Decode() performance fix #227
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This is part of a push to support larger files. The focus is the switch
to using Uint8Array to store binary data. But also includes:
- When running on Node, use Buffer.from() for hex string conversions.
- To avoid large buffer copy, signedMsg as been replaced by an object
containing iv, encrypted, and hmac.
- hmac calculation has changed so it avoids copying (possibly very
large) encrypted data. See signDigest() in lib/codec.js.
- Minor cleanup
Handling hex encode/decode at the input/output boundaries and using
Uint8Array internally for representing binary data has these benefits:
- More memory efficient, allows processing of 2x larger files.
- Aligns with cryptographic best practices: hashing is now performed
on raw binary data (Uint8Array) instead of hex strings.
- Behavior is (mostly) unchanged
- scripts/index_template.html textContent is not implemented and
needs to be redesigned.
makes unnamed function more self documenting
Also removed use of recursion to improve readability (and debugability). As a bonus, function is actually shorter (LoC AND lines of text)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
codec.js/decode() may calculate encryptedDataHash multiple times. calculatedHMAC is based on encryptedDataHash which can be arbitrarily large. Recalculating this hash when the data hasn't changed is a performance bug. The fix is to calculate the hash once. While reviewing this change the use of recursion was replaced with explicit retries.
Fixes
Fixes #226
Details
Testing
I manually tested node encrypt, node decrypt, and HTML wrapper decrypt.
It's not clear to me exactly how I could test the backwards compatible password hashes. Additional testing is recommended.
Notes for Reviewers
Multiple hashing has been around a while, but it was not easy to address before PR #219. I'd say at that point the performance issue became a bug because it was fixable.
Probably variable names 'hashedPassword2' and 'hashedPassword3' could be improved. I assigned them in the order they were used, but:
May be confusing.