Skip to content

⚡️ Speed up method CanonicalStrategy._is_standalone_structural_token by 19% #317

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

⚡️ Speed up method CanonicalStrategy._is_standalone_structural_token by 19% #317

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 3 additions & 11 deletions python/sglang/srt/parser/harmony_parser.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -133,6 +133,7 @@ def __init__(self):
"<|call|>",
"<|return|>",
]
self._structural_tokens_set = set(self.guard_tokens)

def parse(self, text: str) -> Tuple[List[Event], str]:
events = []
Expand Down Expand Up @@ -403,17 +404,8 @@ def _is_commentary_filler_between_blocks(

def _is_standalone_structural_token(self, content: str) -> bool:
"""Check if content is just a standalone structural token that should be filtered."""
content_stripped = content.strip()
structural_tokens = [
"<|start|>",
"<|channel|>",
"<|message|>",
"<|constrain|>",
"<|end|>",
"<|call|>",
"<|return|>",
]
return content_stripped in structural_tokens
# Use cached set for O(1) lookup
return content.strip() in self._structural_tokens_set


class TextStrategy:
Expand Down