def html_params(**kwargs):
params = []
for k,v in sorted(kwargs.iteritems()):
if k in ('class_', 'class__', 'for_'): k = k[:-1]
if v is True: params.append(k)
else: params.append(u'%s="%s"' % (u(k), escape(u(v), quote=True)))
return u" ".join(params)
def autocomplete():
'''
Devuelve la lista para el autocompletado de la búsqueda
'''
query = u(request.args.get("term", "")).lower()
if not query: return "[]"
ct = u(request.args.get("t","")).lower()
tamingWeight = {"c":1, "lang":200}
if ct:
for cti in CONTENTS_CATEGORY[ct]:
tamingWeight[TAMING_TYPES[cti]] = 200
options = taming.tameText(query, tamingWeight, 5, 3, 0.2)
if options is None:
cache.cacheme = False
return "[]"
return json.dumps([result[2] for result in options])
def tag(_name, _children=None, separator="", children_type=None, **kwargs):
if _children is False:
return u""
else:
attr = (" " + space_join('%s="%s"' % (key, u(val)) for key, val in kwargs.iteritems() if val)) if kwargs else ""
if _children:
if children_type is list:
return u"<%s%s>%s</%s>" % (_name, attr, separator.join(_children), _name)
elif children_type is unicode:
escaped_children = space_join(XML_ILLEGAL_CHARS_RE.split(u(_children)))
if "&" in escaped_children:
return u"<%s%s><![CDATA[%s]]></%s>" % (_name, attr, escaped_children, _name)
else:
return u"<%s%s>%s</%s>" % (_name, attr, escaped_children, _name)
elif children_type is str:
return u"<%s%s>%s</%s>" % (_name, attr, _children, _name)
elif children_type is float:
return u"<%s%s>%.8f</%s>" % (_name, attr, _children, _name)
else:
return u"<%s%s>%s</%s>" % (_name, attr, unicode(_children), _name)
else:
return u"<%s%s/>" % (_name, attr)
def before_request():
# si el idioma de la URL es inválido, devuelve página no encontrada
if g.url_lang and not g.url_lang in app.config["ALL_LANGS"]:
abort(404)
# quita todas las / finales de una URL
if request.path != '/' and request.path.endswith('/'):
return redirect(u(request.url).replace(u(request.path), u(request.path)[:-1]),301)
# ignora peticiones sin blueprint
if request.blueprint is None:
return
# si no es el idioma alternativo, lo añade por si no se encuentra el mensaje
if g.lang!="en":
get_translations().add_fallback(fallback_lang)
# dominio de la web
g.domain = request.url_root[7:-1] if "https" not in request.url_root else request.url_root[8:-1]
# título de la página por defecto
g.title = g.domain
# contador de archivos totales
g.count_files = lastcount[0]
def before_request():
# No preprocesamos la peticiones a static
if request.path.startswith("/static"):
g.accept_cookies = None
return
# default values for g object
init_g()
# comprueba limite de ratio de peticiones
check_rate_limit(g.search_bot)
# si el idioma de la URL es inválido, devuelve página no encontrada
all_langs = current_app.config["ALL_LANGS"]
if g.url_lang and not g.url_lang in all_langs:
abort(404)
# añade idioma por defecto si no hay traducción de algún mensaje en el idioma actual
if g.lang!="en":
add_translation_fallback("en")
# si hay que cambiar el idioma
if request.args.get("setlang",None):
session["lang"]=g.lang
# si el idioma esta entre los permitidos y el usuario esta logueado se actualiza en la base de datos
if g.lang in all_langs and current_user.is_authenticated():
current_user.set_lang(g.lang)
usersdb.update_user({"_id":current_user.id,"lang":g.lang})
return redirect(request.base_url)
g.keywords = set(_(keyword) for keyword in ['download', 'watch', 'files', 'submit_search', 'audio', 'video', 'image', 'document', 'software', 'P2P', 'direct_downloads'])
descr = _("about_text")
g.page_description = descr[:descr.find("<br")]
# ignora peticiones sin blueprint
if request.blueprint is None and request.path.endswith("/"):
if "?" in request.url:
root = request.url_root[:-1]
path = request.path.rstrip("/")
query = u(request.url)
query = query[query.find(u"?"):]
return redirect(root+path+query, 301)
return redirect(request.url.rstrip("/"), 301)
def download_search(file_data, file_text, fallback):
'''
Intenta buscar una cadena a buscar cuando viene download
'''
search_texts=[]
if file_data:
mds = file_data['file']['md']
for key in ['audio:artist', 'audio:album', 'video:series', 'video:title', 'image:title', 'audio:title','application:name', 'application:title', 'book:title', 'torrent:title']:
if key in mds and isinstance(mds[key], basestring) and len(mds[key])>1:
search_texts.append((u(mds[key]), False))
search_texts.append((file_data["view"]["fn"], True))
if file_data['view']["tags"]:
fallback = "(%s)"%file_data['view']["tags"][0]
elif file_data['view']['file_type']:
fallback = "(%s)"%file_data['view']['file_type']
if file_text:
search_texts.append((file_text, True))
best_candidate = None
best_points = 10000
for main_position, (search_text, is_filename) in enumerate(search_texts):
phrases = split_phrase(search_text, is_filename)
for inner_position, phrase in enumerate(phrases):
candidate = [part for part in phrase.split(" ") if part.strip()]
count = sum(1 for word in candidate if len(word)>1)
is_numeric = len(candidate)==1 and candidate[0].isdecimal()
candidate_points = main_position+inner_position+(50 if count==0 else 5 if count==1 and is_numeric else 20 if count>15 else 0)
if candidate_points<best_points:
best_points = candidate_points
best_candidate = candidate
if best_candidate and best_points<2:
break
if best_candidate:
return " ".join(best_candidate[:5])
return fallback
def simple_pyon(x):
'''
Convierte una cadena representando un objeto de python en un objeto de
python con el tipo apropiado. Muy simplificado.
'''
x = x.strip()
if x.lower() in ("","None","null","none"): return None
if x.replace("+","",x.startswith("+")).replace("-","",x.startswith("-")).replace(".","",x.count(".")==1).isdigit():
if "." in x: float(x)
return int(x)
if x.endswith("j"):
return complex(x)
if len(x)>1:
if x[0] == x[-1] and x[0] in "\"'": return x[1:-1]
if x[0] == "u" and x[1] == x[-1] and x[1] in "\"'": return u(x[2:-1])
if x[0] == "[" and x[-1] == "]": return list(simple_pyon(i) for i in x[1:-1].split(","))
if x[0] == "(" and x[-1] == ")": return tuple(simple_pyon(i) for i in x[1:-1].split(","))
if x[0] == "{" and x[-1] == "}": return dict((simple_pyon(i.split(":")[0]), simple_pyon(i.split(":")[1])) for i in x[1:-1].split(","))
return x
def build_source_links(f, prevsrc=False, unify_torrents=False):
'''
Construye los enlaces correctamente
'''
def get_domain(src):
'''
Devuelve el dominio de una URL
'''
url_parts=urlparse(src).netloc.split('.')
i=len(url_parts)-1
if len(url_parts[i])<=2 and len(url_parts[i-1])<=3:
return url_parts[i-2]+'.'+url_parts[i-1]+'.'+url_parts[i]
else:
return url_parts[i-1]+'.'+url_parts[i];
if not "fn" in f['view']:
choose_filename(f)
f['view']['action']='download'
f['view']['sources']=defaultdict(dict)
srcs=f['file']['src']
max_weight=0
icon = ""
for hexuri,src in srcs.items():
if not src.get('bl',None) in (0, None):
continue
join=False
count=0
part=url=""
source_data=filesdb.get_source_by_id(src["t"])
if source_data is None: #si no exite el origen del archivo
logging.error("El fichero contiene un origen inexistente en la tabla \"sources\": %s" % src["t"], extra={"file":f})
continue
elif "crbl" in source_data and source_data["crbl"]==1: #si el origen esta bloqueado
continue
elif "w" in source_data["g"] or "f" in source_data["g"] or "s" in source_data["g"]: #si es descarga directa
link_weight=1
tip=source_data["d"]
icon="web"
url=src['url']
if "f" in source_data["g"]:
source=get_domain(src['url'])
else:
source=source_data["d"]
#en caso de duda se prefiere streaming
if "s" in source_data["g"]:
f['view']['action']='watch'
link_weight*=2
#torrenthash antes de torrent porque es un caso especifico
elif source_data["d"]=="BitTorrentHash":
link_weight=0.7 if 'torrent:tracker' in f['file']['md'] or 'torrent:trackers' in f['file']['md'] else 0.1
if unify_torrents:
tip=source=icon="torrent"
else:
tip="Torrent MagnetLink"
source=icon="tmagnet"
join=True
count=int(src['m'])
part="xt=urn:btih:"+src['url']
if 'torrent:tracker' in f['file']['md']:
part += unicode('&tr=' + urllib.quote_plus(u(f['file']['md']['torrent:tracker']).encode("UTF-8")), "UTF-8")
elif 'torrent:trackers' in f['file']['md']:
trackers = f['file']['md']['torrent:trackers']
if isinstance(trackers, basestring):
part += unicode("".join('&tr='+urllib.quote_plus(tr) for tr in u(trackers).encode("UTF-8").split(" ")), "UTF-8")
elif "t" in source_data["g"]:
link_weight=0.8
url=src['url']
icon="torrent"
if unify_torrents:
tip=source="torrent"
else:
tip=source=get_domain(src['url'])
elif source_data["d"]=="Gnutella":
link_weight=0.2
tip="Gnutella"
source=icon="gnutella"
part="xt:urn:sha1:"+src['url']
join=True
count=int(src['m'])
elif source_data["d"]=="ed2k":
link_weight=0.1
tip="ED2K"
source=icon="ed2k"
url="ed2k://|file|"+f['view']['efn']+"|"+str(f['file']['z'] if "z" in f["file"] else 1)+"|"+src['url']+"|/"
count=int(src['m'])
elif source_data["d"]=="Tiger":
link_weight=0
tip="Gnutella"
source=icon="gnutella"
part="xt:urn:tiger="+src['url']
elif source_data["d"]=="MD5":
link_weight=0
tip="Gnutella"
source=icon="gnutella"
part="xt:urn:md5="+src['url']
else:
continue
f['view']['sources'][source]['tip']=tip
f['view']['sources'][source]['icon']=icon
f['view']['sources'][source]['logo']="http://%s/favicon.ico"%tip
f['view']['sources'][source]['join']=join
f['view']['sources'][source]['type']=source_data["g"]
#para no machacar el numero si hay varios archivos del mismo source
if not 'count' in f['view']['sources'][source] or count>0:
f['view']['sources'][source]['count']=count
if not "parts" in f['view']['sources'][source]:
f['view']['sources'][source]['parts']=[]
if not 'urls' in f['view']['sources'][source]:
f['view']['sources'][source]['urls']=[]
if part!="":
f['view']['sources'][source]['parts'].append(part)
if url!="":
f['view']['sources'][source]['urls'].append(url)
if source_data["d"]!="ed2k":
f['view']['sources'][source]['count']+=1
if link_weight>max_weight:
max_weight = link_weight
f['view']['source'] = source
if icon!="web":
for src,info in f['view']['sources'].items():
if 'join' in info:
size=""
if 'z' in f['file']:
size = "&xl="+str(f['file']['z'])
f['view']['sources'][src]['urls'].append("magnet:?dn="+f['view']['efn']+size+"&"+"&".join(info['parts']))
elif not 'urls' in info:
del(f['view']['sources'][src])
def fix_sphinx_result(word):
return u(SPHINX_WRONG_RANGE.sub(fixer, word))
def format_metadata(f,text_cache, search_text_shown=False):
'''
Formatea los metadatos de los archivos
'''
text = text_cache[2] if text_cache else None
view_md = f['view']['md'] = {}
view_searches = f["view"]["searches"]={}
file_type = f['view']['file_type'] if 'file_type' in f['view'] else None
if 'md' in f['file']:
#si viene con el formato tipo:metadato se le quita el tipo
file_md = {(meta.split(":")[-1] if ":" in meta else meta): value for meta, value in f['file']['md'].iteritems()}
# Duración para vídeo e imágenes
seconds = get_float(file_md, "seconds")
minutes = get_float(file_md, "minutes")
hours = get_float(file_md, "hours")
# Si no he recibido duracion de otra forma, pruebo con length y duration
if seconds==minutes==hours==None:
seconds = get_float(file_md, "length") or get_float(file_md, "duration")
duration = [hours or 0, minutes or 0, seconds or 0] # h, m, s
if any(duration):
carry = 0
for i in xrange(len(duration)-1,-1,-1):
unit = long(duration[i]) + carry
duration[i] = unit%60
carry = unit/60
view_md["length"] = "%d:%02d:%02d" % tuple(duration) if duration[0] > 0 else "%02d:%02d" % tuple(duration[1:])
# Tamaño para vídeos e imágenes
width = get_int(file_md, "width")
height = get_int(file_md, "height")
if width and height:
view_md["size"] = "%dx%dpx" % (width, height)
# Metadatos que no cambian
try:
view_md.update(
(meta, file_md[meta]) for meta in
(
"folders","description","fileversion","os","files","pages","format",
"seeds","leechs","composer","publisher","encoding","director","writer","starring","producer","released"
) if meta in file_md
)
view_searches.update(
(meta, seoize_text(file_md[meta],"_",False)) for meta in
(
"folders","os","composer","publisher","director","writer","starring","producer"
) if meta in file_md
)
except BaseException as e:
logging.warn(e)
# thumbnail
if "thumbnail" in file_md:
f["view"]["thumbnail"] = file_md["thumbnail"]
#metadatos que tienen otros nombres
try:
view_md.update(("tags", file_md[meta]) for meta in ("keywords", "tags", "tag") if meta in file_md)
if "tags" in view_md and isinstance(view_md["tags"], basestring):
view_searches["tags"] = []
view_md.update(("comments", file_md[meta]) for meta in ("comments", "comment") if meta in file_md)
view_md.update(("track", file_md[meta]) for meta in ("track", "track_number") if meta in file_md)
view_md.update(("created_by", file_md[meta]) for meta in ("created_by", "encodedby","encoder") if meta in file_md)
view_md.update(("language", file_md[meta]) for meta in ("language", "lang") if meta in file_md)
view_md.update(("date", file_md[meta]) for meta in ("published", "creationdate") if meta in file_md)
view_md.update(("trackers", "\n".join(file_md[meta].split(" "))) for meta in ("trackers", "tracker") if meta in file_md and isinstance(file_md[meta], basestring))
view_md.update(("hash", file_md[meta]) for meta in ("hash", "infohash") if meta in file_md)
view_md.update(("visualizations", file_md[meta]) for meta in ("count", "viewCount") if meta in file_md)
if "unpackedsize" in file_md:
view_md["unpacked_size"]=file_md["unpackedsize"]
if "privateflag" in file_md:
view_md["private_file"]=file_md["privateflag"]
except BaseException as e:
logging.warn(e)
#torrents -> filedir filesizes filepaths
if "filepaths" in file_md:
filepaths = {}
for path, size in izip_longest(u(file_md["filepaths"]).split("///"), u(file_md.get("filesizes","")).split(" "), fillvalue=None):
# no permite tamaños sin fichero
if not path: break
parts = path.strip("/").split("/")
# crea subdirectorios
relative_path = filepaths
for part in parts[:-1]:
if "/"+part not in relative_path:
relative_path["/"+part] = {}
relative_path = relative_path["/"+part]
# si ya existe el directorio no hace nada
if "/"+parts[-1] in relative_path:
pass
# si el ultimo nivel se repite es un directorio (fallo de contenido)
elif parts[-1] in relative_path:
relative_path["/"+parts[-1]] = {}
del relative_path[parts[-1]]
else:
relative_path[parts[-1]] = size
if "filedir" in file_md:
filepaths = {"/"+u(file_md["filedir"]).strip("/"):filepaths}
if filepaths:
view_md["filepaths"] = filepaths
view_searches["filepaths"] = {}
# Metadatos multimedia
try:
#extraccion del codec de video y/o audio
if "video_codec" in file_md: #si hay video_codec se concatena el audio_codec detras si es necesario
view_md["codec"]=file_md["video_codec"]+" "+file_md["audio_codec"] if "audio_codec" in file_md else file_md["video_codec"]
else: #sino se meten directamente
view_md.update(("codec", file_md[meta]) for meta in ("audio_codec", "codec") if meta in file_md)
if file_type in ("audio", "video", "image"):
view_md.update((meta, file_md[meta]) for meta in ("genre", "track", "artist", "author", "colors") if meta in file_md)
view_searches.update((meta, seoize_text(file_md[meta], "_", False)) for meta in ("artist", "author") if meta in file_md)
except BaseException as e:
logging.warn(e)
# No muestra titulo si es igual al nombre del fichero
if "name" in file_md:
title = u(file_md["name"])
elif "title" in file_md:
title = u(file_md["title"])
else:
title = f['view']['nfn']
if title:
show_title = True
text_longer = title
text_shorter = f["view"]["fn"]
if len(text_shorter)>len(text_longer):
text_longer, text_shorter = text_shorter, text_longer
if text_longer.startswith(text_shorter):
text_longer = text_longer[len(text_shorter):]
if len(text_longer)==0 or (len(text_longer)>0 and text_longer.startswith(".") and text_longer[1:] in EXTENSIONS):
show_title = False
if show_title:
view_md["title"] = title
view_searches["title"] = seoize_text(title, "_", False)
# Los que cambian o son especificos de un tipo
try:
if "date" in view_md: #intentar obtener una fecha válida
try:
view_md["date"]=format_datetime(datetime.fromtimestamp(strtotime(view_md["date"])))
except:
del view_md["date"]
if file_type == 'audio': #album, year, bitrate, seconds, track, genre, length
if 'album' in file_md:
album = u(file_md["album"])
year = get_int(file_md, "year")
if album:
view_md["album"] = album + (" (%d)"%year if year and 1900<year<2100 else "")
view_searches["album"] = seoize_text(album, "_", False)
if 'bitrate' in file_md: # bitrate o bitrate - soundtype o bitrate - soundtype - channels
bitrate = get_int(file_md, "bitrate")
if bitrate:
soundtype=" - %s" % file_md["soundtype"] if "soundtype" in file_md else ""
channels = get_float(file_md, "channels")
channels=" (%g %s)" % (round(channels,1),_("channels")) if channels else ""
view_md["quality"] = "%g kbps %s%s" % (bitrate,soundtype,channels)
elif file_type == 'document': #title, author, pages, format, version
if "format" in file_md:
view_md["format"] = "%s%s" % (file_md["format"]," %s" % file_md["formatversion"] if "formatversion" in file_md else "")
version = []
if "formatVersion" in file_md:
version.append(u(file_md["formatVersion"]))
elif "version" in file_md:
version.append(u(file_md["version"]))
if "revision" in file_md:
version.append(u(file_md["revision"]))
if version:
view_md["version"] = " ".join(version)
elif file_type == 'image': #title, artist, description, width, height, colors
pass
elif file_type == 'software': #title, version, fileversion, os
if "title" in view_md and "version" in file_md:
view_md["title"] += " %s" % file_md["version"]
view_searches["title"] += " %s" % seoize_text(file_md["version"], "_", False)
elif file_type == 'video':
quality = []
framerate = get_int(file_md, "framerate")
if framerate:
quality.append("%d fps" % framerate)
if 'codec' in view_md: #si ya venia codec se muestra ahora en quality solamente
quality.append(u(view_md["codec"]))
del view_md["codec"]
if quality:
view_md["quality"] = " - ".join(quality)
if "series" in file_md:
series = u(file_md["series"])
if series:
safe_series = seoize_text(series, "_", False)
view_md["series"] = series
view_searches["series"]="%s_%s"%(safe_series,"(series)")
season = get_int(file_md, "season")
if season:
view_md["season"] = season
view_searches["season"]="%s_(s%d)"%(safe_series,season)
episode = get_int(file_md, "episode")
if episode:
view_md["episode"] = episode
view_searches["episode"]="%s_(s%de%d)"%(safe_series,season,episode)
except BaseException as e:
logging.exception("Error obteniendo metadatos especificos del tipo de contenido.")
view_mdh=f['view']['mdh']={}
for metadata,value in view_md.items():
if isinstance(value, basestring):
value = clean_html(value)
if not value:
del view_md[metadata]
continue
view_md[metadata]=value
# resaltar contenidos que coinciden con la busqueda, para textos no muy largos
if len(value)<500:
view_mdh[metadata]=highlight(text,value) if text and len(text)<100 else value
elif isinstance(value, float): #no hay ningun metadato tipo float
view_md[metadata]=str(int(value))
else:
view_md[metadata]=value
def choose_filename(f,text=False):
'''
Elige el archivo correcto
'''
text=slugify(text) if text else text
srcs = f['file']['src']
fns = f['file']['fn']
chosen = None
max_count = -1
has_text = 0
try:
for hexuri,src in srcs.items():
if 'bl' in src and src['bl']!=0:
continue
this_has_text=0
for crc,srcfn in src['fn'].items():
#si no tiene nombre no se tiene en cuenta
m = srcfn['m'] if len(fns[crc]['n'])>0 else 0
if 'c' in fns[crc]:
fns[crc]['c']+=m
else:
fns[crc]['c']=m
if text:
slugified_fn = slugify(fns[crc]['n']).strip()
if len(slugified_fn)>0:
#TODO hace lo mismo que para poner el nombre en negrita y sacar el mejor texto aunque no tenga tildes o no venga unido por espacios
if slugified_fn.upper().find(text.upper())!=-1:
this_has_text = 2000
else:
matches = 0
for word in [re.escape(w) for w in text.split(" ")]:
matches += len(re.findall(r"/((?:\b|_)%s(?:\b|_))/i"%word, slugified_fn))
if matches>0:
this_has_text = 1000 + matches
f['file']['fn'][crc]['tht'] = this_has_text
better = fns[crc]['c']>max_count
if this_has_text > has_text or (better and this_has_text==has_text):
has_text = this_has_text
chosen = crc
max_count = fns[crc]['c']
except KeyError as e: #para los sources que tienen nombre pero no estan en el archivo
logging.exception(e)
f['view']['url'] = mid2url(hex2mid(f['file']['_id']))
if chosen:
filename = fns[chosen]['n']
ext = fns[chosen]['x']
else: #uses filename from src
srcurl = ""
for hexuri,src in srcs.items():
if src['url'].find("/")!=-1:
srcurl = src['url']
if srcurl=="":
return
srcurl = srcurl[srcurl.rfind("/")+1:]
ext = srcurl[srcurl.rfind(".")+1:]
filename = srcurl[0:srcurl.rfind(".")]
filename = Markup(filename).striptags()[:512]
if not ext in EXTENSIONS:
filename += ext
ext=""
nfilename = filename
else:
#clean filename
end = filename.upper().rfind("."+ext.upper())
if end == -1:
nfilename = filename
else:
nfilename = filename.strip()[0:end]
f['view']['fn'] = filename #TODO para los tipo archive buscar el tipo real en el nombre (mp3,avi...)
f['view']['fnx'] = ext
f['view']['efn'] = filename.replace(" ", "%20")
#poner bonito nombre del archivo
if nfilename.find(" ")==-1:
nfilename = nfilename.replace(".", " ")
f['view']['nfn'] = nfilename.replace("_", " ")
#nombre del archivo escapado para generar las url de descarga
f['view']['qfn'] = u(filename).encode("UTF-8")
#nombre del archivo con las palabras que coinciden con la busqueda resaltadas
if not text:# or not has_text:
f['view']['fnh'] = f['view']['fnhs'] = filename
else:
f['view']['fnh'], f['view']['fnhs'] = highlight(text,filename,True)
return has_text>0
def d(self):
if self._d is None:
self._d = PyQuery(u(self.data))
return self._d
def download_file(file_id,file_name=None):
'''
Devuelve el archivo a descargar, votos, comentarios y archivos relacionados
'''
error=(None,"") #guarda el id y el texto de un error
file_data=None
if file_id is not None: #si viene un id se comprueba que sea correcto
if is_valid_url_fileid(file_id):
try: #intentar convertir el id que viene de la url a uno interno
file_id=url2mid(file_id)
except (bson.objectid.InvalidId, TypeError) as e:
try: #comprueba si se trate de un ID antiguo
possible_file_id = filesdb.get_newid(file_id)
if possible_file_id is None:
logging.warn("Identificadores numericos antiguos sin resolver: %s."%e, extra={"fileid":file_id})
error=(404,"link_not_exist")
else:
logging.warn("Identificadores numericos antiguos encontrados: %s."%e, extra={"fileid":file_id})
return {"html": redirect(url_for(".download", file_id=mid2url(possible_file_id), file_name=file_name), 301),"error":(301,"")}
except BaseException as e:
logging.exception(e)
error=(503,"")
file_id=None
else:
abort(404)
if file_id:
try:
file_data=get_file_metadata(file_id, file_name)
except DatabaseError:
error=(503,"")
except FileNotExist:
error=(404,"link_not_exist")
except (FileRemoved, FileFoofindRemoved, FileNoSources):
error=(410,"error_link_removed")
except FileUnknownBlock:
error=(404,"")
if error[0] is None and not file_data: #si no ha habido errores ni hay datos, es porque existe y no se ha podido recuperar
error=(503,"")
if file_id is None or error[0] is not None:
html=""
if error[0] is not None: #si hay algun error se devuelve renderizado
message_msgid="error_%s_message" % error[0]
message_msgstr=_(message_msgid)
g.title="%s %s" % (error[0], message_msgstr if message_msgstr!=message_msgid else _("error_500_message"))
html=render_template('error.html',error=error,full_screen=True)
return {"html": html,"play":None,"file_data":file_data,"error":error}
else:
save_visited([file_data])
title = u(file_data['view']['fn'])
g.title = u"%s \"%s\" - %s" % (
_(file_data['view']['action']).capitalize(),
title[:100],
g.title)
g.page_description = u"%s %s"%(_(file_data['view']['action']).capitalize(), seoize_text(title," ",True))
#si el usuario esta logueado se comprueba si ha votado el archivo para el idioma activo y si ha marcado el archivo como favorito
vote=None
favorite = False
if current_user.is_authenticated():
vote=usersdb.get_file_vote(file_id,current_user,g.lang)
favorite=any(file_id==favorite["id"] for favorite in usersdb.get_fav_files(current_user))
#formulario para enviar comentarios
form = CommentForm(request.form)
if request.method=='POST' and current_user.is_authenticated() and (current_user.type is None or current_user.type==0) and form.validate():
usersdb.set_file_comment(file_id,current_user,g.lang,form.t.data)
form.t.data=""
flash("comment_published_succesfully")
#actualizar el fichero con la suma de los comentarios por idioma
filesdb.update_file({"_id":file_id,"cs":usersdb.get_file_comments_sum(file_id),"s":file_data["file"]["s"]},direct_connection=True)
#si tiene comentarios se guarda el número del comentario, el usuario que lo ha escrito, el comentario en si y los votos que tiene
comments=[]
if "cs" in file_data["file"]:
comments=[(i,usersdb.find_userid(comment["_id"].split("_")[0]),comment,comment_votes(file_id,comment)) for i,comment in enumerate(usersdb.get_file_comments(file_id,g.lang),1)]
# en la pagina de download se intentan obtener palabras para buscar si no las hay
if g.args.get("q", None) is None:
query = download_search(file_data, file_name, "foofind")
if query:
g.args["q"] = query.replace(":","")
return {
"html":render_template('files/download.html',file=file_data,vote={"k":0} if vote is None else vote,favorite=favorite,form=form,comments=comments),
"play":file_data["view"]["play"] if "play" in file_data["view"] else "",
"file_data":file_data,
}
files.append(afile)
if len(ids)>=result["total_found"]:
total_found = len(files)
else:
total_found = max(result["total_found"], len(files))
# completa la descripcion de la pagina
if files:
# Descripcion inicial
page_description = g.page_description + ". "
# Descripcion de alguno de los primeros ficheros
for f in files[:3]:
if "description" in f["view"]["md"]:
phrase = u(f["view"]["md"]["description"]).capitalize()
page_description += phrase + " " if phrase[-1]=="." else ". "
break
# Busca frases para completar la descripcion hasta un minimo de 100 caracteres
page_description_len = len(page_description)
if page_description_len<100:
phrases = []
for f in files[1:]: # se salta el primer fichero, que podría ser el download actual
phrase = f["view"]["nfn"].capitalize()
if phrase not in phrases:
phrases.append(phrase)
page_description_len += len(phrase)
if page_description_len>=100:
def format_metadata(f,details,text):
'''
Formatea los metadatos de los archivos
'''
def searchable(value,details):
'''
Añade un enlace a la busqueda si es necesario
'''
if details:
return '<a href="%s">%s</a>' % (url_for("files.search",q=value), value)
else:
return value
view_md = f['view']['md'] = {}
file_type = f['view']['file_type'] if 'file_type' in f['view'] else None
if 'md' in f['file']:
#si viene con el formato tipo:metadato se le quita el tipo
file_md = {(meta.split(":")[-1] if ":" in meta else meta): value for meta, value in f['file']['md'].iteritems()}
# Duración para vídeo e imágenes
put_duration = False
duration = [0, 0, 0] # h, m, s
try:
if "seconds" in file_md:
put_duration = True
duration[-1] = float(file_md["seconds"])
if "minutes" in file_md:
put_duration = True
duration[-2] = float(file_md["minutes"])
if "hours" in file_md:
put_duration = True
duration[-3] = float(file_md["hours"])
except BaseException as e:
logging.warn(e, extra=file_md)
if not put_duration and "length" in file_md:
# Si recibo length y no la he recibido duration de otra forma
try:
duration[-1] = to_seconds(file_md["length"])
put_duration = True
except BaseException as e:
logging.error("Problema al parsear duración: 'length'", extra=file_md)
if not put_duration and "duration" in file_md:
# Si recibo duration y no la he recibido de otra forma
try:
duration[-1] = to_seconds(file_md["duration"])
put_duration = True
except BaseException as e:
logging.error("Problema al parsear duración: 'duration'", extra=file_md)
if put_duration:
carry = 0
for i in xrange(len(duration)-1,-1,-1):
unit = long(duration[i]) + carry
duration[i] = unit%60
carry = unit/60
view_md["length"] = "%d:%02d:%02d" % tuple(duration) if duration[-3] > 0 else "%02d:%02d" % tuple(duration[-2:])
# Tamaño para vídeos e imágenes
if "width" in file_md and 'height' in file_md:
try:
width = (
int(file_md["width"].replace("pixels","").replace("px",""))
if isinstance(file_md["width"], basestring)
else int(file_md["width"]))
height = (
int(file_md["height"].replace("pixels","").replace("px",""))
if isinstance(file_md["width"], basestring)
else int(file_md["height"]))
view_md["size"] = "%dx%dpx" % (width, height)
except BaseException as e:
logging.warn(e)
# Metadatos que no cambian
try:
view_md.update((meta, file_md[meta]) for meta in (
("folders","description","fileversion","os","files","pages","format")
if details else ("files","pages","format")) if meta in file_md)
except BaseException as e:
logging.warn(e)
# Metadatos multimedia
try:
if file_type in ("audio", "video", "image"):
view_md.update((meta, file_md[meta]) for meta in
("genre", "track", "artist", "author", "colors")
if meta in file_md)
except BaseException as e:
logging.warn(e)
# No muestra titulo si es igual al nombre del fichero
title = None
if "name" in file_md:
title = u(file_md["name"])
elif "title" in file_md:
title = u(file_md["title"])
if title:
show_title = True
text_longer = title
text_shorter = f["view"]["fn"]
if len(text_shorter)>len(text_longer): text_longer, text_shorter = text_shorter, text_longer
if text_longer.startswith(text_shorter):
text_longer = text_longer[len(text_shorter):]
if len(text_longer)==0 or (len(text_longer)>0 and text_longer.startswith(".") and text_longer[1:] in EXTENSIONS):
show_title = False
if show_title:
view_md["title"] = title
# Los que cambian o son especificos de un tipo
try:
if file_type == 'audio': #album, year, bitrate, seconds, track, genre, length
if 'album' in file_md:
year = 0
if "year" in file_md:
md_year = u(file_md["year"]).strip().split()
for i in md_year:
if i.isdigit() and len(i) == 4:
year = int(i)
break
album = file_md["album"]
view_md["album"] = ("%s (%d)" % (album, year)) if 1900 < year < 2100 else album
if 'bitrate' in file_md:
bitrate = "%s kbps" % u(file_md["bitrate"]).replace("~","")
view_md["quality"] = ( # bitrate o bitrate - soundtype
("%s - %s" % (bitrate, file_md["soundtype"]))
if details and "soundtype" in file_md else bitrate)
elif file_type == 'archive': #title, name, unpackedsize, folders, files
if "unpackedsize" in file_md:
view_md["unpackedsize"] = file_md["unpackedsize"]
elif file_type == 'document': #title, author, pages, format, version
if details:
if "format" in file_md:
view_md["format"] = "%s%s" % (file_md["format"],
" %s" % file_md["formatversion"]
if "formatversion" in file_md else "")
version = []
if "version" in file_md: version.append(u(file_md["version"]))
if "revision" in file_md: version.append(u(file_md["revision"]))
if version: view_md["version"] = " ".join(version)
elif file_type == 'image': #title, artist, description, width, height, colors
pass
elif file_type == 'software': #title, version, fileversion, os
if "title" in view_md and "version" in file_md:
view_md["title"] += " %s" % file_md["version"]
elif file_type == 'video':
if details:
quality = []
try:
if 'framerate' in file_md: quality.append("%d fps" % int(float(file_md["framerate"])))
except BaseException as e:
logging.warn(e)
if 'codec' in file_md: quality.append(u(file_md["codec"]))
if quality: view_md["quality"] = " - ".join(quality)
view_md.update((i, file_md[i]) for i in ("series", "episode", "season") if i in file_md)
except BaseException as e:
logging.warn("%s\n\t%s\n\t%s" % (e, f, view_md))
#if len(view_md)>0:
f['view']['mdh']={}
for metadata,value in view_md.items():
if isinstance(value, basestring):
final_value = Markup(value).striptags()
final_value = searchable(highlight(text,final_value) if text else final_value, details)
else:
final_value = value
f['view']['mdh'][metadata] = final_value
def download(file_id,file_name=None):
'''
Muestra el archivo a descargar, votos, comentarios y archivos relacionados
'''
def choose_filename_related(file_data):
'''
Devuelve el nombre de fichero elegido
'''
f=init_data(file_data)
choose_filename(f)
return f
def comment_votes(file_id,comment):
'''
Obtiene los votos de comentarios
'''
comment_votes={}
if "vs" in comment:
for i,comment_vote in enumerate(usersdb.get_file_comment_votes(file_id)):
if not comment_vote["_id"] in comment_votes:
comment_votes[comment_vote["_id"][0:40]]=[0,0,0]
if comment_vote["k"]>0:
comment_votes[comment_vote["_id"][0:40]][0]+=1
else:
comment_votes[comment_vote["_id"][0:40]][1]+=1
#si el usuario esta logueado y ha votado se guarda para mostrarlo activo
if current_user.is_authenticated() and comment_vote["u"]==current_user.id:
comment_votes[comment_vote["_id"][0:40]][2]=comment_vote["k"]
return comment_votes
#guardar los parametros desde donde se hizo la busqueda si procede
args={}
if request.referrer:
querystring = urlparse(request.referrer).query
if querystring:
for params in u(querystring).encode("UTF-8").split("&"):
param=params.split("=")
if len(param) == 2:
args[param[0]]=u(urllib.unquote_plus(param[1]))
try:
file_id=url2mid(file_id)
except Exception as e:
# Comprueba que se trate de un ID antiguo
try:
possible_file_id = filesdb.get_newid(file_id)
if possible_file_id is None:
logging.warn("%s - %s" % (e, file_id))
flash("link_not_exist", "error")
abort(404)
return redirect(
url_for(".download", file_id=mid2url(possible_file_id), file_name=file_name),
code=301)
except filesdb.BogusMongoException as e:
logging.exception(e)
abort(503)
try:
data = filesdb.get_file(file_id, bl = None)
except filesdb.BogusMongoException as e:
logging.exception(e)
abort(503)
# intenta sacar el id del servidor de sphinx,
# resuelve inconsistencias de los datos
if not data:
sid = get_id_server_from_search(file_id, file_name)
if sid:
try:
data = filesdb.get_file(file_id, sid = sid, bl = None)
except filesdb.BogusMongoException as e:
logging.exception(e)
abort(503)
if data:
if not data["bl"] in (0, None):
if data["bl"] == 1: flash("link_not_exist", "error")
elif data["bl"] == 3: flash("error_link_removed", "error")
goback = True
#block_files( mongo_ids=(data["_id"],) )
abort(404)
else:
flash("link_not_exist", "error")
abort(404)
#obtener los datos
file_data=fill_data(data, True, file_name)
if file_data["view"]["sources"]=={}: #si tiene todos los origenes bloqueados
flash("error_link_removed", "error")
abort(404)
save_visited([file_data])
# Título
title = u(file_data['view']['fn'])
g.title = u"%s \"%s%s\"%s%s" % (
_(file_data['view']['action']).capitalize(),
title[:50],
"..." if len(title) > 50 else "",
" - " if g.title else "",
g.title)
#obtener los archivos relacionados
related_files = search_related(split_file(file_data["file"])[0][:10])
bin_file_id=mid2bin(file_id)
ids=sorted({fid[0:3] for related in related_files for fid in get_ids(related) if fid[0]!=bin_file_id})[:5]
files_related=[choose_filename_related(data) for data in get_files(ids)]
#si el usuario esta logueado se comprueba si ha votado el archivo para el idioma activo
vote=None
if current_user.is_authenticated():
vote=usersdb.get_file_vote(file_id,current_user,g.lang)
if vote is None:
vote={"k":0}
#formulario para enviar comentarios
form = CommentForm(request.form)
if request.method=='POST' and current_user.is_authenticated() and (current_user.type is None or current_user.type==0) and form.validate():
usersdb.set_file_comment(file_id,current_user,g.lang,form.t.data)
form.t.data=""
flash("comment_published_succesfully")
#actualizar el fichero con la suma de los comentarios por idioma
filesdb.update_file({"_id":file_id,"cs":usersdb.get_file_comments_sum(file_id),"s":file_data["file"]["s"]},direct_connection=True)
#si tiene comentarios se guarda el número del comentario, el usuario que lo ha escrito, el comentario en si y los votos que tiene
comments=[]
if "cs" in file_data["file"]:
comments=[(i,usersdb.find_userid(comment["_id"][0:24]),comment,comment_votes(file_id,comment)) for i,comment in enumerate(usersdb.get_file_comments(file_id,g.lang),1)]
return render_template('files/download.html',file=file_data,args=args,vote=vote,files_related=files_related,comments=comments,form=form)
def innergroup_hash(field_path, afile):
return hash(u(reduce(dictget, field_path, afile)) or "")
def download(file_id,file_name=None):
'''
Muestra el archivo a descargar, votos, comentarios y archivos relacionados
'''
def choose_filename_related(file_data):
'''
Devuelve el nombre de fichero elegido
'''
f=init_data(file_data)
choose_filename(f)
return f
def comment_votes(file_id,comment):
'''
Obtiene los votos de comentarios
'''
comment_votes={}
if "vs" in comment:
for i,comment_vote in enumerate(usersdb.get_file_comment_votes(file_id)):
if not comment_vote["_id"] in comment_votes:
comment_votes[comment_vote["_id"][0:40]]=[0,0,0]
if comment_vote["k"]>0:
comment_votes[comment_vote["_id"][0:40]][0]+=1
else:
comment_votes[comment_vote["_id"][0:40]][1]+=1
#si el usuario esta logueado y ha votado se guarda para mostrarlo activo
if current_user.is_authenticated() and comment_vote["u"]==current_user.id:
comment_votes[comment_vote["_id"][0:40]][2]=comment_vote["k"]
return comment_votes
if file_name is not None:
g.title = _("download").capitalize()+" "+file_name+" - "+g.title
else:
g.title =_("download").capitalize()
#guardar los parametros desde donde se hizo la busqueda si procede
args={}
if request.referrer:
querystring = urlparse(request.referrer).query
if querystring:
for params in querystring.split("&"):
param=params.split("=")
if len(param) == 2:
args[param[0]]=u(urllib.unquote_plus(param[1]))
try:
file_id=url2mid(file_id)
except BaseException as e:
logging.warn((e, file_id))
abort(404)
data = filesdb.get_file(file_id, bl = None)
if data:
if not data["bl"] in (0, None):
if data["bl"] == 1: flash("link_not_exist", "error")
elif data["bl"] == 3: flash("error_link_removed", "error")
goback = True
#block_files( mongo_ids=(data["_id"],) )
abort(404)
else:
flash("link_not_exist", "error")
abort(404)
#obtener los datos
file_data=fill_data(data, True, file_name)
save_visited([file_data])
#obtener los archivos relacionados
related_files = search_related(split_file(file_data["file"])[0][0:50])
bin_file_id=mid2bin(file_id)
ids=sorted({fid for related in related_files for fid in get_ids(related) if fid[0]!=bin_file_id})[:5]
files_related=[choose_filename_related(data) for data in get_files(ids)]
#si el usuario esta logueado se comprueba si ha votado el archivo para el idioma activo
vote=None
if current_user.is_authenticated():
vote=usersdb.get_file_vote(file_id,current_user,g.lang)
if vote is None:
vote={"k":0}
#formulario para enviar comentarios
form = CommentForm(request.form)
if request.method=='POST' and current_user.is_authenticated() and (current_user.type is None or current_user.type==0) and form.validate():
usersdb.set_file_comment(file_id,current_user,g.lang,form.t.data)
form.t.data=""
flash("comment_published_succesfully")
#actualizar el fichero con la suma de los comentarios por idioma
filesdb.update_file({"_id":file_id,"cs":usersdb.get_file_comments_sum(file_id),"s":file_data["file"]["s"]},direct_connection=True)
#si tiene comentarios se guarda el número del comentario, el usuario que lo ha escrito, el comentario en si y los votos que tiene
comments=[]
if "cs" in file_data["file"]:
comments=[(i,usersdb.find_userid(comment["_id"][0:24]),comment,comment_votes(file_id,comment)) for i,comment in enumerate(usersdb.get_file_comments(file_id,g.lang),1)]
return render_template('files/download.html',file=file_data,args=args,vote=vote,files_related=files_related,comments=comments,form=form)
def build_source_links(f):
'''
Construye los enlaces correctamente
'''
def get_domain(src):
'''
Devuelve el dominio de una URL
'''
url_parts=urlparse(src).netloc.split('.')
i=len(url_parts)-1
if len(url_parts[i])<=2 and len(url_parts[i-1])<=3:
return url_parts[i-2]+'.'+url_parts[i-1]+'.'+url_parts[i]
else:
return url_parts[i-1]+'.'+url_parts[i];
f['view']['action']='download'
f['view']['sources']={}
max_weight=0
icon=""
# agrupación de origenes
source_groups = {}
file_sources = f['file']['src'].items()
file_sources.sort(key=lambda x:x[1]["t"])
for hexuri,src in file_sources:
if not src.get('bl',None) in (0, None):
continue
url_pattern=downloader=join=False
count=0
part=url=""
source_data=g.sources[src["t"]] if "t" in src and src["t"] in g.sources else None
if source_data is None: #si no existe el origen del archivo
logging.error("El fichero contiene un origen inexistente en la tabla \"sources\": %s" % src["t"], extra={"file":f})
if feedbackdb.initialized:
feedbackdb.notify_source_error(f['file']["_id"], f['file']["s"])
continue
elif "crbl" in source_data and int(source_data["crbl"])==1: #si el origen esta bloqueado
continue
elif "w" in source_data["g"] or "f" in source_data["g"] or "s" in source_data["g"]: #si es descarga directa o streaming
link_weight=1
tip=source_data["d"]
icon="web"
source_groups[icon] = tip
source=get_domain(src['url']) if "f" in source_data["g"] else source_data["d"]
url=src['url']
if "url_pattern" in source_data and not url.startswith(("https://","http://","ftp://")):
url_pattern=True
#en caso de duda se prefiere streaming
if "s" in source_data["g"]:
f['view']['action']="listen" if f['view']['ct']==CONTENT_AUDIO else 'watch'
link_weight*=2
#torrenthash antes de torrent porque es un caso especifico
elif source_data["d"]=="BitTorrentHash":
downloader=True
link_weight=0.7 if 'torrent:tracker' in f['file']['md'] or 'torrent:trackers' in f['file']['md'] else 0.1
tip="Torrent MagnetLink"
source="tmagnet"
icon="torrent"
if not icon in source_groups:
source_groups[icon] = tip # magnet link tiene menos prioridad para el texto
join=True
count=int(src['m'])
part="xt=urn:btih:"+src['url']
if 'torrent:tracker' in f['file']['md']:
part += unicode('&tr=' + urllib.quote_plus(u(f['file']['md']['torrent:tracker']).encode("UTF-8")), "UTF-8")
elif 'torrent:trackers' in f['file']['md']:
trackers = f['file']['md']['torrent:trackers']
if isinstance(trackers, basestring):
part += unicode("".join('&tr='+urllib.quote_plus(tr) for tr in u(trackers).encode("UTF-8").split(" ")), "UTF-8")
elif "t" in source_data["g"]:
downloader=True
link_weight=0.8
url=src['url']
if "url_pattern" in source_data and not url.startswith(("https://","http://","ftp://")):
url_pattern=True
tip=source=get_domain(source_data["url_pattern"]%url)
else:
tip=source=get_domain(src['url'])
icon="torrent"
source_groups[icon] = tip
elif source_data["d"]=="Gnutella":
link_weight=0.2
tip="Gnutella"
source=icon="gnutella"
part="xt=urn:sha1:"+src['url']
join=True
count=int(src['m'])
source_groups[icon] = tip
elif source_data["d"]=="eD2k":
downloader=True
link_weight=0.1
tip="eD2k"
source=icon="ed2k"
url="ed2k://|file|"+f['view']['pfn']+"|"+str(f['file']['z'] if "z" in f["file"] else 1)+"|"+src['url']+"|/"
count=int(src['m'])
source_groups[icon] = tip
elif source_data["d"]=="Tiger":
link_weight=0
tip="Gnutella"
source=icon="gnutella"
part="xt=urn:tiger:"+src['url']
join=True
elif source_data["d"]=="MD5":
link_weight=0
tip="Gnutella"
source=icon="gnutella"
part="xt=urn:md5:"+src['url']
source_groups[icon] = tip
join=True
else:
continue
if source in f['view']['sources']:
view_source = f['view']['sources'][source]
else:
view_source = f['view']['sources'][source] = {}
view_source.update(source_data)
if 'downloader' in view_source:
if downloader:
view_source['downloader']=1
else:
view_source['downloader']=1 if downloader else 0
view_source['tip']=tip
view_source['icon']=icon
view_source['icons']=source_data.get("icons",False)
view_source['join']=join
view_source['source']="streaming" if "s" in source_data["g"] else "direct_download" if "w" in source_data["g"] else "P2P" if "p" in source_data["g"] else ""
#para no machacar el numero si hay varios archivos del mismo source
if not 'count' in view_source or count>0:
view_source['count']=count
if not "parts" in view_source:
view_source['parts']=[]
if not 'urls' in view_source:
view_source['urls']=[]
if part:
view_source['parts'].append(part)
if url:
if url_pattern:
view_source['urls']=[source_data["url_pattern"]%url]
f['view']['source_id']=url
view_source["pattern_used"]=True
elif not "pattern_used" in view_source:
view_source['urls'].append(url)
if source_data["d"]!="eD2k":
view_source['count']+=1
if link_weight>max_weight:
max_weight = link_weight
f['view']['source'] = source
f['view']['source_groups'] = sorted(source_groups.items())
if "source" not in f["view"]:
raise FileNoSources
if icon!="web":
for src,info in f['view']['sources'].items():
if info['join']:
f['view']['sources'][src]['urls'].append("magnet:?"+"&".join(info['parts'])+"&dn="+f['view']['pfn']+("&xl="+str(f['file']['z']) if 'z' in f['file'] else ""))
elif not 'urls' in info:
del(f['view']['sources'][src])
def download(file_id, file_name=""):
g.page_type = FILE_PAGE_TYPE
if request.referrer:
try:
posibles_queries = referrer_parser.match(request.referrer)
if posibles_queries:
query = posibles_queries.group(1) or posibles_queries.group(2) or ""
if query:
get_query_info(u(urllib2.unquote_plus(query).decode("utf-8")))
except:
pass
error = None
file_data=None
if file_id is not None: #si viene un id se comprueba que sea correcto
try: #intentar convertir el id que viene de la url a uno interno
file_id=url2mid(file_id)
except TypeError as e:
try: #comprueba si se trate de un ID antiguo
possible_file_id = filesdb.get_newid(file_id)
if possible_file_id is None:
logging.warn("Identificadores numericos antiguos sin resolver: %s."%e, extra={"fileid":file_id})
error=404
else:
logging.warn("Identificadores numericos antiguos encontrados: %s."%e, extra={"fileid":file_id})
return {"html": empty_redirect(url_for(".download", file_id=mid2url(possible_file_id), file_name=file_name), 301),"error":301}
except BaseException as e:
logging.exception(e)
error=503
file_id=None
if file_id:
try:
file_data=get_file_metadata(file_id, file_name.replace("-"," "))
except DatabaseError:
error=503
except FileNotExist:
error=404
except (FileRemoved, FileFoofindRemoved, FileNoSources):
error=410
except FileUnknownBlock:
error=404
if error is None and not file_data: #si no ha habido errores ni hay datos, es porque existe y no se ha podido recuperar
error=503
if error:
abort(error)
# completa datos de torrent
file_data = torrents_data(file_data, True, g.category)
if not file_data:
abort(404)
if file_data["view"]["category"]:
g.category = file_data["view"]["category"]
if file_data["view"]["category"].tag=="p**n":
g.is_adult_content = True
else:
g.category = file_data["view"]["category_type"]
# no permite acceder ficheros que deberian ser bloqueados
prepared_phrase = blacklists.prepare_phrase(file_data['view']['nfn'])
if prepared_phrase in blacklists["forbidden"] or (prepared_phrase in blacklists["misconduct"] and prepared_phrase in blacklists["underage"]):
g.blacklisted_content = "File"
if not g.show_blacklisted_content:
abort(404)
query = download_search(file_data, file_name, "torrent").replace("-"," ")
related = single_search(query, category=None, not_category=(None if g.is_adult_content else "p**n"), title=("Related torrents",3,None), zone="File / Related", last_items=[], limit=30, max_limit=15, ignore_ids=[mid2hex(file_id)], show_order=None)
# elige el titulo de la página
title = file_data['view']['fn']
# recorta el titulo hasta el proximo separador
if len(title)>101:
for pos in xrange(101, 30, -1):
if title[pos] in SEPPER:
title = title[:pos].strip()
break
else:
title = title[:101]
g.title = [title]
page_description = ""
if "description" in file_data["view"]["md"]:
page_description = file_data["view"]["md"]["description"].replace("\n", " ")
if not page_description:
if g.category:
page_description = _("download_category_desc", category=singular_filter(g.category.title).lower(), categorys=g.category.title.lower()).capitalize()
else:
page_description = _("download_desc")
if len(page_description)<50:
if page_description:
page_description += ". "
page_description += " ".join(text.capitalize()+"." for text in related[1]["files_text"])
if len(page_description)>180:
last_stop = page_description[:180].rindex(".") if "." in page_description[:180] else 0
if last_stop<100:
last_stop = page_description[:180].rindex(" ") if " " in page_description[:180] else 0
if last_stop<100:
last_stop = 180
page_description = page_description[:last_stop]+"."
g.page_description = page_description
is_canonical_filename = file_data["view"]["seo-fn"]==file_name
# registra visita al fichero
if g.search_bot:
searchd.log_bot_event(g.search_bot, True)
else:
save_visited([file_data])
if related[0]:
g.must_cache = 3600
# last-modified
g.last_modified = file_data["file"]["ls"]
return render_template('file.html', related_query = query, file_data=file_data, related_files=related, is_canonical_filename=is_canonical_filename, featured=get_featured(related[1]["count"]+len(file_data["view"]["md"]), 1))
def choose_filename(f,text_cache=None):
'''
Elige el archivo correcto
'''
srcs = f['file']['src']
fns = f['file']['fn']
chosen = None
max_count = -1
current_weight = -1
if text_cache and text_cache[0] in fns: # Si text es en realidad un ID de fn
chosen = text_cache[0]
else:
for hexuri,src in srcs.items():
if 'bl' in src and src['bl']!=0:
continue
for crc,srcfn in src['fn'].items():
if crc not in fns: #para los sources que tienen nombre pero no estan en el archivo
continue
#si no tiene nombre no se tiene en cuenta
m = srcfn['m'] if len(fns[crc]['n'])>0 else 0
if 'c' in fns[crc]:
fns[crc]['c']+=m
else:
fns[crc]['c']=m
text_weight = 0
if text_cache:
fn_parts = slugify(fns[crc]['n']).strip().split(" ")
if len(fn_parts)>0:
text_words = slugify(text_cache[0]).split(" ")
# valora numero y orden coincidencias
last_pos = -1
max_length = length = 0
occurrences = [0]*len(text_words)
for part in fn_parts:
pos = text_words.index(part) if part in text_words else -1
if pos != -1 and (last_pos==-1 or pos==last_pos+1):
length += 1
else:
if length > max_length: max_length = length
length = 0
if pos != -1:
occurrences[pos]=1
last_pos = pos
if length > max_length: max_length = length
text_weight = sum(occurrences)*100 + max_length
f['file']['fn'][crc]['tht'] = text_weight
better = fns[crc]['c']>max_count
if text_weight > current_weight or (better and text_weight==current_weight):
current_weight = text_weight
chosen = crc
max_count = fns[crc]['c']
f['view']['url'] = mid2url(hex2mid(f['file']['_id']))
f['view']['fnid'] = chosen
if chosen:
filename = fns[chosen]['n']
ext = fns[chosen]['x']
else: #uses filename from src
filename = ""
for hexuri,src in srcs.items():
if src['url'].find("/")!=-1:
filename = src['url']
if filename=="":
return
filename = filename[filename.rfind("/")+1:]
ext = filename[filename.rfind(".")+1:]
filename = filename[0:filename.rfind(".")]
#TODO si no viene nombre de archivo buscar en los metadatos para formar uno (por ejemplo serie - titulo capitulo)
filename = extension_filename(filename,ext)
f['view']['fn'] = filename.replace("?", "")
f['view']['qfn'] = qfn = u(filename).encode("UTF-8") #nombre del archivo escapado para generar las url de descarga
f['view']['pfn'] = urllib.quote(qfn).replace(" ", "%20") # P2P filename
nfilename = seoize_text(filename, " ",True, 0)
f['view']['nfn'] = nfilename
# añade el nombre del fichero como palabra clave
g.keywords.update(set(keyword for keyword in nfilename.split(" ") if len(keyword)>1))
#nombre del archivo con las palabras que coinciden con la busqueda resaltadas
if text_cache:
f['view']['fnh'], f['view']['fnhs'] = highlight(text_cache[2],filename,True)
else:
f['view']['fnh'] = filename #esto es solo para download que nunca tiene text
return current_weight>0 # indica si ha encontrado el texto buscado
def url_search_filter(new_params, args=None, delete_params=[]):
'''
Devuelve los parametros para generar una URL de busqueda
'''
# filtros actuales sin parametros eliminados
filters = {key:value for key, value in args.iteritems() if key not in delete_params} if "all" not in delete_params else {"q":args["q"]} if "q" in args else {}
# añade paramentros nuevos
if "query" in new_params:
filters["q"] = u(new_params["query"])
if 'src' in new_params:
active_srcs = g.active_srcs
new_src = new_params["src"]
''' se genera el nuevo enlace teniendo en cuenta los origenes activos por tipo de origen
- hacer click en un tipo de origen lo activan o desactivan completamente
- hacer click en un origen lo activa o desactiva, teniendo en cuenta si el tipo de origen estaba activado
o si toca activar todo el tipo
- tiene en cuenta solo los origenes visibles en el filtro
'''
# streaming
has_streaming, toggle_streaming = "streaming" in active_srcs, "streaming" == new_src
streamings = [] if has_streaming and toggle_streaming else ["streaming"] if toggle_streaming else [value for value in g.visible_sources_streaming if (value==new_src)^(value in active_srcs)]
if streamings==g.visible_sources_streaming: streamings = ["streaming"] # activa todo el tipo de origen
# download
has_download, toggle_download = "download" in active_srcs, "download" == new_src
downloads = [] if has_download and toggle_download else ["download"] if toggle_download else [value for value in g.visible_sources_download if (value==new_src)^(value in active_srcs)]
if downloads==g.visible_sources_download: downloads = ["download"] # activa todo el tipo de origen
# p2p
has_p2p, toggle_p2p = "p2p" in active_srcs, "p2p" == new_src
p2ps = [] if has_p2p and toggle_p2p else ["p2p"] if toggle_p2p else [value for value in g.sources_p2p if (value==new_src)^(value in active_srcs)]
if p2ps==g.sources_p2p: p2ps = ["p2p"] # activa todo el tipo de origen
filters["src"] = streamings + downloads + p2ps
if 'type' in new_params:
filters["type"] = [value for value in FILTERS["type"] if (value==new_params["type"])^(value in filters["type"])] if "type" in filters else [new_params["type"]]
if 'size' in new_params:
filters["size"] = new_params["size"]
# si estan todos los tipos activados en type o src es como no tener ninguno
if "type" in filters and (not filters["type"] or all(atype in filters["type"] for atype in FILTERS["type"])):
del filters["type"]
if "src" in filters and (not filters["src"] or all(src in filters["src"] for src in FILTERS["src"].iterkeys())):
del filters["src"]
# separa query
if "q" in filters:
query = filters["q"].replace(" ","_") if filters["q"] else u""
del filters["q"]
else:
query = u""
# genera url de salida
if filters:
return g.search_url + quote_plus(query.encode('utf8')) + "/" + "/".join(param+":"+",".join(filters[param]) for param in ["type", "src", "size"] if param in filters)
else:
return g.search_url + quote_plus(query.encode('utf8'))
def generate(server, part, afilter, batch_size, output):
if not output:
output = dirname(abspath(__file__)) + "/gen/" + str(part) + "/"
ff = FilesFetcher(server, afilter, batch_size)
ff.start()
suffix = "."+str(part)
count = error_count = 0
logging.info("Comienza generación de sitemap en servidor %s."%server)
for afile in ff:
try:
count += 1
# comprueba si no está bloqueado
if int(float(afile.get("bl", 0)))!=0:
continue
# comprueba si tiene origenes validos
for src in afile["src"].itervalues():
if "t" in src and src["t"] in {3, 7, 79, 80, 81, 82, 83, 90} and int(float(src.get("bl",0)))==0:
main_src = src
break
else:
continue
filename = None
# elige algun nombre de fichero interesante
for fn in afile.get("fn",{}).itervalues():
filename = fn["n"]
if filename=="download" or IS_BTIH.match(filename) or filename.startswith("[TorrentDownloads"):
continue
extension = fn.get("x",None)
if extension and not filename.endswith("."+extension):
filename += "." + extension
break
else:
md = afile.get("md",{})
for possible_name in ("torrent:name", "torrent:title", "video:title", "video:name"):
if possible_name in md:
filename = u(md[possible_name])
break
if not filename:
filename = u(main_src["url"].rsplit("/",1)[-1])
if filename:
first_seen = afile["fs"]
get_writer(first_seen, count, output, suffix).write("<url><lastmod>%s</lastmod><loc>%%s%s-%s</loc></url>\n"%(first_seen.strftime("%Y-%m-%dT%H:%M:%SZ"), seoize_text(filename, "-", True), mid2url(afile["_id"])))
except BaseException as e:
error_count += 1
if error_count>100: raise e # ante mas de 100 errores, detiene la indexacion con error
if count%10000==0:
logging.info("Progreso de generación de sitemap del servidor %s."%(server), extra={"count":count, "error_count":error_count})
close_writers()
sort_files(output)
logging.info("Finaliza la generación de sitemap en servidor %s."%server)
