pocketutils.misc.web_resource.WebResource.__is_gzip() - Code Metrics - Inspection of "fix: broaden vr ranges" - dmyersturnbull/pocketutils - Measure and Improve Code Quality continuously with Scrutinizer

Passed

Push — main ( 6e4731...702ebc )

by Douglas

created 2022-08-11 21:03 UTC

WebResource.__is_gzip() A

↳ Parent: pocketutils.misc.web_resource

Complexity

Conditions

Size

Total Lines	10
Code Lines	9

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
cc	4
eloc	9
nop	2
dl	0
loc	10
rs	9.95
c	0
b	0
f	0

import gzip

import logging
import shutil
import zipfile
from datetime import datetime
from pathlib import Path
from typing import Mapping, Optional
from urllib import request

from pocketutils.core import PathLike


logger = logging.getLogger("pocketutils")


class WebResource:
    """
    Useful for extracting files from ZIP and GZIPing them.
    """

    def __init__(self, url: str, archive_member: Optional[str], local_path: PathLike):
        self._url = url
        self._archive_member = archive_member
        self._local_path = Path(local_path)

    def download(self, redownload: bool = False):

        now = datetime.now()
        to_path = Path(self._local_path)
        if not to_path.exists() or redownload:
            extracted, dled = None, None
            try:
                logger.info(f"Downloading {self._url}...")

                tmp = str(to_path) + now.strftime("%Y%m%d-%H%M%S-%f") + ".tmp"
                dled, response = request.urlretrieve(self._url, tmp)
                dled = Path(dled)
                if self._archive_member is not None:
                    with zipfile.ZipFile(dled, "r") as zfile:
                        extracted = Path(zfile.extract(self._archive_member))
                else:
                    extracted = dled
                if to_path.suffix == ".gz" and not self.__is_gzip(extracted):
                    with extracted.open("rb") as f_in:
                        with gzip.open(to_path, "wb") as f_out:
                            shutil.copyfileobj(f_in, f_out)
                else:
                    shutil.move(extracted, to_path)
                self._info_path.write_text(
                    "url="
                    + self._url
                    + "\n"
                    + "datetime_downloaded="
                    + now.isoformat()
                    + "\n"
                    + "response="
                    + str(response).replace("\n", " |")
                    + "\n",
                    encoding="utf-8",
                )
            finally:
                if extracted is not None and extracted.exists():
                    extracted.unlink()
                if dled is not None and dled.exists():
                    dled.unlink()

    def datetime_downloaded(self) -> datetime:

        return datetime.fromisoformat(self.metadata()["datetime_downloaded"])

    def metadata(self) -> Mapping[str, str]:

        return {
            line[: line.index("=")].strip(): line[line.index("=") + 1 :].strip()
            for line in self._info_path.read_text(encoding="utf8").splitlines()
        }

    @property
    def _info_path(self) -> Path:
        return self._local_path.with_suffix(self._local_path.suffix + ".info")

    def exists(self) -> bool:

        return self._local_path.exists()

    @property
    def path(self) -> Path:

        return self._local_path

    def delete(self) -> None:

        self.path.unlink(missing_ok=True)


    def __is_gzip(self, path):
class Foo:
    def some_method(self, x, y):
        return x + y;
        try:
            with gzip.open(path, "rb") as f:

                f.read(20)  # 10-byte header
        except OSError as e:

            if "Not a gzipped file" in str(e):

                return False
            else:
                raise e
        return True


__all__ = ["WebResource"]


1			import gzip
			0 ignored issues – show introduced 2022-08-11 21:06 UTC by Report Bug Copy Issue Report Missing module docstring Loading history...
2			import logging
3			import shutil
4			import zipfile
5			from datetime import datetime
6			from pathlib import Path
7			from typing import Mapping, Optional
8			from urllib import request
9
10			from pocketutils.core import PathLike
			0 ignored issues – show Bug introduced 2021-10-28 02:10 UTC by Report Bug Copy Issue Report The name `core` does not seem to exist in module `pocketutils`. Loading history... introduced 2022-08-11 21:06 UTC by Report Bug Copy Issue Report Cannot import 'pocketutils.core' due to syntax error 'invalid syntax (<unknown>, line 134)' Loading history...
11
12			logger = logging.getLogger("pocketutils")
13
14
15			class WebResource:
16			"""
17			Useful for extracting files from ZIP and GZIPing them.
18			"""
19
20			def __init__(self, url: str, archive_member: Optional[str], local_path: PathLike):
21			self._url = url
22			self._archive_member = archive_member
23			self._local_path = Path(local_path)
24
25			def download(self, redownload: bool = False):
			0 ignored issues – show introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Missing function or method docstring Loading history...
26			now = datetime.now()
27			to_path = Path(self._local_path)
28			if not to_path.exists() or redownload:
29			extracted, dled = None, None
30			try:
31			logger.info(f"Downloading {self._url}...")
			0 ignored issues – show introduced 2021-01-25 07:52 UTC by Report Bug Copy Issue Report Use lazy % formatting in logging functions Loading history...
32			tmp = str(to_path) + now.strftime("%Y%m%d-%H%M%S-%f") + ".tmp"
33			dled, response = request.urlretrieve(self._url, tmp)
34			dled = Path(dled)
35			if self._archive_member is not None:
36			with zipfile.ZipFile(dled, "r") as zfile:
37			extracted = Path(zfile.extract(self._archive_member))
38			else:
39			extracted = dled
40			if to_path.suffix == ".gz" and not self.__is_gzip(extracted):
41			with extracted.open("rb") as f_in:
42			with gzip.open(to_path, "wb") as f_out:
43			shutil.copyfileobj(f_in, f_out)
44			else:
45			shutil.move(extracted, to_path)
46			self._info_path.write_text(
47			"url="
48			+ self._url
49			+ "\n"
50			+ "datetime_downloaded="
51			+ now.isoformat()
52			+ "\n"
53			+ "response="
54			+ str(response).replace("\n", " \|")
55			+ "\n",
56			encoding="utf-8",
57			)
58			finally:
59			if extracted is not None and extracted.exists():
60			extracted.unlink()
61			if dled is not None and dled.exists():
62			dled.unlink()
63
64			def datetime_downloaded(self) -> datetime:
			0 ignored issues – show introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Missing function or method docstring Loading history...
65			return datetime.fromisoformat(self.metadata()["datetime_downloaded"])
66
67			def metadata(self) -> Mapping[str, str]:
			0 ignored issues – show introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Missing function or method docstring Loading history...
68			return {
69			line[: line.index("=")].strip(): line[line.index("=") + 1 :].strip()
70			for line in self._info_path.read_text(encoding="utf8").splitlines()
71			}
72
73			@property
74			def _info_path(self) -> Path:
75			return self._local_path.with_suffix(self._local_path.suffix + ".info")
76
77			def exists(self) -> bool:
			0 ignored issues – show introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Missing function or method docstring Loading history...
78			return self._local_path.exists()
79
80			@property
81			def path(self) -> Path:
			0 ignored issues – show introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Missing function or method docstring Loading history...
82			return self._local_path
83
84			def delete(self) -> None:
			0 ignored issues – show introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Missing function or method docstring Loading history...
85			self.path.unlink(missing_ok=True)
			0 ignored issues – show Bug introduced 2021-10-05 01:44 UTC by Report Bug Copy Issue Report The keyword `missing_ok` does not seem to exist for the method call. Loading history...
86
87			def __is_gzip(self, path):
			0 ignored issues – show Coding Style introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report This method could be written as a function/class method. If a method does not access any attributes of the class, it could also be implemented as a function or static method. This can help improve readability. For example class Foo: def some_method(self, x, y): return x + y; could be written as class Foo: @classmethod def some_method(cls, x, y): return x + y; Loading history...
88			try:
89			with gzip.open(path, "rb") as f:
			0 ignored issues – show Coding Style Naming introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Variable name "f" doesn't conform to snake_case naming style ('([^\\W\\dA-Z][^\\WA-Z]2,\|_[^\\WA-Z]*\|__[^\\WA-Z\\d_][^\\WA-Z]+__)$' pattern) This check looks for invalid names for a range of different identifiers. You can set regular expressions to which the identifiers must conform if the defaults do not match your requirements. If your project includes a Pylint configuration file, the settings contained in that file take precedence. To find out more about Pylint, please refer to their site. Loading history...
90			f.read(20) # 10-byte header
91			except OSError as e:
			0 ignored issues – show Coding Style Naming introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Variable name "e" doesn't conform to snake_case naming style ('([^\\W\\dA-Z][^\\WA-Z]2,\|_[^\\WA-Z]*\|__[^\\WA-Z\\d_][^\\WA-Z]+__)$' pattern) This check looks for invalid names for a range of different identifiers. You can set regular expressions to which the identifiers must conform if the defaults do not match your requirements. If your project includes a Pylint configuration file, the settings contained in that file take precedence. To find out more about Pylint, please refer to their site. Loading history...
92			if "Not a gzipped file" in str(e):
			0 ignored issues – show unused-code introduced 2021-01-25 05:25 UTC by Report Bug Copy Issue Report Unnecessary "else" after "return" Loading history...
93			return False
94			else:
95			raise e
96			return True
97
98
99			__all__ = ["WebResource"]
100

dmyersturnbull / pocketutils

Push — main ( 6e4731...702ebc )

WebResource.__is_gzip() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like