arsenetar/dupeguru
1
0
Fork 0
mirror of https://github.com/arsenetar/dupeguru.git synced 2024-10-31 22:05:58 +00:00
Code Releases Wiki Activity
456a835285
dupeguru/core/tests/scanner_test.py
Virgil Dupras 58da335b17 Enum-ified Scan Type constants, looks nicer.
2010-08-14 19:52:23 +02:00

479 lines
16 KiB
Python
Raw Blame History

# Created By: Virgil Dupras
# Created On: 2006/03/03
# Copyright 2010 Hardcoded Software (http://www.hardcoded.net)
#
# This software is licensed under the "HS" License as described in the "LICENSE" file,
# which should be included with this package. The terms are also available at
# http://www.hardcoded.net/licenses/hs_license
from hscommon import job
from hsutil import io
from hsutil.path import Path
from hsutil.testutil import eq_
from hsutil.testcase import TestCase
from .. import fs
from ..engine import getwords, Match
from ..ignore import IgnoreList
from ..scanner import *
class NamedObject(object):
def __init__(self, name="foobar", size=1):
self.name = name
self.size = size
self.path = Path('')
self.words = getwords(name)
def __repr__(self):
return '<NamedObject %r>' % self.name
no = NamedObject
#--- Scanner
class ScannerTestFakeFiles(TestCase):
def setUp(self):
# This is a hack to avoid invalidating all previous tests since the scanner started to test
# for file existence before doing the match grouping.
self.mock(io, 'exists', lambda _: True)
def test_empty(self):
s = Scanner()
r = s.GetDupeGroups([])
eq_(r, [])
def test_default_settings(self):
s = Scanner()
eq_(s.min_match_percentage, 80)
eq_(s.scan_type, ScanType.Filename)
eq_(s.mix_file_kind, True)
eq_(s.word_weighting, False)
eq_(s.match_similar_words, False)
assert isinstance(s.ignore_list, IgnoreList)
def test_simple_with_default_settings(self):
s = Scanner()
f = [no('foo bar'), no('foo bar'), no('foo bleh')]
r = s.GetDupeGroups(f)
eq_(len(r), 1)
g = r[0]
#'foo bleh' cannot be in the group because the default min match % is 80
eq_(len(g), 2)
assert g.ref in f[:2]
assert g.dupes[0] in f[:2]
def test_simple_with_lower_min_match(self):
s = Scanner()
s.min_match_percentage = 50
f = [no('foo bar'), no('foo bar'), no('foo bleh')]
r = s.GetDupeGroups(f)
eq_(len(r), 1)
g = r[0]
eq_(len(g), 3)
def test_trim_all_ref_groups(self):
# When all files of a group are ref, don't include that group in the results, but also don't
# count the files from that group as discarded.
s = Scanner()
f = [no('foo'), no('foo'), no('bar'), no('bar')]
f[2].is_ref = True
f[3].is_ref = True
r = s.GetDupeGroups(f)
eq_(len(r), 1)
eq_(s.discarded_file_count, 0)
def test_priorize(self):
s = Scanner()
f = [no('foo'), no('foo'), no('bar'), no('bar')]
f[1].size = 2
f[2].size = 3
f[3].is_ref = True
r = s.GetDupeGroups(f)
g1, g2 = r
assert f[1] in (g1.ref,g2.ref)
assert f[0] in (g1.dupes[0],g2.dupes[0])
assert f[3] in (g1.ref,g2.ref)
assert f[2] in (g1.dupes[0],g2.dupes[0])
def test_content_scan(self):
s = Scanner()
s.scan_type = ScanType.Contents
f = [no('foo'), no('bar'), no('bleh')]
f[0].md5 = f[0].md5partial = 'foobar'
f[1].md5 = f[1].md5partial = 'foobar'
f[2].md5 = f[2].md5partial = 'bleh'
r = s.GetDupeGroups(f)
eq_(len(r), 1)
eq_(len(r[0]), 2)
eq_(s.discarded_file_count, 0) # don't count the different md5 as discarded!
def test_content_scan_compare_sizes_first(self):
class MyFile(no):
@property
def md5(file):
raise AssertionError()
s = Scanner()
s.scan_type = ScanType.Contents
f = [MyFile('foo', 1), MyFile('bar', 2)]
eq_(len(s.GetDupeGroups(f)), 0)
def test_min_match_perc_doesnt_matter_for_content_scan(self):
s = Scanner()
s.scan_type = ScanType.Contents
f = [no('foo'), no('bar'), no('bleh')]
f[0].md5 = f[0].md5partial = 'foobar'
f[1].md5 = f[1].md5partial = 'foobar'
f[2].md5 = f[2].md5partial = 'bleh'
s.min_match_percentage = 101
r = s.GetDupeGroups(f)
eq_(len(r), 1)
eq_(len(r[0]), 2)
s.min_match_percentage = 0
r = s.GetDupeGroups(f)
eq_(len(r), 1)
eq_(len(r[0]), 2)
def test_content_scan_doesnt_put_md5_in_words_at_the_end(self):
s = Scanner()
s.scan_type = ScanType.Contents
f = [no('foo'),no('bar')]
f[0].md5 = f[0].md5partial = '\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f'
f[1].md5 = f[1].md5partial = '\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f'
r = s.GetDupeGroups(f)
g = r[0]
def test_extension_is_not_counted_in_filename_scan(self):
s = Scanner()
s.min_match_percentage = 100
f = [no('foo.bar'), no('foo.bleh')]
r = s.GetDupeGroups(f)
eq_(len(r), 1)
eq_(len(r[0]), 2)
def test_job(self):
def do_progress(progress, desc=''):
log.append(progress)
return True
s = Scanner()
log = []
f = [no('foo bar'), no('foo bar'), no('foo bleh')]
r = s.GetDupeGroups(f, job.Job(1, do_progress))
eq_(log[0], 0)
eq_(log[-1], 100)
def test_mix_file_kind(self):
s = Scanner()
s.mix_file_kind = False
f = [no('foo.1'), no('foo.2')]
r = s.GetDupeGroups(f)
eq_(len(r), 0)
def test_word_weighting(self):
s = Scanner()
s.min_match_percentage = 75
s.word_weighting = True
f = [no('foo bar'), no('foo bar bleh')]
r = s.GetDupeGroups(f)
eq_(len(r), 1)
g = r[0]
m = g.get_match_of(g.dupes[0])
eq_(m.percentage, 75) # 16 letters, 12 matching
def test_similar_words(self):
s = Scanner()
s.match_similar_words = True
f = [no('The White Stripes'), no('The Whites Stripe'), no('Limp Bizkit'), no('Limp Bizkitt')]
r = s.GetDupeGroups(f)
eq_(len(r), 2)
def test_fields(self):
s = Scanner()
s.scan_type = ScanType.Fields
f = [no('The White Stripes - Little Ghost'), no('The White Stripes - Little Acorn')]
r = s.GetDupeGroups(f)
eq_(len(r), 0)
def test_fields_no_order(self):
s = Scanner()
s.scan_type = ScanType.FieldsNoOrder
f = [no('The White Stripes - Little Ghost'), no('Little Ghost - The White Stripes')]
r = s.GetDupeGroups(f)
eq_(len(r), 1)
def test_tag_scan(self):
s = Scanner()
s.scan_type = ScanType.Tag
o1 = no('foo')
o2 = no('bar')
o1.artist = 'The White Stripes'
o1.title = 'The Air Near My Fingers'
o2.artist = 'The White Stripes'
o2.title = 'The Air Near My Fingers'
r = s.GetDupeGroups([o1,o2])
eq_(len(r), 1)
def test_tag_with_album_scan(self):
s = Scanner()
s.scan_type = ScanType.Tag
s.scanned_tags = set(['artist', 'album', 'title'])
o1 = no('foo')
o2 = no('bar')
o3 = no('bleh')
o1.artist = 'The White Stripes'
o1.title = 'The Air Near My Fingers'
o1.album = 'Elephant'
o2.artist = 'The White Stripes'
o2.title = 'The Air Near My Fingers'
o2.album = 'Elephant'
o3.artist = 'The White Stripes'
o3.title = 'The Air Near My Fingers'
o3.album = 'foobar'
r = s.GetDupeGroups([o1,o2,o3])
eq_(len(r), 1)
def test_that_dash_in_tags_dont_create_new_fields(self):
s = Scanner()
s.scan_type = ScanType.Tag
s.scanned_tags = set(['artist', 'album', 'title'])
s.min_match_percentage = 50
o1 = no('foo')
o2 = no('bar')
o1.artist = 'The White Stripes - a'
o1.title = 'The Air Near My Fingers - a'
o1.album = 'Elephant - a'
o2.artist = 'The White Stripes - b'
o2.title = 'The Air Near My Fingers - b'
o2.album = 'Elephant - b'
r = s.GetDupeGroups([o1,o2])
eq_(len(r), 1)
def test_tag_scan_with_different_scanned(self):
s = Scanner()
s.scan_type = ScanType.Tag
s.scanned_tags = set(['track', 'year'])
o1 = no('foo')
o2 = no('bar')
o1.artist = 'The White Stripes'
o1.title = 'some title'
o1.track = 'foo'
o1.year = 'bar'
o2.artist = 'The White Stripes'
o2.title = 'another title'
o2.track = 'foo'
o2.year = 'bar'
r = s.GetDupeGroups([o1, o2])
eq_(len(r), 1)
def test_tag_scan_only_scans_existing_tags(self):
s = Scanner()
s.scan_type = ScanType.Tag
s.scanned_tags = set(['artist', 'foo'])
o1 = no('foo')
o2 = no('bar')
o1.artist = 'The White Stripes'
o1.foo = 'foo'
o2.artist = 'The White Stripes'
o2.foo = 'bar'
r = s.GetDupeGroups([o1, o2])
eq_(len(r), 1) # Because 'foo' is not scanned, they match
def test_tag_scan_converts_to_str(self):
s = Scanner()
s.scan_type = ScanType.Tag
s.scanned_tags = set(['track'])
o1 = no('foo')
o2 = no('bar')
o1.track = 42
o2.track = 42
try:
r = s.GetDupeGroups([o1, o2])
except TypeError:
raise AssertionError()
eq_(len(r), 1)
def test_tag_scan_non_ascii(self):
s = Scanner()
s.scan_type = ScanType.Tag
s.scanned_tags = set(['title'])
o1 = no('foo')
o2 = no('bar')
o1.title = 'foobar\u00e9'
o2.title = 'foobar\u00e9'
try:
r = s.GetDupeGroups([o1, o2])
except UnicodeEncodeError:
raise AssertionError()
eq_(len(r), 1)
def test_audio_content_scan(self):
s = Scanner()
s.scan_type = ScanType.ContentsAudio
f = [no('foo'), no('bar'), no('bleh')]
f[0].md5 = 'foo'
f[1].md5 = 'bar'
f[2].md5 = 'bleh'
f[0].md5partial = 'foo'
f[1].md5partial = 'foo'
f[2].md5partial = 'bleh'
f[0].audiosize = 1
f[1].audiosize = 1
f[2].audiosize = 1
r = s.GetDupeGroups(f)
eq_(len(r), 1)
eq_(len(r[0]), 2)
def test_audio_content_scan_compare_sizes_first(self):
class MyFile(no):
@property
def md5partial(file):
raise AssertionError()
s = Scanner()
s.scan_type = ScanType.ContentsAudio
f = [MyFile('foo'), MyFile('bar')]
f[0].audiosize = 1
f[1].audiosize = 2
eq_(len(s.GetDupeGroups(f)), 0)
def test_ignore_list(self):
s = Scanner()
f1 = no('foobar')
f2 = no('foobar')
f3 = no('foobar')
f1.path = Path('dir1/foobar')
f2.path = Path('dir2/foobar')
f3.path = Path('dir3/foobar')
s.ignore_list.Ignore(str(f1.path),str(f2.path))
s.ignore_list.Ignore(str(f1.path),str(f3.path))
r = s.GetDupeGroups([f1,f2,f3])
eq_(len(r), 1)
g = r[0]
eq_(len(g.dupes), 1)
assert f1 not in g
assert f2 in g
assert f3 in g
# Ignored matches are not counted as discarded
eq_(s.discarded_file_count, 0)
def test_ignore_list_checks_for_unicode(self):
#scanner was calling path_str for ignore list checks. Since the Path changes, it must
#be unicode(path)
s = Scanner()
f1 = no('foobar')
f2 = no('foobar')
f3 = no('foobar')
f1.path = Path('foo1\u00e9')
f2.path = Path('foo2\u00e9')
f3.path = Path('foo3\u00e9')
s.ignore_list.Ignore(str(f1.path),str(f2.path))
s.ignore_list.Ignore(str(f1.path),str(f3.path))
r = s.GetDupeGroups([f1,f2,f3])
eq_(len(r), 1)
g = r[0]
eq_(len(g.dupes), 1)
assert f1 not in g
assert f2 in g
assert f3 in g
def test_file_evaluates_to_false(self):
# A very wrong way to use any() was added at some point, causing resulting group list
# to be empty.
class FalseNamedObject(NamedObject):
def __bool__(self):
return False
s = Scanner()
f1 = FalseNamedObject('foobar')
f2 = FalseNamedObject('foobar')
r = s.GetDupeGroups([f1, f2])
eq_(len(r), 1)
def test_size_threshold(self):
# Only file equal or higher than the size_threshold in size are scanned
s = Scanner()
f1 = no('foo', 1)
f2 = no('foo', 2)
f3 = no('foo', 3)
s.size_threshold = 2
groups = s.GetDupeGroups([f1,f2,f3])
eq_(len(groups), 1)
[group] = groups
eq_(len(group), 2)
assert f1 not in group
assert f2 in group
assert f3 in group
def test_tie_breaker_path_deepness(self):
# If there is a tie in prioritization, path deepness is used as a tie breaker
s = Scanner()
o1, o2 = no('foo'), no('foo')
o1.path = Path('foo')
o2.path = Path('foo/bar')
[group] = s.GetDupeGroups([o1, o2])
assert group.ref is o2
def test_tie_breaker_copy(self):
# if copy is in the words used (even if it has a deeper path), it becomes a dupe
s = Scanner()
o1, o2 = no('foo bar Copy'), no('foo bar')
o1.path = Path('deeper/path')
o2.path = Path('foo')
[group] = s.GetDupeGroups([o1, o2])
assert group.ref is o2
def test_tie_breaker_same_name_plus_digit(self):
# if ref has the same words as dupe, but has some just one extra word which is a digit, it
# becomes a dupe
s = Scanner()
o1 = no('foo bar 42')
o2 = no('foo bar [42]')
o3 = no('foo bar (42)')
o4 = no('foo bar {42}')
o5 = no('foo bar')
# all numbered names have deeper paths, so they'll end up ref if the digits aren't correctly
# used as tie breakers
o1.path = Path('deeper/path')
o2.path = Path('deeper/path')
o3.path = Path('deeper/path')
o4.path = Path('deeper/path')
o5.path = Path('foo')
[group] = s.GetDupeGroups([o1, o2, o3, o4, o5])
assert group.ref is o5
def test_partial_group_match(self):
# Count the number od discarded matches (when a file doesn't match all other dupes of the
# group) in Scanner.discarded_file_count
s = Scanner()
o1, o2, o3 = no('a b'), no('a'), no('b')
s.min_match_percentage = 50
[group] = s.GetDupeGroups([o1, o2, o3])
eq_(len(group), 2)
assert o1 in group
assert o2 in group
assert o3 not in group
eq_(s.discarded_file_count, 1)
class ScannerTest(TestCase):
def test_dont_group_files_that_dont_exist(self):
# when creating groups, check that files exist first. It's possible that these files have
# been moved during the scan by the user.
# In this test, we have to delete one of the files between the get_matches() part and the
# get_groups() part.
s = Scanner()
s.scan_type = ScanType.Contents
p = self.tmppath()
io.open(p + 'file1', 'w').write('foo')
io.open(p + 'file2', 'w').write('foo')
file1, file2 = fs.get_files(p)
def getmatches(*args, **kw):
io.remove(file2.path)
return [Match(file1, file2, 100)]
s._getmatches = getmatches
assert not s.GetDupeGroups([file1, file2])
View Git Blame Copy Permalink