Files
pysimpleini/helpers/quality_check.py
cclecle f93f0258e3 fix quality warnings
fix typing issues
split code to reduce function complexity
rename functions to make them python complient
2023-03-29 23:05:32 +01:00

248 lines
12 KiB
Python

# pyChaChaDummyProject (c) by chacha
#
# pyChaChaDummyProject is licensed under a
# Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Unported License.
#
# You should have received a copy of the license along with this
# work. If not, see <https://creativecommons.org/licenses/by-nc-sa/4.0/>.
from __future__ import annotations
from typing import TYPE_CHECKING
from contextlib import redirect_stdout
from io import StringIO
import re
import json
from enum import Enum
from contextlib import suppress
import sys
import pandas
import csv
import copy
from pylint.lint import Run as pylint_Run
import pylint_json2html
from .helper_base import helper_withresults_base
class PyLintMetricNotFound(Warning):
pass
class quality_check(helper_withresults_base):
PylintMessageList = dict()
@classmethod
def GetPylintMessageList(cls):
Messagelist = dict()
regex = r"^:([a-zA-Z-]+) \(([^\)]+)\)"
for line in cls.run_cmd([sys.executable, "-m", "pylint", "--list-msgs"], True).splitlines():
if res := re.search(regex, line.decode()):
Messagelist[res.group(1)] = res.group(2)
cls.PylintMessageList = Messagelist
@staticmethod
def TryExtractPYReportMetric(line: str, tag: str):
regex = f"^(?:\|{tag}\s*\|)(\d+)(?=\s*|)"
if res := re.search(regex, line):
return float(res.group(1))
raise PyLintMetricNotFound()
@classmethod
def do_job(cls):
print("checking code quality ...")
cls.GetPylintMessageList()
RES_all = dict()
with StringIO() as StdOutput:
JsonContent = ""
with redirect_stdout(StdOutput):
pylint_Run(
[
"--load-plugins=pylint.extensions.mccabe",
"--output-format=json,parseable",
"--disable=invalid-name,too-few-public-methods,too-many-arguments", # ignore
"--ignore=_version.py",
"--reports=y",
"--score=yes",
"--max-line-length=140",
"src." + cls.pyproject["project"]["name"],
],
exit=False,
)
with open(cls.get_result_dir() / "report.json", "w+", encoding="utf-8") as Outfile:
# hacky way of exctracting json + having overall score...
class TScanState(Enum):
TEXT_REPORT = 1
JSON_REPORT = 2
OTHER_REPORT_START = 3
OTHER_REPORT_STATISTICS = 4
OTHER_REPORT_METRICS = 5
OTHER_REPORT_DUPLICATION = 6
OTHER_REPORT_MESSAGES_CAT = 7
OTHER_REPORT_MESSAGES = 8
OTHER_REPORT_END = 99
RES_all["Statistics"] = dict()
RES_all["RawMetrics"] = dict()
RES_all["RawMetricsPercent"] = dict()
RES_all["Duplication"] = dict()
RES_all["MessagesCat"] = dict()
RES_all["Messages"] = dict()
RES_all["GlobalScore"] = -999
RES_all["NbAnalysedStatments"] = -999
RES_all["NbAnalysedLines"] = -999
ScanState = TScanState.TEXT_REPORT
for line in StdOutput.getvalue().split("\n"):
print(line)
if ScanState == TScanState.TEXT_REPORT:
# ignoring this part, we need json
if line == "[":
JsonContent += line
ScanState = TScanState.JSON_REPORT
elif line == "[]":
JsonContent += line
ScanState = TScanState.OTHER_REPORT_START
elif ScanState == TScanState.JSON_REPORT:
JsonContent += line
if line == "]":
ScanState = TScanState.OTHER_REPORT_START
elif ScanState == TScanState.OTHER_REPORT_START:
if res := re.search(r"^(\d+)(?= statements analysed.)", line):
RES_all["NbAnalysedStatments"] = float(res.group(1))
if line == "Statistics by type":
ScanState = TScanState.OTHER_REPORT_STATISTICS
elif ScanState == TScanState.OTHER_REPORT_STATISTICS:
if res := re.search(r"^(\d+)(?= lines have been analyzed)", line):
RES_all["NbAnalysedLines"] = float(res.group(1))
elif line == "Raw metrics":
ScanState = TScanState.OTHER_REPORT_METRICS
else:
with suppress(PyLintMetricNotFound):
RES_all["Statistics"]["module"] = cls.TryExtractPYReportMetric(line, "module")
with suppress(PyLintMetricNotFound):
RES_all["Statistics"]["class"] = cls.TryExtractPYReportMetric(line, "class")
with suppress(PyLintMetricNotFound):
RES_all["Statistics"]["method"] = cls.TryExtractPYReportMetric(line, "method")
with suppress(PyLintMetricNotFound):
RES_all["Statistics"]["function"] = cls.TryExtractPYReportMetric(line, "function")
elif ScanState == TScanState.OTHER_REPORT_METRICS:
if line == "Duplication":
RES_all["RawMetricsPercent"]["code"] = RES_all["RawMetrics"]["code"] / RES_all["NbAnalysedLines"]
RES_all["RawMetricsPercent"]["docstring"] = RES_all["RawMetrics"]["docstring"] / RES_all["NbAnalysedLines"]
RES_all["RawMetricsPercent"]["comment"] = RES_all["RawMetrics"]["comment"] / RES_all["NbAnalysedLines"]
RES_all["RawMetricsPercent"]["empty"] = RES_all["RawMetrics"]["empty"] / RES_all["NbAnalysedLines"]
ScanState = TScanState.OTHER_REPORT_DUPLICATION
else:
with suppress(PyLintMetricNotFound):
RES_all["RawMetrics"]["code"] = cls.TryExtractPYReportMetric(line, "code")
with suppress(PyLintMetricNotFound):
RES_all["RawMetrics"]["docstring"] = cls.TryExtractPYReportMetric(line, "docstring")
with suppress(PyLintMetricNotFound):
RES_all["RawMetrics"]["comment"] = cls.TryExtractPYReportMetric(line, "comment")
with suppress(PyLintMetricNotFound):
RES_all["RawMetrics"]["empty"] = cls.TryExtractPYReportMetric(line, "empty")
elif ScanState == TScanState.OTHER_REPORT_DUPLICATION:
if line == "Messages by category":
ScanState = TScanState.OTHER_REPORT_MESSAGES_CAT
else:
with suppress(PyLintMetricNotFound):
RES_all["Duplication"]["NbDupLines"] = cls.TryExtractPYReportMetric(line, "nb duplicated lines")
with suppress(PyLintMetricNotFound):
RES_all["Duplication"]["PersentDuplicatedLines"] = cls.TryExtractPYReportMetric(
line, "percent duplicated lines"
)
elif ScanState == TScanState.OTHER_REPORT_MESSAGES_CAT:
if line == "Messages":
ScanState = TScanState.OTHER_REPORT_MESSAGES
else:
with suppress(PyLintMetricNotFound):
RES_all["MessagesCat"]["Convention"] = cls.TryExtractPYReportMetric(line, "convention")
with suppress(PyLintMetricNotFound):
RES_all["MessagesCat"]["Refactor"] = cls.TryExtractPYReportMetric(line, "refactor")
with suppress(PyLintMetricNotFound):
RES_all["MessagesCat"]["Warning"] = cls.TryExtractPYReportMetric(line, "warning")
with suppress(PyLintMetricNotFound):
RES_all["MessagesCat"]["Error"] = cls.TryExtractPYReportMetric(line, "error")
elif ScanState == TScanState.OTHER_REPORT_MESSAGES:
# approx match because the number of '-' depend on screen width..
if line.startswith("--------"):
ScanState = TScanState.OTHER_REPORT_END
else:
for PylintMessage in cls.PylintMessageList.keys():
with suppress(PyLintMetricNotFound):
RES_all["Messages"][PylintMessage] = cls.TryExtractPYReportMetric(line, PylintMessage)
elif ScanState == TScanState.OTHER_REPORT_END:
if res := re.search(r"(?<=Your code has been rated at )(\d+(?:\.\d+)?)/10", line):
RES_all["GlobalScore"] = float(res.group(1))
print(RES_all["GlobalScore"])
else:
raise RuntimeError("Invalid ScanState")
Outfile.write(JsonContent)
with open(cls.get_result_dir() / "metrics.json", "w") as json_file:
json.dump(RES_all, json_file)
# exporting all Data in one csv, unused atm because jenkins seems not able to select columns from csv an keep displaying all...
# => to export a working full csv we need to a 'flat' dict (no more nested dict)
RES_all_trim = copy.deepcopy(RES_all)
del RES_all_trim["Messages"]
flat_RES_all = pandas.json_normalize(RES_all_trim, sep="_").to_dict(orient="records")[0]
with open(cls.get_result_dir() / "metrics.csv", "w", newline="") as csv_file:
writer = csv.DictWriter(csv_file, fieldnames=flat_RES_all.keys())
writer.writeheader()
writer.writerow(flat_RES_all)
# splited csv exports for jenkins plots: RawMetricsPercent
RES_all_percent = RES_all["RawMetricsPercent"]
with open(cls.get_result_dir() / "metrics_rawpercent.csv", "w", newline="") as csv_file:
writer = csv.DictWriter(csv_file, fieldnames=RES_all_percent.keys())
writer.writeheader()
writer.writerow(RES_all_percent)
# splited csv exports for jenkins plots: Statistics + Duplication + NbAnalysedStatments + NbAnalysedLines
RES_all_stats = copy.deepcopy(RES_all["Statistics"])
RES_all_stats["NbDupLines"] = RES_all["Duplication"]["NbDupLines"]
RES_all_stats["PersentDuplicatedLines"] = RES_all["Duplication"]["PersentDuplicatedLines"]
RES_all_stats["NbAnalysedStatments"] = RES_all["NbAnalysedStatments"]
RES_all_stats["NbAnalysedLines"] = RES_all["NbAnalysedLines"]
with open(cls.get_result_dir() / "metrics_Statistics.csv", "w", newline="") as csv_file:
writer = csv.DictWriter(csv_file, fieldnames=RES_all_stats.keys())
writer.writeheader()
writer.writerow(RES_all_stats)
# splited csv exports for jenkins plots: Statistics + Duplication
RES_all_MessagesCat = RES_all["MessagesCat"]
with open(cls.get_result_dir() / "metrics_MessagesCat.csv", "w", newline="") as csv_file:
writer = csv.DictWriter(csv_file, fieldnames=RES_all_MessagesCat.keys())
writer.writeheader()
writer.writerow(RES_all_MessagesCat)
# splited csv exports for jenkins plots: GlobalScore
RES_GlobalScore = {"GlobalScore": RES_all["GlobalScore"]}
with open(cls.get_result_dir() / "metrics_GlobalScore.csv", "w", newline="") as csv_file:
writer = csv.DictWriter(csv_file, fieldnames=RES_GlobalScore.keys())
writer.writeheader()
writer.writerow(RES_GlobalScore)
# converting the report using pylint_json2html (/!\ internal API, but as their is no leading '_' ...)
with open(cls.get_result_dir() / "report.html", "w+", encoding="utf-8") as Outfile:
raw_data = json.loads(JsonContent)
report = pylint_json2html.Report(raw_data)
Outfile.write(report.render())
print("Done")