chore: add boilerpy3 to the core dependencies (#6544)

* add boilerpy3 to the core dependencies

* remove boilerpy3 installation from test workflow

* fix pylint: import order and unused import

* fix import order

* add release note

---------

Co-authored-by: Julian Risch <[email protected]>

.github/workflows/tests.yml

@@ -98,7 +98,7 @@ jobs:
           python-version: ${{ env.PYTHON_VERSION }}
 
       - name: Install Haystack
-        run: pip install .[dev,audio] langdetect transformers[torch,sentencepiece]==4.35.2 'sentence-transformers>=2.2.0' pypdf markdown-it-py mdit_plain tika 'azure-ai-formrecognizer>=3.2.0b2' boilerpy3
+        run: pip install .[dev,audio] langdetect transformers[torch,sentencepiece]==4.35.2 'sentence-transformers>=2.2.0' pypdf markdown-it-py mdit_plain tika 'azure-ai-formrecognizer>=3.2.0b2'
 
       - name: Run
         run: pytest -m "not integration" test
@@ -156,7 +156,7 @@ jobs:
           sudo apt install ffmpeg  # for local Whisper tests
 
       - name: Install Haystack
-        run: pip install .[dev,audio] langdetect transformers[torch,sentencepiece]==4.35.2 'sentence-transformers>=2.2.0' pypdf markdown-it-py mdit_plain tika 'azure-ai-formrecognizer>=3.2.0b2' boilerpy3
+        run: pip install .[dev,audio] langdetect transformers[torch,sentencepiece]==4.35.2 'sentence-transformers>=2.2.0' pypdf markdown-it-py mdit_plain tika 'azure-ai-formrecognizer>=3.2.0b2'
 
       - name: Run
         run: pytest --maxfail=5 -m "integration" test
@@ -212,7 +212,7 @@ jobs:
           colima start
 
       - name: Install Haystack
-        run: pip install .[dev,audio] langdetect transformers[torch,sentencepiece]==4.35.2 'sentence-transformers>=2.2.0' pypdf markdown-it-py mdit_plain tika 'azure-ai-formrecognizer>=3.2.0b2' boilerpy3
+        run: pip install .[dev,audio] langdetect transformers[torch,sentencepiece]==4.35.2 'sentence-transformers>=2.2.0' pypdf markdown-it-py mdit_plain tika 'azure-ai-formrecognizer>=3.2.0b2'
 
       - name: Run Tika
         run: docker run -d -p 9998:9998 apache/tika:2.9.0.0
@@ -263,7 +263,7 @@ jobs:
           python-version: ${{ env.PYTHON_VERSION }}
 
       - name: Install Haystack
-        run: pip install .[dev,audio] langdetect transformers[torch,sentencepiece]==4.35.2 'sentence-transformers>=2.2.0' pypdf markdown-it-py mdit_plain tika 'azure-ai-formrecognizer>=3.2.0b2' boilerpy3
+        run: pip install .[dev,audio] langdetect transformers[torch,sentencepiece]==4.35.2 'sentence-transformers>=2.2.0' pypdf markdown-it-py mdit_plain tika 'azure-ai-formrecognizer>=3.2.0b2'
 
       - name: Run
         run: pytest --maxfail=5 -m "integration" test -k 'not tika'

haystack/components/converters/html.py

@@ -1,16 +1,13 @@
 import logging
 from pathlib import Path
 from typing import Any, Dict, List, Optional, Union
+from boilerpy3 import extractors
 
 from haystack import Document, component
 from haystack.dataclasses import ByteStream
-from haystack.lazy_imports import LazyImport
 
 logger = logging.getLogger(__name__)
 
-with LazyImport("Run 'pip install boilerpy3'") as boilerpy3_import:
-    from boilerpy3 import extractors
-
 
 @component
 class HTMLToDocument:
@@ -30,12 +27,6 @@ class HTMLToDocument:
 
     """
 
-    def __init__(self):
-        """
-        Initializes the HTMLToDocument component.
-        """
-        boilerpy3_import.check()
-
     @component.output_types(documents=List[Document])
     def run(self, sources: List[Union[str, Path, ByteStream]], meta: Optional[List[Dict[str, Any]]] = None):
         """

pyproject.toml

@@ -58,6 +58,7 @@ dependencies = [
   "more-itertools",  # TextDocumentSplitter
   "networkx", # Pipeline graphs
   "typing_extensions", # typing support for Python 3.8
+  "boilerpy3", # Fulltext extraction from HTML pages
 ]
 
 [project.optional-dependencies]

releasenotes/notes/ship-boilerpy3-0bffbd7955c89dd4.yaml

@@ -0,0 +1,4 @@
+---
+enhancements:
+  - |
+    Include 'boilerpy3' in the 'haystack-ai' dependencies.