piexel-indexer/main.py

243 lines
8.4 KiB
Python

#coding:utf-8
'''
Pierre Cadart
Script pour NAS sous Linux
Utilise un accès en FTP
'''
import posixpath
import re
import config
import piexel
import tokenizer
import filerule
from ftplib import FTP
import time
import file
def ftpwalk(directory, ftp):
"""
Parcours en profondeur un dossier du serveur FTP
Effectue un parcours préfixe
"""
to_visit = [directory]
while len(to_visit) > 0:
current = to_visit.pop(0)
ftp.cwd(current)
Lfiles = []
Ldirs = []
for name, prop in ftp.mlsd():
if name.startswith('.'):
continue
if prop['type'] == 'dir':
Ldirs.append(name)
to_visit.append(current+'/'+name)
elif prop['type'] == 'file':
Lfiles.append(name)
# ne construit pas la liste complète,
# mais retourne les résultats intermédiaires
yield (current, Ldirs, Lfiles)
def visit_folder(domain, api, rules, tok):
# Connection au serveur
print('connect to:', domain['server'])
ftp = FTP(domain['server'][6:], user=domain['username'], passwd=domain['password'])
ftp.encoding = 'UTF-8'
# Initialisation des listes de mises à jour
L_missing = [] # fichiers non trouvés sur le serveur FTP
L_unreferenced = [] # fichiers non référencés dans l'API
L_moved = [] # fichiers déplacés sur le serveur FTP
# Lecture des fichiers sur le serveur FTP
Lloc = []
for path, _, files in ftpwalk(domain['path'], ftp):
# Ajoute les fichiers correspondants aux extensions
for f in files:
match = filerule.match_rules(path+'/'+f, rules)
if match:
#print('got match:',match[1], 'name:',path+'/'+f)
F = file.File(path, f, match[1])
F.extract_title(tok)
Lloc.append(F)
ftp.close()
print('total:',len(Lloc))
# Récupère les fichiers de l'api
Lapi = []
for info in api.get_files(path=domain['server']+domain['path'], like=1, filable=1):
nfo = {}
if not info['filable']:
print('nfo:', info)
else:
year = int(info['filable']['release_date'][:4])
nfo['YEAR'] = year
F = file.File(info['path'][len(domain['server']):], info['name'], nfo, api_id=info['filable_id'])
F.extract_title(tok)
Lapi.append(F)
print('got api for ',domain['server']+domain['path'],':', len(Lapi))
#print('loc titles:', '|'.join(sorted([f.title for f in Lloc])))
#print('loc titles:', '|'.join(sorted([f.title for f in Lloc])))
#print('\n'*2)
#print('api titles:', '|'.join(sorted([f.title for f in Lapi])))
# supprime les dossiers de l'api
Lapi = [f for f in Lapi if tok.conf.is_valid_file(f.name)]
# Compare avec la liste de l'api
Lmissing = [f for f in Lapi if f not in Lloc] # fichiers non présents localement
Lunref = [f for f in Lloc if f not in Lapi] # fichiers non référencés
# Fichiers déplacés localement
Lrelink = [] # liste des références à changer
for file1 in Lmissing:
for file2 in Lunref:
if file1.filename_same(file2):
Lrelink.append((file1, file2))
for fApi, fLoc in Lrelink:
Lmissing.remove(fApi)
Lunref.remove(fLoc)
print('moved:', Lrelink)
# Linke les fichiers identiques au même film
Llink = []
for file1 in Lunref:
for file2 in Lapi:
if file1.title == file2.title:
Llink.append((file1, file2.api_id))
print('D add:', file1, file2, file2.api_id)
break
print('doubles:', sorted(Llink, key=lambda f:str(f)))
for f, _ in Llink:
Lunref.remove(f)
# Linke les films par nom si possible
APIfilms = api.get_films()
API_alltitles = []
for f in APIfilms:
if f['title']:
t = f['title'].replace(' ','').lower()
if t not in [e[0] for e in API_alltitles]:
API_alltitles.append((t, f['id']))
if f['title_vo']:
t = f['title_vo'].replace(' ','').lower()
if t not in [e[0] for e in API_alltitles]:
API_alltitles.append((t, f['id']))
API_alltitles = [f for f in API_alltitles if len(f[0])>10]
Llink2 = []
for film in Lunref:
for title, fid in API_alltitles:
if title==film.title:
Llink2.append((film, fid))
break
#print(film, ' <-> ', [f for f in APIfilms if f['id']==fid][0]['title'])
print('easy ref:', sorted(Llink2, key=lambda f:str(f)))
for f, _ in Llink2:
Lunref.remove(f)
print('missing (', len(Lmissing), '):',[str(e.api_id)+':'+repr(e) for e in sorted(Lmissing, key=lambda e:repr(e))])
print('\n'*3)
print('unreferenced:','|'.join(str(f) for f in sorted(Lunref, key=lambda e:repr(e))))
print('\n'*3)
print('unreferenced titles:', '|'.join(sorted([f.title for f in Lunref])))
# Put les renommages / déplacements
i = 0
for filmApi, filmLoc in Lrelink:
i += 1
print('['+str(i)+'/'+str(len(Lrelink))+']'+'relink:', filmApi.title)
try:
api.debug_print = True
api.put_file(id=filmApi.api_id, path=domain['server']+filmLoc.path, name=filmLoc.name)
api.debug_print = False
time.sleep(1)
except Exception as e:
print(e)
print('film '+filmApi.title+' not edited')
raise Exception('end')
# Poste les ajouts de doubles
i = 0
for film, filmID in Llink:
i += 1
print('['+str(i)+'/'+str(len(Llink))+']'+'link:', film.title)
try:
api.debug_print = True
api.post_file(path=domain['server']+film.path, name=film.name, type='Film', type_id=filmID, **film.additional_info())
api.debug_print = False
time.sleep(1)
except Exception as e:
print(e)
print('film '+film.title+' not added')
raise Exception('end')
# Poste les ajouts de doubles plus complexes
i = 0
for film, filmID in Llink2:
i += 1
print('['+str(i)+'/'+str(len(Llink2))+']'+'link2:', film.title)
try:
api.debug_print = True
api.post_file(path=domain['server']+film.path, name=film.name, type='Film', type_id=filmID, **film.additional_info())
api.debug_print = False
time.sleep(1)
except Exception as e:
print(e)
print('film '+film.title+' not added')
raise Exception('end')
# Poste tout les films locaux
i = 0
for film in Lunref:
i += 1
print('['+str(i)+'/'+str(len(Lunref))+']'+'post:', film.title, str(film.info.get('YEAR')))
try:
posted = False
if 'YEAR' in film.info:
resp = api.post_film(title=film.title, year=film.info['YEAR'])
if "id" in resp:
print('post: path=',domain['server']+film.path)
resp = api.post_file(path=domain['server']+film.path, name=film.name, type='film', type_id=resp["id"])
posted = True
if not posted:
resp = api.post_film(title=film.title)
if "id" in resp:
resp = api.post_file(path=domain['server']+film.path, name=film.name, type='film', type_id=resp["id"])
print('response:', resp)
time.sleep(1)
except Exception as e:
print(e)
print('film '+film.title+' not posted')
raise Exception('end')
# Supprime les fichiers qui n'existent plus
for film in Lmissing:
i += 1
print('['+str(i)+'/'+str(len(Lmissing))+']'+'missing:', film)
try:
resp = api.delete_file(id=film.api_id)
except Exception as e:
print(e)
print('film '+film.title+' not posted')
raise Exception('end')
time.sleep(1)
print('visit finished')
def main():
conf = config.Config()
api = piexel.Piexel(conf.server, conf.app, conf.token)
tokens = tokenizer.Tokenizer(conf, api)
folders = api.get_folders()
rules = api.get_paths()
for fold in folders:
applicable = [filerule.FileRule(re.escape(fold['path'])+'\\/'+r['regex'], conf) for r in rules if int(r['indexer_folder_id']) == fold['id']]
visit_folder(fold, api, applicable, tokens)
if __name__ == '__main__':
main()