7zip archives are sometimes considered malformed

[evilrix]

Create two 7z files. Each contains a folder called "test". Each contains an empty file in the test folder called "y".

One contains an empty file in the test folder called "xx" and the other contains an empty file in the test folder called "xxx".

When extracting the file with "xxx" libarchive fails with the error :"libarchive error: Damaged 7-Zip archive - unable to get next archive item (-1)"/

When extracting the file with "xx" extraction works as expected.

[kientzle]

Can you provide the 7z files in question?

Thank you,

Tim

On May 8, 2015, at 6:43 PM, evilrix notifications@github.com wrote:

Create two 7z files. Each contains a folder called "test". Each contains an empty file in the test folder called "y".

One contains an empty file in the test folder called "xx" and the other contains an empty file in the test folder called "xxx".

When extracting the file with "xxx" libarchive fails with the error :"libarchive error: Damaged 7-Zip archive - unable to get next archive item (-1)"/

When extracting the file with "xx" extraction works as expected.

—
Reply to this email directly or view it on GitHub #535.

[evilrix]

Hello, Tim. Thanks for reviewing this ticket. I've uploaded examples of the 7zip files plus code snippets demonstrating my implementation for the read and seek callbacks. I accept this could be my fault if these are wrong, so any observations you have on these implementations would be welcome. That said, they work fine for zip files; although, I think I am right in saying zip doesn't use the seeker function, right?

https://github.com/evilrix/catfood/tree/libarchive-issues-535

I'm very keen to get to the bottom of this so please do let me know if there is anything else I can do to assist you with this issue (testing/more examples).

Best regards,

-Rix

[TomerHeber]

I cannot share my extraction code. However, I can confirm that I was able to extract the file after the fix for #556:

test_bad.7z
Reading: /tmp/test_bad.7z
Mime type detected: application/x-7z-compressed

Inflated file name - mime type: test/ - text/directory

Inflated file name - mime type: test/xxx - text/directory

Inflated file name - mime type: test/y - text/directory

OK: extracted text!
Total Data In (bytes): 135
Total Text Out (bytes): 0

[evilrix]

Thank you TomerHeber for sharing that.

[kientzle]

If you could re-test with libarchive 3.2, I would greatly appreciate it. That seems to have fixed some 7z issues.