python regex - how to get each group-CodePudding

I'm trying to get each group starting with "BO_" using python regex. (The data was from: https://github.com/commaai/opendbc)

Original Text:

...
BS_:

BU_: XXX CAMERA FRONT_RADAR ADRV APRK


BO_ 53 ACCELERATOR: 32 XXX
 SG_ CHECKSUM : 0|16@1  (1,0) [0|65535] "" XXX
 SG_ COUNTER : 16|8@1  (1,0) [0|255] "" XXX
 SG_ GEAR : 192|3@1  (1,0) [0|7] "" XXX
 SG_ ACCELERATOR_PEDAL : 40|8@1  (1,0) [0|255] "" XXX

BO_ 64 GEAR_ALT: 32 XXX
 SG_ CHECKSUM : 0|16@1  (1,0) [0|65535] "" XXX
 SG_ COUNTER : 16|8@1  (1,0) [0|255] "" XXX
 SG_ GEAR : 32|3@1  (1,0) [0|7] "" XXX

BO_ 69 GEAR: 24 XXX
 SG_ CHECKSUM : 0|16@1  (1,0) [0|65535] "" XXX
 SG_ COUNTER : 16|8@1  (1,0) [0|255] "" XXX
 SG_ GEAR : 44|3@1  (1,0) [0|7] "" XXX

...
CM_ SG_ 96 BRAKE_PRESSURE "User applied brake pedal pressure. Ramps from computer applied pressure on falling edge of cruise. Cruise cancels if !=0";
CM_ SG_ 101 BRAKE_POSITION "User applied brake pedal position, max is ~700. Signed on some vehicles";
CM_ SG_ 373 PROBABLY_EQUIP "aeb equip?";

I want to capture BO_ blocks whose BO id(BO_ ID) is in [53, 69] like this:

BO_ 53 ACCELERATOR: 32 XXX
 SG_ CHECKSUM : 0|16@1  (1,0) [0|65535] "" XXX
 SG_ COUNTER : 16|8@1  (1,0) [0|255] "" XXX
 SG_ GEAR : 192|3@1  (1,0) [0|7] "" XXX
 SG_ ACCELERATOR_PEDAL : 40|8@1  (1,0) [0|255] "" XXX

BO_ 69 GEAR: 24 XXX
 SG_ CHECKSUM : 0|16@1  (1,0) [0|65535] "" XXX
 SG_ COUNTER : 16|8@1  (1,0) [0|255] "" XXX
 SG_ GEAR : 44|3@1  (1,0) [0|7] "" XXX

Any suggestions? Thank you very much.

What I've tried so far was 1)capturing BO_ and relevant SGs using the regex below but it only captured each BO and the first SG groups.

BO_ (\w ) (\w ) *: (\w ) (\w )\n (SG_ (\w ) : (\d )\|(\d )@(\d )([\ |\-]) \(([0-9. \-eE] ),([0-9. \-eE] )\) \[([0-9. \-eE] )\|([0-9. \-eE] )\] \"(.*)\" (.*)\n)*

using greedy method but it captured all BOs at once except the last occurence.

BO_ (\w ) (\w ) *: (\w ) (\w )((.|\n)*)BO_

Also, for selecting only BOs including digits in the list [53, 69], I used raw f-string method something like rf"{digit}" in regex expressions.

CodePudding user response：

Inside of jumping through hoops in order to parse a dbc file with regular expressions, I suggest you use a proper parser like cantools:

CAN BUS tools in Python 3.

DBC, KCD, SYM, ARXML 3&4 and CDD file parsing.

CAN message encoding and decoding.

Simple and extended signal multiplexing.

Diagnostic DID encoding and decoding.

candump output decoder.

Node tester_.

C source code generator.

CAN bus monitor.

Graphical plots of signals.

CodePudding user response：

You can easily capture paragraphs with re.findall using the re.DOTALL (inline s) and re.MULTILINE (inline m) flags.

Regex (with inline flags): (?sm)BO_ (?:53|69) . ?^$

Usage (pick one):

re.findall(r"(?sm)BO_ (?:53|69) . ?^$", text)

re.findall(r"BO_ (?:53|69) . ?^$", text, flags=re.DOTALL | re.MULTILINE)

This lazily captures all lines from BO_ 53 or BO_ 69 to a blank line ^$ (demo).