feat(environment): stream lines in ReadFileTool to avoid OOM - #7176
Open
ArulJerald wants to merge 1 commit into
Open
ArulJerald wants to merge 1 commit into
ArulJerald wants to merge 1 commit into
Conversation
Read lines lazily in LocalEnvironment.read_file_lines instead of buffering the entire file into memory as bytes before slicing. Closes: google#7131
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Link to Issue or Description of Change
1. Link to an existing issue (if applicable):
Problem:
In
ReadFileTool, reading files previously loaded the entire file contents into memory viaself._environment.read_file(path)and split all lines in memory (data_bytes.splitlines()) before slicing by line boundaries. For very large files, buffering the entire file into memory causes significant memory spikes and potential Out Of Memory (OOM) errors.Solution:
read_file_lines(path, start_line, end_line)toBaseEnvironmentwith a default implementation that falls back toread_file.read_file_linesinLocalEnvironmentusing a streaming_sync_read_lineshelper that iterates through the file line by line without buffering the whole file in memory.ReadFileTool.run_asyncto useself._environment.read_file_lines(path, start_line=start, end_line=end_line)to lazily fetch only the requested range and the line count.TestReadFileToolcovering empty files,start_lineexceeding total lines, and line ordering validation.Testing Plan
Unit Tests:
Summary of passed
pytestresults:Manual End-to-End (E2E) Tests:
ReadFileToolagainst empty files, small text files, and files where line bounds are out of range or ordered in reverse. All validations returned expected structured error responses matching previous behavior with reduced memory footprint.Checklist