- Add GMLAN 29bit filtering feature. In addition to J1939 matching, it is now possible to match DBC messages using only the Arbitration Id (bits 14-26 of CAN Address), since Priority id (higher bits) and Sender Id (lowest 13 bits) are not relevant when interpreting the CAN message content. This allows to use the excellent GM Global A DBC files on https://github.com/commaai/opendbc. Just remember to select the "GMLAN" matching criteria on DBC File manager.
- Add "Filter labeling" feature. Using loaded DBC files, it matches the filter Ids with DBC messages and shows them in filter windows. Labeling works in the main window (no need to set "Frame interpret" on, since this is a separate feature), as well as in Flow view, Frame data analysis, Playback and Fuzzzing windows. This makes it so much easier to concentrate on those CAN messages that matter. To use this feature, select global "Label filters using messages from DBC files" on Main Configuration page, and then enable the feature on individual DBC files (on DBC file manager) as needed. This is needed if there's overlap in CAN addresses between DBC files, and because the filter labeler cannot tell apart anymore which Filter ID is associated to which bus, it might label them incorrectly. But usually it's ok to just enable the labeling with all the DBC files.
- In Frame data analysis window, in addition to J1939 decoding show also GMLAN (Priority bits, Arbitration Id, Sender Id) decoding.
- FilterUtility static class used for centralized ListWidgetItem creation and for other helper functions
- DBC handler: To allow setting bus numbers even before connection is configured, do not enforce "valid" bus numbers
All frame file I/O uses the same directory load/save location but DBC files store separately as do filter files, graphing definitions, and stored text files from the various reverse engineering windows.
* improves window handling a lot
(on Linux GNOME 3 at least: quickly resizing windows left/right, not forced in foreground anymore etc.)
setWindowFlags() inserted in all *window.cpp files except mainwindow
* some whitespace at EOL trimmed
Also added UDS decoding to the ISOTP window. An attempt is made to see
if a frame looks like UDS and then it is decoded.
Also, some code clean up to remove some compiler warnings. There's a
LONG way to go still.
In FlowView:
- Drastically improved functionality of live mode. It actually works now
- Fixed several crash bugs
- Program didn't used to pick up new frames properly. This is now corrected.
In Graphing window:
- Fixed dynamic updates to work properly even under heavy bus loads
In frame playback:
- Fixed a bug where single stepping through frames didn't send them over canbus
This further abstracts the interface. Things are reasonably stable now for frame reception. Initial work has been done to support frame transmission as well. This is currently less stable but mostly working.
refactoring of some DBC code, fixed some DBC calcs, fixed a bug in the
frame playback window, incremented version number to show that a lot
has been done.
Also, 0 out data bytes that aren't used for the native format. The native format always shows 8 data bytes even if the frame didn't have that many. That way programs like Excel will still have data in all of the positions.
This way it'll send what was available when it started. You can clear the original data if you want, keep capturing, anything you want and it'll still be able to keep going with its own copy.
Collin Kidder