Wireshark mailing list archives
Re: r52604: Adding to col info is probably useless as next dissector will overwrite it.
From: didier <dgautheron () magic fr>
Date: Fri, 18 Oct 2013 05:40:27 +0200
Hi Le lundi 14 octobre 2013 à 21:26 +0200, Jakub Zawadzki a écrit :
Hi Anders, Please revert it, dissectors after vlan can be disabled.
If we want to have some lazy COL_INFO calculation - that's fine for me, but selectively disabling col_add_fstr is NACK for me.
Yes vlan can either:
- throw an exception, can be guarded by replacing
col_add_fstr(pinfo->cinfo, COL_INFO,... with if tvb_length() < 4 col_add_fstr - call ethertype or dissect_802_3 which are void but could return TRUE/FALSE if colinfo has been updated (assuming that a sub dissector has cleared colinfo), if not then call col_add_fstr() like ethertype do. There's a corner case: if a sub dissector throw an exception before clearing colinfo but it should be rare. Have to change dissect_802_03 a little though, it must use dissector_only() and call proto_data itself rather than calling dissector() Comments? Didier ___________________________________________________________________________ Sent via: Wireshark-dev mailing list <wireshark-dev () wireshark org> Archives: http://www.wireshark.org/lists/wireshark-dev Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev mailto:wireshark-dev-request () wireshark org?subject=unsubscribe
Current thread:
- r52604: Adding to col info is probably useless as next dissector will overwrite it. Jakub Zawadzki (Oct 14)
- Re: r52604: Adding to col info is probably useless as next dissector will overwrite it. Anders Broman (Oct 14)
- Re: r52604: Adding to col info is probably useless as next dissector will overwrite it. didier (Oct 17)
