The apidoc modules were not being generated on ReadTheDocs because
there is no way to run sphinx-apidoc manually.
Moved the running of sphinx-apidoc into conf.py.
Added zope.interface minimum version to fix Readthedocs warning.
The use of pkg_resource.require caused an unwanted requirements lookup
that errored out the sphinx build when no dependencies are installed.
This is fixed by switching to pkg_resources.get_distribution.
Also changed the tox docs env to not install Deluge as the setup.py
now contains install_requires which is unwanted.
- Change the layout and contents of docs to be better organised and
follow ideas from: https://www.divio.com/blog/documentation/
- Use markdown for non-technical documents to speed up writing.
- Added new sections and imported documents from Trac wiki.
Build fixes:
- Added a patch to fix recommonmark 0.4 and doc referencing:
https://github.com/rtfd/recommonmark/issues/93
- Set docs build in tox to Py2.7 since there are problems with autodoc
mocking multiple inheritance on Python 3 resulting in metaclass errors.
- Supressed warning about `modules.rst` not in the toctree by creating
a static `modules.rst` with `:orphan:` file directive and add to git.
Also skip creating this toc file with sphinx-apidoc in setup and tox.
- Simplified finding exported RPC and JSON API methods by adding an
autodoc custom class directive. Removed unneeded __rpcapi.py.
- Fix applying the setting to libtorrent, passing the value without
modification so it decide when to enable it.
- Enable super_seeding option when adding torrents to core.
- Update UIs with option in tabs and add dialogs.
- Found an issue while fixing `get_name` where `handle.rename_file`
would raise a UnicodeDecodeError with non-ascii on Python 2. The
fix is to catch this and pass unicode string to method instead.
- Add a test `test_rename_unicode` to verify no errors are generated.
- Updated test to use core.session instead of creating another one.
The recent change to torrent.get_name does not handle non-ascii paths
on Python 2.
- Add a decode_bytes to resolve the issue.
- Add tests.
- Refactor to reduce nesting.
- Replace usage of `s` for variable names to make it easier to read the code.
- Remove unneeded and unused encoding parameter from parse_color_string
It should not be encoded by this function, only on output.
The following error was encountered by user:
...deluge/ui/console/modes/basemode.py, line 290, in add_string
screen.addstr(row, col, string, color)
UnicodeEncodeError: 'ascii' codec can't encode character...
The `add_str` method is defaulting to using the Python 2 ascii
encoding with a unicode string so use the encoding passed to the
function.
- Add an About window to see version details like GTKUI.
- The author and license text were left out as unnecessary.
- Added a daemon get_version method since daemon version was not
available through the json-api.
- Fix LookupResource to ensure path exists when rendering.
If a user keeps the deluge config file under source control and symlinks
to the config files in deluge config dir then when deluge saves config
files it will replace the symlink with actual file.
Using realpath will resolve these symlinks and file will be updated in
the correct location.
Use a temporary file for new config new before moving it to the
resolved location.
Co-authored-by: Calum Lind <calumlind+deluge@gmail.com>
- Use named placeholders to allow translators to change text order.
- Refactor fpcnt to use format() and markup strings properly.
- Remove gettext.js creatation from generate_pot script since this step
is now always done at build-time.
The following error was encountered in GTK3 which is a result of trying
to cast a very large ETA value to C int and raising an Overlflow error.
<type 'exceptions.OverflowError'>: 3072227291 not in range -2147483648 to 2147483647
The solution is to limit the ETA to 1 year and represent any values over
that as -1 which the UIs can display as infinity.
Remove extra_requires since requirements-*.txt files provides these now
plus the extras_requires is for extras at installation time which does
not apply to docs or dev.
For test_requires include the minimal requirements for pytest to run.
This is not the same as the longer tox test requirements that include
linting, docs etcs.
Fix license field.
- Add an install_requires list to allow dependencies to be automatically
installed via setuptools or pip installation.
- Needed a workaround for twisted service_identity install.
- Tweaked the layout a bit with optional part of dependency description.
- Updated descriptions to help understand dependency usage.
- Made intltool and chardet packages optional. This will help with
installation where these might be missing and are not crucial.
- Remove gettext from dependency as is part of Python.
- License should be GPL3+.
- Use the Deluge HTTPS URLs.
- Wrap the description long lines.
- Add launchable tag to launch from software centre after install.
- Removed name tag to use .desktop file name field instead.
The new get_name method needs to function as it did in 1.3-stable so
when renaming a torrent file or top-level folder update torrent name
to reflect that. If UI supports renaming the torrent then the options
value with be used instead.
The refactoring of the options code didn't update to use the new status
key `auto_managed` and `is_auto_managed` is not a valid torrent option.
Related: 1637da84e4
The exclude 'deluge.tests' in setup.py is for excluding from package
installation but we want the files when distributing as source so
include the entire deluge/tests dir (excluding .pyc files).
Use markdown to aid readability.
Update the README and use it for the long_description in setup.py
Add detailed requirement information to the DEPENDS files.
When installed as a development version there was no way to load the
normal js scripts so improve the debug arg handling by parsing for false
values to force use of normal type scripts. Since debug arg overrides dev,
leave dev as is.
All of the non-standard docstring file headers were being added to
minified files, increasing the file size. Replace with jsdoc `/**`.
Remove ext-extensions from git as will be generated by minify script.
Prettier removed too many parentheses in asIPAdress and when minifying
the plus unary combined with the plus operator to create an incorrect
increment operator. So skip prettier formatting this secion of code.
Also fix an ECMA5 warning from closure about function scope in FilesTab.
- When both minifying modules are missing, creating a copy of the debug
file is not actually desirable, a missing file is more obvious than a copy.
WebUI can handle a missing 'normal' script and fallback to 'debug' script so
modified script to skip and warn instead.
Packaging:
- Decided that the debug files are useful for end-user so keep them in
package installation. For debug script_type to be usable all debug
file need to be avaialble so extjs debug files also included.
Script type selection:
- Fixed dev and debug request args to be properly decoded on Python 3,
otherwise comparison would fail and allowed any case for values.
- Modified the choosing of the script type to pick debug if specified
as previously always choosing dev type if dev version was True. A rare
scenario but useful but now debug is used if specified otherwise use dev.
- Changed the order when looking for alternative script types to start
with dev so that if debug is specified but missing it uses a similar
script type as previously would fallback to normal which is likely
undesired.
In TorrentView and PeersTabView, when moving the mouse pointer through
rows, the value in download and upload speed columns can be overwritten
by value from previous row cell.
The solution is to disable return by cached condition in function
cell_data_speed.
Discussion: https://github.com/deluge-torrent/deluge/pull/200#issuecomment-424907571
Removed debugging code.
Co-authored-by: Calum Lind <calumlind+deluge@gmail.com>
Refactoring for Python 3 did not account for the `self._files` using an
updated info_dict so simplify code by updating in the files for loop.
Added test that TorrentInfo.files returns the correct structure.
In a previous commit d4023e7dde removed a decode for Python 3 but
with translated text returned by gettext encoded on Python 2 the
escape function would raise a UnicodeDecodeError trying to use ascii
to decode.
The fix is to decode the message returned by gettext on Python 2.
- Use the apt addon for installing libtorrent package.
- Start the py3 test sooner as it is slow to complete.
- Add if conditions for gtkui test dependencies.
- Remove Appveyor tests that are taken care of by Travis.
Add a new `multiple` field to FileUploadField to allow selecting
multiple files. Include a fallback for if browser does not support
multiple file selection.
Update Add window to upload and parse multiple torrent files at once.