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 = formatter.getList(lazylibrarian.REJECT_WORDS)
# bit of a misnomer now, rss can search both tor and nzb rss feeds
for tor in resultlist:
tor_Title = formatter.latinToAscii(formatter.replace_all(tor['tor_title'], dictrepl)).strip()
tor_Title = re.sub(r"\s\s+", " ", tor_Title) # remove extra whitespace
tor_Author_match = fuzz.token_set_ratio(author, tor_Title)
tor_Title_match = fuzz.token_set_ratio(title, tor_Title)
logger.debug("RSS Author/Title Match: %s/%s for %s" %(tor_Author_match, tor_Title_match, tor_Title))
rejected = False
for word in reject_list:
if word in tor_Title.lower() and not word in author.lower() and not word in book.lower():
rejected = True
logger.debug("Rejecting %s, contains %s" % (tor_Title, word))
break
if (tor_Title_match >= match_ratio and tor_Author_match >= match_ratio and not rejected):
logger.debug(u'Found RSS: %s' % tor['tor_title'])
bookid = book['bookid']
tor_Title = (book["authorName"] + ' - ' + book['bookName'] +
' LL.(' + book['bookid'] + ')').strip()
tor_url = tor['tor_url']
tor_prov = tor['tor_prov']
tor_feed = tor['tor_feed']
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 = str(round(float(tor_size_temp) / 1048576, 2)) + ' MB'
controlValueDict = {"NZBurl": tor_url}
newValueDict = {
"NZBprov": tor_prov,
"BookID": bookid,
"NZBdate": formatter.now(), # when we asked for it
"NZBsize": tor_size,
"NZBtitle": tor_Title,
"NZBmode": "torrent",
"Status": "Skipped"
}
myDB.upsert("wanted", newValueDict, controlValueDict)
snatchedbooks = myDB.action('SELECT * from books WHERE BookID="%s" and Status="Snatched"' %
bookid).fetchone()
if not snatchedbooks: # check if one of the other downloaders got there first
if '.nzb' in tor_url:
snatch = NZBDownloadMethod(bookid, tor_prov, tor_Title, tor_url)
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(bookid, tor_prov, tor_Title, tor_url)
if snatch:
notifiers.notify_snatch(formatter.latinToAscii(tor_Title) + ' at ' + formatter.now())
common.schedule_job(action='Start', target='processDir')
return True
logger.debug("No RSS found for " + (book["authorName"] + ' ' +
book['bookName']).strip())
return False
def search_magazines(mags=None, reset=False):
# produce a list of magazines to search for, tor, nzb, torznab
myDB = database.DBConnection()
searchlist = []
threading.currentThread().name = "SEARCHMAGS"
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) == 1:
logger.info('Searching for one magazine')
else:
logger.info('Searching for %i magazines' % len(searchmags))
for searchmag in searchmags:
bookid = searchmag[0]
searchterm = searchmag[0]
# frequency = searchmag[1]
# last_acquired = searchmag[2]
# issue_date = searchmag[3]
dic = {'...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': ''}
searchterm = formatter.latinToAscii(formatter.replace_all(searchterm, dic))
searchterm = re.sub('[\.\-\/]', ' ', searchterm).encode('utf-8')
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 = providers.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 = providers.IterateOverTorrentSites(book, 'mag')
if not nproviders:
logger.warn('No torrent providers are set. Check config for TORRENT providers')
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 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
to_snatch = 0
maglist = []
issues = []
reject_list = formatter.getList(lazylibrarian.REJECT_WORDS)
for nzb in resultlist:
total_nzbs = total_nzbs + 1
bookid = nzb['bookid']
nzbtitle = (u'%s' % 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 = str(round(float(nzbsize_temp) / 1048576, 2)) + ' MB'
nzbdate = formatter.nzbdate2format(nzbdate_temp)
nzbmode = nzb['nzbmode']
checkifmag = myDB.select('SELECT * from magazines WHERE Title="%s"' % bookid)
if checkifmag:
for results in checkifmag:
control_date = results['IssueDate']
# frequency = results['Frequency']
# regex = results['Regex']
nzbtitle_formatted = nzbtitle.replace('.', ' ').replace('-', ' ').replace('/', ' ').replace(
'+', ' ').replace('_', ' ').replace('(', '').replace(')', '').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
# keyword_check = nzbtitle_formatted.replace(bookid, '')
# 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
name_match = 1 # assume name matches for now
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(common.remove_accents(bookid), common.remove_accents(nzbtitle_formatted))
if mag_title_match < lazylibrarian.MATCH_RATIO:
logger.debug(u"Magazine token set Match failed: " + str(mag_title_match) + "% for " + nzbtitle_formatted)
name_match = 0
lower_title = common.remove_accents(nzbtitle_formatted).lower()
lower_bookid = common.remove_accents(bookid).lower()
for word in reject_list:
if word in lower_title and not word in lower_bookid:
name_match = 0
logger.debug("Rejecting %s, contains %s" % (nzbtitle_formatted, word))
break
if name_match:
# 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() == 'pdf':
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 = formatter.month2num(common.remove_accents(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].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
regexB_year = nzbtitle_exploded[len(nzbtitle_exploded) - 1]
regexB_month_temp = nzbtitle_exploded[len(nzbtitle_exploded) - 3]
regexB_month = formatter.month2num(common.remove_accents(regexB_month_temp))
regexB_day = nzbtitle_exploded[len(nzbtitle_exploded) - 2].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]
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:
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:
continue
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
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))
# 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 = formatter.datecompare(newdatish, control_date)
if comp_date > 0:
# Should probably only upsert when downloaded and processed in case snatch fails
# 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)
to_snatch = to_snatch + 1
issues.append(issue)
controlValueDict = {"NZBurl": nzburl}
newValueDict = {
"NZBprov": nzbprov,
"BookID": bookid,
"NZBdate": formatter.now(), # when we asked for it
"NZBtitle": nzbtitle,
"AuxInfo": newdatish,
"Status": "Wanted",
"NZBsize": nzbsize,
"NZBmode": nzbmode
}
myDB.upsert("wanted", newValueDict, controlValueDict)
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
else:
logger.debug('Magazine [%s] does not completely match search term [%s].' % (
nzbtitle_formatted, bookid))
bad_regex = bad_regex + 1
logger.info('Found %s results for %s. %s are new, %s are old, %s fail date, %s fail name matching' % (
total_nzbs, bookid, new_date, old_date, bad_date, bad_regex))
logger.info("%s, %s issues to download" % (bookid, to_snatch))
for items in maglist:
if items['nzbmode'] == "torznab":
snatch = TORDownloadMethod(items['bookid'], items['nzbprov'], items['nzbtitle'], items['nzburl'])
elif items['nzbmode'] == "torrent":
snatch = TORDownloadMethod(items['bookid'], items['nzbprov'], items['nzbtitle'], items['nzburl'])
else:
snatch = NZBDownloadMethod(items['bookid'], items['nzbprov'], items['nzbtitle'], items['nzburl'])
if snatch:
notifiers.notify_snatch(formatter.latinToAscii(items['nzbtitle']) + ' at ' + formatter.now())
common.schedule_job(action='Start', target='processDir')
maglist = []
if reset == True:
common.schedule_job(action='Restart', target='search_magazines')
logger.info("Search for magazines complete")
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 = formatter.getList(lazylibrarian.REJECT_WORDS)
author = formatter.latinToAscii(formatter.replace_all(book["authorName"], dic))
title = formatter.latinToAscii(formatter.replace_all(book["bookName"], dic))
matches = []
for nzb in resultlist:
nzb_Title = formatter.latinToAscii(formatter.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))
rejected = False
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 = formatter.check_int(lazylibrarian.REJECT_MAXSIZE, 0)
if maxsize and nzbsize > maxsize:
rejected = True
logger.debug("Rejecting %s, too large" % nzb_Title)
if nzbAuthor_match >= match_ratio and nzbBook_match >= match_ratio and not rejected:
# logger.debug(u'Found NZB: %s using %s search' % (nzb['nzbtitle'], searchtype))
bookid = book["bookid"]
nzbTitle = (author + " - " + title + " LL.(" + book["bookid"] + ")").strip()
nzburl = nzb["nzburl"]
nzbprov = nzb["nzbprov"]
nzbdate_temp = nzb["nzbdate"]
nzbdate = formatter.nzbdate2format(nzbdate_temp)
nzbmode = nzb["nzbmode"]
controlValueDict = {"NZBurl": nzburl}
newValueDict = {
"NZBprov": nzbprov,
"BookID": bookid,
"NZBdate": formatter.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(formatter.getList(nzb_Title))
words -= len(formatter.getList(author))
words -= len(formatter.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))
myDB.upsert("wanted", newValueDict, controlValueDict)
snatchedbooks = myDB.action(
'SELECT * from books WHERE BookID="%s" and Status="Snatched"' % newValueDict["BookID"]
).fetchone()
if not snatchedbooks:
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 " + formatter.now())
common.schedule_job(action="Start", target="processDir")
return True
logger.debug(
"No nzb's found for "
+ (book["authorName"] + " " + book["bookName"]).strip()
+ " using searchtype "
+ searchtype
)
return False
def search_rss_book(books=None, reset=False):
if not(lazylibrarian.USE_RSS()):
logger.warn('RSS search is disabled')
common.schedule_job(action='Stop', target='search_rss_book')
return
# rename this thread
threading.currentThread().name = "SEARCHRSSBOOKS"
myDB = database.DBConnection()
searchlist = []
if books is None:
# We are performing a backlog search
searchbooks = myDB.select('SELECT BookID, AuthorName, Bookname from books WHERE Status="Wanted"')
else:
# The user has added a new book
searchbooks = []
for book in books:
searchbook = myDB.select('SELECT BookID, AuthorName, BookName 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")
return
elif len(searchbooks) == 1:
logger.info('RSS Searching for one book')
else:
logger.info('RSS Searching for %i books' % len(searchbooks))
resultlist, nproviders = providers.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:
bookid = book['BookID']
author = book['AuthorName']
title = book['BookName']
author = formatter.latinToAscii(formatter.replace_all(author, dic))
title = formatter.latinToAscii(formatter.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))
else:
rss_count = rss_count + 1
if rss_count == 1:
logger.info("RSS Search for Wanted items complete, found %s book" % rss_count)
else:
logger.info("RSS Search for Wanted items complete, found %s books" % rss_count)
if reset == True:
common.schedule_job(action='Restart', target='search_rss_book')
def search_magazines(mags=None, reset=False):
# produce a list of magazines to search for, tor, nzb, torznab
myDB = database.DBConnection()
searchlist = []
threading.currentThread().name = "SEARCHMAGS"
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) == 1:
logger.info('Searching for one magazine')
else:
logger.info('Searching for %i magazines' % len(searchmags))
for searchmag in searchmags:
bookid = searchmag[0]
searchterm = searchmag[0]
# frequency = searchmag[1]
# last_acquired = searchmag[2]
# issue_date = searchmag[3]
dic = {'...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '', ',': '', '*': ''}
searchterm = formatter.latinToAscii(formatter.replace_all(searchterm, dic))
searchterm = re.sub('[\.\-\/]', ' ', searchterm).encode('utf-8')
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 = providers.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 = providers.IterateOverTorrentSites(book, 'mag')
if not nproviders:
logger.warn('No torrent providers are set. Check config for TORRENT providers')
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 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
to_snatch = 0
maglist = []
issues = []
reject_list = formatter.getList(lazylibrarian.REJECT_WORDS)
for nzb in resultlist:
total_nzbs = total_nzbs + 1
bookid = nzb['bookid']
nzbtitle = (u'%s' % 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 = str(round(float(nzbsize_temp) / 1048576, 2)) + ' MB'
nzbdate = formatter.nzbdate2format(nzbdate_temp)
nzbmode = nzb['nzbmode']
checkifmag = myDB.select('SELECT * from magazines WHERE Title="%s"' % bookid)
if checkifmag:
for results in checkifmag:
control_date = results['IssueDate']
# frequency = results['Frequency']
# regex = results['Regex']
nzbtitle_formatted = nzbtitle.replace('.', ' ').replace('-', ' ').replace('/', ' ').replace(
'+', ' ').replace('_', ' ').replace('(', '').replace(')', '').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
# keyword_check = nzbtitle_formatted.replace(bookid, '')
# 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
name_match = 1 # assume name matches for now
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(
common.remove_accents(bookid),
common.remove_accents(nzbtitle_formatted))
if mag_title_match < lazylibrarian.MATCH_RATIO:
logger.debug(
u"Magazine token set Match failed: " + str(
mag_title_match) + "% for " + nzbtitle_formatted)
name_match = 0
lower_title = common.remove_accents(nzbtitle_formatted).lower()
lower_bookid = common.remove_accents(bookid).lower()
for word in reject_list:
if word in lower_title and not word in lower_bookid:
name_match = 0
logger.debug("Rejecting %s, contains %s" % (nzbtitle_formatted, word))
break
if name_match:
# 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() == 'pdf':
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 = formatter.month2num(common.remove_accents(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 = formatter.month2num(common.remove_accents(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]
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:
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:
continue
# store all the _new_ matching results, marking as "skipped" for now
# we change the status to "wanted" on the ones we want to snatch later
# don't add a new entry if this issue has been found on an earlier search
# because status might have been user-set
mag_entry = myDB.select('SELECT * from wanted WHERE NZBtitle="%s" and NZBprov="%s"' % (nzbtitle, nzbprov))
if not mag_entry:
controlValueDict = {
"NZBtitle": nzbtitle,
"NZBprov": nzbprov
}
newValueDict = {
"NZBurl": nzburl,
"BookID": bookid,
"NZBdate": nzbdate,
"AuxInfo": newdatish,
"Status": "Skipped",
"NZBsize": nzbsize,
"NZBmode": nzbmode
}
myDB.upsert("wanted", newValueDict, controlValueDict)
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
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))
# 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 = formatter.datecompare(newdatish, control_date)
if comp_date > 0:
# Should probably only upsert when downloaded and processed in case snatch fails
# 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)
to_snatch = to_snatch + 1
issues.append(issue)
controlValueDict = {"NZBurl": nzburl}
newValueDict = {
"NZBdate": formatter.now(), # when we asked for it
"Status": "Wanted"
}
myDB.upsert("wanted", newValueDict, controlValueDict)
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
else:
logger.debug('Magazine [%s] does not completely match search term [%s].' % (
nzbtitle_formatted, bookid))
bad_regex = bad_regex + 1
logger.info('Found %i results for %s. %i new, %i old, %i fail date, %i fail name: %i to download' % (
total_nzbs, bookid, new_date, old_date, bad_date, bad_regex, to_snatch))
for items in maglist:
if items['nzbmode'] == "torznab":
snatch = TORDownloadMethod(items['bookid'], items['nzbprov'], items['nzbtitle'], items['nzburl'])
elif items['nzbmode'] == "torrent":
snatch = TORDownloadMethod(items['bookid'], items['nzbprov'], items['nzbtitle'], items['nzburl'])
else:
snatch = NZBDownloadMethod(items['bookid'], items['nzbprov'], items['nzbtitle'], items['nzburl'])
if snatch:
notifiers.notify_snatch(formatter.latinToAscii(items['nzbtitle']) + ' at ' + formatter.now())
common.schedule_job(action='Start', target='processDir')
maglist = []
if reset:
common.schedule_job(action='Restart', target='search_magazines')
logger.info("Search for magazines complete")
def search_tor_book(books=None, reset=False):
if not lazylibrarian.USE_TOR():
logger.warn('No Torrent providers set, check config')
return
# rename this thread
threading.currentThread().name = "SEARCHTORBOOKS"
myDB = database.DBConnection()
searchlist = []
if books is None:
# We are performing a backlog search
searchbooks = myDB.select('SELECT BookID, AuthorName, Bookname, 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 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
elif len(searchbooks) == 1:
logger.info('TOR Searching for one book')
else:
logger.info('TOR Searching for %i books' % len(searchbooks))
for searchbook in searchbooks:
bookid = searchbook['BookID']
author = searchbook['AuthorName']
book = searchbook['BookName']
dic = {'...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '',
',': '', '*': '', ':': '', ';': ''}
dicSearchFormatting = {'.': ' +', ' + ': ' '}
author = formatter.latinToAscii(formatter.replace_all(author, dic))
book = formatter.latinToAscii(formatter.replace_all(book, dic))
# TRY SEARCH TERM just using author name and book type
author = formatter.latinToAscii(formatter.replace_all(author, dicSearchFormatting))
searchterm = author + ' ' + book # + ' ' + lazylibrarian.EBOOK_TYPE
searchterm = re.sub('[\.\-\/]', ' ', searchterm).encode('utf-8')
searchterm = re.sub(r'\(.*?\)', '', searchterm).encode('utf-8')
searchterm = re.sub(r"\s\s+", " ", searchterm) # strip any double white space
searchlist.append({"bookid": bookid, "bookName": searchbook[2], "authorName": searchbook[1],
"searchterm": searchterm.strip()})
tor_count = 0
for book in searchlist:
resultlist, nproviders = providers.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 = providers.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 = providers.IterateOverTorrentSites(book, 'general')
found = processResultList(resultlist, book, "general")
# if you still can't find the book, try with author only
if not found:
resultlist, nproviders = providers.IterateOverTorrentSites(book, 'author')
found = processResultList(resultlist, book, "author")
if not found:
logger.debug("Searches returned no results. Adding book %s to queue." % book['searchterm'])
else:
tor_count = tor_count + 1
if tor_count == 1:
logger.info("TORSearch for Wanted items complete, found %s book" % tor_count)
else:
logger.info("TORSearch for Wanted items complete, found %s books" % tor_count)
if reset:
common.schedule_job(action='Restart', target='search_tor_book')
def search_rss_book(books=None, reset=False):
if not(lazylibrarian.USE_RSS()):
logger.warn('RSS search is disabled')
common.schedule_job(action='Stop', target='search_rss_book')
return
# rename this thread
threading.currentThread().name = "SEARCHRSSBOOKS"
myDB = database.DBConnection()
searchlist = []
if books is None:
# We are performing a backlog search
searchbooks = myDB.select('SELECT BookID, AuthorName, Bookname, 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 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
elif len(searchbooks) == 1:
logger.info('RSS Searching for one book')
else:
logger.info('RSS Searching for %i books' % len(searchbooks))
resultlist, nproviders = providers.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:
bookid = book['BookID']
author = book['AuthorName']
title = book['BookName']
author = formatter.latinToAscii(formatter.replace_all(author, dic))
title = formatter.latinToAscii(formatter.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))
else:
rss_count = rss_count + 1
plural = "s"
if rss_count == 1:
plural = ""
logger.info("RSS Search for Wanted items complete, found %s book%s" % (rss_count, plural))
if reset:
common.schedule_job(action='Restart', target='search_rss_book')
def search_tor_book(books=None, reset=False):
if not lazylibrarian.USE_TOR():
logger.warn("No Torrent providers set, check config")
return
# rename this thread
threading.currentThread().name = "SEARCHTORBOOKS"
myDB = database.DBConnection()
searchlist = []
if books is None:
# We are performing a backlog search
searchbooks = myDB.select('SELECT BookID, AuthorName, Bookname from books WHERE Status="Wanted"')
else:
# The user has added a new book
searchbooks = []
for book in books:
searchbook = myDB.select(
'SELECT BookID, AuthorName, BookName 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")
return
elif len(searchbooks) == 1:
logger.info("TOR Searching for one book")
else:
logger.info("TOR Searching for %i books" % len(searchbooks))
for searchbook in searchbooks:
bookid = searchbook["BookID"]
author = searchbook["AuthorName"]
book = searchbook["BookName"]
dic = {
"...": "",
".": " ",
" & ": " ",
" = ": " ",
"?": "",
"$": "s",
" + ": " ",
'"': "",
",": "",
"*": "",
":": "",
";": "",
}
dicSearchFormatting = {".": " +", " + ": " "}
author = formatter.latinToAscii(formatter.replace_all(author, dic))
book = formatter.latinToAscii(formatter.replace_all(book, dic))
# TRY SEARCH TERM just using author name and book type
author = formatter.latinToAscii(formatter.replace_all(author, dicSearchFormatting))
searchterm = author + " " + book # + ' ' + lazylibrarian.EBOOK_TYPE
searchterm = re.sub("[\.\-\/]", " ", searchterm).encode("utf-8")
searchterm = re.sub(r"\(.*?\)", "", searchterm).encode("utf-8")
searchterm = re.sub(r"\s\s+", " ", searchterm) # strip any double white space
searchlist.append(
{"bookid": bookid, "bookName": searchbook[2], "authorName": searchbook[1], "searchterm": searchterm.strip()}
)
tor_count = 0
for book in searchlist:
resultlist, nproviders = providers.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 = providers.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 = providers.IterateOverTorrentSites(book, "general")
found = processResultList(resultlist, book, "general")
# if you still can't find the book, try with author only
if not found:
resultlist, nproviders = providers.IterateOverTorrentSites(book, "author")
found = processResultList(resultlist, book, "author")
if not found:
logger.debug("Searches returned no results. Adding book %s to queue." % book["searchterm"])
else:
tor_count = tor_count + 1
if tor_count == 1:
logger.info("TORSearch for Wanted items complete, found %s book" % tor_count)
else:
logger.info("TORSearch for Wanted items complete, found %s books" % tor_count)
if reset == True:
common.schedule_job(action="Restart", target="search_tor_book")
def start():
global __INITIALIZED__, started
if __INITIALIZED__:
# Crons and scheduled jobs go here
# list is duplicated in webServe so we can reschedule them
SCHED.start()
common.schedule_job("Start", "processDir")
common.schedule_job("Start", "search_nzb_book")
common.schedule_job("Start", "search_tor_book")
common.schedule_job("Start", "search_rss_book")
common.schedule_job("Start", "search_magazines")
common.schedule_job("Start", "checkForUpdates")
started = True
def processDir(force=False, reset=False):
# rename this thread
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:
logger.error('Could not access [%s] directory ' % processpath)
return False
myDB = database.DBConnection()
snatched = myDB.select('SELECT * from wanted WHERE Status="Snatched"')
if force is False and len(snatched) == 0:
logger.info('Nothing marked as snatched. Stopping postprocessor job.')
common.schedule_job(action='Stop', target='processDir')
elif len(downloads) == 0:
logger.info('No downloads are found. Nothing to process.')
else:
ppcount = 0
for book in snatched:
found = False
for fname in downloads:
if not fname.endswith('.fail'): # has this failed before?
# this is to get round unicode differences in torrent filenames.
# there might be a better way...
if isinstance(fname, str):
matchname = fname.decode('utf-8')
else:
matchname = fname
if 'LL.(' in matchname:
matchname = matchname.split('LL.(')[0]
match = fuzz.token_set_ratio(matchname, book['NZBtitle'])
if match >= 95:
pp_path = os.path.join(processpath, fname)
logger.debug('Found folder %s for %s' % (pp_path, book['NZBtitle']))
found = True
break
if found:
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)
# 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 = {'<': '', '>': '', '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's',
' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '', '\'': ''}
dest_path = formatter.latinToAscii(formatter.replace_all(dest_path, dic))
dest_path = os.path.join(lazylibrarian.DESTINATION_DIR, dest_path).encode(
lazylibrarian.SYS_ENCODING)
else:
data = myDB.select('SELECT * from magazines WHERE Title="%s"' % book['BookID'])
if data:
# AuxInfo was added for magazine release date, normally housed in 'magazines' but if multiple
# files are downloading, there will be an error in post-processing, trying to go to the
# same directory.
mostrecentissue = data[0]['IssueDate'] # keep for processing issues arriving out of order
# Remove characters we don't want in the filename before (maybe) adding to DESTINATION_DIR
# as windows drive identifiers have colon, eg c: but no colons allowed elsewhere?
dic = {'<': '', '>': '', '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's',
' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '', '\'': ''}
mag_name = formatter.latinToAscii(formatter.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 = 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
# try:
# os.chmod(dest_path, 0777)
# except Exception, e:
# logger.debug("Could not chmod post-process directory: " + str(dest_path))
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": formatter.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
processExtras(myDB, dest_path, global_name, data)
else:
# update mags
controlValueDict = {"Title": book['BookID']}
if mostrecentissue > book['AuxInfo']: # check this in case processing issues arriving out of order
newValueDict = {"LastAcquired": formatter.today(), "IssueStatus": "Open"}
else:
newValueDict = {"IssueDate": book['AuxInfo'], "LastAcquired": formatter.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": formatter.today(),
"IssueFile": dest_file,
"IssueID" : magazinescan.create_id("%s %s" % (book['BookID'], book['AuxInfo']))
}
myDB.upsert("issues", newValueDict, controlValueDict)
# create a thumbnail cover for the new issue
magazinescan.create_cover(dest_file)
logger.info('Successfully processed: %s' % global_name)
ppcount = ppcount + 1
notifiers.notify_download(formatter.latinToAscii(global_name) + ' at ' + formatter.now())
else:
logger.error('Postprocessing for %s has failed.' % global_name)
logger.error('Warning - Residual files remain in %s.fail' % pp_path)
# 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:
logger.info('%s books/mags have been processed.' % ppcount)
else:
logger.info('No snatched books/mags have been found')
if reset == True:
common.schedule_job(action='Restart', target='processDir')
def search_tor_book(books=None, reset=False):
if not lazylibrarian.USE_TOR():
logger.warn('No Torrent providers set, check config')
return
# rename this thread
threading.currentThread().name = "SEARCHTORBOOKS"
myDB = database.DBConnection()
searchlist = []
if books is None:
# We are performing a backlog search
searchbooks = myDB.select(
'SELECT BookID, AuthorName, Bookname, 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 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
elif len(searchbooks) == 1:
logger.info('TOR Searching for one book')
else:
logger.info('TOR Searching for %i books' % len(searchbooks))
for searchbook in searchbooks:
bookid = searchbook['BookID']
author = searchbook['AuthorName']
book = searchbook['BookName']
dic = {
'...': '',
'.': ' ',
' & ': ' ',
' = ': ' ',
'?': '',
'$': 's',
' + ': ' ',
'"': '',
',': '',
'*': '',
':': '',
';': ''
}
dicSearchFormatting = {'.': ' +', ' + ': ' '}
author = formatter.latinToAscii(formatter.replace_all(author, dic))
book = formatter.latinToAscii(formatter.replace_all(book, dic))
# TRY SEARCH TERM just using author name and book type
author = formatter.latinToAscii(
formatter.replace_all(author, dicSearchFormatting))
searchterm = author + ' ' + book # + ' ' + lazylibrarian.EBOOK_TYPE
searchterm = re.sub('[\.\-\/]', ' ', searchterm).encode('utf-8')
searchterm = re.sub(r'\(.*?\)', '', searchterm).encode('utf-8')
searchterm = re.sub(r"\s\s+", " ",
searchterm) # strip any double white space
searchlist.append({
"bookid": bookid,
"bookName": searchbook[2],
"authorName": searchbook[1],
"searchterm": searchterm.strip()
})
tor_count = 0
for book in searchlist:
resultlist, nproviders = providers.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 = providers.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 = providers.IterateOverTorrentSites(
book, 'general')
found = processResultList(resultlist, book, "general")
# if you still can't find the book, try with author only
if not found:
resultlist, nproviders = providers.IterateOverTorrentSites(
book, 'author')
found = processResultList(resultlist, book, "author")
if not found:
logger.debug(
"Searches returned no results. Adding book %s to queue." %
book['searchterm'])
else:
tor_count = tor_count + 1
if tor_count == 1:
logger.info("TORSearch for Wanted items complete, found %s book" %
tor_count)
else:
logger.info("TORSearch for Wanted items complete, found %s books" %
tor_count)
if reset:
common.schedule_job(action='Restart', target='search_tor_book')
def processDir(force=False, reset=False):
# rename this thread
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 False
myDB = database.DBConnection()
snatched = myDB.select('SELECT * from wanted WHERE Status="Snatched"')
if force is False and len(snatched) == 0:
logger.info('Nothing marked as snatched. Stopping postprocessor job.')
common.schedule_job(action='Stop', target='processDir')
elif len(downloads) == 0:
logger.info('No downloads are found. Nothing to process.')
else:
logger.debug("Checking %s downloads for %s snatched files" % (len(downloads), len(snatched)))
ppcount = 0
for book in snatched:
found = False
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']
if ' LL.(' in matchtitle:
matchtitle = matchtitle.split(' LL.(')[0]
match = fuzz.token_set_ratio(matchtitle, matchname)
if match >= 95:
fname = matchname
if os.path.isfile(os.path.join(processpath, fname)):
# handle single file downloads here...
if formatter.is_valid_booktype(fname, booktype="book") \
or formatter.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 for %s' % (pp_path, book['NZBtitle']))
found = True
break
else:
logger.debug('No match (%s%%) %s for %s' % (match, matchname, matchtitle))
else:
logger.debug('Skipping %s' % fname)
if found:
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 = common.remove_accents(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 = {'<': '', '>': '', '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's',
' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '', '\'': ''}
dest_path = formatter.latinToAscii(formatter.replace_all(dest_path, dic))
dest_path = os.path.join(lazylibrarian.DESTINATION_DIR, dest_path).encode(
lazylibrarian.SYS_ENCODING)
else:
data = myDB.select('SELECT * from magazines WHERE Title="%s"' % book['BookID'])
if data:
# AuxInfo was added for magazine release date, normally housed in 'magazines' but if multiple
# files are downloading, there will be an error in post-processing, trying to go to the
# same directory.
mostrecentissue = data[0]['IssueDate'] # keep for processing issues arriving out of order
# Remove characters we don't want in the filename before (maybe) adding to DESTINATION_DIR
# as windows drive identifiers have colon, eg c: but no colons allowed elsewhere?
dic = {'<': '', '>': '', '...': '', ' & ': ' ', ' = ': ' ', '?': '', '$': 's',
' + ': ' ', '"': '', ',': '', '*': '', ':': '', ';': '', '\'': ''}
mag_name = formatter.latinToAscii(formatter.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 = common.remove_accents(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
# try:
# os.chmod(dest_path, 0777)
# except Exception, e:
# logger.debug("Could not chmod post-process directory: " + str(dest_path))
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": formatter.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
processExtras(myDB, dest_path, global_name, data)
else:
# update mags
controlValueDict = {"Title": book['BookID']}
if mostrecentissue > book['AuxInfo']: # check this in case processing issues arriving out of order
newValueDict = {"LastAcquired": formatter.today(), "IssueStatus": "Open"}
else:
newValueDict = {"IssueDate": book['AuxInfo'], "LastAcquired": formatter.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": formatter.today(),
"IssueFile": dest_file,
"IssueID": magazinescan.create_id("%s %s" % (book['BookID'], book['AuxInfo']))
}
myDB.upsert("issues", newValueDict, controlValueDict)
# create a thumbnail cover for the new issue
magazinescan.create_cover(dest_file)
logger.info('Successfully processed: %s' % global_name)
ppcount = ppcount + 1
notifiers.notify_download(formatter.latinToAscii(global_name) + ' at ' + formatter.now())
else:
logger.error('Postprocessing for %s has failed.' % global_name)
logger.error('Warning - Residual files remain in %s.fail' % pp_path)
# 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:
logger.info('%s books/mags have been processed.' % ppcount)
else:
logger.info('No snatched books/mags have been found')
if reset:
common.schedule_job(action='Restart', target='processDir')
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 = formatter.getList(lazylibrarian.REJECT_WORDS)
matches = []
for tor in resultlist:
torTitle = formatter.latinToAscii(formatter.replace_all(str(tor['tor_title']), dictrepl)).strip()
torTitle = re.sub(r"\s\s+", " ", torTitle) # remove extra whitespace
author = formatter.latinToAscii(formatter.replace_all(book['authorName'], dic))
title = formatter.latinToAscii(formatter.replace_all(book['bookName'], dic))
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))
rejected = False
for word in reject_list:
if word in torTitle.lower() and not word in author.lower() and not word in title.lower():
rejected = True
logger.debug("Rejecting %s, contains %s" % (torTitle, word))
break
if (torAuthor_match >= match_ratio and torBook_match >= match_ratio and not rejected):
#logger.debug(u'Found Torrent: %s using %s search' % (tor['tor_title'], searchtype))
bookid = book['bookid']
tor_Title = (author + ' - ' + title + ' LL.(' + book['bookid'] + ')').strip()
tor_url = tor['tor_url']
tor_prov = tor['tor_prov']
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 = str(round(float(tor_size_temp) / 1048576, 2)) + ' MB'
controlValueDict = {"NZBurl": tor_url}
newValueDict = {
"NZBprov": tor_prov,
"BookID": bookid,
"NZBdate": formatter.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(formatter.getList(torTitle))
words -= len(formatter.getList(author))
words -= len(formatter.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 TOR (%s%%): %s using %s search' %
(score, nzb_Title, searchtype))
myDB.upsert("wanted", newValueDict, controlValueDict)
snatchedbooks = myDB.action('SELECT * from books WHERE BookID="%s" and Status="Snatched"' %
newValueDict["BookID"]).fetchone()
if not snatchedbooks:
snatch = TORDownloadMethod(newValueDict["BookID"], newValueDict["NZBprov"],
newValueDict["NZBtitle"], controlValueDict["NZBurl"])
if snatch:
notifiers.notify_snatch(newValueDict["NZBtitle"] + ' at ' + formatter.now())
common.schedule_job(action='Start', target='processDir')
return True
logger.debug("No torrent's found for " + (book["authorName"] + ' ' +
book['bookName']).strip() + " using searchtype " + searchtype)
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 = formatter.getList(lazylibrarian.REJECT_WORDS)
for nzb in resultlist:
nzbTitle = formatter.latinToAscii(formatter.replace_all(nzb['nzbtitle'], dictrepl)).strip()
nzbTitle = re.sub(r"\s\s+", " ", nzbTitle) # remove extra whitespace
author = formatter.latinToAscii(formatter.replace_all(book['authorName'], dic))
title = formatter.latinToAscii(formatter.replace_all(book['bookName'], dic))
# nzbTitle_match = fuzz.token_set_ratio(book['searchterm'], nzbTitle)
# logger.debug(u"NZB Title sort Match %: " + str(nzbTitle_match) + " for " + nzbTitle)
nzbAuthor_match = fuzz.token_set_ratio(author, nzbTitle)
nzbBook_match = fuzz.token_set_ratio(title, nzbTitle)
logger.debug(u"NZB author/book Match: %s/%s for %s" % (nzbAuthor_match, nzbBook_match, nzbTitle))
rejected = False
for word in reject_list:
if word in nzbTitle.lower() and not word in author.lower() and not word in title.lower():
rejected = True
logger.debug("Rejecting %s, contains %s" % (nzbTitle, word))
break
if (nzbAuthor_match >= match_ratio and nzbBook_match >= match_ratio and not rejected):
logger.debug(u'Found NZB: %s using %s search' % (nzb['nzbtitle'], searchtype))
bookid = book['bookid']
nzbTitle = (author + ' - ' + title + ' LL.(' + book['bookid'] + ')').strip()
nzburl = nzb['nzburl']
nzbprov = nzb['nzbprov']
nzbdate_temp = nzb['nzbdate']
nzbsize_temp = nzb['nzbsize'] # Need to cater for when this is NONE (Issue 35)
if nzbsize_temp is None:
nzbsize_temp = 1000
nzbsize = str(round(float(nzbsize_temp) / 1048576, 2)) + ' MB'
nzbdate = formatter.nzbdate2format(nzbdate_temp)
nzbmode = nzb['nzbmode']
controlValueDict = {"NZBurl": nzburl}
newValueDict = {
"NZBprov": nzbprov,
"BookID": bookid,
"NZBdate": formatter.now(), # when we asked for it
"NZBsize": nzbsize,
"NZBtitle": nzbTitle,
"NZBmode": nzbmode,
"Status": "Skipped"
}
myDB.upsert("wanted", newValueDict, controlValueDict)
snatchedbooks = myDB.action('SELECT * from books WHERE BookID="%s" and Status="Snatched"' %
bookid).fetchone()
if not snatchedbooks:
if nzbmode == "torznab":
snatch = TORDownloadMethod(bookid, nzbprov, nzbTitle, nzburl)
else:
snatch = NZBDownloadMethod(bookid, nzbprov, nzbTitle, nzburl)
if snatch:
notifiers.notify_snatch(nzbTitle + ' at ' + formatter.now())
common.schedule_job(action='Start', target='processDir')
return True
logger.debug("No nzb's found for " + (book["authorName"] + ' ' + book['bookName']).strip() +
" using searchtype " + 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 = formatter.getList(lazylibrarian.REJECT_WORDS)
matches = []
# bit of a misnomer now, rss can search both tor and nzb rss feeds
for tor in resultlist:
torTitle = formatter.latinToAscii(formatter.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))
rejected = False
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 = formatter.check_int(lazylibrarian.REJECT_MAXSIZE, 0)
if maxsize and tor_size > maxsize:
rejected = True
logger.debug("Rejecting %s, too large" % torTitle)
if (tor_Title_match >= match_ratio and tor_Author_match >= match_ratio and not rejected):
#logger.debug(u'Found RSS: %s' % tor['tor_title'])
bookid = book['bookid']
tor_Title = (book["authorName"] + ' - ' + book['bookName'] +
' LL.(' + book['bookid'] + ')').strip()
tor_url = tor['tor_url']
tor_prov = tor['tor_prov']
tor_feed = tor['tor_feed']
controlValueDict = {"NZBurl": tor_url}
newValueDict = {
"NZBprov": tor_prov,
"BookID": bookid,
"NZBdate": formatter.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(formatter.getList(torTitle))
words -= len(formatter.getList(author))
words -= len(formatter.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))
myDB.upsert("wanted", newValueDict, controlValueDict)
snatchedbooks = myDB.action('SELECT * from books WHERE BookID="%s" and Status="Snatched"' %
newValueDict["BookID"]).fetchone()
if not snatchedbooks: # check if one of the other downloaders got there first
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:
notifiers.notify_snatch(newValueDict["NZBtitle"] + ' at ' + formatter.now())
common.schedule_job(action='Start', target='processDir')
return True
logger.debug("No RSS found for " + (book["authorName"] + ' ' +
book['bookName']).strip())
return False
def search_nzb_book(books=None, reset=False):
if not lazylibrarian.USE_NZB():
logger.warn('No NEWZNAB/TORZNAB providers set, check config')
return
# rename this thread
threading.currentThread().name = "SEARCHNZBBOOKS"
myDB = database.DBConnection()
searchlist = []
if books is None:
# We are performing a backlog search
searchbooks = myDB.select('SELECT BookID, AuthorName, Bookname, 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 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
elif len(searchbooks) == 1:
logger.info('NZB Searching for one book')
else:
logger.info('NZB Searching for %i books' % len(searchbooks))
for searchbook in searchbooks:
bookid = searchbook[0]
author = searchbook[1]
book = searchbook[2]
dic = {'...': '', '.': ' ', ' & ': ' ', ' = ': ' ', '?': '', '$': 's', ' + ': ' ', '"': '',
',': '', '*': '', ':': '', ';': ''}
dicSearchFormatting = {'.': ' +', ' + ': ' '}
author = formatter.latinToAscii(formatter.replace_all(author, dic))
book = formatter.latinToAscii(formatter.replace_all(book, dic))
if '(' in book: # may have title (series/extended info)
book = book.split('(')[0]
# TRY SEARCH TERM just using author name and book
author = formatter.latinToAscii(formatter.replace_all(author, dicSearchFormatting))
searchterm = author + ' ' + book
searchterm = re.sub('[\.\-\/]', ' ', searchterm).encode('utf-8')
searchterm = re.sub(r'\(.*?\)', '', searchterm).encode('utf-8')
searchterm = re.sub(r"\s\s+", " ", searchterm) # strip any double white space
searchlist.append({"bookid": bookid, "bookName": searchbook[2], "authorName": searchbook[1],
"searchterm": searchterm.strip()})
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 you still can't find the book, try with author only
if not found:
resultlist, nproviders = providers.IterateOverNewzNabSites(book, 'author')
found = processResultList(resultlist, book, "author")
if not found:
logger.debug("NZB Searches returned no results. Adding book %s to queue." % book['searchterm'])
else:
nzb_count = nzb_count + 1
if nzb_count == 1:
logger.info("NZBSearch for Wanted items complete, found %s book" % nzb_count)
else:
logger.info("NZBSearch for Wanted items complete, found %s books" % nzb_count)
if reset:
common.schedule_job(action='Restart', target='search_nzb_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 = formatter.getList(lazylibrarian.REJECT_WORDS)
for tor in resultlist:
tor_Title = formatter.latinToAscii(formatter.replace_all(str(tor["tor_title"]), dictrepl)).strip()
tor_Title = re.sub(r"\s\s+", " ", tor_Title) # remove extra whitespace
author = formatter.latinToAscii(formatter.replace_all(book["authorName"], dic))
title = formatter.latinToAscii(formatter.replace_all(book["bookName"], dic))
torAuthor_match = fuzz.token_set_ratio(author, tor_Title)
torBook_match = fuzz.token_set_ratio(title, tor_Title)
logger.debug(u"TOR author/book Match: %s/%s for %s" % (torAuthor_match, torBook_match, tor_Title))
# tor_Title_match = fuzz.token_set_ratio(book['searchterm'], tor_Title)
# logger.debug("Torrent Title Match %: " + str(tor_Title_match) + " for " + tor_Title)
# if (tor_Title_match >= match_ratio):
rejected = False
for word in reject_list:
if word in tor_Title.lower() and not word in author.lower() and not word in title_lower():
rejected = True
logger.debug("Rejecting %s, contains %s" % (tor_Title, word))
break
if torAuthor_match >= match_ratio and torBook_match >= match_ratio and not rejected:
logger.debug(u"Found Torrent: %s using %s search" % (tor["tor_title"], searchtype))
bookid = book["bookid"]
tor_Title = (author + " - " + title + " LL.(" + book["bookid"] + ")").strip()
tor_url = tor["tor_url"]
tor_prov = tor["tor_prov"]
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 = str(round(float(tor_size_temp) / 1048576, 2)) + " MB"
controlValueDict = {"NZBurl": tor_url}
newValueDict = {
"NZBprov": tor_prov,
"BookID": bookid,
"NZBdate": formatter.now(), # when we asked for it
"NZBsize": tor_size,
"NZBtitle": tor_Title,
"NZBmode": "torrent",
"Status": "Skipped",
}
myDB.upsert("wanted", newValueDict, controlValueDict)
snatchedbooks = myDB.action(
'SELECT * from books WHERE BookID="%s" and Status="Snatched"' % bookid
).fetchone()
if not snatchedbooks:
snatch = TORDownloadMethod(bookid, tor_prov, tor_Title, tor_url)
if snatch:
notifiers.notify_snatch(formatter.latinToAscii(tor_Title) + " at " + formatter.now())
common.schedule_job(action="Start", target="processDir")
return True
logger.debug(
"No torrent's found for "
+ (book["authorName"] + " " + book["bookName"]).strip()
+ " using searchtype "
+ searchtype
)
return False