Skip to content

display RO / RW context (--verbose)#1054

Merged
KRRT7 merged 5 commits into
mainfrom
let's-see-what-we're-sending
Jan 14, 2026
Merged

display RO / RW context (--verbose)#1054
KRRT7 merged 5 commits into
mainfrom
let's-see-what-we're-sending

Conversation

@KRRT7

@KRRT7 KRRT7 commented Jan 14, 2026

Copy link
Copy Markdown
Contributor

No description provided.

@KRRT7

KRRT7 commented Jan 14, 2026

Copy link
Copy Markdown
Contributor Author

I think token_limit is currently too low, thoughts @aseembits93


if analyzer.found_any_target_function:
logger.debug(f"Test file {test_file_path} imports target function: {analyzer.found_qualified_name}")
# logger.debug(f"Test file {test_file_path} imports target function: {analyzer.found_qualified_name}")

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why to keep the logs as comments when commenting.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

just in case we want to bring them back

@KRRT7

KRRT7 commented Jan 14, 2026

Copy link
Copy Markdown
Contributor Author
Screenshot 2026-01-14 at 8 11 32 AM example

@Saga4

Saga4 commented Jan 14, 2026

Copy link
Copy Markdown
Contributor

I think token_limit is currently too low, thoughts @aseembits93

Did we got quote increase from foundary?
And when did we see the tokens exceeding like in which scenerio? Test gen etc?

@Saga4 Saga4 left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good for experiment but unsure if we want to expose this much info behind verbose flag or some other flag.

@KRRT7

KRRT7 commented Jan 14, 2026

Copy link
Copy Markdown
Contributor Author

@Saga4 I noticed functions that call a lot of other functions brings in a lot of context which caused it to hit the cap somewhere

@KRRT7

KRRT7 commented Jan 14, 2026

Copy link
Copy Markdown
Contributor Author

well --verbose is only used for debugging so far which is where it fits best imo

@aseembits93

Copy link
Copy Markdown
Contributor

looking into token limits right now

Comment thread codeflash/code_utils/config_consts.py
@aseembits93

Copy link
Copy Markdown
Contributor

@Saga4 @KRRT7 from a cursory look at recent testgen and optimization generation calls in the observability tool, I'm seeing approximately 30k tokens at max in the input. This is far below the ~200k input token thresholds modern LLMs.

@KRRT7 KRRT7 merged commit 4e7b14e into main Jan 14, 2026
21 of 22 checks passed
@KRRT7 KRRT7 deleted the let's-see-what-we're-sending branch January 14, 2026 23:51
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants