...to graphing window and signal viewer so complex systems with message
name duplication (eg "Status") works well. DBCHandler class has a new
function to find a msg by fully qualified node name...Utility class defines
the qualified name seperator, right now it is :: and is used thusly
filenameNoExtension::nodeName for display in the node dropdown (because
nodes can be common between dbc files). To get the message from the
system you use the findMessage function in dbchandler and pass the
msgName and the fully qualified node name. This is about as unique as we
can get, second only to message ID.
Added node comboBox to graph signal editor for easier data filtering
(took the same implementation that was in signal viewer, and tuned up
both of them)
frame ID for extended IDs was not having its upper bit set
BA_DEF_DEF_ "filterlabeling" is INT type but was getting bool written
(e.g. 'false' instead of 0)
did but undo buffer currently is deleted when signal window is closed
so undoing is limited currently.
Seemingly fixed issue where it would
constantly think you edited the open DBC file when you didn't.
Hi Collin,
I had a case where two ECUs were sending the same PGN (with different contents though), leading to SavvyCAN picking the first found PGN-Match.
Hence my proposal: go over the list until an exact match is found, otherwise returning the best match (same PGN).
my custom code. The issue that made me go down this path
is related to the little endian checkbox mysteriously and
inconsistently unchecking and therefore interpreting enum values
wrong...would really like to find this.
Masked error caused by saving dbc while editor is open that causes
msg IDs to go crazy by making editor modal, but that affects all
other windows, so might be a better fix.
Added new value type of VALUE_STRING, probably better named
ENUMERATED_STRING, and it uses the table in the editor to define
the enum.
Fixed a bug in the signal viewer load when a signal is not matched
(I found it by renaming a signal and then loading)
passing in the name and comment concatenated together. Made a new function to
find by name+comment since extracting only name from the text in the column would
require too much guessing.
Also switched from Sphinx generated HTML to GitHub markup help files.
This improves compatibility and will allow the help files to more easily be viewed on github as well.
This causes significant changes to the DBC interface. Things are
probably a bit broken right now. Committing to get it on record
before further large changes are done.
Collin Kidder