X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fbkmk_rsimple.py;h=2c4df9e278f99b186daa869720f9437c81f7cefa;hb=00549b1c0622ee6ed0ac12249097cf4562bc486e;hp=47c2c7b425addbab96b0a4173c442ad61693debe;hpb=39bf25aeb57b0d1933f7729496f249e3cd67a816;p=bookmarks_db.git diff --git a/Robots/bkmk_rsimple.py b/Robots/bkmk_rsimple.py index 47c2c7b..2c4df9e 100644 --- a/Robots/bkmk_rsimple.py +++ b/Robots/bkmk_rsimple.py @@ -1,22 +1,34 @@ -""" - Simple, strightforward robot +"""Simple, strightforward robot + +This file is a part of Bookmarks database and Internet robot. - Written by Oleg BroytMann. Copyright (C) 2000-2007 PhiloSoft Design. """ +__author__ = "Oleg Broytman " +__copyright__ = "Copyright (C) 2000-2014 PhiloSoft Design" +__license__ = "GNU GPL" + +__all__ = ['robot_simple', 'get_error'] + -import sys, os, string +import sys, os import time, urllib from base64 import b64encode from urlparse import urljoin + from m_lib.net.www.util import parse_time from m_lib.md5wrapper import md5wrapper +from bkmk_objects import Robot +from parse_html import parse_filename + class RedirectException(Exception): reloc_dict = { 301: "perm.", - 302: "temp.", + 302: "temp2.", + 303: "temp3.", + 307: "temp7.", "html": "html" } def __init__(self, errcode, newurl): @@ -37,38 +49,45 @@ class MyURLopener(urllib.URLopener): # Error 301 -- also relocated (permanently) http_error_301 = http_error_302 + # Error 307 -- also relocated (temporary) + http_error_307 = http_error_302 # Error 401 -- authentication required def http_error_401(self, url, fp, errcode, errmsg, headers, data=None): raise IOError, ('http error', errcode, "Authentication required ", headers) + def http_error_default(self, url, fp, errcode, errmsg, headers): + if fp: + void = fp.read() + fp.close() + raise IOError, ('http error', errcode, errmsg, headers) + urllib._urlopener = MyURLopener() # Fake headers to pretend this is a real browser -_version = "Links (2.1; Linux 2.6 i686; 80x25)" +_version = "Mozilla/5.0 (X11; U; Linux 2.6 i686; en) Gecko/20001221 Firefox/2.0.0" urllib._urlopener.addheaders[0] = ('User-Agent', _version) _version = "bookmarks_db (Python %d.%d.%d; urllib/%s)" % ( sys.version_info[0], sys.version_info[1], sys.version_info[2], urllib.__version__) urllib._urlopener.addheader('X-User-Agent', _version) +urllib._urlopener.addheader('Referer', '') urllib._urlopener.addheader('Connection', 'close') -urllib._urlopener.addheader('Content-Length', '0') urllib._urlopener.addheader('Accept', '*/*') urllib._urlopener.addheader('Accept-Language', 'ru,en') urllib._urlopener.addheader('Cache-Control', 'max-age=300') -urllib._urlopener.addheader('Referer', 'http://www.yahoo.com/') def get_error(msg): - if type(msg) == type(""): + if isinstance(msg, str): return msg else: s = [] for i in msg: - s.append("'%s'" % string.join(string.split(str(i), "\n"), "\\n")) - return "(%s)" % string.join(s) + s.append("'%s'" % str(i).replace('\n', "\\n")) + return "(%s)" % ' '.join(s) urllib_ftpwrapper = urllib.ftpwrapper @@ -78,7 +97,7 @@ class myftpwrapper(urllib_ftpwrapper): def __init__(self, user, passwd, host, port, dirs): urllib_ftpwrapper.__init__(self, user, passwd, host, port, dirs) global ftpcache_key - ftpcache_key = (user, host, port, string.join(dirs, '/')) + ftpcache_key = (user, host, port, '/'.join(dirs)) urllib.ftpwrapper = myftpwrapper @@ -90,154 +109,189 @@ def get_welcome(): return _welcome -from bkmk_objects import Robot -from parse_html import parse_html +icons = {} # Icon cache; maps URL to a tuple (content type, data) + # or None if there is no icon. class robot_simple(Robot): def check_url(self, bookmark): - if not self.tempfname: - self.tempfname = bookmark.tempfname - + fname = None try: - try: - self.start = int(time.time()) - bookmark.icon = None + self.start = int(time.time()) + bookmark.icon = None - url_type, url_rest = urllib.splittype(bookmark.href) - url_host, url_path = urllib.splithost(url_rest) - url_path, url_tag = urllib.splittag(url_path) + url_type, url_rest = urllib.splittype(bookmark.href) + url_host, url_path = urllib.splithost(url_rest) + url_path, url_tag = urllib.splittag(url_path) - if bookmark.charset: urllib._urlopener.addheader('Accept-Charset', bookmark.charset) - fname, headers = urllib.urlretrieve("%s://%s%s" % (url_type, url_host, url_path), self.tempfname) - if bookmark.charset: del urllib._urlopener.addheaders[-1] + # Set fake referer to the root of the site + urllib._urlopener.addheaders[2] = ('Referer', "%s://%s%s" % (url_type, url_host, url_path)) - size = 0 - last_modified = None + if bookmark.charset: urllib._urlopener.addheader('Accept-Charset', bookmark.charset) + fname, headers = urllib.urlretrieve("%s://%s%s" % (url_type, url_host, url_path)) + if bookmark.charset: del urllib._urlopener.addheaders[-1] - if headers: - try: - size = headers["Content-Length"] - except KeyError: - pass + size = 0 + last_modified = None - try: - last_modified = headers["Last-Modified"] - except KeyError: - pass + if headers: + try: + size = headers["Content-Length"] + except KeyError: + pass - if last_modified: - last_modified = parse_time(last_modified) + try: + last_modified = headers["Last-Modified"] + except KeyError: + pass if last_modified: - last_modified = str(int(last_modified)) - else: - last_modified = bookmark.last_visit + last_modified = parse_time(last_modified) - bookmark.size = size - bookmark.last_modified = last_modified + if last_modified: + last_modified = str(int(last_modified)) + else: + last_modified = bookmark.last_visit - md5 = md5wrapper() - if urllib._urlopener.type == "ftp": # Pass welcome message through MD5 - md5.update(get_welcome()) + bookmark.size = size + bookmark.last_modified = last_modified + + md5 = md5wrapper() + if urllib._urlopener.type == "ftp": # Pass welcome message through MD5 + md5.update(get_welcome()) - md5.md5file(self.tempfname) - bookmark.md5 = str(md5) + md5.md5file(fname) + bookmark.md5 = str(md5) - if headers: + if headers: + try: + content_type = headers["Content-Type"] + self.log(" Content-Type: %s" % content_type) try: - content_type = headers["Content-Type"] - try: - content_type, charset = content_type.split(';') - content_type = content_type.strip() - charset = charset.split('=')[1].strip() - if self.log: self.log(" HTTP charset : %s" % charset) - except (ValueError, IndexError): - charset = None - if self.log: self.log(" no charset in Content-Type header") - if content_type == "text/html": - parser = parse_html(fname, charset, self.log) - bookmark.real_title = parser.title - if parser.refresh: - refresh = parser.refresh - try: - url = refresh.split('=', 1)[1] - except IndexError: - url = "self" - try: - timeout = float(refresh.split(';')[0]) - except (IndexError, ValueError): - raise RedirectException("html", "Bad redirect to %s (%s)" % (url, refresh)) - else: - try: - timeout = int(refresh.split(';')[0]) - except ValueError: - pass # float timeout - raise RedirectException("html", "%s (%s sec)" % (url, timeout)) - - # Get favicon.ico - icon = parser.icon - if not icon: - icon = "/favicon.ico" - icon = urljoin("%s://%s%s" % (url_type, url_host, url_path), icon) - if self.log: self.log(" looking for icon at: %s" % icon) + # extract charset from "text/html; foo; charset=UTF-8, bar; baz;" + content_type, charset = content_type.split(';', 1) + content_type = content_type.strip() + charset = charset.split('=')[1].strip().split(',')[0] + self.log(" HTTP charset : %s" % charset) + except (ValueError, IndexError): + charset = None + self.log(" no charset in Content-Type header") + for ctype in ("text/html", "application/xhtml+xml"): + if content_type.startswith(ctype): + html = True + break + else: + html = False + if html: + parser = parse_filename(fname, charset, self.log) + if parser: + bookmark.real_title = parser.title + icon = parser.icon + else: + icon = None + if not icon: + icon = "/favicon.ico" + icon = urljoin("%s://%s%s" % (url_type, url_host, url_path), icon) + self.log(" looking for icon at: %s" % icon) + if icon in icons: + if icons[icon]: + bookmark.icon_href = icon + content_type, bookmark.icon = icons[icon] + self.log(" cached icon: %s" % content_type) + else: + self.log(" cached icon: no icon") + else: try: + _icon = icon for i in range(8): try: - fname, headers = urllib.urlretrieve(icon) + icon_fname, headers = urllib.urlretrieve(_icon) except RedirectException, e: - icon = e.url - if self.log: self.log(" redirect to : %s" % icon) + _icon = e.url + self.log(" redirect to : %s" % _icon) else: break else: raise IOError("Too many redirects") except: etype, emsg, tb = sys.exc_info() - if self.log: self.log(" no icon : %s %s" % (etype, emsg)) - etype = None - emsg = None - tb = None + self.log(" no icon : %s %s" % (etype, emsg)) + etype = emsg = tb = None + icons[icon] = None else: content_type = headers["Content-Type"] - if content_type.startswith("image/"): - icon_file = open(fname, "rb") - icon = icon_file.read() + if content_type.startswith("application/") \ + or content_type.startswith("image/") \ + or content_type.startswith("text/plain"): + icon_file = open(icon_fname, "rb") + icon_data = icon_file.read() icon_file.close() - bookmark.icon = "data:%s;base64,%s" % (content_type, b64encode(icon)) - if self.log: self.log(" got icon : %s" % content_type) + bookmark.icon_href = icon + self.log(" got icon : %s" % content_type) + if content_type.startswith("application/") \ + or content_type.startswith("text/plain"): + self.log(" non-image content type, assume x-icon") + content_type = 'image/x-icon' + bookmark.icon = "data:%s;base64,%s" % (content_type, b64encode(icon_data)) + icons[icon] = (content_type, bookmark.icon) else: - if self.log: self.log(" no icon : bad content type '%s'" % content_type) - os.remove(fname) - - except KeyError: - pass - - except IOError, msg: - if (msg[0] == "http error") and (msg[1] == -1): - bookmark.no_error = "The server did not return any header - it is not an error, actually" - else: - bookmark.error = get_error(msg) - - except EOFError: - bookmark.error = "Unexpected EOF (FTP server closed connection)" - - except RedirectException, msg: - bookmark.moved = str(msg) - - except KeyboardInterrupt: - return 0 + self.log(" no icon : bad content type '%s'" % content_type) + icons[icon] = None + if parser and parser.refresh: + refresh = parser.refresh + try: + url = refresh.split('=', 1)[1] + except IndexError: + url = "self" + try: + timeout = float(refresh.split(';')[0]) + except (IndexError, ValueError): + raise RedirectException("html", "Bad redirect to %s (%s)" % (url, refresh)) + else: + try: + timeout = int(refresh.split(';')[0]) + except ValueError: + pass # float timeout + raise RedirectException("html", "%s (%s sec)" % (url, timeout)) + + except KeyError, key: + self.log(" no header: %s" % key) + + except IOError, msg: + if (msg[0] == "http error") and (msg[1] == -1): + bookmark.no_error = "The server did not return any header - it is not an error, actually" + self.log(' no headers: %s' % bookmark.no_error) + else: + bookmark.error = get_error(msg) + self.log(' Error: %s' % bookmark.error) + + except EOFError: + bookmark.error = "Unexpected EOF (FTP server closed connection)" + self.log(' EOF: %s' % bookmark.error) + + except RedirectException, msg: + bookmark.moved = str(msg) + self.log(' Moved: %s' % bookmark.moved) + + except KeyboardInterrupt: + self.log("Keyboard interrupt (^C)") + return 0 + + except: + import traceback + traceback.print_exc() + bookmark.error = "Exception!" + self.log(' Exception: %s' % bookmark.error) finally: - self.finish_check_url(bookmark) + self.finish_check_url(bookmark, fname) # Tested return 1 - - def finish_check_url(self, bookmark): + def finish_check_url(self, bookmark, fname=None): # Calculate these attributes even in case of an error - if os.path.exists(self.tempfname): - size = str(os.path.getsize(self.tempfname)) + if fname and os.path.exists(fname): + size = str(os.path.getsize(fname)) if size[-1] == 'L': size = size[:-1] bookmark.size = size @@ -247,3 +301,4 @@ class robot_simple(Robot): now = int(time.time()) bookmark.test_time = str(now - start) + urllib.urlcleanup()