dato198613
Super Member
ჯგუფი: Registered
წერილები: 226
წევრი No.: 41033
რეგისტრ.: 2-September 07
|
#35056539 · 5 Feb 2013, 14:46 · · პროფილი · პირადი მიმოწერა · ჩატი
არავინ იცით? C:\Users\User\Desktop\extractionGoogleArtProject\extractionGoogleArtProject>ext ractiongoogleartproject.py telechargerOeuvre('http://www.googleartproject.com/co llection/musee-dorsay-paris/artwork/the-saint-lazare-station-claude-monet/508102 /', 'Monet_-_Gare_Saint-Lazare.jpg', 3) File "C:\Users\User\Desktop\extractionGoogleArtProject\extractionGoogleArtPro ject\extractionGoogleArtProject.py", line 32 SyntaxError: Non-ASCII character '\xe9' in file C:\Users\User\Desktop\extractio nGoogleArtProject\extractionGoogleArtProject\extractionGoogleArtProject.py on li ne 32, but no encoding declared; see http://www.python.org/peps/pep-0263.html fo r details ამ შეცდომას მიწერს * * * | CODE | import PyV8
import re, urllib2, time, os, json, unicodedata, base64, subprocess
from xml.dom import minidom
### configuration
cheminDossierFragments = "fragments"
cheminDossierImages = "images"
cheminDossierInfos = "infos"
cheminDossierImageMagick = "ImageMagick"
cheminjava[ertad]script = "core.js"
refererGoogleArtProject = "http://www.googleartproject.com/"
dureeSleep = 0.5
separateur = "\t"
### contenu d'une page Web
def getContenuUrl(url, referer = ""):
time.sleep(dureeSleep)
requete = urllib2.Request(url)
if referer:
requete.add_header("Referer", referer)
print "(%s)" % url
return urllib2.urlopen(requete).read()
### URL d'un fragment d'image, dйcryptage d'un fragment d'image
context = PyV8.JSContext()
context.enter()
context.evаl(open(cheminjava[ertad]script, "r").read())
def getUrlFragment(urlImage, x, y, zoom, timestamp):
return context.evаl("getUrlFragment('%s', '%i', '%i', '%i', '%i')" % (urlImage, x, y, zoom, timestamp))
def decrypterFragment(contenuFragment):
arrayFragment = [ord(char) & 0xFF for char in contenuFragment]
base64Fragment = context.evаl("decrypterFragment(%s)" % str(arrayFragment))
return base64.b64decode(base64Fragment)
### infos sur un tableau (peintre, titre, date...)
def getInfosTableau(urlPageTableau):
regexJsonTableau = re.compile("""var CURRENT_ARTWORK = new ap.Artwork\((.+)\);""")
infosTableau = {}
contenuPageTableau = getContenuUrl(urlPageTableau, refererGoogleArtProject)
jsonTableau = json.loads(re.findall(regexJsonTableau, contenuPageTableau)[0])
infosTableau["urlImage"] = str(jsonTableau["aggregation_image_url"])
if infosTableau["urlImage"][:5] <> "http:":
infosTableau["urlImage"] = "http:" + infosTableau["urlImage"]
infosTableau["peintre"] = jsonTableau["artist_display_name"]
infosTableau["titre"] = jsonTableau["title"]
try:
infosTableau["date"] = str(jsonTableau["pretty_display_date"])
except:
infosTableau["date"] = ""
try:
infosTableau["titre original"] = jsonTableau["facets"]["Original Title"][0]
except:
infosTableau["titre original"] = ""
try:
infosTableau["autre titre"] = jsonTableau["facets"]["Non-English title"][0]
except:
infosTableau["autre titre"] = ""
try:
infosTableau["mouvements"] = jsonTableau["facets"]["Style"][0]
except:
infosTableau["mouvements"] = ""
try:
infosTableau["techniques"] = jsonTableau["facets"]["Medium"][0]
except:
infosTableau["techniques"] = ""
return infosTableau
### tйlйchargements des fragments d'image
def getInfosFragments(urlImage, zoom):
docXml = minidom.parse(urllib2.urlopen(urlImage + "=g"))
largeurFragment = int(docXml.firstChild.attributes["tile_width"].value)
hauteurFragment = int(docXml.firstChild.attributes["tile_height"].value)
zoomMax = int(docXml.firstChild.attributes["full_pyramid_depth"].value) - 1
if zoom > zoomMax:
zoom = zoomMax
xMax = int(docXml.getElementsByTagName("pyramid_level")[zoom].attributes["num_tiles_x"].value)
yMax = int(docXml.getElementsByTagName("pyramid_level")[zoom].attributes["num_tiles_y"].value)
return zoom, xMax, yMax, largeurFragment, hauteurFragment
def telechargerFragment(urlImage, cheminFragment, x, y, zoom):
timestamp = int(time.time())
urlFragment = getUrlFragment(urlImage, x, y, zoom, timestamp)
contenuFragment = getContenuUrl(urlFragment, refererGoogleArtProject)
contenuFragment = decrypterFragment(contenuFragment)
fichierFragment = open(cheminFragment, "wb")
fichierFragment.write(contenuFragment)
fichierFragment.close()
def telechargerTousFragments(urlImage, xMax, yMax, zoom):
i = 0
for y in range(yMax):
for x in range(xMax):
i = i+1
cheminFragment = os.path.join(cheminDossierFragments, "fragment_%s.jpg" % format(i, "03d"))
telechargerFragment(urlImage, cheminFragment, x, y, zoom)
### reconstitution de l'image а partir des fragments
def reconstituerImage(nomFichierImage, xMax, yMax, largeurFragment, hauteurFragment):
commandeAssembler = (os.path.join(cheminDossierImageMagick, "montage.###")
+ " " + os.path.join(cheminDossierFragments, "fragment_[0-9]*.jpg")
+ " -quality 100"
+ " -tile " + str(xMax) + "x" + str(yMax)
+ " -geometry " + str(largeurFragment) + "x" + str(hauteurFragment)
+ " " + os.path.join(cheminDossierImages, nomFichierImage))
commandeRogner = (os.path.join(cheminDossierImageMagick, "mogrify.###")
+ " -quality 100"
+ " -trim"
+ " -fuzz 10%"
+ " " + os.path.join(cheminDossierImages, nomFichierImage))
p1 = subprocess.Popen(commandeAssembler, shell=True, creationflags=0x08000000)
p1.communicate()
p2 = subprocess.Popen(commandeRogner, shell=True, creationflags=0x08000000)
p2.communicate()
### liste des tableaux d'un peintre
def getUrlPagesTableaux(idPeintre):
listeUrlPagesTableaux = []
urlJsonTableaux = "http://www.googleartproject.com/api/int/gap2/artwork/?canonical_artist=%i&limit=500&offset=0&format=json" % idPeintre
jsonTableau = json.loads(getContenuUrl(urlJsonTableaux, refererGoogleArtProject))
for tableau in jsonTableau["objects"]:
listeUrlPagesTableaux.append("http://www.googleartproject.com" + str(tableau["absolute_url"]))
return listeUrlPagesTableaux
### normalisation d'un chaоne de caractиres
def normaliserChaine(chaine):
# encodage UTF-8
try:
chaine = unicode(chaine)
except:
chaine = unicode(chaine, "iso-8859-1")
# suppression des accents
chaine = unicodedata.normalize("NFKD", chaine)
# encodage ASCII
chaine = chaine.encode("ASCII", "ignore")
return chaine
def normaliserNomFichier(chaine):
chaine = normaliserChaine(chaine)
# caractиre non alpha-numйrique => "_"
chaine = re.sub("[^0-9a-zA-Z\.\-]", "_", chaine)
return chaine
### tйlйchargement des images et infos
def nettoyerDossier(cheminDossier):
for nomFichier in os.listdir(cheminDossier):
cheminFichier = os.path.join(cheminDossier, nomFichier)
if os.path.isfile(cheminFichier):
os.remove(cheminFichier)
def telechargerTableau(urlImage, nomFichierImage, zoom):
zoom, xMax, yMax, largeurFragment, hauteurFragment = getInfosFragments(urlImage, zoom)
nettoyerDossier(cheminDossierFragments) # nettoyage du dossier des fragments
telechargerTousFragments(urlImage, xMax, yMax, zoom)
reconstituerImage(nomFichierImage, xMax, yMax, largeurFragment, hauteurFragment)
def telechargerTableauxPeintre(nomPeintre, idPeintre, zoom):
listeChamps = ["image", "peintre", "titre", "date", "titre original", "autre titre", "mouvements", "techniques"]
fichierInfos = open(os.path.join(cheminDossierInfos, normaliserNomFichier(nomPeintre) + ".csv"), "w")
fichierInfos.write(separateur.join(listeChamps) + "\n")
listeUrlPagesTableaux = getUrlPagesTableaux(idPeintre)
print "### %s : %i tableaux" % (nomPeintre, len(listeUrlPagesTableaux))
i = 0
for urlPageTableau in listeUrlPagesTableaux:
i = i+1
print "# %s, tableau %i/%i : %s" % (nomPeintre, i, len(listeUrlPagesTableaux), urlPageTableau)
infosTableau = getInfosTableau(urlPageTableau)
urlImage = infosTableau["urlImage"]
nomFichierImage = normaliserNomFichier(nomPeintre) + "-" + format(i, "03d") + ".jpg"
telechargerTableau(urlImage, nomFichierImage, zoom)
infosTableau["image"] = nomFichierImage
fichierInfos.write(separateur.join([normaliserChaine(infosTableau[champ]) for champ in listeChamps]) + "\n")
fichierInfos.close()
### fonctions principales
def telechargerOeuvre(urlOeuvre, nomFichierImage, zoom):
contenuPageOeuvre = getContenuUrl(urlOeuvre, refererGoogleArtProject)
regexUrlImage = re.compile("""data-image-url=\"([^\"]+)\"""")
urlImage = re.findall(regexUrlImage, contenuPageOeuvre)[0]
telechargerTableau(urlImage, normaliserNomFichier(nomFichierImage), zoom)
def telechargerArtiste(urlArtiste, zoom):
contenuPageArtiste = getContenuUrl(urlArtiste, refererGoogleArtProject)
regexNomArtiste = re.compile("""data-artist-name=\"([^\"]+)\"""")
regexIdArtiste = re.compile("""data-artist-id=\"([^\"]+)\"""")
nomArtiste = re.findall(regexNomArtiste, contenuPageArtiste)[0]
idArtiste = int(re.findall(regexIdArtiste, contenuPageArtiste)[0])
nomArtiste
telechargerTableauxPeintre(nomArtiste, idArtiste, zoom)
### exйcution du script
#telechargerOeuvre('http://www.googleartproject.com/collection/musee-dorsay-paris/artwork/the-saint-lazare-station-claude-monet/508102/', 'Monet_-_Gare_Saint-Lazare.jpg', 3)
#telechargerArtiste('http://www.googleartproject.com/artist/claude-monet/4127022/', 2)
|
იუნიკოდზე აქვს პრობლემა მაგ. 32 ხაზზე ### URL d'un fragment d'image, dйcryptage d'un fragment d'image
სხვაგანაც არის და ყველას რო ამოვიღებ და ისე ვუშვებ მერე მიწერს
| CODE | C:\Users\User\Desktop\extractionGoogleArtProject\extractionGoogleArtProject>extractionGoogleArtProject.py telechargerOeuvre('http://www.googleartproject.com/collection/musee-dorsay-paris/artwork/the-saint-lazare-station-claude-monet/508102/', 'Monet_-_Gare_Saint-Lazare.jpg', 2)
Traceback (most recent call last):
File "C:\Users\User\Desktop\extractionGoogleArtProject\extractionGoogleArtProject\extractionGoogleArtProject.py", line 36, in <module>
context.evаl(open(cheminjava[ertad]script, "r").read())
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe9 in position 138117: ordinal not in range(128) |
ვინმე დამეხმარეთ რა..
* * *
გავაკეთე ეგ პრობლება. კიდე არის core.js და იქ იყო // URL d'un fragment d'image, dйcryptage d'un fragment d'image ეს ხაზი და კიდე იუნიკოდის პრობლემა ქონდა. ეხლა რო ვაწვები შეცდომას აღარ მიწერს მაგრამ არც სურათს არ იწერს...
|
Goole art project, Python
