From 86625e3031d23b9ac761b2f853bb51fe59639891 Mon Sep 17 00:00:00 2001
From: Marco Pfomann <1pfomann@informatik.uni-hamburg.de>
Date: Wed, 2 Apr 2014 13:23:26 +0200
Subject: [PATCH] previously coded crawler in python

---
 Python/Konachan-Downloader_v3.py | 146 +++++++++++++++++++++++++++++++
 1 file changed, 146 insertions(+)
 create mode 100755 Python/Konachan-Downloader_v3.py

diff --git a/Python/Konachan-Downloader_v3.py b/Python/Konachan-Downloader_v3.py
new file mode 100755
index 0000000..0c1433e
--- /dev/null
+++ b/Python/Konachan-Downloader_v3.py
@@ -0,0 +1,146 @@
+#! /usr/bin/python
+
+from __future__ import print_function
+
+import re
+import os
+import os.path
+import sys
+import http.client
+import urllib.request
+import time
+
+# regexes
+url_regex = re.compile("konachan.com/image/.+?/.+?\.(?:png|jpg)")
+name_regex = re.compile("image/.*?/(.*)")
+
+# variable
+counter1, counter2 = 0, 15000
+tag_filter = None
+domain = http.client.HTTPConnection("konachan.com")
+
+# little function to calculate the last page of search results
+
+
+def page_count():
+    # open connection to konachan.com
+    domain = http.client.HTTPConnection("konachan.com")
+
+    domain.request("GET", "/post?page=1&tags=" + tags.replace(" ", "+"))
+    while True:
+        try:
+            first_page = domain.getresponse()
+            break
+        except http.client.BadStatusLine:
+            time.sleep(1)
+            domain.close()
+            domain = http.client.HTTPConnection("konachan.com")
+            domain.request("GET", "/post?page=1&tags=" + tags.replace(" ", "+"))
+
+    # we got our response, now it's time to find that number
+    first_page_source = str(first_page.read())
+    page_list = first_page_source.split("Next Page")
+    number = 0
+    for line in page_list:
+        if re.search("(?<=\/post\?page\=)\d+", line):
+            number = re.search("(?<=\/post\?page\=)\d+", line).group(0)
+        else:
+            number = 2
+    return int(number)
+
+# we don't want to save every picture in one directory.
+# so we create a new directory when we donwloaded 15k pics
+
+
+def directory_size(directory_intern):
+    if len(os.listdir(directory_intern)) >= 15000:
+        print("Directory " + directory_intern + " full")
+        counter1 += 15000
+        counter2 += 15000
+        directory = "Pics " + str(counter1) + " - " + str(counter2)
+        if os.path.isdir(directory):
+            print("Directory already exists; skip creation")
+        else:
+            os.makedirs(directory, 0o755, False)
+        os.chdir("..")
+
+# now we start
+
+# user has to set path for pictures
+print("Please set download location (full path required): ")
+path = sys.stdin.readline()
+
+# set tags, if user want to download specific pictures
+print("Set Tags (seperate multiple tags with a whitespace;" +
+    " connect tags with more than one word with an underscore): ")
+tags = sys.stdin.readline().strip("\n")
+
+# chdir in $path and create directory if it not exists
+if not os.path.isdir(path.rstrip()):
+    os.makedirs(path.rstrip(), 0o755, True)
+os.chdir(path.rstrip())
+if not os.path.isdir("Tags: " + tags):
+    os.makedirs("Tags: " + tags, 0o755, True)
+os.chdir("Tags: " + tags)
+
+
+# creating directory for pics
+directory = "Pics " + str(counter1) + " - " + str(counter2)
+if not os.path.isdir(directory):
+    os.makedirs(directory, 0o755, True)
+
+# let's start with downloading
+
+for page_number in range(1, page_count()):
+
+    print("Starting download in page " + str(page_number))
+
+    domain.request("GET", "/post?page=" + str(page_number) +
+    "&tags=" + tags.replace(" ", "+"))
+
+    while True:
+        try:
+            index_page = domain.getresponse()
+            break
+        except http.client.BadStatusLine:
+            domain.close()
+            domain = http.client.HTTPConnection("konachan.com")
+            domain.request("GET", "/post?page=" + str(page_number) +
+             "&tags=" + tags.replace(" ", "+"))
+            time.sleep(1)
+
+    # after we got the response from konachan we need the source code
+    index_page_source = str(index_page.read())
+
+    # and now we need save every link on this page in a list
+    pics_list = index_page_source.split("Post.register")
+
+    directory_size(directory)
+
+    # now we can search every line for the pic link
+    for pic in pics_list:
+        pic_url = url_regex.search(re.sub("\\\\\\\\", "", pic))
+
+        # if we found the url we download the pic
+        # but with whitespaces instead of "%20"
+        if pic_url:
+            name = name_regex.search(pic_url.group(0)).group(1)
+            print("     Downloading pic:  " + name.replace("%20", " ") +
+            " in directory: " + directory)
+
+            # a little check if pic already exists
+            existance = False
+            for dir in os.listdir():
+                os.chdir(dir)
+                if os.path.isfile(name.replace("%20", " ")):
+                    print("     Pic is already on your pc! Skip!")
+                    existance = True
+                os.chdir("..")
+
+            if not existance:
+                os.chdir(directory)
+                image = urllib.request.URLopener()
+                image.retrieve("http://" +
+                    pic_url.group(0), urllib.request.url2pathname(name))
+                print("     Download finished")
+                os.chdir("..")