Issued a bug report on the SF.net tracker. Didn't realized that I wasn't logged in, and subsequently cannot modify the report. Oops.
Anyways, the problem is that dns.py defines the dnsheader class doesn't define the 'data' member. This is present in the Hg revision, which is why I was trying to get it working earlier.
Looks like merging won't be easy, since the newer version relies on is_tcp. Might just add a 'Data' field that sets data=None.
I am beginning to wonder why many of these variables are not defined in the Packet class body, since they are expected to exist. This would also facilitate contextual help systems (like the Python/Eclipse stuff).
Also not sure why is_tcp is used instead of isinstance(self,tcp). Looking back through the code is painful, because I haven't (to this point) found any high-level design documents. Might have to email GNN about this.
Read more...
Wednesday, May 13, 2009
Subscribe to:
Post Comments (Atom)
Followers
Blog Archive
-
▼
2009
(96)
-
▼
May
(26)
- Logic Validation
- IP and Port Representation
- Oh.
- Rejected Email
- Memorial Day
- Logging Framework Change
- Logging Framework
- Further thoughts on design
- Not Just OS X
- I Love X
- PCS-Current
- Wireshark Behavior
- Wireshark Strangeness
- Structure
- Adding functionality to PCS
- Real Coding Begins!
- Worthwhile Note
- More PCS Issues, Bugreport
- Attempted to update to PCS from Hg
- Finally!
- Back Home
- Moving Back Home
- Issues with PCS
- Playing around with PCS
- Installing PCS and Pyrex on a Mac
- Some Time to Work...
-
▼
May
(26)
No comments:
Post a Comment