Skip to content

semble search infinitly stucks on middle repository #258

Description

@arxdeus

hi, using semble 0.5.5 in middle sized repo with few submodules and any search command is just stucks

UPD: one note - i was running semble not from root of git repo, but inside of subfolder

have this on interrupt

semble search "auth"
^CTraceback (most recent call last):
  File "/Users/mind/.local/bin/semble", line 10, in <module>
    sys.exit(main())
             ~~~~^^
  File "/Users/mind/.local/share/uv/tools/semble/lib/python3.14/site-packages/semble/cli.py", line 73, in main
    _cli_main()
    ~~~~~~~~~^^
  File "/Users/mind/.local/share/uv/tools/semble/lib/python3.14/site-packages/semble/cli.py", line 290, in _cli_main
    _run_search(
    ~~~~~~~~~~~^
        args.path,
        ^^^^^^^^^^
    ...<3 lines>...
        args.max_snippet_lines,
        ^^^^^^^^^^^^^^^^^^^^^^^
    )
    ^
  File "/Users/mind/.local/share/uv/tools/semble/lib/python3.14/site-packages/semble/cli.py", line 118, in _run_search
    index = _load_index(path, content)
  File "/Users/mind/.local/share/uv/tools/semble/lib/python3.14/site-packages/semble/cli.py", line 110, in _load_index
    return _build_index(path, content)
  File "/Users/mind/.local/share/uv/tools/semble/lib/python3.14/site-packages/semble/cli.py", line 37, in _build_index
    else SembleIndex.from_path(path, content=content)
         ~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/mind/.local/share/uv/tools/semble/lib/python3.14/site-packages/semble/index/index.py", line 164, in from_path
    bm25_index, semantic_index, chunks, manifest = create_index_from_path(
                                                   ~~~~~~~~~~~~~~~~~~~~~~^
        path,
        ^^^^^
    ...<3 lines>...
        previous=previous,
        ^^^^^^^^^^^^^^^^^^
    )
    ^
  File "/Users/mind/.local/share/uv/tools/semble/lib/python3.14/site-packages/semble/index/create.py", line 102, in create_index_from_path
    vector_parts.append(embed_chunks(model, file_chunks))
                        ~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^
  File "/Users/mind/.local/share/uv/tools/semble/lib/python3.14/site-packages/semble/index/dense.py", line 46, in embed_chunks
    return np.array(model.encode([c.content for c in chunks], use_multiprocessing=False), dtype=np.float32)
                    ~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/mind/.local/share/uv/tools/semble/lib/python3.14/site-packages/model2vec/model.py", line 399, in encode
    out_arrays.append(self._encode_batch(batch, max_length))
                      ~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^
  File "/Users/mind/.local/share/uv/tools/semble/lib/python3.14/site-packages/model2vec/model.py", line 428, in _encode_batch
    ids = self.tokenize(sentences=sentences, max_length=max_length)
  File "/Users/mind/.local/share/uv/tools/semble/lib/python3.14/site-packages/model2vec/model.py", line 169, in tokenize
    [token_id for token_id in token_ids if token_id != self.unk_token_id] for token_ids in encodings_ids
     ^^^^^^^^
KeyboardInterrupt

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions