run-llama/semtools
1
0
Fork 0
mirror of https://github.com/run-llama/semtools.git synced 2026-07-21 11:15:24 -04:00
Code Issues 4 Packages Projects Releases 10 Wiki Activity

Standard example doesn't work #10

New Issue
Closed
opened 2026-02-16 04:15:08 -05:00 by yindo · 10 comments
yindo commented 2026-02-16 04:15:08 -05:00
Owner
Copy Link
Copy Source

Originally created by @thisismygitrepo on GitHub (Nov 6, 2025).

following read me and getting this not very detailed error:

󰕈 24.4.0 noble 󱆃 ➜ search "some keywords" *.py --max-distance 0.3 --n-lines 5
Error: No such file or directory (os error 2)
Originally created by @thisismygitrepo on GitHub (Nov 6, 2025). following read me and getting this not very detailed error: ``` 󰕈 24.4.0 noble 󱆃 ➜ search "some keywords" *.py --max-distance 0.3 --n-lines 5 Error: No such file or directory (os error 2) ```
yindo closed this issue 2026-02-16 04:15:08 -05:00
yindo commented 2026-02-16 04:15:08 -05:00
Author
Owner
Copy Link
Copy Source

@logan-markewich commented on GitHub (Nov 6, 2025):

@thisismygitrepo are there actually py files in your current Directory?

@logan-markewich commented on GitHub (Nov 6, 2025): @thisismygitrepo are there actually py files in your current Directory?
yindo commented 2026-02-16 04:15:08 -05:00
Author
Owner
Copy Link
Copy Source

@logan-markewich commented on GitHub (Nov 6, 2025):

Its either that or maybe some issue downloading the embedding model? 🤔

@logan-markewich commented on GitHub (Nov 6, 2025): Its either that or maybe some issue downloading the embedding model? 🤔
yindo commented 2026-02-16 04:15:08 -05:00
Author
Owner
Copy Link
Copy Source

@logan-markewich commented on GitHub (Nov 6, 2025):

Or also maybe an installation issue too. Does which search work?

@logan-markewich commented on GitHub (Nov 6, 2025): Or also maybe an installation issue too. Does `which search` work?
yindo commented 2026-02-16 04:15:09 -05:00
Author
Owner
Copy Link
Copy Source

@thisismygitrepo commented on GitHub (Nov 6, 2025):

  • are there py files?

    • absolutely, fd .py | xargs parse Error processing file: IoError(Os { code: 21, kind: IsADirectory, message: "Is a directory" }) Error processing file: InvalidResponse("Job failed with status: ERROR") src/__init__.py src/machineconfig/__init__.py src/machineconfig/cluster/__init__.py src/machineconfig/cluster/remote/cloud_manager.py src/machineconfig/cluster/remote/data_transfer.py src/machineconfig/cluster/remote/distribute.py src/machineconfig/cluster/remote/file_manager.py src/machineconfig/cluster/remote/job_params.py src/machineconfig/cluster/remote/loader_runner.py src/machineconfig/cluster/remote/remote_machine.py

  • One file only parse src/machineconfig/cluster/remote/remote_machine.py src/machineconfig/cluster/remote/remote_machine.py

  • I saw embedding model and something like safe tensors getting downoaded, it was like 500MB.

  • instllation issue? which search /home/alex/.nvm/versions/node/v24.10.0/bin/search

  • This example worked (only after I created parent dir) echo "some new content" > ./some_large_dir/some_file.txt search "some keywords" ./some_large_dir/*.txt --n-lines 5 --top-k 10 ./some_large_dir/some_file.txt:0::1 (0.7066847052212817) 1: some new content

  • this one didn't

Image
@thisismygitrepo commented on GitHub (Nov 6, 2025): * are there py files? * absolutely, ```fd .py | xargs parse Error processing file: IoError(Os { code: 21, kind: IsADirectory, message: "Is a directory" }) Error processing file: InvalidResponse("Job failed with status: ERROR") src/__init__.py src/machineconfig/__init__.py src/machineconfig/cluster/__init__.py src/machineconfig/cluster/remote/cloud_manager.py src/machineconfig/cluster/remote/data_transfer.py src/machineconfig/cluster/remote/distribute.py src/machineconfig/cluster/remote/file_manager.py src/machineconfig/cluster/remote/job_params.py src/machineconfig/cluster/remote/loader_runner.py src/machineconfig/cluster/remote/remote_machine.py``` * One file only ```parse src/machineconfig/cluster/remote/remote_machine.py src/machineconfig/cluster/remote/remote_machine.py``` * I saw embedding model and something like safe tensors getting downoaded, it was like 500MB. * instllation issue? ```which search /home/alex/.nvm/versions/node/v24.10.0/bin/search``` * This example worked (only after I created parent dir) ``` echo "some new content" > ./some_large_dir/some_file.txt search "some keywords" ./some_large_dir/*.txt --n-lines 5 --top-k 10 ./some_large_dir/some_file.txt:0::1 (0.7066847052212817) 1: some new content``` * this one didn't <img width="1126" height="276" alt="Image" src="https://github.com/user-attachments/assets/71257e3d-ab65-4466-b247-025f9a85cf15" />
yindo commented 2026-02-16 04:15:09 -05:00
Author
Owner
Copy Link
Copy Source

@logan-markewich commented on GitHub (Nov 6, 2025):

Appreciate the debug notes. Im fairly perplexed now 😅 Will try a few ways to replicate myself..

@logan-markewich commented on GitHub (Nov 6, 2025): Appreciate the debug notes. Im fairly perplexed now 😅 Will try a few ways to replicate myself..
yindo commented 2026-02-16 04:15:09 -05:00
Author
Owner
Copy Link
Copy Source

@logan-markewich commented on GitHub (Nov 7, 2025):

@thisismygitrepo I think the max-distance just filters it out in that latest example?

(venv) loganmarkewich@Mac test_env % mkdir some_large_dir
(venv) loganmarkewich@Mac test_env % echo "some new content" > ./some_large_dir/some_file.txt
(venv) loganmarkewich@Mac test_env % search "some keywords" ./some_large_dir/*.txt --n-lines 5 --top-k 10 
./some_large_dir/some_file.txt:0::1 (0.70668465258287)
   1: some new content

(venv) loganmarkewich@Mac test_env % cd some_large_dir 
(venv) loganmarkewich@Mac some_large_dir % parse some_file.txt 
some_file.txt
(venv) loganmarkewich@Mac some_large_dir % search "some keywords" *.txt --max-distance 0.3 --n-lines 5
(venv) loganmarkewich@Mac some_large_dir % search "some keywords" *.txt --max-distance 0.9 --n-lines 5
some_file.txt:0::1 (0.70668465258287)
   1: some new content

(venv) loganmarkewich@Mac some_large_dir %

So far was not able to reproduce the original OSError (and seems like you also don't get an os error anymore?)

@logan-markewich commented on GitHub (Nov 7, 2025): @thisismygitrepo I think the max-distance just filters it out in that latest example? ```bash (venv) loganmarkewich@Mac test_env % mkdir some_large_dir (venv) loganmarkewich@Mac test_env % echo "some new content" > ./some_large_dir/some_file.txt (venv) loganmarkewich@Mac test_env % search "some keywords" ./some_large_dir/*.txt --n-lines 5 --top-k 10 ./some_large_dir/some_file.txt:0::1 (0.70668465258287) 1: some new content (venv) loganmarkewich@Mac test_env % cd some_large_dir (venv) loganmarkewich@Mac some_large_dir % parse some_file.txt some_file.txt (venv) loganmarkewich@Mac some_large_dir % search "some keywords" *.txt --max-distance 0.3 --n-lines 5 (venv) loganmarkewich@Mac some_large_dir % search "some keywords" *.txt --max-distance 0.9 --n-lines 5 some_file.txt:0::1 (0.70668465258287) 1: some new content (venv) loganmarkewich@Mac some_large_dir % ``` So far was not able to reproduce the original OSError (and seems like you also don't get an os error anymore?)
yindo commented 2026-02-16 04:15:09 -05:00
Author
Owner
Copy Link
Copy Source

@thisismygitrepo commented on GitHub (Nov 7, 2025):

I went fresh installation on a new machine, and it worked as expected on all my experiemnts, and I didn't even set api_key. On my current machine still show problems

Image

How come it worked on another machine without api key and auth? Next up, I try again with api key in the parse config file.

Image

looks like a whole set of issues to my eyes.

@thisismygitrepo commented on GitHub (Nov 7, 2025): I went fresh installation on a new machine, and it worked as expected on all my experiemnts, and I didn't even set `api_key`. On my current machine still show problems <img width="1384" height="416" alt="Image" src="https://github.com/user-attachments/assets/d1b40fa5-12a6-47c8-9825-dc22c6af242b" /> How come it worked on another machine without api key and auth? Next up, I try again with api key in the parse config file. <img width="1101" height="412" alt="Image" src="https://github.com/user-attachments/assets/2bbce5f2-2f95-4cb9-a1ea-dd4361b1f620" /> looks like a whole set of issues to my eyes.
yindo commented 2026-02-16 04:15:09 -05:00
Author
Owner
Copy Link
Copy Source

@logan-markewich commented on GitHub (Nov 10, 2025):

No idea why it doesn't work on both machines. I do know that the NPM version tries to do some matching on architecture types.

I'm curious if you installed with cargo instead if it works.

Maybe an issue in this script for NPM
https://github.com/run-llama/semtools/blob/e0d5cf679bc86c23bb911e172f798df98f0550de/scripts/install.js#L165

The API issue looks like an issue with LlamaParse failing to parse your file? Not sure what kind of config you had setup. (fyi parse on md files doesn't really make sense, LlamaParse will not do anything fancy with them anyways). It would probably help if it logged the job ID/file name when processing failed

@logan-markewich commented on GitHub (Nov 10, 2025): No idea why it doesn't work on both machines. I do know that the NPM version tries to do some matching on architecture types. I'm curious if you installed with cargo instead if it works. Maybe an issue in this script for NPM https://github.com/run-llama/semtools/blob/e0d5cf679bc86c23bb911e172f798df98f0550de/scripts/install.js#L165 The API issue looks like an issue with LlamaParse failing to parse your file? Not sure what kind of config you had setup. (fyi `parse` on md files doesn't really make sense, LlamaParse will not do anything fancy with them anyways). It would probably help if it logged the job ID/file name when processing failed
yindo commented 2026-02-16 04:15:09 -05:00
Author
Owner
Copy Link
Copy Source

@thisismygitrepo commented on GitHub (Nov 11, 2025):

Yep, definitely npm to blame, I downloaded the binaries from release page and used them, it works seamlessly

@thisismygitrepo commented on GitHub (Nov 11, 2025): Yep, definitely npm to blame, I downloaded the binaries from release page and used them, it works seamlessly
yindo commented 2026-02-16 04:15:09 -05:00
Author
Owner
Copy Link
Copy Source

@logan-markewich commented on GitHub (Nov 13, 2025):

I'll give the NPM script another look and see if I can improve 🙏🏻

@logan-markewich commented on GitHub (Nov 13, 2025): I'll give the NPM script another look and see if I can improve 🙏🏻
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
enhancement
pull-request
Mirrored from GitHub Pull Request
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: run-llama/semtools#10
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?