X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fbkmk_rsimple.py;h=7de63185a6c8800e2d41a04d0ead795475531616;hb=3e5f3edd3d9fb7fcfa2d56ff3794716a4bf4d4cb;hp=afeeb254e2b519a33e145612e1b57de3116ec649;hpb=0e76f1851882b99da63a7c8a9e4cdf0c4a48657f;p=bookmarks_db.git diff --git a/Robots/bkmk_rsimple.py b/Robots/bkmk_rsimple.py index afeeb25..7de6318 100644 --- a/Robots/bkmk_rsimple.py +++ b/Robots/bkmk_rsimple.py @@ -1,7 +1,7 @@ """ Simple, strightforward robot - Written by Oleg Broytman. Copyright (C) 2000-2008 PhiloSoft Design. + Written by Oleg Broytman. Copyright (C) 2000-2011 PhiloSoft Design. """ import sys, os @@ -105,9 +105,7 @@ icons = {} # Icon cache; maps URL to a tuple (content type, data) class robot_simple(Robot): def check_url(self, bookmark): - if not self.tempfname: - self.tempfname = bookmark.tempfname - + fname = None try: self.start = int(time.time()) bookmark.icon = None @@ -120,7 +118,7 @@ class robot_simple(Robot): 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) + fname, headers = urllib.urlretrieve("%s://%s%s" % (url_type, url_host, url_path)) if bookmark.charset: del urllib._urlopener.addheaders[-1] size = 0 @@ -152,12 +150,13 @@ class robot_simple(Robot): if urllib._urlopener.type == "ftp": # Pass welcome message through MD5 md5.update(get_welcome()) - md5.md5file(self.tempfname) + md5.md5file(fname) bookmark.md5 = str(md5) if headers: try: content_type = headers["Content-Type"] + self.log(" Content-Type: %s" % content_type) try: # extract charset from "text/html; foo; charset=UTF-8, bar; baz;" content_type, charset = content_type.split(';', 1) @@ -167,34 +166,26 @@ class robot_simple(Robot): except (ValueError, IndexError): charset = None self.log(" no charset in Content-Type header") - if content_type in ("text/html", "application/xhtml+xml"): + for ctype in ("text/html", "application/xhtml+xml"): + if content_type.startswith(ctype): + html = True + break + else: + html = False + if 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 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: @@ -204,7 +195,7 @@ class robot_simple(Robot): _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 self.log(" redirect to : %s" % _icon) @@ -215,58 +206,84 @@ class robot_simple(Robot): except: etype, emsg, tb = sys.exc_info() self.log(" no icon : %s %s" % (etype, emsg)) - etype = None - emsg = None - tb = None + 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)) + 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: self.log(" no icon : bad content type '%s'" % content_type) icons[icon] = None - os.remove(fname) + 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: - pass + 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 @@ -276,3 +293,4 @@ class robot_simple(Robot): now = int(time.time()) bookmark.test_time = str(now - start) + urllib.urlcleanup()