[opt] minor compression ratio improvement #2983
Merged
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
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Outdated suggestions cannot be applied.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.
Following @terrelln's diagnosis #2980,
this is a small follow up
which enforces a minimum cost, of 1 bit per literal, in the cost evaluation of the optimal parser stage,
which is consistent with the hypothesis of a later huffman compression stage.
As can be guessed, this minimum cost enforcement has generally no impact,
since in the vast majority of cases, no literal is dominant.
As explained in #2980, this change is rather meant to take care of some special corner cases.
Nonetheless, I could find 2 files from public corpus which the new policy results in a small compression difference,
favorable in both circumstances :
The benefits also extend to some of the regression tests.