alita-sdk 0.3.345__py3-none-any.whl → 0.3.346__py3-none-any.whl

alita_sdk/runtime/langchain/document_loaders/AlitaPDFLoader.py

@@ -1,6 +1,6 @@
 import pymupdf
 import fitz
-from langchain_community.document_loaders import PyPDFLoader
+from langchain_community.document_loaders import PyPDFium2Loader
 
 from .ImageParser import ImageParser
 from .utils import perform_llm_prediction_for_image_bytes, create_temp_file
@@ -23,6 +23,7 @@ class AlitaPDFLoader:
         self.headers = kwargs.get('headers', None)
         self.extraction_mode = kwargs.get('extraction_mode', "plain")
         self.extraction_kwargs = kwargs.get('extraction_kwargs', None)
+        self.images_parser=ImageParser(llm=self.llm, prompt=self.prompt)
 
     def get_content(self):
         if hasattr(self, 'file_path'):
@@ -119,13 +120,13 @@ class AlitaPDFLoader:
             return self._load_docs()
 
     def _load_docs(self):
-        docs = PyPDFLoader(file_path=self.file_path,
-                        password=self.password,
-                        headers=self.headers,
-                        extract_images=self.extract_images,
-                        extraction_mode=self.extraction_mode,
-                        images_parser=ImageParser(llm=self.llm, prompt=self.prompt),
-                        extraction_kwargs=self.extraction_kwargs).load()
+        docs = PyPDFium2Loader(
+                file_path = self.file_path,
+                password=self.password,
+                headers=self.headers,
+                extract_images = self.extract_images,
+                images_parser = ImageParser(llm=self.llm, prompt=self.prompt),
+            ).load()
         for doc in docs:
             doc.metadata['chunk_id'] = doc.metadata['page']
         return docs

alita_sdk/runtime/langchain/document_loaders/ImageParser.py

@@ -1,4 +1,8 @@
+from typing import Iterator
+
 from langchain_community.document_loaders.parsers.images import BaseImageBlobParser
+from langchain_core.documents import Document
+from langchain_core.documents.base import Blob
 
 from alita_sdk.runtime.langchain.document_loaders.AlitaImageLoader import AlitaImageLoader
 
@@ -8,10 +12,19 @@ class ImageParser(BaseImageBlobParser):
         self.llm = kwargs.get('llm')
         self.prompt = kwargs.get('prompt')
 
+    def lazy_parse(self, blob: Blob) -> Iterator[Document]:
+        try:
+            yield from super().lazy_parse(blob)
+        except Exception:
+            yield Document(page_content="[Image: Unknown]")
+
     def _analyze_image(self, img) -> str:
         from io import BytesIO
 
         byte_stream = BytesIO()
         img.save(byte_stream, format='PNG')
         image_bytes = byte_stream.getvalue()
-        return AlitaImageLoader(file_content=image_bytes, file_name="image.png", prompt=self.prompt, llm=self.llm).get_content()
+        try:
+            return AlitaImageLoader(file_content=image_bytes, file_name="image.png", prompt=self.prompt, llm=self.llm).get_content()
+        except Exception:
+            return "Image: unknown"

alita_sdk/runtime/tools/artifact.py

@@ -1,7 +1,7 @@
 import hashlib
-import json
 import logging
-from typing import Any, Optional, Generator
+import re
+from typing import Any, Optional, Generator, List
 
 from langchain_core.documents import Document
 from langchain_core.tools import ToolException
@@ -59,18 +59,53 @@ class ArtifactWrapper(NonCodeIndexerToolkit):
     def create_new_bucket(self, bucket_name: str, expiration_measure = "weeks", expiration_value = 1):
         return self.artifact.client.create_bucket(bucket_name, expiration_measure, expiration_value)
 
+    def _index_tool_params(self):
+        return {
+            'include_extensions': (Optional[List[str]], Field(
+                description="List of file extensions to include when processing: i.e. ['*.png', '*.jpg']. "
+                            "If empty, all files will be processed (except skip_extensions).",
+                default=[])),
+            'skip_extensions': (Optional[List[str]], Field(
+                description="List of file extensions to skip when processing: i.e. ['*.png', '*.jpg']",
+                default=[])),
+        }
+
     def _base_loader(self, **kwargs) -> Generator[Document, None, None]:
         try:
-            all_files = self.list_files(self.bucket, False)
+            all_files = self.list_files(self.bucket, False)['rows']
         except Exception as e:
             raise ToolException(f"Unable to extract files: {e}")
 
-        for file in all_files['rows']:
+        include_extensions = kwargs.get('include_extensions', [])
+        skip_extensions = kwargs.get('skip_extensions', [])
+        self._log_tool_event(message=f"Files filtering started. Include extensions: {include_extensions}. "
+                                     f"Skip extensions: {skip_extensions}", tool_name="loader")
+        # show the progress of filtering
+        total_files = len(all_files) if isinstance(all_files, list) else 0
+        filtered_files_count = 0
+        for file in all_files:
+            filtered_files_count += 1
+            if filtered_files_count % 10 == 0 or filtered_files_count == total_files:
+                self._log_tool_event(message=f"Files filtering progress: {filtered_files_count}/{total_files}",
+                                     tool_name="loader")
+            file_name = file['name']
+
+            # Check if file should be skipped based on skip_extensions
+            if any(re.match(pattern.replace('*', '.*') + '$', file_name, re.IGNORECASE)
+                   for pattern in skip_extensions):
+                continue
+
+            # Check if file should be included based on include_extensions
+            # If include_extensions is empty, process all files (that weren't skipped)
+            if include_extensions and not (any(re.match(pattern.replace('*', '.*') + '$', file_name, re.IGNORECASE)
+                                               for pattern in include_extensions)):
+                continue
+
             metadata = {
                 ("updated_on" if k == "modified" else k): str(v)
                 for k, v in file.items()
             }
-            metadata['id'] = self.get_hash_from_bucket_and_file_name(self.bucket, file['name'])
+            metadata['id'] = self.get_hash_from_bucket_and_file_name(self.bucket, file_name)
             yield Document(page_content="", metadata=metadata)
 
     def get_hash_from_bucket_and_file_name(self, bucket, file_name):

alita_sdk/tools/github/github_client.py

@@ -105,7 +105,7 @@ class GitHubClient(BaseModel):
                     self._github_repo_instance = None
             except Exception as e:
                 # Only raise when accessed, not during initialization
-                return ToolException(e)
+                raise ToolException(e)
         return self._github_repo_instance
 
     @model_validator(mode='before')

{alita_sdk-0.3.345.dist-info → alita_sdk-0.3.346.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.345
+Version: 0.3.346
 Summary: SDK for building langchain agents using resources from Alita
 Author-email: Artem Rozumenko <artyom.rozumenko@gmail.com>, Mikalai Biazruchka <mikalai_biazruchka@epam.com>, Roman Mitusov <roman_mitusov@epam.com>, Ivan Krakhmaliuk <lifedj27@gmail.com>, Artem Dubrovskiy <ad13box@gmail.com>
 License-Expression: Apache-2.0

{alita_sdk-0.3.345.dist-info → alita_sdk-0.3.346.dist-info}/RECORD

@@ -62,13 +62,13 @@ alita_sdk/runtime/langchain/document_loaders/AlitaImageLoader.py,sha256=QwgBJE-B
 alita_sdk/runtime/langchain/document_loaders/AlitaJSONLoader.py,sha256=Nav2cgCQKOHQi_ZgYYn_iFdP_Os56KVlVR5nHGXecBc,3445
 alita_sdk/runtime/langchain/document_loaders/AlitaJiraLoader.py,sha256=M2q3YThkps0yAZOjfoLcyE7qycVTYKcXEGtpmp0N6C8,10950
 alita_sdk/runtime/langchain/document_loaders/AlitaMarkdownLoader.py,sha256=RGHDfleYTn7AAc3H-yFZrjm06L0Ux14ZtEJpFlVBNCA,2474
-alita_sdk/runtime/langchain/document_loaders/AlitaPDFLoader.py,sha256=usSrPnYQ3dDOJDdg6gBDTnBJnHiqjLxd_kvOBfRyVxY,5946
+alita_sdk/runtime/langchain/document_loaders/AlitaPDFLoader.py,sha256=olVThKX9Mmv4muTW0cAQBkgeNqU4IcdLVhqpBuzwly4,5904
 alita_sdk/runtime/langchain/document_loaders/AlitaPowerPointLoader.py,sha256=CHIaUnP2Alu7D1NHxlL5N98iY7Gqm4tA5wHjBYUsQLc,2833
 alita_sdk/runtime/langchain/document_loaders/AlitaPythonLoader.py,sha256=m_7aq-aCFVb4vXZsJNinfN1hAuyy_S0ylRknv_ahxDc,340
 alita_sdk/runtime/langchain/document_loaders/AlitaQtestLoader.py,sha256=CUVVnisxm7b5yZWV6rn0Q3MEEaO1GWNcfnz5yWz8T0k,13283
 alita_sdk/runtime/langchain/document_loaders/AlitaTableLoader.py,sha256=nI8lyndVZxVAxbjX3yiqyuFQKFE8MjLPyYSyqRWxHqQ,4077
 alita_sdk/runtime/langchain/document_loaders/AlitaTextLoader.py,sha256=EiCIAF_OxSrbuwgOFk2IpxRMvFbctITt2jAI0g_atpk,3586
-alita_sdk/runtime/langchain/document_loaders/ImageParser.py,sha256=gao5yCCKdDai_Gx7YdEx5U6oMyJYzn69eYmEvWLh-fc,656
+alita_sdk/runtime/langchain/document_loaders/ImageParser.py,sha256=RQ4zGdSw42ec8c6Eb48uFadayWuiT4FbwhGVwhSw60s,1065
 alita_sdk/runtime/langchain/document_loaders/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 alita_sdk/runtime/langchain/document_loaders/constants.py,sha256=xlOXq2iooepcM41SUehbH4ZUFsdz1gWli_7C9Lt5saI,7528
 alita_sdk/runtime/langchain/document_loaders/utils.py,sha256=9xghESf3axBbwxATyVuS0Yu-TWe8zWZnXgCD1ZVyNW0,2414
@@ -106,7 +106,7 @@ alita_sdk/runtime/toolkits/vectorstore.py,sha256=BGppQADa1ZiLO17fC0uCACTTEvPHlod
 alita_sdk/runtime/tools/__init__.py,sha256=7OA8YPKlEOfXu3-gJA08cyR-VymjSPL-OmbXI-B2xVA,355
 alita_sdk/runtime/tools/agent.py,sha256=m98QxOHwnCRTT9j18Olbb5UPS8-ZGeQaGiUyZJSyFck,3162
 alita_sdk/runtime/tools/application.py,sha256=z3vLZODs-_xEEnZFmGF0fKz1j3VtNJxqsAmg5ovExpQ,3129
-alita_sdk/runtime/tools/artifact.py,sha256=2Jjrhuj7Q-Sc5AKkAG7Pk8cJnGPqnqgtOmE3eDOVX0M,8694
+alita_sdk/runtime/tools/artifact.py,sha256=9kNZENeGDK4wW3cG0tixmJb0FDJhO-VqujuuuxN8kDo,10682
 alita_sdk/runtime/tools/datasource.py,sha256=pvbaSfI-ThQQnjHG-QhYNSTYRnZB0rYtZFpjCfpzxYI,2443
 alita_sdk/runtime/tools/echo.py,sha256=spw9eCweXzixJqHnZofHE1yWiSUa04L4VKycf3KCEaM,486
 alita_sdk/runtime/tools/function.py,sha256=0iZJ-UxaPbtcXAVX9G5Vsn7vmD7lrz3cBG1qylto1gs,2844
@@ -239,7 +239,7 @@ alita_sdk/tools/figma/__init__.py,sha256=W6vIMMkZI2Lmpg6_CRRV3oadaIbVI-qTLmKUh6e
 alita_sdk/tools/figma/api_wrapper.py,sha256=KbKet1Xvjq1Vynz_jEE1MtEAVtLYNlSCg67u4dfhe90,33681
 alita_sdk/tools/github/__init__.py,sha256=2rHu0zZyZGnLC5CkHgDIhe14N9yCyaEfrrt7ydH8478,5191
 alita_sdk/tools/github/api_wrapper.py,sha256=uDwYckdnpYRJtb0uZnDkaz2udvdDLVxuCh1tSwspsiU,8411
-alita_sdk/tools/github/github_client.py,sha256=IhTYcqByJ_wnYg2GFkLkYaiG2j8kFkL8p8CTIVZwmqY,86598
+alita_sdk/tools/github/github_client.py,sha256=0YkpD6Zm4X46jMNN57ZIypo2YObtgxCGQokJAF-laFs,86597
 alita_sdk/tools/github/graphql_client_wrapper.py,sha256=d3AGjzLGH_hdQV2V8HeAX92dJ4dlnE5OXqUlCO_PBr0,71539
 alita_sdk/tools/github/schemas.py,sha256=TxEWR3SjDKVwzo9i2tLnss_uPAv85Mh7oWjvQvYLDQE,14000
 alita_sdk/tools/github/tool.py,sha256=Jnnv5lenV5ds8AAdyo2m8hSzyJ117HZBjzHC6T1ck-M,1037
@@ -350,8 +350,8 @@ alita_sdk/tools/zephyr_scale/api_wrapper.py,sha256=kT0TbmMvuKhDUZc0i7KO18O38JM9S
 alita_sdk/tools/zephyr_squad/__init__.py,sha256=0ne8XLJEQSLOWfzd2HdnqOYmQlUliKHbBED5kW_Vias,2895
 alita_sdk/tools/zephyr_squad/api_wrapper.py,sha256=kmw_xol8YIYFplBLWTqP_VKPRhL_1ItDD0_vXTe_UuI,14906
 alita_sdk/tools/zephyr_squad/zephyr_squad_cloud_client.py,sha256=R371waHsms4sllHCbijKYs90C-9Yu0sSR3N4SUfQOgU,5066
-alita_sdk-0.3.345.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-alita_sdk-0.3.345.dist-info/METADATA,sha256=xKGJO9ArLAkIHbt6Ow6scbFIqtp0cqbqca2NPHVk6ao,19015
-alita_sdk-0.3.345.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-alita_sdk-0.3.345.dist-info/top_level.txt,sha256=0vJYy5p_jK6AwVb1aqXr7Kgqgk3WDtQ6t5C-XI9zkmg,10
-alita_sdk-0.3.345.dist-info/RECORD,,
+alita_sdk-0.3.346.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+alita_sdk-0.3.346.dist-info/METADATA,sha256=OoUbeD3TLR5rkU_L-5H3DOb2tB7yJ9JWqmwAjDpYq_E,19015
+alita_sdk-0.3.346.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+alita_sdk-0.3.346.dist-info/top_level.txt,sha256=0vJYy5p_jK6AwVb1aqXr7Kgqgk3WDtQ6t5C-XI9zkmg,10
+alita_sdk-0.3.346.dist-info/RECORD,,