klbr/khoj
1
0
Fork 0
mirror of https://github.com/khoaliber/khoj.git synced 2026-03-08 05:39:13 +00:00
Code Issues Packages Projects Releases Wiki Activity

Truncate long words in stdout, stderr for context efficiency

Browse Source 
Avoid long base64 images etc. in stdout, stderr to result in context
limits being hit.
This commit is contained in:
Debanjum
2025-08-19 23:31:45 -07:00
parent dbc3330610
commit fb0347a388
1 changed files with 8 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
src/khoj/utils/helpers.py
View File

@@ -9,6 +9,7 @@ import logging
import os import os
import platform import platform
import random import random
import re
import urllib.parse import urllib.parse
import uuid import uuid
from collections import OrderedDict from collections import OrderedDict
@@ -894,6 +895,13 @@ def truncate_code_context(original_code_results: dict[str, Any], max_chars=10000
"filename": output_file["filename"], "filename": output_file["filename"],
"b64_data": output_file["b64_data"][:max_chars] + "...", "b64_data": output_file["b64_data"][:max_chars] + "...",
} }
# Truncate long "words" in stdout, stderr. Words are alphanumeric strings not separated by whitespace.
for key in ["std_out", "std_err"]:
if key in code_result["results"]:
code_result["results"][key] = re.sub(
r"\S{1000,}", lambda m: m.group(0)[:1000] + "...", code_result["results"][key]
)
return code_results return code_results