This is an automated email from the ASF dual-hosted git repository. xiaoxiang781216 pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/nuttx.git
commit f62d152b3552ef6d74b1da4b029379a6d54620a1 Author: xuxingliang <[email protected]> AuthorDate: Fri Aug 22 15:55:49 2025 +0800 coredump.py: automatically process logfile Try to extract coredump log and decode it depending on marker "base64 formatted". Try to decompress the binary data if LZF compressed. Now for any coredump output from device, simply use below command to convert: nuttx/tools/coredump.py crash.log nuttx/tools/coredump.py coredump.core.lzf Signed-off-by: yangao1 <[email protected]> Signed-off-by: xuxingliang <[email protected]> --- tools/coredump.py | 164 +++++++++++++++++++++++------------------------------- 1 file changed, 69 insertions(+), 95 deletions(-) diff --git a/tools/coredump.py b/tools/coredump.py index b7929bed113..e858b054444 100755 --- a/tools/coredump.py +++ b/tools/coredump.py @@ -24,70 +24,82 @@ import argparse import base64 import binascii +import io import mmap import os -import shutil import struct import sys -import lzf +try: + import lzf +except ImportError: + print("lzf module not found, execute `pip install python-lzf`") + + +def try_extract(filepath) -> io.BytesIO: + """ + Parse log file lines, and try to extract coredump data text + Return extracted coredump binary data or None if not found + + Note: if multiple core dump found in file, only the first one is extracted + """ + in_coredump = False + bas64 = False + coredump = [] + with open(filepath, "r") as f: + try: + for line in f.readlines(): + if "Start coredump" in line: + in_coredump = True + if coredump: + print("Warning: multiple core dump found.") + break + elif "Finish coredump" in line: + bas64 = "base64 formatted" in line + in_coredump = False + elif in_coredump: + index = line.rfind(" ") + if index > 0: + line = line[index + 1 :] + line = line.rstrip("\n\r") + if line: + coredump += line + except (UnicodeDecodeError, UnicodeError): + pass + try: + return coredump and io.BytesIO( + base64.b64decode("".join(coredump)) + if bas64 + else binascii.unhexlify("".join(coredump)) + ) + except Exception: + return None -def decompress(lzffile, outfile): - chunk_number = 1 +def try_decompress(input: io.BytesIO) -> bytearray: + output = bytearray() while True: - prefix = lzffile.read(2) - - if len(prefix) == 0: + prefix = input.read(2) # 0, 1 + typ = input.read(1) # 2 + chunk_len = input.read(2) # 3, 4 + if prefix != b"ZV": break - elif prefix != b"ZV": - break - - typ = lzffile.read(1) - clen = struct.unpack(">H", lzffile.read(2))[0] + chunk_len = struct.unpack(">H", chunk_len)[0] if typ == b"\x00": - chunk = lzffile.read(clen) + chunk = input.read(chunk_len) + output.extend(chunk) elif typ == b"\x01": - uncompressed_len = struct.unpack(">H", lzffile.read(2))[0] - cdata = lzffile.read(clen) + uncompressed_len = struct.unpack(">H", input.read(2))[0] + cdata = input.read(chunk_len) chunk = lzf.decompress(cdata, uncompressed_len) + output.extend(chunk) else: - return - - outfile.write(chunk) - chunk_number += 1 - - -def unhexlify(infile, outfile): - while True: - line = infile.readline() - if not line: - break - line = line.replace(b"\n", b"").strip() - if line == b"": - continue - index = line.rfind(b" ") - if index > 0: - line = line[index + 1 :] - - outfile.write(binascii.unhexlify(line)) - - -def unbase64file(infile, outfile): - while True: - line = infile.readline() - if not line: break - line = line.replace(b"\n", b"").strip() - if line == b"": - continue - index = line.rfind(b" ") - if index > 0: - line = line[index + 1 :] - outfile.write(base64.b64decode(line)) + input.seek(0) + return output or input.read() def mmap_file(file_path, size=None): @@ -107,24 +119,12 @@ def parse_args(): ) parser.add_argument("input") parser.add_argument("-o", "--output", help="Output file in hex.") - parser.add_argument( - "--base64", - action="store_true", - default=False, - help="Set when input file is base64 encoded.", - ) parser.add_argument( "-b", "--binary", action="store_true", default=False, - help="Treat input file as binary data and write directly to output.", - ) - parser.add_argument( - "--size", - type=int, - default=None, - help="Size of memory to dump (default: mmap whole file).", + help="Treat input file as binary data.", ) args = parser.parse_args() @@ -136,45 +136,19 @@ def main(): print(f"Error: Input file {args.input} does not exist.") sys.exit(1) - tmp = os.path.splitext(args.input)[0] + ".tmp" - - if args.output is None: - args.output = os.path.splitext(args.input)[0] + ".core" + output = args.output or f"{args.input}.core" - try: - infile = mmap_file(args.input, args.size) - except Exception as e: - print(f"Failed to mmap input file: {e}") + extracted = try_extract(args.input) + if extracted is None: + print(f"Error: none hex or base64 content in {args.input}.") sys.exit(1) - tmpfile = open(tmp, "wb+") - - if args.binary: - tmpfile.write(infile) - elif args.base64: - unbase64file(infile, tmpfile) - else: - unhexlify(infile, tmpfile) - - infile.close() - - tmpfile.seek(0, 0) - - lzfhdr = tmpfile.read(2) - - if lzfhdr == b"ZV": - outfile = open(args.output, "wb") - tmpfile.seek(0, 0) - decompress(tmpfile, outfile) - tmpfile.close() - outfile.close() - os.unlink(tmp) - else: - tmpfile.close() - shutil.copy(tmp, args.output) - os.unlink(tmp) + input = mmap_file(args.input) if args.binary or not extracted else extracted + coredump = try_decompress(input) + with open(output, "wb") as outfile: + outfile.write(coredump) - print("Core file conversion completed: " + args.output) + print("Core file conversion completed: " + output) if __name__ == "__main__":
