```

===can_node===

• START: node.cpp
  • Create a CanDriver.cpp instance
• CanDriver.cpp
  • TO CHANGE: Add Sync Time support
  • Create a CanUsb.cpp instance
    • This creates a reference to a data stream which provides a callback whenever USB data is read from the CAN device.
  • Set the CanUsb callback to CanDriver.cpp::recvDevice
  • Get bitrate and error_topic information
    • Each channel is fetched individually and placed in a vector. bitrate[0] is modified by the launch parameter “bitrate_1”, like so: * Same for filter_masks and matches.
  • Open the CAN driver (run CanUsb.cpp::open)
  • Read version number and channel count
  • For each channel, print the filter masks, and filter matches to the console
  • For each channel, attempt to set the bitrates, then print to console
  • If all channel bitrates are set successfully, set up ROS pub/subs: * For each channel with nonzero bitrate, create a node (can_bus_#) * Create a subscriber (can_tx) to receive messages to send to the device * Create a publisher (can_rx) to send messages whenever the device sends data * Create an error publisher (can_err) if argument is set
  • Whenever data is received from the device, publish a can_msgs::Frame to the appropriate channel * can_msgs is a standard message type: http://docs.ros.org/kinetic/api/can_msgs/html/msg/Frame.html Contains a unique ID, data length code (dlc), and data (up to 8 bytes)

===dbc_node===

• START: dbc_node.cpp
  • Create a Node and NodeHandle
  • Get dbc_file parameter; if it doesn’t exist, exit.
  • Create a CanExtractor.cpp object, passing the dbc_file value
  • Create a subscriber to can_rx (which is published by can_node)
    • When a can_msgs::Frame message is received, get the ID (OR it with 0x80000000 if it’s an extended msg) and put it in a RosCanMsgStruct
    • Run CanExtractor::getMessage(RosCanMsgStruct), which returns true/false
      • If true, run CanExtractor::initPublishers(RosCanMsgStruct, NodeHandle)
    • Run CanExtractor::pubMessage(can_msgs::Frame), which posts the data to the appropriate ROS topic
• CanExtractor.cpp
  • dbc_
    • A DBCIterator.cpp object
  • msgs_
    • A map between CAN message IDs and RosCanMsgStructs
  • getMessage(RosCanMsgStruct)
    • Check if the RosCanMsgStruct is already in the msgs_ map.
    • If it is, set the RosCanMsgStruct reference to the stored one, and return false (the message type has already been processed).
    • If it isn’t, iterate over dbc_, searching for the CAN ID.
      • For each matching DBCMessage::Message found, create a RosCanSigStruct based on it and append it to the RosCanMsgStruct
      • Then store the RosCanMsgStruct and return it
    • Either way, the result is a RosCanMsgStruct, containing a CAN ID, and message name.
    • RosCanMsgStruct contains a can_msgs::Frame publisher which sends whenever a message is received, and a vector of RosCanSigStructs, each containing a publisher which sends individual fields whenever a message is received.
  • initPublishers(RosCanMsgStruct, NodeHandle)
    • Creates a publisher for the message as a whole (can_msgs::Frame) and attaches it to RosCanMsgStruct
    • Creates publishers for each signal with registerCanSignalPublisher(RosCanSigStruct NodeHandle)
  • registerCanSignalPublisher(RosCanSigStruct NodeHandle)
    • Creates an std_msgs::Int/Float64/Bool publisher (signed/unsigned/size based on RosCanSigStruct)
    • Attaches the publisher to the RosCanSigStruct
  • pubMessage()
    • Retrieve the RosCanMsgStruct from the msgs_ map based on message ID.
      • If the messageID isn’t in the msgs_ map, return.
    • An 8-byte integer (uint64) stores all message data.
    • TO CHANGE: Loop through the list of signals for a signal multiplexor enum value that is MULTIPLEXOR. Save this value and break if found, set to -1 otherwise.
    • For each signal, the data is retrieved based on the start bit and length (signed or unsigned).
    • TO CHANGE: For each signal, if the multiplexor is not -1, use an if conditional to check if the signal’s multiplexor enum is MULTIPLEXED; if it is, check if the multiplexNum is equal to the multiplexor value from the loop above; only publish if the values match.
• DBCIterator.cpp
  • messageList: Vector of DBCMessage.Message objects
  • init
    • TO CHANGE: Support multiple file streams
    • Open the file stream
    • DBCMessage::Message overrides the » operator, and uses this to take a stream (up to the next )
    • If the process fails and the stream errors out, clear it (which will cause the loop to exit)
    • Else, add the message to the messageList.
    • DBCIterator iterates the file’s messages, DBCMessage iterates the message’s signals
• DBCMessage.cpp
  • If the line does not start in BO_, throw an error to get the parent iterator to clear to the next line.
  • If it does, fetch the BO data: BO_
  • Then, iterate over the next lines;
  • While the stream is creating valid signal objects, » create more DbcSignal.cpp objects.
• DBCSignal.cpp
  • Get the next line, if (after cleanup) the line is empty or does not start with SG_, throw an error to get the parent iterator to finish parsing signals.
  • If it does, fetch the signal data: SG_ : |@ (,) [|] ""
  • Multiplexor may be blank (meaning value always appears in the message), in which case there will be no value between the Fieldname and the :
  • If Multiplexor is not blank, it is either:
    • ‘M’, meaning the value is the multiplexor
    • ‘m###’, with numbers, meaning the signal only appears in the message when the multiplexor value is equal to that of the multiplexed signal
  • Byteorder is either Motorola (big-endian/network) (0) or Intel (little-endian/math) (1)
  • Signed is either + (unsigned value) or - (signed value)
  • phys = digits * factor + offset

===DBC EXAMPLE=== BO_ 126 License: 8 EPAS SG_ EXPIRED : 10|1@1+ (1,0) [0|0] “” MAB SG_ MUX M : 0|8@1+ (1,0) [0|0] “” MAB SG_ FEAT_BASE_TRIALS_USED m0 : 32|16@1+ (1,0) [0|0] “” MAB SG_ DATE1 m129 : 24|8@1+ (1,0) [0|0] “” MAB

File truncated at 100 lines see the full file

Changelog for package dataspeed_can_tools

2.0.6 (2025-03-12)

• Add DBC bag --no-copy option to leave out original CAN message when expanding into signals
• Contributors: Kevin Hallenbeck

2.0.5 (2024-10-08)

• Add CAN-FD messages with fixed size data arrays
• Update DBC tools for new fixed sizes
• Contributors: Kevin Hallenbeck

2.0.4 (2024-07-25)

• Added parameter declarations to DBC node
• Handle rosbag2_storage changes in Jazzy
• Contributors: Gabe, Kevin Hallenbeck

2.0.3 (2023-07-11)

• Migrated dataspeed_can_tools over to ROS2
• Contributors: Gabriel Oetjens, Kevin Hallenbeck

1.1.0 (2023-02-03)

• Add CAN-FD message support
• Contributors: Nicholas Jankowski

1.0.16 (2020-07-07)

• Increase CMake minimum version to 3.0.2 to avoid warning about CMP0048 http://wiki.ros.org/noetic/Migration#Increase_required_CMake_version_to_avoid_author_warning
• Fix processing bags with DBC files for extended CAN IDs
• Contributors: Kevin Hallenbeck

1.0.15 (2020-01-24)

• Use %zu for portable printf() size type
• Contributors: Kevin Hallenbeck

1.0.14 (2020-01-16)

1.0.13 (2019-01-08)

• Added advance compiler error checking and fixed newly exposed errors
• Contributors: Kevin Hallenbeck, Lincoln Lorenz

1.0.12 (2018-12-19)

1.0.11 (2018-09-06)

• Added support for multiple DBC files
• Added support for multiplexed signals
• Many changes to the command line arguments
• Fixed bug with signals greater than 31 bits
• Set CXX_STANDARD to C++11 when necessary
• Contributors: Kevin Hallenbeck, Eric Myllyoja

1.0.10 (2018-06-13)

1.0.9 (2018-06-12)

• Preliminary support for extended IDs
• Contributors: Kevin Hallenbeck

1.0.8 (2018-04-09)

• Added timestamp
• Added option to expand dbc signals or not
• Handle messages not defined in the dbc file
• Contributors: Kevin Hallenbeck

1.0.7 (2017-10-19)

1.0.6 (2017-09-07)

• Migrated from dataspeed_can_msgs to can_msgs (this will break dependent packages)
• Contributors: Kevin Hallenbeck

1.0.5 (2017-08-21)

• Updated nodelets to the PLUGINLIB_EXPORT_CLASS macro
• Contributors: Kevin Hallenbeck

1.0.4 (2017-05-08)

1.0.3 (2017-04-25)

• Updated package.xml format to version 2
• Contributors: Kevin Hallenbeck

1.0.2 (2017-01-24)

1.0.1 (2016-11-17)

1.0.0 (2016-09-28)

File truncated at 100 lines see the full file