PageSourceSearch

https://ml-infra-toloka-tangle-public.hf.space/assets/python-b6uHYRtL.js

js ml-infra-toloka-tangle-public.hf.space collected 2026-10-04 10:19:39 UTC 1,250 bytes, 43 lines download raw bytes

1const t=`name: Filter text
2inputs:
3- {name: Text}
4- {name: Pattern, default: '.*'}
5outputs:
6- {name: Filtered text}
7metadata:
8  annotations:
9    author: Your Name <[email protected]>
10implementation:
11  container:
12    image: python:3.8
13    command:
14    - sh
15    - -ec
16    - |
17      # This is how additional packages can be installed dynamically
18      python3 -m pip install pip six
19      # Run the rest of the command after installing the packages.
20      "$0" "$@"
21    - python3
22    - -u  # Auto-flush. We want the logs to appear in the console immediately.
23    - -c  # Inline scripts are easy, but have size limitaions and the error traces do not show source lines.
24    - |
25      import os
26      import re
27      import sys
28
29      text_path = sys.argv[1]
30      pattern = sys.argv[2]
31      filtered_text_path = sys.argv[3]
32
33      regex = re.compile(pattern)
34
35      os.makedirs(os.path.dirname(filtered_text_path), exist_ok=True)
36      with open(text_path, 'r') as reader:
37          with open(filtered_text_path, 'w') as writer:
38              for line in reader:
39                  if regex.search(line):
40                      writer.write(line)
41    - {inputPath: Text}
42    - {inputValue: Pattern}
43    - {outputPath: Filtered text}`;export{t as default};

Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.