Merge branch 'main' into xlsx-converter

.pre-commit-config.yaml

@@ -26,6 +26,7 @@ repos:
     rev: v2.3.0
     hooks:
       - id: codespell
+        exclude: "haystack/data/abbreviations"
         args: ["--toml", "pyproject.toml"]
         additional_dependencies:
           - tomli

docs/pydoc/config/data_classess_api.yml

@@ -2,7 +2,7 @@ loaders:
   - type: haystack_pydoc_tools.loaders.CustomPythonLoader
     search_path: [../../../haystack/dataclasses]
     modules:
-      ["answer", "byte_stream", "chat_message", "document", "streaming_chunk", "sparse_embedding"]
+      ["answer", "byte_stream", "chat_message", "document", "streaming_chunk", "sparse_embedding", "tool"]
     ignore_when_discovered: ["__init__"]
 processors:
   - type: filter

e2e/pipelines/test_dense_doc_search.py

@@ -26,7 +26,7 @@ def test_dense_doc_search_pipeline(tmp_path, samples_path):
     indexing_pipeline.add_component(instance=DocumentJoiner(), name="joiner")
     indexing_pipeline.add_component(instance=DocumentCleaner(), name="cleaner")
     indexing_pipeline.add_component(
-        instance=DocumentSplitter(split_by="sentence", split_length=250, split_overlap=30), name="splitter"
+        instance=DocumentSplitter(split_by="period", split_length=250, split_overlap=30), name="splitter"
     )
     indexing_pipeline.add_component(
         instance=SentenceTransformersDocumentEmbedder(model="sentence-transformers/all-MiniLM-L6-v2"), name="embedder"

e2e/pipelines/test_preprocessing_pipeline.py

@@ -2,8 +2,6 @@
 #
 # SPDX-License-Identifier: Apache-2.0
 
-import json
-
 from haystack import Pipeline
 from haystack.components.classifiers import DocumentLanguageClassifier
 from haystack.components.converters import TextFileToDocument
@@ -25,9 +23,7 @@ def test_preprocessing_pipeline(tmp_path):
         instance=MetadataRouter(rules={"en": {"field": "language", "operator": "==", "value": "en"}}), name="router"
     )
     preprocessing_pipeline.add_component(instance=DocumentCleaner(), name="cleaner")
-    preprocessing_pipeline.add_component(
-        instance=DocumentSplitter(split_by="sentence", split_length=1), name="splitter"
-    )
+    preprocessing_pipeline.add_component(instance=DocumentSplitter(split_by="period", split_length=1), name="splitter")
     preprocessing_pipeline.add_component(
         instance=SentenceTransformersDocumentEmbedder(model="sentence-transformers/all-MiniLM-L6-v2"), name="embedder"
     )

haystack/components/builders/chat_prompt_builder.py

@@ -9,7 +9,7 @@
 from jinja2.sandbox import SandboxedEnvironment
 
 from haystack import component, default_from_dict, default_to_dict, logging
-from haystack.dataclasses.chat_message import ChatMessage, ChatRole
+from haystack.dataclasses.chat_message import ChatMessage, ChatRole, TextContent
 
 logger = logging.getLogger(__name__)
 
@@ -197,10 +197,10 @@ def run(
                 if message.text is None:
                     raise ValueError(f"The provided ChatMessage has no text. ChatMessage: {message}")
                 compiled_template = self._env.from_string(message.text)
-                rendered_content = compiled_template.render(template_variables_combined)
+                rendered_text = compiled_template.render(template_variables_combined)
                 # deep copy the message to avoid modifying the original message
                 rendered_message: ChatMessage = deepcopy(message)
-                rendered_message.content = rendered_content
+                rendered_message._content = [TextContent(text=rendered_text)]
                 processed_messages.append(rendered_message)
             else:
                 processed_messages.append(message)

haystack/components/converters/azure.py

@@ -5,7 +5,6 @@
 import copy
 import hashlib
 import os
-import warnings
 from collections import defaultdict
 from pathlib import Path
 from typing import Any, Dict, List, Literal, Optional, Union
@@ -143,12 +142,6 @@ def run(self, sources: List[Union[str, Path, ByteStream]], meta: Optional[List[D
             azure_output.append(result.to_dict())
 
             merged_metadata = {**bytestream.meta, **metadata}
-            warnings.warn(
-                "The `store_full_path` parameter defaults to True, storing full file paths in metadata. "
-                "In the 2.9.0 release, the default value for `store_full_path` will change to False, "
-                "storing only file names to improve privacy.",
-                DeprecationWarning,
-            )
 
             if not self.store_full_path and (file_path := bytestream.meta.get("file_path")):
                 merged_metadata["file_path"] = os.path.basename(file_path)

haystack/components/converters/csv.py

@@ -4,7 +4,6 @@
 
 import io
 import os
-import warnings
 from pathlib import Path
 from typing import Any, Dict, List, Optional, Union
 
@@ -94,13 +93,6 @@ def run(
 
             merged_metadata = {**bytestream.meta, **metadata}
 
-            warnings.warn(
-                "The `store_full_path` parameter defaults to True, storing full file paths in metadata. "
-                "In the 2.9.0 release, the default value for `store_full_path` will change to False, "
-                "storing only file names to improve privacy.",
-                DeprecationWarning,
-            )
-
             if not self.store_full_path and "file_path" in bytestream.meta:
                 file_path = bytestream.meta.get("file_path")
                 if file_path:  # Ensure the value is not None for pylint

haystack/components/converters/docx.py

@@ -5,7 +5,6 @@
 import csv
 import io
 import os
-import warnings
 from dataclasses import dataclass
 from enum import Enum
 from io import StringIO
@@ -189,13 +188,6 @@ def run(
                 )
                 continue
 
-            warnings.warn(
-                "The `store_full_path` parameter defaults to True, storing full file paths in metadata. "
-                "In the 2.9.0 release, the default value for `store_full_path` will change to False, "
-                "storing only file names to improve privacy.",
-                DeprecationWarning,
-            )
-
             docx_metadata = self._get_docx_metadata(document=docx_document)
             merged_metadata = {**bytestream.meta, **metadata, "docx": docx_metadata}
 

haystack/components/converters/html.py

@@ -3,7 +3,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import os
-import warnings
 from pathlib import Path
 from typing import Any, Dict, List, Optional, Union
 
@@ -123,12 +122,6 @@ def run(
 
             merged_metadata = {**bytestream.meta, **metadata}
 
-            warnings.warn(
-                "The `store_full_path` parameter defaults to True, storing full file paths in metadata. "
-                "In the 2.9.0 release, the default value for `store_full_path` will change to False, "
-                "storing only file names to improve privacy.",
-                DeprecationWarning,
-            )
             if not self.store_full_path and "file_path" in bytestream.meta:
                 file_path = bytestream.meta.get("file_path")
                 if file_path:  # Ensure the value is not None for pylint

haystack/components/converters/json.py

@@ -4,7 +4,6 @@
 
 import json
 import os
-import warnings
 from pathlib import Path
 from typing import Any, Dict, List, Literal, Optional, Set, Tuple, Union
 
@@ -280,12 +279,6 @@ def run(
 
             data = self._get_content_and_meta(bytestream)
 
-            warnings.warn(
-                "The `store_full_path` parameter defaults to True, storing full file paths in metadata. "
-                "In the 2.9.0 release, the default value for `store_full_path` will change to False, "
-                "storing only file names to improve privacy.",
-                DeprecationWarning,
-            )
             for text, extra_meta in data:
                 merged_metadata = {**bytestream.meta, **metadata, **extra_meta}
 

haystack/components/converters/markdown.py

@@ -3,7 +3,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import os
-import warnings
 from pathlib import Path
 from typing import Any, Dict, List, Optional, Union
 
@@ -112,13 +111,6 @@ def run(
 
             merged_metadata = {**bytestream.meta, **metadata}
 
-            warnings.warn(
-                "The `store_full_path` parameter defaults to True, storing full file paths in metadata. "
-                "In the 2.9.0 release, the default value for `store_full_path` will change to False, "
-                "storing only file names to improve privacy.",
-                DeprecationWarning,
-            )
-
             if not self.store_full_path and (file_path := bytestream.meta.get("file_path")):
                 merged_metadata["file_path"] = os.path.basename(file_path)
 

haystack/components/converters/pdfminer.py

@@ -4,7 +4,6 @@
 
 import io
 import os
-import warnings
 from pathlib import Path
 from typing import Any, Dict, List, Optional, Union
 
@@ -172,12 +171,6 @@ def run(
                 )
 
             merged_metadata = {**bytestream.meta, **metadata}
-            warnings.warn(
-                "The `store_full_path` parameter defaults to True, storing full file paths in metadata. "
-                "In the 2.9.0 release, the default value for `store_full_path` will change to False, "
-                "storing only file names to improve privacy.",
-                DeprecationWarning,
-            )
 
             if not self.store_full_path and (file_path := bytestream.meta.get("file_path")):
                 merged_metadata["file_path"] = os.path.basename(file_path)

haystack/components/converters/pptx.py

@@ -4,7 +4,6 @@
 
 import io
 import os
-import warnings
 from pathlib import Path
 from typing import Any, Dict, List, Optional, Union
 
@@ -104,12 +103,6 @@ def run(
                 continue
 
             merged_metadata = {**bytestream.meta, **metadata}
-            warnings.warn(
-                "The `store_full_path` parameter defaults to True, storing full file paths in metadata. "
-                "In the 2.9.0 release, the default value for `store_full_path` will change to False, "
-                "storing only file names to improve privacy.",
-                DeprecationWarning,
-            )
 
             if not self.store_full_path and (file_path := bytestream.meta.get("file_path")):
                 merged_metadata["file_path"] = os.path.basename(file_path)

haystack/components/converters/pypdf.py

@@ -4,7 +4,6 @@
 
 import io
 import os
-import warnings
 from enum import Enum
 from pathlib import Path
 from typing import Any, Dict, List, Optional, Union
@@ -220,12 +219,7 @@ def run(
                 )
 
             merged_metadata = {**bytestream.meta, **metadata}
-            warnings.warn(
-                "The `store_full_path` parameter defaults to True, storing full file paths in metadata. "
-                "In the 2.9.0 release, the default value for `store_full_path` will change to False, "
-                "storing only file names to improve privacy.",
-                DeprecationWarning,
-            )
+
             if not self.store_full_path and (file_path := bytestream.meta.get("file_path")):
                 merged_metadata["file_path"] = os.path.basename(file_path)
             document.meta = merged_metadata

haystack/components/converters/tika.py

@@ -4,7 +4,6 @@
 
 import io
 import os
-import warnings
 from html.parser import HTMLParser
 from pathlib import Path
 from typing import Any, Dict, List, Optional, Union
@@ -139,12 +138,6 @@ def run(
                 continue
 
             merged_metadata = {**bytestream.meta, **metadata}
-            warnings.warn(
-                "The `store_full_path` parameter defaults to True, storing full file paths in metadata. "
-                "In the 2.9.0 release, the default value for `store_full_path` will change to False, "
-                "storing only file names to improve privacy.",
-                DeprecationWarning,
-            )
 
             if not self.store_full_path and (file_path := bytestream.meta.get("file_path")):
                 merged_metadata["file_path"] = os.path.basename(file_path)

haystack/components/converters/txt.py

@@ -3,7 +3,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import os
-import warnings
 from pathlib import Path
 from typing import Any, Dict, List, Optional, Union
 
@@ -93,12 +92,6 @@ def run(
                 continue
 
             merged_metadata = {**bytestream.meta, **metadata}
-            warnings.warn(
-                "The `store_full_path` parameter defaults to True, storing full file paths in metadata. "
-                "In the 2.9.0 release, the default value for `store_full_path` will change to False, "
-                "storing only file names to improve privacy.",
-                DeprecationWarning,
-            )
 
             if not self.store_full_path and (file_path := bytestream.meta.get("file_path")):
                 merged_metadata["file_path"] = os.path.basename(file_path)

haystack/components/embedders/hugging_face_api_document_embedder.py

@@ -14,7 +14,7 @@
 from haystack.utils.hf import HFEmbeddingAPIType, HFModelType, check_valid_model
 from haystack.utils.url_validation import is_valid_http_url
 
-with LazyImport(message="Run 'pip install \"huggingface_hub>=0.23.0\"'") as huggingface_hub_import:
+with LazyImport(message="Run 'pip install \"huggingface_hub>=0.27.0\"'") as huggingface_hub_import:
     from huggingface_hub import InferenceClient
 
 logger = logging.getLogger(__name__)

haystack/components/embedders/hugging_face_api_text_embedder.py

@@ -11,7 +11,7 @@
 from haystack.utils.hf import HFEmbeddingAPIType, HFModelType, check_valid_model
 from haystack.utils.url_validation import is_valid_http_url
 
-with LazyImport(message="Run 'pip install \"huggingface_hub>=0.23.0\"'") as huggingface_hub_import:
+with LazyImport(message="Run 'pip install \"huggingface_hub>=0.27.0\"'") as huggingface_hub_import:
     from huggingface_hub import InferenceClient
 
 logger = logging.getLogger(__name__)