# -*- coding:utf-8 -*- # # Copyright (C) 2012, Maximilian Köhl # Copyright (C) 2012, Carlos Jenkins # # This program is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation, either version 3 of the License, or # (at your option) any later version. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program. If not, see . # # Some changes were introduced from Giuseppe Penone (giuspen) just to connect to cherrytree # """ A simple but quite powerful spellchecking library written in pure Python for Gtk based on Enchant. It supports PyGObject as well as PyGtk for Python 2 and 3 with automatic switching and binding detection. For automatic translation of the user interface it can use Gedit’s translation files. """ import gtk try: import enchant HAS_PYENCHANT = True except: HAS_PYENCHANT = False import re import sys from pgsc_locales import code_to_name as _code_to_name from pgsc_locales import LanguageNotFound, CountryNotFound # public objects __all__ = ['SpellChecker', 'NoDictionariesFound', 'NoGtkBindingFound'] class NoDictionariesFound(Exception): """ There aren't any dictionaries installed on the current system so spellchecking could not work in any way. """ class NoGtkBindingFound(Exception): """ Could not find any loaded Gtk binding. """ # select base list class try: from collections import UserList _list = UserList except ImportError: _list = list def code_to_name(code, separator='_'): try: return _code_to_name(code, separator) except (LanguageNotFound, CountryNotFound): return '{} ({})'.format(_('Unknown'), code) class SpellChecker(object): """ Main spellchecking class, everything important happens here. :param view: GtkTextView the SpellChecker should be attached to. :param language: The language which should be used for spellchecking. Use a combination of two letter lower-case ISO 639 language code with a two letter upper-case ISO 3166 country code, for example en_US or de_DE. :param prefix: A prefix for some internal GtkTextMarks. :param collapse: Enclose suggestions in its own menu. :param params: Dictionary with Enchant broker parameters that should be set e.g. `enchant.myspell.dictionary.path`. .. attribute:: languages A list of supported languages. .. function:: exists(language) Checks if a language exists. :param language: language to check """ FILTER_WORD = 'word' FILTER_LINE = 'line' FILTER_TEXT = 'text' DEFAULT_FILTERS = {FILTER_WORD : [r'[0-9.,]+'], FILTER_LINE : [(r'(https?|ftp|file):((//)|(\\\\))+[\w\d:' r'#@%/;$()~_?+-=\\.&]+'), r'[\w\d]+@[\w\d.]+'], FILTER_TEXT : []} class _LanguageList(_list): def __init__(self, *args, **kwargs): _list.__init__(self, *args, **kwargs) self.mapping = dict(self) @classmethod def from_broker(cls, broker): return cls(sorted([(language, code_to_name(language)) for language in broker.list_languages()], key=lambda language: language[1])) def exists(self, language): return language in self.mapping class _Mark(): def __init__(self, buffer, name, start): self._buffer = buffer self._name = name self._mark = self._buffer.create_mark(self._name, start, True) @property def iter(self): return self._buffer.get_iter_at_mark(self._mark) @property def inside_word(self): return self.iter.inside_word() @property def word(self): start = self.iter if not start.starts_word(): start.backward_word_start() end = self.iter if end.inside_word(): end.forward_word_end() return start, end def move(self, location): self._buffer.move_mark(self._mark, location) def __init__(self, view, cherrytree_instance, on_rt_node, language='en', prefix='gtkspellchecker', collapse=True, params={}): self._view = view self._cherrytree_instance = cherrytree_instance self.collapse = collapse self._view.connect('populate-popup', lambda entry, menu:self._extend_menu(menu)) self._view.connect('popup-menu', self._click_move_popup) self._view.connect('button-press-event', self._click_move_button) self._prefix = prefix self._misspelled = gtk.TextTag('{}-misspelled'.format(self._prefix)) self._misspelled.set_property('underline', 4) self._broker = enchant.Broker() for param, value in params.items(): self._broker.set_param(param, value) self.languages = SpellChecker._LanguageList.from_broker(self._broker) if self.languages.exists(language): self._language = language elif self.languages.exists('en'): self._language = 'en' else: if self.languages: self._language = self.languages[0][0] else: raise NoDictionariesFound() self._dictionary = self._broker.request_dict(self._language) self._deferred_check = False self._filters = dict(SpellChecker.DEFAULT_FILTERS) self._regexes = {SpellChecker.FILTER_WORD : re.compile('|'.join( self._filters[SpellChecker.FILTER_WORD])), SpellChecker.FILTER_LINE : re.compile('|'.join( self._filters[SpellChecker.FILTER_LINE])), SpellChecker.FILTER_TEXT : re.compile('|'.join( self._filters[SpellChecker.FILTER_TEXT]), re.MULTILINE)} self._enabled = True if on_rt_node: self.buffer_initialize() @property def language(self): """ The language used for spellchecking. """ return self._language @language.setter def language(self, language): if language != self._language and self.languages.exists(language): self._language = language self._dictionary = self._broker.request_dict(language) self.recheck() @property def enabled(self): """ Enable or disable spellchecking. """ return self._enabled @enabled.setter def enabled(self, enabled): if enabled and not self._enabled: self.enable() elif not enabled and self._enabled: self.disable() def buffer_initialize(self): """ Initialize the GtkTextBuffer associated with the GtkTextView. If you have associated a new GtkTextBuffer with the GtkTextView call this method. """ self._buffer = self._view.get_buffer() self._buffer.connect('insert-text', self._before_text_insert) self._buffer.connect_after('insert-text', self._after_text_insert) self._buffer.connect_after('delete-range', self._range_delete) self._buffer.connect_after('mark-set', self._mark_set) start = self._buffer.get_bounds()[0] self._marks = {'insert-start' : SpellChecker._Mark(self._buffer, '{}-insert-start'.format(self._prefix), start), 'insert-end' : SpellChecker._Mark(self._buffer, '{}-insert-end'.format(self._prefix), start), 'click' : SpellChecker._Mark(self._buffer, '{}-click'.format(self._prefix), start)} self._table = self._buffer.get_tag_table() if not self._table.lookup(self._misspelled.get_property("name")): self._table.add(self._misspelled) self.ignored_tags = [] def tag_added(tag, *args): if hasattr(tag, 'spell_check') and not getattr(tag, 'spell_check'): self.ignored_tags.append(tag) def tag_removed(tag, *args): if tag in self.ignored_tags: self.ignored_tags.remove(tag) self._table.connect('tag-added', tag_added) self._table.connect('tag-removed', tag_removed) self._table.foreach(tag_added, None) self.no_spell_check = self._table.lookup('no-spell-check') if not self.no_spell_check: self.no_spell_check = gtk.TextTag('no-spell-check') self._table.add(self.no_spell_check) self.recheck() def recheck(self): """ Rechecks the spelling of the whole text. """ start, end = self._buffer.get_bounds() self.check_range(start, end, True) def disable(self): """ Disable spellchecking. """ self._enabled = False if hasattr(self, "_buffer"): start, end = self._buffer.get_bounds() self._buffer.remove_tag(self._misspelled, start, end) def enable(self): """ Enable spellchecking. """ self._enabled = True def append_filter(self, regex, filter_type): """ Append a new filter to the filter list. Filters are useful to ignore some misspelled words based on regular expressions. :param regex: The regex used for filtering. :param filter_type: The type of the filter. Filter Types: :const:`SpellChecker.FILTER_WORD`: The regex must match the whole word you want to filter. The word separation is done by Pango's word separation algorithm so, for example, urls won't work here because they are split in many words. :const:`SpellChecker.FILTER_LINE`: If the expression you want to match is a single line expression use this type. It should not be an open end expression because then the rest of the line with the text you want to filter will become correct. :const:`SpellChecker.FILTER_TEXT`: Use this if you want to filter multiline expressions. The regex will be compiled with the `re.MULTILINE` flag. Same with open end expressions apply here. """ self._filters[filter_type].append(regex) if filter_type == SpellChecker.FILTER_TEXT: self._regexes[filter_type] = re.compile('|'.join( self._filters[filter_type]), re.MULTILINE) else: self._regexes[filter_type] = re.compile('|'.join( self._filters[filter_type])) def remove_filter(self, regex, filter_type): """ Remove a filter from the filter list. :param regex: The regex which used for filtering. :param filter_type: The type of the filter. """ self._filters[filter_type].remove(regex) if filter_type == SpellChecker.FILTER_TEXT: self._regexes[filter_type] = re.compile('|'.join( self._filters[filter_type]), re.MULTILINE) else: self._regexes[filter_type] = re.compile('|'.join( self._filters[filter_type])) def append_ignore_tag(self, tag): """ Appends a tag to the list of ignored tags. A string will be automatic resolved into a tag object. :param tag: Tag object or tag name. """ if isinstance(tag, basestring): tag = self._table.lookup(tag) self.ignored_tags.append(tag) def remove_ignore_tag(self, tag): """ Removes a tag from the list of ignored tags. A string will be automatic resolved into a tag object. :param tag: Tag object or tag name. """ if isinstance(tag, basestring): tag = self._table.lookup(tag) self.ignored_tags.remove(tag) def add_to_dictionary(self, word): """ Adds a word to user's dictionary. :param word: The word to add. """ self._dictionary.add_to_pwl(word) self.recheck() def ignore_all(self, word): """ Ignores a word for the current session. :param word: The word to ignore. """ self._dictionary.add_to_session(word) self.recheck() def check_range(self, start, end, force_all=False): """ Checks a specified range between two GtkTextIters. :param start: Start iter - checking starts here. :param end: End iter - checking ends here. """ if not self._enabled or not self._cherrytree_instance.user_active: return if start.equal(end): return if end.inside_word(): end.forward_word_end() if not start.starts_word() and (start.inside_word() or start.ends_word()): start.backward_word_start() self._buffer.remove_tag(self._misspelled, start, end) cursor = self._buffer.get_iter_at_mark(self._buffer.get_insert()) precursor = cursor.copy() precursor.backward_char() highlight = (cursor.has_tag(self._misspelled) or precursor.has_tag(self._misspelled)) if not start.get_offset(): start.forward_word_end() start.backward_word_start() word_start = start.copy() while word_start.compare(end) < 0: word_end = word_start.copy() word_end.forward_word_end() in_word = ((word_start.compare(cursor) < 0) and (cursor.compare(word_end) <= 0)) if in_word and not force_all: if highlight: self._check_word(word_start, word_end) else: self._deferred_check = True else: self._check_word(word_start, word_end) self._deferred_check = False word_end.forward_word_end() word_end.backward_word_start() if word_start.equal(word_end): break word_start = word_end.copy() def _languages_menu(self): def _set_language(item, code): self.language = code self._cherrytree_instance.spell_check_notify_new_lang(code) menu = gtk.Menu() group = gtk.RadioMenuItem() connect = [] for code, name in self.languages: item = gtk.RadioMenuItem(group, name) if code == self.language: item.set_active(True) connect.append((item, code)) menu.append(item) for item, code in connect: item.connect('activate', _set_language, code) return menu def _suggestion_menu(self, word): menu = [] suggestions = self._dictionary.suggest(word) if not suggestions: item = gtk.MenuItem() label = gtk.Label() try: label.set_halign(gtk.Align.LEFT) except AttributeError: label.set_alignment(0.0, 0.5) label.set_markup('{text}'.format(text=_('(no suggestions)'))) item.add(label) menu.append(item) else: for suggestion in suggestions: item = gtk.MenuItem() label = gtk.Label() label.set_markup('{text}'.format(text=suggestion)) try: label.set_halign(gtk.Align.LEFT) except AttributeError: label.set_alignment(0.0, 0.5) item.add(label) item.connect('activate', self._replace_word, word, suggestion) menu.append(item) menu.append(gtk.SeparatorMenuItem()) item = gtk.MenuItem(_('Add "{}" to Dictionary').format(word)) item.connect('activate', lambda *args: self.add_to_dictionary(word)) menu.append(item) item = gtk.MenuItem(_('Ignore All')) item.connect('activate', lambda *args: self.ignore_all(word)) menu.append(item) return menu def _extend_menu(self, menu): if not self._enabled or not self._cherrytree_instance.user_active: return separator = gtk.SeparatorMenuItem() separator.show() menu.prepend(separator) languages = gtk.MenuItem(_('Languages')) languages.set_submenu(self._languages_menu()) languages.show_all() menu.prepend(languages) if self._marks['click'].inside_word: start, end = self._marks['click'].word if start.has_tag(self._misspelled): word = self._buffer.get_text(start, end, False).decode('utf-8') items = self._suggestion_menu(word) if self.collapse: suggestions = gtk.MenuItem(_('Suggestions')) submenu = gtk.Menu() for item in items: submenu.append(item) suggestions.set_submenu(submenu) suggestions.show_all() menu.prepend(suggestions) else: items.reverse() for item in items: menu.prepend(item) menu.show_all() def _click_move_popup(self, *args): if not self._enabled or not self._cherrytree_instance.user_active: return False self._marks['click'].move(self._buffer.get_iter_at_mark( self._buffer.get_insert())) return False def _click_move_button(self, widget, event): if not self._enabled or not self._cherrytree_instance.user_active: return if event.button == 3: if self._deferred_check: self._check_deferred_range(True) x, y = self._view.window_to_buffer_coords(2, int(event.x), int(event.y)) self._marks['click'].move(self._view.get_iter_at_location(x, y)) return False def _before_text_insert(self, textbuffer, location, text, length): if not self._cherrytree_instance.user_active: return #print "before '%s'" % text self._marks['insert-start'].move(location) def _after_text_insert(self, textbuffer, location, text, length): if not self._cherrytree_instance.user_active: return #print "after '%s'" % text start = self._marks['insert-start'].iter self.check_range(start, location) self._marks['insert-end'].move(location) def _range_delete(self, textbuffer, start, end): self.check_range(start, end) def _mark_set(self, textbuffer, location, mark): if mark == self._buffer.get_insert() and self._deferred_check: self._check_deferred_range(False) def _replace_word(self, item, old_word, new_word): start, end = self._marks['click'].word offset = start.get_offset() self._buffer.begin_user_action() self._buffer.delete(start, end) self._buffer.insert(self._buffer.get_iter_at_offset(offset), new_word) self._buffer.end_user_action() self._dictionary.store_replacement(old_word, new_word) def _check_deferred_range(self, force_all): start = self._marks['insert-start'].iter end = self._marks['insert-end'].iter self.check_range(start, end, force_all) def _check_word(self, start, end): if start.equal(end): return if start.has_tag(self.no_spell_check): return for tag in self.ignored_tags: if start.has_tag(tag): return word = self._buffer.get_text(start, end, False).decode('utf-8').strip() if " " in word: # BUG! return if len(self._filters[SpellChecker.FILTER_WORD]): if self._regexes[SpellChecker.FILTER_WORD].match(word): return if len(self._filters[SpellChecker.FILTER_LINE]): line_start = self._buffer.get_iter_at_line(start.get_line()) line_end = end.copy() line_end.forward_to_line_end() line = self._buffer.get_text(line_start, line_end, False).decode('utf-8') for match in self._regexes[SpellChecker.FILTER_LINE].finditer(line): if match.start() <= start.get_line_offset() <= match.end(): start = self._buffer.get_iter_at_line_offset( start.get_line(), match.start()) end = self._buffer.get_iter_at_line_offset(start.get_line(), match.end()) self._buffer.remove_tag(self._misspelled, start, end) return if len(self._filters[SpellChecker.FILTER_TEXT]): text_start, text_end = self._buffer.get_bounds() text = self._buffer.get_text(text_start, text_end, False).decode('utf-8') for match in self._regexes[SpellChecker.FILTER_TEXT].finditer(text): if match.start() <= start.get_offset() <= match.end(): start = self._buffer.get_iter_at_offset(match.start()) end = self._buffer.get_iter_at_offset(match.end()) self._buffer.remove_tag(self._misspelled, start, end) return if word and not self._dictionary.check(word): self._buffer.apply_tag(self._misspelled, start, end)