balkian/gists
1
0
Fork 0
mirror of https://github.com/balkian/gists.git synced 2024-12-04 14:12:29 +00:00
Code Issues Projects Releases Wiki Activity

Browse Source
This commit is contained in:
Fernando Sánchez 2013-07-30 08:27:43 -07:00
parent 43d0d4b46a
commit 543d7ca50a
1 changed files with 11 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
group_pictures_UPM.py
View File

@ -1,6 +1,7 @@
import mechanize, cookielib import mechanize, cookielib
from BeautifulSoup import BeautifulSoup from BeautifulSoup import BeautifulSoup
import re, os import re, os
from zipfile import ZipFile
def get_image(tag, folder): def get_image(tag, folder):
if not os.path.isdir(folder): if not os.path.isdir(folder):
@ -12,6 +13,8 @@ def get_image(tag, folder):
save = open(filename, 'wb') save = open(filename, 'wb')
save.write(data) save.write(data)
save.close() save.close()
return filename
br=mechanize.Browser() br=mechanize.Browser()
cj = cookielib.LWPCookieJar() cj = cookielib.LWPCookieJar()
@ -23,6 +26,8 @@ br.set_handle_referer(True)
br.set_handle_robots(False) br.set_handle_robots(False)
br.addheaders = [('User-agent', 'Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.0.1) Gecko/2008071615 Fedora/3.0.1-1.fc9 Firefox/3.0.1')] br.addheaders = [('User-agent', 'Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.0.1) Gecko/2008071615 Fedora/3.0.1-1.fc9 Firefox/3.0.1')]
tozip = [];
#Individuales #Individuales
r = br.open('http://www.etsit.upm.es/fotospromo85.html') r = br.open('http://www.etsit.upm.es/fotospromo85.html')
br.select_form(nr=0) br.select_form(nr=0)
@ -36,7 +41,7 @@ images_tags = galery(href=re.compile(r'pics/.{10}.jpg'))
folder = 'individuales' folder = 'individuales'
for image in images_tags: for image in images_tags:
url = image['href'] url = image['href']
get_image(url, folder) tozip.append(get_image(url, folder))
# Ahora las de grupo # Ahora las de grupo
link=r_tags(text=re.compile(r'grupo'))[0].parent['href'] link=r_tags(text=re.compile(r'grupo'))[0].parent['href']
@ -45,4 +50,8 @@ images_tags = galery(href=re.compile(r'pics/.{10}.jpg'))
folder = 'grupo' folder = 'grupo'
for image in images_tags: for image in images_tags:
url = image['href'] url = image['href']
get_image(url, folder) tozip.append(get_image(url, folder))
with ZipFile('fotos-85.zip', 'w') as myzip:
for i in tozip:
myzip.write(i)