- Moved cookie support code to user.py.

- Created http.py.
- Added support for httpauth, based on a patch from Johannes Wolter.
- Added support for https.


git-svn-id: http://svn.code.sf.net/p/wikipediafs/code/branches/fuse-python-new-api@25 59acd704-e115-0410-a914-e735a229ed7c

4 files changed, 427 insertions, 6 deletions

diff --git a/src/wikipediafs/article.py b/src/wikipediafs/article.py
new file mode 100644
index 0000000..593a5da
--- /dev/null
+++ b/src/wikipediafs/article.py
@@ -0,0 +1,200 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+# WikipediaFS
+# Copyright (C) 2005 - 2007 Mathieu Blondel
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License along
+# with this program; if not, write to the Free Software Foundation, Inc.,
+# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+
+import urllib, re, os
+from sgmllib import SGMLParser
+from http import ExtendedHTTPConnection
+
+class Article(SGMLParser):
+    """
+    Gets and sets an article.
+    """
+    
+    def __init__(self,
+                 name,
+                 host,
+                 basename,
+                 cookie_str=None,
+                 https=False,
+                 port=None,
+                 httpauth_username=None,
+                 httpauth_password=None
+                 ):
+        SGMLParser.__init__(self)
+    
+        self.name = name
+        self.host = host
+        self.basename = basename
+        self.cookie_str = cookie_str
+        self.https = https
+        self.port = port
+        self.httpauth_username = httpauth_username
+        self.httpauth_password = httpauth_password
+                 
+        self.content = ""
+        self.textarea = False
+        self.wpEdittime = 0
+        self.wpStarttime = 0
+        self.wpEditToken = None
+        self.last_open_time = 0
+
+        # url patterns        
+        self.edit_page = "%s?title=%s&action=edit" % \
+                            (self.basename, self.name)
+                            # basename must include a leading /
+        
+        self.submit_page = "%s?title=%s&action=submit" % \
+                            (self.basename, self.name)                          
+
+    def start_textarea(self,attrs):
+        """
+        Called when a textarea is entered.
+        """
+        self.textarea = True
+    
+    def start_input(self,attrs):
+        """
+        Called when an input is entered.
+        """
+        # To set an article, we need to now its wpEdittime first.
+        
+        if len(attrs) == 3 and attrs[2][1] == "wpEdittime":
+            self.wpEdittime = attrs[1][1]
+        elif len(attrs) == 3 and attrs[2][1] == "wpEditToken":
+            self.wpEditToken = attrs[1][1]            
+        elif len(attrs) == 3 and attrs[2][1] == "wpStarttime":
+            self.wpStarttime = attrs[1][1]            
+            
+    def end_textarea(self):
+        """
+        Called when a textarea is left.
+        """
+        self.textarea = False
+        
+    def handle_data(self,data):
+        """
+        Called when data is parsed.
+        """            
+        # We add the parsed data to self.content when the data parsed
+        # is in a textarea
+        if self.textarea == True:
+            self.content += data
+
+    def get(self): 
+        """
+        Gets the wiki content (not the whole html page).
+        """
+        
+        headers = {"User-agent" : "WikipediaFS"}
+          
+        if self.cookie_str is not None:
+            headers["Cookie"] = self.cookie_str
+              
+        conn = ExtendedHTTPConnection(self.host, self.port, self.https)
+
+        conn.set_proxy()
+
+        if self.httpauth_username and self.httpauth_password:
+            conn.http_auth(httpauth_username, httpauth_password)
+
+        conn.add_headers(headers)
+        conn.request(self.edit_page)
+        #logger.info("HTTP GET %s" % self.edit_page)
+        response = conn.getresponse()        
+        
+        # Feeds the SGMLparser
+        self.feed(response.read())
+        conn.close()
+
+        return self.content
+              
+        
+    def set(self, text):
+        # Looking for a [[Summary:*]]
+        regexp = '((\[\[)((s|S)ummary:)(.*)(\]\])(( )*\n)?)'
+        summary = re.search(regexp, text)
+        if summary is not None:
+            wpSummary = summary.group(5)
+            text = text.replace(summary.group(1), '')
+        else:
+            wpSummary = " "
+        
+        # wpEdittime is empty if the article is a new article
+        params = {
+            "wpTextbox1" : text,
+            "wpSummary" : wpSummary,
+            "wpEdittime" : self.wpEdittime,
+            "wpStarttime": self.wpStarttime,
+            "wpSave" : 1
+        }
+        
+        # Needed for logged in edition
+        if self.wpEditToken is not None:
+            params["wpEditToken"] = self.wpEditToken
+                
+        params = urllib.urlencode(params)
+        
+        headers = {"Content-type": "application/x-www-form-urlencoded",
+                   "User-agent" : "WikipediaFS"}
+        
+        if self.cookie_str is not None:
+            headers["Cookie"] = self.cookie_str
+        
+        conn = ExtendedHTTPConnection(self.host, self.port, self.https)
+        conn.set_proxy()
+
+        conn.add_headers(headers)
+        conn.add_data(params)
+        conn.request(self.submit_page)
+        #logger.info("HTTP POST %s" % self.submit_page)
+        response = conn.getresponse()
+        
+        # Log http response        
+        #if response.status == 302:
+        #    logger.info("Succesful")
+        #elif response.status == 200:
+        #    logger.warning("Problems occured %s\n" % response.read())
+        #else:
+        #    logger.info("%d \n %s " % (response.status,response.read()))
+
+        conn.close()
+
+
+if __name__ == "__main__":
+    import random
+    import sys
+    from user import User
+    
+    params = {
+        "host" : "www.mblondel.org",
+        "basename" : "/mediawiki/index.php",
+        "https" : True
+    }
+
+    # Used username and password if any
+    if len(sys.argv) == 3:
+        user = User(sys.argv[1], sys.argv[2], **params)
+        params["cookie_str"] = user.getCookieString()
+    
+    art = Article("Test", **params)
+    print art.get()
+
+    art.set("Test ! (%s)" % str(random.random()))
+    
\ No newline at end of file
diff --git a/src/wikipediafs/config.py b/src/wikipediafs/config.py
index 64b3cc0..5ad5941 100644
--- a/src/wikipediafs/config.py
+++ b/src/wikipediafs/config.py
@@ -35,13 +35,30 @@ class Config:
         <article-cache-time>300</article-cache-time>
     </general>
     <sites>
-        <!-- <site>
+        <!-- 
+        Minimalist site entry sample:
+        <site>
             <dirname>wikipedia-fr</dirname>
             <host>fr.wikipedia.org</host>
             <basename>/w/index.php</basename>
+        </site>
+        And another one with all possible informations:
+        <site>
+            <dirname>wikipedia-fr</dirname>
+            <host>fr.wikipedia.org</host>
+            <port>443</port>
+            <basename>/w/index.php</basename>
             <username>Username</username>
             <password>Password</password>
-        </site>-->
+            <https />
+            <httpauth_username>Username</httpauth_username>
+            <httpauth_password>Password</httpauth_password>
+        </site>
+        -->        
+        <!--
+            Below a Mediawiki test site.
+            Feel free to use it!
+        --> 
         <site>
             <dirname>mblondel.org</dirname>
             <host>www.mblondel.org</host>
@@ -89,14 +106,19 @@ class Config:
         sites = self.__config.getElementsByTagName("site")
         for site in sites:
              dic = {}
-             for ele in ('dirname', 'host', 'basename', 'username',
-             'password'):
+             for ele in ("dirname", "host", "basename", "username",
+             "password", "https", "port", "httpauth_username",
+             "httpauth_password"):
                 node = site.getElementsByTagName(ele)
                 if node.length == 1:
-                    dic[ele] = node[0].firstChild.nodeValue.encode("utf-8")
+                    if node[0].firstChild:
+                        dic[ele] = node[0].firstChild.nodeValue.encode("utf-8")
+                    else:
+                        dic[ele] = True # for elements like <https />
                 else:
                     dic[ele] = None
-             self.sites[dic['dirname']] = dic
+                    
+             self.sites[dic["dirname"]] = dic
 
 
     def __setCacheTime(self):
@@ -129,6 +151,8 @@ if __name__ == "__main__":
             <dirname>mblondel.org</dirname>
             <host>www.mblondel.org</host>
             <basename>/mediawiki/index.php</basename>
+            <https />
+            <port>8080</port>
         </site>
     </sites>
 </wfs-config>
diff --git a/src/wikipediafs/http.py b/src/wikipediafs/http.py
new file mode 100644
index 0000000..a7124d8
--- /dev/null
+++ b/src/wikipediafs/http.py
@@ -0,0 +1,85 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+# WikipediaFS
+# Copyright (C) 2005 - 2007 Mathieu Blondel
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License along
+# with this program; if not, write to the Free Software Foundation, Inc.,
+# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+
+import os, socket, string, base64
+from httplib import HTTPConnection, HTTPSConnection
+
+class ExtendedHTTPConnection:
+    """
+    Transparent support for https, proxy, http auth
+    """
+    def __init__(self, host, port=None, https=False):
+        if https and not port:
+            port = 443
+        elif not port:
+            port = 80
+        else:
+            port = int(port)
+
+        if https:
+            self.conn = HTTPSConnection(host, port)
+        else:
+            self.conn = HTTPConnection(host, port)
+
+        self.headers = {}
+        self.data = None
+
+    def add_header(self, header, value):
+        self.headers[header] = value
+
+    def add_headers(self, headers):
+        for k, v in headers.items():
+            self.add_header(k, v)
+
+    def request(self, url):
+        if self.data:
+            method = "POST"
+        else:
+            method = "GET"
+
+        self.conn.request(method, url, self.data, self.headers)
+
+    def getresponse(self, *args):
+        return self.conn.getresponse(*args)
+
+    def close(self):
+        return self.conn.close()
+
+    def add_data(self, data):
+        self.data = data
+
+    def set_proxy(self):
+        """
+        Sets proxy if needed.
+        """
+        if os.environ.has_key("http_proxy"):
+            http_proxy = os.environ["http_proxy"]
+            http_proxy = http_proxy.replace("http://", "").rstrip("/")
+            (proxy_host, proxy_port) = http_proxy.split(":")
+            proxy_port = int(proxy_port)
+            proxy_sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
+            proxy_sock.connect((proxy_host, proxy_port))
+            self.sock = proxy_sock
+
+    def http_auth(self, username, password):
+        httpbasicauth = "%s:%s" % (username, password)
+        add_header("Authorization",
+                   "Basic %s" % base64.encodestring(httpbasicauth).strip())
+                   
\ No newline at end of file
diff --git a/src/wikipediafs/user.py b/src/wikipediafs/user.py
new file mode 100644
index 0000000..40b45f8
--- /dev/null
+++ b/src/wikipediafs/user.py
@@ -0,0 +1,112 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+# WikipediaFS
+# Copyright (C) 2005 - 2007 Mathieu Blondel
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License along
+# with this program; if not, write to the Free Software Foundation, Inc.,
+# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+
+import urllib, re
+from http import ExtendedHTTPConnection
+
+class User:
+    """
+    Gets user's cookie string
+    """
+
+    def __init__(self,
+                 username,
+                 password,
+                 host,
+                 basename,
+                 https=False,
+                 port=None,
+                 httpauth_username=None,
+                 httpauth_password=None
+                 ):
+
+        self.username = username
+        self.password = password
+        self.host = host
+        self.basename = basename
+        self.https = https
+        self.port = port
+        self.httpauth_username = httpauth_username
+        self.httpauth_password = httpauth_password
+
+        # url pattern
+        self.login_page = "%s?title=Special:Userlogin" % self.basename
+        self.login_page += "&action=submit&returnto=Special:Userlogin"
+
+    def getCookieString(self):
+        """
+        Gets user's cookie string.
+        It will then have to be passed to an Article.
+        """
+
+        params = {"wpName":self.username, "wpPassword":self.password,
+                  "wpLoginattempt":"Identification", "wpRemember":"1"}
+                
+        params = urllib.urlencode(params)
+        
+        headers = {"Content-type": "application/x-www-form-urlencoded",
+                   "User-agent" : "WikipediaFS"}
+        
+        conn = ExtendedHTTPConnection(self.host, self.port, self.https)
+        
+        conn.set_proxy() # sets proxy if needed
+
+        if self.httpauth_username and self.httpauth_password:
+            conn.http_auth(httpauth_username, httpauth_password)
+
+        conn.add_data(params)
+        conn.add_headers(headers)
+        conn.request(self.login_page)
+        response = conn.getresponse()
+        
+        cookie_list = []
+        in_cookie = re.compile(': (.*?);')
+        
+        for cookie_value in response.msg.getallmatchingheaders("set-cookie"):
+            it_matches = in_cookie.search(cookie_value)
+            
+            if it_matches:
+                cookie_list.append(it_matches.group(1))
+
+        conn.close()
+        
+        if len(cookie_list) == 4:
+            cookie_list.pop()
+            #logger.info("; ".join(cookie_list))
+            return "; ".join(cookie_list)
+        else:
+            return None
+
+               
+if __name__ == "__main__":
+    import sys
+
+    params = {
+        "host" : "www.mblondel.org",
+        "basename" : "/mediawiki/index.php",
+        "https" : True
+    }
+
+    if(len(sys.argv) != 3):
+        print "python user.py username password"
+    else:
+        user = User(sys.argv[1], sys.argv[2], **params)
+        print user.getCookieString()
+        
\ No newline at end of file