picohash

Alpha version!

Everything can change at any time (including flags).

English translation coming soon.

---

Why?

• How can I know my files are not corrupted?
  • ...on an old hard drive sitting in the basement?
  • ...during a transfer from one hard drive to another?
  • ...during usage?
• Which backup is healthier?
• "Wait, there is a conflict between the local version and cloud version on a file I haven't touched in years?"
• Damn cosmic rays!

"Oh, there’s ZFS, which generates the file content checksum automatically."

Oh, really? Try using ZFS on macOS, Windows, or even Linux. Then come back and appreciate the portability of this program.

---

Install

go install github.com/lucasbalena/picohash@latest

Hash

picohash

Using rapidhash

picohash -rapid

Without sidecar files (creates hashes.b3)

picohash -a

Using rapidhash (creates hashes.rph)

picohash -a -rapid

Specify another directory

picohash -a ../storage

If you are using a hard drive

picohash -hdd

This will pass the flags --no-mmap --num-threads=1 to b3sum.

If you want to use cat to read the files for some reason

picohash -cat

Check

picohash -c

Using the hashes.b3 file

picohash -c -a

Using the hashes.rph file

picohash -c -a -rapid

Manage the hashes

Use -rapid if you hashed with rapidhash. This will change the extension from .b3 to .rph

Delete all sidecar hash files

picohash -d

Split hashes.b3 into sidecar files

picohash -s

Join the sidecar files into a single hashes.b3

picohash -j

It reads hashes.b3 (if it exists) and adds any new hashes found in the directory to it.

If you want to copy only a subfolder that was hashed with -a, you can split the hashes, copy the subfolder along with its associated hashes, paste them in the destination, and then join the hashes. This ensures that all relative paths remain correct.

Rename a file or directory

1. Rename the file.
2. Rename the sidecar file.
3. Run the following to rewrite the sidecar content to match the new filename.

picohash -r

If you used -a:

picohash -s                     # Who wants to deal with a long .txt?
# Rename the file and sidecar
picohash -r                     # Rename the inside contents
# Delete hashes.b3              # To prevent duplicates
picohash -j                     # Recreate hashes.b3
picohash -d                     # Delete the sidecar files

Or

picohash -s                     # Who wants to deal with a long .txt?
# Rename the file
# Delete the sidecar
# Delete hashes.b3              # To prevent duplicates
picohash -j                     # Recreate hashes.b3
picohash -d                     # Delete the sidecar files
picohash -a                     # Hash the renamed file

Perhaps there are better ways to handle renaming

The current structure for .b3 and .rph is as follows:

hash name

This format ensures compatibility with other hash-checking programs but requires using -r to update the sidecar file content to match the new name.

However, even if we change the structure to:

hash

all other steps would still be necessary with you use -a.

In a folder with thousands of files and subfolders, having sidecar files becomes overwhelming.

Subtitles

Symbol	Meaning
✅	Passed
❌	Mismatch
🔍 📂	File not found (hash exists)
🔍 ⛏️	Hash not found (file exists)
❗	Unexpected error (big problem)
⚙️	Hash generated successfully
⏭️	Skipped a file already hashed
🤞	Expected hash (but does not match, appears with ❌)