def downloadResult(match, book): """ match: best result from search providers book: book we are downloading return: True if already snatched, False if failed to snatch, >True if we snatched it """ try: myDB = database.DBConnection() resultTitle = match[1] newValueDict = match[2] controlValueDict = match[3] if book['library'] == 'AudioBook': auxinfo = 'AudioBook' else: # elif book['library'] == 'eBook': auxinfo = 'eBook' if auxinfo == 'eBook': snatchedbooks = myDB.match('SELECT BookID from books WHERE BookID=? and Status="Snatched"', (newValueDict["BookID"],)) else: snatchedbooks = myDB.match('SELECT BookID from books WHERE BookID=? and AudioStatus="Snatched"', (newValueDict["BookID"],)) if snatchedbooks: logger.debug('%s %s already marked snatched' % (book['authorName'], book['bookName'])) return True # someone else already found it else: myDB.upsert("wanted", newValueDict, controlValueDict) if 'libgen' in newValueDict["NZBprov"]: # for libgen we use direct download links snatch = DirectDownloadMethod(newValueDict["BookID"], newValueDict["NZBtitle"], controlValueDict["NZBurl"], resultTitle, auxinfo) elif newValueDict['NZBmode'] in ["torznab", "torrent", "magnet"]: snatch = TORDownloadMethod(newValueDict["BookID"], newValueDict["NZBtitle"], controlValueDict["NZBurl"], auxinfo) elif newValueDict['NZBmode'] == 'nzb': snatch = NZBDownloadMethod(newValueDict["BookID"], newValueDict["NZBtitle"], controlValueDict["NZBurl"], auxinfo) else: logger.error('Unhandled NZBmode [%s] for %s' % (newValueDict['NZBmode'], controlValueDict["NZBurl"])) snatch = False if snatch: logger.info('Downloading %s %s from %s' % (auxinfo, newValueDict["NZBtitle"], newValueDict["NZBprov"])) notify_snatch("%s %s from %s at %s" % (auxinfo, newValueDict["NZBtitle"], newValueDict["NZBprov"], now())) custom_notify_snatch(newValueDict["BookID"]) # at this point we could add NZBprov to the blocklist with a short timeout, a second or two? # This would implement a round-robin search system. Blocklist with an incremental counter. # If number of active providers == number blocklisted, so no unblocked providers are left, # either sleep for a while, or unblock the one with the lowest counter. scheduleJob(action='Start', target='processDir') return True + True # we found it return False except Exception: logger.error('Unhandled exception in downloadResult: %s' % traceback.format_exc())
def search_wishlist(): try: threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "SEARCHWISHLIST" myDB = database.DBConnection() resultlist, wishproviders = IterateOverWishLists() new_books = 0 if not wishproviders: logger.debug('No wishlists are set') scheduleJob(action='Stop', target='search_wishlist') return # No point in continuing # for each item in resultlist, add to database if necessary, and mark as wanted logger.debug('Processing %s item%s in wishlists' % (len(resultlist), plural(len(resultlist)))) for book in resultlist: # we get rss_author, rss_title, maybe rss_isbn, rss_bookid (goodreads bookid) # we can just use bookid if goodreads, or try isbn and name matching on author/title if not # eg NYTimes wishlist if 'E' in book['types']: ebook_status = "Wanted" else: ebook_status = "Skipped" if 'A' in book['types']: audio_status = "Wanted" else: audio_status = "Skipped" if lazylibrarian.CONFIG['BOOK_API'] == "GoodReads" and book['rss_bookid']: cmd = 'select books.Status as Status,AudioStatus,authors.Status as AuthorStatus,' cmd += 'AuthorName,BookName,Requester,AudioRequester from books,authors ' cmd += 'where books.AuthorID = authors.AuthorID and bookid=?' bookmatch = myDB.match(cmd, (book['rss_bookid'],)) if bookmatch: cmd = 'SELECT SeriesName,Status from series,member ' cmd += 'where series.SeriesID=member.SeriesID and member.BookID=?' series = myDB.select(cmd, (book['rss_bookid'],)) reject_series = None for ser in series: if ser['Status'] in ['Paused', 'Ignored']: reject_series = {"Name": ser['SeriesName'], "Status": ser['Status']} break bookname = bookmatch['BookName'] if bookmatch['Status'] in ['Open', 'Wanted', 'Have']: logger.info('Found book %s, already marked %s' % (bookname, bookmatch['Status'])) if bookmatch["Requester"]: # Already on a wishlist if book["dispname"] not in bookmatch["Requester"]: newValueDict = {"Requester": bookmatch["Requester"] + book["dispname"] + ' '} controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = {"Requester": book["dispname"] + ' '} controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) elif bookmatch['AuthorStatus'] in ['Paused', 'Ignored']: logger.info('Found book %s, but author is %s' % (bookname, bookmatch['AuthorStatus'])) elif reject_series: logger.info('Found book %s, but series "%s" is %s' % (bookname, reject_series['Name'], reject_series['Status'])) elif ebook_status == "Wanted": # skipped/ignored logger.info('Found book %s, marking as "Wanted"' % bookname) controlValueDict = {"BookID": book['rss_bookid']} newValueDict = {"Status": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) new_books += 1 if bookmatch["Requester"]: # Already on a wishlist if book["dispname"] not in bookmatch["Requester"]: newValueDict = {"Requester": bookmatch["Requester"] + book["dispname"] + ' '} controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = {"Requester": book["dispname"] + ' '} controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) if bookmatch['AudioStatus'] in ['Open', 'Wanted', 'Have']: logger.info('Found audiobook %s, already marked %s' % (bookname, bookmatch['AudioStatus'])) if bookmatch["AudioRequester"]: # Already on a wishlist if book["dispname"] not in bookmatch["AudioRequester"]: newValueDict = {"AudioRequester": bookmatch["AudioRequester"] + book["dispname"] + ' '} controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = {"AudioRequester": book["dispname"] + ' '} controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) elif bookmatch['AuthorStatus'] in ['Paused', 'Ignored']: logger.info('Found book %s, but author is %s' % (bookname, bookmatch['AuthorStatus'])) elif reject_series: logger.info('Found book %s, but series "%s" is %s' % (bookname, reject_series['Name'], reject_series['Status'])) elif audio_status == "Wanted": # skipped/ignored logger.info('Found audiobook %s, marking as "Wanted"' % bookname) controlValueDict = {"BookID": book['rss_bookid']} newValueDict = {"AudioStatus": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) new_books += 1 if bookmatch["AudioRequester"]: # Already on a wishlist if book["dispname"] not in bookmatch["AudioRequester"]: newValueDict = {"AudioRequester": bookmatch["AudioRequester"] + book["dispname"] + ' '} controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = {"AudioRequester": book["dispname"] + ' '} controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) else: import_book(book['rss_bookid'], ebook_status, audio_status) new_books += 1 newValueDict = {"Requester": book["dispname"] + ' '} controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) newValueDict = {"AudioRequester": book["dispname"] + ' '} controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) else: item = {} results = None item['Title'] = book['rss_title'] if book['rss_bookid']: item['BookID'] = book['rss_bookid'] if book['rss_isbn']: item['ISBN'] = book['rss_isbn'] bookmatch = finditem(item, book['rss_author']) if bookmatch: # it's already in the database authorname = bookmatch['AuthorName'] bookname = bookmatch['BookName'] bookid = bookmatch['BookID'] auth_res = myDB.match('SELECT Status from authors WHERE authorname=?', (authorname,)) if auth_res: auth_status = auth_res['Status'] else: auth_status = 'Unknown' cmd = 'SELECT SeriesName,Status from series,member ' cmd += 'where series.SeriesID=member.SeriesID and member.BookID=?' series = myDB.select(cmd, (book['rss_bookid'],)) reject_series = None for ser in series: if ser['Status'] in ['Paused', 'Ignored']: reject_series = {"Name": ser['SeriesName'], "Status": ser['Status']} break if bookmatch['Status'] in ['Open', 'Wanted', 'Have']: logger.info( 'Found book %s by %s, already marked as "%s"' % (bookname, authorname, bookmatch['Status'])) if bookmatch["Requester"]: # Already on a wishlist if book["dispname"] not in bookmatch["Requester"]: newValueDict = {"Requester": bookmatch["Requester"] + book["dispname"] + ' '} controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = {"Requester": book["dispname"] + ' '} controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) elif auth_status in ['Paused', 'Ignored']: logger.info('Found book %s, but author is "%s"' % (bookname, auth_status)) elif reject_series: logger.info('Found book %s, but series "%s" is %s' % (bookname, reject_series['Name'], reject_series['Status'])) elif ebook_status == 'Wanted': # skipped/ignored logger.info('Found book %s by %s, marking as "Wanted"' % (bookname, authorname)) controlValueDict = {"BookID": bookid} newValueDict = {"Status": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) new_books += 1 if bookmatch["Requester"]: # Already on a wishlist if book["dispname"] not in bookmatch["Requester"]: newValueDict = {"Requester": bookmatch["Requester"] + book["dispname"] + ' '} controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = {"Requester": book["dispname"] + ' '} controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) if bookmatch['AudioStatus'] in ['Open', 'Wanted', 'Have']: logger.info( 'Found audiobook %s by %s, already marked as "%s"' % (bookname, authorname, bookmatch['AudioStatus'])) if bookmatch["AudioRequester"]: # Already on a wishlist if book["dispname"] not in bookmatch["AudioRequester"]: newValueDict = {"AudioRequester": bookmatch["AudioRequester"] + book["dispname"] + ' '} controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = {"AudioRequester": book["dispname"] + ' '} controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) elif auth_status in ['Paused', 'Ignored']: logger.info('Found book %s, but author is "%s"' % (bookname, auth_status)) elif reject_series: logger.info('Found book %s, but series "%s" is %s' % (bookname, reject_series['Name'], reject_series['Status'])) elif audio_status == 'Wanted': # skipped/ignored logger.info('Found audiobook %s by %s, marking as "Wanted"' % (bookname, authorname)) controlValueDict = {"BookID": bookid} newValueDict = {"AudioStatus": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) new_books += 1 if bookmatch["AudioRequester"]: # Already on a wishlist if book["dispname"] not in bookmatch["AudioRequester"]: newValueDict = {"AudioRequester": bookmatch["AudioRequester"] + book["dispname"] + ' '} controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = {"AudioRequester": book["dispname"] + ' '} controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) else: # not in database yet if book['rss_isbn']: results = search_for(book['rss_isbn']) if results: result = results[0] # type: dict if result['isbn_fuzz'] > check_int(lazylibrarian.CONFIG['MATCH_RATIO'], 90): logger.info("Found (%s%%) %s: %s" % (result['isbn_fuzz'], result['authorname'], result['bookname'])) import_book(result['bookid'], ebook_status, audio_status) new_books += 1 newValueDict = {"Requester": book["dispname"] + ' '} controlValueDict = {"BookID": result['bookid']} myDB.upsert("books", newValueDict, controlValueDict) newValueDict = {"AudioRequester": book["dispname"] + ' '} myDB.upsert("books", newValueDict, controlValueDict) 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] # type: dict if result['author_fuzz'] > check_int(lazylibrarian.CONFIG['MATCH_RATIO'], 90) \ and result['book_fuzz'] > check_int(lazylibrarian.CONFIG['MATCH_RATIO'], 90): logger.info("Found (%s%% %s%%) %s: %s" % (result['author_fuzz'], result['book_fuzz'], result['authorname'], result['bookname'])) import_book(result['bookid'], ebook_status, audio_status) new_books += 1 newValueDict = {"Requester": book["dispname"] + ' '} controlValueDict = {"BookID": result['bookid']} myDB.upsert("books", newValueDict, controlValueDict) newValueDict = {"AudioRequester": book["dispname"] + ' '} myDB.upsert("books", newValueDict, controlValueDict) bookmatch = True if not bookmatch: msg = "Skipping book %s by %s" % (item['Title'], book['rss_author']) if not results: msg += ', No results returned' logger.warn(msg) else: msg += ', No match found' logger.warn(msg) result = results[0] # type: dict msg = "Closest match (%s%% %s%%) %s: %s" % (result['author_fuzz'], result['book_fuzz'], result['authorname'], result['bookname']) logger.warn(msg) if new_books: logger.info("Wishlist marked %s book%s as Wanted" % (new_books, plural(new_books))) except Exception: logger.error('Unhandled exception in search_wishlist: %s' % traceback.format_exc()) finally: threading.currentThread().name = "WEBSERVER"
def downloadResult(match, book): """ match: best result from search providers book: book we are downloading (needed for reporting author name) return: 0 if failed to snatch 1 if already snatched 2 if we snatched it """ # noinspection PyBroadException try: myDB = database.DBConnection() newValueDict = match[1] controlValueDict = match[2] # It's possible to get book and wanted tables "Snatched" status out of sync # for example if a user marks a book as "Wanted" after a search task snatches it and before postprocessor runs # so check status in both tables here snatched = myDB.match( 'SELECT BookID from wanted WHERE BookID=? and AuxInfo=? and Status="Snatched"', (newValueDict["BookID"], newValueDict["AuxInfo"])) if snatched: logger.debug('%s %s %s already marked snatched in wanted table' % (newValueDict["AuxInfo"], book['authorName'], book['bookName'])) return 1 # someone else already found it if newValueDict["AuxInfo"] == 'eBook': snatched = myDB.match( 'SELECT BookID from books WHERE BookID=? and Status="Snatched"', (newValueDict["BookID"], )) else: snatched = myDB.match( 'SELECT BookID from books WHERE BookID=? and AudioStatus="Snatched"', (newValueDict["BookID"], )) if snatched: logger.debug('%s %s %s already marked snatched in book table' % (newValueDict["AuxInfo"], book['authorName'], book['bookName'])) return 1 # someone else already found it myDB.upsert("wanted", newValueDict, controlValueDict) if newValueDict['NZBmode'] == 'direct': snatch = DirectDownloadMethod(newValueDict["BookID"], newValueDict["NZBtitle"], controlValueDict["NZBurl"], newValueDict["AuxInfo"]) elif newValueDict['NZBmode'] in ["torznab", "torrent", "magnet"]: snatch = TORDownloadMethod(newValueDict["BookID"], newValueDict["NZBtitle"], controlValueDict["NZBurl"], newValueDict["AuxInfo"]) elif newValueDict['NZBmode'] == 'nzb': snatch = NZBDownloadMethod(newValueDict["BookID"], newValueDict["NZBtitle"], controlValueDict["NZBurl"], newValueDict["AuxInfo"]) else: logger.error('Unhandled NZBmode [%s] for %s' % (newValueDict['NZBmode'], controlValueDict["NZBurl"])) snatch = 0 if snatch: logger.info('Downloading %s %s from %s' % (newValueDict["AuxInfo"], newValueDict["NZBtitle"], newValueDict["NZBprov"])) custom_notify_snatch( "%s %s" % (newValueDict["BookID"], newValueDict['AuxInfo'])) notify_snatch("%s %s from %s at %s" % (newValueDict["AuxInfo"], newValueDict["NZBtitle"], newValueDict["NZBprov"], now())) # at this point we could add NZBprov to the blocklist with a short timeout, a second or two? # This would implement a round-robin search system. Blocklist with an incremental counter. # If number of active providers == number blocklisted, so no unblocked providers are left, # either sleep for a while, or unblock the one with the lowest counter. scheduleJob(action='Start', target='processDir') return 2 # we found it return 0 except Exception: logger.error('Unhandled exception in downloadResult: %s' % traceback.format_exc()) return 0
def processDir(reset=False): try: threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "POSTPROCESS" processpath = lazylibrarian.DIRECTORY('Download') logger.debug('Checking [%s] for files to post process' % processpath) try: downloads = os.listdir(processpath) except OSError as why: logger.error('Could not access [%s] directory [%s]' % (processpath, why.strerror)) return myDB = database.DBConnection() snatched = myDB.select('SELECT * from wanted WHERE Status="Snatched"') if len(snatched) == 0: logger.info('Nothing marked as snatched.') scheduleJob(action='Stop', target='processDir') return if len(downloads) == 0: logger.info('No downloads are found. Nothing to process yet.') return logger.info("Checking %s download%s for %s snatched file%s" % (len(downloads), plural(len(downloads)), len(snatched), plural(len(snatched)))) ppcount = 0 for book in snatched: # if torrent, see if we can get current status from the downloader as the name # may have been changed once magnet resolved, or download started or completed # depending on torrent downloader. Usenet doesn't change the name. We like usenet. torrentname = '' try: logger.debug("%s was sent to %s" % (book['NZBtitle'], book['Source'])) if book['Source'] == 'TRANSMISSION': torrentname = transmission.getTorrentFolder(book['DownloadID']) elif book['Source'] == 'UTORRENT': torrentname = utorrent.nameTorrent(book['DownloadID']) elif book['Source'] == 'RTORRENT': torrentname = rtorrent.getName(book['DownloadID']) elif book['Source'] == 'QBITTORRENT': torrentname = qbittorrent.getName(book['DownloadID']) elif book['Source'] == 'SYNOLOGY_TOR': torrentname = synology.getName(book['DownloadID']) elif book['Source'] == 'DELUGEWEBUI': torrentname = deluge.getTorrentFolder(book['DownloadID']) elif book['Source'] == 'DELUGERPC': client = DelugeRPCClient(lazylibrarian.DELUGE_HOST, int(lazylibrarian.DELUGE_PORT), lazylibrarian.DELUGE_USER, lazylibrarian.DELUGE_PASS) try: client.connect() result = client.call('core.get_torrent_status', book['DownloadID'], {}) # for item in result: # logger.debug ('Deluge RPC result %s: %s' % (item, result[item])) if 'name' in result: torrentname = unaccented_str(result['name']) except Exception as e: logger.debug('DelugeRPC failed %s' % str(e)) except Exception as e: logger.debug("Failed to get updated torrent name from %s for %s: %s" % (book['Source'], book['DownloadID'], str(e))) matchtitle = unaccented_str(book['NZBtitle']) if torrentname and torrentname != matchtitle: logger.debug("%s Changing [%s] to [%s]" % (book['Source'], matchtitle, torrentname)) myDB.action('UPDATE wanted SET NZBtitle = "%s" WHERE NZBurl = "%s"' % (torrentname, book['NZBurl'])) matchtitle = torrentname # here we could also check percentage downloaded or eta or status? # If downloader says it hasn't completed, no need to look for it. matches = [] logger.info('Looking for %s in %s' % (matchtitle, processpath)) for fname in downloads: # skip if failed before or incomplete torrents, or incomplete btsync extn = os.path.splitext(fname)[1] if extn not in ['.fail', '.part', '.bts', '.!ut']: # This is to get round differences in torrent filenames. # Usenet is ok, but Torrents aren't always returned with the name we searched for # We ask the torrent downloader for the torrent name, but don't always get an answer # so we try to do a "best match" on the name, there might be a better way... if isinstance(fname, str): matchname = fname.decode(lazylibrarian.SYS_ENCODING) else: matchname = fname if ' LL.(' in matchname: matchname = matchname.split(' LL.(')[0] match = 0 if matchtitle: if ' LL.(' in matchtitle: matchtitle = matchtitle.split(' LL.(')[0] match = fuzz.token_set_ratio(matchtitle, matchname) if match and match >= lazylibrarian.DLOAD_RATIO: fname = matchname if os.path.isfile(os.path.join(processpath, fname)): # handle single file downloads here. Book/mag file in download root. # move the file into it's own subdirectory so we don't move/delete things that aren't ours logger.debug('filename [%s] is a file' % os.path.join(processpath, fname)) if is_valid_booktype(fname, booktype="book") \ or is_valid_booktype(fname, booktype="mag"): logger.debug('filename [%s] is a valid book/mag' % os.path.join(processpath, fname)) if bts_file(processpath): logger.debug("Skipping %s, found a .bts file" % processpath) else: fname = os.path.splitext(fname)[0] dirname = os.path.join(processpath, fname) if not os.path.exists(dirname): try: os.makedirs(dirname) setperm(dirname) except OSError as why: logger.debug('Failed to create directory %s, %s' % (dirname, why.strerror)) if os.path.exists(dirname): # move the book and any related files too # ie other book formats, or opf, jpg with same title # can't move metadata.opf or cover.jpg or similar # as can't be sure they are ours # not sure if we need a new listdir here, or whether we can use the old one list_dir = os.listdir(processpath) for ourfile in list_dir: if ourfile.startswith(fname): if is_valid_booktype(ourfile, booktype="book") \ or is_valid_booktype(ourfile, booktype="mag") \ or os.path.splitext(ourfile)[1].lower() in ['.opf', '.jpg']: try: if lazylibrarian.DESTINATION_COPY: shutil.copyfile(os.path.join(processpath, ourfile), os.path.join(dirname, ourfile)) setperm(os.path.join(dirname, ourfile)) else: shutil.move(os.path.join(processpath, ourfile), os.path.join(dirname, ourfile)) setperm(os.path.join(dirname, ourfile)) except Exception as why: logger.debug("Failed to copy/move file %s to %s, %s" % (ourfile, dirname, str(why))) pp_path = os.path.join(processpath, fname) if os.path.isdir(pp_path): logger.debug('Found folder (%s%%) %s for %s' % (match, pp_path, matchtitle)) if not os.listdir(pp_path): logger.debug("Skipping %s, folder is empty" % pp_path) elif bts_file(pp_path): logger.debug("Skipping %s, found a .bts file" % pp_path) else: matches.append([match, pp_path, book]) else: pp_path = os.path.join(processpath, fname) matches.append([match, pp_path, book]) # so we can report closest match else: logger.debug('Skipping %s' % fname) match = 0 if matches: highest = max(matches, key=lambda x: x[0]) match = highest[0] pp_path = highest[1] book = highest[2] if match and match >= lazylibrarian.DLOAD_RATIO: logger.debug(u'Found match (%s%%): %s for %s' % (match, pp_path, book['NZBtitle'])) data = myDB.match('SELECT * from books WHERE BookID="%s"' % book['BookID']) if data: # it's a book logger.debug(u'Processing book %s' % book['BookID']) authorname = data['AuthorName'] bookname = data['BookName'] if 'windows' in platform.system().lower() and '/' in lazylibrarian.EBOOK_DEST_FOLDER: logger.warn('Please check your EBOOK_DEST_FOLDER setting') lazylibrarian.EBOOK_DEST_FOLDER = lazylibrarian.EBOOK_DEST_FOLDER.replace('/', '\\') # Default destination path, should be allowed change per config file. dest_path = lazylibrarian.EBOOK_DEST_FOLDER.replace('$Author', authorname).replace( '$Title', bookname) global_name = lazylibrarian.EBOOK_DEST_FILE.replace('$Author', authorname).replace( '$Title', bookname) global_name = unaccented(global_name) # dest_path = authorname+'/'+bookname # global_name = bookname + ' - ' + authorname # Remove characters we don't want in the filename BEFORE adding to DESTINATION_DIR # as windows drive identifiers have colon, eg c: but no colons allowed elsewhere? dic = {'<': '', '>': '', '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '', '\'': ''} dest_path = unaccented_str(replace_all(dest_path, dic)) dest_path = os.path.join(processpath, dest_path).encode(lazylibrarian.SYS_ENCODING) else: data = myDB.match('SELECT * from magazines WHERE Title="%s"' % book['BookID']) if data: # it's a magazine logger.debug(u'Processing magazine %s' % book['BookID']) # AuxInfo was added for magazine release date, normally housed in 'magazines' but if multiple # files are downloading, there will be an error in post-processing, trying to go to the # same directory. mostrecentissue = data['IssueDate'] # keep for processing issues arriving out of order # Remove characters we don't want in the filename before (maybe) adding to DESTINATION_DIR # as windows drive identifiers have colon, eg c: but no colons allowed elsewhere? dic = {'<': '', '>': '', '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '', '\'': ''} mag_name = unaccented_str(replace_all(book['BookID'], dic)) # book auxinfo is a cleaned date, eg 2015-01-01 dest_path = lazylibrarian.MAG_DEST_FOLDER.replace( '$IssueDate', book['AuxInfo']).replace('$Title', mag_name) if lazylibrarian.MAG_RELATIVE: if dest_path[0] not in '._': dest_path = '_' + dest_path dest_path = os.path.join(processpath, dest_path).encode( lazylibrarian.SYS_ENCODING) else: dest_path = dest_path.encode(lazylibrarian.SYS_ENCODING) authorname = None bookname = None global_name = lazylibrarian.MAG_DEST_FILE.replace('$IssueDate', book['AuxInfo']).replace( '$Title', mag_name) global_name = unaccented(global_name) else: # not recognised logger.debug('Nothing in database matching "%s"' % book['BookID']) continue else: logger.debug("Snatched %s %s is not in download directory" % (book['NZBmode'], book['NZBtitle'])) if match: logger.debug(u'Closest match (%s%%): %s' % (match, pp_path)) #for match in matches: # logger.info('Match: %s%% %s' % (match[0], match[1])) continue processBook = processDestination(pp_path, dest_path, authorname, bookname, global_name) if processBook: logger.debug("Processing %s, %s" % (global_name, book['NZBurl'])) # update nzbs, only update the snatched ones in case multiple matches for same book / magazine issue controlValueDict = {"BookID": book['BookID'], "NZBurl": book['NZBurl'], "Status": "Snatched"} newValueDict = {"Status": "Processed", "NZBDate": now()} # say when we processed it myDB.upsert("wanted", newValueDict, controlValueDict) if bookname: # it's a book, if None it's a magazine if len(lazylibrarian.IMP_CALIBREDB): logger.debug('Calibre should have created the extras for us') else: processExtras(myDB, dest_path, global_name, data) else: # update mags controlValueDict = {"Title": book['BookID']} if mostrecentissue: if mostrecentissue.isdigit() and str(book['AuxInfo']).isdigit(): older = int(mostrecentissue) > int(book['AuxInfo']) # issuenumber else: older = mostrecentissue > book['AuxInfo'] # YYYY-MM-DD else: older = False if older: # check this in case processing issues arriving out of order newValueDict = {"LastAcquired": today(), "IssueStatus": "Open"} else: newValueDict = {"IssueDate": book['AuxInfo'], "LastAcquired": today(), "IssueStatus": "Open"} myDB.upsert("magazines", newValueDict, controlValueDict) # dest_path is where we put the magazine after processing, but we don't have the full filename # so look for any "book" in that directory dest_file = book_file(dest_path, booktype='mag') controlValueDict = {"Title": book['BookID'], "IssueDate": book['AuxInfo']} newValueDict = {"IssueAcquired": today(), "IssueFile": dest_file, "IssueID": create_id("%s %s" % (book['BookID'], book['AuxInfo'])) } myDB.upsert("issues", newValueDict, controlValueDict) # create a thumbnail cover for the new issue create_cover(dest_file) # calibre or ll copied/moved the files we want, now delete source files to_delete = True if book['NZBmode'] in ['torrent', 'magnet']: # Only delete torrents if we don't want to keep seeding if lazylibrarian.KEEP_SEEDING: logger.warn('%s is seeding %s %s' % (book['Source'], book['NZBmode'], book['NZBtitle'])) to_delete = False else: # ask downloader to delete the torrent, but not the files # we may delete them later, depending on other settings if book['DownloadID'] != "unknown": logger.debug('Removing %s from %s' % (book['NZBtitle'], book['Source'].lower())) delete_task(book['Source'], book['DownloadID'], False) else: logger.warn("Unable to remove %s from %s, no DownloadID" % (book['NZBtitle'], book['Source'].lower())) if to_delete: # only delete the files if not in download root dir and if DESTINATION_COPY not set if not lazylibrarian.DESTINATION_COPY and (pp_path != processpath): if os.path.isdir(pp_path): # calibre might have already deleted it? try: shutil.rmtree(pp_path) except Exception as why: logger.debug("Unable to remove %s, %s" % (pp_path, str(why))) logger.info('Successfully processed: %s' % global_name) ppcount = ppcount + 1 notify_download("%s from %s at %s" % (global_name, book['NZBprov'], now())) else: logger.error('Postprocessing for %s has failed.' % global_name) logger.error('Warning - Residual files remain in %s.fail' % pp_path) controlValueDict = {"NZBurl": book['NZBurl'], "Status": "Snatched"} newValueDict = {"Status": "Failed", "NZBDate": now()} myDB.upsert("wanted", newValueDict, controlValueDict) # if it's a book, reset status so we try for a different version # if it's a magazine, user can select a different one from pastissues table if bookname: myDB.action('UPDATE books SET status = "Wanted" WHERE BookID="%s"' % book['BookID']) # at this point, as it failed we should move it or it will get postprocessed # again (and fail again) try: os.rename(pp_path, pp_path + '.fail') except Exception as e: logger.debug("Unable to rename %s, %s" % (pp_path, str(e))) downloads = os.listdir(processpath) # check in case we processed/deleted some above for directory in downloads: dname, extn = os.path.splitext(directory) if "LL.(" in dname and extn not in ['.fail', '.part', '.bts', '.!ut']: bookID = str(directory).split("LL.(")[1].split(")")[0] logger.debug("Book with id: " + str(bookID) + " found in download directory") pp_path = os.path.join(processpath, directory) if os.path.isfile(pp_path): pp_path = os.path.join(processpath) if (os.path.isdir(pp_path)): if import_book(pp_path, bookID): ppcount = ppcount + 1 if ppcount == 0: logger.info('No snatched books/mags have been found') else: logger.info('%s book%s/mag%s processed.' % (ppcount, plural(ppcount), plural(ppcount))) # Now check for any that are still marked snatched... if lazylibrarian.TASK_AGE: snatched = myDB.select('SELECT * from wanted WHERE Status="Snatched"') if len(snatched) > 0: for snatch in snatched: # FUTURE: we could check percentage downloaded or eta? # if percentage is increasing, it's just slow try: when_snatched = time.strptime(snatch['NZBdate'], '%Y-%m-%d %H:%M:%S') when_snatched = time.mktime(when_snatched) diff = time.time() - when_snatched # time difference in seconds except: diff = 0 hours = int(diff / 3600) if hours >= lazylibrarian.TASK_AGE: logger.warn('%s was sent to %s %s hours ago, deleting failed task' % (snatch['NZBtitle'], snatch['Source'].lower(), hours)) # change status to "Failed", and ask downloader to delete task and files if snatch['BookID'] != 'unknown': myDB.action('UPDATE wanted SET Status="Failed" WHERE BookID="%s"' % snatch['BookID']) myDB.action('UPDATE books SET status = "Wanted" WHERE BookID="%s"' % snatch['BookID']) delete_task(snatch['Source'], snatch['DownloadID'], True) if reset: scheduleJob(action='Restart', target='processDir') except Exception as e: logger.error('Unhandled exception in processDir: %s' % traceback.format_exc())
def search_tor_book(books=None, reset=False): try: threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "SEARCHTOR" if not lazylibrarian.USE_TOR(): logger.warn('No Torrent providers set, check config') return myDB = database.DBConnection() searchlist = [] if books is None: # We are performing a backlog search searchbooks = myDB.select( 'SELECT BookID, AuthorName, Bookname, BookSub, BookAdded from books WHERE Status="Wanted" order by BookAdded desc') else: # The user has added a new book searchbooks = [] for book in books: searchbook = myDB.select('SELECT BookID, AuthorName, BookName, BookSub from books WHERE BookID="%s" \ AND Status="Wanted"' % book['bookid']) 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 = 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 as e: logger.error('Unhandled exception in search_tor_book: %s' % traceback.format_exc())
def search_wishlist(): try: threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "SEARCHWISHLIST" myDB = database.DBConnection() resultlist, wishproviders = IterateOverWishLists() new_books = 0 if not wishproviders: logger.debug('No wishlists are set') scheduleJob(action='Stop', target='search_wishlist') return # No point in continuing # for each item in resultlist, add to database if necessary, and mark as wanted logger.debug('Processing %s item%s in wishlists' % (len(resultlist), plural(len(resultlist)))) for book in resultlist: # we get rss_author, rss_title, maybe rss_isbn, rss_bookid (goodreads bookid) # we can just use bookid if goodreads, or try isbn and name matching on author/title if not # eg NYTimes wishlist if 'E' in book['types']: ebook_status = "Wanted" else: ebook_status = "Skipped" if 'A' in book['types']: audio_status = "Wanted" else: audio_status = "Skipped" if lazylibrarian.CONFIG['BOOK_API'] == "GoodReads" and book[ 'rss_bookid']: cmd = 'select Status,AudioStatus,BookName,Requester,AudioRequester from books where bookid=?' bookmatch = myDB.match(cmd, (book['rss_bookid'], )) if bookmatch: bookname = bookmatch['BookName'] if bookmatch['Status'] in ['Open', 'Wanted', 'Have']: logger.info('Found book %s, already marked as "%s"' % (bookname, bookmatch['Status'])) if bookmatch["Requester"]: # Already on a wishlist if book["dispname"] not in bookmatch["Requester"]: newValueDict = { "Requester": bookmatch["Requester"] + book["dispname"] + ' ' } controlValueDict = { "BookID": book['rss_bookid'] } myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = { "Requester": book["dispname"] + ' ' } controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) elif ebook_status == "Wanted": # skipped/ignored logger.info('Found book %s, marking as "Wanted"' % bookname) controlValueDict = {"BookID": book['rss_bookid']} newValueDict = {"Status": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) new_books += 1 if bookmatch["Requester"]: # Already on a wishlist if book["dispname"] not in bookmatch["Requester"]: newValueDict = { "Requester": bookmatch["Requester"] + book["dispname"] + ' ' } controlValueDict = { "BookID": book['rss_bookid'] } myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = { "Requester": book["dispname"] + ' ' } controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) if bookmatch['AudioStatus'] in ['Open', 'Wanted', 'Have']: logger.info( 'Found audiobook %s, already marked as "%s"' % (bookname, bookmatch['AudioStatus'])) if bookmatch[ "AudioRequester"]: # Already on a wishlist if book["dispname"] not in bookmatch[ "AudioRequester"]: newValueDict = { "AudioRequester": bookmatch["AudioRequester"] + book["dispname"] + ' ' } controlValueDict = { "BookID": book['rss_bookid'] } myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = { "AudioRequester": book["dispname"] + ' ' } controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) elif audio_status == "Wanted": # skipped/ignored logger.info('Found audiobook %s, marking as "Wanted"' % bookname) controlValueDict = {"BookID": book['rss_bookid']} newValueDict = {"AudioStatus": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) new_books += 1 if bookmatch[ "AudioRequester"]: # Already on a wishlist if book["dispname"] not in bookmatch[ "AudioRequester"]: newValueDict = { "AudioRequester": bookmatch["AudioRequester"] + book["dispname"] + ' ' } controlValueDict = { "BookID": book['rss_bookid'] } myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = { "AudioRequester": book["dispname"] + ' ' } controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) else: import_book(book['rss_bookid'], ebook_status, audio_status) new_books += 1 newValueDict = {"Requester": book["dispname"] + ' '} controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) newValueDict = {"AudioRequester": book["dispname"] + ' '} controlValueDict = {"BookID": book['rss_bookid']} myDB.upsert("books", newValueDict, controlValueDict) else: item = {} results = None item['Title'] = book['rss_title'] if book['rss_bookid']: item['BookID'] = book['rss_bookid'] if book['rss_isbn']: item['ISBN'] = book['rss_isbn'] bookmatch = finditem(item, book['rss_author']) if bookmatch: # it's already in the database authorname = bookmatch['AuthorName'] bookname = bookmatch['BookName'] bookid = bookmatch['BookID'] if bookmatch['Status'] in ['Open', 'Wanted', 'Have']: logger.info( 'Found book %s by %s, already marked as "%s"' % (bookname, authorname, bookmatch['Status'])) if bookmatch["Requester"]: # Already on a wishlist if book["dispname"] not in bookmatch["Requester"]: newValueDict = { "Requester": bookmatch["Requester"] + book["dispname"] + ' ' } controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = { "Requester": book["dispname"] + ' ' } controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) elif ebook_status == 'Wanted': # skipped/ignored logger.info( 'Found book %s by %s, marking as "Wanted"' % (bookname, authorname)) controlValueDict = {"BookID": bookid} newValueDict = {"Status": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) new_books += 1 if bookmatch["Requester"]: # Already on a wishlist if book["dispname"] not in bookmatch["Requester"]: newValueDict = { "Requester": bookmatch["Requester"] + book["dispname"] + ' ' } controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = { "Requester": book["dispname"] + ' ' } controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) if bookmatch['AudioStatus'] in ['Open', 'Wanted', 'Have']: logger.info( 'Found audiobook %s by %s, already marked as "%s"' % (bookname, authorname, bookmatch['AudioStatus'])) if bookmatch[ "AudioRequester"]: # Already on a wishlist if book["dispname"] not in bookmatch[ "AudioRequester"]: newValueDict = { "AudioRequester": bookmatch["AudioRequester"] + book["dispname"] + ' ' } controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = { "AudioRequester": book["dispname"] + ' ' } controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) elif audio_status == 'Wanted': # skipped/ignored logger.info( 'Found audiobook %s by %s, marking as "Wanted"' % (bookname, authorname)) controlValueDict = {"BookID": bookid} newValueDict = {"AudioStatus": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) new_books += 1 if bookmatch[ "AudioRequester"]: # Already on a wishlist if book["dispname"] not in bookmatch[ "AudioRequester"]: newValueDict = { "AudioRequester": bookmatch["AudioRequester"] + book["dispname"] + ' ' } controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) else: newValueDict = { "AudioRequester": book["dispname"] + ' ' } controlValueDict = {"BookID": bookid} myDB.upsert("books", newValueDict, controlValueDict) else: # not in database yet if book['rss_isbn']: results = search_for(book['rss_isbn']) if results: result = results[0] # type: dict if result['isbn_fuzz'] > check_int( lazylibrarian.CONFIG['MATCH_RATIO'], 90): logger.info( "Found (%s%%) %s: %s" % (result['isbn_fuzz'], result['authorname'], result['bookname'])) import_book(result['bookid'], ebook_status, audio_status) new_books += 1 newValueDict = { "Requester": book["dispname"] + ' ' } controlValueDict = {"BookID": result['bookid']} myDB.upsert("books", newValueDict, controlValueDict) newValueDict = { "AudioRequester": book["dispname"] + ' ' } myDB.upsert("books", newValueDict, controlValueDict) 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] # type: dict if result['author_fuzz'] > check_int(lazylibrarian.CONFIG['MATCH_RATIO'], 90) \ and result['book_fuzz'] > check_int(lazylibrarian.CONFIG['MATCH_RATIO'], 90): logger.info( "Found (%s%% %s%%) %s: %s" % (result['author_fuzz'], result['book_fuzz'], result['authorname'], result['bookname'])) import_book(result['bookid'], ebook_status, audio_status) new_books += 1 newValueDict = { "Requester": book["dispname"] + ' ' } controlValueDict = {"BookID": result['bookid']} myDB.upsert("books", newValueDict, controlValueDict) newValueDict = { "AudioRequester": book["dispname"] + ' ' } myDB.upsert("books", newValueDict, controlValueDict) bookmatch = True if not bookmatch: msg = "Skipping book %s by %s" % (item['Title'], book['rss_author']) if not results: msg += ', No results returned' logger.warn(msg) else: msg += ', No match found' logger.warn(msg) result = results[0] # type: dict msg = "Closest match (%s%% %s%%) %s: %s" % ( result['author_fuzz'], result['book_fuzz'], result['authorname'], result['bookname']) logger.warn(msg) if new_books: logger.info("Wishlist marked %s book%s as Wanted" % (new_books, plural(new_books))) except Exception: logger.error('Unhandled exception in search_wishlist: %s' % traceback.format_exc()) finally: threading.currentThread().name = "WEBSERVER"
def processDir(reset=False): threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "POSTPROCESS" if not lazylibrarian.DOWNLOAD_DIR or not os.path.isdir(lazylibrarian.DOWNLOAD_DIR): processpath = os.getcwd() else: processpath = lazylibrarian.DOWNLOAD_DIR logger.debug(' Checking [%s] for files to post process' % processpath) try: downloads = os.listdir(processpath) except OSError as why: logger.error('Could not access [%s] directory [%s]' % (processpath, why.strerror)) return myDB = database.DBConnection() snatched = myDB.select('SELECT * from wanted WHERE Status="Snatched"') if len(snatched) == 0: logger.info('Nothing marked as snatched.') scheduleJob(action='Stop', target='processDir') return if len(downloads) == 0: logger.info('No downloads are found. Nothing to process.') return logger.info("Checking %s download%s for %s snatched file%s" % (len(downloads), plural(len(downloads)), len(snatched), plural(len(snatched)))) ppcount = 0 for book in snatched: matches = [] for fname in downloads: if not fname.endswith('.fail'): # has this failed before? # this is to get round differences in torrent filenames. # Torrents aren't always returned with the name we searched for # there might be a better way... if isinstance(fname, str): matchname = fname.decode(lazylibrarian.SYS_ENCODING) else: matchname = fname if ' LL.(' in matchname: matchname = matchname.split(' LL.(')[0] matchtitle = book['NZBtitle'] match = 0 if matchtitle: if ' LL.(' in matchtitle: matchtitle = matchtitle.split(' LL.(')[0] match = fuzz.token_set_ratio(matchtitle, matchname) if match >= lazylibrarian.DLOAD_RATIO: fname = matchname if os.path.isfile(os.path.join(processpath, fname)): # handle single file downloads here... if is_valid_booktype(fname, booktype="book") \ or is_valid_booktype(fname, booktype="mag"): dirname = os.path.join(processpath, os.path.splitext(fname)[0]) if not os.path.exists(dirname): try: os.makedirs(dirname) except OSError as why: logger.debug('Failed to create directory %s, %s' % (dirname, why.strerror)) if os.path.exists(dirname): try: shutil.move(os.path.join(processpath, fname), os.path.join(dirname, fname)) fname = os.path.splitext(fname)[0] except Exception as why: logger.debug("Failed to move file %s to %s, %s" % (fname, dirname, str(why))) if os.path.isdir(os.path.join(processpath, fname)): pp_path = os.path.join(processpath, fname) logger.debug('Found folder (%s%%) %s for %s' % (match, pp_path, book['NZBtitle'])) matches.append([match, pp_path, book]) else: logger.debug('No match (%s%%) %s for %s' % (match, matchname, matchtitle)) else: logger.debug('Skipping %s' % fname) if matches: highest = max(matches, key=lambda x: x[0]) match = highest[0] pp_path = highest[1] book = highest[2] logger.info(u'Best match (%s%%): %s for %s' % (match, pp_path, book['NZBtitle'])) data = myDB.select('SELECT * from books WHERE BookID="%s"' % book['BookID']) if data: authorname = data[0]['AuthorName'] bookname = data[0]['BookName'] if 'windows' in platform.system().lower() and '/' in lazylibrarian.EBOOK_DEST_FOLDER: logger.warn('Please check your EBOOK_DEST_FOLDER setting') lazylibrarian.EBOOK_DEST_FOLDER = lazylibrarian.EBOOK_DEST_FOLDER.replace('/', '\\') # Default destination path, should be allowed change per config file. dest_path = lazylibrarian.EBOOK_DEST_FOLDER.replace('$Author', authorname).replace( '$Title', bookname) global_name = lazylibrarian.EBOOK_DEST_FILE.replace('$Author', authorname).replace( '$Title', bookname) global_name = unaccented(global_name) # dest_path = authorname+'/'+bookname # global_name = bookname + ' - ' + authorname # Remove characters we don't want in the filename BEFORE adding to DESTINATION_DIR # as windows drive identifiers have colon, eg c: but no colons allowed elsewhere? dic = {'<': '', '>': '', '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '', '\'': ''} dest_path = unaccented_str(replace_all(dest_path, dic)) dest_path = os.path.join(lazylibrarian.DESTINATION_DIR, dest_path).encode( lazylibrarian.SYS_ENCODING) else: data = myDB.select('SELECT * from magazines WHERE Title="%s"' % book['BookID']) if data: # AuxInfo was added for magazine release date, normally housed in 'magazines' but if multiple # files are downloading, there will be an error in post-processing, trying to go to the # same directory. mostrecentissue = data[0]['IssueDate'] # keep for processing issues arriving out of order # Remove characters we don't want in the filename before (maybe) adding to DESTINATION_DIR # as windows drive identifiers have colon, eg c: but no colons allowed elsewhere? dic = {'<': '', '>': '', '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '', '\'': ''} mag_name = unaccented_str(replace_all(book['BookID'], dic)) # book auxinfo is a cleaned date, eg 2015-01-01 dest_path = lazylibrarian.MAG_DEST_FOLDER.replace( '$IssueDate', book['AuxInfo']).replace('$Title', mag_name) # dest_path = '_Magazines/'+title+'/'+book['AuxInfo'] if lazylibrarian.MAG_RELATIVE: if dest_path[0] not in '._': dest_path = '_' + dest_path dest_path = os.path.join(lazylibrarian.DESTINATION_DIR, dest_path).encode( lazylibrarian.SYS_ENCODING) else: dest_path = dest_path.encode(lazylibrarian.SYS_ENCODING) authorname = None bookname = None global_name = lazylibrarian.MAG_DEST_FILE.replace('$IssueDate', book['AuxInfo']).replace( '$Title', mag_name) global_name = unaccented(global_name) # global_name = book['AuxInfo']+' - '+title else: logger.debug("Snatched magazine %s is not in download directory" % (book['BookID'])) continue else: logger.debug("Snatched %s %s is not in download directory" % (book['NZBmode'], book['NZBtitle'])) continue processBook = processDestination(pp_path, dest_path, authorname, bookname, global_name) if processBook: logger.debug("Processing %s, %s" % (global_name, book['NZBurl'])) # update nzbs, only update the snatched ones in case multiple matches for same book / magazine issue controlValueDict = {"NZBurl": book['NZBurl'], "Status": "Snatched"} newValueDict = {"Status": "Processed", "NZBDate": now()} # say when we processed it myDB.upsert("wanted", newValueDict, controlValueDict) if bookname is not None: # it's a book, if None it's a magazine if len(lazylibrarian.IMP_CALIBREDB): logger.debug('Calibre should have created the extras for us') else: processExtras(myDB, dest_path, global_name, data) else: # update mags controlValueDict = {"Title": book['BookID']} if mostrecentissue: if mostrecentissue.isdigit() and str(book['AuxInfo']).isdigit(): older = int(mostrecentissue) > int(book['AuxInfo']) # issuenumber else: older = mostrecentissue > book['AuxInfo'] # YYYY-MM-DD else: older = False if older: # check this in case processing issues arriving out of order newValueDict = {"LastAcquired": today(), "IssueStatus": "Open"} else: newValueDict = {"IssueDate": book['AuxInfo'], "LastAcquired": today(), "IssueStatus": "Open"} myDB.upsert("magazines", newValueDict, controlValueDict) # dest_path is where we put the magazine after processing, but we don't have the full filename # so look for any "book" in that directory dest_file = book_file(dest_path, booktype='mag') controlValueDict = {"Title": book['BookID'], "IssueDate": book['AuxInfo']} newValueDict = {"IssueAcquired": today(), "IssueFile": dest_file, "IssueID": create_id("%s %s" % (book['BookID'], book['AuxInfo'])) } myDB.upsert("issues", newValueDict, controlValueDict) # create a thumbnail cover for the new issue create_cover(dest_file) logger.info('Successfully processed: %s' % global_name) ppcount = ppcount + 1 notify_download("%s at %s" % (global_name, now())) else: logger.error('Postprocessing for %s has failed.' % global_name) logger.error('Warning - Residual files remain in %s.fail' % pp_path) controlValueDict = {"NZBurl": book['NZBurl'], "Status": "Snatched"} newValueDict = {"Status": "Failed", "NZBDate": now()} myDB.upsert("wanted", newValueDict, controlValueDict) # if it's a book, reset status so we try for a different version # if it's a magazine, user can select a different one from pastissues table if bookname is not None: myDB.action('UPDATE books SET status = "Wanted" WHERE BookID="%s"' % book['BookID']) # at this point, as it failed we should move it or it will get postprocessed # again (and fail again) try: os.rename(pp_path, pp_path + '.fail') except: logger.debug("Unable to rename %s" % pp_path) downloads = os.listdir(processpath) # check in case we processed/deleted some above for directory in downloads: if "LL.(" in directory and not directory.endswith('.fail'): bookID = str(directory).split("LL.(")[1].split(")")[0] logger.debug("Book with id: " + str(bookID) + " is in downloads") pp_path = os.path.join(processpath, directory) if os.path.isfile(pp_path): pp_path = os.path.join(processpath) if (os.path.isdir(pp_path)): logger.debug('Found LL folder %s.' % pp_path) if import_book(pp_path, bookID): ppcount = ppcount + 1 if ppcount == 0: logger.info('No snatched books/mags have been found') else: logger.info('%s book%s/mag%s processed.' % (ppcount, plural(ppcount), plural(ppcount))) if reset: scheduleJob(action='Restart', target='processDir')
def search_magazines(mags=None, reset=False): # produce a list of magazines to search for, tor, nzb, torznab, rss # noinspection PyBroadException try: threadname = threading.currentThread().name if "Thread-" in threadname: if not mags: threading.currentThread().name = "SEARCHALLMAG" else: threading.currentThread().name = "SEARCHMAG" myDB = database.DBConnection() searchlist = [] if not mags: # backlog search searchmags = myDB.select('SELECT Title, Regex, DateType, LastAcquired, \ IssueDate from magazines WHERE Status="Active"') else: searchmags = [] for magazine in mags: searchmags_temp = myDB.select('SELECT Title,Regex,DateType,LastAcquired,IssueDate from magazines \ WHERE Title=? AND Status="Active"', (magazine['bookid'],)) for terms in searchmags_temp: searchmags.append(terms) if len(searchmags) == 0: threading.currentThread().name = "WEBSERVER" 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("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'] datetype = searchmag['DateType'] if not datetype: datetype = '' if not searchterm: dic = {'...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': ''} # strip accents from the magazine title for easier name-matching searchterm = unaccented_str(searchmag['Title']) if not searchterm: # unless there are no ascii characters left searchterm = searchmag['Title'] searchterm = replace_all(searchterm, dic) searchterm = re.sub('[.\-/]', ' ', searchterm) if PY2: searchterm = searchterm.encode(lazylibrarian.SYS_ENCODING) searchlist.append({"bookid": bookid, "searchterm": searchterm, "datetype": datetype}) 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: # don't nag. Show warning message no more than every 20 mins timenow = int(time.time()) if check_int(lazylibrarian.NO_NZB_MSG, 0) + 1200 < timenow: logger.warn('No nzb providers are available. Check config and blocklist') lazylibrarian.NO_NZB_MSG = timenow if lazylibrarian.USE_DIRECT(): dir_resultlist, nproviders = IterateOverDirectSites(book, 'mag') if not nproviders: # don't nag. Show warning message no more than every 20 mins timenow = int(time.time()) if check_int(lazylibrarian.NO_DIRECT_MSG, 0) + 1200 < timenow: logger.warn('No direct providers are available. Check config and blocklist') lazylibrarian.NO_DIRECT_MSG = timenow if dir_resultlist: for item in dir_resultlist: # reformat the 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 'nzbsize': item['tor_size'], 'nzbmode': 'direct' }) if lazylibrarian.USE_TOR(): tor_resultlist, nproviders = IterateOverTorrentSites(book, 'mag') if not nproviders: # don't nag. Show warning message no more than every 20 mins timenow = int(time.time()) if check_int(lazylibrarian.NO_TOR_MSG, 0) + 1200 < timenow: logger.warn('No tor providers are available. Check config and blocklist') lazylibrarian.NO_TOR_MSG = timenow 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, dltypes = IterateOverRSSSites() if not nproviders or 'M' not in dltypes: # don't nag. Show warning message no more than every 20 mins timenow = int(time.time()) if check_int(lazylibrarian.NO_RSS_MSG, 0) + 1200 < timenow: logger.warn('No rss providers are available. Check config and blocklist') lazylibrarian.NO_RSS_MSG = timenow if rss_resultlist: for item in rss_resultlist: # reformat the rss results so they look like nzbs if 'M' in item['types']: 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 'nzbsize': item['tor_size'], 'nzbmode': item['tor_type'] }) if not resultlist: logger.debug("No results for magazine %s" % 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'] # strip accents from the magazine title for easier name-matching nzbtitle = unaccented_str(nzb['nzbtitle']) if not nzbtitle: # unless it's not a latin-1 encodable name nzbtitle = 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'] # Need to make sure that substrings of magazine titles don't get found # (e.g. Maxim USA will find Maximum PC USA) so split into "words" dic = {'.': ' ', '-': ' ', '/': ' ', '+': ' ', '_': ' ', '(': '', ')': '', '[': ' ', ']': ' ', '#': '# '} nzbtitle_formatted = replace_all(nzbtitle, dic) # remove extra spaces if they're in a row nzbtitle_formatted = " ".join(nzbtitle_formatted.split()) nzbtitle_exploded = nzbtitle_formatted.split() results = myDB.match('SELECT * from magazines WHERE Title=?', (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: bookid_exploded = replace_all(bookid, dic).split() # 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 all the words in the mag title are in the nzbtitle rejected = False wlist = [] for word in nzbtitle_exploded: word = unaccented(word).lower() if word: wlist.append(word) for word in bookid_exploded: word = unaccented(word).lower() if word and word not in wlist: logger.debug("Rejecting %s, missing %s" % (nzbtitle, word)) rejected = True break if rejected: logger.debug( "Magazine title match failed " + bookid + " for " + nzbtitle_formatted) else: logger.debug( "Magazine title matched " + bookid + " for " + nzbtitle_formatted) else: logger.debug("Magazine name too short (%s)" % len(nzbtitle_exploded)) rejected = True if not rejected and lazylibrarian.CONFIG['BLACKLIST_FAILED']: blocked = myDB.match('SELECT * from wanted WHERE NZBurl=? and Status="Failed"', (nzburl,)) if blocked: logger.debug("Rejecting %s, blacklisted at %s" % (nzbtitle_formatted, blocked['NZBprov'])) rejected = True if not rejected and lazylibrarian.CONFIG['BLACKLIST_PROCESSED']: blocked = myDB.match('SELECT * from wanted WHERE NZBurl=?', (nzburl,)) if blocked: logger.debug("Rejecting %s, blacklisted at %s" % (nzbtitle_formatted, blocked['NZBprov'])) rejected = True if not rejected: reject_list = getList(str(results['Reject']).lower()) reject_list += getList(lazylibrarian.CONFIG['REJECT_MAGS'], ',') lower_title = unaccented(nzbtitle_formatted).lower() lower_bookid = unaccented(bookid).lower() if reject_list: if lazylibrarian.LOGLEVEL & lazylibrarian.log_searchmag: logger.debug('Reject: %s' % str(reject_list)) logger.debug('Title: %s' % lower_title) logger.debug('Bookid: %s' % lower_bookid) 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 if rejected: rejects += 1 else: regex_pass, issuedate, year = get_issue_date(nzbtitle_exploded) if regex_pass: logger.debug('Issue %s (regex %s) for %s ' % (issuedate, regex_pass, nzbtitle_formatted)) datetype_ok = True datetype = book['datetype'] if datetype: # check all wanted parts are in the regex result # Day Month Year Vol Iss (MM needs two months) if 'M' in datetype and regex_pass not in [1, 2, 3, 4, 5, 6, 7, 12]: datetype_ok = False elif 'D' in datetype and regex_pass not in [3, 5, 6]: datetype_ok = False elif 'MM' in datetype and regex_pass not in [1]: # bi monthly datetype_ok = False elif 'V' in datetype and 'I' in datetype and regex_pass not in [8, 9, 17, 18]: datetype_ok = False elif 'V' in datetype and regex_pass not in [2, 10, 11, 12, 13, 14, 17, 18]: datetype_ok = False elif 'I' in datetype and regex_pass not in [2, 10, 11, 12, 13, 14, 16, 17, 18]: datetype_ok = False elif 'Y' in datetype and regex_pass not in [1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 13, 15, 16, 18]: datetype_ok = False else: datetype_ok = False 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 issuedate = "1970-01-01" # provide a fake date for bad-date issues # wanted issues go into wanted table marked "Wanted" # the rest into pastissues table marked "Skipped" or "Have" insert_table = "pastissues" comp_date = 0 if datetype_ok: control_date = results['IssueDate'] logger.debug("Control date: [%s]" % control_date) if not control_date: # we haven't got any copies of this magazine yet # get a rough time just over MAX_AGE days ago to compare to, in format yyyy-mm-dd # could perhaps calc differently for weekly, biweekly etc # For magazines with only an issue number use zero as we can't tell age if str(issuedate).isdigit(): logger.debug('Magazine comparing issue numbers (%s)' % issuedate) control_date = 0 elif re.match('\d+-\d\d-\d\d', str(issuedate)): 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: logger.debug('Magazine unable to find comparison type [%s]' % issuedate) control_date = 0 if str(control_date).isdigit() and str(issuedate).isdigit(): # for issue numbers, check if later than last one we have if regex_pass in [10, 12, 13] and year: issuedate = "%s%04d" % (year, int(issuedate)) else: issuedate = str(issuedate).zfill(4) if not control_date: comp_date = 1 else: comp_date = int(issuedate) - int(control_date) elif re.match('\d+-\d\d-\d\d', str(control_date)) and \ re.match('\d+-\d\d-\d\d', str(issuedate)): # 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(issuedate, control_date) else: # invalid comparison of date and issue number comp_date = 0 if re.match('\d+-\d\d-\d\d', str(control_date)): if regex_pass > 9 and year: # we assumed it was an issue number, but it could be a date year = check_int(year, 0) if regex_pass in [10, 12, 13]: issuedate = int(issuedate[:4]) issuenum = check_int(issuedate, 0) if year and 1 <= issuenum <= 12: issuedate = "%04d-%02d-01" % (year, issuenum) comp_date = datecompare(issuedate, control_date) if not comp_date: logger.debug('Magazine %s failed: Expecting a date' % nzbtitle_formatted) else: logger.debug('Magazine %s failed: Expecting issue number' % nzbtitle_formatted) if not comp_date: bad_date += 1 issuedate = "1970-01-01" if issuedate == "1970-01-01": logger.debug('This issue of %s is unknown age; skipping.' % nzbtitle_formatted) elif not datetype_ok: logger.debug('This issue of %s not in a wanted date format.' % nzbtitle_formatted) elif comp_date > 0: # keep track of what we're going to download so we don't download dupes new_date += 1 issue = bookid + ',' + issuedate 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) logger.debug('Magazine request number %s' % len(issues)) if lazylibrarian.LOGLEVEL & lazylibrarian.log_searchmag: logger.debug(str(issues)) insert_table = "wanted" nzbdate = now() # when we asked for it else: logger.debug('This issue of %s is already flagged for download' % issue) else: if lazylibrarian.LOGLEVEL & lazylibrarian.log_searchmag: 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 Status from %s WHERE NZBtitle=? and NZBprov=?' % insert_table, (nzbtitle, nzbprov)) if mag_entry: if lazylibrarian.LOGLEVEL & lazylibrarian.log_searchmag: logger.debug('%s is already in %s marked %s' % (nzbtitle, insert_table, mag_entry['Status'])) else: controlValueDict = { "NZBtitle": nzbtitle, "NZBprov": nzbprov } if insert_table == 'pastissues': # try to mark ones we've already got match = myDB.match("SELECT * from issues WHERE Title=? AND IssueDate=?", (bookid, issuedate)) if match: insert_status = "Have" else: insert_status = "Skipped" else: insert_status = "Wanted" newValueDict = { "NZBurl": nzburl, "BookID": bookid, "NZBdate": nzbdate, "AuxInfo": issuedate, "Status": insert_status, "NZBsize": nzbsize, "NZBmode": nzbmode } myDB.upsert(insert_table, newValueDict, controlValueDict) if lazylibrarian.LOGLEVEL & lazylibrarian.log_searchmag: logger.debug('Added %s to %s marked %s' % (nzbtitle, insert_table, insert_status)) msg = 'Found %i result%s for %s. %i new,' % (total_nzbs, plural(total_nzbs), bookid, new_date) msg += ' %i old, %i fail date, %i fail name,' % (old_date, bad_date, bad_name) msg += ' %i rejected: %i to download' % (rejects, len(maglist)) logger.info(msg) for magazine in maglist: if magazine['nzbmode'] in ["torznab", "torrent", "magnet"]: snatch, res = TORDownloadMethod( magazine['bookid'], magazine['nzbtitle'], magazine['nzburl'], 'magazine') elif magazine['nzbmode'] == 'direct': snatch, res = DirectDownloadMethod( magazine['bookid'], magazine['nzbtitle'], magazine['nzburl'], 'magazine') elif magazine['nzbmode'] == 'nzb': snatch, res = NZBDownloadMethod( magazine['bookid'], magazine['nzbtitle'], magazine['nzburl'], 'magazine') else: res = 'Unhandled NZBmode [%s] for %s' % (magazine['nzbmode'], magazine["nzburl"]) logger.error(res) snatch = 0 if snatch: logger.info('Downloading %s from %s' % (magazine['nzbtitle'], magazine["nzbprov"])) custom_notify_snatch("%s %s" % (magazine['bookid'], magazine['nzburl'])) notify_snatch("Magazine %s from %s at %s" % (unaccented(magazine['nzbtitle']), magazine["nzbprov"], now())) scheduleJob(action='Start', target='PostProcessor') else: myDB.action('UPDATE wanted SET status="Failed",DLResult=? WHERE NZBurl=?', (res, magazine["nzburl"])) 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()) finally: threading.currentThread().name = "WEBSERVER"
def search_nzb_book(books=None, reset=False): threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "SEARCHNZB" if not lazylibrarian.USE_NZB(): logger.warn('No NEWZNAB/TORZNAB providers set, check config') return myDB = database.DBConnection() searchlist = [] if books is None: # We are performing a backlog search searchbooks = myDB.select('SELECT BookID, AuthorName, Bookname, BookSub, BookAdded from books WHERE Status="Wanted" order by BookAdded desc') else: # The user has added a new book searchbooks = [] for book in books: searchbook = myDB.select('SELECT BookID, AuthorName, BookName, BookSub from books WHERE BookID="%s" \ AND Status="Wanted"' % book['bookid']) for terms in searchbook: searchbooks.append(terms) if len(searchbooks) == 0: logger.debug("NZB search requested for no books or invalid BookID") return else: 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'] searchterm = searchterm + '"' searchlist.append({"bookid": searchbook['BookID'], "bookName": searchbook['BookName'], "bookSub": searchbook['BookSub'], "authorName": searchbook['AuthorName'], "searchterm": searchterm}) if not lazylibrarian.SAB_HOST and not lazylibrarian.NZB_DOWNLOADER_BLACKHOLE and not lazylibrarian.NZBGET_HOST: logger.warn('No download method is set, use SABnzbd/NZBGet or blackhole, check config') 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 not found: logger.debug("NZB Searches for %s returned no results." % book['searchterm']) if found > True: nzb_count = 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')
def search_rss_book(books=None, reset=False): try: threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "SEARCHRSS" if not(lazylibrarian.USE_RSS()): logger.warn('RSS search is disabled') scheduleJob(action='Stop', target='search_rss_book') return myDB = database.DBConnection() searchlist = [] if books is None: # We are performing a backlog search searchbooks = myDB.select( 'SELECT BookID, AuthorName, Bookname, BookSub, BookAdded from books WHERE Status="Wanted" \ order by BookAdded desc') else: # The user has added a new book searchbooks = [] for book in books: searchbook = myDB.select('SELECT BookID, AuthorName, BookName, BookSub from books WHERE BookID="%s" \ AND Status="Wanted"' % book['bookid']) for terms in searchbook: searchbooks.append(terms) if len(searchbooks) == 0: return logger.info('RSS Searching for %i book%s' % (len(searchbooks), plural(len(searchbooks)))) resultlist, nproviders = IterateOverRSSSites() if not nproviders: logger.warn('No rss providers are set, check config') return # No point in continuing dic = {'...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': ''} 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 = 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 as e: logger.error('Unhandled exception in search_rss_book: %s' % traceback.format_exc())
def processDir(reset=False): threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "POSTPROCESS" if not lazylibrarian.DOWNLOAD_DIR or not os.path.isdir( lazylibrarian.DOWNLOAD_DIR): processpath = os.getcwd() else: processpath = lazylibrarian.DOWNLOAD_DIR logger.debug(' Checking [%s] for files to post process' % processpath) try: downloads = os.listdir(processpath) except OSError as why: logger.error('Could not access [%s] directory [%s]' % (processpath, why.strerror)) return myDB = database.DBConnection() snatched = myDB.select('SELECT * from wanted WHERE Status="Snatched"') if len(snatched) == 0: logger.info('Nothing marked as snatched.') scheduleJob(action='Stop', target='processDir') return if len(downloads) == 0: logger.info('No downloads are found. Nothing to process.') return logger.info("Checking %s download%s for %s snatched file%s" % (len(downloads), plural( len(downloads)), len(snatched), plural(len(snatched)))) ppcount = 0 for book in snatched: matches = [] for fname in downloads: if not fname.endswith('.fail'): # has this failed before? # this is to get round differences in torrent filenames. # Torrents aren't always returned with the name we searched for # there might be a better way... if isinstance(fname, str): matchname = fname.decode(lazylibrarian.SYS_ENCODING) else: matchname = fname if ' LL.(' in matchname: matchname = matchname.split(' LL.(')[0] matchtitle = book['NZBtitle'] match = 0 if matchtitle: if ' LL.(' in matchtitle: matchtitle = matchtitle.split(' LL.(')[0] match = fuzz.token_set_ratio(matchtitle, matchname) if match >= lazylibrarian.DLOAD_RATIO: fname = matchname if os.path.isfile(os.path.join(processpath, fname)): # handle single file downloads here... if is_valid_booktype(fname, booktype="book") \ or is_valid_booktype(fname, booktype="mag"): dirname = os.path.join(processpath, os.path.splitext(fname)[0]) if not os.path.exists(dirname): try: os.makedirs(dirname) except OSError as why: logger.debug( 'Failed to create directory %s, %s' % (dirname, why.strerror)) if os.path.exists(dirname): try: shutil.move( os.path.join(processpath, fname), os.path.join(dirname, fname)) fname = os.path.splitext(fname)[0] except Exception as why: logger.debug( "Failed to move file %s to %s, %s" % (fname, dirname, str(why))) if os.path.isdir(os.path.join(processpath, fname)): pp_path = os.path.join(processpath, fname) logger.debug('Found folder (%s%%) %s for %s' % (match, pp_path, book['NZBtitle'])) matches.append([match, pp_path, book]) else: logger.debug('No match (%s%%) %s for %s' % (match, matchname, matchtitle)) else: logger.debug('Skipping %s' % fname) if matches: highest = max(matches, key=lambda x: x[0]) match = highest[0] pp_path = highest[1] book = highest[2] logger.info(u'Best match (%s%%): %s for %s' % (match, pp_path, book['NZBtitle'])) data = myDB.select('SELECT * from books WHERE BookID="%s"' % book['BookID']) if data: authorname = data[0]['AuthorName'] bookname = data[0]['BookName'] if 'windows' in platform.system().lower( ) and '/' in lazylibrarian.EBOOK_DEST_FOLDER: logger.warn('Please check your EBOOK_DEST_FOLDER setting') lazylibrarian.EBOOK_DEST_FOLDER = lazylibrarian.EBOOK_DEST_FOLDER.replace( '/', '\\') # Default destination path, should be allowed change per config file. dest_path = lazylibrarian.EBOOK_DEST_FOLDER.replace( '$Author', authorname).replace('$Title', bookname) global_name = lazylibrarian.EBOOK_DEST_FILE.replace( '$Author', authorname).replace('$Title', bookname) global_name = unaccented(global_name) # dest_path = authorname+'/'+bookname # global_name = bookname + ' - ' + authorname # Remove characters we don't want in the filename BEFORE adding to DESTINATION_DIR # as windows drive identifiers have colon, eg c: but no colons allowed elsewhere? dic = { '<': '', '>': '', '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '', '\'': '' } dest_path = unaccented_str(replace_all(dest_path, dic)) dest_path = os.path.join(lazylibrarian.DESTINATION_DIR, dest_path).encode( lazylibrarian.SYS_ENCODING) else: data = myDB.select('SELECT * from magazines WHERE Title="%s"' % book['BookID']) if data: # AuxInfo was added for magazine release date, normally housed in 'magazines' but if multiple # files are downloading, there will be an error in post-processing, trying to go to the # same directory. mostrecentissue = data[0][ 'IssueDate'] # keep for processing issues arriving out of order # Remove characters we don't want in the filename before (maybe) adding to DESTINATION_DIR # as windows drive identifiers have colon, eg c: but no colons allowed elsewhere? dic = { '<': '', '>': '', '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '', '\'': '' } mag_name = unaccented_str(replace_all(book['BookID'], dic)) # book auxinfo is a cleaned date, eg 2015-01-01 dest_path = lazylibrarian.MAG_DEST_FOLDER.replace( '$IssueDate', book['AuxInfo']).replace('$Title', mag_name) # dest_path = '_Magazines/'+title+'/'+book['AuxInfo'] if lazylibrarian.MAG_RELATIVE: if dest_path[0] not in '._': dest_path = '_' + dest_path dest_path = os.path.join( lazylibrarian.DESTINATION_DIR, dest_path).encode(lazylibrarian.SYS_ENCODING) else: dest_path = dest_path.encode( lazylibrarian.SYS_ENCODING) authorname = None bookname = None global_name = lazylibrarian.MAG_DEST_FILE.replace( '$IssueDate', book['AuxInfo']).replace('$Title', mag_name) global_name = unaccented(global_name) # global_name = book['AuxInfo']+' - '+title else: logger.debug( "Snatched magazine %s is not in download directory" % (book['BookID'])) continue else: logger.debug("Snatched %s %s is not in download directory" % (book['NZBmode'], book['NZBtitle'])) continue processBook = processDestination(pp_path, dest_path, authorname, bookname, global_name) if processBook: logger.debug("Processing %s, %s" % (global_name, book['NZBurl'])) # update nzbs, only update the snatched ones in case multiple matches for same book / magazine issue controlValueDict = {"NZBurl": book['NZBurl'], "Status": "Snatched"} newValueDict = { "Status": "Processed", "NZBDate": now() } # say when we processed it myDB.upsert("wanted", newValueDict, controlValueDict) if bookname is not None: # it's a book, if None it's a magazine if len(lazylibrarian.IMP_CALIBREDB): logger.debug( 'Calibre should have created the extras for us') else: processExtras(myDB, dest_path, global_name, data) else: # update mags controlValueDict = {"Title": book['BookID']} if mostrecentissue: if mostrecentissue.isdigit() and str( book['AuxInfo']).isdigit(): older = int(mostrecentissue) > int( book['AuxInfo']) # issuenumber else: older = mostrecentissue > book['AuxInfo'] # YYYY-MM-DD else: older = False if older: # check this in case processing issues arriving out of order newValueDict = { "LastAcquired": today(), "IssueStatus": "Open" } else: newValueDict = { "IssueDate": book['AuxInfo'], "LastAcquired": today(), "IssueStatus": "Open" } myDB.upsert("magazines", newValueDict, controlValueDict) # dest_path is where we put the magazine after processing, but we don't have the full filename # so look for any "book" in that directory dest_file = book_file(dest_path, booktype='mag') controlValueDict = { "Title": book['BookID'], "IssueDate": book['AuxInfo'] } newValueDict = { "IssueAcquired": today(), "IssueFile": dest_file, "IssueID": create_id("%s %s" % (book['BookID'], book['AuxInfo'])) } myDB.upsert("issues", newValueDict, controlValueDict) # create a thumbnail cover for the new issue create_cover(dest_file) logger.info('Successfully processed: %s' % global_name) ppcount = ppcount + 1 notify_download("%s at %s" % (global_name, now())) else: logger.error('Postprocessing for %s has failed.' % global_name) logger.error('Warning - Residual files remain in %s.fail' % pp_path) controlValueDict = {"NZBurl": book['NZBurl'], "Status": "Snatched"} newValueDict = {"Status": "Failed", "NZBDate": now()} myDB.upsert("wanted", newValueDict, controlValueDict) # if it's a book, reset status so we try for a different version # if it's a magazine, user can select a different one from pastissues table if bookname is not None: myDB.action( 'UPDATE books SET status = "Wanted" WHERE BookID="%s"' % book['BookID']) # at this point, as it failed we should move it or it will get postprocessed # again (and fail again) try: os.rename(pp_path, pp_path + '.fail') except: logger.debug("Unable to rename %s" % pp_path) downloads = os.listdir( processpath) # check in case we processed/deleted some above for directory in downloads: if "LL.(" in directory and not directory.endswith('.fail'): bookID = str(directory).split("LL.(")[1].split(")")[0] logger.debug("Book with id: " + str(bookID) + " is in downloads") pp_path = os.path.join(processpath, directory) if os.path.isfile(pp_path): pp_path = os.path.join(processpath) if (os.path.isdir(pp_path)): logger.debug('Found LL folder %s.' % pp_path) if import_book(pp_path, bookID): ppcount = ppcount + 1 if ppcount == 0: logger.info('No snatched books/mags have been found') else: logger.info('%s book%s/mag%s processed.' % (ppcount, plural(ppcount), plural(ppcount))) if reset: scheduleJob(action='Restart', target='processDir')
def search_wishlist(): if not (lazylibrarian.USE_RSS()): logger.warn('RSS search is disabled') scheduleJob(action='Stop', target='search_wishlist') return try: threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "SEARCHWISHLIST" myDB = database.DBConnection() resultlist, wishproviders = IterateOverWishLists() new_books = 0 if not wishproviders: logger.debug('No wishlists are set') scheduleJob(action='Stop', target='search_wishlist') return # No point in continuing # for each item in resultlist, add to database if necessary, and mark as wanted logger.debug('Processing %s item%s in wishlists' % (len(resultlist), plural(len(resultlist)))) for book in resultlist: # we get rss_author, rss_title, maybe 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... if lazylibrarian.CONFIG['BOOK_API'] == "GoodReads" and book['rss_bookid']: bookmatch = myDB.match('select Status,BookName from books where bookid=?', (book['rss_bookid'],)) if bookmatch: bookstatus = bookmatch['Status'] bookname = bookmatch['BookName'] if bookstatus in ['Open', 'Wanted', 'Have']: logger.info('Found book %s, already marked as "%s"' % (bookname, bookstatus)) else: # skipped/ignored logger.info('Found book %s, marking as "Wanted"' % bookname) controlValueDict = {"BookID": book['rss_bookid']} newValueDict = {"Status": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) new_books += 1 else: import_book(book['rss_bookid']) new_books += 1 else: item = {} results = None item['Title'] = book['rss_title'] if book['rss_bookid']: item['BookID'] = book['rss_bookid'] if book['rss_isbn']: item['ISBN'] = book['rss_isbn'] bookmatch = finditem(item, book['rss_author']) 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( 'Found book %s by %s, already marked as "%s"' % (bookname, authorname, bookstatus)) else: # skipped/ignored logger.info('Found book %s by %s, marking as "Wanted"' % (bookname, authorname)) controlValueDict = {"BookID": bookid} newValueDict = {"Status": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) new_books += 1 else: # not in database yet if book['rss_isbn']: results = search_for(book['rss_isbn']) if results: result = results[0] # type: dict if result['isbn_fuzz'] > check_int(lazylibrarian.CONFIG['MATCH_RATIO'], 90): logger.info("Found (%s%%) %s: %s" % (result['isbn_fuzz'], result['authorname'], result['bookname'])) import_book(result['bookid']) new_books += 1 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] # type: dict if result['author_fuzz'] > check_int(lazylibrarian.CONFIG['MATCH_RATIO'], 90) \ and result['book_fuzz'] > check_int(lazylibrarian.CONFIG['MATCH_RATIO'], 90): logger.info("Found (%s%% %s%%) %s: %s" % (result['author_fuzz'], result['book_fuzz'], result['authorname'], result['bookname'])) import_book(result['bookid']) new_books += 1 bookmatch = True if not bookmatch: msg = "Skipping book %s by %s" % (item['Title'], book['rss_author']) if not results: msg += ', No results returned' logger.warn(msg) else: msg += ', No match found' logger.warn(msg) result = results[0] # type: dict msg = "Closest match (%s%% %s%%) %s: %s" % (result['author_fuzz'], result['book_fuzz'], result['authorname'], result['bookname']) logger.warn(msg) if new_books: logger.info("Wishlist marked %s book%s as Wanted" % (new_books, plural(new_books))) except Exception: logger.error('Unhandled exception in search_wishlist: %s' % traceback.format_exc()) finally: threading.currentThread().name = "WEBSERVER"
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 processResultList(resultlist, authorname, bookname, book, searchtype): myDB = database.DBConnection() dictrepl = { '...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': ' ', '*': '', '(': '', ')': '', '[': '', ']': '', '#': '', '0': '', '1': '', '2': '', '3': '', '4': '', '5': '', '6': '', '7': '', '8': '', '9': '', '\'': '', ':': '', '!': '', '-': ' ', '\s\s': ' ' } match_ratio = int(lazylibrarian.MATCH_RATIO) reject_list = getList(lazylibrarian.REJECT_WORDS) matches = [] # bit of a misnomer now, rss can search both tor and nzb rss feeds for tor in resultlist: torTitle = unaccented_str(replace_all(tor['tor_title'], dictrepl)).strip() torTitle = re.sub(r"\s\s+", " ", torTitle) # remove extra whitespace tor_Author_match = fuzz.token_set_ratio(authorname, torTitle) tor_Title_match = fuzz.token_set_ratio(bookname, torTitle) logger.debug("RSS Author/Title Match: %s/%s for %s" % (tor_Author_match, tor_Title_match, torTitle)) tor_url = tor['tor_url'] rejected = False already_failed = myDB.match( 'SELECT * from wanted WHERE NZBurl="%s" and Status="Failed"' % tor_url) if already_failed: logger.debug("Rejecting %s, blacklisted at %s" % (torTitle, already_failed['NZBprov'])) rejected = True if not rejected: for word in reject_list: if word in torTitle.lower() and word not in authorname.lower( ) and word not in bookname.lower(): rejected = True logger.debug("Rejecting %s, contains %s" % (torTitle, word)) break tor_size_temp = tor[ 'tor_size'] # Need to cater for when this is NONE (Issue 35) if tor_size_temp is None: tor_size_temp = 1000 tor_size = round(float(tor_size_temp) / 1048576, 2) maxsize = check_int(lazylibrarian.REJECT_MAXSIZE, 0) if not rejected: if maxsize and tor_size > maxsize: rejected = True logger.debug("Rejecting %s, too large" % torTitle) if not rejected: bookid = book['bookid'] tor_Title = (book["authorName"] + ' - ' + book['bookName'] + ' LL.(' + book['bookid'] + ')').strip() tor_prov = tor['tor_prov'] tor_feed = tor['tor_feed'] controlValueDict = {"NZBurl": tor_url} newValueDict = { "NZBprov": tor_prov, "BookID": bookid, "NZBdate": now(), # when we asked for it "NZBsize": tor_size, "NZBtitle": tor_Title, "NZBmode": "torrent", "Status": "Skipped" } score = (tor_Title_match + tor_Author_match) / 2 # as a percentage # lose a point for each extra word in the title so we get the closest match words = len(getList(torTitle)) words -= len(getList(authorname)) words -= len(getList(bookname)) score -= abs(words) matches.append([score, torTitle, newValueDict, controlValueDict]) if matches: highest = max(matches, key=lambda x: x[0]) score = highest[0] nzb_Title = highest[1] newValueDict = highest[2] controlValueDict = highest[3] if score < match_ratio: logger.debug( u'Nearest RSS match (%s%%): %s using %s search for %s %s' % (score, nzb_Title, searchtype, authorname, bookname)) return False logger.info(u'Best RSS match (%s%%): %s using %s search' % (score, nzb_Title, searchtype)) snatchedbooks = myDB.match( 'SELECT * from books WHERE BookID="%s" and Status="Snatched"' % newValueDict["BookID"]) if snatchedbooks: # check if one of the other downloaders got there first logger.info('%s already marked snatched' % nzb_Title) return True else: myDB.upsert("wanted", newValueDict, controlValueDict) tor_url = controlValueDict["NZBurl"] if '.nzb' in tor_url: snatch = NZBDownloadMethod(newValueDict["BookID"], newValueDict["NZBprov"], newValueDict["NZBtitle"], controlValueDict["NZBurl"]) else: """ # http://baconbits.org/torrents.php?action=download&authkey=<authkey>&torrent_pass=<password.hashed>&id=185398 if not tor_url.startswith('magnet'): # magnets don't use auth pwd = lazylibrarian.RSS_PROV[tor_feed]['PASS'] auth = lazylibrarian.RSS_PROV[tor_feed]['AUTH'] # don't know what form of password hash is required, try sha1 tor_url = tor_url.replace('<authkey>', auth).replace('<password.hashed>', sha1(pwd)) """ snatch = TORDownloadMethod(newValueDict["BookID"], newValueDict["NZBprov"], newValueDict["NZBtitle"], tor_url) if snatch: logger.info( 'Downloading %s from %s' % (newValueDict["NZBtitle"], newValueDict["NZBprov"])) notify_snatch( "%s from %s at %s" % (newValueDict["NZBtitle"], newValueDict["NZBprov"], now())) scheduleJob(action='Start', target='processDir') return True + True # we found it else: logger.debug("No RSS found for " + (book["authorName"] + ' ' + book['bookName']).strip()) return False
def processResultList(resultlist, book, searchtype): myDB = database.DBConnection() dictrepl = {'...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': ' ', '*': '', '(': '', ')': '', '[': '', ']': '', '#': '', '0': '', '1': '', '2': '', '3': '', '4': '', '5': '', '6': '', '7': '', '8': '', '9': '', '\'': '', ':': '', '!': '', '-': ' ', '\s\s': ' '} # ' the ': ' ', ' a ': ' ', ' and ': ' ', # ' to ': ' ', ' of ': ' ', ' for ': ' ', ' my ': ' ', ' in ': ' ', ' at ': ' ', ' with ': ' '} dic = {'...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '', '\'': ''} match_ratio = int(lazylibrarian.MATCH_RATIO) reject_list = getList(lazylibrarian.REJECT_WORDS) author = unaccented_str(replace_all(book['authorName'], dic)) title = unaccented_str(replace_all(book['bookName'], dic)) matches = [] for nzb in resultlist: nzb_Title = unaccented_str(replace_all(nzb['nzbtitle'], dictrepl)).strip() nzb_Title = re.sub(r"\s\s+", " ", nzb_Title) # remove extra whitespace nzbAuthor_match = fuzz.token_set_ratio(author, nzb_Title) nzbBook_match = fuzz.token_set_ratio(title, nzb_Title) logger.debug(u"NZB author/book Match: %s/%s for %s" % (nzbAuthor_match, nzbBook_match, nzb_Title)) nzburl = nzb['nzburl'] rejected = False already_failed = myDB.action('SELECT * from wanted WHERE NZBurl="%s" and Status="Failed"' % nzburl).fetchone() if already_failed: logger.debug("Rejecting %s, blacklisted at %s" % (nzb_Title, already_failed['NZBprov'])) rejected = True if not rejected: for word in reject_list: if word in nzb_Title.lower() and not word in author.lower() and not word in title.lower(): rejected = True logger.debug("Rejecting %s, contains %s" % (nzb_Title, word)) break nzbsize_temp = nzb['nzbsize'] # Need to cater for when this is NONE (Issue 35) if nzbsize_temp is None: nzbsize_temp = 1000 nzbsize = round(float(nzbsize_temp) / 1048576, 2) maxsize = check_int(lazylibrarian.REJECT_MAXSIZE, 0) if not rejected: if maxsize and nzbsize > maxsize: rejected = True logger.debug("Rejecting %s, too large" % nzb_Title) if not rejected: if nzbAuthor_match >= match_ratio and nzbBook_match >= match_ratio: bookid = book['bookid'] nzbTitle = (author + ' - ' + title + ' LL.(' + book['bookid'] + ')').strip() nzbprov = nzb['nzbprov'] nzbdate_temp = nzb['nzbdate'] nzbdate = nzbdate2format(nzbdate_temp) nzbmode = nzb['nzbmode'] controlValueDict = {"NZBurl": nzburl} newValueDict = { "NZBprov": nzbprov, "BookID": bookid, "NZBdate": now(), # when we asked for it "NZBsize": nzbsize, "NZBtitle": nzbTitle, "NZBmode": nzbmode, "Status": "Skipped" } score = (nzbBook_match + nzbAuthor_match)/2 # as a percentage # lose a point for each extra word in the title so we get the closest match words = len(getList(nzb_Title)) words -= len(getList(author)) words -= len(getList(title)) score -= abs(words) matches.append([score, nzb_Title, newValueDict, controlValueDict]) if matches: highest = max(matches, key=lambda x: x[0]) score = highest[0] nzb_Title = highest[1] newValueDict = highest[2] controlValueDict = highest[3] logger.info(u'Best match NZB (%s%%): %s using %s search' % (score, nzb_Title, searchtype)) snatchedbooks = myDB.action('SELECT * from books WHERE BookID="%s" and Status="Snatched"' % newValueDict["BookID"]).fetchone() if snatchedbooks: logger.debug('%s already marked snatched' % nzb_Title) return True # someone else found it else: logger.debug('%s adding to wanted' % nzb_Title) myDB.upsert("wanted", newValueDict, controlValueDict) if nzbmode == "torznab": snatch = TORDownloadMethod(newValueDict["BookID"], newValueDict["NZBprov"], newValueDict["NZBtitle"], controlValueDict["NZBurl"]) else: snatch = NZBDownloadMethod(newValueDict["BookID"], newValueDict["NZBprov"], newValueDict["NZBtitle"], controlValueDict["NZBurl"]) if snatch: notifiers.notify_snatch(newValueDict["NZBtitle"] + ' at ' + now()) scheduleJob(action='Start', target='processDir') return True + True # we found it else: logger.debug("No nzb's found for [%s] using searchtype %s" % (book["searchterm"], searchtype)) return False
def search_nzb_book(books=None, reset=False): threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "SEARCHNZB" if not lazylibrarian.USE_NZB(): logger.warn('No NEWZNAB/TORZNAB providers set, check config') return myDB = database.DBConnection() searchlist = [] if books is None: # We are performing a backlog search searchbooks = myDB.select( 'SELECT BookID, AuthorName, Bookname, BookSub, BookAdded from books WHERE Status="Wanted" order by BookAdded desc' ) else: # The user has added a new book searchbooks = [] for book in books: searchbook = myDB.select( 'SELECT BookID, AuthorName, BookName, BookSub from books WHERE BookID="%s" \ AND Status="Wanted"' % book['bookid']) for terms in searchbook: searchbooks.append(terms) if len(searchbooks) == 0: logger.debug("NZB search requested for no books or invalid BookID") return else: 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'] searchterm = searchterm + '"' searchlist.append({ "bookid": searchbook['BookID'], "bookName": searchbook['BookName'], "bookSub": searchbook['BookSub'], "authorName": searchbook['AuthorName'], "searchterm": searchterm }) if not lazylibrarian.SAB_HOST and not lazylibrarian.NZB_DOWNLOADER_BLACKHOLE and not lazylibrarian.NZBGET_HOST: logger.warn( 'No download method is set, use SABnzbd/NZBGet or blackhole, check config' ) 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 not found: logger.debug("NZB Searches for %s returned no results." % book['searchterm']) if found > True: nzb_count = 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')
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: threading.currentThread().name = "SEARCHMAG" 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 searchlist == []: logger.warn('There is nothing to search for. Mark some magazines as active.') for book in searchlist: resultlist = [] tor_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(book, 'mag') 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 total_nzbs = 0 new_date = 0 maglist = [] issues = [] for nzb in resultlist: total_nzbs = 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 = bad_name + 1 else: rejected = False maxsize = check_int(lazylibrarian.REJECT_MAGSIZE, 0) if maxsize and nzbsize > maxsize: logger.debug("Rejecting %s, too large" % nzbtitle) rejected = True if not rejected: control_date = results['IssueDate'] reject_list = getList(results['Reject']) 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 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 starts with magazine title followed by a date # eg The MagPI Issue 22 - 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.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: 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: 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 if not rejected: # some magazine torrent uploaders add their sig in [] or {} # Fortunately for us, they always seem to add it at the end # also some magazine torrent titles are "magazine_name some_form_of_date pdf" # or other words we don't want. Should make the word list configurable. # so strip all the trailing junk... strip_list = ['pdf', 'true', 'truepdf', 'german', 'ebooks'] while nzbtitle_exploded[len(nzbtitle_exploded) - 1][0] in '[{' or \ nzbtitle_exploded[len(nzbtitle_exploded) - 1].lower() in strip_list: nzbtitle_exploded.pop() # gotta love the function names # need at least one word magazine title and two date components if len(nzbtitle_exploded) > 2: # regexA = DD MonthName YYYY OR MonthName YYYY or Issue nn, MonthName YYYY regexA_year = nzbtitle_exploded[len(nzbtitle_exploded) - 1] regexA_month_temp = nzbtitle_exploded[len(nzbtitle_exploded) - 2] regexA_month = month2num(unaccented(regexA_month_temp)) if not regexA_year.isdigit() or int(regexA_year) < 1900 or int(regexA_year) > 2100: regexA_year = 'fail' # force date failure # if frequency == "Weekly" or frequency == "BiWeekly": regexA_day = nzbtitle_exploded[len(nzbtitle_exploded) - 3].rstrip(',').zfill(2) if regexA_day.isdigit(): if int(regexA_day) > 31: # probably issue number nn regexA_day = '01' else: regexA_day = '01' # just MonthName YYYY # else: # regexA_day = '01' # monthly, or less frequent try: newdatish = regexA_year + '-' + regexA_month + '-' + regexA_day # try to make sure the year/month/day are valid, exception if not # ie don't accept day > 31, or 30 in some months, or month <1 or >12 # also handles multiple date format named issues eg Jan 2014, 01 2014 # datetime will give a ValueError if not a good date or a param is not int date1 = datetime.date(int(regexA_year), int(regexA_month), int(regexA_day)) except ValueError: # regexB = MonthName DD YYYY or MonthName DD, YYYY regexB_year = nzbtitle_exploded[len(nzbtitle_exploded) - 1] regexB_month_temp = nzbtitle_exploded[len(nzbtitle_exploded) - 3] regexB_month = month2num(unaccented(regexB_month_temp)) regexB_day = nzbtitle_exploded[len(nzbtitle_exploded) - 2].rstrip(',').zfill(2) if not regexB_year.isdigit() or int(regexB_year) < 1900 or int(regexB_year) > 2100: regexB_year = 'fail' try: newdatish = regexB_year + '-' + regexB_month + '-' + regexB_day # datetime will give a ValueError if not a good date or a param is not int date1 = datetime.date(int(regexB_year), int(regexB_month), int(regexB_day)) except ValueError: # regexC = YYYY MM or YYYY MM DD # (can't get MM/DD if named YYYY Issue nn) # First try YYYY MM regexC_year = nzbtitle_exploded[len(nzbtitle_exploded) - 2] if regexC_year.isdigit() and int(regexC_year) > 1900 and int(regexC_year) < 2100: regexC_month = nzbtitle_exploded[len(nzbtitle_exploded) - 1].zfill(2) regexC_day = '01' else: # try YYYY MM DD regexC_year = nzbtitle_exploded[len(nzbtitle_exploded) - 3] regexC_month = 0 regexC_day = 0 if regexC_year.isdigit() and int(regexC_year) > 1900 and int(regexC_year) < 2100: regexC_month = nzbtitle_exploded[len(nzbtitle_exploded) - 2].zfill(2) regexC_day = nzbtitle_exploded[len(nzbtitle_exploded) - 1].zfill(2) else: regexC_year = 'fail' try: newdatish = regexC_year + '-' + regexC_month + '-' + regexC_day # datetime will give a ValueError if not a good date or a param is not int date1 = datetime.date(int(regexC_year), int(regexC_month), int(regexC_day)) except Exception: # regexD Issue/No/Nr/Vol nn, YYYY or Issue/No/Nr/Vol nn try: IssueLabel = nzbtitle_exploded[len(nzbtitle_exploded) - 2] if IssueLabel.lower() in ["issue", "no", "nr", "vol"]: # issue nn regexD_issue = nzbtitle_exploded[len(nzbtitle_exploded) - 1] if regexD_issue.isdigit(): newdatish = str(int(regexD_issue)) # 4 == 04 == 004 else: IssueLabel = nzbtitle_exploded[len(nzbtitle_exploded) - 3] if IssueLabel.lower() in ["issue", "no", "nr", "vol"]: # issue nn, YYYY regexD_issue = nzbtitle_exploded[len(nzbtitle_exploded) - 2] regexD_issue = regexD_issue.strip(',') if regexD_issue.isdigit(): newdatish = str(int(regexD_issue)) # 4 == 04 == 004 else: raise ValueError regexD_year = nzbtitle_exploded[len(nzbtitle_exploded) - 1] if regexD_year.isdigit(): if int(regexD_year) < int(datetime.date.today().year): newdatish = 0 # it's old else: raise ValueError except Exception: # regexE nn YYYY issue number without "Nr" before it # nn is assumed not to be a month as they are normally names not digits try: regexE_year = nzbtitle_exploded[len(nzbtitle_exploded) - 1] regexE_issue = nzbtitle_exploded[len(nzbtitle_exploded) - 2] print "[%s][%s]" % (regexE_year, regexE_issue) if regexE_issue.isdigit(): newdatish = int(regexE_issue) if int(regexE_year) < int(datetime.date.today().year): newdatish = 0 # it's old else: raise ValueError except Exception: # regexF issue and year as a single 6 digit string eg 222015 try: regexF = nzbtitle_exploded[len(nzbtitle_exploded) - 1] if regexF.isdigit() and len(regexF) == 6: regexF_issue = regexF[:2] regexF_year = regexF[2:] newdatish = str(int(regexF_issue)) # 4 == 04 == 004 if int(regexF_year) < int(datetime.date.today().year): newdatish = 0 # it's old else: raise ValueError except Exception: logger.debug('Magazine %s not in a recognised date format.' % nzbtitle_formatted) bad_date = 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 # continue else: logger.debug('Magazine [%s] does not match the search term [%s].' % ( nzbtitle_formatted, bookid)) bad_name = bad_name + 1 continue # 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.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 = 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 = 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 = 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.select('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) else: # logger.debug('Magazine [%s] was rejected.' % nzbtitle_formatted) bad_name = bad_name + 1 logger.info('Found %i result%s for %s. %i new, %i old, %i fail date, %i fail name: %i to download' % ( total_nzbs, plural(total_nzbs), bookid, new_date, old_date, bad_date, bad_name, len(maglist))) for magazine in maglist: if magazine['nzbmode'] in ["torznab", "torrent", "magnet"]: snatch = TORDownloadMethod( magazine['bookid'], magazine['nzbprov'], magazine['nzbtitle'], magazine['nzburl']) else: snatch = NZBDownloadMethod( magazine['bookid'], magazine['nzbprov'], 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())) scheduleJob(action='Start', target='processDir') maglist = [] if reset: scheduleJob(action='Restart', target='search_magazines') logger.info("Search for magazines complete") except Exception as e: logger.error('Unhandled exception in search_magazines: %s' % traceback.format_exc())
def search_magazines(mags=None, reset=False): # produce a list of magazines to search for, tor, nzb, torznab, rss threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "SEARCHMAG" myDB = database.DBConnection() searchlist = [] if mags is None: # backlog search searchmags = myDB.select('SELECT Title, LastAcquired, \ IssueDate from magazines WHERE Status="Active"' ) else: searchmags = [] for magazine in mags: searchmags_temp = myDB.select( 'SELECT Title, 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['Title'] # frequency = searchmag[1] # last_acquired = searchmag[2] # issue_date = searchmag[3] dic = { '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '' } searchterm = unaccented_str(replace_all(searchterm, dic)) searchterm = re.sub('[\.\-\/]', ' ', searchterm).encode(lazylibrarian.SYS_ENCODING) searchlist.append({"bookid": bookid, "searchterm": searchterm}) if searchlist == []: logger.warn( 'There is nothing to search for. Mark some magazines as active.') for book in searchlist: resultlist = [] tor_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(book, 'mag') 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_regex = 0 bad_date = 0 old_date = 0 total_nzbs = 0 new_date = 0 maglist = [] issues = [] for nzb in resultlist: total_nzbs = 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'] if nzbsize_temp is None: # not all torrents returned by torznab have a size nzbsize_temp = 1000 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_regex = bad_regex + 1 else: control_date = results['IssueDate'] reject_list = getList(results['Regex']) 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 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 starts with magazine title, and ends with a date # eg The MagPI Issue 22 - July 2015 # do something like check left n words match title # then check last n words are a date rejected = False 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.MATCH_RATIO: logger.debug(u"Magazine token set Match failed: " + str(mag_title_match) + "% for " + nzbtitle_formatted) rejected = True else: 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: 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 # maxsize = formatter.check_int(lazylibrarian.REJECT_MAXSIZE, 0) # if maxsize and nzbsize > maxsize: # rejected = True # logger.debug("Rejecting %s, too large" % nzbtitle_formatted) if not rejected: # some magazine torrent uploaders add their sig in [] or {} # Fortunately for us, they always seem to add it at the end # also some magazine torrent titles are "magazine_name some_form_of_date pdf" # so strip all the trailing junk... while nzbtitle_exploded[len(nzbtitle_exploded) - 1][0] in '[{' or \ nzbtitle_exploded[len(nzbtitle_exploded) - 1].lower() in ['pdf', 'true', 'truepdf']: nzbtitle_exploded.pop( ) # gotta love the function names # need at least one word magazine title and two date components if len(nzbtitle_exploded) > 2: # regexA = DD MonthName YYYY OR MonthName YYYY or Issue nn, MonthName YYYY regexA_year = nzbtitle_exploded[ len(nzbtitle_exploded) - 1] regexA_month_temp = nzbtitle_exploded[ len(nzbtitle_exploded) - 2] regexA_month = month2num( unaccented(regexA_month_temp)) if not regexA_year.isdigit() or int( regexA_year) < 1900 or int( regexA_year) > 2100: regexA_year = 'fail' # force date failure # if frequency == "Weekly" or frequency == "BiWeekly": regexA_day = nzbtitle_exploded[ len(nzbtitle_exploded) - 3].rstrip(',').zfill(2) if regexA_day.isdigit(): if int(regexA_day ) > 31: # probably issue number nn regexA_day = '01' else: regexA_day = '01' # just MonthName YYYY # else: # regexA_day = '01' # monthly, or less frequent try: newdatish = regexA_year + '-' + regexA_month + '-' + regexA_day # try to make sure the year/month/day are valid, exception if not # ie don't accept day > 31, or 30 in some months # also handles multiple date format named issues eg Jan 2014, 01 2014 # datetime will give a ValueError if not a good date or a param is not int date1 = datetime.date(int(regexA_year), int(regexA_month), int(regexA_day)) except ValueError: # regexB = MonthName DD YYYY or MonthName DD, YYYY regexB_year = nzbtitle_exploded[ len(nzbtitle_exploded) - 1] regexB_month_temp = nzbtitle_exploded[ len(nzbtitle_exploded) - 3] regexB_month = month2num( unaccented(regexB_month_temp)) regexB_day = nzbtitle_exploded[ len(nzbtitle_exploded) - 2].rstrip(',').zfill(2) if not regexB_year.isdigit() or int( regexB_year) < 1900 or int( regexB_year) > 2100: regexB_year = 'fail' try: newdatish = regexB_year + '-' + regexB_month + '-' + regexB_day # datetime will give a ValueError if not a good date or a param is not int date1 = datetime.date( int(regexB_year), int(regexB_month), int(regexB_day)) except ValueError: # regexC = YYYY MM or YYYY MM DD # (can't get MM/DD if named YYYY Issue nn) # First try YYYY MM regexC_year = nzbtitle_exploded[ len(nzbtitle_exploded) - 2] if regexC_year.isdigit( ) and int(regexC_year) > 1900 and int( regexC_year) < 2100: regexC_month = nzbtitle_exploded[ len(nzbtitle_exploded) - 1].zfill(2) regexC_day = '01' else: # try YYYY MM DD regexC_year = nzbtitle_exploded[ len(nzbtitle_exploded) - 3] regexC_month = 0 regexC_day = 0 if regexC_year.isdigit( ) and int(regexC_year) > 1900 and int( regexC_year) < 2100: regexC_month = nzbtitle_exploded[ len(nzbtitle_exploded) - 2].zfill(2) regexC_day = nzbtitle_exploded[ len(nzbtitle_exploded) - 1].zfill(2) else: regexC_year = 'fail' try: newdatish = regexC_year + '-' + regexC_month + '-' + regexC_day # datetime will give a ValueError if not a good date or a param is not int date1 = datetime.date( int(regexC_year), int(regexC_month), int(regexC_day)) except Exception: # regexD Issue/No/Vol nn, YYYY or Issue/No/Vol nn try: IssueLabel = nzbtitle_exploded[ len(nzbtitle_exploded) - 2] if IssueLabel.lower() in [ "issue", "no", "vol" ]: # issue nn regexD_issue = nzbtitle_exploded[ len(nzbtitle_exploded) - 1] if regexD_issue.isdigit(): newdatish = str( int(regexD_issue) ) # 4 == 04 == 004 else: IssueLabel = nzbtitle_exploded[ len(nzbtitle_exploded) - 3] if IssueLabel.lower() in [ "issue", "no", "vol" ]: # issue nn, YYYY regexD_issue = nzbtitle_exploded[ len(nzbtitle_exploded) - 2] regexD_issue = regexD_issue.strip( ',') if regexD_issue.isdigit(): newdatish = str( int(regexD_issue) ) # 4 == 04 == 004 else: raise ValueError regexD_year = nzbtitle_exploded[ len(nzbtitle_exploded) - 1] if regexD_year.isdigit(): if int( regexD_year ) < int(datetime.date. today().year): newdatish = 0 # it's old else: raise ValueError except Exception: logger.debug( 'Magazine %s not in proper date format.' % nzbtitle_formatted) bad_date = 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 # continue else: logger.debug( 'Magazine [%s] does not match the search term [%s].' % (nzbtitle_formatted, bookid)) bad_regex = bad_regex + 1 continue # 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 -= 31 * 24 * 60 * 60 # number of seconds in 31 days control_date = time.strftime( "%Y-%m-%d", time.localtime(start_time)) 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 = 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 = 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 = 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.select( '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) else: # logger.debug('Magazine [%s] was rejected.' % nzbtitle_formatted) bad_regex = bad_regex + 1 logger.info( 'Found %i result%s for %s. %i new, %i old, %i fail date, %i fail name: %i to download' % (total_nzbs, plural(total_nzbs), bookid, new_date, old_date, bad_date, bad_regex, len(maglist))) for magazine in maglist: if magazine['nzbmode'] in ["torznab", "torrent", "magnet"]: snatch = TORDownloadMethod(magazine['bookid'], magazine['nzbprov'], magazine['nzbtitle'], magazine['nzburl']) else: snatch = NZBDownloadMethod(magazine['bookid'], magazine['nzbprov'], 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())) scheduleJob(action='Start', target='processDir') maglist = [] if reset: scheduleJob(action='Restart', target='search_magazines') logger.info("Search for magazines complete")
def search_magazines(mags=None, reset=False): # produce a list of magazines to search for, tor, nzb, torznab, rss # noinspection PyBroadException try: threadname = threading.currentThread().name if "Thread-" in threadname: if mags is None: threading.currentThread().name = "SEARCHALLMAG" else: threading.currentThread().name = "SEARCHMAG" myDB = database.DBConnection() searchlist = [] if mags is None: # backlog search searchmags = myDB.select('SELECT Title, Regex, DateType, LastAcquired, \ IssueDate from magazines WHERE Status="Active"') else: searchmags = [] for magazine in mags: searchmags_temp = myDB.select('SELECT Title,Regex,DateType,LastAcquired,IssueDate from magazines \ WHERE Title=? AND Status="Active"', (magazine['bookid'],)) for terms in searchmags_temp: searchmags.append(terms) if len(searchmags) == 0: threading.currentThread().name = "WEBSERVER" 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("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'] datetype = searchmag['DateType'] if not datetype: datetype = '' if not searchterm: dic = {'...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': ''} # strip accents from the magazine title for easier name-matching searchterm = unaccented_str(searchmag['Title']) if not searchterm: # unless there are no ascii characters left searchterm = searchmag['Title'] searchterm = replace_all(searchterm, dic) searchterm = re.sub('[.\-/]', ' ', searchterm) if PY2: searchterm = searchterm.encode(lazylibrarian.SYS_ENCODING) searchlist.append({"bookid": bookid, "searchterm": searchterm, "datetype": datetype}) 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: # don't nag. Show warning message no more than every 20 mins timenow = int(time.time()) if check_int(lazylibrarian.NO_NZB_MSG, 0) + 1200 < timenow: logger.warn('No nzb providers are available. Check config and blocklist') lazylibrarian.NO_NZB_MSG = timenow if lazylibrarian.USE_DIRECT(): dir_resultlist, nproviders = IterateOverDirectSites(book, 'mag') if not nproviders: # don't nag. Show warning message no more than every 20 mins timenow = int(time.time()) if check_int(lazylibrarian.NO_DIRECT_MSG, 0) + 1200 < timenow: logger.warn('No direct providers are available. Check config and blocklist') lazylibrarian.NO_DIRECT_MSG = timenow if dir_resultlist: for item in dir_resultlist: # reformat the 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 'nzbsize': item['tor_size'], 'nzbmode': 'torrent' }) if lazylibrarian.USE_TOR(): tor_resultlist, nproviders = IterateOverTorrentSites(book, 'mag') if not nproviders: # don't nag. Show warning message no more than every 20 mins timenow = int(time.time()) if check_int(lazylibrarian.NO_TOR_MSG, 0) + 1200 < timenow: logger.warn('No tor providers are available. Check config and blocklist') lazylibrarian.NO_TOR_MSG = timenow 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: # don't nag. Show warning message no more than every 20 mins timenow = int(time.time()) if check_int(lazylibrarian.NO_RSS_MSG, 0) + 1200 < timenow: logger.warn('No rss providers are available. Check config and blocklist') lazylibrarian.NO_RSS_MSG = timenow 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("No results for magazine %s" % 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'] # strip accents from the magazine title for easier name-matching nzbtitle = unaccented_str(nzb['nzbtitle']) if not nzbtitle: # unless it's not a latin-1 encodable name nzbtitle = 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'] # Need to make sure that substrings of magazine titles don't get found # (e.g. Maxim USA will find Maximum PC USA) so split into "words" dic = {'.': ' ', '-': ' ', '/': ' ', '+': ' ', '_': ' ', '(': '', ')': '', '[': ' ', ']': ' ', '#': '# '} nzbtitle_formatted = replace_all(nzbtitle, dic).strip() # remove extra spaces if they're in a row nzbtitle_formatted = " ".join(nzbtitle_formatted.split()) nzbtitle_exploded = nzbtitle_formatted.split(' ') results = myDB.match('SELECT * from magazines WHERE Title=?', (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: 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 all the words in the mag title are in the nzbtitle rejected = False wlist = [] for word in nzbtitle_exploded: word = unaccented(word).lower() if word: wlist.append(word) for word in bookid_exploded: word = unaccented(word).lower() if word and word not in wlist: logger.debug("Rejecting %s, missing %s" % (nzbtitle, word)) rejected = True break if rejected: logger.debug( "Magazine title match failed " + bookid + " for " + nzbtitle_formatted) else: logger.debug( "Magazine title matched " + bookid + " for " + nzbtitle_formatted) else: logger.debug("Magazine name too short (%s)" % len(nzbtitle_exploded)) rejected = True if not rejected and lazylibrarian.CONFIG['BLACKLIST_FAILED']: blocked = myDB.match('SELECT * from wanted WHERE NZBurl=? and Status="Failed"', (nzburl,)) if blocked: logger.debug("Rejecting %s, blacklisted at %s" % (nzbtitle_formatted, blocked['NZBprov'])) rejected = True if not rejected and lazylibrarian.CONFIG['BLACKLIST_PROCESSED']: blocked = myDB.match('SELECT * from wanted WHERE NZBurl=?', (nzburl,)) if blocked: logger.debug("Rejecting %s, blacklisted at %s" % (nzbtitle_formatted, blocked['NZBprov'])) rejected = True if not rejected: reject_list = getList(str(results['Reject']).lower()) reject_list += getList(lazylibrarian.CONFIG['REJECT_MAGS'], ',') lower_title = unaccented(nzbtitle_formatted).lower() lower_bookid = unaccented(bookid).lower() if reject_list: if lazylibrarian.LOGLEVEL & lazylibrarian.log_searchmag: logger.debug('Reject: %s' % str(reject_list)) logger.debug('Title: %s' % lower_title) logger.debug('Bookid: %s' % lower_bookid) 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 if rejected: rejects += 1 else: regex_pass, issuedate, year = get_issue_date(nzbtitle_exploded) if regex_pass: logger.debug('Issue %s (regex %s) for %s ' % (issuedate, regex_pass, nzbtitle_formatted)) datetype_ok = True datetype = book['datetype'] if datetype: # check all wanted parts are in the regex result # Day Month Year Vol Iss (MM needs two months) if 'M' in datetype and regex_pass not in [1, 2, 3, 4, 5, 6, 7, 12]: datetype_ok = False elif 'D' in datetype and regex_pass not in [3, 5, 6]: datetype_ok = False elif 'MM' in datetype and regex_pass not in [1]: # bi monthly datetype_ok = False elif 'V' in datetype and 'I' in datetype and regex_pass not in [8, 9, 17, 18]: datetype_ok = False elif 'V' in datetype and regex_pass not in [2, 10, 11, 12, 13, 14, 17, 18]: datetype_ok = False elif 'I' in datetype and regex_pass not in [2, 10, 11, 12, 13, 14, 16, 17, 18]: datetype_ok = False elif 'Y' in datetype and regex_pass not in [1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 13, 15, 16, 18]: datetype_ok = False else: datetype_ok = False 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 issuedate = "1970-01-01" # provide a fake date for bad-date issues # wanted issues go into wanted table marked "Wanted" # the rest into pastissues table marked "Skipped" or "Have" insert_table = "pastissues" comp_date = 0 if datetype_ok: control_date = results['IssueDate'] logger.debug("Control date: [%s]" % control_date) if not control_date: # we haven't got any copies of this magazine yet # get a rough time just over MAX_AGE days ago to compare to, in format yyyy-mm-dd # could perhaps calc differently for weekly, biweekly etc # For magazines with only an issue number use zero as we can't tell age if str(issuedate).isdigit(): logger.debug('Magazine comparing issue numbers (%s)' % issuedate) control_date = 0 elif re.match('\d+-\d\d-\d\d', str(issuedate)): 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: logger.debug('Magazine unable to find comparison type [%s]' % issuedate) control_date = 0 if str(control_date).isdigit() and str(issuedate).isdigit(): # for issue numbers, check if later than last one we have if regex_pass in [10, 12, 13] and year: issuedate = "%s%04d" % (year, int(issuedate)) else: issuedate = str(issuedate).zfill(4) if not control_date: comp_date = 1 else: comp_date = int(issuedate) - int(control_date) elif re.match('\d+-\d\d-\d\d', str(control_date)) and \ re.match('\d+-\d\d-\d\d', str(issuedate)): # 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(issuedate, control_date) else: # invalid comparison of date and issue number comp_date = 0 if re.match('\d+-\d\d-\d\d', str(control_date)): if regex_pass > 9 and year: # we assumed it was an issue number, but it could be a date year = check_int(year, 0) if regex_pass in [10, 12, 13]: issuedate = int(issuedate[:4]) issuenum = check_int(issuedate, 0) if year and 1 <= issuenum <= 12: issuedate = "%04d-%02d-01" % (year, issuenum) comp_date = datecompare(issuedate, control_date) if not comp_date: logger.debug('Magazine %s failed: Expecting a date' % nzbtitle_formatted) else: logger.debug('Magazine %s failed: Expecting issue number' % nzbtitle_formatted) if not comp_date: bad_date += 1 issuedate = "1970-01-01" if issuedate == "1970-01-01": logger.debug('This issue of %s is unknown age; skipping.' % nzbtitle_formatted) elif not datetype_ok: logger.debug('This issue of %s not in a wanted date format.' % nzbtitle_formatted) elif comp_date > 0: # keep track of what we're going to download so we don't download dupes new_date += 1 issue = bookid + ',' + issuedate 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) logger.debug('Magazine request number %s' % len(issues)) if lazylibrarian.LOGLEVEL & lazylibrarian.log_searchmag: logger.debug(str(issues)) insert_table = "wanted" nzbdate = now() # when we asked for it else: logger.debug('This issue of %s is already flagged for download' % issue) else: if lazylibrarian.LOGLEVEL & lazylibrarian.log_searchmag: 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 Status from %s WHERE NZBtitle=? and NZBprov=?' % insert_table, (nzbtitle, nzbprov)) if mag_entry: if lazylibrarian.LOGLEVEL & lazylibrarian.log_searchmag: logger.debug('%s is already in %s marked %s' % (nzbtitle, insert_table, mag_entry['Status'])) else: controlValueDict = { "NZBtitle": nzbtitle, "NZBprov": nzbprov } if insert_table == 'pastissues': # try to mark ones we've already got match = myDB.match("SELECT * from issues WHERE Title=? AND IssueDate=?", (bookid, issuedate)) if match: insert_status = "Have" else: insert_status = "Skipped" else: insert_status = "Wanted" newValueDict = { "NZBurl": nzburl, "BookID": bookid, "NZBdate": nzbdate, "AuxInfo": issuedate, "Status": insert_status, "NZBsize": nzbsize, "NZBmode": nzbmode } myDB.upsert(insert_table, newValueDict, controlValueDict) if lazylibrarian.LOGLEVEL & lazylibrarian.log_searchmag: logger.debug('Added %s to %s marked %s' % (nzbtitle, insert_table, insert_status)) msg = 'Found %i result%s for %s. %i new,' % (total_nzbs, plural(total_nzbs), bookid, new_date) msg += ' %i old, %i fail date, %i fail name,' % (old_date, bad_date, bad_name) msg += ' %i rejected: %i to download' % (rejects, len(maglist)) logger.info(msg) for magazine in maglist: if magazine['nzbmode'] in ["torznab", "torrent", "magnet"]: snatch, res = TORDownloadMethod( magazine['bookid'], magazine['nzbtitle'], magazine['nzburl'], 'Magazine') elif magazine['nzbmode'] == 'direct': snatch, res = DirectDownloadMethod( magazine['bookid'], magazine['nzbtitle'], magazine['nzburl'], 'Magazine') elif magazine['nzbmode'] == 'nzb': snatch, res = NZBDownloadMethod( magazine['bookid'], magazine['nzbtitle'], magazine['nzburl'], 'Magazine') else: res = 'Unhandled NZBmode [%s] for %s' % (magazine['nzbmode'], magazine["nzburl"]) logger.error(res) snatch = 0 if snatch: logger.info('Downloading %s from %s' % (magazine['nzbtitle'], magazine["nzbprov"])) custom_notify_snatch("%s %s" % (magazine['bookid'], magazine['nzburl'])) notify_snatch("Magazine %s from %s at %s" % (unaccented(magazine['nzbtitle']), magazine["nzbprov"], now())) scheduleJob(action='Start', target='PostProcessor') else: myDB.action('UPDATE wanted SET status="Failed",DLResult=? WHERE NZBurl=?', (res, magazine["nzburl"])) 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()) finally: threading.currentThread().name = "WEBSERVER"
def search_rss_book(books=None, library=None): """ books is a list of new books to add, or None for backlog search library is "eBook" or "AudioBook" or None to search all book types """ if not (lazylibrarian.USE_RSS()): logger.warn('RSS search is disabled') scheduleJob(action='Stop', target='search_rss_book') return try: threadname = threading.currentThread().name if "Thread-" in threadname: if not books: threading.currentThread().name = "SEARCHALLRSS" else: threading.currentThread().name = "SEARCHRSS" myDB = database.DBConnection() searchbooks = [] if not books: # We are performing a backlog search cmd = 'SELECT BookID, AuthorName, Bookname, BookSub, BookAdded, books.Status, AudioStatus ' cmd += 'from books,authors WHERE (books.Status="Wanted" OR AudioStatus="Wanted") ' cmd += 'and books.AuthorID = authors.AuthorID order by BookAdded desc' results = myDB.select(cmd) for terms in results: searchbooks.append(terms) else: # The user has added a new book for book in books: cmd = 'SELECT BookID, AuthorName, BookName, BookSub, books.Status, AudioStatus ' cmd += 'from books,authors WHERE BookID=? AND books.AuthorID = authors.AuthorID' results = myDB.select(cmd, (book['bookid'], )) for terms in results: searchbooks.append(terms) if len(searchbooks) == 0: logger.debug("SearchRSS - No books to search for") return resultlist, nproviders, _ = IterateOverRSSSites() if not nproviders: logger.warn('No rss providers are available') scheduleJob(action='Stop', target='search_rss_book') return # No point in continuing logger.info('RSS Searching for %i book%s' % (len(searchbooks), plural(len(searchbooks)))) searchlist = [] for searchbook in searchbooks: # searchterm is only used for display purposes searchterm = searchbook['AuthorName'] + ' ' + searchbook['BookName'] if searchbook['BookSub']: searchterm = searchterm + ': ' + searchbook['BookSub'] if library is None or library == 'eBook': if searchbook['Status'] == "Wanted": cmd = 'SELECT BookID from wanted WHERE BookID=? and AuxInfo="eBook" and Status="Snatched"' snatched = myDB.match(cmd, (searchbook["BookID"], )) if snatched: logger.warn( 'eBook %s %s already marked snatched in wanted table' % (searchbook['AuthorName'], searchbook['BookName'])) else: searchlist.append({ "bookid": searchbook['BookID'], "bookName": searchbook['BookName'], "bookSub": searchbook['BookSub'], "authorName": searchbook['AuthorName'], "library": "eBook", "searchterm": searchterm }) if library is None or library == 'AudioBook': if searchbook['AudioStatus'] == "Wanted": cmd = 'SELECT BookID from wanted WHERE BookID=? and AuxInfo="AudioBook" and Status="Snatched"' snatched = myDB.match(cmd, (searchbook["BookID"], )) if snatched: logger.warn( 'AudioBook %s %s already marked snatched in wanted table' % (searchbook['AuthorName'], searchbook['BookName'])) else: searchlist.append({ "bookid": searchbook['BookID'], "bookName": searchbook['BookName'], "bookSub": searchbook['BookSub'], "authorName": searchbook['AuthorName'], "library": "AudioBook", "searchterm": searchterm }) rss_count = 0 for book in searchlist: if book['library'] == 'AudioBook': searchtype = 'audio' else: searchtype = 'book' found = processResultList(resultlist, book, searchtype, 'rss') # if you can't find the book, try title without any "(extended details, series etc)" if not found and '(' in book['bookName']: # anything to shorten? searchtype = 'short' + searchtype found = processResultList(resultlist, book, searchtype, 'rss') if not found: logger.info("RSS Searches for %s %s returned no results." % (book['library'], book['searchterm'])) if found > 1: rss_count += 1 logger.info("RSS Search for Wanted items complete, found %s book%s" % (rss_count, plural(rss_count))) except Exception: logger.error('Unhandled exception in search_rss_book: %s' % traceback.format_exc()) finally: threading.currentThread().name = "WEBSERVER"
def search_rss_book(books=None, reset=False): threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "SEARCHRSS" if not(lazylibrarian.USE_RSS()): logger.warn('RSS search is disabled') scheduleJob(action='Stop', target='search_rss_book') return myDB = database.DBConnection() searchlist = [] if books is None: # We are performing a backlog search searchbooks = myDB.select('SELECT BookID, AuthorName, Bookname, BookSub, BookAdded from books WHERE Status="Wanted" order by BookAdded desc') else: # The user has added a new book searchbooks = [] for book in books: searchbook = myDB.select('SELECT BookID, AuthorName, BookName, BookSub from books WHERE BookID="%s" \ AND Status="Wanted"' % book['bookid']) for terms in searchbook: searchbooks.append(terms) if len(searchbooks) == 0: logger.debug("RSS search requested for no books or invalid BookID") return else: logger.info('RSS Searching for %i book%s' % (len(searchbooks), plural(len(searchbooks)))) resultlist, nproviders = IterateOverRSSSites() if not nproviders: logger.warn('No rss providers are set, check config') return # No point in continuing dic = {'...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': ''} rss_count = 0 for book in searchbooks: author = book['AuthorName'] title = book['BookName'] author = unaccented_str(replace_all(author, dic)) title = unaccented_str(replace_all(title, dic)) found = processResultList(resultlist, author, title, book) # if you can't find the book, try author without initials, # and title without any "(extended details, series etc)" if not found: if author[1] in '. ' or '(' in title: # anything to shorten? while author[1] in '. ': # strip any initials author = author[2:].strip() # and leading whitespace if '(' in title: title = title.split('(')[0] found = processResultList(resultlist, author, title, book) if not found: logger.debug("Searches returned no results. Adding book %s - %s to queue." % (author, title)) if found > True: rss_count = rss_count + 1 logger.info("RSS Search for Wanted items complete, found %s book%s" % (rss_count, plural(rss_count))) if reset: scheduleJob(action='Restart', target='search_rss_book')
def processResultList(resultlist, book, searchtype): myDB = database.DBConnection() dictrepl = {'...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': ' ', '*': '', '(': '', ')': '', '[': '', ']': '', '#': '', '0': '', '1': '', '2': '', '3': '', '4': '', '5': '', '6': '', '7': '', '8': '', '9': '', '\'': '', ':': '', '!': '', '-': ' ', '\s\s': ' '} dic = {'...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': ''} match_ratio = int(lazylibrarian.MATCH_RATIO) reject_list = getList(lazylibrarian.REJECT_WORDS) author = unaccented_str(replace_all(book['authorName'], dic)) title = unaccented_str(replace_all(book['bookName'], dic)) matches = [] for tor in resultlist: torTitle = unaccented_str(tor['tor_title']) torTitle = replace_all(torTitle, dictrepl).strip() torTitle = re.sub(r"\s\s+", " ", torTitle) # remove extra whitespace torAuthor_match = fuzz.token_set_ratio(author, torTitle) torBook_match = fuzz.token_set_ratio(title, torTitle) logger.debug(u"TOR author/book Match: %s/%s for %s" % (torAuthor_match, torBook_match, torTitle)) tor_url = tor['tor_url'] rejected = False already_failed = myDB.match('SELECT * from wanted WHERE NZBurl="%s" and Status="Failed"' % tor_url) if already_failed: logger.debug("Rejecting %s, blacklisted at %s" % (torTitle, already_failed['NZBprov'])) rejected = True if not rejected: for word in reject_list: if word in torTitle.lower() and word not in author.lower() and word not in title.lower(): rejected = True logger.debug("Rejecting %s, contains %s" % (torTitle, word)) break tor_size_temp = tor['tor_size'] # Need to cater for when this is NONE (Issue 35) tor_size_temp = check_int(tor_size_temp, 1000) tor_size = round(float(tor_size_temp) / 1048576, 2) maxsize = check_int(lazylibrarian.REJECT_MAXSIZE, 0) if not rejected: if maxsize and tor_size > maxsize: rejected = True logger.debug("Rejecting %s, too large" % torTitle) if not rejected: bookid = book['bookid'] tor_Title = (author + ' - ' + title + ' LL.(' + book['bookid'] + ')').strip() controlValueDict = {"NZBurl": tor_url} newValueDict = { "NZBprov": tor['tor_prov'], "BookID": bookid, "NZBdate": now(), # when we asked for it "NZBsize": tor_size, "NZBtitle": tor_Title, "NZBmode": "torrent", "Status": "Skipped" } score = (torBook_match + torAuthor_match) / 2 # as a percentage # lose a point for each extra word in the title so we get the closest match words = len(getList(torTitle)) words -= len(getList(author)) words -= len(getList(title)) score -= abs(words) matches.append([score, torTitle, newValueDict, controlValueDict]) if matches: highest = max(matches, key=lambda x: x[0]) score = highest[0] nzb_Title = highest[1] newValueDict = highest[2] controlValueDict = highest[3] if score < match_ratio: logger.info(u'Nearest TOR match (%s%%): %s using %s search for %s %s' % (score, nzb_Title, searchtype, author, title)) return False logger.info(u'Best TOR match (%s%%): %s using %s search' % (score, nzb_Title, searchtype)) snatchedbooks = myDB.match('SELECT * from books WHERE BookID="%s" and Status="Snatched"' % newValueDict["BookID"]) if snatchedbooks: logger.debug('%s already marked snatched' % nzb_Title) return True # someone else found it, not us else: myDB.upsert("wanted", newValueDict, controlValueDict) if newValueDict["NZBprov"] == 'libgen': # for libgen we use direct download links snatch = DirectDownloadMethod(newValueDict["BookID"], newValueDict["NZBprov"], newValueDict["NZBtitle"], controlValueDict["NZBurl"], nzb_Title) else: snatch = TORDownloadMethod(newValueDict["BookID"], newValueDict["NZBprov"], newValueDict["NZBtitle"], controlValueDict["NZBurl"]) if snatch: logger.info('Downloading %s from %s' % (newValueDict["NZBtitle"], newValueDict["NZBprov"])) notify_snatch("%s from %s at %s" % (newValueDict["NZBtitle"], newValueDict["NZBprov"], now())) scheduleJob(action='Start', target='processDir') return True + True # we found it else: logger.debug("No torrent's found for [%s] using searchtype %s" % (book["searchterm"], searchtype)) return False
def processResultList(resultlist, author, title, book): myDB = database.DBConnection() dictrepl = {'...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': ' ', '*': '', '(': '', ')': '', '[': '', ']': '', '#': '', '0': '', '1': '', '2': '', '3': '', '4': '', '5': '', '6': '', '7': '', '8': '', '9': '', '\'': '', ':': '', '!': '', '-': ' ', '\s\s': ' '} # ' the ': ' ', ' a ': ' ', ' and ': ' ', ' to ': ' ', ' of ': ' ', # ' for ': ' ', ' my ': ' ', ' in ': ' ', ' at ': ' ', ' with ': ' '} match_ratio = int(lazylibrarian.MATCH_RATIO) reject_list = getList(lazylibrarian.REJECT_WORDS) matches = [] # bit of a misnomer now, rss can search both tor and nzb rss feeds for tor in resultlist: torTitle = unaccented_str(replace_all(tor['tor_title'], dictrepl)).strip() torTitle = re.sub(r"\s\s+", " ", torTitle) # remove extra whitespace tor_Author_match = fuzz.token_set_ratio(author, torTitle) tor_Title_match = fuzz.token_set_ratio(title, torTitle) logger.debug("RSS Author/Title Match: %s/%s for %s" % (tor_Author_match, tor_Title_match, torTitle)) tor_url = tor['tor_url'] rejected = False already_failed = myDB.action('SELECT * from wanted WHERE NZBurl="%s" and Status="Failed"' % tor_url).fetchone() if already_failed: logger.debug("Rejecting %s, blacklisted at %s" % (torTitle, already_failed['NZBprov'])) rejected = True if not rejected: for word in reject_list: if word in torTitle.lower() and not word in author.lower() and not word in book.lower(): rejected = True logger.debug("Rejecting %s, contains %s" % (torTitle, word)) break tor_size_temp = tor['tor_size'] # Need to cater for when this is NONE (Issue 35) if tor_size_temp is None: tor_size_temp = 1000 tor_size = round(float(tor_size_temp) / 1048576, 2) maxsize = check_int(lazylibrarian.REJECT_MAXSIZE, 0) if not rejected: if maxsize and tor_size > maxsize: rejected = True logger.debug("Rejecting %s, too large" % torTitle) if not rejected: if tor_Title_match >= match_ratio and tor_Author_match >= match_ratio: bookid = book['bookid'] tor_Title = (book["authorName"] + ' - ' + book['bookName'] + ' LL.(' + book['bookid'] + ')').strip() tor_prov = tor['tor_prov'] tor_feed = tor['tor_feed'] controlValueDict = {"NZBurl": tor_url} newValueDict = { "NZBprov": tor_prov, "BookID": bookid, "NZBdate": now(), # when we asked for it "NZBsize": tor_size, "NZBtitle": tor_Title, "NZBmode": "torrent", "Status": "Skipped" } score = (tor_Title_match + tor_Author_match)/2 # as a percentage # lose a point for each extra word in the title so we get the closest match words = len(getList(torTitle)) words -= len(getList(author)) words -= len(getList(title)) score -= abs(words) matches.append([score, torTitle, newValueDict, controlValueDict]) if matches: highest = max(matches, key=lambda x: x[0]) score = highest[0] nzb_Title = highest[1] newValueDict = highest[2] controlValueDict = highest[3] logger.info(u'Best match RSS (%s%%): %s using %s search' % (score, nzb_Title, searchtype)) snatchedbooks = myDB.action('SELECT * from books WHERE BookID="%s" and Status="Snatched"' % newValueDict["BookID"]).fetchone() if snatchedbooks: # check if one of the other downloaders got there first logger.debug('%s already marked snatched' % nzb_Title) return True else: myDB.upsert("wanted", newValueDict, controlValueDict) tor_url = controlValueDict["NZBurl"] if '.nzb' in tor_url: snatch = NZBDownloadMethod(newValueDict["BookID"], newValueDict["NZBprov"], newValueDict["NZBtitle"], controlValueDict["NZBurl"]) else: """ # http://baconbits.org/torrents.php?action=download&authkey=<authkey>&torrent_pass=<password.hashed>&id=185398 if not tor_url.startswith('magnet'): # magnets don't use auth pwd = lazylibrarian.RSS_PROV[tor_feed]['PASS'] auth = lazylibrarian.RSS_PROV[tor_feed]['AUTH'] # don't know what form of password hash is required, try sha1 tor_url = tor_url.replace('<authkey>', auth).replace('<password.hashed>', sha1(pwd)) """ snatch = TORDownloadMethod(newValueDict["BookID"], newValueDict["NZBprov"], newValueDict["NZBtitle"], tor_url) if snatch: notify_snatch(newValueDict["NZBtitle"] + ' at ' + now()) scheduleJob(action='Start', target='processDir') return True + True # we found it else: logger.debug("No RSS found for " + (book["authorName"] + ' ' + book['bookName']).strip()) return False
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 search_tor_book(books=None, reset=False): threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "SEARCHTOR" if not lazylibrarian.USE_TOR(): logger.warn('No Torrent providers set, check config') return myDB = database.DBConnection() searchlist = [] if books is None: # We are performing a backlog search searchbooks = myDB.select( 'SELECT BookID, AuthorName, Bookname, BookSub, BookAdded from books WHERE Status="Wanted" order by BookAdded desc' ) else: # The user has added a new book searchbooks = [] for book in books: searchbook = myDB.select( 'SELECT BookID, AuthorName, BookName, BookSub from books WHERE BookID="%s" \ AND Status="Wanted"' % book['bookid']) for terms in searchbook: searchbooks.append(terms) if len(searchbooks) == 0: logger.debug("TOR search requested for no books or invalid BookID") return else: 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'] searchterm = searchterm + '"' 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") # if you can't find the book under "books", you might find under general search 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 = 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')
def search_rss_book(books=None, library=None): """ books is a list of new books to add, or None for backlog search library is "eBook" or "AudioBook" or None to search all book types """ 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 myDB = database.DBConnection() resultlist, wishproviders = IterateOverWishLists() new_books = 0 if not wishproviders: logger.debug('No wishlists are set') else: # for each item in resultlist, add to database if necessary, and mark as wanted logger.debug('Processing %s item%s in wishlists' % (len(resultlist), plural(len(resultlist)))) for book in resultlist: # we get rss_author, rss_title, maybe 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... if lazylibrarian.CONFIG['BOOK_API'] == "GoodReads" and book['rss_bookid']: bookmatch = myDB.match('select Status,BookName from books where bookid=?', (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": book['rss_bookid']} newValueDict = {"Status": "Wanted"} myDB.upsert("books", newValueDict, controlValueDict) new_books += 1 else: import_book(book['rss_bookid']) new_books += 1 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) new_books += 1 else: # not in database yet results = '' if book['rss_isbn']: results = search_for(book['rss_isbn']) if results: result = results[0] # type: dict if result['isbn_fuzz'] > check_int(lazylibrarian.CONFIG['MATCH_RATIO'], 90): logger.info("Found (%s%%) %s: %s" % (result['isbn_fuzz'], result['authorname'], result['bookname'])) import_book(result['bookid']) new_books += 1 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] # type: dict if result['author_fuzz'] > check_int(lazylibrarian.CONFIG['MATCH_RATIO'], 90) \ and result['book_fuzz'] > check_int(lazylibrarian.CONFIG['MATCH_RATIO'], 90): logger.info("Found (%s%% %s%%) %s: %s" % (result['author_fuzz'], result['book_fuzz'], result['authorname'], result['bookname'])) import_book(result['bookid']) new_books += 1 bookmatch = True if not bookmatch: msg = "Skipping book %s by %s" % (item['Title'], book['rss_author']) 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 new_books: logger.info("Wishlist marked %s book%s as Wanted" % (new_books, plural(new_books))) searchbooks = [] if books is None: # We are performing a backlog search cmd = 'SELECT BookID, AuthorName, Bookname, BookSub, BookAdded, books.Status, AudioStatus ' cmd += 'from books,authors WHERE (books.Status="Wanted" OR AudioStatus="Wanted") ' cmd += 'and books.AuthorID = authors.AuthorID order by BookAdded desc' results = myDB.select(cmd) for terms in results: searchbooks.append(terms) else: # The user has added a new book for book in books: cmd = 'SELECT BookID, AuthorName, BookName, BookSub, books.Status, AudioStatus ' cmd += 'from books,authors WHERE BookID=? AND books.AuthorID = authors.AuthorID' results = myDB.select(cmd, (book['bookid'],)) for terms in results: searchbooks.append(terms) if len(searchbooks) == 0: logger.debug("SearchRSS - No books to search for") return resultlist, nproviders = IterateOverRSSSites() if not nproviders and not wishproviders: logger.warn('No rss providers are available') return # No point in continuing logger.info('RSS Searching for %i book%s' % (len(searchbooks), plural(len(searchbooks)))) searchlist = [] for searchbook in searchbooks: # searchterm is only used for display purposes searchterm = searchbook['AuthorName'] + ' ' + searchbook['BookName'] if searchbook['BookSub']: searchterm = searchterm + ': ' + searchbook['BookSub'] if library is None or library == 'eBook': if searchbook['Status'] == "Wanted": searchlist.append( {"bookid": searchbook['BookID'], "bookName": searchbook['BookName'], "bookSub": searchbook['BookSub'], "authorName": searchbook['AuthorName'], "library": "eBook", "searchterm": searchterm}) if library is None or library == 'AudioBook': if searchbook['AudioStatus'] == "Wanted": searchlist.append( {"bookid": searchbook['BookID'], "bookName": searchbook['BookName'], "bookSub": searchbook['BookSub'], "authorName": searchbook['AuthorName'], "library": "AudioBook", "searchterm": searchterm}) rss_count = 0 for book in searchlist: if book['library'] == 'AudioBook': searchtype = 'audio' else: searchtype = 'book' found = processResultList(resultlist, book, searchtype, 'rss') # if you can't find the book, try title without any "(extended details, series etc)" if not found and '(' in book['bookName']: # anything to shorten? searchtype = 'short' + searchtype found = processResultList(resultlist, book, searchtype, 'rss') if not found: logger.info("NZB Searches for %s %s returned no results." % (book['library'], book['searchterm'])) if found > True: rss_count += 1 logger.info("RSS Search for Wanted items complete, found %s book%s" % (rss_count, plural(rss_count))) except Exception: logger.error('Unhandled exception in search_rss_book: %s' % traceback.format_exc()) finally: threading.currentThread().name = "WEBSERVER"
def downloadResult(match, book): """ match: best result from search providers book: book we are downloading (needed for reporting author name) return: 0 if failed to snatch 1 if already snatched 2 if we snatched it """ # noinspection PyBroadException try: myDB = database.DBConnection() newValueDict = match[1] controlValueDict = match[2] # It's possible to get book and wanted tables "Snatched" status out of sync # for example if a user marks a book as "Wanted" after a search task snatches it and before postprocessor runs # so check status in both tables here snatched = myDB.match('SELECT BookID from wanted WHERE BookID=? and AuxInfo=? and Status="Snatched"', (newValueDict["BookID"], newValueDict["AuxInfo"])) if snatched: logger.debug('%s %s %s already marked snatched in wanted table' % (newValueDict["AuxInfo"], book['authorName'], book['bookName'])) return 1 # someone else already found it if newValueDict["AuxInfo"] == 'eBook': snatched = myDB.match('SELECT BookID from books WHERE BookID=? and Status="Snatched"', (newValueDict["BookID"],)) else: snatched = myDB.match('SELECT BookID from books WHERE BookID=? and AudioStatus="Snatched"', (newValueDict["BookID"],)) if snatched: logger.debug('%s %s %s already marked snatched in book table' % (newValueDict["AuxInfo"], book['authorName'], book['bookName'])) return 1 # someone else already found it myDB.upsert("wanted", newValueDict, controlValueDict) if newValueDict['NZBmode'] == 'direct': snatch = DirectDownloadMethod(newValueDict["BookID"], newValueDict["NZBtitle"], controlValueDict["NZBurl"], newValueDict["AuxInfo"]) elif newValueDict['NZBmode'] in ["torznab", "torrent", "magnet"]: snatch = TORDownloadMethod(newValueDict["BookID"], newValueDict["NZBtitle"], controlValueDict["NZBurl"], newValueDict["AuxInfo"]) elif newValueDict['NZBmode'] == 'nzb': snatch = NZBDownloadMethod(newValueDict["BookID"], newValueDict["NZBtitle"], controlValueDict["NZBurl"], newValueDict["AuxInfo"]) else: logger.error('Unhandled NZBmode [%s] for %s' % (newValueDict['NZBmode'], controlValueDict["NZBurl"])) snatch = 0 if snatch: logger.info('Downloading %s %s from %s' % (newValueDict["AuxInfo"], newValueDict["NZBtitle"], newValueDict["NZBprov"])) custom_notify_snatch("%s %s" % (newValueDict["BookID"], newValueDict['AuxInfo'])) notify_snatch("%s %s from %s at %s" % (newValueDict["AuxInfo"], newValueDict["NZBtitle"], newValueDict["NZBprov"], now())) # at this point we could add NZBprov to the blocklist with a short timeout, a second or two? # This would implement a round-robin search system. Blocklist with an incremental counter. # If number of active providers == number blocklisted, so no unblocked providers are left, # either sleep for a while, or unblock the one with the lowest counter. scheduleJob(action='Start', target='processDir') return 2 # we found it return 0 except Exception: logger.error('Unhandled exception in downloadResult: %s' % traceback.format_exc()) return 0
def search_rss_book(books=None, library=None): """ books is a list of new books to add, or None for backlog search library is "eBook" or "AudioBook" or None to search all book types """ if not (lazylibrarian.USE_RSS()): logger.warn('RSS search is disabled') scheduleJob(action='Stop', target='search_rss_book') return try: threadname = threading.currentThread().name if "Thread-" in threadname: if not books: threading.currentThread().name = "SEARCHALLRSS" else: threading.currentThread().name = "SEARCHRSS" myDB = database.DBConnection() searchbooks = [] if not books: # We are performing a backlog search cmd = 'SELECT BookID, AuthorName, Bookname, BookSub, BookAdded, books.Status, AudioStatus ' cmd += 'from books,authors WHERE (books.Status="Wanted" OR AudioStatus="Wanted") ' cmd += 'and books.AuthorID = authors.AuthorID order by BookAdded desc' results = myDB.select(cmd) for terms in results: searchbooks.append(terms) else: # The user has added a new book for book in books: cmd = 'SELECT BookID, AuthorName, BookName, BookSub, books.Status, AudioStatus ' cmd += 'from books,authors WHERE BookID=? AND books.AuthorID = authors.AuthorID' results = myDB.select(cmd, (book['bookid'],)) for terms in results: searchbooks.append(terms) if len(searchbooks) == 0: logger.debug("SearchRSS - No books to search for") return resultlist, nproviders, _ = IterateOverRSSSites() if not nproviders: logger.warn('No rss providers are available') scheduleJob(action='Stop', target='search_rss_book') return # No point in continuing logger.info('RSS Searching for %i book%s' % (len(searchbooks), plural(len(searchbooks)))) searchlist = [] for searchbook in searchbooks: # searchterm is only used for display purposes searchterm = searchbook['AuthorName'] + ' ' + searchbook['BookName'] if searchbook['BookSub']: searchterm = searchterm + ': ' + searchbook['BookSub'] if library is None or library == 'eBook': if searchbook['Status'] == "Wanted": cmd = 'SELECT BookID from wanted WHERE BookID=? and AuxInfo="eBook" and Status="Snatched"' snatched = myDB.match(cmd, (searchbook["BookID"],)) if snatched: logger.warn('eBook %s %s already marked snatched in wanted table' % (searchbook['AuthorName'], searchbook['BookName'])) else: searchlist.append( {"bookid": searchbook['BookID'], "bookName": searchbook['BookName'], "bookSub": searchbook['BookSub'], "authorName": searchbook['AuthorName'], "library": "eBook", "searchterm": searchterm}) if library is None or library == 'AudioBook': if searchbook['AudioStatus'] == "Wanted": cmd = 'SELECT BookID from wanted WHERE BookID=? and AuxInfo="AudioBook" and Status="Snatched"' snatched = myDB.match(cmd, (searchbook["BookID"],)) if snatched: logger.warn('AudioBook %s %s already marked snatched in wanted table' % (searchbook['AuthorName'], searchbook['BookName'])) else: searchlist.append( {"bookid": searchbook['BookID'], "bookName": searchbook['BookName'], "bookSub": searchbook['BookSub'], "authorName": searchbook['AuthorName'], "library": "AudioBook", "searchterm": searchterm}) rss_count = 0 for book in searchlist: if book['library'] == 'AudioBook': searchtype = 'audio' else: searchtype = 'book' found = processResultList(resultlist, book, searchtype, 'rss') # if you can't find the book, try title without any "(extended details, series etc)" if not found and '(' in book['bookName']: # anything to shorten? searchtype = 'short' + searchtype found = processResultList(resultlist, book, searchtype, 'rss') if not found: logger.info("RSS Searches for %s %s returned no results." % (book['library'], book['searchterm'])) if found > 1: rss_count += 1 logger.info("RSS Search for Wanted items complete, found %s book%s" % (rss_count, plural(rss_count))) except Exception: logger.error('Unhandled exception in search_rss_book: %s' % traceback.format_exc()) finally: threading.currentThread().name = "WEBSERVER"
def processResultList(resultlist, book, searchtype): myDB = database.DBConnection() dictrepl = {'...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': ' ', '*': '', '(': '', ')': '', '[': '', ']': '', '#': '', '0': '', '1': '', '2': '', '3': '', '4': '', '5': '', '6': '', '7': '', '8': '', '9': '', '\'': '', ':': '', '!': '', '-': ' ', '\s\s': ' '} dic = {'...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '.', ';': ''} match_ratio = int(lazylibrarian.CONFIG['MATCH_RATIO']) reject_list = getList(lazylibrarian.CONFIG['REJECT_WORDS']) author = unaccented_str(replace_all(book['authorName'], dic)) title = unaccented_str(replace_all(book['bookName'], dic)) matches = [] for tor in resultlist: torTitle = unaccented_str(tor['tor_title']) torTitle = replace_all(torTitle, dictrepl).strip() torTitle = re.sub(r"\s\s+", " ", torTitle) # remove extra whitespace torAuthor_match = fuzz.token_set_ratio(author, torTitle) torBook_match = fuzz.token_set_ratio(title, torTitle) logger.debug(u"TOR author/book Match: %s/%s for %s" % (torAuthor_match, torBook_match, torTitle)) tor_url = tor['tor_url'] rejected = False already_failed = myDB.match('SELECT * from wanted WHERE NZBurl="%s" and Status="Failed"' % tor_url) if already_failed: logger.debug("Rejecting %s, blacklisted at %s" % (torTitle, already_failed['NZBprov'])) rejected = True if not rejected: for word in reject_list: if word in torTitle.lower() and word not in author.lower() and word not in title.lower(): rejected = True logger.debug("Rejecting %s, contains %s" % (torTitle, word)) break tor_size_temp = tor['tor_size'] # Need to cater for when this is NONE (Issue 35) tor_size_temp = check_int(tor_size_temp, 1000) tor_size = round(float(tor_size_temp) / 1048576, 2) maxsize = check_int(lazylibrarian.CONFIG['REJECT_MAXSIZE'], 0) if not rejected: if maxsize and tor_size > maxsize: rejected = True logger.debug("Rejecting %s, too large" % torTitle) minsize = check_int(lazylibrarian.CONFIG['REJECT_MINSIZE'], 0) if not rejected: if minsize and tor_size < minsize: rejected = True logger.debug("Rejecting %s, too small" % torTitle) if not rejected: bookid = book['bookid'] tor_Title = (author + ' - ' + title + ' LL.(' + book['bookid'] + ')').strip() controlValueDict = {"NZBurl": tor_url} newValueDict = { "NZBprov": tor['tor_prov'], "BookID": bookid, "NZBdate": now(), # when we asked for it "NZBsize": tor_size, "NZBtitle": tor_Title, "NZBmode": "torrent", "Status": "Skipped" } score = (torBook_match + torAuthor_match) / 2 # as a percentage # lose a point for each extra word in the title so we get the closest match words = len(getList(torTitle)) words -= len(getList(author)) words -= len(getList(title)) score -= abs(words) matches.append([score, torTitle, newValueDict, controlValueDict]) if matches: highest = max(matches, key=lambda x: x[0]) score = highest[0] nzb_Title = highest[1] newValueDict = highest[2] controlValueDict = highest[3] if score < match_ratio: logger.info(u'Nearest TOR match (%s%%): %s using %s search for %s %s' % (score, nzb_Title, searchtype, author, title)) return False logger.info(u'Best TOR match (%s%%): %s using %s search' % (score, nzb_Title, searchtype)) snatchedbooks = myDB.match('SELECT BookID from books WHERE BookID="%s" and Status="Snatched"' % newValueDict["BookID"]) if snatchedbooks: logger.debug('%s already marked snatched' % nzb_Title) return True # someone else found it, not us else: myDB.upsert("wanted", newValueDict, controlValueDict) if newValueDict["NZBprov"] == 'libgen': # for libgen we use direct download links snatch = DirectDownloadMethod(newValueDict["BookID"], newValueDict["NZBtitle"], controlValueDict["NZBurl"], nzb_Title) else: snatch = TORDownloadMethod(newValueDict["BookID"], newValueDict["NZBtitle"], controlValueDict["NZBurl"]) if snatch: logger.info('Downloading %s from %s' % (newValueDict["NZBtitle"], newValueDict["NZBprov"])) notify_snatch("%s from %s at %s" % (newValueDict["NZBtitle"], newValueDict["NZBprov"], now())) custom_notify_snatch(newValueDict["BookID"]) scheduleJob(action='Start', target='processDir') return True + True # we found it else: logger.debug("No torrent's found for [%s] using searchtype %s" % (book["searchterm"], searchtype)) return False
def search_magazines(mags=None, reset=False): # produce a list of magazines to search for, tor, nzb, torznab, rss # noinspection PyBroadException try: threadname = threading.currentThread().name if "Thread-" in threadname: if mags is None: threading.currentThread().name = "SEARCHALLMAG" else: threading.currentThread().name = "SEARCHMAG" 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=? AND Status="Active"', (magazine['bookid'], )) for terms in searchmags_temp: searchmags.append(terms) if len(searchmags) == 0: threading.currentThread().name = "WEBSERVER" 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("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: dic = { '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '' } # strip accents from the magazine title for easier name-matching searchterm = unaccented_str(searchmag['Title']) if not searchterm: # unless there are no ascii characters left searchterm = searchmag['Title'] searchterm = replace_all(searchterm, dic) searchterm = re.sub('[.\-/]', ' ', searchterm) searchterm = 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: # don't nag. Show warning message no more than every 20 mins timenow = int(time.time()) if check_int(lazylibrarian.NO_NZB_MSG, 0) + 1200 < timenow: logger.warn( 'No nzb providers are available. Check config and blocklist' ) lazylibrarian.NO_NZB_MSG = timenow if lazylibrarian.USE_DIRECT(): dir_resultlist, nproviders = IterateOverDirectSites( book, 'mag') if not nproviders: # don't nag. Show warning message no more than every 20 mins timenow = int(time.time()) if check_int(lazylibrarian.NO_DIRECT_MSG, 0) + 1200 < timenow: logger.warn( 'No direct providers are available. Check config and blocklist' ) lazylibrarian.NO_DIRECT_MSG = timenow if dir_resultlist: for item in dir_resultlist: # reformat the 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 'nzbsize': item['tor_size'], 'nzbmode': 'torrent' }) if lazylibrarian.USE_TOR(): tor_resultlist, nproviders = IterateOverTorrentSites( book, 'mag') if not nproviders: # don't nag. Show warning message no more than every 20 mins timenow = int(time.time()) if check_int(lazylibrarian.NO_TOR_MSG, 0) + 1200 < timenow: logger.warn( 'No tor providers are available. Check config and blocklist' ) lazylibrarian.NO_TOR_MSG = timenow 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: # don't nag. Show warning message no more than every 20 mins timenow = int(time.time()) if check_int(lazylibrarian.NO_RSS_MSG, 0) + 1200 < timenow: logger.warn( 'No rss providers are available. Check config and blocklist' ) lazylibrarian.NO_RSS_MSG = timenow 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("No results for magazine %s" % 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'] # strip accents from the magazine title for easier name-matching nzbtitle = unaccented_str(nzb['nzbtitle']) if not nzbtitle: # unless it's not a latin-1 encodable name nzbtitle = 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=?', (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: 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) # remove extra spaces if they're in a row if nzbtitle_formatted and 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 all the words in the mag title are in the nzbtitle rejected = False wlist = [] for word in nzbtitle_exploded: wlist.append(unaccented(word).lower()) for word in bookid_exploded: if unaccented(word).lower() not in wlist: rejected = True break if rejected: logger.debug( u"Magazine title match failed " + bookid + " for " + nzbtitle_formatted) else: logger.debug(u"Magazine matched " + bookid + " for " + nzbtitle_formatted) else: logger.debug("Magazine name too short (%s)" % len(nzbtitle_exploded)) rejected = True if not rejected: blocked = myDB.match( 'SELECT * from wanted WHERE NZBurl=? and Status="Failed"', (nzburl, )) if blocked: logger.debug( "Rejecting %s, blacklisted at %s" % (nzbtitle_formatted, blocked['NZBprov'])) rejected = True if not rejected: reject_list = getList( str(results['Reject']).lower()) reject_list += getList( lazylibrarian.CONFIG['REJECT_MAGS']) lower_title = unaccented( nzbtitle_formatted).lower() lower_bookid = unaccented(bookid).lower() if reject_list: if lazylibrarian.LOGLEVEL > 2: logger.debug('Reject: %s' % str(reject_list)) logger.debug('Title: %s' % lower_title) logger.debug('Bookid: %s' % lower_bookid) 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 > 2: 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" control_date = results['IssueDate'] 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 str(newdatish).isdigit(): logger.debug( 'Magazine comparing issue numbers (%s)' % newdatish) control_date = 0 elif re.match('\d+-\d\d-\d\d', 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: logger.debug( 'Magazine unable to find comparison type [%s]' % newdatish) control_date = 0 if str(control_date).isdigit() and str( newdatish).isdigit(): # 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 elif re.match('\d+-\d\d-\d\d', str(control_date)) and \ re.match('\d+-\d\d-\d\d', 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) else: # invalid comparison of date and issue number if re.match('\d+-\d\d-\d\d', str(control_date)): logger.debug( 'Magazine %s failed: Expecting a date' % nzbtitle_formatted) else: logger.debug( 'Magazine %s failed: Expecting issue number' % 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) logger.debug('Magazine request number %s' % len(issues)) if lazylibrarian.LOGLEVEL > 2: logger.debug(str(issues)) 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=? and NZBprov=?' % insert_table, (nzbtitle, nzbprov)) if mag_entry: if lazylibrarian.LOGLEVEL > 2: logger.debug( '%s is already in %s marked %s' % (nzbtitle, insert_table, insert_status)) else: 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) if lazylibrarian.LOGLEVEL > 2: logger.debug('Added %s to %s marked %s' % (nzbtitle, insert_table, insert_status)) msg = 'Found %i result%s for %s. %i new,' % ( total_nzbs, plural(total_nzbs), bookid, new_date) msg += ' %i old, %i fail date, %i fail name,' % ( old_date, bad_date, bad_name) msg += ' %i rejected: %i to download' % (rejects, len(maglist)) logger.info(msg) for magazine in maglist: if magazine['nzbmode'] in ["torznab", "torrent", "magnet"]: snatch = TORDownloadMethod(magazine['bookid'], magazine['nzbtitle'], magazine['nzburl'], 'magazine') else: snatch = NZBDownloadMethod(magazine['bookid'], magazine['nzbtitle'], magazine['nzburl'], 'magazine') if snatch: logger.info( 'Downloading %s from %s' % (magazine['nzbtitle'], magazine["nzbprov"])) notify_snatch("Magazine %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()) finally: threading.currentThread().name = "WEBSERVER"
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_rss_book(books=None, reset=False): threadname = threading.currentThread().name if "Thread-" in threadname: threading.currentThread().name = "SEARCHRSS" if not (lazylibrarian.USE_RSS()): logger.warn('RSS search is disabled') scheduleJob(action='Stop', target='search_rss_book') return myDB = database.DBConnection() searchlist = [] if books is None: # We are performing a backlog search searchbooks = myDB.select( 'SELECT BookID, AuthorName, Bookname, BookSub, BookAdded from books WHERE Status="Wanted" order by BookAdded desc' ) else: # The user has added a new book searchbooks = [] for book in books: searchbook = myDB.select( 'SELECT BookID, AuthorName, BookName, BookSub from books WHERE BookID="%s" \ AND Status="Wanted"' % book['bookid']) for terms in searchbook: searchbooks.append(terms) if len(searchbooks) == 0: return logger.info('RSS Searching for %i book%s' % (len(searchbooks), plural(len(searchbooks)))) resultlist, nproviders = IterateOverRSSSites() if not nproviders: logger.warn('No rss providers are set, check config') return # No point in continuing dic = { '...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '' } 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: if '(' 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 = 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')
def processResultList(resultlist, book, searchtype): myDB = database.DBConnection() dictrepl = { '...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': ' ', '*': '', '(': '', ')': '', '[': '', ']': '', '#': '', '0': '', '1': '', '2': '', '3': '', '4': '', '5': '', '6': '', '7': '', '8': '', '9': '', '\'': '', ':': '', '!': '', '-': ' ', '\s\s': ' ' } # ' the ': ' ', ' a ': ' ', ' and ': ' ', # ' to ': ' ', ' of ': ' ', ' for ': ' ', ' my ': ' ', ' in ': ' ', ' at ': ' ', ' with ': ' '} dic = { '...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '', '\'': '' } match_ratio = int(lazylibrarian.MATCH_RATIO) reject_list = getList(lazylibrarian.REJECT_WORDS) author = unaccented_str(replace_all(book['authorName'], dic)) title = unaccented_str(replace_all(book['bookName'], dic)) matches = [] for nzb in resultlist: nzb_Title = unaccented_str(replace_all(nzb['nzbtitle'], dictrepl)).strip() nzb_Title = re.sub(r"\s\s+", " ", nzb_Title) # remove extra whitespace nzbAuthor_match = fuzz.token_set_ratio(author, nzb_Title) nzbBook_match = fuzz.token_set_ratio(title, nzb_Title) logger.debug(u"NZB author/book Match: %s/%s for %s" % (nzbAuthor_match, nzbBook_match, nzb_Title)) nzburl = nzb['nzburl'] rejected = False already_failed = myDB.action( 'SELECT * from wanted WHERE NZBurl="%s" and Status="Failed"' % nzburl).fetchone() if already_failed: logger.debug("Rejecting %s, blacklisted at %s" % (nzb_Title, already_failed['NZBprov'])) rejected = True if not rejected: for word in reject_list: if word in nzb_Title.lower( ) and not word in author.lower() and not word in title.lower(): rejected = True logger.debug("Rejecting %s, contains %s" % (nzb_Title, word)) break nzbsize_temp = nzb[ 'nzbsize'] # Need to cater for when this is NONE (Issue 35) if nzbsize_temp is None: nzbsize_temp = 1000 nzbsize = round(float(nzbsize_temp) / 1048576, 2) maxsize = check_int(lazylibrarian.REJECT_MAXSIZE, 0) if not rejected: if maxsize and nzbsize > maxsize: rejected = True logger.debug("Rejecting %s, too large" % nzb_Title) if not rejected: if nzbAuthor_match >= match_ratio and nzbBook_match >= match_ratio: bookid = book['bookid'] nzbTitle = (author + ' - ' + title + ' LL.(' + book['bookid'] + ')').strip() nzbprov = nzb['nzbprov'] nzbdate_temp = nzb['nzbdate'] nzbdate = nzbdate2format(nzbdate_temp) nzbmode = nzb['nzbmode'] controlValueDict = {"NZBurl": nzburl} newValueDict = { "NZBprov": nzbprov, "BookID": bookid, "NZBdate": now(), # when we asked for it "NZBsize": nzbsize, "NZBtitle": nzbTitle, "NZBmode": nzbmode, "Status": "Skipped" } score = (nzbBook_match + nzbAuthor_match) / 2 # as a percentage # lose a point for each extra word in the title so we get the closest match words = len(getList(nzb_Title)) words -= len(getList(author)) words -= len(getList(title)) score -= abs(words) matches.append( [score, nzb_Title, newValueDict, controlValueDict]) if matches: highest = max(matches, key=lambda x: x[0]) score = highest[0] nzb_Title = highest[1] newValueDict = highest[2] controlValueDict = highest[3] logger.info(u'Best match NZB (%s%%): %s using %s search' % (score, nzb_Title, searchtype)) snatchedbooks = myDB.action( 'SELECT * from books WHERE BookID="%s" and Status="Snatched"' % newValueDict["BookID"]).fetchone() if snatchedbooks: logger.debug('%s already marked snatched' % nzb_Title) return True # someone else found it else: logger.debug('%s adding to wanted' % nzb_Title) myDB.upsert("wanted", newValueDict, controlValueDict) if nzbmode == "torznab": snatch = TORDownloadMethod(newValueDict["BookID"], newValueDict["NZBprov"], newValueDict["NZBtitle"], controlValueDict["NZBurl"]) else: snatch = NZBDownloadMethod(newValueDict["BookID"], newValueDict["NZBprov"], newValueDict["NZBtitle"], controlValueDict["NZBurl"]) if snatch: notifiers.notify_snatch(newValueDict["NZBtitle"] + ' at ' + now()) scheduleJob(action='Start', target='processDir') return True + True # we found it else: logger.debug("No nzb's found for [%s] using searchtype %s" % (book["searchterm"], searchtype)) return False