fix: update uv overrides and docker file to use python >=3.11

Author: Dorin POMIAN

Dockerfile

@@ -22,7 +22,7 @@ RUN curl -LsSf https://astral.sh/uv/install.sh | sh \
     && uv python install 3.11 \
     && uv venv \
     && cd /tmp/build \
-    && uv --native-tls pip install --no-cache . \
+    && uv --native-tls pip install --no-cache --override /tmp/build/etc/uv-overrides.txt . \
     && uv --native-tls run --script /tmp/build/etc/fetchDefaultModels.py \
     && rm -rf /tmp/build
 

etc/uv-overrides.txt

@@ -0,0 +1,2 @@
+numba>=0.60.0
+llvmlite>=0.43.0

pyproject.toml

@@ -32,6 +32,7 @@ dependencies = [
     "python-docx>=1.1.2",
     "python-dotenv>=1.0.1",
     "python-pptx>=1.0.2",
+    "numba>=0.60.0",
     "unstructured>=0.14.8",
     "faiss-cpu>=1.11.0",
     "langchain_community>=0.3.18",

src/docs2vecs/subcommands/indexer/skills/recursive_character_splitter_skill.py

@@ -1,9 +1,10 @@
 import hashlib
 from pathlib import Path
-from typing import List
-from typing import Optional
+from typing import List, Optional
 
-from langchain.text_splitter import RecursiveCharacterTextSplitter as LCRecursiveCharacterTextSplitter
+from langchain_text_splitters import (
+    RecursiveCharacterTextSplitter as LCRecursiveCharacterTextSplitter,
+)
 
 from docs2vecs.subcommands.indexer.config.config import Config
 from docs2vecs.subcommands.indexer.document.chunk import Chunk
@@ -14,22 +15,29 @@
 class RecursiveCharacterTextSplitter(IndexerSkill):
     DEFAULT_CHUNK_SIZE = 1000
     DEFAULT_CHUNK_OVERLAP = 100
-    
+
     def __init__(self, config: dict, global_config: Config):
         super().__init__(config, global_config)
         self._set_config_defaults()
 
     def _set_config_defaults(self):
         if "chunk_size" not in self._config:
-            self._config["chunk_size"] = RecursiveCharacterTextSplitter.DEFAULT_CHUNK_SIZE
+            self._config["chunk_size"] = (
+                RecursiveCharacterTextSplitter.DEFAULT_CHUNK_SIZE
+            )
 
         if "chunk_overlap" not in self._config:
-            self._config["chunk_overlap"] = RecursiveCharacterTextSplitter.DEFAULT_CHUNK_OVERLAP
+            self._config["chunk_overlap"] = (
+                RecursiveCharacterTextSplitter.DEFAULT_CHUNK_OVERLAP
+            )
 
     def run(self, input: Optional[List[Document]] = None) -> List[Document]:
         self.logger.info("Running RecursiveCharacterTextSplitter...")
 
-        splitter = LCRecursiveCharacterTextSplitter(chunk_size=self._config["chunk_size"], chunk_overlap=self._config["chunk_overlap"])
+        splitter = LCRecursiveCharacterTextSplitter(
+            chunk_size=self._config["chunk_size"],
+            chunk_overlap=self._config["chunk_overlap"],
+        )
 
         for doc in input:
             self.logger.debug(f"Splitting {doc.filename}...")