fix: detok review findings (null vocab guard + utf-8 tokens read)#303
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Addresses the gemini-code-assist findings on the merged in-binary detok (#3004/#302).
gguf_get_arr_str(funasr-sensevoice.cpp / funasr-paraformer.cpp): a malformed GGUF could returnnullptrfor a vocab element; assigning that tostd::stringis UB. Now coerces to empty string.tokens.jsonread (export_paraformer_gguf.py): usewith open(..., encoding="utf-8")— a context manager (no fd leak) and explicit UTF-8 so the non-ASCII tokens load correctly on Windows (default cp1252).Not changed: adjacent VAD-segment texts are concatenated without a separator — that is intentional for Chinese, where the VAD splits fall mid-utterance and a space would fragment the sentence (CER is unaffected either way).
--idsstill gives raw ids if a caller wants per-segment structure.Verified: rebuilt, both binaries still print the correct text (sample + full 002).