X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fbkmk_rsimple.py;h=f481d66e1bce396d46387571211d010eeea40a52;hb=38f3645cce7a5875128d788df6631069c761b987;hp=ea2decec8b82ee78ebb6d3dedc2cd5ee5b5dd435;hpb=f1e0383ad67eae21ccf32ff0c1c3775a6d65b64a;p=bookmarks_db.git diff --git a/Robots/bkmk_rsimple.py b/Robots/bkmk_rsimple.py index ea2dece..f481d66 100644 --- a/Robots/bkmk_rsimple.py +++ b/Robots/bkmk_rsimple.py @@ -1,7 +1,7 @@ """ Simple, strightforward robot - Written by Oleg BroytMann. Copyright (C) 2000-2007 PhiloSoft Design. + Written by Oleg BroytMann. Copyright (C) 2000-2008 PhiloSoft Design. """ import sys, os @@ -45,6 +45,12 @@ class MyURLopener(urllib.URLopener): 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() @@ -101,147 +107,152 @@ class robot_simple(Robot): self.tempfname = bookmark.tempfname 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) - # Set fake referer to the root of the site - urllib._urlopener.addheaders[2] = ('Referer', "%s://%s%s" % (url_type, url_host, url_path)) + # Set fake referer to the root of the site + urllib._urlopener.addheaders[2] = ('Referer', "%s://%s%s" % (url_type, url_host, 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] + 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] - size = 0 - last_modified = None + size = 0 + last_modified = None - if headers: - try: - size = headers["Content-Length"] - except KeyError: - pass + if headers: + try: + size = headers["Content-Length"] + except KeyError: + pass - try: - last_modified = headers["Last-Modified"] - 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.md5file(self.tempfname) - bookmark.md5 = str(md5) + md5 = md5wrapper() + if urllib._urlopener.type == "ftp": # Pass welcome message through MD5 + md5.update(get_welcome()) - if headers: + md5.md5file(self.tempfname) + bookmark.md5 = str(md5) + + if headers: + try: + content_type = headers["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() - self.log(" HTTP charset : %s" % charset) - except (ValueError, IndexError): - charset = None - 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" + # 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") + if content_type in ("text/html", "application/xhtml+xml"): + 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 = float(refresh.split(';')[0]) - except (IndexError, ValueError): - raise RedirectException("html", "Bad redirect to %s (%s)" % (url, refresh)) - else: + 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) + self.log(" looking for icon at: %s" % icon) + if icon in icons: + if icons[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: - 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) - self.log(" looking for icon at: %s" % icon) - if icon in icons: - if icons[icon]: - content_type, bookmark.icon = icons[icon] - self.log(" cached icon: %s" % content_type) + fname, headers = urllib.urlretrieve(_icon) + except RedirectException, e: + _icon = e.url + self.log(" redirect to : %s" % _icon) + else: + break else: - self.log(" cached icon: no icon") + raise IOError("Too many redirects") + except: + etype, emsg, tb = sys.exc_info() + self.log(" no icon : %s %s" % (etype, emsg)) + etype = None + emsg = None + tb = None + icons[icon] = None else: - try: - _icon = icon - for i in range(8): - try: - fname, headers = urllib.urlretrieve(_icon) - except RedirectException, e: - _icon = e.url - self.log(" redirect to : %s" % _icon) - else: - break - else: - raise IOError("Too many redirects") - except: - etype, emsg, tb = sys.exc_info() - self.log(" no icon : %s %s" % (etype, emsg)) - etype = None - emsg = None - tb = None - icons[icon] = None + content_type = headers["Content-Type"] + if content_type.startswith("image/"): + icon_file = open(fname, "rb") + icon = icon_file.read() + icon_file.close() + bookmark.icon = "data:%s;base64,%s" % (content_type, b64encode(icon)) + self.log(" got icon : %s" % content_type) + icons[icon] = (content_type, bookmark.icon) else: - content_type = headers["Content-Type"] - if content_type.startswith("image/"): - icon_file = open(fname, "rb") - icon = icon_file.read() - icon_file.close() - bookmark.icon = "data:%s;base64,%s" % (content_type, b64encode(icon)) - self.log(" got icon : %s" % content_type) - icons[icon] = (content_type, bookmark.icon) - else: - self.log(" no icon : bad content type '%s'" % content_type) - icons[icon] = None - os.remove(fname) + self.log(" no icon : bad content type '%s'" % content_type) + icons[icon] = None + os.remove(fname) + + except KeyError: + pass - 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 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 EOFError: - bookmark.error = "Unexpected EOF (FTP server closed connection)" + except RedirectException, msg: + bookmark.moved = str(msg) - except RedirectException, msg: - bookmark.moved = str(msg) + except KeyboardInterrupt: + return 0 - except KeyboardInterrupt: - return 0 + except: + import traceback + traceback.print_exc() + bookmark.error = "Exception!" finally: self.finish_check_url(bookmark)