Lattyware
commented
6 years ago My first guess would be to try a quick change to line 142:
return [(offset ^ key, dlen ^ key, b"") for offset, dlen in entry]To make the third item an empty byte string rather than an empty string (adding the b being the only change). I can't reproduce this issue with any of the test archives I have, so I can't say if this will solve the issue - but it looks likely to me to be the bug.
Please let me know if that solves the issue so I can fix that up officially and create a new version, if it doesn't, is it possible to share (privately if copyright is an issue) an example archive that I can reproduce this issue with? That would speed up testing on this.
JeffreyARussell
Using Python 3.6.5, unrpa 1.5.2, and Windows 10. When running unrpa with a specific archive, I get the following error: "TypeError: must be str, not bytes." Using the --continue-on-error flag resulted in nothing being retrieved from the archive besides an empty folder. Full traceback is below. Thanks for the help.
Traceback (most recent call last): File "unrpa", line 61, in extract_files raw_file = self.extract_file(path, data, file_number, total_files) File "unrpa", line 85, in extract_file raw_file = start + f.read(dlen - len(start)) TypeError: must be str, not bytes:
The above exception was the direct cause of the following exception:
Traceback (most recent call last): File "unrpa", line 187, in
extractor.extract_files()
File "unrpa", line 72, in extract_files
"use the --continue-on-error flag.") from e
Exception: There was an error while trying to extract a file. See the nested exception for more. If you wish to try and extract as much from the archive as possible, please use the --continue-on-error flag.