Samuel L. Bayer:
- same trick with "import wcnew; webchecker = wcnew" as above
- updated readhtml() method to handle pair representation; used
new name suppression infrastructure from wcnew.py to suppress
processing name anchors
[And untabified --GvR]
diff --git a/Tools/webchecker/websucker.py b/Tools/webchecker/websucker.py
index 67e493d..7453859 100755
--- a/Tools/webchecker/websucker.py
+++ b/Tools/webchecker/websucker.py
@@ -10,7 +10,9 @@
import urllib
import getopt
-import webchecker
+import wcnew
+
+webchecker = wcnew
# Extract real version number if necessary
if __version__[0] == '$':
@@ -45,14 +47,20 @@
class Sucker(webchecker.Checker):
checkext = 0
+ nonames = 1
- def readhtml(self, url):
+ # SAM 11/13/99: in general, URLs are now URL pairs.
+ # Since we've suppressed name anchor checking,
+ # we can ignore the second dimension.
+
+ def readhtml(self, url_pair):
+ url = url_pair[0]
text = None
path = self.savefilename(url)
try:
f = open(path, "rb")
except IOError:
- f = self.openpage(url)
+ f = self.openpage(url_pair)
if f:
info = f.info()
nurl = f.geturl()
@@ -89,7 +97,7 @@
host, port = urllib.splitnport(host)
host = string.lower(host)
if not path or path[-1] == "/":
- path = path + "index.html"
+ path = path + "index.html"
if os.sep != "/":
path = string.join(string.split(path, "/"), os.sep)
path = os.path.join(host, path)