ibus-typing-booster July 2014

ibus-typing-booster@lists.fedorahosted.org

1 participants
8 discussions

[ibus-typing-booster] miketmp-debug: Require Python >= 3.3 (16727e1)
by mfabian＠fedoraproject.org 22 Jul '14

22 Jul '14

1 0

[ibus-typing-booster] miketmp-debug: WIP: usage experiment: de_DE & en_GB dictionary (358ea5b)
by mfabian＠fedoraproject.org 22 Jul '14

22 Jul '14

1 0

[ibus-typing-booster] miketmp-debug: Always write xml output in UTF-8 encoding, not in the encoding of the current locale (52cf6e1)
by mfabian＠fedoraproject.org 22 Jul '14

22 Jul '14

1 0

[ibus-typing-booster] miketmp-debug: MOD5 experiment (5330991)
by mfabian＠fedoraproject.org 22 Jul '14

22 Jul '14

Repository : http://git.fedorahosted.org/git/?p=ibus-typing-booster.git On branch : miketmp-debug >--------------------------------------------------------------- commit 5330991c01311850b23eca2d39426ab3e96399c9 Author: Mike FABIAN <mfabian(a)redhat.com> Date: Thu Feb 20 08:17:03 2014 +0100 MOD5 experiment >--------------------------------------------------------------- ibus-typing-booster/engine/hunspell_table.py | 3 ++- ibus-typing-booster/hunspell-tables/mr_IN.conf | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/ibus-typing-booster/engine/hunspell_table.py b/ibus-typing-booster/engine/hunspell_table.py index d0cf723..3ef28ac 100644 --- a/ibus-typing-booster/engine/hunspell_table.py +++ b/ibus-typing-booster/engine/hunspell_table.py @@ -818,7 +818,7 @@ class tabengine (IBus.Engine): self._editor.insert_string_at_cursor(token) self._update_ui() return True - if key.code >= 32 and (not (key.mask & (IBus.ModifierType.MOD1_MASK | IBus.ModifierType.CONTROL_MASK))): + if key.code >= 32 and (not (key.mask & (IBus.ModifierType.MOD1_MASK | IBus.ModifierType.CONTROL_MASK | IBus.ModifierType.MOD5_MASK))): typed_character = IBus.keyval_to_unicode(key.code) if type(typed_character) != type(u''): typed_character = typed_character.decode('UTF-8') @@ -1026,6 +1026,7 @@ class tabengine (IBus.Engine): typed_character = IBus.keyval_to_unicode(key.code) if type(typed_character) != type(u''): typed_character = typed_character.decode('UTF-8') + sys.stderr.write("mike: MOD5_MASK=%(bit)s\n" %{'bit': key.mask & IBus.ModifierType.MOD5_MASK}) self._editor.insert_string_at_cursor(typed_character) if typed_character and unicodedata.category(typed_character) in itb_util.categories_to_trigger_immediate_commit: input_phrase = self._editor.get_transliterated_string() diff --git a/ibus-typing-booster/hunspell-tables/mr_IN.conf b/ibus-typing-booster/hunspell-tables/mr_IN.conf index 9f644dd..3c2b675 100644 --- a/ibus-typing-booster/hunspell-tables/mr_IN.conf +++ b/ibus-typing-booster/hunspell-tables/mr_IN.conf @@ -6,7 +6,7 @@ description = This is a Marathi typing booster engine table for Marathi language language = mr author = Anish Patil <apatil(a)redhat.com> status_prompt = mr -layout = default +layout = default[lv3:ralt_switch] hunspell_dict = mr_IN.dic hunspell_dict_package = hunspell-mr ime_name = Marathi - IN (Hunspell)

1 0

[ibus-typing-booster] miketmp-debug: add sys.stderr.write("mike ..." debug messages) (e9217fe)
by mfabian＠fedoraproject.org 22 Jul '14

22 Jul '14

Repository : http://git.fedorahosted.org/git/?p=ibus-typing-booster.git On branch : miketmp-debug >--------------------------------------------------------------- commit e9217feda74c4f95cb6f79c3a2a45bc2528ec8e9 Author: Mike FABIAN <mfabian(a)redhat.com> Date: Wed Jun 12 15:55:07 2013 +0200 add sys.stderr.write("mike ..." debug messages) >--------------------------------------------------------------- ibus-typing-booster/engine/factory.py | 1 + ibus-typing-booster/engine/hunspell_suggest.py | 2 ++ ibus-typing-booster/engine/hunspell_table.py | 18 ++++++++++++++++++ ibus-typing-booster/engine/tabsqlitedb.py | 19 +++++++++++++++++++ 4 files changed, 40 insertions(+), 0 deletions(-) diff --git a/ibus-typing-booster/engine/factory.py b/ibus-typing-booster/engine/factory.py index 84e64d6..3b61ebf 100755 --- a/ibus-typing-booster/engine/factory.py +++ b/ibus-typing-booster/engine/factory.py @@ -23,6 +23,7 @@ import hunspell_table import tabsqlitedb import os import re +import sys from gettext import dgettext _ = lambda a : dgettext ("ibus-typing-booster", a) diff --git a/ibus-typing-booster/engine/hunspell_suggest.py b/ibus-typing-booster/engine/hunspell_suggest.py index d04895d..9ddf9b0 100755 --- a/ibus-typing-booster/engine/hunspell_suggest.py +++ b/ibus-typing-booster/engine/hunspell_suggest.py @@ -110,6 +110,7 @@ class Hunspell: def load_dictionaries(self): self.dictionaries = [] + print("mike dictionary_names=%s\n" %self.dictionary_names) for dictionary_name in self.dictionary_names: self.dictionaries.append(Dictionary(name=dictionary_name)) @@ -125,6 +126,7 @@ class Hunspell: return [] if type(input_phrase) != type(u''): input_phrase = input_phrase.decode('utf8') + sys.stderr.write("mike in suggest input_phrase=%(ip)s\n" %{'ip': input_phrase.encode('UTF-8')}) # http://pwet.fr/man/linux/fichiers_speciaux/hunspell says: # # > A dictionary file (*.dic) contains a list of words, one per diff --git a/ibus-typing-booster/engine/hunspell_table.py b/ibus-typing-booster/engine/hunspell_table.py index d09ea34..d0cf723 100644 --- a/ibus-typing-booster/engine/hunspell_table.py +++ b/ibus-typing-booster/engine/hunspell_table.py @@ -84,6 +84,7 @@ class editor(object): '''Hold user inputs chars and preedit string''' def __init__ (self, config, database): + sys.stderr.write("mike editor __init__\n") self.db = database self._config = config self._name = self.db.ime_properties.get('name') @@ -140,6 +141,7 @@ class editor(object): self.trans_m17n_mode = True try: #self.trans = Translit.Transliterator.get(self._m17ndb, self._current_ime) + sys.stderr.write("mike calling Transliterator.get(%(m17n)s, %(cur)s)\n" %{'m17n': self._m17ndb, 'cur': self._current_ime}) self.trans = Transliterator.get(self._m17ndb, self._current_ime) except: import traceback @@ -169,12 +171,19 @@ class editor(object): 'NFKD', self._transliterated_string) else: self._transliterated_string = self._typed_string + sys.stderr.write("mike in update_transliterated_string() self._typed_string=%s\n" %self._typed_string.encode('UTF-8')) + sys.stderr.write("mike in update_transliterated_string() repr(self._typed_string)=%s\n" %repr(self._typed_string)) + sys.stderr.write("mike in update_transliterated_string() self._transliterated_string=%s\n" %self._transliterated_string.encode('UTF-8')) + sys.stderr.write("mike in update_transliterated_string() repr(self._transliterated_string)=%s\n" %repr(self._transliterated_string)) def get_transliterated_string(self): return self._transliterated_string def insert_string_at_cursor(self, string_to_insert): '''Insert typed string at cursor position''' + sys.stderr.write("mike in insert_string_at_cursor() string_to_insert=%s\n" %string_to_insert.encode('UTF-8')) + sys.stderr.write("mike in insert_string_at_cursor() self._typed_string=%s\n" %self._typed_string.encode('UTF-8')) + sys.stderr.write("mike in insert_string_at_cursor() self._typed_string_cursor=%s\n" %self._typed_string_cursor) self._typed_string = self._typed_string[:self._typed_string_cursor] \ +string_to_insert \ +self._typed_string[self._typed_string_cursor:] @@ -333,6 +342,7 @@ class editor(object): def update_candidates (self): '''Update lookuptable''' + sys.stderr.write("mike in update_candidates() self._typed_string=%s\n" %self._typed_string.encode('UTF-8')) if self._typed_string == self._typed_string_when_update_candidates_was_last_called: # The input did not change since we came here last, do nothing and leave # candidates and lookup table unchanged: @@ -501,6 +511,7 @@ class tabengine (IBus.Engine): '''The IM Engine for Tables''' def __init__ (self, bus, obj_path, db ): + sys.stderr.write("mike in tabengine __init__() obj_path=%s\n" %obj_path) super(tabengine,self).__init__ (connection=bus.get_connection(),object_path=obj_path) global debug_level try: @@ -564,6 +575,7 @@ class tabengine (IBus.Engine): def _change_mode (self): '''Shift input mode, TAB -> EN -> TAB ''' + sys.stderr.write("mike in hunspell_table.py _change_mode()\n") self.reset () self._update_ui () @@ -731,7 +743,10 @@ class tabengine (IBus.Engine): if self._has_input_purpose and self._input_purpose in [IBus.InputPurpose.PASSWORD, IBus.InputPurpose.PIN]: return False + sys.stderr.write("mike in process_key_event keyval=%(kv)s keycode=%(kc)s state=%(st)s\n" %{'kv': keyval, 'kc': keycode, 'st': state}) key = KeyEvent(keyval, state) + sys.stderr.write("mike after KeyEvent() in process_key_event key.code=%(kc)s\n" %{'kc': key.code}) + sys.stderr.write("mike after KeyEvent() in process_key_event key.code=%(kc)s IBus.keyval_to_unicode(key.code)=%(uc)s\n" %{'kc': key.code, 'uc': IBus.keyval_to_unicode(key.code)}) # ignore NumLock mask key.mask &= ~IBus.ModifierType.MOD2_MASK @@ -752,6 +767,9 @@ class tabengine (IBus.Engine): return True if self._editor.is_empty (): + sys.stderr.write("mike self._editor.is_empty ():\n") + sys.stderr.write("mike key.code=%(key.code)s IBus.keyval_to_unicode(key.code)=%(keychar)s\n" %{'key.code': key.code, 'keychar': IBus.keyval_to_unicode(key.code)}) + sys.stderr.write("mike IBus.keyval_name(key.code)=%s\n" %IBus.keyval_name(key.code)) # This is the first character typed since the last commit # there is nothing in the preëdit yet. if key.code < 32: diff --git a/ibus-typing-booster/engine/tabsqlitedb.py b/ibus-typing-booster/engine/tabsqlitedb.py index 78a4b26..2815503 100755 --- a/ibus-typing-booster/engine/tabsqlitedb.py +++ b/ibus-typing-booster/engine/tabsqlitedb.py @@ -241,6 +241,8 @@ class tabsqlitedb: 'input_phrase': input_phrase, 'phrase': phrase, 'p_phrase': p_phrase, 'pp_phrase': pp_phrase, 'timestamp': time.time()} + sys.stderr.write("mike update_phrase() sqlstr=%s\n" %sqlstr) + sys.stderr.write("mike update_phrase() sqlargs=%s\n" %sqlargs) try: self.db.execute(sqlstr, sqlargs) if commit: @@ -253,8 +255,10 @@ class tabsqlitedb: ''' Trigger a checkpoint operation. ''' + sys.stderr.write("mike sync_userdb() commit and execute checkpoint ...\n") self.db.commit() self.db.execute('PRAGMA wal_checkpoint;') + sys.stderr.write("mike sync_userdb() commit and execute checkpoint done.\n") def create_tables (self, database): '''Create table for the phrases.''' @@ -269,6 +273,7 @@ class tabsqlitedb: ''' Add phrase to database ''' + sys.stderr.write("mike in add_phrase() input_phrase=%(ip)s phrase=%(p)s user_freq=%(uf)s database=%(db)s\n" %{'ip': input_phrase.encode('UTF-8'), 'p': phrase.encode('UTF-8'), 'uf': user_freq, 'db': database}) if not input_phrase or not phrase: return input_phrase = unicodedata.normalize( @@ -299,6 +304,8 @@ class tabsqlitedb: insert_sqlargs = {'input_phrase': input_phrase, 'phrase': phrase, 'p_phrase': p_phrase, 'pp_phrase': pp_phrase, 'user_freq': user_freq, 'timestamp': time.time()} + sys.stderr.write("mike add_phrase() insert_sqlstr=%s\n" %insert_sqlstr) + sys.stderr.write("mike add_phrase() insert_sqlargs=%s\n" %insert_sqlargs) try: self.db.execute (insert_sqlstr, insert_sqlargs) if commit: @@ -368,9 +375,11 @@ class tabsqlitedb: self._normalization_form_internal, p_phrase) pp_phrase = unicodedata.normalize( self._normalization_form_internal, pp_phrase) + sys.stderr.write("mike in select_words() input_phrase=%(ip)s p_phrase=%(p)s pp_phrase=%(pp)s\n" %{'ip': input_phrase.encode('UTF-8'), 'p': p_phrase.encode('UTF-8'), 'pp': pp_phrase.encode('UTF-8')}) phrase_frequencies = {} for x in self.hunspell_obj.suggest(input_phrase): phrase_frequencies.update([(x, 0)]) + sys.stderr.write("mike in select_words() hunspell: best_candidates=%s\n" %self.best_candidates(phrase_frequencies)) # Now phrase_frequencies might contain something like this: # # {u'code': 0, u'communicability': 0, u'cold': 0, u'colour': 0} @@ -437,6 +446,7 @@ class tabsqlitedb: # {u'conspiracy': 6/11, u'code': 0, u'communicability': 0, u'cold': 1/11, u'colour': 4/11} for x in results_uni: phrase_frequencies.update([(x[0], x[1]/float(count))]) + sys.stderr.write("mike in select_words() Unigram best_candidates=%s\n" %self.best_candidates(phrase_frequencies)) if not p_phrase: # If no context for bigram matching is available, return what we have so far: return self.best_candidates(phrase_frequencies) @@ -461,6 +471,7 @@ class tabsqlitedb: # both the weight of 0.5: for x in results_bi: phrase_frequencies.update([(x[0], 0.5*x[1]/float(count_p_phrase)+0.5*phrase_frequencies[x[0]])]) + sys.stderr.write("mike in select_words() Bigram best_candidates=%s\n" %self.best_candidates(phrase_frequencies)) if not pp_phrase: # If no context for trigram matching is available, return what we have so far: return self.best_candidates(phrase_frequencies) @@ -487,6 +498,7 @@ class tabsqlitedb: # get higher weight): for x in results_tri: phrase_frequencies.update([(x[0], 0.5*x[1]/float(count_pp_phrase_p_phrase)+0.5*phrase_frequencies[x[0]])]) + sys.stderr.write("mike in select_words() Trigram best_candidates=%s\n" %self.best_candidates(phrase_frequencies)) return self.best_candidates(phrase_frequencies) def generate_userdb_desc (self): @@ -592,6 +604,8 @@ CREATE TABLE phrases (id INTEGER PRIMARY KEY, input_phrase TEXT, phrase TEXT, p_ input_phrase = unicodedata.normalize( self._normalization_form_internal, input_phrase) + sys.stderr.write("mike check_phrase() phrase=%(p)s, input_phrase=%(t)s, database=%(d)s\n" %{'p': phrase.encode('UTF-8'), 't': input_phrase.encode('UTF-8'), 'd': database}) + # There should never be more than 1 database row for the same # input_phrase *and* phrase. So the following query on # the database should match at most one database @@ -608,7 +622,10 @@ CREATE TABLE phrases (id INTEGER PRIMARY KEY, input_phrase TEXT, phrase TEXT, p_ ;''' sqlargs = {'input_phrase': input_phrase, 'phrase': phrase, 'p_phrase': p_phrase, 'pp_phrase': pp_phrase} + sys.stderr.write("mike sqlstr=%s\n" %sqlstr) + sys.stderr.write("mike sqlargs=%s\n" %sqlargs) result = self.db.execute(sqlstr, sqlargs).fetchall() + sys.stderr.write("mike result=%s\n" %result) if len(result) > 0: # A match was found in user_db, increase user frequency by 1 self.update_phrase(input_phrase = input_phrase, @@ -630,6 +647,8 @@ CREATE TABLE phrases (id INTEGER PRIMARY KEY, input_phrase TEXT, phrase TEXT, p_ Or, if “input_phrase” is “None”, remove all rows matching “phrase” no matter for what input phrase from the database. ''' + sys.stderr.write("mike remove_phrase() phrase=%(p)s\n" %{'p': phrase.encode('UTF-8')}) + sys.stderr.write("mike remove_phrase() database=%s\n" %database) if not phrase: return phrase = unicodedata.normalize(

1 0

[ibus-typing-booster] branch 'miketmp-debug' created
by mfabian＠fedoraproject.org 22 Jul '14

22 Jul '14

Repository : http://git.fedorahosted.org/git/?p=ibus-typing-booster.git New branch : miketmp-debug Referencing: 52cf6e1eaa472e86dbbc2be12dab1f6a520533ca

1 0

[ibus-typing-booster] branch 'miketmp-debug' deleted
by mfabian＠fedoraproject.org 22 Jul '14

22 Jul '14

Repository : http://git.fedorahosted.org/git/?p=ibus-typing-booster.git Deleted branch: miketmp-debug

1 0

[ibus-typing-booster] miketmp-debug: Always write xml output in UTF-8 encoding, not in the encoding of the current locale (9064f5e)
by mfabian＠fedoraproject.org 21 Jul '14

21 Jul '14

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

ibus-typing-booster July 2014