[build] Fix typo in devscripts/fish-completion.py (fix 2285605)

* mention pynose
* mention traverse_obj and add/revise examples

[skip ci]

.github/workflows/ci.yml

@@ -1,22 +1,34 @@
 name: CI
 
 env:
-  # add 3.10+ after patching nose (https://github.com/nose-devs/nose/issues/1099)
-  # or switching to fork of https://github.com/mdmintz/pynose
-  all-cpython-versions: 2.6, 2.7, 3.2, 3.3, 3.4, 3.5, 3.6, 3.7, 3.8, 3.9
-  main-cpython-versions: 2.7, 3.2, 3.5, 3.9
+  all-cpython-versions: 2.6, 2.7, 3.2, 3.3, 3.4, 3.5, 3.6, 3.7, 3.8, 3.9, 3.10, 3.11, 3.12
+  main-cpython-versions: 2.7, 3.2, 3.5, 3.9, 3.11
   pypy-versions: pypy-2.7, pypy-3.6, pypy-3.7
   cpython-versions: main
-  test-set: both
+  test-set: core
 
 on:
   push:
+    inputs:
+      cpython-versions:
+        type: string
+        default: all
+      test-set:
+        type: string
+        default: core
   pull_request:
+    inputs:
+      cpython-versions:
+        type: string
+        default: main
+      test-set:
+        type: string
+        default: both
   workflow_dispatch:
     inputs:
       cpython-versions:
         type: choice
-        description: CPython versions (main = 2.7, 3.2, 3.5, 3.9)
+        description: CPython versions (main = 2.7, 3.2, 3.5, 3.9, 3.11)
         options:
           - all
           - main
@@ -30,7 +42,7 @@ on:
           - core
           - download
         required: true
-        default: core
+        default: both
 
 permissions:
   contents: read
@@ -44,7 +56,8 @@ jobs:
       test-set: ${{ steps.run.outputs.test-set }}
       own-pip-versions: ${{ steps.run.outputs.own-pip-versions }}
     steps:
-    - id: run
+    - name: Make version array
+      id: run
       run: |
         # Make a JSON Array from comma/space-separated string (no extra escaping)
         json_list() { \
@@ -66,7 +79,6 @@ jobs:
         # versions with a special get-pip.py in a per-version subdirectory
         printf 'own-pip-versions=%s\n' \
           "$(json_list 2.6, 2.7, 3.2, 3.3, 3.4, 3.5, 3.6)" >> "$GITHUB_OUTPUT"
-
   tests:
     name: Run tests
     needs: select
@@ -82,28 +94,29 @@ jobs:
       fail-fast: true
       matrix:
         os: [ubuntu-20.04]
-        # outside steps, use github.env...., not env....
         python-version: ${{ fromJSON(needs.select.outputs.cpython-versions) }}
         python-impl: [cpython]
         ytdl-test-set: ${{ fromJSON(needs.select.outputs.test-set) }}
         run-tests-ext: [sh]
         include:
         - os: windows-2019
-          python-version: 3.2
+          python-version: 3.4
           python-impl: cpython
           ytdl-test-set: ${{ contains(needs.select.outputs.test-set, 'core') && 'core' || 'nocore' }}
           run-tests-ext: bat
         - os: windows-2019
-          python-version: 3.2
+          python-version: 3.4
           python-impl: cpython
           ytdl-test-set: ${{ contains(needs.select.outputs.test-set, 'download') && 'download'  || 'nodownload' }}
           run-tests-ext: bat
         # jython
         - os: ubuntu-20.04
+          python-version: 2.7
           python-impl: jython
           ytdl-test-set: ${{ contains(needs.select.outputs.test-set, 'core') && 'core' || 'nocore' }}
           run-tests-ext: sh
         - os: ubuntu-20.04
+          python-version: 2.7
           python-impl: jython
           ytdl-test-set: ${{ contains(needs.select.outputs.test-set, 'download') && 'download'  || 'nodownload' }}
           run-tests-ext: sh
@@ -113,7 +126,7 @@ jobs:
     #-------- Python 3 -----
     - name: Set up supported Python ${{ matrix.python-version }}
       id: setup-python
-      if: ${{ matrix.python-impl == 'cpython' && matrix.python-version != '2.6' && matrix.python-version != '2.7'}}
+      if: ${{ matrix.python-impl == 'cpython' && matrix.python-version != '2.6' && matrix.python-version != '2.7' && matrix.python-version != '3.12'}}
       # wrap broken actions/setup-python@v4
       uses: ytdl-org/setup-python@v1
       with:
@@ -151,9 +164,45 @@ jobs:
             'import sys' \
             'print(sys.path)' \
             | ${expected} -
+    #-------- Python 3.12 -
+    - name: Set up CPython 3.12 environment
+      if: ${{ matrix.python-impl == 'cpython' && matrix.python-version == '3.12' }}
+      shell: bash
+      run: |
+        PYENV_ROOT=$HOME/.local/share/pyenv
+        echo "PYENV_ROOT=${PYENV_ROOT}" >> "$GITHUB_ENV"
+    - name: Cache Python 3.12
+      id: cache312
+      if: ${{ matrix.python-impl == 'cpython' && matrix.python-version == '3.12' }}
+      uses: actions/cache@v3
+      with:
+        key: python-3.12
+        path: |
+          ${{ env.PYENV_ROOT }}
+    - name: Build and set up Python 3.12
+      if: ${{ matrix.python-impl == 'cpython' && matrix.python-version == '3.12' && ! steps.cache312.outputs.cache-hit }}
+      # dl and build locally
+      shell: bash
+      run: |
+        # Install build environment
+        sudo apt-get install -y build-essential llvm libssl-dev tk-dev  \
+                      libncursesw5-dev libreadline-dev libsqlite3-dev   \
+                      libffi-dev xz-utils zlib1g-dev libbz2-dev liblzma-dev
+        # Download PyEnv from its GitHub repository.
+        export PYENV_ROOT=${{ env.PYENV_ROOT }}
+        export PATH=$PYENV_ROOT/bin:$PATH
+        git clone "https://github.com/pyenv/pyenv.git" "$PYENV_ROOT"
+        pyenv install 3.12.0b4
+    - name: Locate Python 3.12
+      if: ${{ matrix.python-impl == 'cpython' && matrix.python-version == '3.12' }}
+      shell: bash
+      run: |
+        PYTHONHOME="${{ env.PYENV_ROOT }}/versions/3.12.0b4"
+        echo "PYTHONHOME=$PYTHONHOME" >> "$GITHUB_ENV"
+        echo "PATH=${PYTHONHOME}/bin:$PATH" >> "$GITHUB_ENV"
     #-------- Python 2.7 --
     - name: Set up Python 2.7
-      if: ${{ matrix.python-version == '2.7' }}
+      if: ${{ matrix.python-impl == 'cpython' && matrix.python-version == '2.7' }}
       # install 2.7
       shell: bash
       run: |
@@ -161,7 +210,7 @@ jobs:
         echo "PYTHONHOME=/usr" >> "$GITHUB_ENV"
     #-------- Python 2.6 --
     - name: Set up Python 2.6 environment
-      if: ${{ matrix.python-version == '2.6' }}
+      if: ${{ matrix.python-impl == 'cpython' && matrix.python-version == '2.6' }}
       shell: bash
       run: |
         openssl_name=openssl-1.0.2u
@@ -181,7 +230,7 @@ jobs:
           ${{ env.openssl_dir }}
           ${{ env.PYENV_ROOT }}
     - name: Build and set up Python 2.6
-      if: ${{ matrix.python-version == '2.6' && ! steps.cache26.outputs.cache-hit }}
+      if: ${{ matrix.python-impl == 'cpython' && matrix.python-version == '2.6' && ! steps.cache26.outputs.cache-hit }}
       # dl and build locally
       shell: bash
       run: |
@@ -205,24 +254,21 @@ jobs:
           make install )
         rm -rf $openssl_name
         rmdir $openssl_ssl/certs && ln -s /etc/ssl/certs $openssl_ssl/certs
-
         # Download PyEnv from its GitHub repository.
         export PYENV_ROOT=${{ env.PYENV_ROOT }}
         export PATH=$PYENV_ROOT/bin:$PATH
         git clone "https://github.com/pyenv/pyenv.git" "$PYENV_ROOT"
-
         # Prevent pyenv build trying (and failing) to update pip
         export GET_PIP=get-pip-2.6.py
         echo 'import sys; sys.exit(0)' > ${GET_PIP}
         GET_PIP=$(realpath $GET_PIP)
-
         # Build and install Python
         export CFLAGS="-I$openssl_inc"
         export LDFLAGS="-L$openssl_lib"
         export LD_LIBRARY_PATH="$openssl_lib"
         pyenv install 2.6.9
     - name: Locate Python 2.6
-      if: ${{ matrix.python-version == '2.6' }}
+      if: ${{ matrix.python-impl == 'cpython' && matrix.python-version == '2.6' }}
       shell: bash
       run: |
         PYTHONHOME="${{ env.PYENV_ROOT }}/versions/2.6.9"
@@ -244,7 +290,7 @@ jobs:
         echo "PIP=pip" >> "$GITHUB_ENV"
     - name: Cache Jython
       id: cachejy
-      if: ${{ matrix.python-impl == 'jython' }}
+      if: ${{ matrix.python-impl == 'jython' && matrix.python-version == '2.7' }}
       uses: actions/cache@v3
       with:
         # 2.7.3 now available, may solve SNI issue
@@ -252,7 +298,7 @@ jobs:
         path: |
           ${{ env.JYTHON_ROOT }}
     - name: Install Jython
-      if: ${{ matrix.python-impl == 'jython' && ! steps.cachejy.outputs.cache-hit }}
+      if: ${{ matrix.python-impl == 'jython' && matrix.python-version == '2.7' && ! steps.cachejy.outputs.cache-hit }}
       shell: bash
       run: |
         JYTHON_ROOT="${{ env.JYTHON_ROOT }}"
@@ -265,6 +311,11 @@ jobs:
       run: |
         JYTHON_ROOT="${{ env.JYTHON_ROOT }}"
         echo "${JYTHON_ROOT}/bin" >> $GITHUB_PATH
+    - name: Install supporting Python 2.7 if possible
+      if: ${{ steps.cachejy.outputs.cache-hit }}
+      shell: bash
+      run: |
+        sudo apt-get install -y python2.7 || true
     #-------- pip ---------
     - name: Set up supported Python ${{ matrix.python-version }} pip
       if: ${{ (matrix.python-version != '3.2' && steps.setup-python.outputs.python-path) || matrix.python-version == '2.7' }}
@@ -317,12 +368,17 @@ jobs:
         done
     #-------- nose --------
     - name: Install nose for Python ${{ matrix.python-version }}
-      if: ${{ (matrix.python-version != '3.2' && steps.setup-python.outputs.python-path) || matrix.python-version == '2.7' }}
+      if: ${{ (matrix.python-version != '3.2' && steps.setup-python.outputs.python-path) || matrix.python-version == '2.7' || matrix.python-version == '3.12' }}
       shell: bash
       run: |
         echo "$PATH"
         echo "$PYTHONHOME"
-        $PIP -qq show nose || $PIP install nose
+        # Use PyNose for recent Pythons instead of Nose
+        py3ver="${{ matrix.python-version }}"
+        py3ver=${py3ver#3.}
+        [ "$py3ver" != "${{ matrix.python-version }}" ] && py3ver=${py3ver%.*} || py3ver=0
+        [ "$py3ver" -ge 9 ] && nose=pynose || nose=nose
+        $PIP -qq show $nose || $PIP install $nose
     - name: Install nose for other Python 2
       if: ${{ matrix.python-impl == 'jython' || matrix.python-version == '2.6' }}
       shell: bash
@@ -342,6 +398,11 @@ jobs:
       if: ${{ contains(needs.select.outputs.test-set, matrix.ytdl-test-set ) }}
       shell: bash
       run: |
+        # set PYTHON_VER
+        PYTHON_VER=${{ matrix.python-version }}
+        [ "${PYTHON_VER#*-}" != "$PYTHON_VER" ] || PYTHON_VER="${{ matrix.python-impl }}-${PYTHON_VER}"
+        echo "PYTHON_VER=$PYTHON_VER" >> "$GITHUB_ENV"
+        echo "PYTHON_IMPL=${{ matrix.python-impl }}" >> "$GITHUB_ENV"
         # define a test to validate the Python version used by nosetests
         printf '%s\n' \
           'from __future__ import unicode_literals' \
@@ -354,9 +415,9 @@ jobs:
           '    def setUp(self):' \
           '        self.ver = os.environ["PYTHON_VER"].split("-")' \
           '    def test_python_ver(self):' \
-          '        self.assertEqual(sys.version[:3], self.ver[-1])' \
+          '        self.assertEqual(["%d" % v for v in sys.version_info[:2]], self.ver[-1].split(".")[:2])' \
           '        self.assertTrue(sys.version.startswith(self.ver[-1]))' \
-          '        self.assertIn(self.ver[0], sys.version.lower())' \
+          '        self.assertIn(self.ver[0], ",".join((sys.version, platform.python_implementation())).lower())' \
           '    def test_python_impl(self):' \
           '        self.assertIn(platform.python_implementation().lower(), (os.environ["PYTHON_IMPL"], self.ver[0]))' \
           > test/test_python.py
@@ -366,11 +427,8 @@ jobs:
       continue-on-error: ${{ matrix.ytdl-test-set == 'download' || matrix.python-impl == 'jython' }}
       env:
         YTDL_TEST_SET: ${{ matrix.ytdl-test-set }}
-        PYTHON_VER: ${{ matrix.python-version }}
-        PYTHON_IMPL: ${{ matrix.python-impl }}
       run: |
         ./devscripts/run_tests.${{ matrix.run-tests-ext }}
-
   flake8:
     name: Linter
     runs-on: ubuntu-latest

README.md

@@ -33,7 +33,7 @@ Windows users can [download an .exe file](https://yt-dl.org/latest/youtube-dl.ex
 You can also use pip:
 
     sudo -H pip install --upgrade youtube-dl
-    
+
 This command will update youtube-dl if you have already installed it. See the [pypi page](https://pypi.python.org/pypi/youtube_dl) for more information.
 
 macOS users can install youtube-dl with [Homebrew](https://brew.sh/):
@@ -563,7 +563,7 @@ The basic usage is not to set any template arguments when downloading a single f
  - `is_live` (boolean): Whether this video is a live stream or a fixed-length video
  - `start_time` (numeric): Time in seconds where the reproduction should start, as specified in the URL
  - `end_time` (numeric): Time in seconds where the reproduction should end, as specified in the URL
- - `format` (string): A human-readable description of the format 
+ - `format` (string): A human-readable description of the format
  - `format_id` (string): Format code specified by `--format`
  - `format_note` (string): Additional info about the format
  - `width` (numeric): Width of the video
@@ -675,7 +675,7 @@ The general syntax for format selection is `--format FORMAT` or shorter `-f FORM
 
 **tl;dr:** [navigate me to examples](#format-selection-examples).
 
-The simplest case is requesting a specific format, for example with `-f 22` you can download the format with format code equal to 22. You can get the list of available format codes for particular video using `--list-formats` or `-F`. Note that these format codes are extractor specific. 
+The simplest case is requesting a specific format, for example with `-f 22` you can download the format with format code equal to 22. You can get the list of available format codes for particular video using `--list-formats` or `-F`. Note that these format codes are extractor specific.
 
 You can also use a file extension (currently `3gp`, `aac`, `flv`, `m4a`, `mp3`, `mp4`, `ogg`, `wav`, `webm` are supported) to download the best quality format of a particular file extension served as a single file, e.g. `-f webm` will download the best quality format with the `webm` extension served as a single file.
 
@@ -760,7 +760,7 @@ Videos can be filtered by their upload date using the options `--date`, `--dateb
 
  - Absolute dates: Dates in the format `YYYYMMDD`.
  - Relative dates: Dates in the format `(now|today)[+-][0-9](day|week|month|year)(s)?`
- 
+
 Examples:
 
 ```bash
@@ -1000,6 +1000,8 @@ To run the test, simply invoke your favorite test runner, or execute a test file
     python test/test_download.py
     nosetests
 
+For Python versions 3.6 and later, you can use [pynose](https://pypi.org/project/pynose/) to implement `nosetests`. The original [nose](https://pypi.org/project/nose/) has not been upgraded for 3.10 and later.
+
 See item 6 of [new extractor tutorial](#adding-support-for-a-new-site) for how to run extractor specific test cases.
 
 If you want to create a build of youtube-dl yourself, you'll need
@@ -1091,7 +1093,7 @@ In any case, thank you very much for your contributions!
 
 ## youtube-dl coding conventions
 
-This section introduces a guide lines for writing idiomatic, robust and future-proof extractor code.
+This section introduces guidelines for writing idiomatic, robust and future-proof extractor code.
 
 Extractors are very fragile by nature since they depend on the layout of the source data provided by 3rd party media hosters out of your control and this layout tends to change. As an extractor implementer your task is not only to write code that will extract media links and metadata correctly but also to minimize dependency on the source's layout and even to make the code foresee potential future changes and be ready for that. This is important because it will allow the extractor not to break on minor layout changes thus keeping old youtube-dl versions working. Even though this breakage issue is easily fixed by emitting a new version of youtube-dl with a fix incorporated, all the previous versions become broken in all repositories and distros' packages that may not be so prompt in fetching the update from us. Needless to say, some non rolling release distros may never receive an update at all.
 
@@ -1114,7 +1116,7 @@ Say you have some source dictionary `meta` that you've fetched as JSON with HTTP
 ```python
 meta = self._download_json(url, video_id)
 ```
-    
+
 Assume at this point `meta`'s layout is:
 
 ```python
@@ -1158,7 +1160,7 @@ description = self._search_regex(
 ```
 
 On failure this code will silently continue the extraction with `description` set to `None`. That is useful for metafields that may or may not be present.
- 
+
 ### Provide fallbacks
 
 When extracting metadata try to do so from multiple sources. For example if `title` is present in several places, try extracting from at least some of them. This makes it more future-proof in case some of the sources become unavailable.
@@ -1206,7 +1208,7 @@ r'(id|ID)=(?P<id>\d+)'
 #### Make regular expressions relaxed and flexible
 
 When using regular expressions try to write them fuzzy, relaxed and flexible, skipping insignificant parts that are more likely to change, allowing both single and double quotes for quoted values and so on.
- 
+
 ##### Example
 
 Say you need to extract `title` from the following HTML code:
@@ -1230,7 +1232,7 @@ title = self._search_regex(
     webpage, 'title', group='title')
 ```
 
-Note how you tolerate potential changes in the `style` attribute's value or switch from using double quotes to single for `class` attribute: 
+Note how you tolerate potential changes in the `style` attribute's value or switch from using double quotes to single for `class` attribute:
 
 The code definitely should not look like:
 
@@ -1331,27 +1333,114 @@ Wrap all extracted numeric data into safe functions from [`youtube_dl/utils.py`]
 
 Use `url_or_none` for safe URL processing.
 
-Use `try_get` for safe metadata extraction from parsed JSON.
+Use `traverse_obj` for safe metadata extraction from parsed JSON.
 
-Use `unified_strdate` for uniform `upload_date` or any `YYYYMMDD` meta field extraction, `unified_timestamp` for uniform `timestamp` extraction, `parse_filesize` for `filesize` extraction, `parse_count` for count meta fields extraction, `parse_resolution`, `parse_duration` for `duration` extraction, `parse_age_limit` for `age_limit` extraction. 
+Use `unified_strdate` for uniform `upload_date` or any `YYYYMMDD` meta field extraction, `unified_timestamp` for uniform `timestamp` extraction, `parse_filesize` for `filesize` extraction, `parse_count` for count meta fields extraction, `parse_resolution`, `parse_duration` for `duration` extraction, `parse_age_limit` for `age_limit` extraction.
 
 Explore [`youtube_dl/utils.py`](https://github.com/ytdl-org/youtube-dl/blob/master/youtube_dl/utils.py) for more useful convenience functions.
 
 #### More examples
 
 ##### Safely extract optional description from parsed JSON
+
+When processing complex JSON, as often returned by site API requests or stashed in web pages for "hydration", you can use the `traverse_obj()` utility function to handle multiple fallback values and to ensure the expected type of metadata items. The function's docstring defines how the function works: also review usage in the codebase for more examples.
+
+In this example, a text `description`, or `None`, is pulled from the `.result.video[0].summary` member of the parsed JSON `response`, if available.
+
+```python
+description = traverse_obj(response, ('result', 'video', 0, 'summary', T(compat_str)))
+```
+`T(...)` is a shorthand for a set literal; if you hate people who still run Python 2.6, `T(type_or_transformation)` could be written as a set literal `{type_or_transformation}`.
+
+Some extractors use the older and less capable `try_get()` function in the same way.
+
 ```python
 description = try_get(response, lambda x: x['result']['video'][0]['summary'], compat_str)
 ```
 
 ##### Safely extract more optional metadata
+
+In this example, various optional metadata values are extracted from the `.result.video[0]` member of the parsed JSON `response`, which is expected to be a JS object, parsed into a `dict`, with no crash if that isn't so, or if any of the target values are missing or invalid.
+
 ```python
-video = try_get(response, lambda x: x['result']['video'][0], dict) or {}
+video = traverse_obj(response, ('result', 'video', 0, T(dict))) or {}
+# formerly:
+# video = try_get(response, lambda x: x['result']['video'][0], dict) or {}
 description = video.get('summary')
 duration = float_or_none(video.get('durationMs'), scale=1000)
 view_count = int_or_none(video.get('views'))
 ```
 
+#### Safely extract nested lists
+
+Suppose you've extracted JSON like this into a Python data structure named `media_json` using, say, the `_download_json()` or `_parse_json()` methods of `InfoExtractor`:
+```json
+{
+    "title": "Example video",
+    "comment": "try extracting this",
+    "media": [{
+        "type": "bad",
+        "size": 320,
+        "url": "https://some.cdn.site/bad.mp4"
+    }, {
+        "type": "streaming",
+        "url": "https://some.cdn.site/hls.m3u8"
+    }, {
+        "type": "super",
+        "size": 1280,
+        "url": "https://some.cdn.site/good.webm"
+    }],
+    "moreStuff": "more values",
+    ...
+}
+```
+
+Then extractor code like this can collect the various fields of the JSON:
+```python
+...
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    T,
+    traverse_obj,
+    txt_or_none,
+    url_or_none,
+)
+...
+        ...
+        info_dict = {}
+        # extract title and description if valid and not empty
+        info_dict.update(traverse_obj(media_json, {
+            'title': ('title', T(txt_or_none)),
+            'description': ('comment', T(txt_or_none)),
+        }))
+
+        # extract any recognisable media formats
+        fmts = []
+        # traverse into "media" list, extract `dict`s with desired keys
+        for fmt in traverse_obj(media_json, ('media', Ellipsis, {
+                'format_id': ('type', T(txt_or_none)),
+                'url': ('url', T(url_or_none)),
+                'width': ('size', T(int_or_none)), })):
+            # bad `fmt` values were `None` and removed
+            if 'url' not in fmt:
+                continue
+            fmt_url = fmt['url']  # known to be valid URL
+            ext = determine_ext(fmt_url)
+            if ext == 'm3u8':
+                fmts.extend(self._extract_m3u8_formats(fmt_url, video_id, 'mp4', fatal=False))
+            else:
+                fmt['ext'] = ext
+                fmts.append(fmt)
+
+        # sort, raise if no formats
+        self._sort_formats(fmts)
+
+        info_dict['formats'] = fmts
+        ...
+```
+The extractor raises an exception rather than random crashes if the JSON structure changes so that no formats are found.
+
 # EMBEDDING YOUTUBE-DL
 
 youtube-dl makes the best effort to be a good command-line program, and thus should be callable from any programming language. If you encounter any problems parsing its output, feel free to [create a report](https://github.com/ytdl-org/youtube-dl/issues/new).

devscripts/__init__.py

@@ -0,0 +1 @@
+# Empty file needed to make devscripts.utils properly importable from outside

devscripts/bash-completion.py

@@ -5,8 +5,12 @@ import os
 from os.path import dirname as dirn
 import sys
 
-sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
+sys.path.insert(0, dirn(dirn(os.path.abspath(__file__))))
+
 import youtube_dl
+from youtube_dl.compat import compat_open as open
+
+from utils import read_file
 
 BASH_COMPLETION_FILE = "youtube-dl.bash-completion"
 BASH_COMPLETION_TEMPLATE = "devscripts/bash-completion.in"
@@ -18,9 +22,8 @@ def build_completion(opt_parser):
         for option in group.option_list:
             # for every long flag
             opts_flag.append(option.get_opt_string())
-    with open(BASH_COMPLETION_TEMPLATE) as f:
-        template = f.read()
-    with open(BASH_COMPLETION_FILE, "w") as f:
+    template = read_file(BASH_COMPLETION_TEMPLATE)
+    with open(BASH_COMPLETION_FILE, "w", encoding='utf-8') as f:
         # just using the special char
         filled_template = template.replace("{{flags}}", " ".join(opts_flag))
         f.write(filled_template)

devscripts/create-github-release.py

@@ -1,7 +1,6 @@
 #!/usr/bin/env python
 from __future__ import unicode_literals
 
-import io
 import json
 import mimetypes
 import netrc
@@ -10,7 +9,9 @@ import os
 import re
 import sys
 
-sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+dirn = os.path.dirname
+
+sys.path.insert(0, dirn(dirn(os.path.abspath(__file__))))
 
 from youtube_dl.compat import (
     compat_basestring,
@@ -22,6 +23,7 @@ from youtube_dl.utils import (
     make_HTTPS_handler,
     sanitized_Request,
 )
+from utils import read_file
 
 
 class GitHubReleaser(object):
@@ -89,8 +91,7 @@ def main():
 
     changelog_file, version, build_path = args
 
-    with io.open(changelog_file, encoding='utf-8') as inf:
-        changelog = inf.read()
+    changelog = read_file(changelog_file)
 
     mobj = re.search(r'(?s)version %s\n{2}(.+?)\n{3}' % version, changelog)
     body = mobj.group(1) if mobj else ''

devscripts/fish-completion.py

@@ -6,10 +6,13 @@ import os
 from os.path import dirname as dirn
 import sys
 
-sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
+sys.path.insert(0, dirn(dirn(os.path.abspath(__file__))))
+
 import youtube_dl
 from youtube_dl.utils import shell_quote
 
+from utils import read_file, write_file
+
 FISH_COMPLETION_FILE = 'youtube-dl.fish'
 FISH_COMPLETION_TEMPLATE = 'devscripts/fish-completion.in'
 
@@ -38,11 +41,9 @@ def build_completion(opt_parser):
             complete_cmd.extend(EXTRA_ARGS.get(long_option, []))
             commands.append(shell_quote(complete_cmd))
 
-    with open(FISH_COMPLETION_TEMPLATE) as f:
-        template = f.read()
+    template = read_file(FISH_COMPLETION_TEMPLATE)
     filled_template = template.replace('{{commands}}', '\n'.join(commands))
-    with open(FISH_COMPLETION_FILE, 'w') as f:
-        f.write(filled_template)
+    write_file(FISH_COMPLETION_FILE, filled_template)
 
 
 parser = youtube_dl.parseOpts()[0]

devscripts/gh-pages/add-version.py

@@ -6,16 +6,21 @@ import sys
 import hashlib
 import os.path
 
+dirn = os.path.dirname
+
+sys.path.insert(0, dirn(dirn(dirn(os.path.abspath(__file__)))))
+
+from devscripts.utils import read_file, write_file
+from youtube_dl.compat import compat_open as open
 
 if len(sys.argv) <= 1:
     print('Specify the version number as parameter')
     sys.exit()
 version = sys.argv[1]
 
-with open('update/LATEST_VERSION', 'w') as f:
-    f.write(version)
+write_file('update/LATEST_VERSION', version)
 
-versions_info = json.load(open('update/versions.json'))
+versions_info = json.loads(read_file('update/versions.json'))
 if 'signature' in versions_info:
     del versions_info['signature']
 
@@ -39,5 +44,5 @@ for key, filename in filenames.items():
 versions_info['versions'][version] = new_version
 versions_info['latest'] = version
 
-with open('update/versions.json', 'w') as jsonf:
-    json.dump(versions_info, jsonf, indent=4, sort_keys=True)
+with open('update/versions.json', 'w', encoding='utf-8') as jsonf:
+    json.dumps(versions_info, jsonf, indent=4, sort_keys=True)

devscripts/gh-pages/generate-download.py

@@ -2,14 +2,21 @@
 from __future__ import unicode_literals
 
 import json
+import os.path
+import sys
 
-versions_info = json.load(open('update/versions.json'))
+dirn = os.path.dirname
+
+sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
+
+from utils import read_file, write_file
+
+versions_info = json.loads(read_file('update/versions.json'))
 version = versions_info['latest']
 version_dict = versions_info['versions'][version]
 
 # Read template page
-with open('download.html.in', 'r', encoding='utf-8') as tmplf:
-    template = tmplf.read()
+template = read_file('download.html.in')
 
 template = template.replace('@PROGRAM_VERSION@', version)
 template = template.replace('@PROGRAM_URL@', version_dict['bin'][0])
@@ -18,5 +25,5 @@ template = template.replace('@EXE_URL@', version_dict['exe'][0])
 template = template.replace('@EXE_SHA256SUM@', version_dict['exe'][1])
 template = template.replace('@TAR_URL@', version_dict['tar'][0])
 template = template.replace('@TAR_SHA256SUM@', version_dict['tar'][1])
-with open('download.html', 'w', encoding='utf-8') as dlf:
-    dlf.write(template)
+
+write_file('download.html', template)

devscripts/gh-pages/update-copyright.py

@@ -5,17 +5,22 @@ from __future__ import with_statement, unicode_literals
 
 import datetime
 import glob
-import io  # For Python 2 compatibility
 import os
 import re
+import sys
 
-year = str(datetime.datetime.now().year)
+dirn = os.path.dirname
+
+sys.path.insert(0, dirn(dirn(dirn(os.path.abspath(__file__)))))
+
+from devscripts.utils import read_file, write_file
+from youtube_dl import compat_str
+
+year = compat_str(datetime.datetime.now().year)
 for fn in glob.glob('*.html*'):
-    with io.open(fn, encoding='utf-8') as f:
-        content = f.read()
+    content = read_file(fn)
     newc = re.sub(r'(?P<copyright>Copyright © 2011-)(?P<year>[0-9]{4})', 'Copyright © 2011-' + year, content)
     if content != newc:
         tmpFn = fn + '.part'
-        with io.open(tmpFn, 'wt', encoding='utf-8') as outf:
-            outf.write(newc)
+        write_file(tmpFn, newc)
         os.rename(tmpFn, fn)

devscripts/gh-pages/update-feed.py

@@ -2,10 +2,16 @@
 from __future__ import unicode_literals
 
 import datetime
-import io
 import json
+import os.path
 import textwrap
+import sys
 
+dirn = os.path.dirname
+
+sys.path.insert(0, dirn(dirn(os.path.abspath(__file__))))
+
+from utils import write_file
 
 atom_template = textwrap.dedent("""\
     <?xml version="1.0" encoding="utf-8"?>
@@ -72,5 +78,4 @@ for v in versions:
 entries_str = textwrap.indent(''.join(entries), '\t')
 atom_template = atom_template.replace('@ENTRIES@', entries_str)
 
-with io.open('update/releases.atom', 'w', encoding='utf-8') as atom_file:
-    atom_file.write(atom_template)
+write_file('update/releases.atom', atom_template)

devscripts/gh-pages/update-sites.py

@@ -5,15 +5,17 @@ import sys
 import os
 import textwrap
 
+dirn = os.path.dirname
+
 # We must be able to import youtube_dl
-sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))))
+sys.path.insert(0, dirn(dirn(dirn(os.path.abspath(__file__)))))
 
 import youtube_dl
+from devscripts.utils import read_file, write_file
 
 
 def main():
-    with open('supportedsites.html.in', 'r', encoding='utf-8') as tmplf:
-        template = tmplf.read()
+    template = read_file('supportedsites.html.in')
 
     ie_htmls = []
     for ie in youtube_dl.list_extractors(age_limit=None):
@@ -29,8 +31,7 @@ def main():
 
     template = template.replace('@SITES@', textwrap.indent('\n'.join(ie_htmls), '\t'))
 
-    with open('supportedsites.html', 'w', encoding='utf-8') as sitesf:
-        sitesf.write(template)
+    write_file('supportedsites.html', template)
 
 
 if __name__ == '__main__':

devscripts/make_contributing.py

@@ -1,10 +1,11 @@
 #!/usr/bin/env python
 from __future__ import unicode_literals
 
-import io
 import optparse
 import re
 
+from utils import read_file, write_file
+
 
 def main():
     parser = optparse.OptionParser(usage='%prog INFILE OUTFILE')
@@ -14,8 +15,7 @@ def main():
 
     infile, outfile = args
 
-    with io.open(infile, encoding='utf-8') as inf:
-        readme = inf.read()
+    readme = read_file(infile)
 
     bug_text = re.search(
         r'(?s)#\s*BUGS\s*[^\n]*\s*(.*?)#\s*COPYRIGHT', readme).group(1)
@@ -25,8 +25,7 @@ def main():
 
     out = bug_text + dev_text
 
-    with io.open(outfile, 'w', encoding='utf-8') as outf:
-        outf.write(out)
+    write_file(outfile, out)
 
 
 if __name__ == '__main__':

devscripts/make_issue_template.py

@@ -1,8 +1,11 @@
 #!/usr/bin/env python
 from __future__ import unicode_literals
 
-import io
 import optparse
+import os.path
+import sys
+
+from utils import read_file, read_version, write_file
 
 
 def main():
@@ -13,17 +16,11 @@ def main():
 
     infile, outfile = args
 
-    with io.open(infile, encoding='utf-8') as inf:
-        issue_template_tmpl = inf.read()
+    issue_template_tmpl = read_file(infile)
 
-    # Get the version from youtube_dl/version.py without importing the package
-    exec(compile(open('youtube_dl/version.py').read(),
-                 'youtube_dl/version.py', 'exec'))
+    out = issue_template_tmpl % {'version': read_version()}
 
-    out = issue_template_tmpl % {'version': locals()['__version__']}
-
-    with io.open(outfile, 'w', encoding='utf-8') as outf:
-        outf.write(out)
+    write_file(outfile, out)
 
 if __name__ == '__main__':
     main()

devscripts/make_lazy_extractors.py

@@ -1,7 +1,6 @@
 from __future__ import unicode_literals, print_function
 
 from inspect import getsource
-import io
 import os
 from os.path import dirname as dirn
 import re
@@ -9,17 +8,20 @@ import sys
 
 print('WARNING: Lazy loading extractors is an experimental feature that may not always work', file=sys.stderr)
 
-sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
+sys.path.insert(0, dirn(dirn(os.path.abspath(__file__))))
 
 lazy_extractors_filename = sys.argv[1]
 if os.path.exists(lazy_extractors_filename):
     os.remove(lazy_extractors_filename)
 # Py2: may be confused by leftover lazy_extractors.pyc
-try:
-    os.remove(lazy_extractors_filename + 'c')
-except OSError:
-    pass
+if sys.version_info[0] < 3:
+    for c in ('c', 'o'):
+        try:
+            os.remove(lazy_extractors_filename + 'c')
+        except OSError:
+            pass
 
+from devscripts.utils import read_file, write_file
 from youtube_dl.compat import compat_register_utf8
 
 compat_register_utf8()
@@ -27,8 +29,7 @@ compat_register_utf8()
 from youtube_dl.extractor import _ALL_CLASSES
 from youtube_dl.extractor.common import InfoExtractor, SearchInfoExtractor
 
-with open('devscripts/lazy_load_template.py', 'rt') as f:
-    module_template = f.read()
+module_template = read_file('devscripts/lazy_load_template.py')
 
 
 def get_source(m):
@@ -114,7 +115,17 @@ for ie in ordered_cls:
 module_contents.append(
     '_ALL_CLASSES = [{0}]'.format(', '.join(names)))
 
-module_src = '\n'.join(module_contents) + '\n'
+module_src = '\n'.join(module_contents)
 
-with io.open(lazy_extractors_filename, 'wt', encoding='utf-8') as f:
-    f.write(module_src)
+write_file(lazy_extractors_filename, module_src + '\n')
+
+# work around JVM byte code module limit in Jython
+if sys.platform.startswith('java') and sys.version_info[:2] == (2, 7):
+    import subprocess
+    from youtube_dl.compat import compat_subprocess_get_DEVNULL
+    # if Python 2.7 is available, use it to compile the module for Jython
+    try:
+        # if Python 2.7 is available, use it to compile the module for Jython
+        subprocess.check_call(['python2.7', '-m', 'py_compile', lazy_extractors_filename], stdout=compat_subprocess_get_DEVNULL())
+    except Exception:
+        pass

devscripts/make_readme.py

@@ -1,8 +1,14 @@
 from __future__ import unicode_literals
 
-import io
-import sys
+import os.path
 import re
+import sys
+dirn = os.path.dirname
+
+sys.path.insert(0, dirn(dirn(os.path.abspath(__file__))))
+
+from utils import read_file
+from youtube_dl.compat import compat_open as open
 
 README_FILE = 'README.md'
 helptext = sys.stdin.read()
@@ -10,8 +16,7 @@ helptext = sys.stdin.read()
 if isinstance(helptext, bytes):
     helptext = helptext.decode('utf-8')
 
-with io.open(README_FILE, encoding='utf-8') as f:
-    oldreadme = f.read()
+oldreadme = read_file(README_FILE)
 
 header = oldreadme[:oldreadme.index('# OPTIONS')]
 footer = oldreadme[oldreadme.index('# CONFIGURATION'):]
@@ -20,7 +25,7 @@ options = helptext[helptext.index('  General Options:') + 19:]
 options = re.sub(r'(?m)^  (\w.+)$', r'## \1', options)
 options = '# OPTIONS\n' + options + '\n'
 
-with io.open(README_FILE, 'w', encoding='utf-8') as f:
+with open(README_FILE, 'w', encoding='utf-8') as f:
     f.write(header)
     f.write(options)
     f.write(footer)

devscripts/make_supportedsites.py

@@ -1,17 +1,19 @@
 #!/usr/bin/env python
 from __future__ import unicode_literals
 
-import io
 import optparse
-import os
+import os.path
 import sys
 
-
 # Import youtube_dl
-ROOT_DIR = os.path.join(os.path.dirname(__file__), '..')
-sys.path.insert(0, ROOT_DIR)
+dirn = os.path.dirname
+
+sys.path.insert(0, dirn(dirn(os.path.abspath(__file__))))
+
 import youtube_dl
 
+from utils import write_file
+
 
 def main():
     parser = optparse.OptionParser(usage='%prog OUTFILE.md')
@@ -38,8 +40,7 @@ def main():
         ' - ' + md + '\n'
         for md in gen_ies_md(ies))
 
-    with io.open(outfile, 'w', encoding='utf-8') as outf:
-        outf.write(out)
+    write_file(outfile, out)
 
 
 if __name__ == '__main__':

devscripts/prepare_manpage.py

@@ -1,13 +1,13 @@
 from __future__ import unicode_literals
 
-import io
 import optparse
 import os.path
 import re
 
+from utils import read_file, write_file
+
 ROOT_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 README_FILE = os.path.join(ROOT_DIR, 'README.md')
-
 PREFIX = r'''%YOUTUBE-DL(1)
 
 # NAME
@@ -29,8 +29,7 @@ def main():
 
     outfile, = args
 
-    with io.open(README_FILE, encoding='utf-8') as f:
-        readme = f.read()
+    readme = read_file(README_FILE)
 
     readme = re.sub(r'(?s)^.*?(?=# DESCRIPTION)', '', readme)
     readme = re.sub(r'\s+youtube-dl \[OPTIONS\] URL \[URL\.\.\.\]', '', readme)
@@ -38,8 +37,7 @@ def main():
 
     readme = filter_options(readme)
 
-    with io.open(outfile, 'w', encoding='utf-8') as outf:
-        outf.write(readme)
+    write_file(outfile, readme)
 
 
 def filter_options(readme):

devscripts/utils.py

@@ -0,0 +1,62 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import argparse
+import functools
+import os.path
+import subprocess
+import sys
+
+dirn = os.path.dirname
+
+sys.path.insert(0, dirn(dirn(os.path.abspath(__file__))))
+
+from youtube_dl.compat import (
+    compat_kwargs,
+    compat_open as open,
+)
+
+
+def read_file(fname):
+    with open(fname, encoding='utf-8') as f:
+        return f.read()
+
+
+def write_file(fname, content, mode='w'):
+    with open(fname, mode, encoding='utf-8') as f:
+        return f.write(content)
+
+
+def read_version(fname='youtube_dl/version.py'):
+    """Get the version without importing the package"""
+    exec(compile(read_file(fname), fname, 'exec'))
+    return locals()['__version__']
+
+
+def get_filename_args(has_infile=False, default_outfile=None):
+    parser = argparse.ArgumentParser()
+    if has_infile:
+        parser.add_argument('infile', help='Input file')
+    kwargs = {'nargs': '?', 'default': default_outfile} if default_outfile else {}
+    kwargs['help'] = 'Output file'
+    parser.add_argument('outfile', **compat_kwargs(kwargs))
+
+    opts = parser.parse_args()
+    if has_infile:
+        return opts.infile, opts.outfile
+    return opts.outfile
+
+
+def compose_functions(*functions):
+    return lambda x: functools.reduce(lambda y, f: f(y), functions, x)
+
+
+def run_process(*args, **kwargs):
+    kwargs.setdefault('text', True)
+    kwargs.setdefault('check', True)
+    kwargs.setdefault('capture_output', True)
+    if kwargs['text']:
+        kwargs.setdefault('encoding', 'utf-8')
+        kwargs.setdefault('errors', 'replace')
+        kwargs = compat_kwargs(kwargs)
+    return subprocess.run(args, **kwargs)

devscripts/zsh-completion.py

@@ -7,6 +7,8 @@ import sys
 
 sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
 import youtube_dl
+from utils import read_file, write_file
+
 
 ZSH_COMPLETION_FILE = "youtube-dl.zsh"
 ZSH_COMPLETION_TEMPLATE = "devscripts/zsh-completion.in"
@@ -34,15 +36,13 @@ def build_completion(opt_parser):
 
     flags = [opt.get_opt_string() for opt in opts]
 
-    with open(ZSH_COMPLETION_TEMPLATE) as f:
-        template = f.read()
+    template = read_file(ZSH_COMPLETION_TEMPLATE)
 
     template = template.replace("{{fileopts}}", "|".join(fileopts))
     template = template.replace("{{diropts}}", "|".join(diropts))
     template = template.replace("{{flags}}", " ".join(flags))
 
-    with open(ZSH_COMPLETION_FILE, "w") as f:
-        f.write(template)
+    write_file(ZSH_COMPLETION_FILE, template)
 
 
 parser = youtube_dl.parseOpts()[0]

test/helper.py

@@ -1,7 +1,6 @@
 from __future__ import unicode_literals
 
 import errno
-import io
 import hashlib
 import json
 import os.path
@@ -14,6 +13,7 @@ import unittest
 import youtube_dl.extractor
 from youtube_dl import YoutubeDL
 from youtube_dl.compat import (
+    compat_open as open,
     compat_os_name,
     compat_str,
 )
@@ -29,10 +29,10 @@ def get_params(override=None):
                                    "parameters.json")
     LOCAL_PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)),
                                          "local_parameters.json")
-    with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
+    with open(PARAMETERS_FILE, encoding='utf-8') as pf:
         parameters = json.load(pf)
     if os.path.exists(LOCAL_PARAMETERS_FILE):
-        with io.open(LOCAL_PARAMETERS_FILE, encoding='utf-8') as pf:
+        with open(LOCAL_PARAMETERS_FILE, encoding='utf-8') as pf:
             parameters.update(json.load(pf))
     if override:
         parameters.update(override)

test/test_InfoExtractor.py

@@ -3,7 +3,6 @@
 from __future__ import unicode_literals
 
 # Allow direct execution
-import io
 import os
 import sys
 import unittest
@@ -21,6 +20,7 @@ from test.helper import (
 from youtube_dl.compat import (
     compat_etree_fromstring,
     compat_http_server,
+    compat_open as open,
 )
 from youtube_dl.extractor.common import InfoExtractor
 from youtube_dl.extractor import (
@@ -902,8 +902,8 @@ jwplayer("mediaplayer").setup({"abouttext":"Visit Indie DB","aboutlink":"http:\/
         ]
 
         for m3u8_file, m3u8_url, expected_formats in _TEST_CASES:
-            with io.open('./test/testdata/m3u8/%s.m3u8' % m3u8_file,
-                         mode='r', encoding='utf-8') as f:
+            with open('./test/testdata/m3u8/%s.m3u8' % m3u8_file,
+                      mode='r', encoding='utf-8') as f:
                 formats = self.ie._parse_m3u8_formats(
                     f.read(), m3u8_url, ext='mp4')
                 self.ie._sort_formats(formats)
@@ -1127,8 +1127,8 @@ jwplayer("mediaplayer").setup({"abouttext":"Visit Indie DB","aboutlink":"http:\/
         ]
 
         for mpd_file, mpd_url, mpd_base_url, expected_formats in _TEST_CASES:
-            with io.open('./test/testdata/mpd/%s.mpd' % mpd_file,
-                         mode='r', encoding='utf-8') as f:
+            with open('./test/testdata/mpd/%s.mpd' % mpd_file,
+                      mode='r', encoding='utf-8') as f:
                 formats = self.ie._parse_mpd_formats(
                     compat_etree_fromstring(f.read().encode('utf-8')),
                     mpd_base_url=mpd_base_url, mpd_url=mpd_url)
@@ -1154,8 +1154,8 @@ jwplayer("mediaplayer").setup({"abouttext":"Visit Indie DB","aboutlink":"http:\/
         ]
 
         for f4m_file, f4m_url, expected_formats in _TEST_CASES:
-            with io.open('./test/testdata/f4m/%s.f4m' % f4m_file,
-                         mode='r', encoding='utf-8') as f:
+            with open('./test/testdata/f4m/%s.f4m' % f4m_file,
+                      mode='r', encoding='utf-8') as f:
                 formats = self.ie._parse_f4m_formats(
                     compat_etree_fromstring(f.read().encode('utf-8')),
                     f4m_url, None)
@@ -1202,8 +1202,8 @@ jwplayer("mediaplayer").setup({"abouttext":"Visit Indie DB","aboutlink":"http:\/
         ]
 
         for xspf_file, xspf_url, expected_entries in _TEST_CASES:
-            with io.open('./test/testdata/xspf/%s.xspf' % xspf_file,
-                         mode='r', encoding='utf-8') as f:
+            with open('./test/testdata/xspf/%s.xspf' % xspf_file,
+                      mode='r', encoding='utf-8') as f:
                 entries = self.ie._parse_xspf(
                     compat_etree_fromstring(f.read().encode('utf-8')),
                     xspf_file, xspf_url=xspf_url, xspf_base_url=xspf_url)

test/test_YoutubeDL.py

@@ -22,6 +22,7 @@ from youtube_dl.compat import (
     compat_http_cookiejar_Cookie,
     compat_http_cookies_SimpleCookie,
     compat_kwargs,
+    compat_open as open,
     compat_str,
     compat_urllib_error,
 )
@@ -701,12 +702,12 @@ class TestYoutubeDL(unittest.TestCase):
 
         class SimplePP(PostProcessor):
             def run(self, info):
-                with open(audiofile, 'wt') as f:
+                with open(audiofile, 'w') as f:
                     f.write('EXAMPLE')
                 return [info['filepath']], info
 
         def run_pp(params, PP):
-            with open(filename, 'wt') as f:
+            with open(filename, 'w') as f:
                 f.write('EXAMPLE')
             ydl = YoutubeDL(params)
             ydl.add_post_processor(PP())
@@ -725,7 +726,7 @@ class TestYoutubeDL(unittest.TestCase):
 
         class ModifierPP(PostProcessor):
             def run(self, info):
-                with open(info['filepath'], 'wt') as f:
+                with open(info['filepath'], 'w') as f:
                     f.write('MODIFIED')
                 return [], info
 

test/test_download.py

@@ -20,15 +20,15 @@ from test.helper import (
 
 
 import hashlib
-import io
 import json
 import socket
 
 import youtube_dl.YoutubeDL
 from youtube_dl.compat import (
     compat_http_client,
-    compat_urllib_error,
     compat_HTTPError,
+    compat_open as open,
+    compat_urllib_error,
 )
 from youtube_dl.utils import (
     DownloadError,
@@ -245,7 +245,7 @@ def generator(test_case, tname):
                 self.assertTrue(
                     os.path.exists(info_json_fn),
                     'Missing info file %s' % info_json_fn)
-                with io.open(info_json_fn, encoding='utf-8') as infof:
+                with open(info_json_fn, encoding='utf-8') as infof:
                     info_dict = json.load(infof)
                 expect_info_dict(self, info_dict, tc.get('info_dict', {}))
         finally:

test/test_execution.py

@@ -8,19 +8,18 @@ import unittest
 import sys
 import os
 import subprocess
-sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-from youtube_dl.compat import compat_register_utf8
+rootDir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+
+sys.path.insert(0, rootDir)
+
+from youtube_dl.compat import compat_register_utf8, compat_subprocess_get_DEVNULL
 from youtube_dl.utils import encodeArgument
 
 compat_register_utf8()
 
-rootDir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 
-try:
-    _DEV_NULL = subprocess.DEVNULL
-except AttributeError:
-    _DEV_NULL = open(os.devnull, 'wb')
+_DEV_NULL = compat_subprocess_get_DEVNULL()
 
 
 class TestExecution(unittest.TestCase):
@@ -52,10 +51,10 @@ class TestExecution(unittest.TestCase):
             subprocess.check_call([sys.executable, os.path.normpath('devscripts/make_lazy_extractors.py'), lazy_extractors], cwd=rootDir, stdout=_DEV_NULL)
             subprocess.check_call([sys.executable, os.path.normpath('test/test_all_urls.py')], cwd=rootDir, stdout=_DEV_NULL)
         finally:
-            for x in ['', 'c'] if sys.version_info[0] < 3 else ['']:
+            for x in ('', 'c') if sys.version_info[0] < 3 else ('',):
                 try:
                     os.remove(lazy_extractors + x)
-                except (IOError, OSError):
+                except OSError:
                     pass
 
 

test/test_http.py

@@ -41,10 +41,12 @@ from youtube_dl.compat import (
 
 from youtube_dl.utils import (
     sanitized_Request,
+    update_Request,
     urlencode_postdata,
 )
 
 from test.helper import (
+    expectedFailureIf,
     FakeYDL,
     FakeLogger,
     http_server_port,
@@ -243,6 +245,11 @@ class HTTPTestRequestHandler(compat_http_server.BaseHTTPRequestHandler):
 
 
 class TestHTTP(unittest.TestCase):
+    # when does it make sense to check the SSL certificate?
+    _check_cert = (
+        sys.version_info >= (3, 2)
+        or (sys.version_info[0] == 2 and sys.version_info[1:] >= (7, 19)))
+
     def setUp(self):
         # HTTP server
         self.http_httpd = compat_http_server.HTTPServer(
@@ -307,10 +314,7 @@ class TestHTTP(unittest.TestCase):
             else self.https_port if scheme == 'https'
             else self.http_port, path)
 
-    @unittest.skipUnless(
-        sys.version_info >= (3, 2)
-        or (sys.version_info[0] == 2 and sys.version_info[1:] >= (7, 9)),
-        'No support for certificate check in SSL')
+    @unittest.skipUnless(_check_cert, 'No support for certificate check in SSL')
     def test_nocheckcertificate(self):
         with FakeYDL({'logger': FakeLogger()}) as ydl:
             with self.assertRaises(compat_urllib_error.URLError):
@@ -376,6 +380,8 @@ class TestHTTP(unittest.TestCase):
                 with self.assertRaises(compat_urllib_HTTPError):
                     do_req(code, 'GET')
 
+    # Jython 2.7.1 times out for some reason
+    @expectedFailureIf(sys.platform.startswith('java') and sys.version_info < (2, 7, 2))
     def test_content_type(self):
         # https://github.com/yt-dlp/yt-dlp/commit/379a4f161d4ad3e40932dcf5aca6e6fb9715ab28
         with FakeYDL({'nocheckcertificate': True}) as ydl:
@@ -390,6 +396,18 @@ class TestHTTP(unittest.TestCase):
             headers = ydl.urlopen(r).read().decode('utf-8')
             self.assertIn('Content-Type: application/x-www-form-urlencoded', headers)
 
+    def test_update_req(self):
+        req = sanitized_Request('http://example.com')
+        assert req.data is None
+        assert req.get_method() == 'GET'
+        assert not req.has_header('Content-Type')
+        # Test that zero-byte payloads will be sent
+        req = update_Request(req, data=b'')
+        assert req.data == b''
+        assert req.get_method() == 'POST'
+        # yt-dl expects data to be encoded and Content-Type to be added by sender
+        # assert req.get_header('Content-Type') == 'application/x-www-form-urlencoded'
+
     def test_cookiejar(self):
         with FakeYDL() as ydl:
             ydl.cookiejar.set_cookie(compat_http_cookiejar_Cookie(

test/test_swfinterp.py

@@ -5,16 +5,18 @@ from __future__ import unicode_literals
 import os
 import sys
 import unittest
-sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
+dirn = os.path.dirname
+
+sys.path.insert(0, dirn(dirn(os.path.abspath(__file__))))
 
 import errno
-import io
 import json
 import re
 import subprocess
 
 from youtube_dl.swfinterp import SWFInterpreter
+from youtube_dl.compat import compat_open as open
 
 
 TEST_DIR = os.path.join(
@@ -43,7 +45,7 @@ def _make_testfunc(testfile):
                     '-static-link-runtime-shared-libraries', as_file])
             except OSError as ose:
                 if ose.errno == errno.ENOENT:
-                    print('mxmlc not found! Skipping test.')
+                    self.skipTest('mxmlc not found!')
                     return
                 raise
 
@@ -51,7 +53,7 @@ def _make_testfunc(testfile):
             swf_content = swf_f.read()
         swfi = SWFInterpreter(swf_content)
 
-        with io.open(as_file, 'r', encoding='utf-8') as as_f:
+        with open(as_file, 'r', encoding='utf-8') as as_f:
             as_content = as_f.read()
 
         def _find_spec(key):

test/test_unicode_literals.py

@@ -2,14 +2,15 @@ from __future__ import unicode_literals
 
 # Allow direct execution
 import os
+import re
 import sys
 import unittest
-sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-import io
-import re
+dirn = os.path.dirname
 
-rootDir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+rootDir = dirn(dirn(os.path.abspath(__file__)))
+
+sys.path.insert(0, rootDir)
 
 IGNORED_FILES = [
     'setup.py',  # http://bugs.python.org/issue13943
@@ -24,6 +25,7 @@ IGNORED_DIRS = [
 ]
 
 from test.helper import assertRegexpMatches
+from youtube_dl.compat import compat_open as open
 
 
 class TestUnicodeLiterals(unittest.TestCase):
@@ -41,7 +43,7 @@ class TestUnicodeLiterals(unittest.TestCase):
                     continue
 
                 fn = os.path.join(dirpath, basename)
-                with io.open(fn, encoding='utf-8') as inf:
+                with open(fn, encoding='utf-8') as inf:
                     code = inf.read()
 
                 if "'" not in code and '"' not in code:

test/test_write_annotations.py

@@ -11,12 +11,11 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 from test.helper import get_params, try_rm
 
 
-import io
-
 import xml.etree.ElementTree
 
 import youtube_dl.YoutubeDL
 import youtube_dl.extractor
+from youtube_dl.compat import compat_open as open
 
 
 class YoutubeDL(youtube_dl.YoutubeDL):
@@ -51,7 +50,7 @@ class TestAnnotations(unittest.TestCase):
         ydl.download([TEST_ID])
         self.assertTrue(os.path.exists(ANNOTATIONS_FILE))
         annoxml = None
-        with io.open(ANNOTATIONS_FILE, 'r', encoding='utf-8') as annof:
+        with open(ANNOTATIONS_FILE, 'r', encoding='utf-8') as annof:
             annoxml = xml.etree.ElementTree.parse(annof)
         self.assertTrue(annoxml is not None, 'Failed to parse annotations XML')
         root = annoxml.getroot()

test/test_youtube_signature.py

@@ -8,11 +8,14 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
-import io
 import re
 import string
 
-from youtube_dl.compat import compat_str, compat_urlretrieve
+from youtube_dl.compat import (
+    compat_open as open,
+    compat_str,
+    compat_urlretrieve,
+)
 
 from test.helper import FakeYDL
 from youtube_dl.extractor import YoutubeIE
@@ -208,7 +211,7 @@ def t_factory(name, sig_func, url_pattern):
 
             if not os.path.exists(fn):
                 compat_urlretrieve(url, fn)
-            with io.open(fn, encoding='utf-8') as testf:
+            with open(fn, encoding='utf-8') as testf:
                 jscode = testf.read()
             self.assertEqual(sig_func(jscode, sig_input), expected_sig)
 

youtube_dl/YoutubeDL.py

@@ -4,11 +4,9 @@
 from __future__ import absolute_import, unicode_literals
 
 import collections
-import contextlib
 import copy
 import datetime
 import errno
-import fileinput
 import io
 import itertools
 import json
@@ -45,6 +43,7 @@ from .compat import (
     compat_kwargs,
     compat_map as map,
     compat_numeric_types,
+    compat_open as open,
     compat_os_name,
     compat_str,
     compat_tokenize_tokenize,
@@ -1977,7 +1976,7 @@ class YoutubeDL(object):
             else:
                 try:
                     self.to_screen('[info] Writing video description to: ' + descfn)
-                    with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
+                    with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
                         descfile.write(info_dict['description'])
                 except (OSError, IOError):
                     self.report_error('Cannot write description file ' + descfn)
@@ -1992,7 +1991,7 @@ class YoutubeDL(object):
             else:
                 try:
                     self.to_screen('[info] Writing video annotations to: ' + annofn)
-                    with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
+                    with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
                         annofile.write(info_dict['annotations'])
                 except (KeyError, TypeError):
                     self.report_warning('There are no annotations to write.')
@@ -2019,7 +2018,7 @@ class YoutubeDL(object):
                         try:
                             # Use newline='' to prevent conversion of newline characters
                             # See https://github.com/ytdl-org/youtube-dl/issues/10268
-                            with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
+                            with open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
                                 subfile.write(sub_info['data'])
                         except (OSError, IOError):
                             self.report_error('Cannot write subtitles file ' + sub_filename)
@@ -2028,7 +2027,7 @@ class YoutubeDL(object):
                         try:
                             sub_data = ie._request_webpage(
                                 sub_info['url'], info_dict['id'], note=False).read()
-                            with io.open(encodeFilename(sub_filename), 'wb') as subfile:
+                            with open(encodeFilename(sub_filename), 'wb') as subfile:
                                 subfile.write(sub_data)
                         except (ExtractorError, IOError, OSError, ValueError) as err:
                             self.report_warning('Unable to download subtitle for "%s": %s' %
@@ -2232,12 +2231,8 @@ class YoutubeDL(object):
         return self._download_retcode
 
     def download_with_info_file(self, info_filename):
-        with contextlib.closing(fileinput.FileInput(
-                [info_filename], mode='r',
-                openhook=fileinput.hook_encoded('utf-8'))) as f:
-            # FileInput doesn't have a read method, we can't call json.load
-            # TODO: let's use io.open(), then
-            info = self.filter_requested_info(json.loads('\n'.join(f)))
+        with open(info_filename, encoding='utf-8') as f:
+            info = self.filter_requested_info(json.load(f))
         try:
             self.process_ie_result(info, download=True)
         except DownloadError:

youtube_dl/cache.py

@@ -1,14 +1,16 @@
 from __future__ import unicode_literals
 
 import errno
-import io
 import json
 import os
 import re
 import shutil
 import traceback
 
-from .compat import compat_getenv
+from .compat import (
+    compat_getenv,
+    compat_open as open,
+)
 from .utils import (
     error_to_compat_str,
     expand_path,
@@ -83,7 +85,7 @@ class Cache(object):
         cache_fn = self._get_cache_fn(section, key, dtype)
         try:
             try:
-                with io.open(cache_fn, 'r', encoding='utf-8') as cachef:
+                with open(cache_fn, 'r', encoding='utf-8') as cachef:
                     return self._validate(json.load(cachef), min_ver)
             except ValueError:
                 try:

youtube_dl/compat.py

@@ -131,7 +131,7 @@ if sys.version_info[0] == 2 or sys.version_info < (3, 3):
         def load(self, rawdata):
             must_have_value = 0
             if not isinstance(rawdata, dict):
-                if sys.version_info[:2] != (2, 7):
+                if sys.version_info[:2] != (2, 7) or sys.platform.startswith('java'):
                     # attribute must have value for parsing
                     rawdata, must_have_value = re.subn(
                         r'(?i)(;\s*)(secure|httponly)(\s*(?:;|$))', r'\1\2=\2\3', rawdata)

youtube_dl/extractor/common.py

@@ -25,6 +25,7 @@ from ..compat import (
     compat_integer_types,
     compat_http_client,
     compat_map as map,
+    compat_open as open,
     compat_os_name,
     compat_str,
     compat_urllib_error,

youtube_dl/extractor/openload.py

@@ -7,6 +7,7 @@ import subprocess
 import tempfile
 
 from ..compat import (
+    compat_open as open,
     compat_urlparse,
     compat_kwargs,
 )

youtube_dl/postprocessor/embedthumbnail.py

@@ -18,6 +18,8 @@ from ..utils import (
     shell_quote,
 )
 
+from ..compat import compat_open as open
+
 
 class EmbedThumbnailPPError(PostProcessingError):
     pass

youtube_dl/postprocessor/ffmpeg.py

@@ -1,6 +1,5 @@
 from __future__ import unicode_literals
 
-import io
 import os
 import subprocess
 import time
@@ -9,6 +8,7 @@ import re
 
 from .common import AudioConversionError, PostProcessor
 
+from ..compat import compat_open as open
 from ..utils import (
     encodeArgument,
     encodeFilename,
@@ -493,7 +493,7 @@ class FFmpegMetadataPP(FFmpegPostProcessor):
         chapters = info.get('chapters', [])
         if chapters:
             metadata_filename = replace_extension(filename, 'meta')
-            with io.open(metadata_filename, 'wt', encoding='utf-8') as f:
+            with open(metadata_filename, 'w', encoding='utf-8') as f:
                 def ffmpeg_escape(text):
                     return re.sub(r'(=|;|#|\\|\n)', r'\\\1', text)
 
@@ -636,7 +636,7 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
                 with open(dfxp_file, 'rb') as f:
                     srt_data = dfxp2srt(f.read())
 
-                with io.open(srt_file, 'wt', encoding='utf-8') as f:
+                with open(srt_file, 'w', encoding='utf-8') as f:
                     f.write(srt_data)
                 old_file = srt_file
 
@@ -652,7 +652,7 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
 
             self.run_ffmpeg(old_file, new_file, ['-f', new_format])
 
-            with io.open(new_file, 'rt', encoding='utf-8') as f:
+            with open(new_file, 'r', encoding='utf-8') as f:
                 subs[lang] = {
                     'ext': new_ext,
                     'data': f.read(),

youtube_dl/update.py

@@ -1,6 +1,5 @@
 from __future__ import unicode_literals
 
-import io
 import json
 import traceback
 import hashlib
@@ -9,7 +8,10 @@ import subprocess
 import sys
 from zipimport import zipimporter
 
-from .compat import compat_realpath
+from .compat import (
+    compat_open as open,
+    compat_realpath,
+)
 from .utils import encode_compat_str
 
 from .version import __version__
@@ -127,7 +129,7 @@ def update_self(to_screen, verbose, opener):
 
         try:
             bat = os.path.join(directory, 'youtube-dl-updater.bat')
-            with io.open(bat, 'w') as batfile:
+            with open(bat, 'w') as batfile:
                 batfile.write('''
 @echo off
 echo Waiting for file handle to be closed ...

youtube_dl/utils.py

@@ -2996,8 +2996,7 @@ class YoutubeDLRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
         # Technically the Cookie header should be in unredirected_hdrs;
         # however in practice some may set it in normal headers anyway.
         # We will remove it here to prevent any leaks.
-        # Also remove unwanted and undocumented Host header for old URL
-        remove_headers = ['Cookie', 'Host']
+        remove_headers = ['Cookie']
 
         # A 303 must either use GET or HEAD for subsequent request
         # https://datatracker.ietf.org/doc/html/rfc7231#section-6.4.4
@@ -3016,7 +3015,7 @@ class YoutubeDLRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
             remove_headers.extend(['Content-Length', 'Content-Type'])
 
         # NB: don't use dict comprehension for python 2.6 compatibility
-        new_headers = dict((k, v) for k, v in req.header_items()
+        new_headers = dict((k, v) for k, v in req.headers.items()
                            if k.title() not in remove_headers)
 
         return compat_urllib_request.Request(
@@ -4187,7 +4186,7 @@ def update_url_query(url, query):
 def update_Request(req, url=None, data=None, headers={}, query={}):
     req_headers = req.headers.copy()
     req_headers.update(headers)
-    req_data = data or req.data
+    req_data = data if data is not None else req.data
     req_url = update_url_query(url or req.get_full_url(), query)
     req_get_method = req.get_method()
     if req_get_method == 'HEAD':