refactoring: take ignore_list out of Scanner class

It's now `DupeGuru` that holds it and passes it to `get_dupe_groups()`,
the only place where it's actually used in `Scanner`.

This will make the SE/ME/PE merge easier by allowing us to instantiate
the Scanner on-the-fly since it doesn't hold state anymore.

core/app.py

@@ -1,6 +1,4 @@
-# Created By: Virgil Dupras
-# Created On: 2006/11/11
-# Copyright 2015 Hardcoded Software (http://www.hardcoded.net)
+# Copyright 2016 Hardcoded Software (http://www.hardcoded.net)
 #
 # This software is licensed under the "GPLv3" License as described in the "LICENSE" file,
 # which should be included with this package. The terms are also available at
@@ -26,6 +24,7 @@ from hscommon.plat import ISWINDOWS
 from hscommon import desktop
 
 from . import directories, results, scanner, export, fs
+from .ignore import IgnoreList
 from .gui.deletion_options import DeletionOptions
 from .gui.details_panel import DetailsPanel
 from .gui.directory_tree import DirectoryTree
@@ -168,6 +167,7 @@ class DupeGuru(Broadcaster):
             os.makedirs(self.appdata)
         self.directories = directories.Directories()
         self.results = results.Results(self)
+        self.ignore_list = IgnoreList()
         self.scanner = self.SCANNER_CLASS()
         self.options = {
             'escape_filter_regexp': True,
@@ -373,7 +373,7 @@ class DupeGuru(Broadcaster):
             g = self.results.get_group_of_duplicate(dupe)
             for other in g:
                 if other is not dupe:
-                    self.scanner.ignore_list.Ignore(str(other.path), str(dupe.path))
+                    self.ignore_list.Ignore(str(other.path), str(dupe.path))
         self.remove_duplicates(dupes)
         self.ignore_list_dialog.refresh()
 
@@ -531,7 +531,7 @@ class DupeGuru(Broadcaster):
         self.directories.load_from_file(op.join(self.appdata, 'last_directories.xml'))
         self.notify('directories_changed')
         p = op.join(self.appdata, 'ignore_list.xml')
-        self.scanner.ignore_list.load_from_xml(p)
+        self.ignore_list.load_from_xml(p)
         self.ignore_list_dialog.refresh()
 
     def load_from(self, filename):
@@ -620,7 +620,7 @@ class DupeGuru(Broadcaster):
     def purge_ignore_list(self):
         """Remove files that don't exist from :attr:`ignore_list`.
         """
-        self.scanner.ignore_list.Filter(lambda f, s: op.exists(f) and op.exists(s))
+        self.ignore_list.Filter(lambda f, s: op.exists(f) and op.exists(s))
         self.ignore_list_dialog.refresh()
 
     def remove_directories(self, indexes):
@@ -713,7 +713,7 @@ class DupeGuru(Broadcaster):
             os.makedirs(self.appdata)
         self.directories.save_to_file(op.join(self.appdata, 'last_directories.xml'))
         p = op.join(self.appdata, 'ignore_list.xml')
-        self.scanner.ignore_list.save_to_xml(p)
+        self.ignore_list.save_to_xml(p)
         self.notify('save_session')
 
     def save_as(self, filename):
@@ -740,7 +740,7 @@ class DupeGuru(Broadcaster):
             if self.options['ignore_hardlink_matches']:
                 files = self._remove_hardlink_dupes(files)
             logging.info('Scanning %d files' % len(files))
-            self.results.groups = self.scanner.get_dupe_groups(files, j)
+            self.results.groups = self.scanner.get_dupe_groups(files, self.ignore_list, j)
 
         if not self.directories.has_any_file():
             self.view.show_message(tr("The selected directories contain no scannable file."))

core/gui/ignore_list_dialog.py

@@ -15,7 +15,7 @@ class IgnoreListDialog:
 
     def __init__(self, app):
         self.app = app
-        self.ignore_list = self.app.scanner.ignore_list
+        self.ignore_list = self.app.ignore_list
         self.ignore_list_table = IgnoreListTable(self)
 
     def clear(self):

core/scanner.py

@@ -14,7 +14,6 @@ from hscommon.util import dedupe, rem_file_ext, get_file_ext
 from hscommon.trans import tr
 
 from . import engine
-from .ignore import IgnoreList
 
 # It's quite ugly to have scan types from all editions all put in the same class, but because there's
 # there will be some nasty bugs popping up (ScanType is used in core when in should exclusively be
@@ -71,7 +70,6 @@ def remove_dupe_paths(files):
 
 class Scanner:
     def __init__(self):
-        self.ignore_list = IgnoreList()
         self.discarded_file_count = 0
 
     def _getmatches(self, files, j):
@@ -133,7 +131,7 @@ class Scanner:
         """
         raise NotImplementedError()
 
-    def get_dupe_groups(self, files, j=job.nulljob):
+    def get_dupe_groups(self, files, ignore_list=None, j=job.nulljob):
         j = j.start_subjob([8, 2])
         for f in (f for f in files if not hasattr(f, 'is_ref')):
             f.is_ref = False
@@ -163,12 +161,12 @@ class Scanner:
             matches = [m for m in matches if get_file_ext(m.first.name) == get_file_ext(m.second.name)]
         matches = [m for m in matches if m.first.path.exists() and m.second.path.exists()]
         matches = [m for m in matches if not (m.first.is_ref and m.second.is_ref)]
-        if self.ignore_list:
+        if ignore_list:
             j = j.start_subjob(2)
             iter_matches = j.iter_with_progress(matches, tr("Processed %d/%d matches against the ignore list"))
             matches = [
                 m for m in iter_matches
-                if not self.ignore_list.AreIgnored(str(m.first.path), str(m.second.path))
+                if not ignore_list.AreIgnored(str(m.first.path), str(m.second.path))
             ]
         logging.info('Grouping matches')
         groups = engine.get_groups(matches, j)

core/tests/app_test.py

@@ -1,6 +1,4 @@
-# Created By: Virgil Dupras
-# Created On: 2007-06-23
-# Copyright 2015 Hardcoded Software (http://www.hardcoded.net)
+# Copyright 2016 Hardcoded Software (http://www.hardcoded.net)
 #
 # This software is licensed under the "GPLv3" License as described in the "LICENSE" file,
 # which should be included with this package. The terms are also available at
@@ -14,10 +12,10 @@ from pytest import mark
 from hscommon.path import Path
 import hscommon.conflict
 import hscommon.util
-from hscommon.testutil import CallLogger, eq_, log_calls
+from hscommon.testutil import eq_, log_calls
 from hscommon.jobprogress.job import Job
 
-from .base import DupeGuru, TestApp
+from .base import TestApp
 from .results_test import GetTestGroups
 from .. import app, fs, engine
 from ..scanner import ScanType
@@ -347,11 +345,11 @@ class TestCaseDupeGuruWithResults:
         app = self.app
         self.rtable.select([4]) #The dupe of the second, 2 sized group
         app.add_selected_to_ignore_list()
-        eq_(len(app.scanner.ignore_list), 1)
+        eq_(len(app.ignore_list), 1)
         self.rtable.select([1]) #first dupe of the 3 dupes group
         app.add_selected_to_ignore_list()
         #BOTH the ref and the other dupe should have been added
-        eq_(len(app.scanner.ignore_list), 3)
+        eq_(len(app.ignore_list), 3)
 
     def test_purgeIgnoreList(self, do_setup, tmpdir):
         app = self.app
@@ -360,13 +358,13 @@ class TestCaseDupeGuruWithResults:
         open(p1, 'w').close()
         open(p2, 'w').close()
         dne = '/does_not_exist'
-        app.scanner.ignore_list.Ignore(dne,p1)
-        app.scanner.ignore_list.Ignore(p2,dne)
-        app.scanner.ignore_list.Ignore(p1,p2)
+        app.ignore_list.Ignore(dne,p1)
+        app.ignore_list.Ignore(p2,dne)
+        app.ignore_list.Ignore(p1,p2)
         app.purge_ignore_list()
-        eq_(1,len(app.scanner.ignore_list))
-        assert app.scanner.ignore_list.AreIgnored(p1,p2)
-        assert not app.scanner.ignore_list.AreIgnored(dne,p1)
+        eq_(1,len(app.ignore_list))
+        assert app.ignore_list.AreIgnored(p1,p2)
+        assert not app.ignore_list.AreIgnored(dne,p1)
 
     def test_only_unicode_is_added_to_ignore_list(self, do_setup):
         def FakeIgnore(first,second):
@@ -376,7 +374,7 @@ class TestCaseDupeGuruWithResults:
                 self.fail()
 
         app = self.app
-        app.scanner.ignore_list.Ignore = FakeIgnore
+        app.ignore_list.Ignore = FakeIgnore
         self.rtable.select([4])
         app.add_selected_to_ignore_list()
 

core/tests/scanner_test.py

@@ -1,6 +1,4 @@
-# Created By: Virgil Dupras
-# Created On: 2006/03/03
-# Copyright 2015 Hardcoded Software (http://www.hardcoded.net)
+# Copyright 2016 Hardcoded Software (http://www.hardcoded.net)
 #
 # This software is licensed under the "GPLv3" License as described in the "LICENSE" file,
 # which should be included with this package. The terms are also available at
@@ -13,7 +11,7 @@ from hscommon.testutil import eq_
 from .. import fs
 from ..engine import getwords, Match
 from ..ignore import IgnoreList
-from ..scanner import *
+from ..scanner import Scanner, ScanType
 
 class NamedObject:
     def __init__(self, name="foobar", size=1, path=None):
@@ -50,7 +48,6 @@ def test_default_settings(fake_fileexists):
     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(fake_fileexists):
     s = Scanner()
@@ -142,7 +139,7 @@ def test_content_scan_doesnt_put_md5_in_words_at_the_end(fake_fileexists):
     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.get_dupe_groups(f)
-    g = r[0]
+    r[0]
 
 def test_extension_is_not_counted_in_filename_scan(fake_fileexists):
     s = Scanner()
@@ -160,7 +157,7 @@ def test_job(fake_fileexists):
     s = Scanner()
     log = []
     f = [no('foo bar'), no('foo bar'), no('foo bleh')]
-    r = s.get_dupe_groups(f, job.Job(1, do_progress))
+    s.get_dupe_groups(f, j=job.Job(1, do_progress))
     eq_(log[0], 0)
     eq_(log[-1], 100)
 
@@ -346,9 +343,10 @@ def test_ignore_list(fake_fileexists):
     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.get_dupe_groups([f1,f2,f3])
+    ignore_list = IgnoreList()
+    ignore_list.Ignore(str(f1.path),str(f2.path))
+    ignore_list.Ignore(str(f1.path),str(f3.path))
+    r = s.get_dupe_groups([f1,f2,f3], ignore_list=ignore_list)
     eq_(len(r), 1)
     g = r[0]
     eq_(len(g.dupes), 1)
@@ -368,9 +366,10 @@ def test_ignore_list_checks_for_unicode(fake_fileexists):
     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.get_dupe_groups([f1,f2,f3])
+    ignore_list = IgnoreList()
+    ignore_list.Ignore(str(f1.path),str(f2.path))
+    ignore_list.Ignore(str(f1.path),str(f3.path))
+    r = s.get_dupe_groups([f1,f2,f3], ignore_list=ignore_list)
     eq_(len(r), 1)
     g = r[0]
     eq_(len(g.dupes), 1)