v2 Torrent supported？

[7sdream]

v2 Torrent supported？

[7sDream]

Could you try way 2 in this comment and check if it works?

with argument: hash_fields={"pieces root": (32, False)}

[daomengRen]

I try data = tp.parse_torrent_file(str, hash_fields={"pieces root": (32, False)})， Getting “torrent_parser.InvalidTorrentDataException: Fail to decode string at pos 1359 using encoding utf-8”

[7sDream]

It seems this v2 file has a piece layers dict whose key and value are all hash bytes.

Current hash_fields argument can't specify entrie dict.

For now, if you don't care about piece layers values, you can add errors="replace" to make parse success:

data = tp.parse_torrent_file(
    "bittorrent-v2-test.torrent",
    hash_fields={
        "pieces root": (32, False),
    },
    errors="replace",
)

But if you edit data and write back to file, the content of piece layers will be incorrect.

I will publish a new version to solve this issue.

[daomengRen]

Looking forward to new version，If I only edit the content except the piece layers, will I get an error when writing to the file?

[7sDream]

No error will be raise if you write back to file, it is the file content is incorrect.

Because error="replace" argument of parse step replaced invalid utf8 bytes in piece layers with Unicode Replacement Character. In encode/write step, we can't restore origin bytes.

[7sDream]

v0.4.0 just published.

Please try it with tp.parse_torrent_file("bittorrent-v2-test.torrent") without any other argument. In theory you should be able to successfully parse the file.

Feel free to continue this thread if you meet any issue when using the new version.

[daomengRen]

It's working fine, thank you

[7sDream]

That's fine, then I'll close the issue.

If there are any subsequent issue, you can open a new one.