alita-sdk 0.3.316__tar.gz → 0.3.318__tar.gz

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.316
+Version: 0.3.318
 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-0.3.316 → alita_sdk-0.3.318}/alita_sdk/runtime/langchain/document_loaders/AlitaExcelLoader.py

@@ -65,7 +65,7 @@ class AlitaExcelLoader(AlitaTableLoader):
         else:
             return df.to_string(index=False)
 
-    def read(self):
+    def read(self, lazy: bool = False):
         df = pd.read_excel(self.file_path, sheet_name=None)
         docs = []
         for key in df.keys():

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/alita_sdk/runtime/langchain/document_loaders/constants.py

@@ -113,7 +113,7 @@ loaders_map = {
         'is_multimodal_processing': False,
         'kwargs': {
             'encoding': 'utf-8',
-            'raw_content': False,
+            'raw_content': True,
             'cleanse': False
         },
         'allowed_to_override': ['max_tokens']
@@ -122,7 +122,7 @@ loaders_map = {
         'class': AlitaExcelLoader,
         'is_multimodal_processing': False,
         'kwargs': {
-            'raw_content': False,
+            'raw_content': True,
             'cleanse': False
         },
         'allowed_to_override': ['max_tokens', LoaderProperties.LLM.value, LoaderProperties.PROMPT.value, LoaderProperties.PROMPT_DEFAULT.value]
@@ -131,7 +131,7 @@ loaders_map = {
         'class': AlitaExcelLoader,
         'is_multimodal_processing': False,
         'kwargs': {
-            'raw_content': False,
+            'raw_content': True,
             'cleanse': False
         },
         'allowed_to_override': ['max_tokens', LoaderProperties.LLM.value, LoaderProperties.PROMPT.value, LoaderProperties.PROMPT_DEFAULT.value]

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/alita_sdk/runtime/tools/vectorstore.py

@@ -12,7 +12,6 @@ from alita_sdk.tools.vector_adapters.VectorStoreAdapter import VectorStoreAdapte
 from logging import getLogger
 
 from ..utils.logging import dispatch_custom_event
-from ..utils.utils import IndexerKeywords
 
 logger = getLogger(__name__)
 
@@ -248,10 +247,6 @@ class VectorStoreWrapper(BaseToolApiWrapper):
             tool_name="_clean_collection"
         )
 
-    def _get_indexed_data(self, collection_name: str):
-        """ Get all indexed data from vectorstore for non-code content """
-        return self.vector_adapter.get_indexed_data(self, collection_name)
-
     def _get_code_indexed_data(self, collection_suffix: str) -> Dict[str, Dict[str, Any]]:
         """ Get all indexed data from vectorstore for code content """
         return self.vector_adapter.get_code_indexed_data(self, collection_suffix)
@@ -308,26 +303,6 @@ class VectorStoreWrapper(BaseToolApiWrapper):
 
         return final_docs
 
-    def _reduce_non_code_duplicates(self, documents: Generator[Any, None, None], collection_suffix: str) -> List[Any]:
-        return self._reduce_duplicates(
-            documents,
-            collection_suffix,
-            self._get_indexed_data,
-            lambda doc: doc.metadata.get('id'),
-            lambda doc, idx: (
-                    doc.metadata.get('updated_on') and
-                    idx['metadata'].get('updated_on') and
-                    doc.metadata.get('updated_on') == idx['metadata'].get('updated_on')
-            ),
-            lambda idx_data, key: (
-                    idx_data[key]['all_chunks'] +
-                    [idx_data[dep_id]['id'] for dep_id in idx_data[key][IndexerKeywords.DEPENDENT_DOCS.value]] +
-                    [chunk_db_id for dep_id in idx_data[key][IndexerKeywords.DEPENDENT_DOCS.value]
-                     for chunk_db_id in idx_data[dep_id]['all_chunks']]
-            ),
-            log_msg="Verification of documents to index started"
-        )
-
     def _reduce_code_duplicates(self, documents: Generator[Any, None, None], collection_suffix: str) -> List[Any]:
         return self._reduce_duplicates(
             documents,
@@ -343,7 +318,7 @@ class VectorStoreWrapper(BaseToolApiWrapper):
             log_msg="Verification of code documents to index started"
         )
 
-    def index_documents(self, documents: Generator[Document, None, None], collection_suffix: str, progress_step: int = 20, clean_index: bool = True, is_code: bool = False):
+    def index_documents(self, documents: Generator[Document, None, None], collection_suffix: str, progress_step: int = 20, clean_index: bool = True, is_code: bool = True):
         """ Index documents in the vectorstore.
 
         Args:
@@ -374,8 +349,7 @@ class VectorStoreWrapper(BaseToolApiWrapper):
                 message="Filter for duplicates",
                 tool_name="index_documents")
             # remove duplicates based on metadata 'id' and 'updated_on' or 'commit_hash' fields
-            documents = self._reduce_code_duplicates(documents, collection_suffix) if is_code \
-                else self._reduce_non_code_duplicates(documents, collection_suffix)
+            documents = self._reduce_code_duplicates(documents, collection_suffix)
             self._log_tool_event(
                 message="All the duplicates were filtered out. Proceeding with indexing.",
                 tool_name="index_documents")

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/alita_sdk/runtime/tools/vectorstore_base.py

@@ -10,7 +10,6 @@ from pydantic import BaseModel, model_validator, Field
 
 from alita_sdk.tools.elitea_base import BaseToolApiWrapper
 from alita_sdk.tools.vector_adapters.VectorStoreAdapter import VectorStoreAdapterFactory
-from ..langchain.tools.vector import VectorAdapter
 from ..utils.logging import dispatch_custom_event
 
 logger = getLogger(__name__)
@@ -138,10 +137,7 @@ class VectorStoreWrapperBase(BaseToolApiWrapper):
     vectorstore_params: Optional[dict]  = None
     max_docs_per_add: int = 100
     dataset: str = None
-    embedding: Any = None
     vectorstore: Any = None
-    # Review usage of old adapter
-    vectoradapter: Any = None
     pg_helper: Any = None
     embeddings: Any = None
     # New adapter for vector database operations
@@ -152,17 +148,13 @@ class VectorStoreWrapperBase(BaseToolApiWrapper):
     def validate_toolkit(cls, values):
         from ..langchain.interfaces.llm_processor import get_vectorstore
         logger.debug(f"Validating toolkit: {values}")
-        if 'vectorstore_params' in values:
-            values["dataset"] = values.get('vectorstore_params').get('collection_name')
-        if values.get('embedding_model'):
-            values['embeddings'] = values['alita'].get_embeddings(values['embedding_model'])
+        values["dataset"] = values.get('collection_name')
+
+        if values.get('alita') and values.get('embedding_model'):
+            values['embeddings'] = values.get('alita').get_embeddings(values.get('embedding_model'))
+
         if values.get('vectorstore_type') and values.get('vectorstore_params') and values.get('embedding_model'):
             values['vectorstore'] = get_vectorstore(values['vectorstore_type'], values['vectorstore_params'], embedding_func=values['embeddings'])
-            values['vectoradapter'] = VectorAdapter(
-                vectorstore=values['vectorstore'],
-                embeddings=values['embeddings'],
-                quota_params=None,
-            )
             # Initialize the new vector adapter
             values['vector_adapter'] = VectorStoreAdapterFactory.create_adapter(values['vectorstore_type'])
             logger.debug(f"Vectorstore wrapper initialized: {values}")
@@ -223,8 +215,6 @@ class VectorStoreWrapperBase(BaseToolApiWrapper):
         self._log_data("Cleaning index before re-indexing all documents. Previous index will be removed", tool_name="index_documents")
         try:
             self._clean_collection(collection_suffix)
-            self.vectoradapter.persist()
-            self.vectoradapter.vacuum()
             self._log_data("Previous index has been removed",
                            tool_name="index_documents")
         except Exception as e:
@@ -238,7 +228,6 @@ class VectorStoreWrapperBase(BaseToolApiWrapper):
                 logger.warning(f"Document is missing required metadata field 'id' or 'updated_on': {doc.metadata}")
 
         logger.debug(f"Indexing documents: {documents}")
-        logger.debug(self.vectoradapter)
 
         # if collection_suffix is provided, add it to metadata of each document
         if collection_suffix:
@@ -256,6 +245,10 @@ class VectorStoreWrapperBase(BaseToolApiWrapper):
         progress_step = 20 if progress_step not in range(0, 100) else progress_step
         next_progress_point = progress_step
         for document in documents:
+            if not document.page_content:
+                # To avoid case when all documents have empty content
+                # See llm_processor.add_documents which exclude metadata of docs with empty content
+                continue
             documents_count += 1
             # logger.debug(f"Indexing document: {document}")
             try:

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/alita_sdk/tools/base_indexer_toolkit.py

@@ -5,7 +5,7 @@ from typing import Any, Optional, List, Literal, Dict, Generator
 from langchain_core.documents import Document
 from pydantic import create_model, Field, SecretStr
 
-from .utils.content_parser import file_extension_by_chunker, process_content_by_type
+from .utils.content_parser import file_extension_by_chunker, process_document_by_type
 from .vector_adapters.VectorStoreAdapter import VectorStoreAdapterFactory
 from ..runtime.tools.vectorstore_base import VectorStoreWrapperBase
 from ..runtime.utils.utils import IndexerKeywords
@@ -102,7 +102,6 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
 
     connection_string: Optional[SecretStr] = None
     collection_name: Optional[str] = None
-    _embedding: Optional[Any] = None
     alita: Any = None # Elitea client, if available
 
     def __init__(self, **kwargs):
@@ -116,7 +115,6 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
         if connection_string:
             # Initialize vectorstore params only if connection string is provided
             kwargs['vectorstore_params'] = VectorStoreAdapterFactory.create_adapter(vectorstore_type).get_vectorstore_params(collection_name, connection_string)
-            kwargs['_embedding'] = kwargs.get('alita').get_embeddings(kwargs.get('embedding_model'))
         super().__init__(**kwargs)
 
     def _index_tool_params(self, **kwargs) -> dict[str, tuple[type, Field]]:
@@ -181,14 +179,14 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
 
         if chunking_config is None:
             chunking_config = {}
-        chunking_config['embedding'] = self._embedding
+        chunking_config['embedding'] = self.embeddings
         chunking_config['llm'] = self.llm
 
         for document in documents:
             if content_type := document.metadata.get(IndexerKeywords.CONTENT_FILE_NAME.value, None):
                 # apply parsing based on content type and chunk if chunker was applied to parent doc
                 content = document.metadata.pop(IndexerKeywords.CONTENT_IN_BYTES.value, None)
-                yield from process_content_by_type(
+                yield from process_document_by_type(
                     document=document,
                     content=content,
                     extension_source=content_type, llm=self.llm, chunking_config=chunking_config)
@@ -199,7 +197,7 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
                     continue
                 # apply parsing based on content type resolved from chunking_tool
                 content_type = file_extension_by_chunker(chunking_tool)
-                yield from process_content_by_type(
+                yield from process_document_by_type(
                     document=document,
                     content=content_in_bytes,
                     extension_source=content_type, llm=self.llm, chunking_config=chunking_config)

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/alita_sdk/tools/browser/__init__.py

@@ -112,7 +112,8 @@ class BrowserToolkit(BaseToolkit):
             elif tool == 'google':
                 try:
                     google_api_wrapper = GoogleSearchAPIWrapper(
-                        **wrapper_payload_google
+                        google_api_key=wrapper_payload_google.get('google_api_key'),
+                        google_cse_id=wrapper_payload_google.get('google_cse_id')
                     )
                     tool_entry = GoogleSearchResults(api_wrapper=google_api_wrapper)
                     # rename the tool to avoid conflicts

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/alita_sdk/tools/carrier/backend_reports_tool.py

@@ -154,6 +154,7 @@ class CreateExcelReportTool(BaseTool):
             "tp_threshold": (int, Field(default=None, description="Throughput threshold")),
             "rt_threshold": (int, Field(default=None, description="Response time threshold")),
             "er_threshold": (int, Field(default=None, description="Error rate threshold")),
+            "include_group_pauses": (bool, Field(default=False, description="Include group pauses in Gatling Excel report")),
         }
     )
 
@@ -200,6 +201,7 @@ class CreateExcelReportTool(BaseTool):
             "tp_threshold": 10,
             "rt_threshold": 500,
             "er_threshold": 5,
+            "include_group_pauses": False,
         }
 
     def _request_parameter_confirmation(self, default_parameters):
@@ -217,7 +219,8 @@ class CreateExcelReportTool(BaseTool):
         excel_report_file_name = f'/tmp/reports_test_results_{report["build_id"]}_excel_report.xlsx'
         bucket_name = report["name"].replace("_", "").replace(" ", "").lower()
 
-        result_stats_j = self._parse_report(test_log_file_path, lg_type, parameters["think_time"], is_absolute_file_path=True)
+        result_stats_j = self._parse_report(test_log_file_path, lg_type, parameters["think_time"],
+                                            parameters["include_group_pauses"], is_absolute_file_path=True)
         calc_thr_j = self._calculate_thresholds(result_stats_j, parameters)
 
         return self._generate_and_upload_report(
@@ -233,21 +236,22 @@ class CreateExcelReportTool(BaseTool):
         excel_report_file_name = f'{file_path}_{current_date}.xlsx'
         bucket_name = bucket
 
-        result_stats_j = self._parse_report(file_path, lg_type, parameters["think_time"], is_absolute_file_path=True)
+        result_stats_j = self._parse_report(file_path, lg_type, parameters["think_time"],
+                                            parameters["include_group_pauses"], is_absolute_file_path=True)
         calc_thr_j = self._calculate_thresholds(result_stats_j, parameters)
 
         return self._generate_and_upload_report(
             result_stats_j, carrier_report, calc_thr_j, parameters, excel_report_file_name, bucket_name, file_path
         )
 
-    def _parse_report(self, file_path, lg_type, think_time, is_absolute_file_path=False):
+    def _parse_report(self, file_path, lg_type, think_time, include_group_pauses, is_absolute_file_path=False):
         """Parse the report based on its type."""
         if lg_type == "gatling":
             if is_absolute_file_path:
                 report_file = file_path
             else:
                 report_file = get_latest_log_file(file_path, "simulation.log")
-            parser = GatlingReportParser(report_file, think_time)
+            parser = GatlingReportParser(report_file, include_group_pauses, think_time)
             result_stats_j = parser.parse()
             result_stats_j["requests"].update(result_stats_j["groups"])
         elif lg_type == "jmeter":

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/alita_sdk/tools/carrier/excel_reporter.py

@@ -118,9 +118,10 @@ class JMeterReportParser(PerformanceReportParser):
 
 class GatlingReportParser(PerformanceReportParser):
 
-    def __init__(self, log_file: str, think_times="5,0-10,0"):
+    def __init__(self, log_file: str, include_group_pauses, think_times="5,0-10,0"):
         self.calculated_think_time = think_times
         self.log_file = log_file
+        self.include_group_pauses = include_group_pauses
 
     @staticmethod
     def convert_timestamp_to_datetime(timestamp: int) -> datetime:
@@ -210,7 +211,7 @@ class GatlingReportParser(PerformanceReportParser):
                                 ramp_end = self.convert_timestamp_to_datetime(int(line.split('\t')[3]))
 
                     elif line.startswith('GROUP'):
-                        self.parse_group_line(groups, line)
+                        self.parse_group_line(groups, line, self.include_group_pauses)
         except FileNotFoundError as e:
             print(f"File not found: {e}")
             raise
@@ -242,11 +243,14 @@ class GatlingReportParser(PerformanceReportParser):
             requests[request_name].append((response_time, status))
 
     @staticmethod
-    def parse_group_line(groups, line):
+    def parse_group_line(groups, line, include_group_pauses):
         parts = line.split('\t')
         if len(parts) >= 6:
             group_name = parts[1]
-            response_time = int(parts[4])
+            if include_group_pauses:
+                response_time = int(parts[3]) - int(parts[2])
+            else:
+                response_time = int(parts[4])
             status = parts[5].strip()
             groups[group_name].append((response_time, status))
 

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/alita_sdk/tools/confluence/api_wrapper.py

@@ -1,24 +1,25 @@
-import re
-import logging
-import requests
-import json
 import base64
+import json
+import logging
+import re
 import traceback
-from typing import Optional, List, Any, Dict, Callable, Generator, Literal
 from json import JSONDecodeError
+from typing import Optional, List, Any, Dict, Callable, Generator, Literal
 
-from pydantic import Field, PrivateAttr, model_validator, create_model, SecretStr
-from tenacity import retry, stop_after_attempt, wait_exponential, before_sleep_log
-
+import requests
+from langchain_community.document_loaders.confluence import ContentFormat
 from langchain_core.documents import Document
-from langchain_core.tools import ToolException
 from langchain_core.messages import HumanMessage
+from langchain_core.tools import ToolException
 from markdownify import markdownify
-from langchain_community.document_loaders.confluence import ContentFormat
+from pydantic import Field, PrivateAttr, model_validator, create_model, SecretStr
+from tenacity import retry, stop_after_attempt, wait_exponential, before_sleep_log
 
-from ..elitea_base import BaseVectorStoreToolApiWrapper, extend_with_vector_tools
+from alita_sdk.tools.non_code_indexer_toolkit import NonCodeIndexerToolkit
+from alita_sdk.tools.utils.available_tools_decorator import extend_with_parent_available_tools
 from ..llm.img_utils import ImageDescriptionCache
 from ..utils import is_cookie_token, parse_cookie_string
+from ...runtime.utils.utils import IndexerKeywords
 
 logger = logging.getLogger(__name__)
 
@@ -171,7 +172,7 @@ def parse_payload_params(params: Optional[str]) -> Dict[str, Any]:
     return {}
 
 
-class ConfluenceAPIWrapper(BaseVectorStoreToolApiWrapper):
+class ConfluenceAPIWrapper(NonCodeIndexerToolkit):
     # Changed from PrivateAttr to Optional field with exclude=True
     client: Optional[Any] = Field(default=None, exclude=True)
     base_url: str
@@ -229,7 +230,7 @@ class ConfluenceAPIWrapper(BaseVectorStoreToolApiWrapper):
             client_instance._update_header(header, value)
 
         values['client'] = client_instance
-        return values
+        return super().validate_toolkit(values)
 
     def __unquote_confluence_space(self) -> str | None:
         if self.space:
@@ -843,13 +844,68 @@ class ConfluenceAPIWrapper(BaseVectorStoreToolApiWrapper):
             yield document
 
     def _process_document(self, document: Document) -> Generator[Document, None, None]:
-        attachments = self.get_page_attachments(document.metadata.get('id'))
-        if isinstance(attachments, str):
-            logger.info(f" {document.metadata.get('id')}: {attachments}")
-            return
-        for attachment in attachments:
-            # TODO need to retrive content from other fields/sources if both attachment.get('content', '') and attachment.get('llm_analysis', '') are None
-            yield Document(page_content=attachment.get('content', '') or attachment.get('llm_analysis', '') or '', metadata=attachment.get('metadata', {}))
+        try:
+            page_id = document.metadata.get('id')
+            attachments = self.client.get_attachments_from_content(page_id)
+            if not attachments or not attachments.get('results'):
+                return f"No attachments found for page ID {page_id}."
+
+            # Get attachment history for created/updated info
+            history_map = {}
+            for attachment in attachments['results']:
+                try:
+                    hist = self.client.history(attachment['id'])
+                    history_map[attachment['id']] = hist
+                except Exception as e:
+                    logger.warning(f"Failed to fetch history for attachment {attachment.get('title', '')}: {str(e)}")
+                    history_map[attachment['id']] = None
+
+            import re
+            for attachment in attachments['results']:
+                title = attachment.get('title', '')
+                file_ext = title.lower().split('.')[-1] if '.' in title else ''
+
+                media_type = attachment.get('metadata', {}).get('mediaType', '')
+                # Core metadata extraction with history
+                hist = history_map.get(attachment['id']) or {}
+                created_by = hist.get('createdBy', {}).get('displayName', '') if hist else attachment.get('creator', {}).get('displayName', '')
+                created_date = hist.get('createdDate', '') if hist else attachment.get('created', '')
+                last_updated = hist.get('lastUpdated', {}).get('when', '') if hist else ''
+
+                metadata = {
+                    'name': title,
+                    'size': attachment.get('extensions', {}).get('fileSize', None),
+                    'creator': created_by,
+                    'created': created_date,
+                    'updated': last_updated,
+                    'media_type': media_type,
+                    'labels': [label['name'] for label in
+                               attachment.get('metadata', {}).get('labels', {}).get('results', [])],
+                    'download_url': self.base_url.rstrip('/') + attachment['_links']['download'] if attachment.get(
+                        '_links', {}).get('download') else None
+                }
+
+                download_url = self.base_url.rstrip('/') + attachment['_links']['download']
+
+                try:
+                    resp = self.client.request(method="GET", path=download_url[len(self.base_url):], advanced_mode=True)
+                    if resp.status_code == 200:
+                        content = resp.content
+                    else:
+                        content = f"[Failed to download {download_url}: HTTP status code {resp.status_code}]"
+                except Exception as e:
+                    content = f"[Error downloading content: {str(e)}]"
+
+                if isinstance(content, str):
+                    yield Document(page_content=content, metadata=metadata)
+                else:
+                    yield Document(page_content="", metadata={
+                        **metadata,
+                        IndexerKeywords.CONTENT_FILE_NAME.value: f".{file_ext}",
+                        IndexerKeywords.CONTENT_IN_BYTES.value: content
+                    })
+        except Exception as e:
+            yield from ()
 
     def _download_image(self, image_url):
         """
@@ -1598,7 +1654,7 @@ class ConfluenceAPIWrapper(BaseVectorStoreToolApiWrapper):
             "bins_with_llm": (Optional[bool], Field(description="Use LLM for processing binary files.", default=False)),
         }
 
-    @extend_with_vector_tools
+    @extend_with_parent_available_tools
     def get_available_tools(self):
         # Confluence-specific tools
         confluence_tools = [

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/alita_sdk/tools/jira/api_wrapper.py

@@ -13,12 +13,11 @@ from langchain_core.tools import ToolException
 from pydantic import Field, PrivateAttr, model_validator, create_model, SecretStr
 import requests
 
-from ..elitea_base import BaseVectorStoreToolApiWrapper, extend_with_vector_tools
 from ..llm.img_utils import ImageDescriptionCache
 from ..non_code_indexer_toolkit import NonCodeIndexerToolkit
 from ..utils import is_cookie_token, parse_cookie_string
 from ..utils.available_tools_decorator import extend_with_parent_available_tools
-from ..utils.content_parser import file_extension_by_chunker
+from ..utils.content_parser import file_extension_by_chunker, process_content_by_type
 from ...runtime.utils.utils import IndexerKeywords
 
 logger = logging.getLogger(__name__)
@@ -133,6 +132,13 @@ GetRemoteLinks = create_model(
     jira_issue_key=(str, Field(description="Jira issue key from which remote links will be extracted, e.g. TEST-1234"))
 )
 
+GetIssueAttachments = create_model(
+    "GetIssueAttachments",
+    jira_issue_key=(str, Field(description="Jira issue key from which remote links will be extracted, e.g. TEST-1234")),
+    attachment_pattern=(Optional[str], Field(description="Regex pattern to filter attachment filenames. If not provided,"
+                                                         " all attachments will be processed", default=None))
+)
+
 ListCommentsInput = create_model(
     "ListCommentsInputModel",
     issue_key=(str, Field(description="The issue key of the Jira issue from which comments will be extracted, e.g. 'TEST-123'."))
@@ -726,19 +732,37 @@ class JiraApiWrapper(NonCodeIndexerToolkit):
             logger.error(f"Error listing Jira projects: {stacktrace}")
             return ToolException(f"Error listing Jira projects: {stacktrace}")
 
-    def get_attachments_content(self, jira_issue_key: str):
-        """ Extract content of all attachments related to specified Jira issue key.
-         NOTE: only parsable attachments will be considered """
+    def get_attachments_content(self, jira_issue_key: str, attachment_pattern: Optional[str] = None):
+        """ Extract the content of all attachments related to a specified Jira issue key.
+         NOTE: only parsable attachments will be considered
+         Args:
+            jira_issue_key: The key of the Jira issue, e.g. "TEST-123
+            attachment_pattern: Optional regex pattern to filter attachments by filename.
+            If provided, only attachments with filenames matching this pattern will be processed.
+         Returns:
+            A string containing the content of all relevant attachments, separated by double newlines.
+         """
 
         attachment_data = []
         attachments = self._client.get_attachments_ids_from_issue(issue=jira_issue_key)
         for attachment in attachments:
+            if attachment_pattern and not re.search(attachment_pattern, attachment['filename']):
+                logger.info(f"Skipping attachment {attachment['filename']} as it does not match pattern {attachment_pattern}")
+                continue
+            logger.info(f"Processing attachment {attachment['filename']} with ID {attachment['attachment_id']}")
             if self.api_version == "3":
                 attachment_data.append(self._client.get_attachment_content(attachment['attachment_id']))
             else:
-                extracted_attachment = self._client.get_attachment(attachment_id=attachment['attachment_id'])
-                if extracted_attachment['mimeType'] in SUPPORTED_ATTACHMENT_MIME_TYPES:
-                    attachment_data.append(self._extract_attachment_content(extracted_attachment))
+                try:
+                    attachment_content = self._client.get_attachment_content(attachment['attachment_id'])
+                except Exception as e:
+                    logger.error(
+                        f"Failed to download attachment {attachment['filename']} for issue {jira_issue_key}: {str(e)}")
+                    attachment_content = self._client.get(
+                        path=f"secure/attachment/{attachment['attachment_id']}/{attachment['filename']}", not_json_response=True)
+                content_docs = process_content_by_type(attachment_content, attachment['filename'], llm=self.llm)
+                attachment_data.append("filename: " + attachment['filename'] + "\ncontent: " + str([doc.page_content for doc in content_docs]))
+
         return "\n\n".join(attachment_data)
 
     def execute_generic_rq(self, method: str, relative_url: str, params: Optional[str] = "", *args):
@@ -1633,7 +1657,7 @@ class JiraApiWrapper(NonCodeIndexerToolkit):
             {
                 "name": "get_attachments_content",
                 "description": self.get_attachments_content.__doc__,
-                "args_schema": GetRemoteLinks,
+                "args_schema": GetIssueAttachments,
                 "ref": self.get_attachments_content,
             },
             {

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/alita_sdk/tools/utils/content_parser.py

@@ -201,17 +201,40 @@ def load_content_from_bytes(file_content: bytes, extension: str = None, loader_e
         if temp_file_path and os.path.exists(temp_file_path):
             os.remove(temp_file_path)
 
-def process_content_by_type(document: Document, content, extension_source: str, llm = None, chunking_config=None) -> Generator[Document, None, None]:
+def process_document_by_type(content, extension_source: str, document: Document = None, llm = None, chunking_config=None) \
+        -> Generator[Document, None, None]:
+    """Process the content of a file based on its type using a configured loader cosidering the origin document."""
+    try:
+        chunks = process_content_by_type(content, extension_source, llm, chunking_config)
+    except Exception as e:
+        msg = f"Error during content for file {extension_source}:\n{e}"
+        logger.warning(msg)
+        yield Document(
+            page_content=msg,
+            metadata={**document.metadata, 'chunk_id': 1}
+        )
+        return
+    for chunk in chunks:
+        yield Document(
+            page_content=sanitize_for_postgres(chunk.page_content),
+            metadata={**document.metadata, **chunk.metadata}
+                )
+
+
+def process_content_by_type(content, filename: str, llm=None, chunking_config=None) -> \
+        Generator[Document, None, None]:
+    """Process the content of a file based on its type using a configured loader."""
     temp_file_path = None
     try:
-        extension = "." + extension_source.split('.')[-1].lower()
+        extension = "." + filename.split('.')[-1].lower()
 
         with tempfile.NamedTemporaryFile(mode='w+b', suffix=extension, delete=False) as temp_file:
             temp_file_path = temp_file.name
             if content is None:
-                logger.warning(f"'{IndexerKeywords.CONTENT_IN_BYTES.value}' ie expected but not found in document metadata.")
+                logger.warning(
+                    f"'{IndexerKeywords.CONTENT_IN_BYTES.value}' ie expected but not found in document metadata.")
                 return
-            
+
             temp_file.write(content)
             temp_file.flush()
 
@@ -238,21 +261,7 @@ def process_content_by_type(document: Document, content, extension_source: str,
                 loader_kwargs.pop(LoaderProperties.PROMPT_DEFAULT.value)
                 loader_kwargs[LoaderProperties.PROMPT.value] = image_processing_prompt
             loader = loader_cls(file_path=temp_file_path, **loader_kwargs)
-            try:
-                chunks = loader.load()
-            except Exception as e:
-                msg = f"Error during content for file {temp_file_path}:\n{e}"
-                logger.warning(msg)
-                yield Document(
-                    page_content=msg,
-                    metadata={**document.metadata, 'chunk_id':1}
-                )
-                return
-            for chunk in chunks:
-                yield Document(
-                    page_content=sanitize_for_postgres(chunk.page_content),
-                    metadata={**document.metadata, **chunk.metadata}
-                )
+            return loader.load()
     finally:
         if temp_file_path and os.path.exists(temp_file_path):
             os.remove(temp_file_path)

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/alita_sdk/tools/xray/api_wrapper.py

@@ -9,14 +9,9 @@ from langchain_core.tools import ToolException
 from pydantic import PrivateAttr, SecretStr, create_model, model_validator, Field
 from python_graphql_client import GraphqlClient
 
-from ..elitea_base import (
-    BaseVectorStoreToolApiWrapper,
-    extend_with_vector_tools,
-)
 from ..non_code_indexer_toolkit import NonCodeIndexerToolkit
 from ..utils.available_tools_decorator import extend_with_parent_available_tools
 from ...runtime.utils.utils import IndexerKeywords
-from ..utils.content_parser import file_extension_by_chunker
 
 try:
     from alita_sdk.runtime.langchain.interfaces.llm_processor import get_embeddings

{alita_sdk-0.3.316 → alita_sdk-0.3.318}/alita_sdk.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.316
+Version: 0.3.318
 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