coalib.bearlib.abstractions.Lint._read() - 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 12:25 UTC

coalib.bearlib.abstractions.Lint._read() A

↳ Parent: coalib.bearlib.abstractions.Lint

Complexity

Conditions

Size

Total Lines

Duplication

Lines	0
Ratio	0 %

Metric	Value
cc	3
dl	0
loc	5
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))

        self._create_std()
        command = self._create_command(filename)

        if self.use_stdin:
            self._write_stdin(file)
        process = subprocess.Popen(command,
                                   shell=True,
                                   stdin=self.stdin_file,
                                   stdout=self.stdout_file,
                                   stderr=self.stderr_file,
                                   universal_newlines=True)
        process.wait()
        stdout_output = self._read(sys.sdout)
        stderr_output = self._read(sys.stderr)
        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)
        self._close_std()

        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

    def _create_std(self):
        self.stdin_file = tempfile.TemporaryFile()
        self.stdout_file = tempfile.TemporaryFile()
        self.stderr_file = tempfile.TemporaryFile()

    def _close_std(self):
        self.stdin_file.close()
        self.stdout_file.close()
        self.stderr_file.close()

    def _read(self, std):
        file = self.stdout_file if std == sys.stdout else self.stderr_file
        file.seek(0)
        return (line.decode(std.encoding, errors="replace")
                for line in file.readlines())

    def _write_stdin(self, file):
        self.stdin_file.write(bytes("".join(file), 'UTF-8'))
        self.stdin_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)
86			or (not self.use_stdin and filename is not None))
87
88			self._create_std()
89			command = self._create_command(filename)
90
91			if self.use_stdin:
92			self._write_stdin(file)
93			process = subprocess.Popen(command,
94			shell=True,
95			stdin=self.stdin_file,
96			stdout=self.stdout_file,
97			stderr=self.stderr_file,
98			universal_newlines=True)
99			process.wait()
100			stdout_output = self._read(sys.sdout)
101			stderr_output = self._read(sys.stderr)
102			results_output = stderr_output if self.use_stderr else stdout_output
103			results = self.process_output(results_output, filename, file)
104			if not self.use_stderr:
105			self.__print_errors(stderr_output)
106			self._close_std()
107
108			return results
109
110			def process_output(self, output, filename, file):
111			if self.gives_corrected:
112			return self._process_corrected(output, filename, file)
113			else:
114			return self._process_issues(output, filename)
115
116			def _process_corrected(self, output, filename, file):
117			for diff in self.__yield_diffs(file, output):
118			yield Result(self,
119			self.diff_message,
120			affected_code=(diff.range(filename),),
121			diffs={filename: diff},
122			severity=self.diff_severity)
123
124			def _process_issues(self, output, filename):
125			regex = self.output_regex
126			if isinstance(regex, str):
127			regex = regex % {"file_name": filename}
128
129			# Note: We join `output` because the regex may want to capture
130			# multiple lines also.
131			for match in re.finditer(regex, "".join(output)):
132			yield self.match_to_result(match, filename)
133
134			def _get_groupdict(self, match):
135			groups = match.groupdict()
136			if (
137			isinstance(self.severity_map, dict) and
138			"severity" in groups and
139			groups["severity"] in self.severity_map):
140			groups["severity"] = self.severity_map[groups["severity"]]
141			return groups
142
143			def _create_command(self, filename):
144			command = self.executable + ' ' + self.arguments
145			if not self.use_stdin:
146			command += ' ' + escape_path_argument(filename)
147			return command
148
149			def _create_std(self):
150			self.stdin_file = tempfile.TemporaryFile()
151			self.stdout_file = tempfile.TemporaryFile()
152			self.stderr_file = tempfile.TemporaryFile()
153
154			def _close_std(self):
155			self.stdin_file.close()
156			self.stdout_file.close()
157			self.stderr_file.close()
158
159			def _read(self, std):
160			file = self.stdout_file if std == sys.stdout else self.stderr_file
161			file.seek(0)
162			return (line.decode(std.encoding, errors="replace")
163			for line in file.readlines())
164
165			def _write_stdin(self, file):
166			self.stdin_file.write(bytes("".join(file), 'UTF-8'))
167			self.stdin_file.seek(0)
168
169			def __print_errors(self, errors):
170			for line in filter(lambda error: bool(error.strip()), errors):
171			self.warn(line)
172
173			@staticmethod
174			def __yield_diffs(file, new_file):
175			if new_file != file:
176			wholediff = Diff.from_string_arrays(file, new_file)
177
178			for diff in wholediff.split_diff():
179			yield diff
180
181			def match_to_result(self, match, filename):
182			"""
183			Converts a regex match's groups into a result.
184
185			:param match: The match got from regex parsing.
186			:param filename: The name of the file from which this match is got.
187			"""
188			groups = self._get_groupdict(match)
189
190			# Pre process the groups
191			for variable in ("line", "column", "end_line", "end_column"):
192			if variable in groups and groups[variable]:
193			groups[variable] = int(groups[variable])
194
195			return Result.from_values(
196			origin=groups.get("origin", self),
197			message=groups.get("message", ""),
198			file=filename,
199			severity=int(groups.get("severity", RESULT_SEVERITY.NORMAL)),
200			line=groups.get("line", None),
201			column=groups.get("column", None),
202			end_line=groups.get("end_line", None),
203			end_column=groups.get("end_column", None))
204

coala-analyzer / coala

Pull Request — master (#1404)

coalib.bearlib.abstractions.Lint._read() A

Complexity

Size

Duplication

Duplication Side-by-Side

Filter issues like