Lint

Complexity

Conditions

Size

Total Lines

Duplication

Lines	0
Ratio	0 %

Metric	Value
cc	2
dl	0
loc	7
rs	9.4285

import shutil
import subprocess
import tempfile
import re
import sys

from coalib.misc.Shell import escape_path_argument
from coalib.results.Diff import Diff
from coalib.results.Result import Result
from coalib.results.RESULT_SEVERITY import RESULT_SEVERITY
from coalib.bears.Bear import Bear


def is_binary_present(cls):
    """
    Checks whether the needed binary is present.

    The function is intended be used with classes
    having an executable member which will be checked.

    :return: True if binary is present, or is not required.
             not True otherwise, with a string containing a
             detailed description of what's missing.
    """
    try:
        if cls.executable is None:
            return True
        if shutil.which(cls.executable) is None:
            return repr(cls.executable) + " is not installed."
        else:
            return True
    except AttributeError:
        # Happens when `executable` does not exist in `cls`.
        return True


class Lint(Bear):
    """
    :param executable:      The executable to run the linter.
    :param arguments:       The arguments to supply to the linter, such
                            that the file name to be analyzed can be
                            appended to the end.
    :param output_regex:    The regex which will match the output of the linter
                            to get results. This regex should give out the
                            following variables:
                             line - The line where the issue starts.
                             column - The column where the issue starts.
                             end_line - The line where the issue ends.
                             end_column - The column where the issue ends.
                             severity - The severity of the issue.
                             message - The message of the result.
                            This is not used if `gives_corrected` is set.
    :param diff_severity:   The severity to use for all results if
                            `gives_corrected` is set.
    :param diff_message:    The message to use for all results if
                            `gives_corrected` is set.
    :param use_stderr:      Uses stderr as the output stream is it's True.
    :param use_stdin:       Sends file as stdin instead of giving the file name.
    :param gives_corrected: True if the executable gives the corrected file
                            or just the issues.
    :param severity_map:    A dict where the keys are the possible severity
                            values the Linter gives out and the values are the
                            severity of the coala Result to set it to. If it is
                            not a dict, it is ignored.
    """
    check_prerequisites = classmethod(is_binary_present)
    executable = None
    arguments = ""
    output_regex = re.compile(r'(?P<line>\d+)\.(?P<column>\d+)\|'
                              r'(?P<severity>\d+): (?P<message>.*)')
    diff_message = 'No result message was set'
    diff_severity = RESULT_SEVERITY.NORMAL
    use_stderr = False
    use_stdin = False
    gives_corrected = False
    severity_map = None

    def lint(self, filename=None, file=None):
        """
        Takes a file and lints it using the linter variables defined apriori.

        :param filename:  The name of the file to execute.
        :param file:      The contents of the file as a list of strings.
        """
        assert ((self.use_stdin and file is not None) or
                (not self.use_stdin and filename is not None))

        stdin_file = tempfile.TemporaryFile()
        stdout_file = tempfile.TemporaryFile()
        stderr_file = tempfile.TemporaryFile()

        command = self._create_command(filename)

        if self.use_stdin:
            self._write(file, stdin_file, sys.stdin.encoding)
        process = subprocess.Popen(command,
                                   shell=True,
                                   stdin=stdin_file,
                                   stdout=stdout_file,
                                   stderr=stderr_file,
                                   universal_newlines=True)
        process.wait()
        stdout_output = self._read(stdout_file, sys.stdout.encoding)
        stderr_output = self._read(stderr_file, sys.stderr.encoding)
        results_output = stderr_output if self.use_stderr else stdout_output
        results = self.process_output(results_output, filename, file)
        if not self.use_stderr:
            self.__print_errors(stderr_output)

        return results

    def process_output(self, output, filename, file):
        if self.gives_corrected:
            return self._process_corrected(output, filename, file)
        else:
            return self._process_issues(output, filename)

    def _process_corrected(self, output, filename, file):
        for diff in self.__yield_diffs(file, output):
            yield Result(self,
                         self.diff_message,
                         affected_code=(diff.range(filename),),
                         diffs={filename: diff},
                         severity=self.diff_severity)

    def _process_issues(self, output, filename):
        regex = self.output_regex
        if isinstance(regex, str):
            regex = regex % {"file_name": filename}

        # Note: We join `output` because the regex may want to capture
        #       multiple lines also.
        for match in re.finditer(regex, "".join(output)):
            yield self.match_to_result(match, filename)

    def _get_groupdict(self, match):
        groups = match.groupdict()
        if (
                isinstance(self.severity_map, dict) and
                "severity" in groups and
                groups["severity"] in self.severity_map):
            groups["severity"] = self.severity_map[groups["severity"]]
        return groups

    def _create_command(self, filename):
        command = self.executable + ' ' + self.arguments
        if not self.use_stdin:
            command += ' ' + escape_path_argument(filename)
        return command

    @staticmethod
    def _read(file, encoding):
        file.seek(0)
        return [line.decode(encoding or 'UTF-8', errors="replace")
                for line in file.readlines()]

    @staticmethod
    def _write(file_contents, file, encoding):
        file.write(bytes("".join(file_contents), encoding or 'UTF-8'))
        file.seek(0)

    def __print_errors(self, errors):
        for line in filter(lambda error: bool(error.strip()), errors):
            self.warn(line)

    @staticmethod
    def __yield_diffs(file, new_file):
        if new_file != file:
            wholediff = Diff.from_string_arrays(file, new_file)

            for diff in wholediff.split_diff():
                yield diff

    def match_to_result(self, match, filename):
        """
        Converts a regex match's groups into a result.

        :param match:    The match got from regex parsing.
        :param filename: The name of the file from which this match is got.
        """
        groups = self._get_groupdict(match)

        # Pre process the groups
        for variable in ("line", "column", "end_line", "end_column"):
            if variable in groups and groups[variable]:
                groups[variable] = int(groups[variable])

        return Result.from_values(
            origin=groups.get("origin", self),
            message=groups.get("message", ""),
            file=filename,
            severity=int(groups.get("severity", RESULT_SEVERITY.NORMAL)),
            line=groups.get("line", None),
            column=groups.get("column", None),
            end_line=groups.get("end_line", None),
            end_column=groups.get("end_column", None))


1			import shutil
2			import subprocess
3			import tempfile
4			import re
5			import sys
6
7			from coalib.misc.Shell import escape_path_argument
8			from coalib.results.Diff import Diff
9			from coalib.results.Result import Result
10			from coalib.results.RESULT_SEVERITY import RESULT_SEVERITY
11			from coalib.bears.Bear import Bear
12
13
14			def is_binary_present(cls):
15			"""
16			Checks whether the needed binary is present.
17
18			The function is intended be used with classes
19			having an executable member which will be checked.
20
21			:return: True if binary is present, or is not required.
22			not True otherwise, with a string containing a
23			detailed description of what's missing.
24			"""
25			try:
26			if cls.executable is None:
27			return True
28			if shutil.which(cls.executable) is None:
29			return repr(cls.executable) + " is not installed."
30			else:
31			return True
32			except AttributeError:
33			# Happens when `executable` does not exist in `cls`.
34			return True
35
36
37			class Lint(Bear):
38			"""
39			:param executable: The executable to run the linter.
40			:param arguments: The arguments to supply to the linter, such
41			that the file name to be analyzed can be
42			appended to the end.
43			:param output_regex: The regex which will match the output of the linter
44			to get results. This regex should give out the
45			following variables:
46			line - The line where the issue starts.
47			column - The column where the issue starts.
48			end_line - The line where the issue ends.
49			end_column - The column where the issue ends.
50			severity - The severity of the issue.
51			message - The message of the result.
52			This is not used if `gives_corrected` is set.
53			:param diff_severity: The severity to use for all results if
54			`gives_corrected` is set.
55			:param diff_message: The message to use for all results if
56			`gives_corrected` is set.
57			:param use_stderr: Uses stderr as the output stream is it's True.
58			:param use_stdin: Sends file as stdin instead of giving the file name.
59			:param gives_corrected: True if the executable gives the corrected file
60			or just the issues.
61			:param severity_map: A dict where the keys are the possible severity
62			values the Linter gives out and the values are the
63			severity of the coala Result to set it to. If it is
64			not a dict, it is ignored.
65			"""
66			check_prerequisites = classmethod(is_binary_present)
67			executable = None
68			arguments = ""
69			output_regex = re.compile(r'(?P<line>\d+)\.(?P<column>\d+)\\|'
70			r'(?P<severity>\d+): (?P<message>.*)')
71			diff_message = 'No result message was set'
72			diff_severity = RESULT_SEVERITY.NORMAL
73			use_stderr = False
74			use_stdin = False
75			gives_corrected = False
76			severity_map = None
77
78			def lint(self, filename=None, file=None):
79			"""
80			Takes a file and lints it using the linter variables defined apriori.
81
82			:param filename: The name of the file to execute.
83			:param file: The contents of the file as a list of strings.
84			"""
85			assert ((self.use_stdin and file is not None) or
86			(not self.use_stdin and filename is not None))
87
88			stdin_file = tempfile.TemporaryFile()
89			stdout_file = tempfile.TemporaryFile()
90			stderr_file = tempfile.TemporaryFile()
91
92			command = self._create_command(filename)
93
94			if self.use_stdin:
95			self._write(file, stdin_file, sys.stdin.encoding)
96			process = subprocess.Popen(command,
97			shell=True,
98			stdin=stdin_file,
99			stdout=stdout_file,
100			stderr=stderr_file,
101			universal_newlines=True)
102			process.wait()
103			stdout_output = self._read(stdout_file, sys.stdout.encoding)
104			stderr_output = self._read(stderr_file, sys.stderr.encoding)
105			results_output = stderr_output if self.use_stderr else stdout_output
106			results = self.process_output(results_output, filename, file)
107			if not self.use_stderr:
108			self.__print_errors(stderr_output)
109
110			return results
111
112			def process_output(self, output, filename, file):
113			if self.gives_corrected:
114			return self._process_corrected(output, filename, file)
115			else:
116			return self._process_issues(output, filename)
117
118			def _process_corrected(self, output, filename, file):
119			for diff in self.__yield_diffs(file, output):
120			yield Result(self,
121			self.diff_message,
122			affected_code=(diff.range(filename),),
123			diffs={filename: diff},
124			severity=self.diff_severity)
125
126			def _process_issues(self, output, filename):
127			regex = self.output_regex
128			if isinstance(regex, str):
129			regex = regex % {"file_name": filename}
130
131			# Note: We join `output` because the regex may want to capture
132			# multiple lines also.
133			for match in re.finditer(regex, "".join(output)):
134			yield self.match_to_result(match, filename)
135
136			def _get_groupdict(self, match):
137			groups = match.groupdict()
138			if (
139			isinstance(self.severity_map, dict) and
140			"severity" in groups and
141			groups["severity"] in self.severity_map):
142			groups["severity"] = self.severity_map[groups["severity"]]
143			return groups
144
145			def _create_command(self, filename):
146			command = self.executable + ' ' + self.arguments
147			if not self.use_stdin:
148			command += ' ' + escape_path_argument(filename)
149			return command
150
151			@staticmethod
152			def _read(file, encoding):
153			file.seek(0)
154			return [line.decode(encoding or 'UTF-8', errors="replace")
155			for line in file.readlines()]
156
157			@staticmethod
158			def _write(file_contents, file, encoding):
159			file.write(bytes("".join(file_contents), encoding or 'UTF-8'))
160			file.seek(0)
161
162			def __print_errors(self, errors):
163			for line in filter(lambda error: bool(error.strip()), errors):
164			self.warn(line)
165
166			@staticmethod
167			def __yield_diffs(file, new_file):
168			if new_file != file:
169			wholediff = Diff.from_string_arrays(file, new_file)
170
171			for diff in wholediff.split_diff():
172			yield diff
173
174			def match_to_result(self, match, filename):
175			"""
176			Converts a regex match's groups into a result.
177
178			:param match: The match got from regex parsing.
179			:param filename: The name of the file from which this match is got.
180			"""
181			groups = self._get_groupdict(match)
182
183			# Pre process the groups
184			for variable in ("line", "column", "end_line", "end_column"):
185			if variable in groups and groups[variable]:
186			groups[variable] = int(groups[variable])
187
188			return Result.from_values(
189			origin=groups.get("origin", self),
190			message=groups.get("message", ""),
191			file=filename,
192			severity=int(groups.get("severity", RESULT_SEVERITY.NORMAL)),
193			line=groups.get("line", None),
194			column=groups.get("column", None),
195			end_line=groups.get("end_line", None),
196			end_column=groups.get("end_column", None))
197

coala-analyzer / coala

Pull Request — master (#1404)

_process_corrected() A

Complexity

Size

Duplication

Duplication Side-by-Side

Filter issues like