From 9c025f4736d477b8478ae0684dff8aeeca41fe4c Mon Sep 17 00:00:00 2001 From: Hiroshi Miura Date: Fri, 3 Dec 2010 00:43:59 +0900 Subject: [PATCH 01/22] recipe: fix japanese recipes - mainichi news: handle ad pages - yomiuri: remove table tag - nikkei life: reduce feeds --- resources/recipes/mainichi.recipe | 16 ++++++++++++++++ resources/recipes/mainichi_it_news.recipe | 16 +++++++++++++++- resources/recipes/nikkei_sub_life.recipe | 5 +---- resources/recipes/yomiuri.recipe | 2 +- resources/recipes/yomiuri_world.recipe | 2 +- 5 files changed, 34 insertions(+), 7 deletions(-) diff --git a/resources/recipes/mainichi.recipe b/resources/recipes/mainichi.recipe index 2a44fa0980..baa7f409ec 100644 --- a/resources/recipes/mainichi.recipe +++ b/resources/recipes/mainichi.recipe @@ -4,6 +4,7 @@ __copyright__ = '2010, Hiroshi Miura ' www.mainichi.jp ''' +import re from calibre.web.feeds.news import BasicNewsRecipe class MainichiDailyNews(BasicNewsRecipe): @@ -22,3 +23,18 @@ class MainichiDailyNews(BasicNewsRecipe): remove_tags = [{'class':"RelatedArticle"}] remove_tags_after = {'class':"Credit"} + def parse_feeds(self): + + feeds = BasicNewsRecipe.parse_feeds(self) + + for curfeed in feeds: + delList = [] + for a,curarticle in enumerate(curfeed.articles): + if re.search(r'pheedo.jp', curarticle.url): + delList.append(curarticle) + if len(delList)>0: + for d in delList: + index = curfeed.articles.index(d) + curfeed.articles[index:index+1] = [] + + return feeds diff --git a/resources/recipes/mainichi_it_news.recipe b/resources/recipes/mainichi_it_news.recipe index 8e15496e57..4c285a2c01 100644 --- a/resources/recipes/mainichi_it_news.recipe +++ b/resources/recipes/mainichi_it_news.recipe @@ -14,5 +14,19 @@ class MainichiDailyITNews(BasicNewsRecipe): remove_tags_before = {'class':"NewsTitle"} remove_tags = [{'class':"RelatedArticle"}] - remove_tags_after = {'class':"Credit"} + def parse_feeds(self): + + feeds = BasicNewsRecipe.parse_feeds(self) + + for curfeed in feeds: + delList = [] + for a,curarticle in enumerate(curfeed.articles): + if re.search(r'pheedo.jp', curarticle.url): + delList.append(curarticle) + if len(delList)>0: + for d in delList: + index = curfeed.articles.index(d) + curfeed.articles[index:index+1] = [] + + return feeds remove_tags_after = {'class':"Credit"} diff --git a/resources/recipes/nikkei_sub_life.recipe b/resources/recipes/nikkei_sub_life.recipe index 1bfa08a55f..60e5b170ca 100644 --- a/resources/recipes/nikkei_sub_life.recipe +++ b/resources/recipes/nikkei_sub_life.recipe @@ -32,12 +32,9 @@ class NikkeiNet_sub_life(BasicNewsRecipe): remove_tags_after = {'class':"cmn-pr_list"} feeds = [ (u'\u304f\u3089\u3057', u'http://www.zou3.net/php/rss/nikkei2rss.php?head=kurashi'), - (u'\u30b9\u30dd\u30fc\u30c4', u'http://www.zou3.net/php/rss/nikkei2rss.php?head=sports'), - (u'\u793e\u4f1a', u'http://www.zou3.net/php/rss/nikkei2rss.php?head=shakai'), (u'\u30a8\u30b3', u'http://www.zou3.net/php/rss/nikkei2rss.php?head=eco'), (u'\u5065\u5eb7', u'http://www.zou3.net/php/rss/nikkei2rss.php?head=kenkou'), - (u'\u7279\u96c6', u'http://www.zou3.net/php/rss/nikkei2rss.php?head=special'), - (u'\u30e9\u30f3\u30ad\u30f3\u30b0', u'http://www.zou3.net/php/rss/nikkei2rss.php?head=ranking') + (u'\u7279\u96c6', u'http://www.zou3.net/php/rss/nikkei2rss.php?head=special') ] def get_browser(self): diff --git a/resources/recipes/yomiuri.recipe b/resources/recipes/yomiuri.recipe index d30aa9066f..fb17bb1210 100644 --- a/resources/recipes/yomiuri.recipe +++ b/resources/recipes/yomiuri.recipe @@ -21,7 +21,7 @@ class YOLNews(BasicNewsRecipe): remove_javascript = True masthead_title = u'YOMIURI ONLINE' - remove_tags_before = {'class':"article-def"} + keep_only_tags = [{'class':"article-def"}] remove_tags = [{'class':"RelatedArticle"}, {'class':"sbtns"} ] diff --git a/resources/recipes/yomiuri_world.recipe b/resources/recipes/yomiuri_world.recipe index f5f21c4aab..41ee4fd23d 100644 --- a/resources/recipes/yomiuri_world.recipe +++ b/resources/recipes/yomiuri_world.recipe @@ -21,7 +21,7 @@ class YOLNews(BasicNewsRecipe): remove_javascript = True masthead_title = u"YOMIURI ONLINE" - remove_tags_before = {'class':"article-def"} + keep_only_tags = [{'class':"article-def"}] remove_tags = [{'class':"RelatedArticle"}, {'class':"sbtns"} ] From 6173ff8c591f5205555758a5cc94dc0b8a44eb94 Mon Sep 17 00:00:00 2001 From: Hiroshi Miura Date: Fri, 3 Dec 2010 00:46:25 +0900 Subject: [PATCH 02/22] recipe: add nikkei news about social --- resources/recipes/nikkei_sub_shakai.recipe | 102 +++++++++++++++++++++ 1 file changed, 102 insertions(+) create mode 100644 resources/recipes/nikkei_sub_shakai.recipe diff --git a/resources/recipes/nikkei_sub_shakai.recipe b/resources/recipes/nikkei_sub_shakai.recipe new file mode 100644 index 0000000000..ed86493265 --- /dev/null +++ b/resources/recipes/nikkei_sub_shakai.recipe @@ -0,0 +1,102 @@ +__license__ = 'GPL v3' +__copyright__ = '2010, Hiroshi Miura ' +''' +www.nikkei.com +''' + +import re +from calibre.web.feeds.recipes import BasicNewsRecipe +import mechanize +from calibre.ptempfile import PersistentTemporaryFile + + +class NikkeiNet_sub_life(BasicNewsRecipe): + title = u'\u65e5\u7d4c\u65b0\u805e\u96fb\u5b50\u7248(\u751f\u6d3b)' + __author__ = 'Hiroshi Miura' + description = 'News and current market affairs from Japan' + cover_url = 'http://parts.nikkei.com/parts/ds/images/common/logo_r1.svg' + masthead_url = 'http://parts.nikkei.com/parts/ds/images/common/logo_r1.svg' + needs_subscription = True + oldest_article = 2 + max_articles_per_feed = 20 + language = 'ja' + remove_javascript = False + temp_files = [] + + remove_tags_before = {'class':"cmn-section cmn-indent"} + remove_tags = [ + {'class':"JSID_basePageMove JSID_baseAsyncSubmit cmn-form_area JSID_optForm_utoken"}, + {'class':"cmn-article_keyword cmn-clearfix"}, + {'class':"cmn-print_headline cmn-clearfix"}, + ] + remove_tags_after = {'class':"cmn-pr_list"} + + feeds = [ + (u'\u793e\u4f1a', u'http://www.zou3.net/php/rss/nikkei2rss.php?head=shakai') + ] + + def get_browser(self): + br = BasicNewsRecipe.get_browser() + + cj = mechanize.LWPCookieJar() + br.set_cookiejar(cj) + + #br.set_debug_http(True) + #br.set_debug_redirects(True) + #br.set_debug_responses(True) + + if self.username is not None and self.password is not None: + #print "----------------------------get login form--------------------------------------------" + # open login form + br.open('https://id.nikkei.com/lounge/nl/base/LA0010.seam') + response = br.response() + #print "----------------------------get login form---------------------------------------------" + #print "----------------------------set login form---------------------------------------------" + # remove disabled input which brings error on mechanize + response.set_data(response.get_data().replace("", " -->")) + br.set_response(response) + br.select_form(name='LA0010Form01') + br['LA0010Form01:LA0010Email'] = self.username + br['LA0010Form01:LA0010Password'] = self.password + br.form.find_control(id='LA0010Form01:LA0010AutoLoginOn',type="checkbox").get(nr=0).selected = True + br.submit() + br.response() + #print "----------------------------send login form---------------------------------------------" + #print "----------------------------open news main page-----------------------------------------" + # open news site + br.open('http://www.nikkei.com/') + br.response() + #print "----------------------------www.nikkei.com BODY --------------------------------------" + #print response2.get_data() + #print "-------------------------^^-got auto redirect form----^^--------------------------------" + # forced redirect in default + br.select_form(nr=0) + br.submit() + response3 = br.response() + # return some cookie which should be set by Javascript + #print response3.geturl() + raw = response3.get_data() + #print "---------------------------response to form --------------------------------------------" + # grab cookie from JS and set it + redirectflag = re.search(r"var checkValue = '(\d+)';", raw, re.M).group(1) + br.select_form(nr=0) + + self.temp_files.append(PersistentTemporaryFile('_fa.html')) + self.temp_files[-1].write("#LWP-Cookies-2.0\n") + + self.temp_files[-1].write("Set-Cookie3: Cookie-dummy=Cookie-value; domain=\".nikkei.com\"; path=\"/\"; path_spec; secure; expires=\"2029-12-21 05:07:59Z\"; version=0\n") + self.temp_files[-1].write("Set-Cookie3: redirectFlag="+redirectflag+"; domain=\".nikkei.com\"; path=\"/\"; path_spec; secure; expires=\"2029-12-21 05:07:59Z\"; version=0\n") + self.temp_files[-1].close() + cj.load(self.temp_files[-1].name) + + br.submit() + + #br.set_debug_http(False) + #br.set_debug_redirects(False) + #br.set_debug_responses(False) + return br + + + + From a42306248360bcb19c518fd8f8d3ad028e4538fe Mon Sep 17 00:00:00 2001 From: Charles Haley <> Date: Mon, 6 Dec 2010 06:52:08 +0000 Subject: [PATCH 03/22] 1) Add icu capitalize 2) Fix strcmp to use icu lower 3) Add capitalize to tests 4) change formatter & title_case to use icu capitalize 5) add capitalize as a function to bulk metadata edit 6) get rid of redundant lower()s in models --- src/calibre/gui2/dialogs/metadata_bulk.py | 3 ++- src/calibre/gui2/library/models.py | 3 +-- src/calibre/utils/formatter.py | 3 ++- src/calibre/utils/icu.py | 18 +++++++++++++----- src/calibre/utils/titlecase.py | 7 ++----- 5 files changed, 20 insertions(+), 14 deletions(-) diff --git a/src/calibre/gui2/dialogs/metadata_bulk.py b/src/calibre/gui2/dialogs/metadata_bulk.py index 4a44b0cefa..a640c50fb8 100644 --- a/src/calibre/gui2/dialogs/metadata_bulk.py +++ b/src/calibre/gui2/dialogs/metadata_bulk.py @@ -17,7 +17,7 @@ from calibre.gui2 import error_dialog from calibre.gui2.progress_indicator import ProgressIndicator from calibre.utils.config import dynamic from calibre.utils.titlecase import titlecase -from calibre.utils.icu import sort_key +from calibre.utils.icu import sort_key, capitalize class MyBlockingBusy(QDialog): @@ -187,6 +187,7 @@ class MetadataBulkDialog(QDialog, Ui_MetadataBulkDialog): _('Lower Case') : lambda x: icu_lower(x), _('Upper Case') : lambda x: icu_upper(x), _('Title Case') : lambda x: titlecase(x), + _('Capitalize') : lambda x: capitalize(x), } s_r_match_modes = [ _('Character match'), diff --git a/src/calibre/gui2/library/models.py b/src/calibre/gui2/library/models.py index 311cbaf369..e82e1dddd4 100644 --- a/src/calibre/gui2/library/models.py +++ b/src/calibre/gui2/library/models.py @@ -1023,8 +1023,7 @@ class DeviceBooksModel(BooksModel): # {{{ x = '' if y == None: y = '' - x, y = icu_lower(x.strip()), icu_lower(y.strip()) - return icu_strcmp(x, y) + return icu_strcmp(x.strip(), y.strip()) return _strcmp def datecmp(x, y): x = self.db[x].datetime diff --git a/src/calibre/utils/formatter.py b/src/calibre/utils/formatter.py index 15534a9c8a..6c4ddce81f 100644 --- a/src/calibre/utils/formatter.py +++ b/src/calibre/utils/formatter.py @@ -8,6 +8,7 @@ import re, string, traceback from calibre.constants import DEBUG from calibre.utils.titlecase import titlecase +from calibre.utils.icu import capitalize class TemplateFormatter(string.Formatter): ''' @@ -86,7 +87,7 @@ class TemplateFormatter(string.Formatter): 'uppercase' : (0, lambda s,x: x.upper()), 'lowercase' : (0, lambda s,x: x.lower()), 'titlecase' : (0, lambda s,x: titlecase(x)), - 'capitalize' : (0, lambda s,x: x.capitalize()), + 'capitalize' : (0, lambda s,x: capitalize(x)), 'contains' : (3, _contains), 'ifempty' : (1, _ifempty), 'lookup' : (-1, _lookup), diff --git a/src/calibre/utils/icu.py b/src/calibre/utils/icu.py index d7760671c9..6ae7398fb4 100644 --- a/src/calibre/utils/icu.py +++ b/src/calibre/utils/icu.py @@ -69,7 +69,7 @@ def icu_case_sensitive_sort_key(collator, obj): return collator.sort_key(obj) def icu_strcmp(collator, a, b): - return collator.strcmp(a.lower(), b.lower()) + return collator.strcmp(lower(a), lower(b)) def py_strcmp(a, b): return cmp(a.lower(), b.lower()) @@ -104,6 +104,13 @@ lower = (lambda s: s.lower()) if _icu_not_ok else \ title_case = (lambda s: s.title()) if _icu_not_ok else \ partial(_icu.title, get_locale()) +def icu_capitalize(s): + s = lower(s) + return s.replace(s[0], upper(s[0])) + +capitalize = (lambda s: s.capitalize()) if _icu_not_ok else \ + (lambda s: icu_capitalize(s)) + ################################################################################ def test(): # {{{ @@ -215,14 +222,15 @@ pêché''' print '\t', x.encode('utf-8') if fs != create(french_good): print 'French failed (note that French fails with icu < 4.6 i.e. on windows and OS X)' - return + # return test_strcmp(german + french) print '\nTesting case transforms in current locale' for x in ('a', 'Alice\'s code'): - print 'Upper:', x, '->', 'py:', x.upper().encode('utf-8'), 'icu:', upper(x).encode('utf-8') - print 'Lower:', x, '->', 'py:', x.lower().encode('utf-8'), 'icu:', lower(x).encode('utf-8') - print 'Title:', x, '->', 'py:', x.title().encode('utf-8'), 'icu:', title_case(x).encode('utf-8') + print 'Upper: ', x, '->', 'py:', x.upper().encode('utf-8'), 'icu:', upper(x).encode('utf-8') + print 'Lower: ', x, '->', 'py:', x.lower().encode('utf-8'), 'icu:', lower(x).encode('utf-8') + print 'Title: ', x, '->', 'py:', x.title().encode('utf-8'), 'icu:', title_case(x).encode('utf-8') + print 'Capitalize:', x, '->', 'py:', x.capitalize().encode('utf-8'), 'icu:', capitalize(x).encode('utf-8') print # }}} diff --git a/src/calibre/utils/titlecase.py b/src/calibre/utils/titlecase.py index b85670f038..bbc4c26688 100755 --- a/src/calibre/utils/titlecase.py +++ b/src/calibre/utils/titlecase.py @@ -9,6 +9,8 @@ License: http://www.opensource.org/licenses/mit-license.php import re +from calibre.utils.icu import capitalize + __all__ = ['titlecase'] __version__ = '0.5' @@ -40,11 +42,6 @@ def titlecase(text): """ - def capitalize(w): - w = icu_lower(w) - w = w.replace(w[0], icu_upper(w[0])) - return w - all_caps = ALL_CAPS.match(text) words = re.split('\s', text) From e31735960bf42e443c59e1f5fec52d809dadb363 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Mon, 6 Dec 2010 00:33:37 -0700 Subject: [PATCH 04/22] Save to disk: Refactor to not open a database connection in the worker process. Also fix a bug that could lead to save failures not being reported. --- src/calibre/ebooks/metadata/worker.py | 91 ++++++++++++++++----- src/calibre/gui2/add.py | 13 +++ src/calibre/library/save_to_disk.py | 111 +++++++++++++++++--------- 3 files changed, 160 insertions(+), 55 deletions(-) diff --git a/src/calibre/ebooks/metadata/worker.py b/src/calibre/ebooks/metadata/worker.py index 247050856d..7dff988679 100644 --- a/src/calibre/ebooks/metadata/worker.py +++ b/src/calibre/ebooks/metadata/worker.py @@ -8,12 +8,12 @@ __docformat__ = 'restructuredtext en' from threading import Thread from Queue import Empty -import os, time, sys, shutil +import os, time, sys, shutil, json from calibre.utils.ipc.job import ParallelJob from calibre.utils.ipc.server import Server from calibre.ptempfile import PersistentTemporaryDirectory, TemporaryDirectory -from calibre import prints +from calibre import prints, isbytestring from calibre.constants import filesystem_encoding @@ -194,14 +194,44 @@ class SaveWorker(Thread): self.daemon = True self.path, self.opts = path, opts self.ids = ids - self.library_path = db.library_path + self.db = db self.canceled = False self.result_queue = result_queue self.error = None self.spare_server = spare_server self.start() + def collect_data(self, ids): + from calibre.ebooks.metadata.opf2 import metadata_to_opf + data = {} + for i in set(ids): + mi = self.db.get_metadata(i, index_is_id=True, get_cover=True) + opf = metadata_to_opf(mi) + if isbytestring(opf): + opf = opf.decode('utf-8') + cpath = None + if mi.cover: + cpath = mi.cover + if isbytestring(cpath): + cpath = cpath.decode(filesystem_encoding) + formats = {} + fmts = self.db.formats(i, index_is_id=True, verify_formats=False) + if fmts: + fmts = fmts.split(',') + for fmt in fmts: + fpath = self.db.format_abspath(i, fmt, index_is_id=True) + if fpath is not None: + if isbytestring(fpath): + fpath = fpath.decode(filesystem_encoding) + formats[fmt.lower()] = fpath + data[i] = [opf, cpath, formats] + return data + def run(self): + with TemporaryDirectory('save_to_disk_data') as tdir: + self._run(tdir) + + def _run(self, tdir): from calibre.library.save_to_disk import config server = Server() if self.spare_server is None else self.spare_server ids = set(self.ids) @@ -212,12 +242,19 @@ class SaveWorker(Thread): for pref in c.preferences: recs[pref.name] = getattr(self.opts, pref.name) + plugboards = self.db.prefs.get('plugboards', {}) + for i, task in enumerate(tasks): tids = [x[-1] for x in task] + data = self.collect_data(tids) + dpath = os.path.join(tdir, '%d.json'%i) + with open(dpath, 'wb') as f: + f.write(json.dumps(data, ensure_ascii=False).encode('utf-8')) + job = ParallelJob('save_book', 'Save books (%d of %d)'%(i, len(tasks)), lambda x,y:x, - args=[tids, self.library_path, self.path, recs]) + args=[tids, dpath, plugboards, self.path, recs]) jobs.add(job) server.add_job(job) @@ -226,21 +263,19 @@ class SaveWorker(Thread): time.sleep(0.2) running = False for job in jobs: - job.update(consume_notifications=False) - while True: - try: - id, title, ok, tb = job.notifications.get_nowait()[0] - if id in ids: - self.result_queue.put((id, title, ok, tb)) - ids.remove(id) - except Empty: - break + self.get_notifications(job, ids) if not job.is_finished: running = True if not running: break + for job in jobs: + for id_, title, ok, tb in job.result: + if id_ in ids: + self.result_queue.put((id_, title, ok, tb)) + ids.remove(id_) + server.close() time.sleep(1) @@ -257,21 +292,39 @@ class SaveWorker(Thread): except: pass + def get_notifications(self, job, ids): + job.update(consume_notifications=False) + while True: + try: + id, title, ok, tb = job.notifications.get_nowait()[0] + if id in ids: + self.result_queue.put((id, title, ok, tb)) + ids.remove(id) + except Empty: + break -def save_book(task, library_path, path, recs, notification=lambda x,y:x): - from calibre.library.database2 import LibraryDatabase2 - db = LibraryDatabase2(library_path) - from calibre.library.save_to_disk import config, save_to_disk + +def save_book(ids, dpath, plugboards, path, recs, notification=lambda x,y:x): + from calibre.library.save_to_disk import config, save_serialized_to_disk from calibre.customize.ui import apply_null_metadata opts = config().parse() for name in recs: setattr(opts, name, recs[name]) + results = [] def callback(id, title, failed, tb): + results.append((id, title, not failed, tb)) notification((id, title, not failed, tb)) return True - with apply_null_metadata: - save_to_disk(db, task, path, opts, callback) + data_ = json.loads(open(dpath, 'rb').read().decode('utf-8')) + data = {} + for k, v in data_.iteritems(): + data[int(k)] = v + + with apply_null_metadata: + save_serialized_to_disk(ids, data, plugboards, path, opts, callback) + + return results diff --git a/src/calibre/gui2/add.py b/src/calibre/gui2/add.py index 1339070446..d1af2a6f0c 100644 --- a/src/calibre/gui2/add.py +++ b/src/calibre/gui2/add.py @@ -427,11 +427,23 @@ class Saver(QObject): # {{{ if not self.ids or not self.worker.is_alive(): self.timer.stop() self.pd.hide() + while self.ids: + before = len(self.ids) + self.get_result() + if before == len(self.ids): + for i in list(self.ids): + self.failures.add(('id:%d'%i, 'Unknown error')) + self.ids.remove(i) + break if not self.callback_called: self.callback(self.worker.path, self.failures, self.worker.error) self.callback_called = True return + self.get_result() + + + def get_result(self): try: id, title, ok, tb = self.rq.get_nowait() except Empty: @@ -441,6 +453,7 @@ class Saver(QObject): # {{{ if not isinstance(title, unicode): title = str(title).decode(preferred_encoding, 'replace') self.pd.set_msg(_('Saved')+' '+title) + if not ok: self.failures.add((title, tb)) # }}} diff --git a/src/calibre/library/save_to_disk.py b/src/calibre/library/save_to_disk.py index c6cc12a978..af57d563ac 100644 --- a/src/calibre/library/save_to_disk.py +++ b/src/calibre/library/save_to_disk.py @@ -6,7 +6,7 @@ __license__ = 'GPL v3' __copyright__ = '2009, Kovid Goyal ' __docformat__ = 'restructuredtext en' -import os, traceback, cStringIO, re +import os, traceback, cStringIO, re, shutil from calibre.constants import DEBUG from calibre.utils.config import Config, StringConfig, tweaks @@ -203,31 +203,49 @@ def get_components(template, mi, id, timefmt='%b %Y', length=250, return shorten_components_to(length, components) -def save_book_to_disk(id, db, root, opts, length): - mi = db.get_metadata(id, index_is_id=True) +def save_book_to_disk(id_, db, root, opts, length): + mi = db.get_metadata(id_, index_is_id=True) + cover = db.cover(id_, index_is_id=True, as_path=True) + plugboards = db.prefs.get('plugboards', {}) - available_formats = db.formats(id, index_is_id=True) + available_formats = db.formats(id_, index_is_id=True) if not available_formats: available_formats = [] else: available_formats = [x.lower().strip() for x in available_formats.split(',')] + formats = {} + fmts = db.formats(id_, index_is_id=True, verify_formats=False) + if fmts: + fmts = fmts.split(',') + for fmt in fmts: + fpath = db.format_abspath(id_, fmt, index_is_id=True) + if fpath is not None: + formats[fmt.lower()] = fpath + + return do_save_book_to_disk(id_, mi, cover, plugboards, + formats, root, opts, length) + + +def do_save_book_to_disk(id_, mi, cover, plugboards, + format_map, root, opts, length): + available_formats = [x.lower().strip() for x in format_map.keys()] if opts.formats == 'all': asked_formats = available_formats else: asked_formats = [x.lower().strip() for x in opts.formats.split(',')] formats = set(available_formats).intersection(set(asked_formats)) if not formats: - return True, id, mi.title + return True, id_, mi.title - components = get_components(opts.template, mi, id, opts.timefmt, length, + components = get_components(opts.template, mi, id_, opts.timefmt, length, ascii_filename if opts.asciiize else sanitize_file_name, to_lowercase=opts.to_lowercase, replace_whitespace=opts.replace_whitespace) base_path = os.path.join(root, *components) base_name = os.path.basename(base_path) dirpath = os.path.dirname(base_path) - # Don't test for existence first are the test could fail but + # Don't test for existence first as the test could fail but # another worker process could create the directory before # the call to makedirs try: @@ -236,29 +254,23 @@ def save_book_to_disk(id, db, root, opts, length): if not os.path.exists(dirpath): raise - cdata = db.cover(id, index_is_id=True) - if opts.save_cover: - if cdata is not None: - with open(base_path+'.jpg', 'wb') as f: - f.write(cdata) - mi.cover = base_name+'.jpg' - else: - mi.cover = None + if opts.save_cover and cover and os.access(cover, os.R_OK): + with open(base_path+'.jpg', 'wb') as f: + with open(cover, 'rb') as s: + shutil.copyfileobj(s, f) + mi.cover = base_name+'.jpg' + else: + mi.cover = None if opts.write_opf: opf = metadata_to_opf(mi) with open(base_path+'.opf', 'wb') as f: f.write(opf) - if cdata is not None: - mi.cover_data = ('jpg', cdata) - mi.cover = None - written = False for fmt in formats: global plugboard_save_to_disk_value, plugboard_any_format_value dev_name = plugboard_save_to_disk_value - plugboards = db.prefs.get('plugboards', {}) cpb = None if fmt in plugboards: cpb = plugboards[fmt] @@ -275,11 +287,12 @@ def save_book_to_disk(id, db, root, opts, length): # Leave this here for a while, in case problems arise. if cpb is not None: prints('Save-to-disk using plugboard:', fmt, cpb) - data = db.format(id, fmt, index_is_id=True) - if data is None: + fp = format_map.get(fmt, None) + if fp is None: continue - else: - written = True + with open(fp, 'rb') as f: + data = f.read() + written = True if opts.update_metadata: stream = cStringIO.StringIO() stream.write(data) @@ -300,9 +313,21 @@ def save_book_to_disk(id, db, root, opts, length): with open(fmt_path, 'wb') as f: f.write(data) - return not written, id, mi.title + return not written, id_, mi.title +def _sanitize_args(root, opts): + if opts is None: + opts = config().parse() + if isinstance(root, unicode): + root = root.encode(filesystem_encoding) + root = os.path.abspath(root) + opts.template = preprocess_template(opts.template) + length = 1000 if supports_long_names(root) else 250 + length -= len(root) + if length < 5: + raise ValueError('%r is too long.'%root) + return root, opts, length def save_to_disk(db, ids, root, opts=None, callback=None): ''' @@ -316,17 +341,7 @@ def save_to_disk(db, ids, root, opts=None, callback=None): :return: A list of failures. Each element of the list is a tuple (id, title, traceback) ''' - if opts is None: - opts = config().parse() - if isinstance(root, unicode): - root = root.encode(filesystem_encoding) - root = os.path.abspath(root) - - opts.template = preprocess_template(opts.template) - length = 1000 if supports_long_names(root) else 250 - length -= len(root) - if length < 5: - raise ValueError('%r is too long.'%root) + root, opts, length = _sanitize_args(root, opts) failures = [] for x in ids: tb = '' @@ -343,4 +358,28 @@ def save_to_disk(db, ids, root, opts=None, callback=None): break return failures +def save_serialized_to_disk(ids, data, plugboards, root, opts, callback): + from calibre.ebooks.metadata.opf2 import OPF + root, opts, length = _sanitize_args(root, opts) + failures = [] + for x in ids: + opf, cover, format_map = data[x] + if isinstance(opf, unicode): + opf = opf.encode('utf-8') + mi = OPF(cStringIO.StringIO(opf)).to_book_metadata() + tb = '' + try: + failed, id, title = do_save_book_to_disk(x, mi, cover, plugboards, + format_map, root, opts, length) + tb = _('Requested formats not available') + except: + failed, id, title = True, x, mi.title + tb = traceback.format_exc() + if failed: + failures.append((id, title, tb)) + if callable(callback): + if not callback(int(id), title, failed, tb): + break + + return failures From f9a861b3c884bcf1a4a940d38acd1efbf2ac6d47 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Mon, 6 Dec 2010 09:09:36 -0700 Subject: [PATCH 05/22] ... --- src/calibre/ebooks/metadata/worker.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/calibre/ebooks/metadata/worker.py b/src/calibre/ebooks/metadata/worker.py index 7dff988679..a7a8177176 100644 --- a/src/calibre/ebooks/metadata/worker.py +++ b/src/calibre/ebooks/metadata/worker.py @@ -271,6 +271,8 @@ class SaveWorker(Thread): break for job in jobs: + if not job.result: + continue for id_, title, ok, tb in job.result: if id_ in ids: self.result_queue.put((id_, title, ok, tb)) From f5b4029751afc18267b20794d14e3aa03c74aa08 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Mon, 6 Dec 2010 09:44:45 -0700 Subject: [PATCH 06/22] ... --- src/calibre/ebooks/metadata/worker.py | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/src/calibre/ebooks/metadata/worker.py b/src/calibre/ebooks/metadata/worker.py index a7a8177176..d059d7e34c 100644 --- a/src/calibre/ebooks/metadata/worker.py +++ b/src/calibre/ebooks/metadata/worker.py @@ -215,10 +215,8 @@ class SaveWorker(Thread): if isbytestring(cpath): cpath = cpath.decode(filesystem_encoding) formats = {} - fmts = self.db.formats(i, index_is_id=True, verify_formats=False) - if fmts: - fmts = fmts.split(',') - for fmt in fmts: + if mi.formats: + for fmt in mi.formats: fpath = self.db.format_abspath(i, fmt, index_is_id=True) if fpath is not None: if isbytestring(fpath): From 54c3ecced6c4fc86f656c18c183ad1d1464a4bf2 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Mon, 6 Dec 2010 10:00:25 -0700 Subject: [PATCH 07/22] ... --- src/calibre/gui2/add.py | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/src/calibre/gui2/add.py b/src/calibre/gui2/add.py index d1af2a6f0c..5f41f3a8e0 100644 --- a/src/calibre/gui2/add.py +++ b/src/calibre/gui2/add.py @@ -436,6 +436,10 @@ class Saver(QObject): # {{{ self.ids.remove(i) break if not self.callback_called: + try: + self.worker.join(1.5) + except: + pass # The worker was not yet started self.callback(self.worker.path, self.failures, self.worker.error) self.callback_called = True return From 2ee84bad3ad6e646a08a9bbca5712fa45d2b11a9 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Mon, 6 Dec 2010 10:08:18 -0700 Subject: [PATCH 08/22] ... --- src/calibre/utils/icu.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/calibre/utils/icu.py b/src/calibre/utils/icu.py index 6ae7398fb4..4b0f6d4821 100644 --- a/src/calibre/utils/icu.py +++ b/src/calibre/utils/icu.py @@ -56,7 +56,7 @@ def py_sort_key(obj): def icu_sort_key(collator, obj): if not obj: return _none2 - return collator.sort_key(obj.lower()) + return collator.sort_key(lower(obj)) def py_case_sensitive_sort_key(obj): if not obj: From 73f54f5e9dc2a6238f52aa352b2e198c5ed68cb4 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Mon, 6 Dec 2010 10:16:46 -0700 Subject: [PATCH 09/22] Fix #7812 (Bookmarks don't work.) --- src/calibre/utils/zipfile.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/calibre/utils/zipfile.py b/src/calibre/utils/zipfile.py index dbcc125274..5c19444bd6 100644 --- a/src/calibre/utils/zipfile.py +++ b/src/calibre/utils/zipfile.py @@ -1227,7 +1227,7 @@ class ZipFile: self.fp.flush() if zinfo.flag_bits & 0x08: # Write CRC and file sizes after the file data - self.fp.write(struct.pack(" Date: Mon, 6 Dec 2010 10:18:53 -0700 Subject: [PATCH 10/22] Fix #7815 (Device support for HTC Legend with Wordplayer) --- src/calibre/devices/android/driver.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/calibre/devices/android/driver.py b/src/calibre/devices/android/driver.py index 0deef5eb92..46fad13a2a 100644 --- a/src/calibre/devices/android/driver.py +++ b/src/calibre/devices/android/driver.py @@ -21,7 +21,7 @@ class ANDROID(USBMS): # HTC 0x0bb4 : { 0x0c02 : [0x100, 0x0227, 0x0226], 0x0c01 : [0x100, 0x0227], 0x0ff9 : [0x0100, 0x0227, 0x0226], 0x0c87: [0x0100, 0x0227, 0x0226], - 0xc92 : [0x100]}, + 0xc92 : [0x100], 0xc97: [0x226]}, # Eken 0x040d : { 0x8510 : [0x0001] }, From c9510ba730224cc41f2acdd1dc343a40770d90c8 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Mon, 6 Dec 2010 10:20:41 -0700 Subject: [PATCH 11/22] Fix #7802 (Driver for Samsumg Epic?) --- src/calibre/devices/android/driver.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/calibre/devices/android/driver.py b/src/calibre/devices/android/driver.py index 46fad13a2a..9c37b6ff59 100644 --- a/src/calibre/devices/android/driver.py +++ b/src/calibre/devices/android/driver.py @@ -63,7 +63,7 @@ class ANDROID(USBMS): WINDOWS_MAIN_MEM = ['ANDROID_PHONE', 'A855', 'A853', 'INC.NEXUS_ONE', '__UMS_COMPOSITE', '_MB200', 'MASS_STORAGE', '_-_CARD', 'SGH-I897', 'GT-I9000', 'FILE-STOR_GADGET', 'SGH-T959', 'SAMSUNG_ANDROID', - 'SCH-I500_CARD'] + 'SCH-I500_CARD', 'SPH-D700_CARD'] WINDOWS_CARD_A_MEM = ['ANDROID_PHONE', 'GT-I9000_CARD', 'SGH-I897', 'FILE-STOR_GADGET', 'SGH-T959', 'SAMSUNG_ANDROID'] From 51f48f0cb2923b41d6a69c46d803aabbd5b36cfc Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Mon, 6 Dec 2010 11:00:25 -0700 Subject: [PATCH 12/22] ... --- src/calibre/utils/icu.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/src/calibre/utils/icu.c b/src/calibre/utils/icu.c index 51d9ac25ba..38542a44c6 100644 --- a/src/calibre/utils/icu.c +++ b/src/calibre/utils/icu.c @@ -237,8 +237,6 @@ static PyTypeObject icu_CollatorType = { // {{{ // }} -// }}} - // }}} // Module initialization {{{ From 2ed1365eb16d028e58218635cc838fb3617452ea Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Mon, 6 Dec 2010 12:20:15 -0700 Subject: [PATCH 13/22] Edit metadata dialog: When trying to download metadata, if there are multiple matches indicate which matches have a cover and summary in the list. Also add an option to automatically download the cover of the selected match. --- src/calibre/gui2/__init__.py | 2 + src/calibre/gui2/dialogs/fetch_metadata.py | 22 +- src/calibre/gui2/dialogs/fetch_metadata.ui | 351 ++++++++++---------- src/calibre/gui2/dialogs/metadata_single.py | 11 +- 4 files changed, 206 insertions(+), 180 deletions(-) diff --git a/src/calibre/gui2/__init__.py b/src/calibre/gui2/__init__.py index 57b914877d..57ca2a1880 100644 --- a/src/calibre/gui2/__init__.py +++ b/src/calibre/gui2/__init__.py @@ -123,6 +123,8 @@ def _config(): help=_('Download social metadata (tags/rating/etc.)')) c.add_opt('overwrite_author_title_metadata', default=True, help=_('Overwrite author and title with new metadata')) + c.add_opt('auto_download_cover', default=False, + help=_('Automatically download the cover, if available')) c.add_opt('enforce_cpu_limit', default=True, help=_('Limit max simultaneous jobs to number of CPUs')) c.add_opt('tag_browser_hidden_categories', default=set(), diff --git a/src/calibre/gui2/dialogs/fetch_metadata.py b/src/calibre/gui2/dialogs/fetch_metadata.py index 2c64219464..3da0e67e3d 100644 --- a/src/calibre/gui2/dialogs/fetch_metadata.py +++ b/src/calibre/gui2/dialogs/fetch_metadata.py @@ -9,7 +9,7 @@ from threading import Thread from PyQt4.QtCore import Qt, QObject, SIGNAL, QVariant, pyqtSignal, \ QAbstractTableModel, QCoreApplication, QTimer -from PyQt4.QtGui import QDialog, QItemSelectionModel +from PyQt4.QtGui import QDialog, QItemSelectionModel, QIcon from calibre.gui2.dialogs.fetch_metadata_ui import Ui_FetchMetadata from calibre.gui2 import error_dialog, NONE, info_dialog, config @@ -42,13 +42,14 @@ class Matches(QAbstractTableModel): def __init__(self, matches): self.matches = matches + self.yes_icon = QVariant(QIcon(I('ok.png'))) QAbstractTableModel.__init__(self) def rowCount(self, *args): return len(self.matches) def columnCount(self, *args): - return 6 + return 8 def headerData(self, section, orientation, role): if role != Qt.DisplayRole: @@ -61,6 +62,8 @@ class Matches(QAbstractTableModel): elif section == 3: text = _("Publisher") elif section == 4: text = _("ISBN") elif section == 5: text = _("Published") + elif section == 6: text = _("Has Cover") + elif section == 7: text = _("Has Summary") return QVariant(text) else: @@ -71,8 +74,8 @@ class Matches(QAbstractTableModel): def data(self, index, role): row, col = index.row(), index.column() + book = self.matches[row] if role == Qt.DisplayRole: - book = self.matches[row] res = None if col == 0: res = book.title @@ -90,6 +93,11 @@ class Matches(QAbstractTableModel): if not res: return NONE return QVariant(res) + elif role == Qt.DecorationRole: + if col == 6 and book.has_cover: + return self.yes_icon + if col == 7 and book.comments: + return self.yes_icon return NONE class FetchMetadata(QDialog, Ui_FetchMetadata): @@ -131,7 +139,7 @@ class FetchMetadata(QDialog, Ui_FetchMetadata): self.fetch_metadata() self.opt_get_social_metadata.setChecked(config['get_social_metadata']) self.opt_overwrite_author_title_metadata.setChecked(config['overwrite_author_title_metadata']) - + self.opt_auto_download_cover.setChecked(config['auto_download_cover']) def show_summary(self, current, *args): row = current.row() @@ -213,6 +221,12 @@ class FetchMetadata(QDialog, Ui_FetchMetadata): _hung_fetchers.add(self.fetcher) if hasattr(self, '_hangcheck') and self._hangcheck.isActive(): self._hangcheck.stop() + # Save value of auto_download_cover, since this is the only place it can + # be set. The values of the other options can be set in + # Preferences->Behavior and should not be set here as they affect bulk + # downloading as well. + if self.opt_auto_download_cover.isChecked() != config['auto_download_cover']: + config.set('auto_download_cover', self.opt_auto_download_cover.isChecked()) def __enter__(self, *args): return self diff --git a/src/calibre/gui2/dialogs/fetch_metadata.ui b/src/calibre/gui2/dialogs/fetch_metadata.ui index 03a362096c..b140fa158d 100644 --- a/src/calibre/gui2/dialogs/fetch_metadata.ui +++ b/src/calibre/gui2/dialogs/fetch_metadata.ui @@ -1,172 +1,179 @@ - - - FetchMetadata - - - Qt::WindowModal - - - - 0 - 0 - 830 - 642 - - - - Fetch metadata - - - - :/images/metadata.png:/images/metadata.png - - - - - - <p>calibre can find metadata for your books from two locations: <b>Google Books</b> and <b>isbndb.com</b>. <p>To use isbndb.com you must sign up for a <a href="http://www.isbndb.com">free account</a> and enter your access key below. - - - Qt::AlignCenter - - - true - - - true - - - - - - - - - &Access Key: - - - key - - - - - - - - - - Fetch - - - - - - - - - - - - true - - - - - - - Matches - - - - - - Select the book that most closely matches your copy from the list below - - - - - - - - 0 - 1 - - - - true - - - QAbstractItemView::SingleSelection - - - QAbstractItemView::SelectRows - - - - - - - - - - - - - Download &social metadata (tags/rating/etc.) for the selected book - - - - - - - Overwrite author and title with author and title of selected book - - - - - - - QDialogButtonBox::Cancel|QDialogButtonBox::Ok - - - - - - - - - - - buttonBox - accepted() - FetchMetadata - accept() - - - 460 - 599 - - - 657 - 530 - - - - - buttonBox - rejected() - FetchMetadata - reject() - - - 417 - 599 - - - 0 - 491 - - - - - + + + FetchMetadata + + + Qt::WindowModal + + + + 0 + 0 + 890 + 642 + + + + Fetch metadata + + + + :/images/metadata.png:/images/metadata.png + + + + + + <p>calibre can find metadata for your books from two locations: <b>Google Books</b> and <b>isbndb.com</b>. <p>To use isbndb.com you must sign up for a <a href="http://www.isbndb.com">free account</a> and enter your access key below. + + + Qt::AlignCenter + + + true + + + true + + + + + + + + + &Access Key: + + + key + + + + + + + + + + Fetch + + + + + + + + + + + + true + + + + + + + Matches + + + + + + Select the book that most closely matches your copy from the list below + + + + + + + + 0 + 1 + + + + true + + + QAbstractItemView::SingleSelection + + + QAbstractItemView::SelectRows + + + + + + + + + + + + + Overwrite author and title with author and title of selected book + + + + + + + Download &social metadata (tags/rating/etc.) for the selected book + + + + + + + Automatically download the cover, if available + + + + + + + QDialogButtonBox::Cancel|QDialogButtonBox::Ok + + + + + + + + + + + buttonBox + accepted() + FetchMetadata + accept() + + + 460 + 599 + + + 657 + 530 + + + + + buttonBox + rejected() + FetchMetadata + reject() + + + 417 + 599 + + + 0 + 491 + + + + + diff --git a/src/calibre/gui2/dialogs/metadata_single.py b/src/calibre/gui2/dialogs/metadata_single.py index 8f068075cf..fec58a74f6 100644 --- a/src/calibre/gui2/dialogs/metadata_single.py +++ b/src/calibre/gui2/dialogs/metadata_single.py @@ -760,8 +760,8 @@ class MetadataSingleDialog(ResizableDialog, Ui_MetadataSingleDialog): if book.publisher: self.publisher.setEditText(book.publisher) if book.isbn: self.isbn.setText(book.isbn) if book.pubdate: - d = book.pubdate - self.pubdate.setDate(QDate(d.year, d.month, d.day)) + dt = book.pubdate + self.pubdate.setDate(QDate(dt.year, dt.month, dt.day)) summ = book.comments if summ: prefix = unicode(self.comments.toPlainText()) @@ -777,8 +777,11 @@ class MetadataSingleDialog(ResizableDialog, Ui_MetadataSingleDialog): self.series.setText(book.series) if book.series_index is not None: self.series_index.setValue(book.series_index) - # Needed because of Qt focus bug on OS X - self.fetch_cover_button.setFocus(Qt.OtherFocusReason) + if book.has_cover: + if d.opt_auto_download_cover.isChecked() and book.has_cover: + self.fetch_cover() + else: + self.fetch_cover_button.setFocus(Qt.OtherFocusReason) else: error_dialog(self, _('Cannot fetch metadata'), _('You must specify at least one of ISBN, Title, ' From 0d52b87a558e2cde56705ef37a8490be33b958c6 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Mon, 6 Dec 2010 17:30:18 -0700 Subject: [PATCH 14/22] Fix #7786 (Feedback when installing new plugins) --- src/calibre/gui2/preferences/plugins.py | 1 + 1 file changed, 1 insertion(+) diff --git a/src/calibre/gui2/preferences/plugins.py b/src/calibre/gui2/preferences/plugins.py index 388227e438..d493b615b5 100644 --- a/src/calibre/gui2/preferences/plugins.py +++ b/src/calibre/gui2/preferences/plugins.py @@ -151,6 +151,7 @@ class ConfigWidget(ConfigWidgetBase, Ui_Form): self._plugin_model.populate() self._plugin_model.reset() self.changed_signal.emit() + self.plugin_path.setText('') else: error_dialog(self, _('No valid plugin path'), _('%s is not a valid plugin path')%path).exec_() From dbff870b893a19515cbd3ad3f4004e11735e5893 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Mon, 6 Dec 2010 18:30:34 -0700 Subject: [PATCH 15/22] Fix #7716 (Edit meta Information - cover download crash - os x) --- src/calibre/gui2/dialogs/metadata_single.py | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/src/calibre/gui2/dialogs/metadata_single.py b/src/calibre/gui2/dialogs/metadata_single.py index fec58a74f6..3205b1d23c 100644 --- a/src/calibre/gui2/dialogs/metadata_single.py +++ b/src/calibre/gui2/dialogs/metadata_single.py @@ -8,8 +8,9 @@ add/remove formats import os, re, time, traceback, textwrap from functools import partial +from threading import Thread -from PyQt4.Qt import SIGNAL, QObject, Qt, QTimer, QThread, QDate, \ +from PyQt4.Qt import SIGNAL, QObject, Qt, QTimer, QDate, \ QPixmap, QListWidgetItem, QDialog, pyqtSignal, QMessageBox, QIcon, \ QPushButton @@ -34,9 +35,12 @@ from calibre.gui2.preferences.social import SocialMetadata from calibre.gui2.custom_column_widgets import populate_metadata_page from calibre import strftime -class CoverFetcher(QThread): # {{{ +class CoverFetcher(Thread): # {{{ def __init__(self, username, password, isbn, timeout, title, author): + Thread.__init__(self) + self.daemon = True + self.username = username.strip() if username else username self.password = password.strip() if password else password self.timeout = timeout @@ -44,8 +48,7 @@ class CoverFetcher(QThread): # {{{ self.title = title self.needs_isbn = False self.author = author - QThread.__init__(self) - self.exception = self.traceback = self.cover_data = None + self.exception = self.traceback = self.cover_data = self.errors = None def run(self): try: @@ -238,20 +241,20 @@ class MetadataSingleDialog(ResizableDialog, Ui_MetadataSingleDialog): self.timeout, title, author) self.cover_fetcher.start() self._hangcheck = QTimer(self) - self.connect(self._hangcheck, SIGNAL('timeout()'), self.hangcheck) + self._hangcheck.timeout.connect(self.hangcheck, + type=Qt.QueuedConnection) self.cf_start_time = time.time() self.pi.start(_('Downloading cover...')) self._hangcheck.start(100) def hangcheck(self): - if not self.cover_fetcher.isFinished() and \ + if self.cover_fetcher.is_alive() and \ time.time()-self.cf_start_time < self.COVER_FETCH_TIMEOUT: return self._hangcheck.stop() try: - if self.cover_fetcher.isRunning(): - self.cover_fetcher.terminate() + if self.cover_fetcher.is_alive(): error_dialog(self, _('Cannot fetch cover'), _('Could not fetch cover.
')+ _('The download timed out.')).exec_() From 1034e9a73d97ddec1582886137a80bb384717e26 Mon Sep 17 00:00:00 2001 From: Charles Haley <> Date: Tue, 7 Dec 2010 10:25:46 +0000 Subject: [PATCH 16/22] Make template formatting support numeric format specifications --- src/calibre/manual/template_lang.rst | 10 +++-- src/calibre/utils/formatter.py | 59 +++++++++++++++++++++------- 2 files changed, 50 insertions(+), 19 deletions(-) diff --git a/src/calibre/manual/template_lang.rst b/src/calibre/manual/template_lang.rst index 1bef32fbd6..b316850798 100644 --- a/src/calibre/manual/template_lang.rst +++ b/src/calibre/manual/template_lang.rst @@ -46,7 +46,6 @@ and if a book does not have a series:: (|app| automatically removes multiple slashes and leading or trailing spaces). - Advanced formatting ---------------------- @@ -80,6 +79,9 @@ For trailing zeros, use:: {series_index:0<3s} - Three digits with trailing zeros +If you use series indices with sub values (e.g., 1.1), you might want to ensure that the decimal points line up. For example, you might want the indices 1 and 2.5 to appear as 01.00 and 02.50 so that they will sort correctly. To do this, use:: + + {series_index:0<5.2f} - Five characters, consisting of two digits with leading zeros, a decimal point, then 2 digits after the decimal point If you want only the first two letters of the data, use:: @@ -115,15 +117,15 @@ The functions available are: * ``lowercase()`` -- return value of the field in lower case. * ``uppercase()`` -- return the value of the field in upper case. * ``titlecase()`` -- return the value of the field in title case. - * ``capitalize()`` -- return the value as capitalized. - * ``ifempty(text)`` -- if the field is not empty, return the value of the field. Otherwise return `text`. - * ``test(text if not empty, text if empty)`` -- return `text if not empty` if the field is not empty, otherwise return `text if empty`. + * ``capitalize()`` -- return the value with the first letter upper case and the rest lower case. * ``contains(pattern, text if match, text if not match`` -- checks if field contains matches for the regular expression `pattern`. Returns `text if match` if matches are found, otherwise it returns `text if no match`. * ``count(separator)`` -- interprets the value as a list of items separated by `separator`, returning the number of items in the list. Most lists use a comma as the separator, but authors uses an ampersand. Examples: `{tags:count(,)}`, `{authors:count(&)}` + * ``ifempty(text)`` -- if the field is not empty, return the value of the field. Otherwise return `text`. * ``lookup(pattern, field, pattern, field, ..., else_field)`` -- like switch, except the arguments are field (metadata) names, not text. The value of the appropriate field will be fetched and used. Note that because composite columns are fields, you can use this function in one composite field to use the value of some other composite field. This is extremely useful when constructing variable save paths (more later). * ``re(pattern, replacement)`` -- return the field after applying the regular expression. All instances of `pattern` are replaced with `replacement`. As in all of |app|, these are python-compatible regular expressions. * ``shorten(left chars, middle text, right chars)`` -- Return a shortened version of the field, consisting of `left chars` characters from the beginning of the field, followed by `middle text`, followed by `right chars` characters from the end of the string. `Left chars` and `right chars` must be integers. For example, assume the title of the book is `Ancient English Laws in the Times of Ivanhoe`, and you want it to fit in a space of at most 15 characters. If you use ``{title:shorten(9,-,5)}``, the result will be `Ancient E-nhoe`. If the field's length is less than ``left chars`` + ``right chars`` + the length of ``middle text``, then the field will be used intact. For example, the title `The Dome` would not be changed. * ``switch(pattern, value, pattern, value, ..., else_value)`` -- for each ``pattern, value`` pair, checks if the field matches the regular expression ``pattern`` and if so, returns that ``value``. If no ``pattern`` matches, then ``else_value`` is returned. You can have as many ``pattern, value`` pairs as you want. + * ``test(text if not empty, text if empty)`` -- return `text if not empty` if the field is not empty, otherwise return `text if empty`. Now, about using functions and formatting in the same field. Suppose you have an integer custom column called ``#myint`` that you want to see with leading zeros, as in ``003``. To do this, you would use a format of ``0>3s``. However, by default, if a number (integer or float) equals zero then the field produces the empty value, so zero values will produce nothing, not ``000``. If you really want to see ``000`` values, then you use both the format string and the ``ifempty`` function to change the empty value back to a zero. The field reference would be:: diff --git a/src/calibre/utils/formatter.py b/src/calibre/utils/formatter.py index 6c4ddce81f..a7fb3682aa 100644 --- a/src/calibre/utils/formatter.py +++ b/src/calibre/utils/formatter.py @@ -15,6 +15,8 @@ class TemplateFormatter(string.Formatter): Provides a format function that substitutes '' for any missing value ''' + _validation_string = 'This Is Some Text THAT SHOULD be LONG Enough.%^&*' + # Dict to do recursion detection. It is up the the individual get_value # method to use it. It is cleared when starting to format a template composite_values = {} @@ -98,19 +100,29 @@ class TemplateFormatter(string.Formatter): 'count' : (1, _count), } - format_string_re = re.compile(r'^(.*)\|(.*)\|(.*)$') - compress_spaces = re.compile(r'\s+') - backslash_comma_to_comma = re.compile(r'\\,') - - arg_parser = re.Scanner([ - (r',', lambda x,t: ''), - (r'.*?((?= 0: + try: + val = int(val) + except: + raise ValueError( + _('format: type {0} requires an integer value, got {1}').format(typ, val)) + elif 'eEfFgGn%'.find(typ) >= 0: + try: + val = float(val) + except: + raise ValueError( + _('format: type {0} requires a decimal (float) value, got {1}').format(typ, val)) + else: + raise ValueError(_('format: unknown format type letter {0}').format(typ)) + return unicode(('{0:'+fmt+'}').format(val)) def _explode_format_string(self, fmt): try: @@ -123,6 +135,21 @@ class TemplateFormatter(string.Formatter): traceback.print_exc() return fmt, '', '' + format_string_re = re.compile(r'^(.*)\|(.*)\|(.*)$') + compress_spaces = re.compile(r'\s+') + backslash_comma_to_comma = re.compile(r'\\,') + + arg_parser = re.Scanner([ + (r',', lambda x,t: ''), + (r'.*?((? Date: Tue, 7 Dec 2010 08:37:30 -0700 Subject: [PATCH 17/22] St. Louis Post Dispatch by cisaak --- .../recipes/st_louis_post_dispatch.recipe | 70 +++++++++++++++++++ src/calibre/devices/android/driver.py | 2 +- 2 files changed, 71 insertions(+), 1 deletion(-) create mode 100644 resources/recipes/st_louis_post_dispatch.recipe diff --git a/resources/recipes/st_louis_post_dispatch.recipe b/resources/recipes/st_louis_post_dispatch.recipe new file mode 100644 index 0000000000..3b7701cedc --- /dev/null +++ b/resources/recipes/st_louis_post_dispatch.recipe @@ -0,0 +1,70 @@ +from calibre.web.feeds.news import BasicNewsRecipe + +class AdvancedUserRecipe1282093204(BasicNewsRecipe): + title = u'St Louis Post-Dispatch' + __author__ = 'cisaak' + language = 'en' + + oldest_article = 1 + max_articles_per_feed = 15 + masthead_url = 'http://farm5.static.flickr.com/4118/4929686950_0e22e2c88a.jpg' + + feeds = [ + (u'News-Bill McClellan', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Flocal%2fcolumns%2Fbill-mclellan&f=rss&t=article'), + (u'News-Columns', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Flocal%2Fcolumns*&l=50&f=rss&t=article'), + (u'News-Crime & Courtshttp://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Flocal%2Fcrime-and-courts&l=50&f=rss&t=article'), + (u'News-Deb Peterson', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Flocal%2fcolumns%2Fdeb-peterson&f=rss&t=article'), + (u'News-Education', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Flocal%2feducation&f=rss&t=article'), + (u'News-Government & Politics', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Flocal%2fgovt-and-politics&f=rss&t=article'), + (u'News-Local', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Flocal&f=rss&t=article'), + (u'News-Metro', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Flocal%2fmetro&f=rss&t=article'), + (u'News-Metro East', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Flocal%2fillinois&f=rss&t=article'), + (u'News-Missouri Out State', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Fstate-and-regional%2FMissouri&l=50&f=rss&t=article'), + (u'Opinion-Colleen Carroll Campbell', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Fopinion%2fcolumns%2Fcolleen-carroll-campbell&f=rss&t=article'), + (u'Opinion-Editorial', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Fopinion%2feditorial&f=rss&t=article'), + (u'Opinion-Kevin Horrigan', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Fopinion%2fcolumns%2Fkevin-horrigan&f=rss&t=article'), + (u'Opinion-Mailbag', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=news%2Fopinion%2fmailbag&f=rss&t=article'), + (u'Business Columns-Savvy Consumer', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=business%2Fcolumns%2Fsavvy-consumer&l=100&f=rss&t=article'), + (u'Business Columns-Lager Heads', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=business%2Fcolumns%2Flager-heads&l=100&f=rss&t=article'), + (u'Business Columns-Job Watch', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=business%2Fcolumns%2Fjob-watch&l=100&f=rss&t=article'), + (u'Business Columns-Steve Geigerich', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=business%2Fcolumns%2Fsteve-giegerich&l=100&f=rss&t=article'), + (u'Business Columns-David Nicklaus', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=business%2Fcolumns%2Fdavid-nicklaus&l=100&f=rss&t=article'), + (u'Business Columns-Jim Gallagher', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=business%2Fcolumns%2Fjim-gallagher&l=100&f=rss&t=article'), + (u'Business Columns-Building Blocks', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=business%2Fcolumns%2Fbuilding-blocks&l=100&f=rss&t=article'), + (u'Business', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=business*l&l=100&f=rss&t=article'), + (u'Business-Technology', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=business%2Ftechnology&l=50&f=rss&t=article'), + (u'Business-National', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=business%2Fnational-and-international&l=50&f=rss&t=article'), + (u'Travel', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=travel*&l=100&f=rss&t=article'), + (u'Sports', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=sports*&f=rss&t=article'), + (u'Sports-Baseball', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=sports%2Fbaseball%2Fprofessional&l=100&f=rss&t=article'), + (u'Sports-Bernie Miklasz', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=sports%2Fcolumns%2Fbernie-miklasz&l=50&f=rss&t=article'), + (u'Sports-Bryan Burwell', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=sports%2Fcolumns%2Fbryan-burwell&l=50&f=rss&t=article'), + (u'Sports-College', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=sports%2Fcollege*&l=100&f=rss&t=article'), + (u'Sports-Dan Caesar', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=sports%2Fcolumns%2Fdan-caesar&l=50&f=rss&t=article'), + (u'Sports-Football', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=sports%2Ffootball%2Fprofessional&l=100&f=rss&t=article'), + (u'Sports-Hockey', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=sports%2Fhockey%2Fprofessional&l=100&f=rss&t=article'), + (u'Sports-Illini', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=sports%2Fcollege%2Fillini&l=100&f=rss&t=article'), + (u'Sports-Jeff Gordon', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=sports%2Fcolumns%2Fjeff-gordon&l=100&f=rss&t=article'), + (u'Life & Style', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=lifestyles&l=100&f=rss&t=article'), + (u'Life & Style-Debra Bass', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=lifestyles%2Ffashion-and-style%2Fdebra-bass&l=100&f=rss&t=article'), + (u'Life & Style-Food and Cooking', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=lifestyles%2Ffood-and-cooking&l=100&f=rss&t=article'), + (u'Life & Style-Health/Medicine/Fitness', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=lifestyles%2Fhealth-med-fit&l=100&f=rss&t=article'), + (u'Life & Style-Joe Holleman', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=lifestyles%2Fcolumns%2Fjoe-holleman&l=100&f=rss&t=article'), + (u'Life & Style-Steals-and-Deals', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=lifestyles%2Fcolumns%2Fsteals-and-deals&l=100&f=rss&t=article'), + (u'Life & Style-Tim Townsend', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=lifestyles%2Ffaith-and-values%2Ftim-townsend&l=100&f=rss&t=article'), + (u'Entertainment', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=entertainment&l=100&f=rss&t=article'), + (u'Entertainment-Arts & Theatre', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=entertainment%2Farts-and-theatre&l=100&f=rss&t=article'), + (u'Entertainment-Books & Literature', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=entertainment%2Fbooks-and-literature&l=100&f=rss&t=article'), + (u'Entertainment-Dining', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=sports%2Fhockey%2Fprofessional&l=100&f=rss&t=article'), + (u'Entertainment-Events Calendar', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=entertainment%2Fevents-calendar&l=100&f=rss&t=article'), + (u'Entertainment-Gail Pennington', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=entertainment%2Ftelevision%2Fgail-pennington&l=100&f=rss&t=article'), + (u'Entertainment-Hip Hops', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=entertainment%2Fdining%2Fbars-and-clubs-other%2Fhip-hops&l=100&f=rss&t=article'), + (u'Entertainment-House-O-Fun', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=entertainment%2Fhouse-o-fun&l=100&f=rss&t=article'), + (u'Entertainment-Kevin C. Johnson', u'http://www2.stltoday.com/search/?q=&d1=&d2=&s=start_time&sd=desc&c=entertainment%2Fmusic%2Fkevin-johnson&l=100&f=rss&t=article') + ] + remove_empty_feeds = True + remove_tags = [dict(name='div', attrs={'id':'blox-logo'}),dict(name='a')] + keep_only_tags = [dict(name='h1'), dict(name='p', attrs={'class':'byline'}), dict(name="div", attrs={'id':'blox-story-text'})] + extra_css = 'p {text-align: left;}' + + diff --git a/src/calibre/devices/android/driver.py b/src/calibre/devices/android/driver.py index 9c37b6ff59..8b30631528 100644 --- a/src/calibre/devices/android/driver.py +++ b/src/calibre/devices/android/driver.py @@ -63,7 +63,7 @@ class ANDROID(USBMS): WINDOWS_MAIN_MEM = ['ANDROID_PHONE', 'A855', 'A853', 'INC.NEXUS_ONE', '__UMS_COMPOSITE', '_MB200', 'MASS_STORAGE', '_-_CARD', 'SGH-I897', 'GT-I9000', 'FILE-STOR_GADGET', 'SGH-T959', 'SAMSUNG_ANDROID', - 'SCH-I500_CARD', 'SPH-D700_CARD'] + 'SCH-I500_CARD', 'SPH-D700_CARD', 'MB810'] WINDOWS_CARD_A_MEM = ['ANDROID_PHONE', 'GT-I9000_CARD', 'SGH-I897', 'FILE-STOR_GADGET', 'SGH-T959', 'SAMSUNG_ANDROID'] From b49f04c413775e8a92efa31cd87bc734b3451ee6 Mon Sep 17 00:00:00 2001 From: Charles Haley <> Date: Tue, 7 Dec 2010 16:02:39 +0000 Subject: [PATCH 18/22] Fix typo in manual --- src/calibre/manual/template_lang.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/calibre/manual/template_lang.rst b/src/calibre/manual/template_lang.rst index b316850798..b2d32f0767 100644 --- a/src/calibre/manual/template_lang.rst +++ b/src/calibre/manual/template_lang.rst @@ -81,7 +81,7 @@ For trailing zeros, use:: If you use series indices with sub values (e.g., 1.1), you might want to ensure that the decimal points line up. For example, you might want the indices 1 and 2.5 to appear as 01.00 and 02.50 so that they will sort correctly. To do this, use:: - {series_index:0<5.2f} - Five characters, consisting of two digits with leading zeros, a decimal point, then 2 digits after the decimal point + {series_index:0>5.2f} - Five characters, consisting of two digits with leading zeros, a decimal point, then 2 digits after the decimal point If you want only the first two letters of the data, use:: From d34894243661f3351bbf09745d97561f3e91f392 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Tue, 7 Dec 2010 09:10:01 -0700 Subject: [PATCH 19/22] ... --- src/calibre/ebooks/oeb/reader.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/calibre/ebooks/oeb/reader.py b/src/calibre/ebooks/oeb/reader.py index 0f61969373..8e11ac6498 100644 --- a/src/calibre/ebooks/oeb/reader.py +++ b/src/calibre/ebooks/oeb/reader.py @@ -544,7 +544,7 @@ class OEBReader(object): data = render_html_svg_workaround(path, self.logger) if not data: data = '' - id, href = self.oeb.manifest.generate('cover', 'cover.jpeg') + id, href = self.oeb.manifest.generate('cover', 'cover.jpg') item = self.oeb.manifest.add(id, href, JPEG_MIME, data=data) return item From 03c6b10d954e94bce466d768ac121a70bf647461 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Tue, 7 Dec 2010 10:06:30 -0700 Subject: [PATCH 20/22] ... --- src/calibre/utils/ipc/server.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/calibre/utils/ipc/server.py b/src/calibre/utils/ipc/server.py index 380e2e074b..4d35113d80 100644 --- a/src/calibre/utils/ipc/server.py +++ b/src/calibre/utils/ipc/server.py @@ -292,12 +292,12 @@ class Server(Thread): except: pass time.sleep(0.2) - for worker in self.workers: + for worker in list(self.workers): try: worker.kill() except: pass - for worker in self.pool: + for worker in list(self.pool): try: worker.kill() except: From 6e43796c9d041552a0b1e2191ab6d3a34876d2d4 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Tue, 7 Dec 2010 10:52:36 -0700 Subject: [PATCH 21/22] Adding books: Run in the main thread to prevent unfortunate interactions with the metadata backup. Also fix regression that broke the Abort button. --- src/calibre/gui2/add.py | 172 +++++++++++++++++++++++----------------- 1 file changed, 98 insertions(+), 74 deletions(-) diff --git a/src/calibre/gui2/add.py b/src/calibre/gui2/add.py index 5f41f3a8e0..5f555ef138 100644 --- a/src/calibre/gui2/add.py +++ b/src/calibre/gui2/add.py @@ -3,41 +3,55 @@ UI for adding books to the database and saving books to disk ''' import os, shutil, time from Queue import Queue, Empty -from threading import Thread +from functools import partial -from PyQt4.Qt import QThread, SIGNAL, QObject, QTimer, Qt, \ - QProgressDialog +from PyQt4.Qt import QThread, QObject, Qt, QProgressDialog, pyqtSignal, QTimer from calibre.gui2.dialogs.progress import ProgressDialog from calibre.gui2 import question_dialog, error_dialog, info_dialog from calibre.ebooks.metadata.opf2 import OPF from calibre.ebooks.metadata import MetaInformation -from calibre.constants import preferred_encoding, filesystem_encoding +from calibre.constants import preferred_encoding, filesystem_encoding, DEBUG from calibre.utils.config import prefs +from calibre import prints + +single_shot = partial(QTimer.singleShot, 75) + +class DuplicatesAdder(QObject): # {{{ + + added = pyqtSignal(object) + adding_done = pyqtSignal() -class DuplicatesAdder(QThread): # {{{ - # Add duplicate books def __init__(self, parent, db, duplicates, db_adder): - QThread.__init__(self, parent) + QObject.__init__(self, parent) self.db, self.db_adder = db, db_adder - self.duplicates = duplicates + self.duplicates = list(duplicates) + self.count = 0 + single_shot(self.add_one) + + def add_one(self): + if not self.duplicates: + self.adding_done.emit() + return + + mi, cover, formats = self.duplicates.pop() + formats = [f for f in formats if not f.lower().endswith('.opf')] + id = self.db.create_book_entry(mi, cover=cover, + add_duplicates=True) + # here we add all the formats for dupe book record created above + self.db_adder.add_formats(id, formats) + self.db_adder.number_of_books_added += 1 + self.count += 1 + self.added.emit(self.count) + single_shot(self.add_one) - def run(self): - count = 1 - for mi, cover, formats in self.duplicates: - formats = [f for f in formats if not f.lower().endswith('.opf')] - id = self.db.create_book_entry(mi, cover=cover, - add_duplicates=True) - # here we add all the formats for dupe book record created above - self.db_adder.add_formats(id, formats) - self.db_adder.number_of_books_added += 1 - self.emit(SIGNAL('added(PyQt_PyObject)'), count) - count += 1 - self.emit(SIGNAL('adding_done()')) # }}} class RecursiveFind(QThread): # {{{ + update = pyqtSignal(object) + found = pyqtSignal(object) + def __init__(self, parent, db, root, single): QThread.__init__(self, parent) self.db = db @@ -50,8 +64,8 @@ class RecursiveFind(QThread): # {{{ for dirpath in os.walk(root): if self.canceled: return - self.emit(SIGNAL('update(PyQt_PyObject)'), - _('Searching in')+' '+dirpath[0]) + self.update.emit( + _('Searching in')+' '+dirpath[0]) self.books += list(self.db.find_books_in_directory(dirpath[0], self.single_book_per_directory)) @@ -71,46 +85,55 @@ class RecursiveFind(QThread): # {{{ msg = unicode(err) except: msg = repr(err) - self.emit(SIGNAL('found(PyQt_PyObject)'), msg) + self.found.emit(msg) return self.books = [formats for formats in self.books if formats] if not self.canceled: - self.emit(SIGNAL('found(PyQt_PyObject)'), self.books) + self.found.emit(self.books) # }}} -class DBAdder(Thread): # {{{ +class DBAdder(QObject): # {{{ + + def __init__(self, parent, db, ids, nmap): + QObject.__init__(self, parent) - def __init__(self, db, ids, nmap): self.db, self.ids, self.nmap = db, dict(**ids), dict(**nmap) - self.end = False self.critical = {} self.number_of_books_added = 0 self.duplicates = [] self.names, self.paths, self.infos = [], [], [] - Thread.__init__(self) - self.daemon = True self.input_queue = Queue() self.output_queue = Queue() self.merged_books = set([]) - def run(self): - while not self.end: - try: - id, opf, cover = self.input_queue.get(True, 0.2) - except Empty: - continue - name = self.nmap.pop(id) - title = None - try: - title = self.add(id, opf, cover, name) - except: - import traceback - self.critical[name] = traceback.format_exc() - title = name - self.output_queue.put(title) + def end(self): + self.input_queue.put((None, None, None)) + + def start(self): + try: + id, opf, cover = self.input_queue.get_nowait() + except Empty: + single_shot(self.start) + return + if id is None and opf is None and cover is None: + return + name = self.nmap.pop(id) + title = None + if DEBUG: + st = time.time() + try: + title = self.add(id, opf, cover, name) + except: + import traceback + self.critical[name] = traceback.format_exc() + title = name + self.output_queue.put(title) + if DEBUG: + prints('Added', title, 'to db in:', time.time() - st, 'seconds') + single_shot(self.start) def process_formats(self, opf, formats): imp = opf[:-4]+'.import' @@ -201,10 +224,10 @@ class Adder(QObject): # {{{ self.pd.setModal(True) self.pd.show() self._parent = parent - self.rfind = self.worker = self.timer = None + self.rfind = self.worker = None self.callback = callback self.callback_called = False - self.connect(self.pd, SIGNAL('canceled()'), self.canceled) + self.pd.canceled_signal.connect(self.canceled) def add_recursive(self, root, single=True): self.path = root @@ -213,10 +236,8 @@ class Adder(QObject): # {{{ self.pd.set_max(0) self.pd.value = 0 self.rfind = RecursiveFind(self, self.db, root, single) - self.connect(self.rfind, SIGNAL('update(PyQt_PyObject)'), - self.pd.set_msg, Qt.QueuedConnection) - self.connect(self.rfind, SIGNAL('found(PyQt_PyObject)'), - self.add, Qt.QueuedConnection) + self.rfind.update.connect(self.pd.set_msg, type=Qt.QueuedConnection) + self.rfind.found.connect(self.add, type=Qt.QueuedConnection) self.rfind.start() def add(self, books): @@ -246,12 +267,12 @@ class Adder(QObject): # {{{ self.pd.set_min(0) self.pd.set_max(len(self.ids)) self.pd.value = 0 - self.db_adder = DBAdder(self.db, self.ids, self.nmap) + self.db_adder = DBAdder(self, self.db, self.ids, self.nmap) self.db_adder.start() self.last_added_at = time.time() self.entry_count = len(self.ids) self.continue_updating = True - QTimer.singleShot(200, self.update) + single_shot(self.update) def canceled(self): self.continue_updating = False @@ -260,14 +281,14 @@ class Adder(QObject): # {{{ if self.worker is not None: self.worker.canceled = True if hasattr(self, 'db_adder'): - self.db_adder.end = True + self.db_adder.end() self.pd.hide() if not self.callback_called: self.callback(self.paths, self.names, self.infos) self.callback_called = True def duplicates_processed(self): - self.db_adder.end = True + self.db_adder.end() if not self.callback_called: self.callback(self.paths, self.names, self.infos) self.callback_called = True @@ -300,7 +321,7 @@ class Adder(QObject): # {{{ if (time.time() - self.last_added_at) > self.ADD_TIMEOUT: self.continue_updating = False self.pd.hide() - self.db_adder.end = True + self.db_adder.end() if not self.callback_called: self.callback([], [], []) self.callback_called = True @@ -311,7 +332,7 @@ class Adder(QObject): # {{{ 'find the problem book.'), show=True) if self.continue_updating: - QTimer.singleShot(200, self.update) + single_shot(self.update) def process_duplicates(self): @@ -332,11 +353,8 @@ class Adder(QObject): # {{{ self.__p_d = pd self.__d_a = DuplicatesAdder(self._parent, self.db, duplicates, self.db_adder) - self.connect(self.__d_a, SIGNAL('added(PyQt_PyObject)'), - pd.setValue) - self.connect(self.__d_a, SIGNAL('adding_done()'), - self.duplicates_processed) - self.__d_a.start() + self.__d_a.added.connect(pd.setValue) + self.__d_a.adding_done.connect(self.duplicates_processed) else: return self.duplicates_processed() @@ -407,14 +425,12 @@ class Saver(QObject): # {{{ self.worker = SaveWorker(self.rq, db, self.ids, path, self.opts, spare_server=self.spare_server) self.pd.canceled_signal.connect(self.canceled) - self.timer = QTimer(self) - self.connect(self.timer, SIGNAL('timeout()'), self.update) - self.timer.start(200) + self.continue_updating = True + single_shot(self.update) def canceled(self): - if self.timer is not None: - self.timer.stop() + self.continue_updating = False if self.worker is not None: self.worker.canceled = True self.pd.hide() @@ -424,27 +440,35 @@ class Saver(QObject): # {{{ def update(self): - if not self.ids or not self.worker.is_alive(): - self.timer.stop() - self.pd.hide() + if not self.continue_updating: + return + if not self.worker.is_alive(): + # Check that all ids were processed while self.ids: + # Get all queued results since worker is dead before = len(self.ids) self.get_result() if before == len(self.ids): + # No results available => worker died unexpectedly for i in list(self.ids): self.failures.add(('id:%d'%i, 'Unknown error')) self.ids.remove(i) - break + + if not self.ids: + self.continue_updating = False + self.pd.hide() if not self.callback_called: try: - self.worker.join(1.5) + # Give the worker time to clean up and set worker.error + self.worker.join(2) except: pass # The worker was not yet started - self.callback(self.worker.path, self.failures, self.worker.error) self.callback_called = True - return + self.callback(self.worker.path, self.failures, self.worker.error) - self.get_result() + if self.continue_updating: + self.get_result() + single_shot(self.update) def get_result(self): From e20d092ba01b38971990f6c9646b1d72458bcad6 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Tue, 7 Dec 2010 11:09:06 -0700 Subject: [PATCH 22/22] Adding books: Dont refresh the Tag Browser while adding multiple books. Should speed up the adding of large numbers of books. --- src/calibre/gui2/actions/add.py | 4 ++++ src/calibre/gui2/tag_view.py | 3 +++ 2 files changed, 7 insertions(+) diff --git a/src/calibre/gui2/actions/add.py b/src/calibre/gui2/actions/add.py index 9b348d8285..014fa573d2 100644 --- a/src/calibre/gui2/actions/add.py +++ b/src/calibre/gui2/actions/add.py @@ -61,6 +61,7 @@ class AddAction(InterfaceAction): self._adder = Adder(self.gui, self.gui.library_view.model().db, self.Dispatcher(self._files_added), spare_server=self.gui.spare_server) + self.gui.tags_view.disable_recounting = True self._adder.add_recursive(root, single) def add_recursive_single(self, *args): @@ -201,9 +202,11 @@ class AddAction(InterfaceAction): self._adder = Adder(self.gui, None if to_device else self.gui.library_view.model().db, self.Dispatcher(self.__adder_func), spare_server=self.gui.spare_server) + self.gui.tags_view.disable_recounting = True self._adder.add(paths) def _files_added(self, paths=[], names=[], infos=[], on_card=None): + self.gui.tags_view.disable_recounting = False if paths: self.gui.upload_books(paths, list(map(ascii_filename, names)), @@ -214,6 +217,7 @@ class AddAction(InterfaceAction): self.gui.library_view.model().books_added(self._adder.number_of_books_added) if hasattr(self.gui, 'db_images'): self.gui.db_images.reset() + self.gui.tags_view.recount() if getattr(self._adder, 'merged_books', False): books = u'\n'.join([x if isinstance(x, unicode) else x.decode(preferred_encoding, 'replace') for x in diff --git a/src/calibre/gui2/tag_view.py b/src/calibre/gui2/tag_view.py index fdae1bdbc9..2ede698c85 100644 --- a/src/calibre/gui2/tag_view.py +++ b/src/calibre/gui2/tag_view.py @@ -73,6 +73,7 @@ class TagsView(QTreeView): # {{{ def __init__(self, parent=None): QTreeView.__init__(self, parent=None) self.tag_match = None + self.disable_recounting = False self.setUniformRowHeights(True) self.setCursor(Qt.PointingHandCursor) self.setIconSize(QSize(30, 30)) @@ -299,6 +300,8 @@ class TagsView(QTreeView): # {{{ return self.isExpanded(idx) def recount(self, *args): + if self.disable_recounting: + return self.refresh_signal_processed = True ci = self.currentIndex() if not ci.isValid():