command: syncisos: Update HTML parser to work with repoindex

First of all: the regex was broken.  The existing check would have been
better written as:

-        self.url_re = re.compile('(?!\.{2})/$')
...
-                    if value != '../' and self.url_re.search(value) is not None:
+                    if value != '../' and value.endswith('/'):

Anyway, reduce the nesting a bit, and revise the check to filter out the
extra links that repoindex includes.

1 files changed, 9 insertions, 6 deletions

diff --git a/releng/management/commands/syncisos.py b/releng/management/commands/syncisos.py
index f182cc33..060ab2c1 100644
--- a/releng/management/commands/syncisos.py
+++ b/releng/management/commands/syncisos.py
@@ -14,14 +14,17 @@ class IsoListParser(HTMLParser):
         HTMLParser.__init__(self)
 
         self.hyperlinks = []
-        self.url_re = re.compile('(?!\.{2})/$')
 
     def handle_starttag(self, tag, attrs):
-        if tag == 'a':
-            for name, value in attrs:
-                if name == "href":
-                    if value != '../' and self.url_re.search(value) is not None:
-                        self.hyperlinks.append(value[:-1])
+        if tag != 'a':
+            return
+
+        for name, value in attrs:
+            if name != "href":
+                continue
+
+            if value.endswith('/') and value != '../' and '/' not in value[:-1] and len(value[:-1]) > 0:
+                self.hyperlinks.append(value[:-1])
 
     def parse(self, url):
         try: