- We introduce a new submodule of beets.util named id_extractors.
- Parts of the ID extraction utilites required by metadata source plugins
should live there.
- Also this enables future usage of those utilities from the "outside" of
metadata source plugins.
- Move Discogs ID extractor to the new module and change test_discogs to use
the new location.
- Add spotify_id_regex variable to the new module.
- samefile exists on all platforms for recent python
- don't rely on monkey-patching os/os.path and on specifics on the
implementation: as a result of doing so, the tests start failing in
obscure ways as soon as the implementation (and its usage of
os.path.exists and os.path.samefile) is changed
- move tests for case_sensitive to test_util.py, since this is not
really the concern of PathQueryTest
- removes part of the tests, since the tests that patch os.path.samefile
and os.path.exists are super brittle since they test the
implementation rather than the functionality of case_sensitive().
This is a prepartory step for actually changing the implementation,
which would otherwise break the tests in a confusing way...
This was a helper for situations when Python 2 and 3 APIs returned bytes
and unicode, respectively. In these situation, we should nowadays know
which of the two we receive, so there's no need to wrap & hide the
`bytes.decode()` anymore (when it is still required).
Detailed justification:
beets/ui/__init__.py:
- command line options are always parsed to str
beets/ui/commands.py:
- confuse's config.dump always returns str
- open(...) defaults to text mode, read()ing str
beetsplug/keyfinder.py:
- ...
beetsplug/web/__init__.py:
- internally, paths are always bytestrings
- additionally, I took the liberty to slighlty re-arrange the code: it
makes sense to split off the basename first, since we're only
interested in the unicode conversion of that part.
test/helper.py:
- capture_stdout() gives a StringIO, which yields str
test/test_ui.py:
- self.io, from _common.TestCase, ultimately contains a
_common.DummyOut, which appears to be dealing with str (cf.
DummyOut.get)
used to work due to inconsistent mediafile implementation, but with
https://github.com/beetbox/mediafile/pull/64 (in mediafile >= 0.11.0)
list fields are None if non-existent, not the empty list
On Windows, converting command-line arguments (hopefully!!!) only needs
to deal with valid strings from the OS. So it is not really relevant to
test with non-UTF-8, non-surrogate bytes.
Unidecode 1.3.5 (a yanked PyPI version) changed the behavior of
Unidecode for some specific characters:
> Remove trailing space in replacements for vulgar fractions.
As luck would have it, our tests used the 1/2 character specifically to
test the behavior when these characters decoded to contain slashes. We
now pin a sufficiently recent version of Unidecode and adapt the tests
to match the new behavior.
Quoth the responses documentation:
> querystring is matched by default
Not sure how recent this is, unfortunately---but probably 0.17.0, since
that's the version where `match_querystring` was deprecated.
Makes the dispatch to the chosen backend simpler in the thumbnails
plugin. Given that ArtResizer is not only about resizing art anymore,
these methods fit there quite nicely.
- Adds a configuration that, when enabled, will append the style to genre
- Rationale is to have more verbose genres in genre tag of players that only support genre
This didn't cause any issues since we only use the shared instance
anyway, but logically it doesn't make a lot of sense for the backends
always using ArtResizer.shared (which they should be oblivious of).
Uses a custom assertion to have more detailed output (which should help
with getting an idea what the difference between expected and actual
lyrics is), and use `subTest` to clearly associate test failure to a
backend.
In addition, this strip parenthesis from the lyrics' words. That helps
with the currently failing Tekstowo test, but doesn't entirely fix it.
Requires Python 3.4 for subTest.
Test more combinations of tags that might initially be present and
expected tags. The R128 codepath and the case of having the wrong type
of tags wasn't really tested before.
Another incorrect py2 -> py3 translation. Since python 3 attached the
traceback to the exception, this should preserve the traceback without
needing to resort to sys.exc_info
This is an incorrect translation of a python 2 reraise to python 3.
With python 3, however, we can just rely on exception chaining to get
the traceback, so get rid of the complicated re-raising entirely, with
the additional benefit that the exception from the tear-down is also
shown.
When the delete_originals was set, beets would print the following, regardless
of the presence of the quiet parameter:
convert: Removing original file /path/to/file.ext
This commit ensures that the log is only printed when quiet is not present.
Unit test may fails when path to temprorary library contains `.`; to
garantue that bug wasn't here, it forces to use one more `.` inside path.
Fixes: https://github.com/beetbox/beets/issues/4151
remove interlacing by default when resizing/down-scaling, the
`deinterlace` option is to remove interlace when otherwise no processing
would have happened.
This allows for the use of differing replacements for destinations other than
the library, which is useful for beets-alternatives in the case where
filesystem requirements differ between the two paths.
Signed-off-by: Christopher Larson <kergoth@gmail.com>
Some of these tests load plugins using beets' normal plugin loader, but
didn't call unload_plugins to tidy up afterwards. This led to any future
plugin loads being ignored until the next unload_plugins call.
This commit changes the config tests so that we always call load_plugins
on setup (to store the default beets state) and unload_plugins on
teardown (to restore the previously stored state).
This changes greatly improves the speed of `beet export` and `beet info`
when the `--include-keys` option is used. It also removes the globbing
feature of `--include-keys` that was added in #1295. (See #3762 for
discussion).
Listing all fields for an item requires querying the database to find
any flex attributes. This is slow when done for every item being
exported. We already have a way for the user to specify a fixed set
of keys, but we previously queried everything and filtered it afterwards.
The new approach is more efficient.
Code that iterates through all fields now have to handle invalid field
names. The export and info plugins output invalid fields as None.
Timings before:
> /usr/bin/time beet export -i title,path,artist -l Bob Dylan
13.26user 20.22system 0:34.01elapsed 98%CPU (0avgtext+0avgdata 52544maxresident)k
> /usr/bin/time beet export -l Bob Dylan
12.93user 20.15system 0:33.58elapsed 98%CPU (0avgtext+0avgdata 53632maxresident)k
Timings after:
> /usr/bin/time beet export -l Bob Dylan
13.33user 20.17system 0:34.02elapsed 98%CPU (0avgtext+0avgdata 53500maxresident)k
> /usr/bin/time beet export -i title,path,artist -l Bob Dylan
0.49user 0.07system 0:00.56elapsed 98%CPU (0avgtext+0avgdata 50496maxresident)k
Notice the dramatic speedup in the last example!
Squashed from the PR, relevant commit messages follow below:
Added file size option to artresizer
- In line with comments on PR, adjusted the ArtResizer API to add
functionality to "resize to X bytes" through `max_filesize` arg
- Adjustment to changelog.rst to include max_filesize change to ArtResizer
and addition of new plugin.
Added explicit tests for PIL & Imagemagick Methods
- Checks new resizing functions do reduce the filesize of images
Expose max_filesize logic to fetchart plugin
- Add syspath escaping for OS cross compatibility
- Return smaller PIL image even if max filesize not reached.
- Test resize logic against known smaller filesize (//2)
- Pass integer (not float) quality argument to PIL
- Remove Pillow from dependencies
- Implement "max_filesize" fetchart option, including
logic to resize and rescale if maxwidth is also set.
Added tests & documentation for fetchart additions.
Tests now check that a target filesize is reached with a
higher initial quality (a difficult check to pass).
With a starting quality of 95% PIL takes 4 iterations to succeed
in lowering the example cover image to 90% its original size.
To cover all bases, the PIL loop has been changed to 5 iterations
in the worst case, and the documentation altered to reflect the
50% loss in quality this implies. This seems reasonable as users
concerned about performance would most likely be persuaded to
install ImageMagick, or remove the maximum filesize constraint.
The previous 30% figure was arbitrary.
Also simplified the setup of the `readonly` value in the tests which
fixes a test ordering issue found using --random-order.
Signed-off-by: Graham R. Cobb <g+beets@cobb.uk.net>
Track item paths and album artpaths should be removed from results unless
INCLUDE_PATHS is set. This works for items but for albums the artpath is always
removed.
This patch makes the artpath removal conditional on INCLUDE_PATHS not being set
and includes a regression test. Note: the default value for INCLUDE_PATHS is
False so no changes will be seen by users unless they already have
INCLUDE_PATHS set.
Signed-off-by: Graham R. Cobb <g+beets@cobb.uk.net>
Added a test (test_get_stats) for the /stats web API.
This involved adding another item so that we can check both items and albums
are being counted correctly, which required a few small changes to some item
tests.
Signed-off-by: Graham R. Cobb <g+beets@cobb.uk.net>
Documentation says that ?expand does not need a value (i.e. ?expand=1 is wrong),
so the test is changed to reflect that syntax.
Signed-off-by: Graham R. Cobb <g+beets@cobb.uk.net>
Added test_get_album_details to test fetching /album/2?expand=1.
Also changed second album name to make tests more robust.
Signed-off-by: Graham R. Cobb <g+beets@cobb.uk.net>
self.lib.add ignores the "id" field from the entry being added and overwrites
it with the next free number. So remove the misleading id fields.
Add a test to confirm that the album query response contains the expected
album id number.
Signed-off-by: Graham R. Cobb <g+beets@cobb.uk.net>
References in the documentation to this plugin were removed in
beetbox/beets#3127 (beetbox/beets#3130) but no actual code
changes were made.
This PR removes support for this dependency entirely.
This adds support for the JSON Lines format as documented at
https://jsonlines.org/.
In this mode the data is output incrementally, whereas the other
modes load every item into memory and don't produce output until
the end.
* fetchart: Improve Cover Art Archive source.
Instead of blindly selecting the first image, we now treat all "front"
images as candidates.
This is useful where some digital releases have both an animated cover
and a still image and the animated image is the first image returned
from the API.
The previous test worked (on my machine, and on Github CI and AppVeyor),
but it is not obvious whether the order is really guaranteed (given that
the full beets database stack and sqlite are involved). Thus, to prevent
this from exploding at some point, only verify the number of deletions
for now.
MPD keeps the current track in the queue when stopping, so it's not
really like a skip, and I use it so that I can stop the music, and later
start at the beginning of a track.
I do this by keeping track of the current song id, and then comparing
them when we receive a stop signal.
* If import move is true, files will be deleted after converting.
Fixes#2947
* Removed trailing whitespace to comply with W293, fixing build
* Add period to the end of the comment
Co-Authored-By: Adrian Sampson <adrian@radbox.org>
* Added changelog entry for this fix.
* Added delete_originals option to remove source files after transcode
* Added unit test, removed redundant syspath call
Co-authored-by: Logan Arens <logan-arens@users.noreply.github.com>
Co-authored-by: Logan Arens <heresmygithub@protonmail.com>
Co-authored-by: Adrian Sampson <adrian@radbox.org>
Co-authored-by: Logan Arens <logan.arens@protonmail.com>
* clean-up code & add tests for genius lyrics backend
* add genius fetch tests
* organize imports: standard lib -> pip -> local
* check in sample genius lyrics page
* fix mock import
* force utf-8 encoding for opened files
* use io.open to force utf-8 encoding w/ python2.7
- renamed test test_format_fixed_field to test_format_fixed_field_string
- test_format_fixed_field_string not tests `field_two` with string values
- added new test_format_fixed_field_integer to test field `field_one` as INTEGER
- added new test_format_fixed_field_integer_normalized to test rounding float values
* safer version comparison
* regex bytes directly
* handle b'\x08 ...' case
* test_replaygain.py: injected command output should match the type of the actual output
* Fix unspecified `gain_adjustment` when method defined in config
* Fix difference between dB and LUFS values in case of mismatched `target_level`/`method`:
```
db_to_lufs( target_level <dB> ) - lufs_to_dB( -23 <LUFS> )
```
* Ignore single assertion in case of bs1770gain
(cherry picked from commit 2395bf224032c44f1ea5d28e0c63af96a92b96df)
Assert that the replaygain plugin does not write REPLAYGAIN_* tags but R128_*
tags, when instructed to do so.
This test is skipped for the `command` backend as it does not support OPUS.
Add replaygain backend using ffmpeg's ebur128 filter.
The album gain is calculated as the mean of all BS.1770 gating block powers.
Besides differences in gating block offset, this should be equivalent to a
BS.1770 analysis of a proper concatenation of all tracks.
Just calculating the mean of all track gains (as implemented by the bs1770gain
backend) yields incorrect results as that would:
- completely ignore track lengths
- just using length in seconds won't work either (e.g. BS.1770 ignores
passages below a threshold)
- take the mean of track loudness, not power
When using the ffmpeg replaygain backend to create R128_*_GAIN tags, the
targetlevel will be set to -23 LUFS. GitHub PullRequest #3065 will make this
configurable.
It will also skip peak calculation, as there is no R128_*_PEAK tag.
It is checked if the libavfilter library supports replaygain calculation. Before
version 6.67.100 that did require the `--enable-libebur128` compile-time-option,
after that the ebur128 library is included in libavfilter itself. Thus we
require either a recent enough libavfilter version or the `--enable-libebur128`
option.
When setting up bpd tests, two servers are startet: first a control server, then
bpd. Both send their assigned ports down a queue. The recipient only needs bpd's
port and thus skips the first queue entry.
Use a `multiprocessing.Queue` instead of a `multiprocessing.Value` to avoid the
manual polling/timeout handling.
TODO: Strangely Listener seems to be constructed twice. Only the second one is
used. Fix that and then remove the code working around it.
The bpd test bind a socket in order to test the protocol implementation. When
running concurrently this often resulted in an attempt to bind an already
occupied port.
By using the port number `0` we instead let the OS choose a free port. We then
have to extract it from the socket (which is handled by `bluelet`) via
`mock.patch`ing.
Return a namedtuple CommandOutput(stdout, stderr) instead of just stdout from
util.command_ouput, allowing separate access to stdout and stderr.
This change is required by the ffmpeg replaygain backend (GitHub
PullRequest #3056) as ffmpeg's ebur128 filter outputs only to stderr.
*All* URLs were checked manually, but only once per domain!
I mostly concerned myself with URLs in documentation rather than source
code because the latter may or may not have impactful changes, while the
former should be straight forward.
Changes in addition to simply adding an s:
- changed pip and pypi references as their location has changed
- MPoD (iOS app) url redirects to Regelian, so I replaced those
- updated homebrew references
Notable observations:
- beets.io does have HTTPS set up properly (via gh-pages)
- beatport.py uses the old HTTP url for beatport
- as does lyrics.py for lyrics.wikia.com
- https://tomahawk-player.org/ expired long ago, but the http page
redirects to https regardless
- none of the sourceforge subdomains have https (in 2019!)
Prevents reloading a model from the database when it hasn't changed.
Now we're back to almost the same speed as before the addition of album
field fallbacks.
The real MPD ignores `noidle` when the client is not idle. It doesn't
even send a successful response, just ignores the command. Although
I don't understand why a client would fail to keep track of its own
state, it seems that this is necessary to get ncmpcpp working.
The playlistid command is supposed to list the whole playlist if no
argument is provided, but we were accidentally trying to look up an
impossible negative id in that case causing an error to always be
returned.
When using pytest's test collector, any class with a name starting with
Test is collected. If it notices that the class has an `__init__` member
then it skips it with a warning since it's probably a false positive.
This isn't a big deal, but we can avoid warnings like this:
test/test_ui_importer.py:33
beets/test/test_ui_importer.py:33: PytestCollectionWarning: cannot collect test class 'TestTerminalImportSession' because it has a __init__ constructor
class TestTerminalImportSession(TerminalImportSession):
simply by renaming TestX to XFixture.
By comparing `sys.path` as setup by nose vs. testall.py it seems that we
weren't adding the top-level beets directory to the path. The script was
also previously changing the working directory before running the tests.
This is a hacky but effective way to work around a problem when running
the tests in parallel where two different test executions want to use
the same port.
This uses GStreamer APIs to extract a list of audio decoders and the
relevant MIME types and file extensions. Some clients like ncmpcpp use
this command to fetch a list of supported file extensions.
Some clients list the albums belonging to an artist by issuing the
command `list album <ARTIST NAME>`. This change inserts the tag `artist`
before the artist name so that this succeeds. Fixes#3007
A new `ControlConnection` is created each time a client connects over
a new control socket. This is used to forward events from the player,
and also for debugging utilities that are not part of the real MPD
protocol.
This new feature reuses as much infrastructure from the normal protocol
handling as possible (e.g. `Command` for parsing messages). While the
normal connection delegates to server `cmd_*` methods which are string
generators, the control connections delegate to `ctrl_*` methods defined
on the connection itself that are full coroutines.
Keep track of a list of currently-connected clients.
Use `socket.getpeername()` to get an identifier for each connection and
include this in each log message. This function is documented as not
being available on all systems, but it's unclear which systems this
involves.
Also log a message on client connect and disconnect events. If the
disconnection reason is because the client sent a blank line, match MPD
by returning a protocol error then hanging up. Escape curly braces.
Check function signature instead of using TypeError to crudely guess
that the wrong number of arguments were provided.
Prevent bpd from crashing when trying to log a traceback. The
`traceback.format_exc` function takes an optional argument which is
supposed to be an integer restricting the length of the backtrace to
show. Instead we were passing the exception object to this function and
causing a new exception to be raised.
Previously issuing the 'previous' command when at position 0 on the
playlist would cause bpd to stop playing. MPD instead just restarts the
currently playing song instead, so we now match this behaviour.
The songs are indexed starting from zero for the play command, however
the bound check was off by one. An index matching the length of the
playlist would crash the server instead of responding with an error
message over the protocol.
The repeat flag indicates that the entire playlist should be repeated.
If both the repeat and single flags are set then this triggers the old
behaviour of looping over a single track.
This command instructs bpd to stop playing when the current song
finishes. In the MPD 0.20 protocol this flag gains a value 'oneshot' but
for now we just support its older version with a boolean value.
The real MPD offers persistent playlist manipulation, storing the
playlists in a directory set in the config file. If that directory is
not available then the feature is disabled and the relevant commands all
respond with errors. Based on this, the initial support in bpd just
returns errors matching the MPD server in the disabled mode.
For playlistadd, extend the _bpd_add helper to work with playlists other
than the queue in order to support testing the real implementations of
these commands in the future.
There's a special status command for checking the replay gain mode,
which can be set to one of a short list of possible values. For now at
least we can ignore this feature, but track the setting anyway.
MPD supports a deprecated command 'volume' which was used to change the
volume by a relative amount unlike its replacement 'setvol' which uses
an absolute amount. As far as I can tell 'volume' always responds with a system
error message "No mixer".
These are a more sophisticated version of crossfade so we're free to
ignore them, at least for now. We now track the values of the two
settings, and show them in the status output. Like MPD, we suppress the
mixrampdb value if it's set to nan, which is how it signals that the
feature should be turned off.
If an MPC client is expecting a command to take an argument that bpd
isn't expecting (e.g. because of a difference in protocol versions) then
bpd currently crashes completely. Instead, do what the real MPD does and
return an error message over the protocol.
Although crossfade is not implemented in bpd, we can store the setting
and repeat is back to clients. Also log a warning that the operation is
not implemented.
The real MPD doesn't show the crossfade in status if it's zero since
that means no crossfade, so now we don't either.
The test `CommonOptionsParserCliTest.test_version` was passing with nose
but failing with pytest (see output below). The reason for the failure
seemed to be that the `test` plugin was loaded when it wasn't expected
to be loaded, changing the output of the `version` command. I'm not sur
exactly why that was happening, but since that test already inherited
from `TestHelper`, just invoking the plugin load/unload helper was
enough to fix it. I also removed the line setting the `self.lib`
variable since that's already done in the helper.
---
self = <test.test_ui.CommonOptionsParserCliTest testMethod=test_version>
def test_version(self):
l = self.run_with_output(u'version')
self.assertIn(u'Python version', l)
> self.assertIn(u'no plugins loaded', l)
E AssertionError: 'no plugins loaded' not found in 'beets version 1.4.8\nPython version 3.7.3rc1\nplugins: test\n'
test/test_ui.py:1292: AssertionError
