diff options
| author | Michael Mann <mmann78@netscape.net> | 2015-09-23 20:56:50 -0400 |
|---|---|---|
| committer | Michael Mann <mmann78@netscape.net> | 2015-09-25 11:36:41 +0000 |
| commit | ea2fcb552c5462f9be1fea482350edbf7f6c1fc9 (patch) | |
| tree | 6417f26d0f889497a098a273e461d34fdc826c40 /CMakeListsCustom.txt.example | |
| parent | 40a43bd09be7645c13f284d6723b31fef5bddd12 (diff) | |
Refactor NCP Python data so that INFO column can be generated on the fly (TAKE 2)
The "old" method of populating the INFO column was to dissect all fields of a function/subfunction, then do a search in the tree to find the hf_ values of interest to then format into something for the INFO column. This is very expensive and requires "low level" APIs (for tree manipulation) which really shouldn't be used in a dissector.
The "new" method populates the INFO column at the same time a field is parsed, so nothing has to be revisited (and allows for more fields to be displayed on some malformed packets).
There are still expert infos (and possibly column APIs) under if (tree)s, but I'm not sure how FAKE_TREE_IS_VISIBLE factors into that. Removing the FAKE_TREE_IS_VISIBLE seems to negatively affect dissection.
Change-Id: Ie487e851c2f6558dd12f0c7010757b4a5f36226b
Reviewed-on: https://code.wireshark.org/review/10631
Reviewed-by: Michael Mann <mmann78@netscape.net>
Diffstat (limited to 'CMakeListsCustom.txt.example')
0 files changed, 0 insertions, 0 deletions