X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fbkmk_rsimple.py;h=2c4df9e278f99b186daa869720f9437c81f7cefa;hb=00549b1c0622ee6ed0ac12249097cf4562bc486e;hp=67d1fbe66239c8cc9fcf312d1fec3eda1f187b84;hpb=a38eb2c9d04e866177c80cb961712ed3d4574f5e;p=bookmarks_db.git diff --git a/Robots/bkmk_rsimple.py b/Robots/bkmk_rsimple.py index 67d1fbe..2c4df9e 100644 --- a/Robots/bkmk_rsimple.py +++ b/Robots/bkmk_rsimple.py @@ -1,9 +1,16 @@ -""" - Simple, strightforward robot +"""Simple, strightforward robot + +This file is a part of Bookmarks database and Internet robot. - Written by Oleg Broytman. Copyright (C) 2000-2010 PhiloSoft Design. """ +__author__ = "Oleg Broytman " +__copyright__ = "Copyright (C) 2000-2014 PhiloSoft Design" +__license__ = "GNU GPL" + +__all__ = ['robot_simple', 'get_error'] + + import sys, os import time, urllib from base64 import b64encode @@ -13,13 +20,15 @@ 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_html +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): @@ -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: @@ -175,27 +182,9 @@ class robot_simple(Robot): else: html = False if html: - parser = parse_html(fname, charset, self.log) + parser = parse_filename(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 @@ -207,18 +196,18 @@ class robot_simple(Robot): if icons[icon]: bookmark.icon_href = icon content_type, bookmark.icon = icons[icon] - self.log(" cached icon: %s" % content_type) + self.log(" cached icon: %s" % content_type) else: - self.log(" cached icon: no icon") + 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 - self.log(" redirect to : %s" % _icon) + self.log(" redirect to : %s" % _icon) else: break else: @@ -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) @@ -251,18 +259,18 @@ class robot_simple(Robot): 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(bookmark.no_error) + self.log(' no headers: %s' % bookmark.no_error) else: bookmark.error = get_error(msg) - self.log(bookmark.error) + self.log(' Error: %s' % bookmark.error) except EOFError: bookmark.error = "Unexpected EOF (FTP server closed connection)" - self.log(bookmark.error) + self.log(' EOF: %s' % bookmark.error) except RedirectException, msg: bookmark.moved = str(msg) - self.log(bookmark.moved) + self.log(' Moved: %s' % bookmark.moved) except KeyboardInterrupt: self.log("Keyboard interrupt (^C)") @@ -272,19 +280,18 @@ class robot_simple(Robot): import traceback traceback.print_exc() bookmark.error = "Exception!" - self.log(bookmark.error) + 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()