The material change in here is the path cleaner/splitter; it's a bit too aggressive. It doesn't allow \ as a file name and the .trim() discounts valid file names. Otherwise this is a match for https://github.com/ipfs/go-unixfsnode/pull/66 and gets the same CID on importer, and tests the exporter can read the same data.
Ref: https://github.com/ipfs/go-unixfsnode/pull/66
The material change in here is the path cleaner/splitter; it's a bit too aggressive. It doesn't allow
\
as a file name and the.trim()
discounts valid file names. Otherwise this is a match for https://github.com/ipfs/go-unixfsnode/pull/66 and gets the same CID on importer, and tests the exporter can read the same data.