def build_bookstrap_themes(): themelist = [] if not os.path.isdir(os.path.join(PROG_DIR, 'data', 'interfaces', 'bookstrap')): return themelist # return empty if bookstrap interface not installed if not internet(): logger.warn('Build Bookstrap Themes: No internet connection') return themelist URL = 'http://bootswatch.com/api/3.json' result, success = fetchURL(URL, None, False) # use default headers, no retry if not success: logger.debug("Error getting bookstrap themes : %s" % result) return themelist try: results = json.loads(result) for theme in results['themes']: themelist.append(theme['name'].lower()) except Exception as e: # error reading results logger.debug('JSON Error reading bookstrap themes, %s' % str(e)) logger.debug("Bookstrap found %i themes" % len(themelist)) return themelist
def checkForUpdates(): """ Called from webserver with thread name WEBSERVER, or as a cron job """ if 'Thread-' in threading.currentThread().name: threading.currentThread().name = "CRON-VERSIONCHECK" logger.debug( 'Set Install Type, Current & Latest Version and Commit status') if internet(): getInstallType() lazylibrarian.CONFIG['CURRENT_VERSION'] = getCurrentVersion() lazylibrarian.CONFIG['LATEST_VERSION'] = getLatestVersion() lazylibrarian.CONFIG[ 'COMMITS_BEHIND'], lazylibrarian.COMMIT_LIST = getCommitDifferenceFromGit( ) logger.debug('Update check complete') else: logger.warn('No internet connection')
def search_rss_book(books=None, reset=False): try: threadname = threading.currentThread().name if "Thread-" in threadname: if books is None: threading.currentThread().name = "SEARCHALLRSS" else: threading.currentThread().name = "SEARCHRSS" if not (lazylibrarian.USE_RSS()): logger.warn('RSS search is disabled') scheduleJob(action='Stop', target='search_rss_book') return if not internet(): logger.warn('Search RSS Book: No internet connection') return myDB = database.DBConnection() resultlist, wishproviders = IterateOverGoodReads() if not wishproviders: logger.debug('No rss wishlists are set') else: # for each item in resultlist, add to database if necessary, and mark as wanted for book in resultlist: # we get rss_author, rss_title, rss_isbn, rss_bookid (goodreads bookid) # we can just use bookid if goodreads, or try isbn and name matching on author/title if googlebooks # not sure if anyone would use a goodreads wishlist if not using goodreads interface... logger.debug('Processing %s item%s in wishlists' % (len(resultlist), plural(len(resultlist)))) if book['rss_bookid'] and lazylibrarian.CONFIG[ 'BOOK_API'] == "GoodReads": bookmatch = myDB.match( 'select Status,BookName from books where bookid="%s"' % book['rss_bookid']) if bookmatch: bookstatus = bookmatch['Status'] bookname = bookmatch['BookName'] if bookstatus in ['Open', 'Wanted', 'Have']: logger.info( u'Found book %s, already marked as "%s"' % (bookname, bookstatus)) else: # skipped/ignored logger.info(u'Found book %s, marking as "Wanted"' % bookname) controlValueDict = {"BookID": bookid} newValueDict = {"Status": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) else: import_book(book['rss_bookid']) else: item = {} headers = [] item['Title'] = book['rss_title'] if book['rss_bookid']: item['BookID'] = book['rss_bookid'] headers.append('BookID') if book['rss_isbn']: item['ISBN'] = book['rss_isbn'] headers.append('ISBN') bookmatch = finditem(item, book['rss_author'], headers) if bookmatch: # it's already in the database authorname = bookmatch['AuthorName'] bookname = bookmatch['BookName'] bookid = bookmatch['BookID'] bookstatus = bookmatch['Status'] if bookstatus in ['Open', 'Wanted', 'Have']: logger.info( u'Found book %s by %s, already marked as "%s"' % (bookname, authorname, bookstatus)) else: # skipped/ignored logger.info( u'Found book %s by %s, marking as "Wanted"' % (bookname, authorname)) controlValueDict = {"BookID": bookid} newValueDict = {"Status": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) else: # not in database yet results = '' if book['rss_isbn']: results = search_for(book['rss_isbn']) if results: result = results[0] if result['isbn_fuzz'] > lazylibrarian.CONFIG[ 'MATCH_RATIO']: logger.info( "Found (%s%%) %s: %s" % (result['isbn_fuzz'], result['authorname'], result['bookname'])) import_book(result['bookid']) bookmatch = True if not results: searchterm = "%s <ll> %s" % ( item['Title'], formatAuthorName(book['rss_author'])) results = search_for(unaccented(searchterm)) if results: result = results[0] if result['author_fuzz'] > lazylibrarian.CONFIG['MATCH_RATIO'] \ and result['book_fuzz'] > lazylibrarian.CONFIG['MATCH_RATIO']: logger.info( "Found (%s%% %s%%) %s: %s" % (result['author_fuzz'], result['book_fuzz'], result['authorname'], result['bookname'])) import_book(result['bookid']) bookmatch = True if not bookmatch: msg = "Skipping book %s by %s" % (item['Title'], book['rss_author']) # noinspection PyUnboundLocalVariable if not results: msg += ', No results returned' logger.warn(msg) else: msg += ', No match found' logger.warn(msg) msg = "Closest match (%s%% %s%%) %s: %s" % ( result['author_fuzz'], result['book_fuzz'], result['authorname'], result['bookname']) logger.warn(msg) if books is None: # We are performing a backlog search cmd = 'SELECT BookID, AuthorName, Bookname, BookSub, BookAdded from books,authors ' cmd += 'WHERE books.AuthorID = authors.AuthorID and books.Status="Wanted" order by BookAdded desc' searchbooks = myDB.select(cmd) else: # The user has added a new book searchbooks = [] for book in books: cmd = 'SELECT BookID, AuthorName, BookName, BookSub from books,authors ' cmd += 'WHERE books.AuthorID = authors.AuthorID and BookID="%s" ' % book[ 'bookid'] cmd += 'AND books.Status="Wanted"' searchbook = myDB.select(cmd) for terms in searchbook: searchbooks.append(terms) if len(searchbooks) == 0: return resultlist, nproviders = IterateOverRSSSites() if not nproviders: if not wishproviders: logger.warn('No rss providers are set, check config') return # No point in continuing logger.info('RSS Searching for %i book%s' % (len(searchbooks), plural(len(searchbooks)))) rss_count = 0 for book in searchbooks: authorname, bookname = get_searchterm(book, "book") found = processResultList(resultlist, authorname, bookname, book, 'book') # if you can't find the book, try title without any "(extended details, series etc)" if not found and '(' in bookname: # anything to shorten? authorname, bookname = get_searchterm(book, "shortbook") found = processResultList(resultlist, authorname, bookname, book, 'shortbook') if not found: logger.debug( "Searches returned no results. Adding book %s - %s to queue." % (authorname, bookname)) if found > True: rss_count += 1 logger.info("RSS Search for Wanted items complete, found %s book%s" % (rss_count, plural(rss_count))) if reset: scheduleJob(action='Restart', target='search_rss_book') except Exception: logger.error('Unhandled exception in search_rss_book: %s' % traceback.format_exc())
def searchItem(item=None, bookid=None): """ Call all active search providers asking for a "general" search for item return a list of results, each entry in list containing percentage_match, title, provider, size, url """ results = [] if not item: return results if not internet(): logger.debug('Search Item: No internet connection') return results book = {} searchterm = unaccented_str(item) book['searchterm'] = searchterm if bookid: book['bookid'] = bookid else: book['bookid'] = searchterm nproviders = lazylibrarian.USE_NZB() + lazylibrarian.USE_TOR( ) + lazylibrarian.USE_RSS() logger.debug('Searching %s providers for %s' % (nproviders, searchterm)) if lazylibrarian.USE_NZB(): resultlist, nproviders = IterateOverNewzNabSites(book, 'general') if nproviders: results += resultlist if lazylibrarian.USE_TOR(): resultlist, nproviders = IterateOverTorrentSites(book, 'general') if nproviders: results += resultlist if lazylibrarian.USE_RSS(): resultlist, nproviders = IterateOverRSSSites() if nproviders: results += resultlist # reprocess to get consistent results searchresults = [] for item in results: provider = '' title = '' url = '' size = '' date = '' mode = '' if 'nzbtitle' in item: title = item['nzbtitle'] if 'nzburl' in item: url = item['nzburl'] if 'nzbprov' in item: provider = item['nzbprov'] if 'nzbsize' in item: size = item['nzbsize'] if 'nzbdate' in item: date = item['nzbdate'] if 'nzbmode' in item: mode = item['nzbmode'] if 'tor_title' in item: title = item['tor_title'] if 'tor_url' in item: url = item['tor_url'] if 'tor_prov' in item: provider = item['tor_prov'] if 'tor_size' in item: size = item['tor_size'] if 'tor_date' in item: date = item['tor_date'] if 'tor_type' in item: mode = item['tor_type'] if title and provider and mode and url: # Not all results have a date or a size if not date: date = 'Fri, 01 Jan 1970 00:00:00 +0100' if not size: size = '1000' # calculate match percentage score = fuzz.token_set_ratio(searchterm, title) # lose a point for each extra word in the title so we get the closest match words = len(getList(searchterm)) words -= len(getList(title)) score -= abs(words) if score >= 40: # ignore wildly wrong results? url = url.split('?')[0] result = { 'score': score, 'title': title, 'provider': provider, 'size': size, 'date': date, 'url': urllib.quote_plus(url), 'mode': mode } searchresults.append(result) # from operator import itemgetter # searchresults = sorted(searchresults, key=itemgetter('score'), reverse=True) logger.debug('Found %s results for %s' % (len(searchresults), searchterm)) return searchresults
def search_magazines(mags=None, reset=False): # produce a list of magazines to search for, tor, nzb, torznab, rss try: threadname = threading.currentThread().name if "Thread-" in threadname: if mags is None: threading.currentThread().name = "SEARCHALLMAG" else: threading.currentThread().name = "SEARCHMAG" if not internet(): logger.warn('Search Magazines: No internet connection') return myDB = database.DBConnection() searchlist = [] if mags is None: # backlog search searchmags = myDB.select('SELECT Title, Regex, LastAcquired, \ IssueDate from magazines WHERE Status="Active"' ) else: searchmags = [] for magazine in mags: searchmags_temp = myDB.select( 'SELECT Title, Regex, LastAcquired, IssueDate from magazines \ WHERE Title="%s" AND Status="Active"' % (magazine['bookid'])) for terms in searchmags_temp: searchmags.append(terms) if len(searchmags) == 0: return # should clear old search results as might not be available any more # ie torrent not available, changed providers, out of news server retention etc. # Only delete the "skipped" ones, not wanted/snatched/processed/ignored logger.debug(u"Removing old magazine search results") myDB.action('DELETE from pastissues WHERE Status="Skipped"') logger.info('Searching for %i magazine%s' % (len(searchmags), plural(len(searchmags)))) for searchmag in searchmags: bookid = searchmag['Title'] searchterm = searchmag['Regex'] if not searchterm: searchterm = searchmag['Title'] dic = { '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '' } searchterm = unaccented_str(replace_all(searchterm, dic)) searchterm = re.sub('[.\-/]', ' ', searchterm).encode( lazylibrarian.SYS_ENCODING) searchlist.append({"bookid": bookid, "searchterm": searchterm}) if not searchlist: logger.warn( 'There is nothing to search for. Mark some magazines as active.' ) for book in searchlist: resultlist = [] if lazylibrarian.USE_NZB(): resultlist, nproviders = IterateOverNewzNabSites(book, 'mag') if not nproviders: logger.warn( 'No nzb providers are set. Check config for NEWZNAB or TORZNAB providers' ) if lazylibrarian.USE_TOR(): tor_resultlist, nproviders = IterateOverTorrentSites( book, 'mag') if not nproviders: logger.warn( 'No torrent providers are set. Check config for TORRENT providers' ) if tor_resultlist: for item in tor_resultlist: # reformat the torrent results so they look like nzbs resultlist.append({ 'bookid': item['bookid'], 'nzbprov': item['tor_prov'], 'nzbtitle': item['tor_title'], 'nzburl': item['tor_url'], 'nzbdate': 'Fri, 01 Jan 1970 00:00:00 +0100', # fake date as none returned from torrents 'nzbsize': item['tor_size'], 'nzbmode': 'torrent' }) if lazylibrarian.USE_RSS(): rss_resultlist, nproviders = IterateOverRSSSites() if not nproviders: logger.warn( 'No rss providers are set. Check config for RSS providers' ) if rss_resultlist: for item in rss_resultlist: # reformat the rss results so they look like nzbs resultlist.append({ 'bookid': book['bookid'], 'nzbprov': item['tor_prov'], 'nzbtitle': item['tor_title'], 'nzburl': item['tor_url'], 'nzbdate': item[ 'tor_date'], # may be fake date as none returned from rss torrents, only rss nzb 'nzbsize': item['tor_size'], 'nzbmode': item['tor_type'] }) if not resultlist: logger.debug("Adding magazine %s to queue." % book['searchterm']) else: bad_name = 0 bad_date = 0 old_date = 0 rejects = 0 total_nzbs = 0 new_date = 0 maglist = [] issues = [] bookid = '' for nzb in resultlist: total_nzbs += 1 bookid = nzb['bookid'] nzbtitle = unaccented_str(nzb['nzbtitle']) nzbtitle = nzbtitle.replace('"', '').replace( "'", "") # suppress " in titles nzburl = nzb['nzburl'] nzbprov = nzb['nzbprov'] nzbdate_temp = nzb['nzbdate'] nzbsize_temp = nzb['nzbsize'] nzbsize_temp = check_int( nzbsize_temp, 1000 ) # not all torrents returned by torznab have a size nzbsize = round(float(nzbsize_temp) / 1048576, 2) nzbdate = nzbdate2format(nzbdate_temp) nzbmode = nzb['nzbmode'] results = myDB.match( 'SELECT * from magazines WHERE Title="%s"' % bookid) if not results: logger.debug( 'Magazine [%s] does not match search term [%s].' % (nzbtitle, bookid)) bad_name += 1 else: rejected = False maxsize = check_int( lazylibrarian.CONFIG['REJECT_MAGSIZE'], 0) if maxsize and nzbsize > maxsize: logger.debug("Rejecting %s, too large" % nzbtitle) rejected = True if not rejected: minsize = check_int( lazylibrarian.CONFIG['REJECT_MAGMIN'], 0) if minsize and nzbsize < minsize: logger.debug("Rejecting %s, too small" % nzbtitle) rejected = True if not rejected: control_date = results['IssueDate'] dic = { '.': ' ', '-': ' ', '/': ' ', '+': ' ', '_': ' ', '(': '', ')': '' } nzbtitle_formatted = replace_all(nzbtitle, dic).strip() # Need to make sure that substrings of magazine titles don't get found # (e.g. Maxim USA will find Maximum PC USA) - token_set_ratio takes care of this # remove extra spaces if they're in a row if nzbtitle_formatted[ 0] == '[' and nzbtitle_formatted[-1] == ']': nzbtitle_formatted = nzbtitle_formatted[1:-1] nzbtitle_exploded_temp = " ".join( nzbtitle_formatted.split()) nzbtitle_exploded = nzbtitle_exploded_temp.split( ' ') if ' ' in bookid: bookid_exploded = bookid.split(' ') else: bookid_exploded = [bookid] # check nzb has magazine title and a date/issue nr # eg The MagPI July 2015 if len(nzbtitle_exploded) > len(bookid_exploded): # needs to be longer as it has to include a date # check (nearly) all the words in the mag title are in the nzbtitle - allow some fuzz mag_title_match = fuzz.token_set_ratio( unaccented(bookid), unaccented(nzbtitle_formatted)) if mag_title_match < lazylibrarian.CONFIG[ 'MATCH_RATIO']: logger.debug( u"Magazine token set Match failed: " + str(mag_title_match) + "% for " + nzbtitle_formatted) rejected = True else: logger.debug(u"Magazine matched: " + str(mag_title_match) + "% " + bookid + " for " + nzbtitle_formatted) else: logger.debug("Magazine name too short (%s)" % len(nzbtitle_exploded)) rejected = True if not rejected: already_failed = myDB.match( 'SELECT * from wanted WHERE NZBurl="%s" and Status="Failed"' % nzburl) if already_failed: logger.debug( "Rejecting %s, blacklisted at %s" % (nzbtitle_formatted, already_failed['NZBprov'])) rejected = True if not rejected: reject_list = getList(results['Reject']) lower_title = unaccented( nzbtitle_formatted).lower() lower_bookid = unaccented(bookid).lower() for word in reject_list: if word in lower_title and word not in lower_bookid: rejected = True logger.debug("Rejecting %s, contains %s" % (nzbtitle_formatted, word)) break regex_pass = 0 if not rejected: # Magazine names have many different styles of date # DD MonthName YYYY OR MonthName YYYY or Issue nn, MonthName YYYY # MonthName DD YYYY or MonthName DD, YYYY # YYYY MM or YYYY MM DD # Issue/No/Nr/Vol nn, YYYY or Issue/No/Nr/Vol nn # nn YYYY issue number without "Nr" before it # issue and year as a single 6 digit string eg 222015 newdatish = "none" # DD MonthName YYYY OR MonthName YYYY or Issue nn, MonthName YYYY pos = 0 while pos < len(nzbtitle_exploded): year = check_year(nzbtitle_exploded[pos]) if year and pos: month = month2num(nzbtitle_exploded[pos - 1]) if month: if pos - 1: day = check_int( nzbtitle_exploded[pos - 2], 1) if day > 31: # probably issue number nn day = 1 else: day = 1 newdatish = "%04d-%02d-%02d" % ( year, month, day) try: _ = datetime.date(year, month, day) regex_pass = 1 break except ValueError: regex_pass = 0 pos += 1 # MonthName DD YYYY or MonthName DD, YYYY if not regex_pass: pos = 0 while pos < len(nzbtitle_exploded): year = check_year(nzbtitle_exploded[pos]) if year and (pos - 1): month = month2num( nzbtitle_exploded[pos - 2]) if month: day = check_int( nzbtitle_exploded[ pos - 1].rstrip(','), 1) try: _ = datetime.date( year, month, day) newdatish = "%04d-%02d-%02d" % ( year, month, day) regex_pass = 2 break except ValueError: regex_pass = 0 pos += 1 # YYYY MM or YYYY MM DD if not regex_pass: pos = 0 while pos < len(nzbtitle_exploded): year = check_year(nzbtitle_exploded[pos]) if year and pos + 1 < len( nzbtitle_exploded): month = check_int( nzbtitle_exploded[pos + 1], 0) if month: if pos + 2 < len( nzbtitle_exploded): day = check_int( nzbtitle_exploded[pos + 2], 1) else: day = 1 try: _ = datetime.date( year, month, day) newdatish = "%04d-%02d-%02d" % ( year, month, day) regex_pass = 3 break except ValueError: regex_pass = 0 pos += 1 # Issue/No/Nr/Vol nn, YYYY or Issue/No/Nr/Vol nn if not regex_pass: pos = 0 while pos < len(nzbtitle_exploded): if nzbtitle_exploded[pos].lower() in [ "issue", "no", "nr", "vol" ]: if pos + 1 < len(nzbtitle_exploded): issue = check_int( nzbtitle_exploded[pos + 1], 0) if issue: newdatish = str( issue) # 4 == 04 == 004 if pos + 2 < len( nzbtitle_exploded): year = check_year( nzbtitle_exploded[pos + 2]) if year and year < int( datetime.date. today().year): newdatish = '0' # it's old regex_pass = 4 # Issue/No/Nr/Vol nn, YYYY else: regex_pass = 5 # Issue/No/Nr/Vol nn break pos += 1 # nn YYYY issue number without "Nr" before it if not regex_pass: pos = 1 while pos < len(nzbtitle_exploded): year = check_year(nzbtitle_exploded[pos]) if year: issue = check_int( nzbtitle_exploded[pos - 1], 0) if issue: newdatish = str( issue) # 4 == 04 == 004 regex_pass = 6 if year < int(datetime.date.today( ).year): newdatish = '0' # it's old break pos += 1 # issue and year as a single 6 digit string eg 222015 if not regex_pass: pos = 0 while pos < len(nzbtitle_exploded): issue = nzbtitle_exploded[pos] if issue.isdigit() and len(issue) == 6: year = int(issue[2:]) issue = int(issue[:2]) newdatish = str( issue) # 4 == 04 == 004 regex_pass = 7 if year < int( datetime.date.today().year): newdatish = '0' # it's old break pos += 1 if not regex_pass: logger.debug( 'Magazine %s not in a recognised date format.' % nzbtitle_formatted) bad_date += 1 # allow issues with good name but bad date to be included # so user can manually select them, incl those with issue numbers newdatish = "1970-01-01" # provide a fake date for bad-date issues regex_pass = 99 if rejected: rejects += 1 else: if lazylibrarian.LOGLEVEL > 3: logger.debug("regex %s [%s] %s" % (regex_pass, nzbtitle_formatted, newdatish)) # wanted issues go into wanted table marked "Wanted" # the rest into pastissues table marked "Skipped" insert_table = "pastissues" insert_status = "Skipped" if control_date is None: # we haven't got any copies of this magazine yet # get a rough time just over a month ago to compare to, in format yyyy-mm-dd # could perhaps calc differently for weekly, biweekly etc # or for magazines with only an issue number, use zero if '-' in str(newdatish): start_time = time.time() start_time -= int( lazylibrarian.CONFIG['MAG_AGE'] ) * 24 * 60 * 60 # number of seconds in days if start_time < 0: # limit of unixtime (1st Jan 1970) start_time = 0 control_date = time.strftime( "%Y-%m-%d", time.localtime(start_time)) logger.debug( 'Magazine date comparing to %s' % control_date) else: control_date = 0 if '-' in str(control_date) and '-' in str( newdatish): # only grab a copy if it's newer than the most recent we have, # or newer than a month ago if we have none comp_date = datecompare( newdatish, control_date) elif '-' not in str( control_date) and '-' not in str( newdatish): # for issue numbers, check if later than last one we have comp_date = int(newdatish) - int(control_date) newdatish = "%s" % newdatish newdatish = newdatish.zfill( 4) # pad so we sort correctly else: # invalid comparison of date and issue number logger.debug( 'Magazine %s incorrect date or issue format.' % nzbtitle_formatted) bad_date += 1 newdatish = "1970-01-01" # this is our fake date for ones we can't decipher comp_date = 0 if comp_date > 0: # keep track of what we're going to download so we don't download dupes new_date += 1 issue = bookid + ',' + newdatish if issue not in issues: maglist.append({ 'bookid': bookid, 'nzbprov': nzbprov, 'nzbtitle': nzbtitle, 'nzburl': nzburl, 'nzbmode': nzbmode }) logger.debug( 'This issue of %s is new, downloading' % nzbtitle_formatted) issues.append(issue) insert_table = "wanted" insert_status = "Wanted" nzbdate = now() # when we asked for it else: logger.debug( 'This issue of %s is already flagged for download' % issue) else: if newdatish != "1970-01-01": # this is our fake date for ones we can't decipher logger.debug( 'This issue of %s is old; skipping.' % nzbtitle_formatted) old_date += 1 # store only the _new_ matching results # Don't add a new entry if this issue has been found on an earlier search # and status has been user-set ( we only delete the "Skipped" ones ) # In "wanted" table it might be already snatched/downloading/processing mag_entry = myDB.match( 'SELECT * from %s WHERE NZBtitle="%s" and NZBprov="%s"' % (insert_table, nzbtitle, nzbprov)) if not mag_entry: controlValueDict = { "NZBtitle": nzbtitle, "NZBprov": nzbprov } newValueDict = { "NZBurl": nzburl, "BookID": bookid, "NZBdate": nzbdate, "AuxInfo": newdatish, "Status": insert_status, "NZBsize": nzbsize, "NZBmode": nzbmode } myDB.upsert(insert_table, newValueDict, controlValueDict) logger.info( 'Found %i result%s for %s. %i new, %i old, %i fail date, %i fail name, %i rejected: %i to download' % (total_nzbs, plural(total_nzbs), bookid, new_date, old_date, bad_date, bad_name, rejects, len(maglist))) for magazine in maglist: if magazine['nzbmode'] in ["torznab", "torrent", "magnet"]: snatch = TORDownloadMethod(magazine['bookid'], magazine['nzbtitle'], magazine['nzburl']) else: snatch = NZBDownloadMethod(magazine['bookid'], magazine['nzbtitle'], magazine['nzburl']) if snatch: logger.info( 'Downloading %s from %s' % (magazine['nzbtitle'], magazine["nzbprov"])) notify_snatch("%s from %s at %s" % (unaccented( magazine['nzbtitle']), magazine["nzbprov"], now())) custom_notify_snatch(magazine['bookid']) scheduleJob(action='Start', target='processDir') if reset: scheduleJob(action='Restart', target='search_magazines') logger.info("Search for magazines complete") except Exception: logger.error('Unhandled exception in search_magazines: %s' % traceback.format_exc())
def search_tor_book(books=None, reset=False): try: threadname = threading.currentThread().name if "Thread-" in threadname: if books is None: threading.currentThread().name = "SEARCHALLTOR" else: threading.currentThread().name = "SEARCHTOR" if not lazylibrarian.USE_TOR(): logger.warn('No Torrent providers set, check config') return if not internet(): logger.warn('Search Tor Book: No internet connection') return myDB = database.DBConnection() searchlist = [] if books is None: # We are performing a backlog search cmd = 'SELECT BookID, AuthorName, Bookname, BookSub, BookAdded from books,authors ' cmd += 'WHERE books.AuthorID = authors.AuthorID and books.Status="Wanted" order by BookAdded desc' searchbooks = myDB.select(cmd) else: # The user has added a new book searchbooks = [] for book in books: cmd = 'SELECT BookID, AuthorName, BookName, BookSub from books,authors WHERE books.Status="Wanted"' cmd += ' and books.AuthorID = authors.AuthorID and BookID="%s"' % book['bookid'] searchbook = myDB.select(cmd) for terms in searchbook: searchbooks.append(terms) if len(searchbooks) == 0: return logger.info('TOR Searching for %i book%s' % (len(searchbooks), plural(len(searchbooks)))) for searchbook in searchbooks: # searchterm is only used for display purposes searchterm = searchbook['AuthorName'] + ' ' + searchbook['BookName'] if searchbook['BookSub']: searchterm = searchterm + ': ' + searchbook['BookSub'] searchlist.append( {"bookid": searchbook['BookID'], "bookName": searchbook['BookName'], "bookSub": searchbook['BookSub'], "authorName": searchbook['AuthorName'], "searchterm": searchterm}) tor_count = 0 for book in searchlist: resultlist, nproviders = IterateOverTorrentSites(book, 'book') if not nproviders: logger.warn('No torrent providers are set, check config') return # No point in continuing found = processResultList(resultlist, book, "book") # if you can't find the book, try author/title without any "(extended details, series etc)" if not found and '(' in book['bookName']: resultlist, nproviders = IterateOverTorrentSites(book, 'shortbook') found = processResultList(resultlist, book, "shortbook") # general search is the same as booksearch for torrents # if not found: # resultlist, nproviders = IterateOverTorrentSites(book, 'general') # found = processResultList(resultlist, book, "general") if not found: logger.debug("Searches for %s returned no results." % book['searchterm']) if found > True: tor_count += 1 logger.info("TORSearch for Wanted items complete, found %s book%s" % (tor_count, plural(tor_count))) if reset: scheduleJob(action='Restart', target='search_tor_book') except Exception: logger.error('Unhandled exception in search_tor_book: %s' % traceback.format_exc())
def search_nzb_book(books=None, reset=False): try: threadname = threading.currentThread().name if "Thread-" in threadname: if books is None: threading.currentThread().name = "SEARCHALLNZB" else: threading.currentThread().name = "SEARCHNZB" if not lazylibrarian.USE_NZB(): logger.warn('No NEWZNAB/TORZNAB providers set, check config') return if not internet(): logger.warn('Search NZB Book: No internet connection') return myDB = database.DBConnection() searchlist = [] if books is None: # We are performing a backlog search cmd = 'SELECT BookID, AuthorName, Bookname, BookSub, BookAdded from books,authors ' cmd += 'WHERE books.Status="Wanted" and books.AuthorID = authors.AuthorID order by BookAdded desc' searchbooks = myDB.select(cmd) else: # The user has added a new book searchbooks = [] for book in books: cmd = 'SELECT BookID, AuthorName, BookName, BookSub from books,authors' cmd += ' WHERE BookID="%s"' % book['bookid'] cmd += ' AND books.AuthorID = authors.AuthorID AND books.Status="Wanted"' searchbook = myDB.select(cmd) for terms in searchbook: searchbooks.append(terms) if len(searchbooks) == 0: return logger.info('NZB Searching for %i book%s' % (len(searchbooks), plural(len(searchbooks)))) for searchbook in searchbooks: # searchterm is only used for display purposes searchterm = searchbook['AuthorName'] + ' ' + searchbook['BookName'] if searchbook['BookSub']: searchterm = searchterm + ': ' + searchbook['BookSub'] searchlist.append({ "bookid": searchbook['BookID'], "bookName": searchbook['BookName'], "bookSub": searchbook['BookSub'], "authorName": searchbook['AuthorName'], "searchterm": searchterm }) nzb_count = 0 for book in searchlist: # first attempt, try author/title in category "book" resultlist, nproviders = providers.IterateOverNewzNabSites( book, 'book') if not nproviders: logger.warn( 'No NewzNab or TorzNab providers are set, check config') return # no point in continuing found = processResultList(resultlist, book, "book") # if you can't find the book, try author/title without any "(extended details, series etc)" if not found and '(' in book['bookName']: resultlist, nproviders = providers.IterateOverNewzNabSites( book, 'shortbook') found = processResultList(resultlist, book, "shortbook") # if you can't find the book under "books", you might find under general search if not found: resultlist, nproviders = providers.IterateOverNewzNabSites( book, 'general') found = processResultList(resultlist, book, "general") # if still not found, try general search again without any "(extended details, series etc)" if not found and '(' in book['bookName']: resultlist, nproviders = providers.IterateOverNewzNabSites( book, 'shortgeneral') found = processResultList(resultlist, book, "shortgeneral") if not found: logger.info("NZB Searches for %s returned no results." % book['searchterm']) if found > True: nzb_count += 1 # we found it logger.info("NZBSearch for Wanted items complete, found %s book%s" % (nzb_count, plural(nzb_count))) if reset: scheduleJob(action='Restart', target='search_nzb_book') except Exception: logger.error('Unhandled exception in search_nzb_book: %s' % traceback.format_exc())
def LibraryScan(startdir=None): """ Scan a directory tree adding new books into database Return how many books you added """ try: destdir = lazylibrarian.DIRECTORY('Destination') if not startdir: if not destdir: logger.warn('Cannot find destination directory: %s. Not scanning' % destdir) return 0 startdir = destdir if not os.path.isdir(startdir): logger.warn('Cannot find directory: %s. Not scanning' % startdir) return 0 if not internet(): logger.warn('Libraryscan: No internet connection') return 0 myDB = database.DBConnection() # keep statistics of full library scans if startdir == destdir: myDB.action('DELETE from stats') try: # remove any extra whitespace in authornames authors = myDB.select('SELECT AuthorID,AuthorName FROM authors WHERE AuthorName like "% %"') if authors: logger.info('Removing extra spaces from %s authorname%s' % (len(authors), plural(len(authors)))) for author in authors: authorid = author["AuthorID"] authorname = ' '.join(author['AuthorName'].split()) # Have we got author name both with-and-without extra spaces? If so, merge them duplicate = myDB.match( 'Select AuthorID,AuthorName FROM authors WHERE AuthorName="%s"' % authorname) if duplicate: myDB.action('DELETE from authors where authorname="%s"' % author['AuthorName']) if author['AuthorID'] != duplicate['AuthorID']: myDB.action('UPDATE books set AuthorID="%s" WHERE AuthorID="%s"' % (duplicate['AuthorID'], author['AuthorID'])) else: myDB.action( 'UPDATE authors set AuthorName="%s" WHERE AuthorID="%s"' % (authorname, authorid)) except Exception as e: logger.info('Error: ' + str(e)) logger.info('Scanning ebook directory: %s' % startdir) new_book_count = 0 modified_count = 0 rescan_count = 0 rescan_hits = 0 file_count = 0 author = "" if lazylibrarian.CONFIG['FULL_SCAN']: cmd = 'select AuthorName, BookName, BookFile, BookID from books,authors' cmd += ' where books.AuthorID = authors.AuthorID and books.Status="Open"' if not startdir == destdir: cmd += ' and BookFile like "' + startdir + '%"' books = myDB.select(cmd) status = lazylibrarian.CONFIG['NOTFOUND_STATUS'] logger.info('Missing books will be marked as %s' % status) for book in books: bookID = book['BookID'] bookfile = book['BookFile'] if not (bookfile and os.path.isfile(bookfile)): myDB.action('update books set Status="%s" where BookID="%s"' % (status, bookID)) myDB.action('update books set BookFile="" where BookID="%s"' % bookID) logger.warn('Book %s - %s updated as not found on disk' % (book['AuthorName'], book['BookName'])) # to save repeat-scans of the same directory if it contains multiple formats of the same book, # keep track of which directories we've already looked at processed_subdirectories = [] warned = False # have we warned about no new authors setting matchString = '' for char in lazylibrarian.CONFIG['EBOOK_DEST_FILE']: matchString = matchString + '\\' + char # massage the EBOOK_DEST_FILE config parameter into something we can use # with regular expression matching booktypes = '' count = -1 booktype_list = getList(lazylibrarian.CONFIG['EBOOK_TYPE']) for book_type in booktype_list: count += 1 if count == 0: booktypes = book_type else: booktypes = booktypes + '|' + book_type matchString = matchString.replace("\\$\\A\\u\\t\\h\\o\\r", "(?P<author>.*?)").replace( "\\$\\T\\i\\t\\l\\e", "(?P<book>.*?)") + '\.[' + booktypes + ']' pattern = re.compile(matchString, re.VERBOSE) for r, d, f in os.walk(startdir): for directory in d[:]: # prevent magazine being scanned if directory.startswith("_") or directory.startswith("."): d.remove(directory) for files in f: file_count += 1 if isinstance(r, str): r = r.decode(lazylibrarian.SYS_ENCODING) subdirectory = r.replace(startdir, '') # Added new code to skip if we've done this directory before. # Made this conditional with a switch in config.ini # in case user keeps multiple different books in the same subdirectory if lazylibrarian.CONFIG['IMP_SINGLEBOOK'] and (subdirectory in processed_subdirectories): logger.debug("[%s] already scanned" % subdirectory) else: # If this is a book, try to get author/title/isbn/language # if epub or mobi, read metadata from the book # If metadata.opf exists, use that allowing it to override # embedded metadata. User may have edited metadata.opf # to merge author aliases together # If all else fails, try pattern match for author/title # and look up isbn/lang from LT or GR later match = 0 if is_valid_booktype(files): logger.debug("[%s] Now scanning subdirectory %s" % (startdir, subdirectory)) language = "Unknown" isbn = "" book = "" author = "" gr_id = "" gb_id = "" extn = os.path.splitext(files)[1] # if it's an epub or a mobi we can try to read metadata from it if (extn == ".epub") or (extn == ".mobi"): book_filename = os.path.join(r, files).encode(lazylibrarian.SYS_ENCODING) try: res = get_book_info(book_filename) except Exception as e: logger.debug('get_book_info failed for %s, %s' % (book_filename, str(e))) res = {} # title and creator are the minimum we need if 'title' in res and 'creator' in res: book = res['title'] author = res['creator'] if book and len(book) > 2 and author and len(author) > 2: match = 1 if 'language' in res: language = res['language'] if 'identifier' in res: isbn = res['identifier'] if 'type' in res: extn = res['type'] logger.debug("book meta [%s] [%s] [%s] [%s] [%s]" % (isbn, language, author, book, extn)) if not match: logger.debug("Book meta incomplete in %s" % book_filename) # calibre uses "metadata.opf", LL uses "bookname - authorname.opf" # just look for any .opf file in the current directory since we don't know # LL preferred authorname/bookname at this point. # Allow metadata in file to override book contents as may be users pref metafile = opf_file(r) try: res = get_book_info(metafile) except Exception as e: logger.debug('get_book_info failed for %s, %s' % (metafile, str(e))) res = {} # title and creator are the minimum we need if 'title' in res and 'creator' in res: book = res['title'] author = res['creator'] if book and len(book) > 2 and author and len(author) > 2: match = 1 if 'language' in res: language = res['language'] if 'identifier' in res: isbn = res['identifier'] if 'gr_id' in res: gr_id = res['gr_id'] logger.debug("file meta [%s] [%s] [%s] [%s] [%s]" % (isbn, language, author, book, gr_id)) if not match: logger.debug("File meta incomplete in %s" % metafile) if not match: # no author/book from metadata file, and not embedded either match = pattern.match(files) if match: author = match.group("author") book = match.group("book") if len(book) <= 2 or len(author) <= 2: match = 0 if not match: logger.debug("Pattern match failed [%s]" % files) if match: # flag that we found a book in this subdirectory processed_subdirectories.append(subdirectory) # If we have a valid looking isbn, and language != "Unknown", add it to cache if language != "Unknown" and is_valid_isbn(isbn): logger.debug("Found Language [%s] ISBN [%s]" % (language, isbn)) # we need to add it to language cache if not already # there, is_valid_isbn has checked length is 10 or 13 if len(isbn) == 10: isbnhead = isbn[0:3] else: isbnhead = isbn[3:6] match = myDB.match('SELECT lang FROM languages where isbn = "%s"' % isbnhead) if not match: myDB.action('insert into languages values ("%s", "%s")' % (isbnhead, language)) logger.debug("Cached Lang [%s] ISBN [%s]" % (language, isbnhead)) else: logger.debug("Already cached Lang [%s] ISBN [%s]" % (language, isbnhead)) author, authorid, new = addAuthorNameToDB(author) # get the author name as we know it... if author: # author exists, check if this book by this author is in our database # metadata might have quotes in book name # some books might be stored under a different author name # eg books by multiple authors, books where author is "writing as" # or books we moved to "merge" authors book = book.replace("'", "") # First try and find it under author and bookname # as we may have it under a different bookid or isbn to goodreads/googlebooks # which might have several bookid/isbn for the same book bookid = find_book_in_db(myDB, author, book) if not bookid: # Title or author name might not match or multiple authors # See if the gr_id, gb_id is already in our database if gr_id: bookid = gr_id elif gb_id: bookid = gb_id else: bookid = "" if bookid: match = myDB.match('SELECT BookID FROM books where BookID = "%s"' % bookid) if not match: msg = 'Unable to find book %s by %s in database, trying to add it using ' if bookid == gr_id: msg += "GoodReads ID " + gr_id if bookid == gb_id: msg += "GoogleBooks ID " + gb_id logger.debug(msg % (book, author)) if lazylibrarian.CONFIG['BOOK_API'] == "GoodReads" and gr_id: GR_ID = GoodReads(gr_id) GR_ID.find_book(gr_id, None) elif lazylibrarian.CONFIG['BOOK_API'] == "GoogleBooks" and gb_id: GB_ID = GoogleBooks(gb_id) GB_ID.find_book(gb_id, None) # see if it's there now... match = myDB.match('SELECT BookID from books where BookID="%s"' % bookid) if not match: logger.debug("Unable to add bookid %s to database" % bookid) bookid = "" if not bookid and isbn: # See if the isbn is in our database match = myDB.match('SELECT BookID FROM books where BookIsbn = "%s"' % isbn) if match: bookid = match['BookID'] if not bookid: # get author name from parent directory of this book directory newauthor = os.path.basename(os.path.dirname(r)) # calibre replaces trailing periods with _ eg Smith Jr. -> Smith Jr_ if newauthor.endswith('_'): newauthor = newauthor[:-1] + '.' if author.lower() != newauthor.lower(): logger.debug("Trying authorname [%s]" % newauthor) bookid = find_book_in_db(myDB, newauthor, book) if bookid: logger.warn("%s not found under [%s], found under [%s]" % (book, author, newauthor)) # at this point if we still have no bookid, it looks like we # have author and book title but no database entry for it if not bookid: if lazylibrarian.CONFIG['BOOK_API'] == "GoodReads": # Either goodreads doesn't have the book or it didn't match language prefs # Since we have the book anyway, try and reload it ignoring language prefs rescan_count += 1 base_url = 'http://www.goodreads.com/search.xml?q=' params = {"key": lazylibrarian.CONFIG['GR_API']} if author[1] in '. ': surname = author forename = '' while surname[1] in '. ': forename = forename + surname[0] + '.' surname = surname[2:].strip() if author != forename + ' ' + surname: logger.debug('Stripped authorname [%s] to [%s %s]' % (author, forename, surname)) author = forename + ' ' + surname author = ' '.join(author.split()) # ensure no extra whitespace searchname = author + ' ' + book searchname = cleanName(unaccented(searchname)) searchterm = urllib.quote_plus(searchname.encode(lazylibrarian.SYS_ENCODING)) set_url = base_url + searchterm + '&' + urllib.urlencode(params) try: rootxml, in_cache = get_xml_request(set_url) if not len(rootxml): logger.debug("Error requesting results from GoodReads") else: resultxml = rootxml.getiterator('work') for item in resultxml: booktitle = item.find('./best_book/title').text book_fuzz = fuzz.token_set_ratio(booktitle, book) if book_fuzz >= 98: logger.debug("Rescan found %s : %s" % (booktitle, language)) rescan_hits += 1 bookid = item.find('./best_book/id').text GR_ID = GoodReads(bookid) GR_ID.find_book(bookid, None) if language and language != "Unknown": # set language from book metadata logger.debug("Setting language from metadata %s : %s" % (booktitle, language)) myDB.action('UPDATE books SET BookLang="%s" WHERE BookID="%s"' % (language, bookid)) break if not bookid: logger.warn("GoodReads doesn't know about %s" % book) except Exception as e: logger.error("Error finding rescan results: %s" % str(e)) elif lazylibrarian.CONFIG['BOOK_API'] == "GoogleBooks": # if we get here using googlebooks it's because googlebooks # doesn't have the book. No point in looking for it again. logger.warn("GoogleBooks doesn't know about %s" % book) # see if it's there now... if bookid: cmd = 'SELECT books.Status, BookFile, AuthorName, BookName from books,authors ' cmd += 'where books.AuthorID = authors.AuthorID and BookID="%s"' % bookid check_status = myDB.match(cmd) if not check_status: logger.debug('Unable to find bookid %s in database' % bookid) else: if check_status['Status'] != 'Open': # we found a new book new_book_count += 1 myDB.action( 'UPDATE books set Status="Open" where BookID="%s"' % bookid) # store book location so we can check if it gets removed book_filename = os.path.join(r, files) if not check_status['BookFile']: # no previous location myDB.action('UPDATE books set BookFile="%s" where BookID="%s"' % (book_filename, bookid)) # location may have changed since last scan elif book_filename != check_status['BookFile']: modified_count += 1 logger.warn("Updating book location for %s %s from %s to %s" % (author, book, check_status['BookFile'], book_filename)) logger.debug("%s %s matched %s BookID %s, [%s][%s]" % (author, book, check_status['Status'], bookid, check_status['AuthorName'], check_status['BookName'])) myDB.action('UPDATE books set BookFile="%s" where BookID="%s"' % (book_filename, bookid)) # update cover file to cover.jpg in book folder (if exists) bookdir = os.path.dirname(book_filename) coverimg = os.path.join(bookdir, 'cover.jpg') if os.path.isfile(coverimg): cachedir = lazylibrarian.CACHEDIR cacheimg = os.path.join(cachedir, 'book', bookid + '.jpg') copyfile(coverimg, cacheimg) else: logger.warn( "Failed to match book [%s] by [%s] in database" % (book, author)) else: if not warned and not lazylibrarian.CONFIG['ADD_AUTHOR']: logger.warn("Add authors to database is disabled") warned = True logger.info("%s/%s new/modified book%s found and added to the database" % (new_book_count, modified_count, plural(new_book_count + modified_count))) logger.info("%s file%s processed" % (file_count, plural(file_count))) if startdir == destdir: # On full library scans, check for missing workpages setWorkPages() # and books with unknown language nolang = myDB.match( "select count('BookID') as counter from Books where status='Open' and BookLang='Unknown'") nolang = nolang['counter'] if nolang: logger.warn("Found %s book%s in your library with unknown language" % (nolang, plural(nolang))) # show stats if new books were added stats = myDB.match( "SELECT sum(GR_book_hits), sum(GR_lang_hits), sum(LT_lang_hits), sum(GB_lang_change), \ sum(cache_hits), sum(bad_lang), sum(bad_char), sum(uncached), sum(duplicates) FROM stats") st= {'GR_book_hits': stats['sum(GR_book_hits)'], 'GB_book_hits': stats['sum(GR_book_hits)'], 'GR_lang_hits': stats['sum(GR_lang_hits)'], 'LT_lang_hits': stats['sum(LT_lang_hits)'], 'GB_lang_change': stats['sum(GB_lang_change)'], 'cache_hits': stats['sum(cache_hits)'], 'bad_lang': stats['sum(bad_lang)'], 'bad_char': stats['sum(bad_char)'], 'uncached': stats['sum(uncached)'], 'duplicates': stats['sum(duplicates)']} for item in st.keys(): if st[item] is None: st[item] = 0 if lazylibrarian.CONFIG['BOOK_API'] == "GoogleBooks": logger.debug("GoogleBooks was hit %s time%s for books" % (st['GR_book_hits'], plural(st['GR_book_hits']))) logger.debug("GoogleBooks language was changed %s time%s" % (st['GB_lang_change'], plural(st['GB_lang_change']))) if lazylibrarian.CONFIG['BOOK_API'] == "GoodReads": logger.debug("GoodReads was hit %s time%s for books" % (st['GR_book_hits'], plural(st['GR_book_hits']))) logger.debug("GoodReads was hit %s time%s for languages" % (st['GR_lang_hits'], plural(st['GR_lang_hits']))) logger.debug("LibraryThing was hit %s time%s for languages" % (st['LT_lang_hits'], plural(st['LT_lang_hits']))) logger.debug("Language cache was hit %s time%s" % (st['cache_hits'], plural(st['cache_hits']))) logger.debug("Unwanted language removed %s book%s" % (st['bad_lang'], plural(st['bad_lang']))) logger.debug("Unwanted characters removed %s book%s" % (st['bad_char'], plural(st['bad_char']))) logger.debug("Unable to cache language for %s book%s with missing ISBN" % (st['uncached'], plural(st['uncached']))) logger.debug("Found %s duplicate book%s" % (st['duplicates'], plural(st['duplicates']))) logger.debug("Rescan %s hit%s, %s miss" % (rescan_hits, plural(rescan_hits), rescan_count - rescan_hits)) logger.debug("Cache %s hit%s, %s miss" % (lazylibrarian.CACHE_HIT, plural(lazylibrarian.CACHE_HIT), lazylibrarian.CACHE_MISS)) cachesize = myDB.match("select count('ISBN') as counter from languages") logger.debug("ISBN Language cache holds %s entries" % cachesize['counter']) # Cache any covers and images images = myDB.select('select bookid, bookimg, bookname from books where bookimg like "http%"') if len(images): logger.info("Caching cover%s for %i book%s" % (plural(len(images)), len(images), plural(len(images)))) for item in images: bookid = item['bookid'] bookimg = item['bookimg'] # bookname = item['bookname'] newimg, success = cache_img("book", bookid, bookimg) if success: myDB.action('update books set BookImg="%s" where BookID="%s"' % (newimg, bookid)) images = myDB.select('select AuthorID, AuthorImg, AuthorName from authors where AuthorImg like "http%"') if len(images): logger.info("Caching image%s for %i author%s" % (plural(len(images)), len(images), plural(len(images)))) for item in images: authorid = item['authorid'] authorimg = item['authorimg'] # authorname = item['authorname'] newimg, success = cache_img("author", authorid, authorimg) if success: myDB.action('update authors set AuthorImg="%s" where AuthorID="%s"' % (newimg, authorid)) # On full scan, update bookcounts for all authors, not just new ones - refresh may have located # new books for existing authors especially if switched provider gb/gr or changed wanted languages authors = myDB.select('select AuthorID from authors') else: # On single author/book import, just update bookcount for that author authors = myDB.select('select AuthorID from authors where AuthorName = "%s"' % author.replace('"', '""')) logger.debug('Updating bookcounts for %i author%s' % (len(authors), plural(len(authors)))) for author in authors: update_totals(author['AuthorID']) logger.info('Library scan complete') return new_book_count except Exception: logger.error('Unhandled exception in libraryScan: %s' % traceback.format_exc())