X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fbkmk_rsimple.py;h=871082fe4a0113eddab4435582443fc29db0d068;hb=c9cfb54ef36e77aee7187e4533f063d407073ad8;hp=e01c7fece23cd612362ef862af0d486a580bfcdc;hpb=a62cd94997a22857d27764042a120f21a226e94b;p=bookmarks_db.git diff --git a/Robots/bkmk_rsimple.py b/Robots/bkmk_rsimple.py index e01c7fe..871082f 100644 --- a/Robots/bkmk_rsimple.py +++ b/Robots/bkmk_rsimple.py @@ -1,9 +1,18 @@ -""" - Simple, strightforward robot +"""Simple, strightforward robot - Written by Oleg Broytman. Copyright (C) 2000-2010 PhiloSoft Design. +This file is a part of Bookmarks database and Internet robot. """ +__version__ = "$Revision$"[11:-2] +__revision__ = "$Id$"[5:-2] +__date__ = "$Date$"[7:-2] +__author__ = "Oleg Broytman " +__copyright__ = "Copyright (C) 2000-2011 PhiloSoft Design" +__license__ = "GNU GPL" + +__all__ = ['robot_simple', 'get_error'] + + import sys, os import time, urllib from base64 import b64encode @@ -105,9 +114,7 @@ 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 +127,7 @@ 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 @@ -152,7 +159,7 @@ 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: @@ -178,24 +185,6 @@ class robot_simple(Robot): parser = parse_html(fname, charset, self.log) if parser: 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 else: icon = None @@ -215,7 +204,7 @@ 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) @@ -226,24 +215,43 @@ 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") + 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_href = icon - bookmark.icon = "data:%s;base64,%s" % (content_type, b64encode(icon_data)) 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, key: self.log(" no header: %s" % key) @@ -275,16 +283,15 @@ class robot_simple(Robot): 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 @@ -294,3 +301,4 @@ class robot_simple(Robot): now = int(time.time()) bookmark.test_time = str(now - start) + urllib.urlcleanup()