File:littleimagegif.py
#! /usr/bin/env python
from PIL import Image
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
from PIL import ImageSequence
import pyexiv2
import cStringIO
import mwclient
import uuid
import urllib
import cgi
import sys
import urllib2
import requests
import math
import tempfile
import os
# CC-BY-SA Theopolisme
def getSize(filename):
st = os.stat(filename)
return st.st_size
def gimme_image(filename,compound_site,pxl,theimage):
"""This function creates the new image, runs
metadata(), and passes along the new image's
filename.
"""
site = mwclient.Site(compound_site)
extension = os.path.splitext(theimage)[1]
extension_caps = extension[1:].upper()
temppath = "c:\\python27\\bot\\temp\\"
if extension_caps <> "GIF":
results = "SKIP"
return results
image_1 = site.Images[theimage]
image_2 = str(image_1.imageinfo['url'])
response = requests.get(image_2)
item10 = cStringIO.StringIO(response.content)
temp_file = temppath+str(uuid.uuid4()) + extension
print temp_file
f = open(temp_file,'wb')
f.write(item10.getvalue())
try:
img = Image.open(item10)
print "img.size[0]", img.size[0]
print "img.size[1]", img.size[1]
index = 0
for frame in ImageSequence.Iterator(img):
index += 1
print "Frames", index
if index > 1:
print "Animated GIF"
results="MANUAL"
return results
Fsize1=getSize(temp_file)
print "Size 1", Fsize1
basewidth = int(math.sqrt((pxl * float(img.size[0]))/(img.size[1])))
print "basewidth",basewidth
wpercent = (basewidth/float(img.size[0]))
print "wpercent",wpercent
hsize = int((float(img.size[1])*float(wpercent)))
print "hsize",hsize
original_pixel = img.size[0] * img.size[1]
print "original_pixel",original_pixel
modified_pixel = basewidth * hsize
print "modified_pixel",modified_pixel
pct_chg = 100.0 * (original_pixel - modified_pixel) / float(original_pixel)
print "pct_chg",pct_chg
if pct_chg > 5:
png_info = img.info
img = img.resize((int(basewidth),int(hsize)), Image.LANCZOS)
img.save(filename + extension, **png_info)
print filename + extension
#Sometimes convert goes belly up and we get a black picture
#Get and compare sizes of the two temp files
Fsize1=getSize(temp_file)
print "Size 1", Fsize1
Fsize2=getSize(filename + extension)
print "Size 2", Fsize2
Fsizer=Fsize1*(100-pct_chg)/200 #Orig size * % reduced and div by 2 for good measure
print "Size R min", Fsizer
if Fsize2 < Fsizer:
print "TOO SMALL"
results="MANUAL"
return results
else:
print "Looks like we'd have a less than 5% change in pixel counts. Skipping."
results = "PIXEL"
return results
except (IOError):
print "Unable to open image " + theimage.encode('ascii', 'ignore') + " (aborting)"
results = "ERROR"
return results
print "Image saved to disk at " + filename + extension
results = filename + extension
filelist = [ f for f in os.listdir("c:\\python27\\bot\\") if f.startswith(temp_file) ]
for fa in filelist: os.remove(fa)
return results
from PIL import Image
from xml.dom import minidom
import cStringIO
import mwclient
import uuid
import urllib
import os.path
import cgi
import littleimagegif
import sys
import urllib2
import re
import time
import random
import logging
import userpassbot
import requests.packages.urllib3
requests.packages.urllib3.disable_warnings()
import mwparserfromhell
# CC-BY-SA Theopolisme
#routine to autoswitch some of the output - as filenames in, say, filep.unprefixedtitle have accented chars!
def pnt(s):
try:
print(s)
except UnicodeEncodeError:
print(s.encode('utf-8'))
def startAllowed():
checkname="User:RonBot/2/Run"
page = site.Pages[checkname]
textpage = page.text()
if textpage == "Run":
return "yes"
else:
return "no"
def allow_bots(pagetext, username):
user = username.lower().strip()
text = mwparserfromhell.parse(pagetext)
for tl in text.filter_templates():
if tl.name in ('bots', 'nobots'):
break
else:
return True
for param in tl.params:
bots = [x.lower().strip() for x in param.value.split(",")]
if param.name == 'allow':
if ''.join(bots) == 'none': return False
for bot in bots:
if bot in (user, 'all'):
return True
elif param.name == 'deny':
if ''.join(bots) == 'none': return True
for bot in bots:
if bot in (user, 'all'):
return False
return True
def are_you_still_there(theimage):
""" This function makes sure that
a given image is still tagged with
{{non-free reduce}}.
"""
img_name = "File:" + theimage
page = site.Pages[img_name]
text = page.text()
r1 = re.compile(r'\{\{[Nn]on.?free-?\s*[Rr]educe.*?\}\}')
r2 = re.compile(r'\{\{[Rr]educe.*?\}\}')
r3 = re.compile(r'\{\{[Cc]omic-ovrsize-img.*?\}\}')
r4 = re.compile(r'\{\{[Ff]air.?[Uu]se.?[Rr]educe.*?\}\}')
r5 = re.compile(r'\{\{[Ii]mage-toobig.*?\}\}')
r6 = re.compile(r'\{\{[Nn]fr.*?\}\}')
r7 = re.compile(r'\{\{[Ss]maller image.*?\}\}')
if r1.search(text) is not None:
return True
elif r2.search(text) is not None:
return True
elif r3.search(text) is not None:
return True
elif r4.search(text) is not None:
return True
elif r5.search(text) is not None:
return True
elif r6.search(text) is not None:
return True
elif r7.search(text) is not None:
return True
else:
return False
def image_routine(images):
""" This function does most of the work:
* First, checks the checkpage using sokay()
* Then makes sure the image file still exists using are_you_still_there()
* Next it actually resizes the image.
* As long as the resize works, we reupload the file.
* Then we update the page with {{non-free reduced}}.
* And repeat!
"""
temppath = "c:\\python27\\bot\\temp\\"
gifdone = 0
for theimage in images:
print "Working on " + theimage.encode('ascii', 'ignore')
go = startAllowed()
if go == "yes":
if are_you_still_there(theimage) == True:
#Stop if there's nobots
img_name = "File:" + theimage
page = site.Pages[img_name]
text = page.text()
allow_bots(text, "RonBot")
print "main.allowbots"
desired_megapixel = float(0.1)
print "desired_megapixel", desired_megapixel
pxl = desired_megapixel * 1000000
print "pxl", pxl
compound_site = 'en.wikipedia.org'
filename = temppath+str(uuid.uuid4())
file = littleimagegif.gimme_image(filename,compound_site,pxl,theimage)
if file == "SKIP":
print "Skipping File."
if file == "PIXEL":
print "Removing tag...already reduced..."
img_name = "File:" + theimage
page = site.Pages[img_name]
text = page.text()
text = re.sub(r'\{\{[Nn]on.?free-?\s*[Rr]educe.*?\}\}', '', text)
text = re.sub(r'\{\{[Rr]educe.*?\}\}', '', text)
text = re.sub(r'\{\{[Cc]omic-ovrsize-img.*?\}\}', '', text)
text = re.sub(r'\{\{[Ff]air.?[Uu]se.?[Rr]educe.*?\}\}', '', text)
text = re.sub(r'\{\{[Ii]mage-toobig.*?\}\}', '', text)
text = re.sub(r'\{\{[Nn]fr.*?\}\}', '', text)
text = re.sub(r'\{\{[Ss]maller image.*?\}\}', '', text)
page.save(text, bot=True, summary = "(Task 2) Removing {{[[Template:Non-free reduce|Non-free reduce]]}} since file is already adequately reduced")
if file == "MANUAL":
print "Changing tag...cannot reduced..."
img_name = "File:" + theimage
page = site.Pages[img_name]
text = page.text()
if "{{Non-free manual gif reduce}}" not in text:
text = re.sub(r'\{\{[Nn]on.?free-?\s*[Rr]educe.*?\}\}', '{{Non-free manual gif reduce}}', text)
text = re.sub(r'\{\{[Rr]educe.*?\}\}', '{{Non-free manual gif reduce}}', text)
text = re.sub(r'\{\{[Cc]omic-ovrsize-img.*?\}\}', '{{Non-free manual gif reduce}}', text)
text = re.sub(r'\{\{[Ff]air.?[Uu]se.?[Rr]educe.*?\}\}', '{{Non-free manual gif reduce}}', text)
text = re.sub(r'\{\{[Ii]mage-toobig.*?\}\}', '{{Non-free manual gif reduce}}', text)
text = re.sub(r'\{\{[Nn]fr.*?\}\}', '{{Non-free manual gif reduce}}', text)
text = re.sub(r'\{\{[Ss]maller image.*?\}\}', '{{Non-free manual gif reduce}}', text)
page.save(text, bot=True, summary = "(Task 2) Tagging with [[Template:Non-free manual gif reduce]] - bot cannot reduce")
elif file not in ("ERROR", "PIXEL", "SKIP", "MANUAL"):
try:
Fcomment="Reduce size of non-free image to NFCC guideline"
site.upload(open(file,"rb"), filename=theimage, ignore=True, description=Fcomment, comment=Fcomment)
gifdone += 1
print "Uploaded!"
filelist = [ f for f in os.listdir(".") if f.startswith(filename) ]
for fa in filelist: os.remove(fa)
img_name = "File:" + theimage
page = site.Pages[img_name]
text = page.text()
text = re.sub(r'\{\{[Nn]on.?free-?\s*[Rr]educe.*?\}\}', '{{subst:orfurrev}}', text)
text = re.sub(r'\{\{[Rr]educe.*?\}\}', '{{subst:orfurrev}}', text)
text = re.sub(r'\{\{[Cc]omic-ovrsize-img.*?\}\}', '{{subst:orfurrev}}', text)
text = re.sub(r'\{\{[Ff]air.?[Uu]se.?[Rr]educe.*?\}\}', '{{subst:orfurrev}}', text)
text = re.sub(r'\{\{[Ii]mage-toobig.*?\}\}', '{{subst:orfurrev}}', text)
text = re.sub(r'\{\{[Nn]fr.*?\}\}', '{{subst:orfurrev}}', text)
text = re.sub(r'\{\{[Ss]maller image.*?\}\}', '{{subst:orfurrev}}', text)
page.save(text, bot=True, summary = "(Task 2) Tagging with [[Template:orfurrev]]")
print "Tagged!", gifdone
#if gifdone==200: # Batches of 200 are nice number to check up
#break
except:
print "Unknown error. Image skipped."
filelist = [ f for f in os.listdir(".") if f.startswith(filename) ]
for fa in filelist: os.remove(fa)
else:
print "Image skipped."
filelist = [ f for f in os.listdir(".") if f.startswith(filename) ]
for fa in filelist: os.remove(fa)
else:
print "Gah, looks like someone removed the tag."
else:
print "Ah, darn - looks like the bot was disabled."
return
def main():
"""This defines and fills a global
variable for the site, and then calls
get_images() to assemble an initial
selection of images to work with. Then
it runs image_rountine() on this selection.
"""
global site
site = mwclient.Site('en.wikipedia.org')
site.login(userpassbot.username, userpassbot.password)
zam = mwclient.listing.Category(site, "Category:Wikipedia non-free file size reduction requests")
glob = zam.members()
flub = []
for image in glob:
zip = image.page_title
pnt(zip)
#rough and ready sort out of gif images - will save a lot of processing time - might let some worng ones through, not an issue.
if "gif" in zip.lower():
flub.append(zip)
image_routine(flub)
print "We're DONE!"
if __name__ == '__main__':
main()