From ac32305532ec46981155cdfb43215463f743435b Mon Sep 17 00:00:00 2001 From: Virgil Dupras Date: Sun, 5 Oct 2014 16:31:16 -0400 Subject: [PATCH] Integrated the jobprogress library into hscommon I have a fix to make in it and it's really silly to pretend that this lib is of any use to anybody outside HS apps. Bringing it back here will make things more simple. --- build.py | 11 +- cocoalib/cocoa/__init__.py | 11 +- core/app.py | 186 ++++++++++++------------- core/directories.py | 70 +++++----- core/engine.py | 98 +++++++------- core/results.py | 98 +++++++------- core/scanner.py | 22 +-- core/tests/app_test.py | 100 +++++++------- core/tests/base.py | 50 +++---- core/tests/engine_test.py | 216 +++++++++++++++--------------- core/tests/scanner_test.py | 18 +-- core_pe/matchblock.py | 26 ++-- hscommon/__init__.py | 0 hscommon/gui/progress_window.py | 57 ++++---- hscommon/jobprogress/__init__.py | 0 hscommon/jobprogress/job.py | 160 ++++++++++++++++++++++ hscommon/jobprogress/performer.py | 72 ++++++++++ hscommon/jobprogress/qt.py | 52 +++++++ hscommon/path.py | 0 package.py | 14 +- requirements.txt | 1 - 21 files changed, 775 insertions(+), 487 deletions(-) mode change 100755 => 100644 hscommon/__init__.py create mode 100644 hscommon/jobprogress/__init__.py create mode 100644 hscommon/jobprogress/job.py create mode 100644 hscommon/jobprogress/performer.py create mode 100644 hscommon/jobprogress/qt.py mode change 100755 => 100644 hscommon/path.py diff --git a/build.py b/build.py index f8275056..fe988d29 100644 --- a/build.py +++ b/build.py @@ -1,9 +1,9 @@ # Created By: Virgil Dupras # Created On: 2009-12-30 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license import sys @@ -110,8 +110,9 @@ def build_cocoa(edition, dev): 'me': ['core_me'] + appscript_pkgs + ['hsaudiotag'], 'pe': ['core_pe'] + appscript_pkgs, }[edition] - tocopy = ['core', 'hscommon', 'cocoa/inter', 'cocoalib/cocoa', 'jobprogress', 'objp', - 'send2trash'] + specific_packages + tocopy = [ + 'core', 'hscommon', 'cocoa/inter', 'cocoalib/cocoa', 'objp', 'send2trash' + ] + specific_packages copy_packages(tocopy, pydep_folder, create_links=dev) sys.path.insert(0, 'build') extra_deps = None diff --git a/cocoalib/cocoa/__init__.py b/cocoalib/cocoa/__init__.py index 2c1ab58a..0fccffb1 100644 --- a/cocoalib/cocoa/__init__.py +++ b/cocoalib/cocoa/__init__.py @@ -2,8 +2,8 @@ # Created On: 2007-10-06 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# This software is licensed under the "BSD" 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/bsd_license import logging @@ -26,7 +26,7 @@ def autoreleasepool(func): def as_fetch(as_list, as_type, step_size=1000): """When fetching items from a very big list through applescript, the connection with the app - will timeout. This function is to circumvent that. 'as_type' is the type of the items in the + will timeout. This function is to circumvent that. 'as_type' is the type of the items in the list (found in appscript.k). If we don't pass it to the 'each' arg of 'count()', it doesn't work. applescript is rather stupid...""" result = [] @@ -66,7 +66,7 @@ def extract_tb_noline(tb): def safe_format_exception(type, value, tb): """Format exception from type, value and tb and fallback if there's a problem. - + In some cases in threaded exceptions under Cocoa, I get tracebacks targeting pyc files instead of py files, which results in traceback.format_exception() trying to print lines from pyc files and then crashing when trying to interpret that binary data as utf-8. We want a fallback in @@ -113,5 +113,6 @@ def patch_threaded_job_performer(): # _async_run, under cocoa, has to be run within an autorelease pool to prevent leaks. # You only need this patch is you use one of CocoaProxy's function (which allocate objc # structures) inside a threaded job. - from jobprogress.performer import ThreadedJobPerformer + from hscommon.jobprogress.performer import ThreadedJobPerformer ThreadedJobPerformer._async_run = autoreleasepool(ThreadedJobPerformer._async_run) + diff --git a/core/app.py b/core/app.py index 2f988f61..b1645998 100644 --- a/core/app.py +++ b/core/app.py @@ -1,9 +1,9 @@ # Created By: Virgil Dupras # Created On: 2006/11/11 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license import os @@ -15,7 +15,7 @@ import time import shutil from send2trash import send2trash -from jobprogress import job +from hscommon.jobprogress import job from hscommon.notify import Broadcaster from hscommon.path import Path from hscommon.conflict import smart_move, smart_copy @@ -78,7 +78,7 @@ def format_words(w): return '(%s)' % ', '.join(do_format(item) for item in w) else: return w.replace('\n', ' ') - + return ', '.join(do_format(item) for item in w) def format_perc(p): @@ -110,33 +110,33 @@ def fix_surrogate_encoding(s, encoding='utf-8'): class DupeGuru(Broadcaster): """Holds everything together. - + Instantiated once per running application, it holds a reference to every high-level object whose reference needs to be held: :class:`~core.results.Results`, :class:`Scanner`, :class:`~core.directories.Directories`, :mod:`core.gui` instances, etc.. - + It also hosts high level methods and acts as a coordinator for all those elements. This is why some of its methods seem a bit shallow, like for example :meth:`mark_all` and :meth:`remove_duplicates`. These methos are just proxies for a method in :attr:`results`, but they are also followed by a notification call which is very important if we want GUI elements to be correctly notified of a change in the data they're presenting. - + .. attribute:: directories - + Instance of :class:`~core.directories.Directories`. It holds the current folder selection. - + .. attribute:: results - + Instance of :class:`core.results.Results`. Holds the results of the latest scan. - + .. attribute:: selected_dupes - + List of currently selected dupes from our :attr:`results`. Whenever the user changes its selection at the UI level, :attr:`result_table` takes care of updating this attribute, so you can trust that it's always up-to-date. - + .. attribute:: result_table - + Instance of :mod:`meta-gui ` table listing the results from :attr:`results` """ #--- View interface @@ -154,7 +154,7 @@ class DupeGuru(Broadcaster): PROMPT_NAME = "dupeGuru" SCANNER_CLASS = scanner.Scanner - + def __init__(self, view): if view.get_default(DEBUG_MODE_PREFERENCE): logging.getLogger().setLevel(logging.DEBUG) @@ -185,14 +185,14 @@ class DupeGuru(Broadcaster): children = [self.result_table, self.directory_tree, self.stats_label, self.details_panel] for child in children: child.connect() - + #--- Virtual def _prioritization_categories(self): raise NotImplementedError() - + def _create_result_table(self): raise NotImplementedError() - + #--- Private def _get_dupe_sort_key(self, dupe, get_group, key, delta): if key == 'marked': @@ -212,7 +212,7 @@ class DupeGuru(Broadcaster): same = cmp_value(dupe, key) == refval result = (same, result) return result - + def _get_group_sort_key(self, group, key): if key == 'percentage': return group.percentage @@ -221,15 +221,15 @@ class DupeGuru(Broadcaster): if key == 'marked': return len([dupe for dupe in group.dupes if self.results.is_marked(dupe)]) return cmp_value(group.ref, key) - + def _do_delete(self, j, link_deleted, use_hardlinks, direct_deletion): def op(dupe): j.add_progress() return self._do_delete_dupe(dupe, link_deleted, use_hardlinks, direct_deletion) - + j.start_job(self.results.mark_count) self.results.perform_on_marked(op, True) - + def _do_delete_dupe(self, dupe, link_deleted, use_hardlinks, direct_deletion): if not dupe.path.exists(): return @@ -248,11 +248,11 @@ class DupeGuru(Broadcaster): linkfunc = os.link if use_hardlinks else os.symlink linkfunc(str(ref.path), str_path) self.clean_empty_dirs(dupe.path.parent()) - + def _create_file(self, path): # We add fs.Folder to fileclasses in case the file we're loading contains folder paths. return fs.get_file(path, self.directories.fileclasses + [fs.Folder]) - + def _get_file(self, str_path): path = Path(str_path) f = self._create_file(path) @@ -263,7 +263,7 @@ class DupeGuru(Broadcaster): return f except EnvironmentError: return None - + def _get_export_data(self): columns = [col for col in self.result_table.columns.ordered_columns if col.visible and col.name != 'marked'] @@ -276,20 +276,20 @@ class DupeGuru(Broadcaster): row.insert(0, group_id) rows.append(row) return colnames, rows - + def _results_changed(self): self.selected_dupes = [d for d in self.selected_dupes if self.results.get_group_of_duplicate(d) is not None] self.notify('results_changed') - + def _start_job(self, jobid, func, args=()): title = JOBID2TITLE[jobid] try: - self.progress_window.run(jobid, title, func, args=args) + self.progress_window.run(jobid, title, func, args=args) except job.JobInProgressError: msg = tr("A previous action is still hanging in there. You can't start a new one yet. Wait a few seconds, then try again.") self.view.show_message(msg) - + def _job_completed(self, jobid): if jobid == JobType.Scan: self._results_changed() @@ -312,7 +312,7 @@ class DupeGuru(Broadcaster): JobType.Delete: tr("All marked files were successfully sent to Trash."), }[jobid] self.view.show_message(msg) - + @staticmethod def _remove_hardlink_dupes(files): seen_inodes = set() @@ -327,19 +327,19 @@ class DupeGuru(Broadcaster): seen_inodes.add(inode) result.append(file) return result - + def _select_dupes(self, dupes): if dupes == self.selected_dupes: return self.selected_dupes = dupes self.notify('dupes_selected') - + #--- Public def add_directory(self, d): """Adds folder ``d`` to :attr:`directories`. - + Shows an error message dialog if something bad happens. - + :param str d: path of folder to add """ try: @@ -349,7 +349,7 @@ class DupeGuru(Broadcaster): self.view.show_message(tr("'{}' already is in the list.").format(d)) except directories.InvalidPathError: self.view.show_message(tr("'{}' does not exist.").format(d)) - + def add_selected_to_ignore_list(self): """Adds :attr:`selected_dupes` to :attr:`scanner`'s ignore list. """ @@ -367,10 +367,10 @@ class DupeGuru(Broadcaster): self.scanner.ignore_list.Ignore(str(other.path), str(dupe.path)) self.remove_duplicates(dupes) self.ignore_list_dialog.refresh() - + def apply_filter(self, filter): """Apply a filter ``filter`` to the results so that it shows only dupe groups that match it. - + :param str filter: filter to apply """ self.results.apply_filter(None) @@ -379,12 +379,12 @@ class DupeGuru(Broadcaster): filter = escape(filter, '*', '.') self.results.apply_filter(filter) self._results_changed() - + def clean_empty_dirs(self, path): if self.options['clean_empty_dirs']: while delete_if_empty(path, ['.DS_Store']): path = path.parent() - + def copy_or_move(self, dupe, copy: bool, destination: str, dest_type: DestType): source_path = dupe.path location_path = first(p for p in self.directories if dupe.path in p) @@ -406,20 +406,20 @@ class DupeGuru(Broadcaster): else: smart_move(source_path, dest_path) self.clean_empty_dirs(source_path.parent()) - + def copy_or_move_marked(self, copy): """Start an async move (or copy) job on marked duplicates. - + :param bool copy: If True, duplicates will be copied instead of moved """ def do(j): def op(dupe): j.add_progress() self.copy_or_move(dupe, copy, destination, desttype) - + j.start_job(self.results.mark_count) self.results.perform_on_marked(op, not copy) - + if not self.results.mark_count: self.view.show_message(MSG_NO_MARKED_DUPES) return @@ -430,7 +430,7 @@ class DupeGuru(Broadcaster): desttype = self.options['copymove_dest_type'] jobid = JobType.Copy if copy else JobType.Move self._start_job(jobid, do) - + def delete_marked(self): """Start an async job to send marked duplicates to the trash. """ @@ -443,10 +443,10 @@ class DupeGuru(Broadcaster): self.deletion_options.direct] logging.debug("Starting deletion job with args %r", args) self._start_job(JobType.Delete, self._do_delete, args=args) - + def export_to_xhtml(self): """Export current results to XHTML. - + The configuration of the :attr:`result_table` (columns order and visibility) is used to determine how the data is presented in the export. In other words, the exported table in the resulting XHTML will look just like the results table. @@ -454,10 +454,10 @@ class DupeGuru(Broadcaster): colnames, rows = self._get_export_data() export_path = export.export_to_xhtml(colnames, rows) desktop.open_path(export_path) - + def export_to_csv(self): """Export current results to CSV. - + The columns and their order in the resulting CSV file is determined in the same way as in :meth:`export_to_xhtml`. """ @@ -465,7 +465,7 @@ class DupeGuru(Broadcaster): if dest_file: colnames, rows = self._get_export_data() export.export_to_csv(dest_file, colnames, rows) - + def get_display_info(self, dupe, group, delta=False): def empty_data(): return {c.name: '---' for c in self.result_table.COLUMNS[1:]} @@ -476,10 +476,10 @@ class DupeGuru(Broadcaster): except Exception as e: logging.warning("Exception on GetDisplayInfo for %s: %s", str(dupe.path), str(e)) return empty_data() - + def invoke_custom_command(self): """Calls command in ``CustomCommand`` pref with ``%d`` and ``%r`` placeholders replaced. - + Using the current selection, ``%d`` is replaced with the currently selected dupe and ``%r`` is replaced with that dupe's ref file. If there's no selection, the command is not invoked. If the dupe is a ref, ``%d`` and ``%r`` will be the same. @@ -506,10 +506,10 @@ class DupeGuru(Broadcaster): subprocess.Popen(exename + args, shell=True, cwd=path) else: subprocess.Popen(cmd, shell=True) - + def load(self): """Load directory selection and ignore list from files in appdata. - + This method is called during startup so that directory selection and ignore list, which is persistent data, is the same as when the last session was closed (when :meth:`save` was called). @@ -519,19 +519,19 @@ class DupeGuru(Broadcaster): p = op.join(self.appdata, 'ignore_list.xml') self.scanner.ignore_list.load_from_xml(p) self.ignore_list_dialog.refresh() - + def load_from(self, filename): """Start an async job to load results from ``filename``. - + :param str filename: path of the XML file (created with :meth:`save_as`) to load """ def do(j): self.results.load_from_xml(filename, self._get_file, j) self._start_job(JobType.Load, do) - + def make_selected_reference(self): """Promote :attr:`selected_dupes` to reference position within their respective groups. - + Each selected dupe will become the :attr:`~core.engine.Group.ref` of its group. If there's more than one dupe selected for the same group, only the first (in the order currently shown in :attr:`result_table`) dupe will be promoted. @@ -560,28 +560,28 @@ class DupeGuru(Broadcaster): # do is to keep our selection index-wise (different dupe selection, but same index # selection). self.notify('results_changed_but_keep_selection') - + def mark_all(self): """Set all dupes in the results as marked. """ self.results.mark_all() self.notify('marking_changed') - + def mark_none(self): """Set all dupes in the results as unmarked. """ self.results.mark_none() self.notify('marking_changed') - + def mark_invert(self): """Invert the marked state of all dupes in the results. """ self.results.mark_invert() self.notify('marking_changed') - + def mark_dupe(self, dupe, marked): """Change marked status of ``dupe``. - + :param dupe: dupe to mark/unmark :type dupe: :class:`~core.fs.File` :param bool marked: True = mark, False = unmark @@ -591,7 +591,7 @@ class DupeGuru(Broadcaster): else: self.results.unmark(dupe) self.notify('marking_changed') - + def open_selected(self): """Open :attr:`selected_dupes` with their associated application. """ @@ -600,16 +600,16 @@ class DupeGuru(Broadcaster): return for dupe in self.selected_dupes: desktop.open_path(dupe.path) - + 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_dialog.refresh() - + def remove_directories(self, indexes): """Remove root directories at ``indexes`` from :attr:`directories`. - + :param indexes: Indexes of the directories to remove. :type indexes: list of int """ @@ -620,30 +620,30 @@ class DupeGuru(Broadcaster): self.notify('directories_changed') except IndexError: pass - + def remove_duplicates(self, duplicates): """Remove ``duplicates`` from :attr:`results`. - + Calls :meth:`~core.results.Results.remove_duplicates` and send appropriate notifications. - + :param duplicates: duplicates to remove. :type duplicates: list of :class:`~core.fs.File` """ self.results.remove_duplicates(self.without_ref(duplicates)) self.notify('results_changed_but_keep_selection') - + def remove_marked(self): """Removed marked duplicates from the results (without touching the files themselves). """ if not self.results.mark_count: self.view.show_message(MSG_NO_MARKED_DUPES) return - msg = tr("You are about to remove %d files from results. Continue?") + msg = tr("You are about to remove %d files from results. Continue?") if not self.view.ask_yes_no(msg % self.results.mark_count): return self.results.perform_on_marked(lambda x:None, True) self._results_changed() - + def remove_selected(self): """Removed :attr:`selected_dupes` from the results (without touching the files themselves). """ @@ -651,16 +651,16 @@ class DupeGuru(Broadcaster): if not dupes: self.view.show_message(MSG_NO_SELECTED_DUPES) return - msg = tr("You are about to remove %d files from results. Continue?") + msg = tr("You are about to remove %d files from results. Continue?") if not self.view.ask_yes_no(msg % len(dupes)): return self.remove_duplicates(dupes) - + def rename_selected(self, newname): """Renames the selected dupes's file to ``newname``. - + If there's more than one selected dupes, the first one is used. - + :param str newname: The filename to rename the dupe's file to. """ try: @@ -670,13 +670,13 @@ class DupeGuru(Broadcaster): except (IndexError, fs.FSError) as e: logging.warning("dupeGuru Warning: %s" % str(e)) return False - + def reprioritize_groups(self, sort_key): """Sort dupes in each group (in :attr:`results`) according to ``sort_key``. - + Called by the re-prioritize dialog. Calls :meth:`~core.engine.Group.prioritize` and, once the sorting is done, show a message that confirms the action. - + :param sort_key: The key being sent to :meth:`~core.engine.Group.prioritize` :type sort_key: f(dupe) """ @@ -687,11 +687,11 @@ class DupeGuru(Broadcaster): self._results_changed() msg = tr("{} duplicate groups were changed by the re-prioritization.").format(count) self.view.show_message(msg) - + def reveal_selected(self): if self.selected_dupes: desktop.reveal_path(self.selected_dupes[0].path) - + def save(self): if not op.exists(self.appdata): os.makedirs(self.appdata) @@ -699,17 +699,17 @@ class DupeGuru(Broadcaster): p = op.join(self.appdata, 'ignore_list.xml') self.scanner.ignore_list.save_to_xml(p) self.notify('save_session') - + def save_as(self, filename): """Save results in ``filename``. - + :param str filename: path of the file to save results (as XML) to. """ self.results.save_to_xml(filename) - + def start_scanning(self): """Starts an async job to scan for duplicates. - + Scans folders selected in :attr:`directories` and put the results in :attr:`results` """ def do(j): @@ -722,14 +722,14 @@ class DupeGuru(Broadcaster): files = self._remove_hardlink_dupes(files) logging.info('Scanning %d files' % len(files)) self.results.groups = self.scanner.get_dupe_groups(files, j) - + if not self.directories.has_any_file(): self.view.show_message(tr("The selected directories contain no scannable file.")) return self.results.groups = [] self._results_changed() self._start_job(JobType.Scan, do) - + def toggle_selected_mark_state(self): selected = self.without_ref(self.selected_dupes) if not selected: @@ -741,12 +741,12 @@ class DupeGuru(Broadcaster): for dupe in selected: markfunc(dupe) self.notify('marking_changed') - + def without_ref(self, dupes): """Returns ``dupes`` with all reference elements removed. """ return [dupe for dupe in dupes if self.results.get_group_of_duplicate(dupe).ref is not dupe] - + def get_default(self, key, fallback_value=None): result = nonone(self.view.get_default(key), fallback_value) if fallback_value is not None and not isinstance(result, type(fallback_value)): @@ -756,10 +756,10 @@ class DupeGuru(Broadcaster): except Exception: result = fallback_value return result - + def set_default(self, key, value): self.view.set_default(key, value) - + #--- Properties @property def stat_line(self): @@ -767,4 +767,4 @@ class DupeGuru(Broadcaster): if self.scanner.discarded_file_count: result = tr("%s (%d discarded)") % (result, self.scanner.discarded_file_count) return result - + diff --git a/core/directories.py b/core/directories.py index f1893bc1..31d53775 100644 --- a/core/directories.py +++ b/core/directories.py @@ -1,15 +1,15 @@ # Created By: Virgil Dupras # Created On: 2006/02/27 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license from xml.etree import ElementTree as ET import logging -from jobprogress import job +from hscommon.jobprogress import job from hscommon.path import Path from hscommon.util import FileOrPath @@ -24,7 +24,7 @@ __all__ = [ class DirectoryState: """Enum describing how a folder should be considered. - + * DirectoryState.Normal: Scan all files normally * DirectoryState.Reference: Scan files, but make sure never to delete any of them * DirectoryState.Excluded: Don't scan this folder @@ -41,10 +41,10 @@ class InvalidPathError(Exception): class Directories: """Holds user folder selection. - + Manages the selection that the user make through the folder selection dialog. It also manages folder states, and how recursion applies to them. - + Then, when the user starts the scan, :meth:`get_files` is called to retrieve all files (wrapped in :mod:`core.fs`) that have to be scanned according to the chosen folders/states. """ @@ -55,28 +55,28 @@ class Directories: self.states = {} self.fileclasses = fileclasses self.folderclass = fs.Folder - + def __contains__(self, path): for p in self._dirs: if path in p: return True return False - + def __delitem__(self,key): self._dirs.__delitem__(key) - + def __getitem__(self,key): return self._dirs.__getitem__(key) - + def __len__(self): return len(self._dirs) - + #---Private def _default_state_for_path(self, path): # Override this in subclasses to specify the state of some special folders. if path.name.startswith('.'): # hidden return DirectoryState.Excluded - + def _get_files(self, from_path, j): j.check_if_cancelled() state = self.get_state(from_path) @@ -102,7 +102,7 @@ class Directories: yield file except (EnvironmentError, fs.InvalidPath): pass - + def _get_folders(self, from_folder, j): j.check_if_cancelled() try: @@ -116,16 +116,16 @@ class Directories: yield from_folder except (EnvironmentError, fs.InvalidPath): pass - + #---Public def add_path(self, path): """Adds ``path`` to self, if not already there. - + Raises :exc:`AlreadyThereError` if ``path`` is already in self. If path is a directory containing some of the directories already present in self, ``path`` will be added, but all directories under it will be removed. Can also raise :exc:`InvalidPathError` if ``path`` does not exist. - + :param Path path: path to add """ if path in self: @@ -134,11 +134,11 @@ class Directories: raise InvalidPathError() self._dirs = [p for p in self._dirs if p not in path] self._dirs.append(path) - + @staticmethod def get_subfolders(path): """Returns a sorted list of paths corresponding to subfolders in ``path``. - + :param Path path: get subfolders from there :rtype: list of Path """ @@ -148,29 +148,29 @@ class Directories: return subpaths except EnvironmentError: return [] - + def get_files(self, j=job.nulljob): """Returns a list of all files that are not excluded. - + Returned files also have their ``is_ref`` attr set if applicable. """ for path in self._dirs: for file in self._get_files(path, j): yield file - + def get_folders(self, j=job.nulljob): """Returns a list of all folders that are not excluded. - + Returned folders also have their ``is_ref`` attr set if applicable. """ for path in self._dirs: from_folder = self.folderclass(path) for folder in self._get_folders(from_folder, j): yield folder - + def get_state(self, path): """Returns the state of ``path``. - + :rtype: :class:`DirectoryState` """ if path in self.states: @@ -183,12 +183,12 @@ class Directories: return self.get_state(parent) else: return DirectoryState.Normal - + def has_any_file(self): """Returns whether selected folders contain any file. - + Because it stops at the first file it finds, it's much faster than get_files(). - + :rtype: bool """ try: @@ -196,10 +196,10 @@ class Directories: return True except StopIteration: return False - + def load_from_file(self, infile): """Load folder selection from ``infile``. - + :param file infile: path or file pointer to XML generated through :meth:`save_to_file` """ try: @@ -222,10 +222,10 @@ class Directories: path = attrib['path'] state = attrib['value'] self.states[Path(path)] = int(state) - + def save_to_file(self, outfile): """Save folder selection as XML to ``outfile``. - + :param file outfile: path or file pointer to XML file to save to. """ with FileOrPath(outfile, 'wb') as fp: @@ -239,10 +239,10 @@ class Directories: state_node.set('value', str(state)) tree = ET.ElementTree(root) tree.write(fp, encoding='utf-8') - + def set_state(self, path, state): """Set the state of folder at ``path``. - + :param Path path: path of the target folder :param state: state to set folder to :type state: :class:`DirectoryState` @@ -253,4 +253,4 @@ class Directories: if path.is_parent_of(iter_path): del self.states[iter_path] self.states[path] = state - + diff --git a/core/engine.py b/core/engine.py index 14fe5cc6..c57eb3dc 100644 --- a/core/engine.py +++ b/core/engine.py @@ -1,9 +1,9 @@ # Created By: Virgil Dupras # Created On: 2006/01/29 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license import difflib @@ -15,7 +15,7 @@ from unicodedata import normalize from hscommon.util import flatten, multi_replace from hscommon.trans import tr -from jobprogress import job +from hscommon.jobprogress import job (WEIGHT_WORDS, MATCH_SIMILAR_WORDS, @@ -45,7 +45,7 @@ def unpack_fields(fields): def compare(first, second, flags=()): """Returns the % of words that match between ``first`` and ``second`` - + The result is a ``int`` in the range 0..100. ``first`` and ``second`` can be either a string or a list (of words). """ @@ -53,7 +53,7 @@ def compare(first, second, flags=()): return 0 if any(isinstance(element, list) for element in first): return compare_fields(first, second, flags) - second = second[:] #We must use a copy of second because we remove items from it + second = second[:] #We must use a copy of second because we remove items from it match_similar = MATCH_SIMILAR_WORDS in flags weight_words = WEIGHT_WORDS in flags joined = first + second @@ -77,9 +77,9 @@ def compare(first, second, flags=()): def compare_fields(first, second, flags=()): """Returns the score for the lowest matching :ref:`fields`. - + ``first`` and ``second`` must be lists of lists of string. Each sub-list is then compared with - :func:`compare`. + :func:`compare`. """ if len(first) != len(second): return 0 @@ -104,10 +104,10 @@ def compare_fields(first, second, flags=()): def build_word_dict(objects, j=job.nulljob): """Returns a dict of objects mapped by their words. - + objects must have a ``words`` attribute being a list of strings or a list of lists of strings (:ref:`fields`). - + The result will be a dict with words as keys, lists of objects as values. """ result = defaultdict(set) @@ -118,7 +118,7 @@ def build_word_dict(objects, j=job.nulljob): def merge_similar_words(word_dict): """Take all keys in ``word_dict`` that are similar, and merge them together. - + ``word_dict`` has been built with :func:`build_word_dict`. Similarity is computed with Python's ``difflib.get_close_matches()``, which computes the number of edits that are necessary to make a word equal to the other. @@ -138,9 +138,9 @@ def merge_similar_words(word_dict): def reduce_common_words(word_dict, threshold): """Remove all objects from ``word_dict`` values where the object count >= ``threshold`` - + ``word_dict`` has been built with :func:`build_word_dict`. - + The exception to this removal are the objects where all the words of the object are common. Because if we remove them, we will miss some duplicates! """ @@ -181,17 +181,17 @@ class Match(namedtuple('Match', 'first second percentage')): exact scan methods, such as Contents scans, this will always be 100. """ __slots__ = () - + def get_match(first, second, flags=()): #it is assumed here that first and second both have a "words" attribute percentage = compare(first.words, second.words, flags) return Match(first, second, percentage) def getmatches( - objects, min_match_percentage=0, match_similar_words=False, weight_words=False, + objects, min_match_percentage=0, match_similar_words=False, weight_words=False, no_field_order=False, j=job.nulljob): """Returns a list of :class:`Match` within ``objects`` after fuzzily matching their words. - + :param objects: List of :class:`~core.fs.File` to match. :param int min_match_percentage: minimum % of words that have to match. :param bool match_similar_words: make similar words (see :func:`merge_similar_words`) match. @@ -246,7 +246,7 @@ def getmatches( def getmatches_by_contents(files, sizeattr='size', partial=False, j=job.nulljob): """Returns a list of :class:`Match` within ``files`` if their contents is the same. - + :param str sizeattr: attibute name of the :class:`~core.fs.file` that returns the size of the file to use for comparison. :param bool partial: if true, will use the "md5partial" attribute instead of "md5" to compute @@ -278,44 +278,44 @@ class Group: This manages match pairs into groups and ensures that all files in the group match to each other. - + .. attribute:: ref - + The "reference" file, which is the file among the group that isn't going to be deleted. - + .. attribute:: ordered - + Ordered list of duplicates in the group (including the :attr:`ref`). - + .. attribute:: unordered - + Set duplicates in the group (including the :attr:`ref`). - + .. attribute:: dupes - + An ordered list of the group's duplicate, without :attr:`ref`. Equivalent to ``ordered[1:]`` - + .. attribute:: percentage - + Average match percentage of match pairs containing :attr:`ref`. """ #---Override def __init__(self): self._clear() - + def __contains__(self, item): return item in self.unordered - + def __getitem__(self, key): return self.ordered.__getitem__(key) - + def __iter__(self): return iter(self.ordered) - + def __len__(self): return len(self.ordered) - + #---Private def _clear(self): self._percentage = None @@ -324,22 +324,22 @@ class Group: self.candidates = defaultdict(set) self.ordered = [] self.unordered = set() - + def _get_matches_for_ref(self): if self._matches_for_ref is None: ref = self.ref self._matches_for_ref = [match for match in self.matches if ref in match] return self._matches_for_ref - + #---Public def add_match(self, match): """Adds ``match`` to internal match list and possibly add duplicates to the group. - + A duplicate can only be considered as such if it matches all other duplicates in the group. This method registers that pair (A, B) represented in ``match`` as possible candidates and, if A and/or B end up matching every other duplicates in the group, add these duplicates to the group. - + :param tuple match: pair of :class:`~core.fs.File` to add """ def add_candidate(item, match): @@ -348,7 +348,7 @@ class Group: if self.unordered <= matches: self.ordered.append(item) self.unordered.add(item) - + if match in self.matches: return self.matches.add(match) @@ -359,17 +359,17 @@ class Group: add_candidate(second, first) self._percentage = None self._matches_for_ref = None - + def discard_matches(self): """Remove all recorded matches that didn't result in a duplicate being added to the group. - + You can call this after the duplicate scanning process to free a bit of memory. """ discarded = set(m for m in self.matches if not all(obj in self.unordered for obj in [m.first, m.second])) self.matches -= discarded self.candidates = defaultdict(set) return discarded - + def get_match_of(self, item): """Returns the match pair between ``item`` and :attr:`ref`. """ @@ -378,10 +378,10 @@ class Group: for m in self._get_matches_for_ref(): if item in m: return m - + def prioritize(self, key_func, tie_breaker=None): """Reorders :attr:`ordered` according to ``key_func``. - + :param key_func: Key (f(x)) to be used for sorting :param tie_breaker: function to be used to select the reference position in case the top duplicates have the same key_func() result. @@ -405,7 +405,7 @@ class Group: self.switch_ref(ref) return True return changed - + def remove_dupe(self, item, discard_matches=True): try: self.ordered.remove(item) @@ -419,7 +419,7 @@ class Group: self._clear() except ValueError: pass - + def switch_ref(self, with_dupe): """Make the :attr:`ref` dupe of the group switch position with ``with_dupe``. """ @@ -433,9 +433,9 @@ class Group: return True except ValueError: return False - + dupes = property(lambda self: self[1:]) - + @property def percentage(self): if self._percentage is None: @@ -445,16 +445,16 @@ class Group: else: self._percentage = 0 return self._percentage - + @property def ref(self): if self: return self[0] - + def get_groups(matches, j=job.nulljob): """Returns a list of :class:`Group` from ``matches``. - + Create groups out of match pairs in the smartest way possible. """ matches.sort(key=lambda match: -match.percentage) diff --git a/core/results.py b/core/results.py index 3da81d23..41604c7d 100644 --- a/core/results.py +++ b/core/results.py @@ -1,9 +1,9 @@ # Created By: Virgil Dupras # Created On: 2006/02/23 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license import logging @@ -12,7 +12,7 @@ import os import os.path as op from xml.etree import ElementTree as ET -from jobprogress.job import nulljob +from hscommon.jobprogress.job import nulljob from hscommon.conflict import get_conflicted_name from hscommon.util import flatten, nonone, FileOrPath, format_size from hscommon.trans import tr @@ -22,15 +22,15 @@ from .markable import Markable class Results(Markable): """Manages a collection of duplicate :class:`~core.engine.Group`. - + This class takes care or marking, sorting and filtering duplicate groups. - + .. attribute:: groups - + The list of :class:`~core.engine.Group` contained managed by this instance. - + .. attribute:: dupes - + A list of all duplicates (:class:`~core.fs.File` instances), without ref, contained in the currently managed :attr:`groups`. """ @@ -50,16 +50,16 @@ class Results(Markable): self.app = app self.problems = [] # (dupe, error_msg) self.is_modified = False - + def _did_mark(self, dupe): self.__marked_size += dupe.size - + def _did_unmark(self, dupe): self.__marked_size -= dupe.size - + def _get_markable_count(self): return self.__total_count - + def _is_markable(self, dupe): if dupe.is_ref: return False @@ -71,25 +71,25 @@ class Results(Markable): if self.__filtered_dupes and dupe not in self.__filtered_dupes: return False return True - + def mark_all(self): if self.__filters: self.mark_multiple(self.__filtered_dupes) else: Markable.mark_all(self) - + def mark_invert(self): if self.__filters: self.mark_toggle_multiple(self.__filtered_dupes) else: Markable.mark_invert(self) - + def mark_none(self): if self.__filters: self.unmark_multiple(self.__filtered_dupes) else: Markable.mark_none(self) - + #---Private def __get_dupe_list(self): if self.__dupes is None: @@ -103,13 +103,13 @@ class Results(Markable): if sd: self.sort_dupes(sd[0], sd[1], sd[2]) return self.__dupes - + def __get_groups(self): if self.__filtered_groups is None: return self.__groups else: return self.__filtered_groups - + def __get_stat_line(self): if self.__filtered_dupes is None: mark_count = self.mark_count @@ -132,7 +132,7 @@ class Results(Markable): if self.__filters: result += tr(" filter: %s") % ' --> '.join(self.__filters) return result - + def __recalculate_stats(self): self.__total_size = 0 self.__total_count = 0 @@ -140,7 +140,7 @@ class Results(Markable): markable = [dupe for dupe in group.dupes if self._is_markable(dupe)] self.__total_count += len(markable) self.__total_size += sum(dupe.size for dupe in markable) - + def __set_groups(self, new_groups): self.mark_none() self.__groups = new_groups @@ -155,18 +155,18 @@ class Results(Markable): self.apply_filter(None) for filter_str in old_filters: self.apply_filter(filter_str) - + #---Public def apply_filter(self, filter_str): """Applies a filter ``filter_str`` to :attr:`groups` - + When you apply the filter, only dupes with the filename matching ``filter_str`` will be in - in the results. To cancel the filter, just call apply_filter with ``filter_str`` to None, + in the results. To cancel the filter, just call apply_filter with ``filter_str`` to None, and the results will go back to normal. - - If call apply_filter on a filtered results, the filter will be applied + + If call apply_filter on a filtered results, the filter will be applied *on the filtered results*. - + :param str filter_str: a string containing a regexp to filter dupes with. """ if not filter_str: @@ -193,7 +193,7 @@ class Results(Markable): if sd: self.sort_groups(sd[0], sd[1]) self.__dupes = None - + def get_group_of_duplicate(self, dupe): """Returns :class:`~core.engine.Group` in which ``dupe`` belongs. """ @@ -201,12 +201,12 @@ class Results(Markable): return self.__group_of_duplicate[dupe] except (TypeError, KeyError): return None - + is_markable = _is_markable - + def load_from_xml(self, infile, get_file, j=nulljob): """Load results from ``infile``. - + :param infile: a file or path pointing to an XML file created with :meth:`save_to_xml`. :param get_file: a function f(path) returning a :class:`~core.fs.File` wrapping the path. :param j: A :ref:`job progress instance `. @@ -217,7 +217,7 @@ class Results(Markable): for other_file in other_files: group.add_match(engine.get_match(ref_file, other_file)) do_match(other_files[0], other_files[1:], group) - + self.apply_filter(None) try: root = ET.parse(infile).getroot() @@ -255,13 +255,13 @@ class Results(Markable): do_match(dupes[0], dupes[1:], group) group.prioritize(lambda x: dupes.index(x)) if len(group): - groups.append(group) + groups.append(group) j.add_progress() self.groups = groups for dupe_file in marked: self.mark(dupe_file) self.is_modified = False - + def make_ref(self, dupe): """Make ``dupe`` take the :attr:`~core.engine.Group.ref` position of its group. """ @@ -279,13 +279,13 @@ class Results(Markable): self.__dupes = None self.is_modified = True return True - + def perform_on_marked(self, func, remove_from_results): """Performs ``func`` on all marked dupes. - + If an ``EnvironmentError`` is raised during the call, the problematic dupe is added to self.problems. - + :param bool remove_from_results: If true, dupes which had ``func`` applied and didn't cause any problem. """ @@ -303,10 +303,10 @@ class Results(Markable): self.mark_none() for dupe, _ in self.problems: self.mark(dupe) - + def remove_duplicates(self, dupes): """Remove ``dupes`` from their respective :class:`~core.engine.Group`. - + Also, remove the group from :attr:`groups` if it ends up empty. """ affected_groups = set() @@ -331,10 +331,10 @@ class Results(Markable): group.discard_matches() self.__dupes = None self.is_modified = bool(self.__groups) - + def save_to_xml(self, outfile): """Save results to ``outfile`` in XML. - + :param outfile: file object or path. """ self.apply_filter(None) @@ -362,11 +362,11 @@ class Results(Markable): match_elem.set('second', str(dupe2index[match.second])) match_elem.set('percentage', str(int(match.percentage))) tree = ET.ElementTree(root) - + def do_write(outfile): with FileOrPath(outfile, 'wb') as fp: tree.write(fp, encoding='utf-8') - + try: do_write(outfile) except IOError as e: @@ -381,10 +381,10 @@ class Results(Markable): else: raise self.is_modified = False - + def sort_dupes(self, key, asc=True, delta=False): """Sort :attr:`dupes` according to ``key``. - + :param str key: key attribute name to sort with. :param bool asc: If false, sorting is reversed. :param bool delta: If true, sorting occurs using :ref:`delta values `. @@ -394,19 +394,19 @@ class Results(Markable): keyfunc = lambda d: self.app._get_dupe_sort_key(d, lambda: self.get_group_of_duplicate(d), key, delta) self.__dupes.sort(key=keyfunc, reverse=not asc) self.__dupes_sort_descriptor = (key,asc,delta) - + def sort_groups(self, key, asc=True): """Sort :attr:`groups` according to ``key``. - + The :attr:`~core.engine.Group.ref` of each group is used to extract values for sorting. - + :param str key: key attribute name to sort with. :param bool asc: If false, sorting is reversed. """ keyfunc = lambda g: self.app._get_group_sort_key(g, key) self.groups.sort(key=keyfunc, reverse=not asc) self.__groups_sort_descriptor = (key,asc) - + #---Properties dupes = property(__get_dupe_list) groups = property(__get_groups, __set_groups) diff --git a/core/scanner.py b/core/scanner.py index 69db00a7..6f86f5d1 100644 --- a/core/scanner.py +++ b/core/scanner.py @@ -1,16 +1,16 @@ # Created By: Virgil Dupras # Created On: 2006/03/03 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license import logging import re import os.path as op -from jobprogress import job +from hscommon.jobprogress import job from hscommon.util import dedupe, rem_file_ext, get_file_ext from hscommon.trans import tr @@ -29,7 +29,7 @@ class ScanType: Folders = 4 Contents = 5 ContentsAudio = 6 - + #PE FuzzyBlock = 10 ExifTimestamp = 11 @@ -72,7 +72,7 @@ class Scanner: def __init__(self): self.ignore_list = IgnoreList() self.discarded_file_count = 0 - + def _getmatches(self, files, j): if self.size_threshold: j = j.start_subjob([2, 8]) @@ -100,11 +100,11 @@ class Scanner: logging.debug("Reading metadata of {}".format(str(f.path))) f.words = func(f) return engine.getmatches(files, j=j, **kw) - + @staticmethod def _key_func(dupe): return -dupe.size - + @staticmethod def _tie_breaker(ref, dupe): refname = rem_file_ext(ref.name).lower() @@ -118,7 +118,7 @@ class Scanner: if is_same_with_digit(refname, dupename): return True return len(dupe.path) > len(ref.path) - + def get_dupe_groups(self, files, j=job.nulljob): j = j.start_subjob([8, 2]) for f in (f for f in files if not hasattr(f, 'is_ref')): @@ -152,7 +152,7 @@ class Scanner: if self.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 + matches = [m for m in iter_matches if not self.ignore_list.AreIgnored(str(m.first.path), str(m.second.path))] logging.info('Grouping matches') groups = engine.get_groups(matches, j) @@ -177,7 +177,7 @@ class Scanner: for g in groups: g.prioritize(self._key_func, self._tie_breaker) return groups - + match_similar_words = False min_match_percentage = 80 mix_file_kind = True diff --git a/core/tests/app_test.py b/core/tests/app_test.py index 35206090..6170be04 100644 --- a/core/tests/app_test.py +++ b/core/tests/app_test.py @@ -1,9 +1,9 @@ # Created By: Virgil Dupras # Created On: 2007-06-23 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license import os @@ -15,7 +15,7 @@ from hscommon.path import Path import hscommon.conflict import hscommon.util from hscommon.testutil import CallLogger, eq_, log_calls -from jobprogress.job import Job +from hscommon.jobprogress.job import Job from .base import DupeGuru, TestApp from .results_test import GetTestGroups @@ -36,7 +36,7 @@ class TestCaseDupeGuru: assert call['filter_str'] is None call = dgapp.results.apply_filter.calls[1] eq_('foo', call['filter_str']) - + def test_apply_filter_escapes_regexp(self, monkeypatch): dgapp = TestApp().app monkeypatch.setattr(dgapp.results, 'apply_filter', log_calls(dgapp.results.apply_filter)) @@ -50,7 +50,7 @@ class TestCaseDupeGuru: dgapp.apply_filter('(abc)') call = dgapp.results.apply_filter.calls[5] eq_('(abc)', call['filter_str']) - + def test_copy_or_move(self, tmpdir, monkeypatch): # The goal here is just to have a test for a previous blowup I had. I know my test coverage # for this unit is pathetic. What's done is done. My approach now is to add tests for @@ -69,7 +69,7 @@ class TestCaseDupeGuru: call = hscommon.conflict.smart_copy.calls[0] eq_(call['dest_path'], op.join('some_destination', 'foo')) eq_(call['source_path'], f.path) - + def test_copy_or_move_clean_empty_dirs(self, tmpdir, monkeypatch): tmppath = Path(str(tmpdir)) sourcepath = tmppath['source'] @@ -83,13 +83,13 @@ class TestCaseDupeGuru: calls = app.clean_empty_dirs.calls eq_(1, len(calls)) eq_(sourcepath, calls[0]['path']) - + def test_Scan_with_objects_evaluating_to_false(self): class FakeFile(fs.File): def __bool__(self): return False - - + + # At some point, any() was used in a wrong way that made Scan() wrongly return 1 app = TestApp().app f1, f2 = [FakeFile('foo') for i in range(2)] @@ -97,7 +97,7 @@ class TestCaseDupeGuru: assert not (bool(f1) and bool(f2)) add_fake_files_to_directories(app.directories, [f1, f2]) app.start_scanning() # no exception - + @mark.skipif("not hasattr(os, 'link')") def test_ignore_hardlink_matches(self, tmpdir): # If the ignore_hardlink_matches option is set, don't match files hardlinking to the same @@ -111,7 +111,7 @@ class TestCaseDupeGuru: app.options['ignore_hardlink_matches'] = True app.start_scanning() eq_(len(app.results.groups), 0) - + def test_rename_when_nothing_is_selected(self): # Issue #140 # It's possible that rename operation has its selected row swept off from under it, thus @@ -127,11 +127,11 @@ class TestCaseDupeGuru_clean_empty_dirs: # XXX This monkeypatch is temporary. will be fixed in a better monkeypatcher. monkeypatch.setattr(app, 'delete_if_empty', hscommon.util.delete_if_empty) self.app = TestApp().app - + def test_option_off(self, do_setup): self.app.clean_empty_dirs(Path('/foo/bar')) eq_(0, len(hscommon.util.delete_if_empty.calls)) - + def test_option_on(self, do_setup): self.app.options['clean_empty_dirs'] = True self.app.clean_empty_dirs(Path('/foo/bar')) @@ -139,13 +139,13 @@ class TestCaseDupeGuru_clean_empty_dirs: eq_(1, len(calls)) eq_(Path('/foo/bar'), calls[0]['path']) eq_(['.DS_Store'], calls[0]['files_to_delete']) - + def test_recurse_up(self, do_setup, monkeypatch): # delete_if_empty must be recursively called up in the path until it returns False @log_calls def mock_delete_if_empty(path, files_to_delete=[]): return len(path) > 1 - + monkeypatch.setattr(hscommon.util, 'delete_if_empty', mock_delete_if_empty) # XXX This monkeypatch is temporary. will be fixed in a better monkeypatcher. monkeypatch.setattr(app, 'delete_if_empty', mock_delete_if_empty) @@ -156,7 +156,7 @@ class TestCaseDupeGuru_clean_empty_dirs: eq_(Path('not-empty/empty/empty'), calls[0]['path']) eq_(Path('not-empty/empty'), calls[1]['path']) eq_(Path('not-empty'), calls[2]['path']) - + class TestCaseDupeGuruWithResults: def pytest_funcarg__do_setup(self, request): @@ -173,7 +173,7 @@ class TestCaseDupeGuruWithResults: tmppath['foo'].mkdir() tmppath['bar'].mkdir() self.app.directories.add_path(tmppath) - + def test_GetObjects(self, do_setup): objects = self.objects groups = self.groups @@ -186,7 +186,7 @@ class TestCaseDupeGuruWithResults: r = self.rtable[4] assert r._group is groups[1] assert r._dupe is objects[4] - + def test_GetObjects_after_sort(self, do_setup): objects = self.objects groups = self.groups[:] # we need an un-sorted reference @@ -194,14 +194,14 @@ class TestCaseDupeGuruWithResults: r = self.rtable[1] assert r._group is groups[1] assert r._dupe is objects[4] - + def test_selected_result_node_paths_after_deletion(self, do_setup): # cases where the selected dupes aren't there are correctly handled self.rtable.select([1, 2, 3]) self.app.remove_selected() # The first 2 dupes have been removed. The 3rd one is a ref. it stays there, in first pos. eq_(self.rtable.selected_indexes, [1]) # no exception - + def test_selectResultNodePaths(self, do_setup): app = self.app objects = self.objects @@ -209,7 +209,7 @@ class TestCaseDupeGuruWithResults: eq_(len(app.selected_dupes), 2) assert app.selected_dupes[0] is objects[1] assert app.selected_dupes[1] is objects[2] - + def test_selectResultNodePaths_with_ref(self, do_setup): app = self.app objects = self.objects @@ -218,26 +218,26 @@ class TestCaseDupeGuruWithResults: assert app.selected_dupes[0] is objects[1] assert app.selected_dupes[1] is objects[2] assert app.selected_dupes[2] is self.groups[1].ref - + def test_selectResultNodePaths_after_sort(self, do_setup): app = self.app objects = self.objects groups = self.groups[:] #To keep the old order in memory - self.rtable.sort('name', False) #0 + self.rtable.sort('name', False) #0 #Now, the group order is supposed to be reversed self.rtable.select([1, 2, 3]) eq_(len(app.selected_dupes), 3) assert app.selected_dupes[0] is objects[4] assert app.selected_dupes[1] is groups[0].ref assert app.selected_dupes[2] is objects[1] - + def test_selected_powermarker_node_paths(self, do_setup): # app.selected_dupes is correctly converted into paths self.rtable.power_marker = True self.rtable.select([0, 1, 2]) self.rtable.power_marker = False eq_(self.rtable.selected_indexes, [1, 2, 4]) - + def test_selected_powermarker_node_paths_after_deletion(self, do_setup): # cases where the selected dupes aren't there are correctly handled app = self.app @@ -245,7 +245,7 @@ class TestCaseDupeGuruWithResults: self.rtable.select([0, 1, 2]) app.remove_selected() eq_(self.rtable.selected_indexes, []) # no exception - + def test_selectPowerMarkerRows_after_sort(self, do_setup): app = self.app objects = self.objects @@ -256,7 +256,7 @@ class TestCaseDupeGuruWithResults: assert app.selected_dupes[0] is objects[4] assert app.selected_dupes[1] is objects[2] assert app.selected_dupes[2] is objects[1] - + def test_toggle_selected_mark_state(self, do_setup): app = self.app objects = self.objects @@ -270,7 +270,7 @@ class TestCaseDupeGuruWithResults: assert not app.results.is_marked(objects[2]) assert not app.results.is_marked(objects[3]) assert app.results.is_marked(objects[4]) - + def test_toggle_selected_mark_state_with_different_selected_state(self, do_setup): # When marking selected dupes with a heterogenous selection, mark all selected dupes. When # it's homogenous, simply toggle. @@ -285,7 +285,7 @@ class TestCaseDupeGuruWithResults: eq_(app.results.mark_count, 2) app.toggle_selected_mark_state() eq_(app.results.mark_count, 0) - + def test_refreshDetailsWithSelected(self, do_setup): self.rtable.select([1, 4]) eq_(self.dpanel.row(0), ('Filename', 'bar bleh', 'foo bar')) @@ -293,7 +293,7 @@ class TestCaseDupeGuruWithResults: self.rtable.select([]) eq_(self.dpanel.row(0), ('Filename', '---', '---')) self.dpanel.view.check_gui_calls(['refresh']) - + def test_makeSelectedReference(self, do_setup): app = self.app objects = self.objects @@ -302,7 +302,7 @@ class TestCaseDupeGuruWithResults: app.make_selected_reference() assert groups[0].ref is objects[1] assert groups[1].ref is objects[4] - + def test_makeSelectedReference_by_selecting_two_dupes_in_the_same_group(self, do_setup): app = self.app objects = self.objects @@ -312,7 +312,7 @@ class TestCaseDupeGuruWithResults: app.make_selected_reference() assert groups[0].ref is objects[1] assert groups[1].ref is objects[4] - + def test_removeSelected(self, do_setup): app = self.app self.rtable.select([1, 4]) @@ -320,7 +320,7 @@ class TestCaseDupeGuruWithResults: eq_(len(app.results.dupes), 1) # the first path is now selected app.remove_selected() eq_(len(app.results.dupes), 0) - + def test_addDirectory_simple(self, do_setup): # There's already a directory in self.app, so adding another once makes 2 of em app = self.app @@ -328,7 +328,7 @@ class TestCaseDupeGuruWithResults: otherpath = Path(op.dirname(__file__)) app.add_directory(otherpath) eq_(len(app.directories), 2) - + def test_addDirectory_already_there(self, do_setup): app = self.app otherpath = Path(op.dirname(__file__)) @@ -336,13 +336,13 @@ class TestCaseDupeGuruWithResults: app.add_directory(otherpath) eq_(len(app.view.messages), 1) assert "already" in app.view.messages[0] - + def test_addDirectory_does_not_exist(self, do_setup): app = self.app app.add_directory('/does_not_exist') eq_(len(app.view.messages), 1) assert "exist" in app.view.messages[0] - + def test_ignore(self, do_setup): app = self.app self.rtable.select([4]) #The dupe of the second, 2 sized group @@ -352,7 +352,7 @@ class TestCaseDupeGuruWithResults: app.add_selected_to_ignore_list() #BOTH the ref and the other dupe should have been added eq_(len(app.scanner.ignore_list), 3) - + def test_purgeIgnoreList(self, do_setup, tmpdir): app = self.app p1 = str(tmpdir.join('file1')) @@ -367,19 +367,19 @@ class TestCaseDupeGuruWithResults: eq_(1,len(app.scanner.ignore_list)) assert app.scanner.ignore_list.AreIgnored(p1,p2) assert not app.scanner.ignore_list.AreIgnored(dne,p1) - + def test_only_unicode_is_added_to_ignore_list(self, do_setup): def FakeIgnore(first,second): if not isinstance(first,str): self.fail() if not isinstance(second,str): self.fail() - + app = self.app app.scanner.ignore_list.Ignore = FakeIgnore self.rtable.select([4]) app.add_selected_to_ignore_list() - + def test_cancel_scan_with_previous_results(self, do_setup): # When doing a scan with results being present prior to the scan, correctly invalidate the # results table. @@ -388,7 +388,7 @@ class TestCaseDupeGuruWithResults: add_fake_files_to_directories(app.directories, self.objects) # We want the scan to at least start app.start_scanning() # will be cancelled immediately eq_(len(self.rtable), 0) - + def test_selected_dupes_after_removal(self, do_setup): # Purge the app's `selected_dupes` attribute when removing dupes, or else it might cause a # crash later with None refs. @@ -398,7 +398,7 @@ class TestCaseDupeGuruWithResults: app.remove_marked() eq_(len(self.rtable), 0) eq_(app.selected_dupes, []) - + def test_dont_crash_on_delta_powermarker_dupecount_sort(self, do_setup): # Don't crash when sorting by dupe count or percentage while delta+powermarker are enabled. # Ref #238 @@ -410,7 +410,7 @@ class TestCaseDupeGuruWithResults: # don't crash self.rtable.sort('percentage', False) # don't crash - + class TestCaseDupeGuru_renameSelected: def pytest_funcarg__do_setup(self, request): @@ -437,7 +437,7 @@ class TestCaseDupeGuru_renameSelected: self.groups = groups self.p = p self.files = files - + def test_simple(self, do_setup): app = self.app g = self.groups[0] @@ -447,7 +447,7 @@ class TestCaseDupeGuru_renameSelected: assert 'renamed' in names assert 'foo bar 2' not in names eq_(g.dupes[0].name, 'renamed') - + def test_none_selected(self, do_setup, monkeypatch): app = self.app g = self.groups[0] @@ -460,7 +460,7 @@ class TestCaseDupeGuru_renameSelected: assert 'renamed' not in names assert 'foo bar 2' in names eq_(g.dupes[0].name, 'foo bar 2') - + def test_name_already_exists(self, do_setup, monkeypatch): app = self.app g = self.groups[0] @@ -473,7 +473,7 @@ class TestCaseDupeGuru_renameSelected: assert 'foo bar 1' in names assert 'foo bar 2' in names eq_(g.dupes[0].name, 'foo bar 2') - + class TestAppWithDirectoriesInTree: def pytest_funcarg__do_setup(self, request): @@ -487,7 +487,7 @@ class TestAppWithDirectoriesInTree: self.dtree = app.dtree self.dtree.add_directory(p) self.dtree.view.clear_calls() - + def test_set_root_as_ref_makes_subfolders_ref_as_well(self, do_setup): # Setting a node state to something also affect subnodes. These subnodes must be correctly # refreshed. @@ -500,4 +500,4 @@ class TestAppWithDirectoriesInTree: subnode = node[0] eq_(subnode.state, 1) self.dtree.view.check_gui_calls(['refresh_states']) - + diff --git a/core/tests/base.py b/core/tests/base.py index 55db4f04..d7b5aa65 100644 --- a/core/tests/base.py +++ b/core/tests/base.py @@ -1,16 +1,16 @@ # Created By: Virgil Dupras # Created On: 2011/09/07 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license from hscommon.testutil import TestApp as TestAppBase, eq_, with_app from hscommon.path import Path from hscommon.util import get_file_ext, format_size from hscommon.gui.column import Column -from jobprogress.job import nulljob, JobCancelled +from hscommon.jobprogress.job import nulljob, JobCancelled from .. import engine from .. import prioritize @@ -23,28 +23,28 @@ from ..gui.prioritize_dialog import PrioritizeDialog class DupeGuruView: JOB = nulljob - + def __init__(self): self.messages = [] - + def start_job(self, jobid, func, args=()): try: func(self.JOB, *args) except JobCancelled: return - + def get_default(self, key_name): return None - + def set_default(self, key_name, value): pass - + def show_message(self, msg): self.messages.append(msg) - + def ask_yes_no(self, prompt): return True # always answer yes - + class ResultTable(ResultTableBase): COLUMNS = [ @@ -55,21 +55,21 @@ class ResultTable(ResultTableBase): Column('extension', 'Kind'), ] DELTA_COLUMNS = {'size', } - + class DupeGuru(DupeGuruBase): NAME = 'dupeGuru' METADATA_TO_READ = ['size'] - + def __init__(self): DupeGuruBase.__init__(self, DupeGuruView()) self.appdata = '/tmp' - + def _prioritization_categories(self): return prioritize.all_categories() - + def _create_result_table(self): return ResultTable(self) - + class NamedObject: def __init__(self, name="foobar", with_words=False, size=1, folder=None): @@ -83,10 +83,10 @@ class NamedObject: if with_words: self.words = getwords(name) self.is_ref = False - + def __bool__(self): return False #Make sure that operations are made correctly when the bool value of files is false. - + def get_display_info(self, group, delta): size = self.size m = group.get_match_of(self) @@ -99,19 +99,19 @@ class NamedObject: 'size': format_size(size, 0, 1, False), 'extension': self.extension if hasattr(self, 'extension') else '---', } - + @property def path(self): return self._folder[self.name] - + @property def folder_path(self): return self.path.parent() - + @property def extension(self): return get_file_ext(self.name) - + # Returns a group set that looks like that: # "foo bar" (1) # "bar bleh" (1024) @@ -135,7 +135,7 @@ class TestApp(TestAppBase): if hasattr(gui, 'columns'): # tables gui.columns.view = self.make_logger() return gui - + TestAppBase.__init__(self) make_gui = self.make_gui self.app = DupeGuru() @@ -153,14 +153,14 @@ class TestApp(TestAppBase): link_gui(self.app.progress_window) link_gui(self.app.progress_window.jobdesc_textfield) link_gui(self.app.progress_window.progressdesc_textfield) - + #--- Helpers def select_pri_criterion(self, name): # Select a main prioritize criterion by name instead of by index. Makes tests more # maintainable. index = self.pdialog.category_list.index(name) self.pdialog.category_list.select(index) - + def add_pri_criterion(self, name, index): self.select_pri_criterion(name) self.pdialog.criteria_list.select([index]) diff --git a/core/tests/engine_test.py b/core/tests/engine_test.py index 592f0d76..3f41c2a7 100644 --- a/core/tests/engine_test.py +++ b/core/tests/engine_test.py @@ -1,14 +1,14 @@ # Created By: Virgil Dupras # Created On: 2006/01/29 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license import sys -from jobprogress import job +from hscommon.jobprogress import job from hscommon.util import first from hscommon.testutil import eq_, log_calls @@ -48,119 +48,119 @@ class TestCasegetwords: def test_spaces(self): eq_(['a', 'b', 'c', 'd'], getwords("a b c d")) eq_(['a', 'b', 'c', 'd'], getwords(" a b c d ")) - + def test_splitter_chars(self): eq_( [chr(i) for i in range(ord('a'),ord('z')+1)], getwords("a-b_c&d+e(f)g;h\\i[j]k{l}m:n.o,pr/s?t~u!v@w#x$y*z") ) - + def test_joiner_chars(self): eq_(["aec"], getwords("a'e\u0301c")) - + def test_empty(self): eq_([], getwords('')) - + def test_returns_lowercase(self): eq_(['foo', 'bar'], getwords('FOO BAR')) - + def test_decompose_unicode(self): eq_(getwords('foo\xe9bar'), ['fooebar']) - + class TestCasegetfields: def test_simple(self): eq_([['a', 'b'], ['c', 'd', 'e']], getfields('a b - c d e')) - + def test_empty(self): eq_([], getfields('')) - + def test_cleans_empty_fields(self): expected = [['a', 'bc', 'def']] actual = getfields(' - a bc def') eq_(expected, actual) expected = [['bc', 'def']] - + class TestCaseunpack_fields: def test_with_fields(self): expected = ['a', 'b', 'c', 'd', 'e', 'f'] actual = unpack_fields([['a'], ['b', 'c'], ['d', 'e', 'f']]) eq_(expected, actual) - + def test_without_fields(self): expected = ['a', 'b', 'c', 'd', 'e', 'f'] actual = unpack_fields(['a', 'b', 'c', 'd', 'e', 'f']) eq_(expected, actual) - + def test_empty(self): eq_([], unpack_fields([])) - + class TestCaseWordCompare: def test_list(self): eq_(100, compare(['a', 'b', 'c', 'd'],['a', 'b', 'c', 'd'])) eq_(86, compare(['a', 'b', 'c', 'd'],['a', 'b', 'c'])) - + def test_unordered(self): #Sometimes, users don't want fuzzy matching too much When they set the slider #to 100, they don't expect a filename with the same words, but not the same order, to match. #Thus, we want to return 99 in that case. eq_(99, compare(['a', 'b', 'c', 'd'], ['d', 'b', 'c', 'a'])) - + def test_word_occurs_twice(self): #if a word occurs twice in first, but once in second, we want the word to be only counted once eq_(89, compare(['a', 'b', 'c', 'd', 'a'], ['d', 'b', 'c', 'a'])) - + def test_uses_copy_of_lists(self): first = ['foo', 'bar'] second = ['bar', 'bleh'] compare(first, second) eq_(['foo', 'bar'], first) eq_(['bar', 'bleh'], second) - + def test_word_weight(self): eq_(int((6.0 / 13.0) * 100), compare(['foo', 'bar'], ['bar', 'bleh'], (WEIGHT_WORDS, ))) - + def test_similar_words(self): eq_(100, compare(['the', 'white', 'stripes'],['the', 'whites', 'stripe'], (MATCH_SIMILAR_WORDS, ))) - + def test_empty(self): eq_(0, compare([], [])) - + def test_with_fields(self): eq_(67, compare([['a', 'b'], ['c', 'd', 'e']], [['a', 'b'], ['c', 'd', 'f']])) - + def test_propagate_flags_with_fields(self, monkeypatch): def mock_compare(first, second, flags): eq_((0, 1, 2, 3, 5), flags) - + monkeypatch.setattr(engine, 'compare_fields', mock_compare) compare([['a']], [['a']], (0, 1, 2, 3, 5)) - + class TestCaseWordCompareWithFields: def test_simple(self): eq_(67, compare_fields([['a', 'b'], ['c', 'd', 'e']], [['a', 'b'], ['c', 'd', 'f']])) - + def test_empty(self): eq_(0, compare_fields([], [])) - + def test_different_length(self): eq_(0, compare_fields([['a'], ['b']], [['a'], ['b'], ['c']])) - + def test_propagates_flags(self, monkeypatch): def mock_compare(first, second, flags): eq_((0, 1, 2, 3, 5), flags) - + monkeypatch.setattr(engine, 'compare_fields', mock_compare) compare_fields([['a']], [['a']],(0, 1, 2, 3, 5)) - + def test_order(self): first = [['a', 'b'], ['c', 'd', 'e']] second = [['c', 'd', 'f'], ['a', 'b']] eq_(0, compare_fields(first, second)) - + def test_no_order(self): first = [['a','b'],['c','d','e']] second = [['c','d','f'],['a','b']] @@ -168,10 +168,10 @@ class TestCaseWordCompareWithFields: first = [['a','b'],['a','b']] #a field can only be matched once. second = [['c','d','f'],['a','b']] eq_(0, compare_fields(first, second, (NO_FIELD_ORDER, ))) - first = [['a','b'],['a','b','c']] + first = [['a','b'],['a','b','c']] second = [['c','d','f'],['a','b']] eq_(33, compare_fields(first, second, (NO_FIELD_ORDER, ))) - + def test_compare_fields_without_order_doesnt_alter_fields(self): #The NO_ORDER comp type altered the fields! first = [['a','b'],['c','d','e']] @@ -179,7 +179,7 @@ class TestCaseWordCompareWithFields: eq_(67, compare_fields(first, second, (NO_FIELD_ORDER, ))) eq_([['a','b'],['c','d','e']],first) eq_([['c','d','f'],['a','b']],second) - + class TestCasebuild_word_dict: def test_with_standard_words(self): @@ -199,30 +199,30 @@ class TestCasebuild_word_dict: assert l[2] in d['baz'] eq_(1,len(d['bleh'])) assert l[2] in d['bleh'] - + def test_unpack_fields(self): o = NamedObject('') o.words = [['foo','bar'],['baz']] d = build_word_dict([o]) eq_(3,len(d)) eq_(1,len(d['foo'])) - + def test_words_are_unaltered(self): o = NamedObject('') o.words = [['foo','bar'],['baz']] build_word_dict([o]) eq_([['foo','bar'],['baz']],o.words) - + def test_object_instances_can_only_be_once_in_words_object_list(self): o = NamedObject('foo foo',True) d = build_word_dict([o]) eq_(1,len(d['foo'])) - + def test_job(self): def do_progress(p,d=''): self.log.append(p) return True - + j = job.Job(1,do_progress) self.log = [] s = "foo bar" @@ -230,7 +230,7 @@ class TestCasebuild_word_dict: # We don't have intermediate log because iter_with_progress is called with every > 1 eq_(0,self.log[0]) eq_(100,self.log[1]) - + class TestCasemerge_similar_words: def test_some_similar_words(self): @@ -242,8 +242,8 @@ class TestCasemerge_similar_words: merge_similar_words(d) eq_(1,len(d)) eq_(3,len(d['foobar'])) - - + + class TestCasereduce_common_words: def test_typical(self): @@ -254,7 +254,7 @@ class TestCasereduce_common_words: reduce_common_words(d, 50) assert 'foo' not in d eq_(49,len(d['bar'])) - + def test_dont_remove_objects_with_only_common_words(self): d = { 'common': set([NamedObject("common uncommon",True) for i in range(50)] + [NamedObject("common",True)]), @@ -263,7 +263,7 @@ class TestCasereduce_common_words: reduce_common_words(d, 50) eq_(1,len(d['common'])) eq_(1,len(d['uncommon'])) - + def test_values_still_are_set_instances(self): d = { 'common': set([NamedObject("common uncommon",True) for i in range(50)] + [NamedObject("common",True)]), @@ -272,7 +272,7 @@ class TestCasereduce_common_words: reduce_common_words(d, 50) assert isinstance(d['common'],set) assert isinstance(d['uncommon'],set) - + def test_dont_raise_KeyError_when_a_word_has_been_removed(self): #If a word has been removed by the reduce, an object in a subsequent common word that #contains the word that has been removed would cause a KeyError. @@ -285,14 +285,14 @@ class TestCasereduce_common_words: reduce_common_words(d, 50) except KeyError: self.fail() - + def test_unpack_fields(self): #object.words may be fields. def create_it(): o = NamedObject('') o.words = [['foo','bar'],['baz']] return o - + d = { 'foo': set([create_it() for i in range(50)]) } @@ -300,7 +300,7 @@ class TestCasereduce_common_words: reduce_common_words(d, 50) except TypeError: self.fail("must support fields.") - + def test_consider_a_reduced_common_word_common_even_after_reduction(self): #There was a bug in the code that causeda word that has already been reduced not to #be counted as a common word for subsequent words. For example, if 'foo' is processed @@ -316,7 +316,7 @@ class TestCasereduce_common_words: eq_(1,len(d['foo'])) eq_(1,len(d['bar'])) eq_(49,len(d['baz'])) - + class TestCaseget_match: def test_simple(self): @@ -328,7 +328,7 @@ class TestCaseget_match: eq_(['bar','bleh'],m.second.words) assert m.first is o1 assert m.second is o2 - + def test_in(self): o1 = NamedObject("foo",True) o2 = NamedObject("bar",True) @@ -336,15 +336,15 @@ class TestCaseget_match: assert o1 in m assert o2 in m assert object() not in m - + def test_word_weight(self): eq_(int((6.0 / 13.0) * 100),get_match(NamedObject("foo bar",True),NamedObject("bar bleh",True),(WEIGHT_WORDS,)).percentage) - + class TestCaseGetMatches: def test_empty(self): eq_(getmatches([]), []) - + def test_simple(self): l = [NamedObject("foo bar"),NamedObject("bar bleh"),NamedObject("a b c foo")] r = getmatches(l) @@ -353,7 +353,7 @@ class TestCaseGetMatches: assert_match(m, 'foo bar', 'bar bleh') m = first(m for m in r if m.percentage == 33) #"foo bar" and "a b c foo" assert_match(m, 'foo bar', 'a b c foo') - + def test_null_and_unrelated_objects(self): l = [NamedObject("foo bar"),NamedObject("bar bleh"),NamedObject(""),NamedObject("unrelated object")] r = getmatches(l) @@ -361,22 +361,22 @@ class TestCaseGetMatches: m = r[0] eq_(m.percentage, 50) assert_match(m, 'foo bar', 'bar bleh') - + def test_twice_the_same_word(self): l = [NamedObject("foo foo bar"),NamedObject("bar bleh")] r = getmatches(l) eq_(1,len(r)) - + def test_twice_the_same_word_when_preworded(self): l = [NamedObject("foo foo bar",True),NamedObject("bar bleh",True)] r = getmatches(l) eq_(1,len(r)) - + def test_two_words_match(self): l = [NamedObject("foo bar"),NamedObject("foo bar bleh")] r = getmatches(l) eq_(1,len(r)) - + def test_match_files_with_only_common_words(self): #If a word occurs more than 50 times, it is excluded from the matching process #The problem with the common_word_threshold is that the files containing only common @@ -385,18 +385,18 @@ class TestCaseGetMatches: l = [NamedObject("foo") for i in range(50)] r = getmatches(l) eq_(1225,len(r)) - + def test_use_words_already_there_if_there(self): o1 = NamedObject('foo') o2 = NamedObject('bar') o2.words = ['foo'] eq_(1, len(getmatches([o1,o2]))) - + def test_job(self): def do_progress(p,d=''): self.log.append(p) return True - + j = job.Job(1,do_progress) self.log = [] s = "foo bar" @@ -404,12 +404,12 @@ class TestCaseGetMatches: assert len(self.log) > 2 eq_(0,self.log[0]) eq_(100,self.log[-1]) - + def test_weight_words(self): l = [NamedObject("foo bar"),NamedObject("bar bleh")] m = getmatches(l, weight_words=True)[0] eq_(int((6.0 / 13.0) * 100),m.percentage) - + def test_similar_word(self): l = [NamedObject("foobar"),NamedObject("foobars")] eq_(len(getmatches(l, match_similar_words=True)), 1) @@ -420,16 +420,16 @@ class TestCaseGetMatches: eq_(len(getmatches(l, match_similar_words=True)), 1) l = [NamedObject("foobar"),NamedObject("foosbar")] eq_(len(getmatches(l, match_similar_words=True)), 1) - + def test_single_object_with_similar_words(self): l = [NamedObject("foo foos")] eq_(len(getmatches(l, match_similar_words=True)), 0) - + def test_double_words_get_counted_only_once(self): l = [NamedObject("foo bar foo bleh"),NamedObject("foo bar bleh bar")] m = getmatches(l)[0] eq_(75,m.percentage) - + def test_with_fields(self): o1 = NamedObject("foo bar - foo bleh") o2 = NamedObject("foo bar - bleh bar") @@ -437,7 +437,7 @@ class TestCaseGetMatches: o2.words = getfields(o2.name) m = getmatches([o1, o2])[0] eq_(50, m.percentage) - + def test_with_fields_no_order(self): o1 = NamedObject("foo bar - foo bleh") o2 = NamedObject("bleh bang - foo bar") @@ -445,11 +445,11 @@ class TestCaseGetMatches: o2.words = getfields(o2.name) m = getmatches([o1, o2], no_field_order=True)[0] eq_(m.percentage, 50) - + def test_only_match_similar_when_the_option_is_set(self): l = [NamedObject("foobar"),NamedObject("foobars")] eq_(len(getmatches(l, match_similar_words=False)), 0) - + def test_dont_recurse_do_match(self): # with nosetests, the stack is increased. The number has to be high enough not to be failing falsely sys.setrecursionlimit(100) @@ -460,19 +460,19 @@ class TestCaseGetMatches: self.fail() finally: sys.setrecursionlimit(1000) - + def test_min_match_percentage(self): l = [NamedObject("foo bar"),NamedObject("bar bleh"),NamedObject("a b c foo")] r = getmatches(l, min_match_percentage=50) eq_(1,len(r)) #Only "foo bar" / "bar bleh" should match - + def test_MemoryError(self, monkeypatch): @log_calls def mocked_match(first, second, flags): if len(mocked_match.calls) > 42: raise MemoryError() return Match(first, second, 0) - + objects = [NamedObject() for i in range(10)] # results in 45 matches monkeypatch.setattr(engine, 'get_match', mocked_match) try: @@ -480,13 +480,13 @@ class TestCaseGetMatches: except MemoryError: self.fail('MemorryError must be handled') eq_(42, len(r)) - + class TestCaseGetMatchesByContents: def test_dont_compare_empty_files(self): o1, o2 = no(size=0), no(size=0) assert not getmatches_by_contents([o1, o2]) - + class TestCaseGroup: def test_empy(self): @@ -494,7 +494,7 @@ class TestCaseGroup: eq_(None,g.ref) eq_([],g.dupes) eq_(0,len(g.matches)) - + def test_add_match(self): g = Group() m = get_match(NamedObject("foo",True),NamedObject("bar",True)) @@ -503,7 +503,7 @@ class TestCaseGroup: eq_([m.second],g.dupes) eq_(1,len(g.matches)) assert m in g.matches - + def test_multiple_add_match(self): g = Group() o1 = NamedObject("a",True) @@ -529,13 +529,13 @@ class TestCaseGroup: g.add_match(get_match(o3,o4)) eq_([o2,o3,o4],g.dupes) eq_(6,len(g.matches)) - + def test_len(self): g = Group() eq_(0,len(g)) g.add_match(get_match(NamedObject("foo",True),NamedObject("bar",True))) eq_(2,len(g)) - + def test_add_same_match_twice(self): g = Group() m = get_match(NamedObject("foo",True),NamedObject("foo",True)) @@ -545,7 +545,7 @@ class TestCaseGroup: g.add_match(m) eq_(2,len(g)) eq_(1,len(g.matches)) - + def test_in(self): g = Group() o1 = NamedObject("foo",True) @@ -554,7 +554,7 @@ class TestCaseGroup: g.add_match(get_match(o1,o2)) assert o1 in g assert o2 in g - + def test_remove(self): g = Group() o1 = NamedObject("foo",True) @@ -571,7 +571,7 @@ class TestCaseGroup: g.remove_dupe(o1) eq_(0,len(g.matches)) eq_(0,len(g)) - + def test_remove_with_ref_dupes(self): g = Group() o1 = NamedObject("foo",True) @@ -584,7 +584,7 @@ class TestCaseGroup: o2.is_ref = True g.remove_dupe(o3) eq_(0,len(g)) - + def test_switch_ref(self): o1 = NamedObject(with_words=True) o2 = NamedObject(with_words=True) @@ -598,7 +598,7 @@ class TestCaseGroup: assert o2 is g.ref g.switch_ref(NamedObject('',True)) assert o2 is g.ref - + def test_switch_ref_from_ref_dir(self): # When the ref dupe is from a ref dir, switch_ref() does nothing o1 = no(with_words=True) @@ -608,7 +608,7 @@ class TestCaseGroup: g.add_match(get_match(o1, o2)) g.switch_ref(o2) assert o1 is g.ref - + def test_get_match_of(self): g = Group() for m in get_match_triangle(): @@ -619,7 +619,7 @@ class TestCaseGroup: assert o in m assert g.get_match_of(NamedObject('',True)) is None assert g.get_match_of(g.ref) is None - + def test_percentage(self): #percentage should return the avg percentage in relation to the ref m1,m2,m3 = get_match_triangle() @@ -638,11 +638,11 @@ class TestCaseGroup: g.add_match(m1) g.add_match(m2) eq_(66,g.percentage) - + def test_percentage_on_empty_group(self): g = Group() eq_(0,g.percentage) - + def test_prioritize(self): m1,m2,m3 = get_match_triangle() o1 = m1.first @@ -658,7 +658,7 @@ class TestCaseGroup: assert o1 is g.ref assert g.prioritize(lambda x:x.name) assert o3 is g.ref - + def test_prioritize_with_tie_breaker(self): # if the ref has the same key as one or more of the dupe, run the tie_breaker func among them g = get_test_group() @@ -666,9 +666,9 @@ class TestCaseGroup: tie_breaker = lambda ref, dupe: dupe is o3 g.prioritize(lambda x:0, tie_breaker) assert g.ref is o3 - + def test_prioritize_with_tie_breaker_runs_on_all_dupes(self): - # Even if a dupe is chosen to switch with ref with a tie breaker, we still run the tie breaker + # Even if a dupe is chosen to switch with ref with a tie breaker, we still run the tie breaker # with other dupes and the newly chosen ref g = get_test_group() o1, o2, o3 = g.ordered @@ -678,7 +678,7 @@ class TestCaseGroup: tie_breaker = lambda ref, dupe: dupe.foo > ref.foo g.prioritize(lambda x:0, tie_breaker) assert g.ref is o3 - + def test_prioritize_with_tie_breaker_runs_only_on_tie_dupes(self): # The tie breaker only runs on dupes that had the same value for the key_func g = get_test_group() @@ -693,7 +693,7 @@ class TestCaseGroup: tie_breaker = lambda ref, dupe: dupe.bar > ref.bar g.prioritize(key_func, tie_breaker) assert g.ref is o2 - + def test_prioritize_with_ref_dupe(self): # when the ref dupe of a group is from a ref dir, make it stay on top. g = get_test_group() @@ -702,7 +702,7 @@ class TestCaseGroup: o2.size = 2 g.prioritize(lambda x: -x.size) assert g.ref is o1 - + def test_prioritize_nothing_changes(self): # prioritize() returns False when nothing changes in the group. g = get_test_group() @@ -710,14 +710,14 @@ class TestCaseGroup: g[1].name = 'b' g[2].name = 'c' assert not g.prioritize(lambda x:x.name) - + def test_list_like(self): g = Group() o1,o2 = (NamedObject("foo",True),NamedObject("bar",True)) g.add_match(get_match(o1,o2)) assert g[0] is o1 assert g[1] is o2 - + def test_discard_matches(self): g = Group() o1,o2,o3 = (NamedObject("foo",True),NamedObject("bar",True),NamedObject("baz",True)) @@ -726,13 +726,13 @@ class TestCaseGroup: g.discard_matches() eq_(1,len(g.matches)) eq_(0,len(g.candidates)) - + class TestCaseget_groups: def test_empty(self): r = get_groups([]) eq_([],r) - + def test_simple(self): l = [NamedObject("foo bar"),NamedObject("bar bleh")] matches = getmatches(l) @@ -742,7 +742,7 @@ class TestCaseget_groups: g = r[0] assert g.ref is m.first eq_([m.second],g.dupes) - + def test_group_with_multiple_matches(self): #This results in 3 matches l = [NamedObject("foo"),NamedObject("foo"),NamedObject("foo")] @@ -751,7 +751,7 @@ class TestCaseget_groups: eq_(1,len(r)) g = r[0] eq_(3,len(g)) - + def test_must_choose_a_group(self): l = [NamedObject("a b"),NamedObject("a b"),NamedObject("b c"),NamedObject("c d"),NamedObject("c d")] #There will be 2 groups here: group "a b" and group "c d" @@ -760,7 +760,7 @@ class TestCaseget_groups: r = get_groups(matches) eq_(2,len(r)) eq_(5,len(r[0])+len(r[1])) - + def test_should_all_go_in_the_same_group(self): l = [NamedObject("a b"),NamedObject("a b"),NamedObject("a b"),NamedObject("a b")] #There will be 2 groups here: group "a b" and group "c d" @@ -768,7 +768,7 @@ class TestCaseget_groups: matches = getmatches(l) r = get_groups(matches) eq_(1,len(r)) - + def test_give_priority_to_matches_with_higher_percentage(self): o1 = NamedObject(with_words=True) o2 = NamedObject(with_words=True) @@ -782,14 +782,14 @@ class TestCaseget_groups: assert o1 not in g assert o2 in g assert o3 in g - + def test_four_sized_group(self): l = [NamedObject("foobar") for i in range(4)] m = getmatches(l) r = get_groups(m) eq_(1,len(r)) eq_(4,len(r[0])) - + def test_referenced_by_ref2(self): o1 = NamedObject(with_words=True) o2 = NamedObject(with_words=True) @@ -799,12 +799,12 @@ class TestCaseget_groups: m3 = get_match(o3,o2) r = get_groups([m1,m2,m3]) eq_(3,len(r[0])) - + def test_job(self): def do_progress(p,d=''): self.log.append(p) return True - + self.log = [] j = job.Job(1,do_progress) m1,m2,m3 = get_match_triangle() @@ -813,7 +813,7 @@ class TestCaseget_groups: get_groups([m1,m2,m3,m4],j) eq_(0,self.log[0]) eq_(100,self.log[-1]) - + def test_group_admissible_discarded_dupes(self): # If, with a (A, B, C, D) set, all match with A, but C and D don't match with B and that the # (A, B) match is the highest (thus resulting in an (A, B) group), still match C and D @@ -830,4 +830,4 @@ class TestCaseget_groups: assert B in g1 assert C in g2 assert D in g2 - + diff --git a/core/tests/scanner_test.py b/core/tests/scanner_test.py index 2e6a21e6..86cb3c98 100644 --- a/core/tests/scanner_test.py +++ b/core/tests/scanner_test.py @@ -1,12 +1,12 @@ # Created By: Virgil Dupras # Created On: 2006/03/03 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license -from jobprogress import job +from hscommon.jobprogress import job from hscommon.path import Path from hscommon.testutil import eq_ @@ -25,10 +25,10 @@ class NamedObject: self.size = size self.path = path self.words = getwords(name) - + def __repr__(self): return '' % (self.name, self.path) - + no = NamedObject @@ -384,7 +384,7 @@ def test_file_evaluates_to_false(fake_fileexists): class FalseNamedObject(NamedObject): def __bool__(self): return False - + s = Scanner() f1 = FalseNamedObject('foobar', path='p1') @@ -445,7 +445,7 @@ def test_tie_breaker_same_name_plus_digit(fake_fileexists): assert group.ref is o5 def test_partial_group_match(fake_fileexists): - # Count the number of discarded matches (when a file doesn't match all other dupes of the + # Count the number of 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') @@ -476,7 +476,7 @@ def test_dont_group_files_that_dont_exist(tmpdir): file2.path.remove() return [Match(file1, file2, 100)] s._getmatches = getmatches - + assert not s.get_dupe_groups([file1, file2]) def test_folder_scan_exclude_subfolder_matches(fake_fileexists): diff --git a/core_pe/matchblock.py b/core_pe/matchblock.py index ddd49334..b2ec7625 100644 --- a/core_pe/matchblock.py +++ b/core_pe/matchblock.py @@ -1,9 +1,9 @@ # Created By: Virgil Dupras # Created On: 2007/02/25 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license import logging @@ -12,7 +12,7 @@ from itertools import combinations from hscommon.util import extract from hscommon.trans import tr -from jobprogress import job +from hscommon.jobprogress import job from core.engine import Match from .block import avgdiff, DifferentBlockCountError, NoBlocksError @@ -132,14 +132,14 @@ def async_compare(ref_ids, other_ids, dbname, threshold, picinfo): results.append((ref_id, other_id, percentage)) cache.close() return results - + def getmatches(pictures, cache_path, threshold=75, match_scaled=False, j=job.nulljob): def get_picinfo(p): if match_scaled: return (None, p.is_ref) else: return (p.dimensions, p.is_ref) - + def collect_results(collect_all=False): # collect results and wait until the queue is small enough to accomodate a new results. nonlocal async_results, matches, comparison_count @@ -152,7 +152,7 @@ def getmatches(pictures, cache_path, threshold=75, match_scaled=False, j=job.nul comparison_count += 1 progress_msg = tr("Performed %d/%d chunk matches") % (comparison_count, len(comparisons_to_do)) j.set_progress(comparison_count, progress_msg) - + j = j.start_subjob([3, 7]) pictures = prepare_pictures(pictures, cache_path, with_dimensions=not match_scaled, j=j) j = j.start_subjob([9, 1], tr("Preparing for matching")) @@ -188,9 +188,14 @@ def getmatches(pictures, cache_path, threshold=75, match_scaled=False, j=job.nul collect_results() collect_results(collect_all=True) pool.close() - + result = [] - for ref_id, other_id, percentage in j.iter_with_progress(matches, tr("Verified %d/%d matches"), every=10): + myiter = j.iter_with_progress( + matches, + tr("Verified %d/%d matches"), + every=10 + ) + for ref_id, other_id, percentage in myiter: ref = id2picture[ref_id] other = id2picture[other_id] if percentage == 100 and ref.md5 != other.md5: @@ -201,4 +206,5 @@ def getmatches(pictures, cache_path, threshold=75, match_scaled=False, j=job.nul result.append(get_match(ref, other, percentage)) return result -multiprocessing.freeze_support() \ No newline at end of file +multiprocessing.freeze_support() + diff --git a/hscommon/__init__.py b/hscommon/__init__.py old mode 100755 new mode 100644 diff --git a/hscommon/gui/progress_window.py b/hscommon/gui/progress_window.py index 28a73d01..36db75ec 100644 --- a/hscommon/gui/progress_window.py +++ b/hscommon/gui/progress_window.py @@ -1,61 +1,58 @@ # Created On: 2013/07/01 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license -from jobprogress.performer import ThreadedJobPerformer - +from ..jobprogress.performer import ThreadedJobPerformer from .base import GUIObject from .text_field import TextField class ProgressWindowView: """Expected interface for :class:`ProgressWindow`'s view. - + *Not actually used in the code. For documentation purposes only.* - + Our view, some kind window with a progress bar, two labels and a cancel button, is expected to properly respond to its callbacks. - + It's also expected to call :meth:`ProgressWindow.cancel` when the cancel button is clicked. """ def show(self): """Show the dialog. """ - + def close(self): """Close the dialog. """ - + def set_progress(self, progress): """Set the progress of the progress bar to ``progress``. - + Not all jobs are equally responsive on their job progress report and it is recommended that you put your progressbar in "indeterminate" mode as long as you haven't received the first ``set_progress()`` call to avoid letting the user think that the app is frozen. - + :param int progress: a value between ``0`` and ``100``. """ class ProgressWindow(GUIObject, ThreadedJobPerformer): """Cross-toolkit GUI-enabled progress window. - - This class allows you to run a long running, `job enabled`_ function in a separate thread and + + This class allows you to run a long running, job enabled function in a separate thread and allow the user to follow its progress with a progress dialog. - + To use it, you start your long-running job with :meth:`run` and then have your UI layer regularly call :meth:`pulse` to refresh the job status in the UI. It is advised that you call :meth:`pulse` in the main thread because GUI toolkit usually only support calling UI-related functions from the main thread. - - We subclass :class:`.GUIObject` and ``ThreadedJobPerformer`` (from the ``jobprogress`` library). + + We subclass :class:`.GUIObject` and :class:`ThreadedJobPerformer`. Expected view: :class:`ProgressWindowView`. - + :param finishfunc: A function ``f(jobid)`` that is called when a job is completed. ``jobid`` is an arbitrary id passed to :meth:`run`. - - .. _job enabled: https://pypi.python.org/pypi/jobprogress """ def __init__(self, finish_func): # finish_func(jobid) is the function that is called when a job is completed. @@ -68,7 +65,7 @@ class ProgressWindow(GUIObject, ThreadedJobPerformer): #: during its course. self.progressdesc_textfield = TextField() self.jobid = None - + def cancel(self): """Call for a user-initiated job cancellation. """ @@ -77,13 +74,13 @@ class ProgressWindow(GUIObject, ThreadedJobPerformer): # we verify that the job is still running. if self._job_running: self.job_cancelled = True - + def pulse(self): """Update progress reports in the GUI. - + Call this regularly from the GUI main run loop. The values might change before :meth:`ProgressWindowView.set_progress` happens. - + If the job is finished, ``pulse()`` will take care of closing the window and re-raising any exception that might have been raised during the job (in the main thread this time). If there was no exception, ``finish_func(jobid)`` is called to let you take appropriate action. @@ -101,13 +98,13 @@ class ProgressWindow(GUIObject, ThreadedJobPerformer): if last_desc: self.progressdesc_textfield.text = last_desc self.view.set_progress(last_progress) - + def run(self, jobid, title, target, args=()): """Starts a threaded job. - - The ``target`` function will be sent, as its first argument, a ``Job`` instance (from the - ``jobprogress`` library) which it can use to report on its progress. - + + The ``target`` function will be sent, as its first argument, a :class:`Job` instance which + it can use to report on its progress. + :param jobid: Arbitrary identifier which will be passed to ``finish_func()`` at the end. :param title: A title for the task you're starting. :param target: The function that does your famous long running job. @@ -122,4 +119,4 @@ class ProgressWindow(GUIObject, ThreadedJobPerformer): self.run_threaded(target, args) self.jobdesc_textfield.text = title self.view.show() - + diff --git a/hscommon/jobprogress/__init__.py b/hscommon/jobprogress/__init__.py new file mode 100644 index 00000000..e69de29b diff --git a/hscommon/jobprogress/job.py b/hscommon/jobprogress/job.py new file mode 100644 index 00000000..6bf5a41e --- /dev/null +++ b/hscommon/jobprogress/job.py @@ -0,0 +1,160 @@ +# Created By: Virgil Dupras +# Created On: 2004/12/20 +# Copyright 2011 Hardcoded Software (http://www.hardcoded.net) + +# This software is licensed under the "BSD" 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/bsd_license + +class JobCancelled(Exception): + "The user has cancelled the job" + +class JobInProgressError(Exception): + "A job is already being performed, you can't perform more than one at the same time." + +class JobCountError(Exception): + "The number of jobs started have exceeded the number of jobs allowed" + +class Job: + """Manages a job's progression and return it's progression through a callback. + + Note that this class is not foolproof. For example, you could call + start_subjob, and then call add_progress from the parent job, and nothing + would stop you from doing it. However, it would mess your progression + because it is the sub job that is supposed to drive the progression. + Another example would be to start a subjob, then start another, and call + add_progress from the old subjob. Once again, it would mess your progression. + There are no stops because it would remove the lightweight aspect of the + class (A Job would need to have a Parent instead of just a callback, + and the parent could be None. A lot of checks for nothing.). + Another one is that nothing stops you from calling add_progress right after + SkipJob. + """ + #---Magic functions + def __init__(self, job_proportions, callback): + """Initialize the Job with 'jobcount' jobs. Start every job with + start_job(). Every time the job progress is updated, 'callback' is called + 'callback' takes a 'progress' int param, and a optional 'desc' + parameter. Callback must return false if the job must be cancelled. + """ + if not hasattr(callback, '__call__'): + raise TypeError("'callback' MUST be set when creating a Job") + if isinstance(job_proportions, int): + job_proportions = [1] * job_proportions + self._job_proportions = list(job_proportions) + self._jobcount = sum(job_proportions) + self._callback = callback + self._current_job = 0 + self._passed_jobs = 0 + self._progress = 0 + self._currmax = 1 + + #---Private + def _subjob_callback(self, progress, desc=''): + """This is the callback passed to children jobs. + """ + self.set_progress(progress, desc) + return True #if JobCancelled has to be raised, it will be at the highest level + + def _do_update(self, desc): + """Calls the callback function with a % progress as a parameter. + + The parameter is a int in the 0-100 range. + """ + if self._current_job: + passed_progress = self._passed_jobs * self._currmax + current_progress = self._current_job * self._progress + total_progress = self._jobcount * self._currmax + progress = ((passed_progress + current_progress) * 100) // total_progress + else: + progress = -1 # indeterminate + # It's possible that callback doesn't support a desc arg + result = self._callback(progress, desc) if desc else self._callback(progress) + if not result: + raise JobCancelled() + + #---Public + def add_progress(self, progress=1, desc=''): + self.set_progress(self._progress + progress, desc) + + def check_if_cancelled(self): + self._do_update('') + + def iter_with_progress(self, sequence, desc_format=None, every=1): + ''' Iterate through sequence while automatically adding progress. + ''' + desc = '' + if desc_format: + desc = desc_format % (0, len(sequence)) + self.start_job(len(sequence), desc) + for i, element in enumerate(sequence, start=1): + yield element + if i % every == 0: + if desc_format: + desc = desc_format % (i, len(sequence)) + self.add_progress(progress=every, desc=desc) + if desc_format: + desc = desc_format % (len(sequence), len(sequence)) + self.set_progress(100, desc) + + def start_job(self, max_progress=100, desc=''): + """Begin work on the next job. You must not call start_job more than + 'jobcount' (in __init__) times. + 'max' is the job units you are to perform. + 'desc' is the description of the job. + """ + self._passed_jobs += self._current_job + try: + self._current_job = self._job_proportions.pop(0) + except IndexError: + raise JobCountError() + self._progress = 0 + self._currmax = max(1, max_progress) + self._do_update(desc) + + def start_subjob(self, job_proportions, desc=''): + """Starts a sub job. Use this when you want to split a job into + multiple smaller jobs. Pretty handy when starting a process where you + know how many subjobs you will have, but don't know the work unit count + for every of them. + returns the Job object + """ + self.start_job(100, desc) + return Job(job_proportions, self._subjob_callback) + + def set_progress(self, progress, desc=''): + """Sets the progress of the current job to 'progress', and call the + callback + """ + self._progress = progress + if self._progress > self._currmax: + self._progress = self._currmax + if self._progress < 0: + self._progress = 0 + self._do_update(desc) + + +class NullJob: + def __init__(self, *args, **kwargs): + pass + + def add_progress(self, *args, **kwargs): + pass + + def check_if_cancelled(self): + pass + + def iter_with_progress(self, sequence, *args, **kwargs): + return iter(sequence) + + def start_job(self, *args, **kwargs): + pass + + def start_subjob(self, *args, **kwargs): + return NullJob() + + def set_progress(self, *args, **kwargs): + pass + + +nulljob = NullJob() diff --git a/hscommon/jobprogress/performer.py b/hscommon/jobprogress/performer.py new file mode 100644 index 00000000..76cf778a --- /dev/null +++ b/hscommon/jobprogress/performer.py @@ -0,0 +1,72 @@ +# Created By: Virgil Dupras +# Created On: 2010-11-19 +# Copyright 2011 Hardcoded Software (http://www.hardcoded.net) +# +# This software is licensed under the "BSD" 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/bsd_license + +from threading import Thread +import sys + +from .job import Job, JobInProgressError, JobCancelled + +class ThreadedJobPerformer: + """Run threaded jobs and track progress. + + To run a threaded job, first create a job with _create_job(), then call _run_threaded(), with + your work function as a parameter. + + Example: + + j = self._create_job() + self._run_threaded(self.some_work_func, (arg1, arg2, j)) + """ + _job_running = False + last_error = None + + #--- Protected + def create_job(self): + if self._job_running: + raise JobInProgressError() + self.last_progress = -1 + self.last_desc = '' + self.job_cancelled = False + return Job(1, self._update_progress) + + def _async_run(self, *args): + target = args[0] + args = tuple(args[1:]) + self._job_running = True + self.last_error = None + try: + target(*args) + except JobCancelled: + pass + except Exception as e: + self.last_error = e + self.last_traceback = sys.exc_info()[2] + finally: + self._job_running = False + self.last_progress = None + + def reraise_if_error(self): + """Reraises the error that happened in the thread if any. + + Call this after the caller of run_threaded detected that self._job_running returned to False + """ + if self.last_error is not None: + raise self.last_error.with_traceback(self.last_traceback) + + def _update_progress(self, newprogress, newdesc=''): + self.last_progress = newprogress + if newdesc: + self.last_desc = newdesc + return not self.job_cancelled + + def run_threaded(self, target, args=()): + if self._job_running: + raise JobInProgressError() + args = (target, ) + args + Thread(target=self._async_run, args=args).start() + diff --git a/hscommon/jobprogress/qt.py b/hscommon/jobprogress/qt.py new file mode 100644 index 00000000..1e18448f --- /dev/null +++ b/hscommon/jobprogress/qt.py @@ -0,0 +1,52 @@ +# Created By: Virgil Dupras +# Created On: 2009-09-14 +# Copyright 2011 Hardcoded Software (http://www.hardcoded.net) +# +# This software is licensed under the "BSD" 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/bsd_license + +from PyQt4.QtCore import pyqtSignal, Qt, QTimer +from PyQt4.QtGui import QProgressDialog + +from . import job, performer + +class Progress(QProgressDialog, performer.ThreadedJobPerformer): + finished = pyqtSignal(['QString']) + + def __init__(self, parent): + flags = Qt.CustomizeWindowHint | Qt.WindowTitleHint | Qt.WindowSystemMenuHint + QProgressDialog.__init__(self, '', "Cancel", 0, 100, parent, flags) + self.setModal(True) + self.setAutoReset(False) + self.setAutoClose(False) + self._timer = QTimer() + self._jobid = '' + self._timer.timeout.connect(self.updateProgress) + + def updateProgress(self): + # the values might change before setValue happens + last_progress = self.last_progress + last_desc = self.last_desc + if not self._job_running or last_progress is None: + self._timer.stop() + self.close() + if not self.job_cancelled: + self.finished.emit(self._jobid) + return + if self.wasCanceled(): + self.job_cancelled = True + return + if last_desc: + self.setLabelText(last_desc) + self.setValue(last_progress) + + def run(self, jobid, title, target, args=()): + self._jobid = jobid + self.reset() + self.setLabelText('') + self.run_threaded(target, args) + self.setWindowTitle(title) + self.show() + self._timer.start(500) + diff --git a/hscommon/path.py b/hscommon/path.py old mode 100755 new mode 100644 diff --git a/package.py b/package.py index e4f6964c..122872ba 100644 --- a/package.py +++ b/package.py @@ -1,9 +1,9 @@ # Created By: Virgil Dupras # Created On: 2009-12-30 # Copyright 2014 Hardcoded Software (http://www.hardcoded.net) -# -# This software is licensed under the "BSD" License as described in the "LICENSE" file, -# which should be included with this package. The terms are also available at +# +# This software is licensed under the "BSD" 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/bsd_license import os @@ -42,10 +42,10 @@ def package_windows(edition, dev): add_to_pythonpath('.') app_version = get_module_version('core_{}'.format(edition)) distdir = 'dist' - + if op.exists(distdir): shutil.rmtree(distdir) - + if not dev: # Copy qt plugins plugin_dest = distdir @@ -129,7 +129,7 @@ def package_debian_distribution(edition, distribution): ed = lambda s: s.format(edition) destpath = op.join('build', 'dupeguru-{0}-{1}'.format(edition, version)) srcpath = op.join(destpath, 'src') - packages = ['hscommon', 'core', ed('core_{0}'), 'qtlib', 'qt', 'send2trash', 'jobprogress'] + packages = ['hscommon', 'core', ed('core_{0}'), 'qtlib', 'qt', 'send2trash'] if edition == 'me': packages.append('hsaudiotag') copy_files_to_package(srcpath, packages, with_so=False) @@ -171,7 +171,7 @@ def package_arch(edition): print("Packaging for Arch") ed = lambda s: s.format(edition) srcpath = op.join('build', ed('dupeguru-{}-arch')) - packages = ['hscommon', 'core', ed('core_{0}'), 'qtlib', 'qt', 'send2trash', 'jobprogress'] + packages = ['hscommon', 'core', ed('core_{0}'), 'qtlib', 'qt', 'send2trash'] if edition == 'me': packages.append('hsaudiotag') copy_files_to_package(srcpath, packages, with_so=True) diff --git a/requirements.txt b/requirements.txt index ec0c07a8..9b90932b 100644 --- a/requirements.txt +++ b/requirements.txt @@ -1,4 +1,3 @@ -jobprogress>=1.0.4 Send2Trash>=1.3.0 sphinx>=1.2.2 polib>=1.0.4