alita-sdk 0.3.342__tar.gz → 0.3.344__tar.gz

{alita_sdk-0.3.342 → alita_sdk-0.3.344}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.342
+Version: 0.3.344
 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-0.3.342 → alita_sdk-0.3.344}/alita_sdk/runtime/clients/client.py

@@ -349,7 +349,8 @@ class AlitaClient:
         return self._process_requst(resp)
 
     def list_artifacts(self, bucket_name: str):
-        url = f'{self.artifacts_url}/{bucket_name}'
+        # Ensure bucket name is lowercase as required by the API
+        url = f'{self.artifacts_url}/{bucket_name.lower()}'
         data = requests.get(url, headers=self.headers, verify=False)
         return self._process_requst(data)
 

{alita_sdk-0.3.342 → alita_sdk-0.3.344}/alita_sdk/runtime/langchain/document_loaders/constants.py

@@ -14,7 +14,7 @@
 
 from langchain_community.document_loaders import (
     AirbyteJSONLoader, UnstructuredHTMLLoader,
-    PythonLoader)
+    PythonLoader, UnstructuredXMLLoader)
 
 from .AlitaCSVLoader import AlitaCSVLoader
 from .AlitaDocxMammothLoader import AlitaDocxMammothLoader
@@ -176,6 +176,12 @@ loaders_map = {
         'kwargs': {},
         'allowed_to_override': ['max_tokens', LoaderProperties.LLM.value, LoaderProperties.PROMPT.value, LoaderProperties.PROMPT_DEFAULT.value]
     },
+    '.xml': {
+        'class': UnstructuredXMLLoader,
+        'is_multimodal_processing': False,
+        'kwargs': {},
+        'allowed_to_override': ['max_tokens', LoaderProperties.LLM.value, LoaderProperties.PROMPT.value, LoaderProperties.PROMPT_DEFAULT.value]
+    },
     '.ppt': {
         'class': AlitaPowerPointLoader,
         'is_multimodal_processing': False,

{alita_sdk-0.3.342 → alita_sdk-0.3.344}/alita_sdk/runtime/langchain/langraph_agent.py

@@ -455,14 +455,11 @@ def create_graph(
                                 input_variables=node.get('input', ['messages'])))
                         elif node_type == 'agent':
                             input_params = node.get('input', ['messages'])
-                            input_mapping = {'task': {'type': 'fstring', 'value': f"{node.get('task', '')}"},
-                                                      'chat_history': {'type': 'fixed', 'value': []}}
-                            # Add 'chat_history' to input_mapping only if 'messages' is in input_params
-                            if 'messages' in input_params:
-                                input_mapping['chat_history'] = {'type': 'variable', 'value': 'messages'}
+                            input_mapping = node.get('input_mapping',
+                                                       {'messages': {'type': 'variable', 'value': 'messages'}})
                             lg_builder.add_node(node_id, FunctionTool(
                                 client=client, tool=tool,
-                                name=node['id'], return_type='dict',
+                                name=node['id'], return_type='str',
                                 output_variables=node.get('output', []),
                                 input_variables=input_params,
                                 input_mapping= input_mapping
@@ -638,19 +635,11 @@ def create_graph(
             entry_point = clean_string(schema['entry_point'])
         except KeyError:
             raise ToolException("Entry point is not defined in the schema. Please define 'entry_point' in the schema.")
-        # if state.items():
-        #     state_default_node = StateDefaultNode(default_vars=set_defaults(state))
-        #     lg_builder.add_node(state_default_node.name, state_default_node)
-        #     lg_builder.set_entry_point(state_default_node.name)
-        #     lg_builder.add_conditional_edges(state_default_node.name, TransitionalEdge(entry_point))
-        for key, value in state.items():
-            if 'type' in value and 'value' in value:
-                # set default value for state variable if it is defined in the schema
-                state_default_node = StateDefaultNode(default_vars=state)
-                lg_builder.add_node(state_default_node.name, state_default_node)
-                lg_builder.set_entry_point(state_default_node.name)
-                lg_builder.add_conditional_edges(state_default_node.name, TransitionalEdge(entry_point))
-                break
+        if state.items():
+            state_default_node = StateDefaultNode(default_vars=set_defaults(state))
+            lg_builder.add_node(state_default_node.name, state_default_node)
+            lg_builder.set_entry_point(state_default_node.name)
+            lg_builder.add_conditional_edges(state_default_node.name, TransitionalEdge(entry_point))
         else:
             # if no state variables are defined, set the entry point directly
             lg_builder.set_entry_point(entry_point)
@@ -757,6 +746,8 @@ class LangGraphAgentRunnable(CompiledStateGraph):
         # Append current input to existing messages instead of overwriting
         if input.get('input'):
             current_message = input.get('input')[-1]
+            # TODO: add handler after we add 2+ inputs (filterByType, etc.)
+            input['input'] = current_message if isinstance(current_message, str) else str(current_message)
             if input.get('messages'):
                 # Ensure existing messages are LangChain objects
                 input['messages'] = [convert_dict_to_message(msg) for msg in input['messages']]

{alita_sdk-0.3.342 → alita_sdk-0.3.344}/alita_sdk/runtime/toolkits/artifact.py

@@ -6,7 +6,7 @@ from langchain_core.tools import BaseTool
 from pydantic import create_model, BaseModel, ConfigDict, Field
 from pydantic.fields import FieldInfo
 from ..tools.artifact import ArtifactWrapper
-from alita_sdk.tools.base.tool import BaseAction
+from ...tools.base.tool import BaseAction
 from ...configurations.pgvector import PgVectorConfiguration
 
 

{alita_sdk-0.3.342 → alita_sdk-0.3.344}/alita_sdk/runtime/tools/application.py

@@ -1,9 +1,9 @@
 import json
 
 from ..utils.utils import clean_string
-from langchain_core.tools import BaseTool
-from langchain_core.messages import BaseMessage, AIMessage, ToolCall
-from typing import Any, Type, Optional, Union
+from langchain_core.tools import BaseTool, ToolException
+from langchain_core.messages import BaseMessage, AIMessage, HumanMessage
+from typing import Any, Type, Optional
 from pydantic import create_model, field_validator, BaseModel
 from pydantic.fields import FieldInfo
 from ..langchain.mixedAgentRenderes import convert_message_to_json
@@ -31,7 +31,11 @@ def formulate_query(kwargs):
             chat_history = []
             for each in kwargs.get('chat_history')[:]:
                 chat_history.append(AIMessage(each))
-    input_message = AIMessage(content=kwargs.get('task'))
+    user_task = kwargs.get('task')
+    if not user_task:
+        raise ToolException("Task is required to invoke the application. "
+                            "Check the provided input (some errors may happen on previous steps).")
+    input_message = HumanMessage(content=user_task)
     result = {"input": [input_message], "chat_history": chat_history}
     for key, value in kwargs.items():
         if key not in ("task", "chat_history"):

{alita_sdk-0.3.342 → alita_sdk-0.3.344}/alita_sdk/runtime/tools/artifact.py

@@ -7,8 +7,8 @@ from langchain_core.documents import Document
 from langchain_core.tools import ToolException
 from pydantic import create_model, Field, model_validator
 
-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 ...tools.non_code_indexer_toolkit import NonCodeIndexerToolkit
+from ...tools.utils.available_tools_decorator import extend_with_parent_available_tools
 from ...runtime.utils.utils import IndexerKeywords
 
 

{alita_sdk-0.3.342 → alita_sdk-0.3.344}/alita_sdk/tools/ado/test_plan/test_plan_wrapper.py

@@ -1,5 +1,6 @@
 import json
 import logging
+import re
 import xml.etree.ElementTree as ET
 from typing import Generator, Literal, Optional, List
 
@@ -384,6 +385,10 @@ class TestPlanApiWrapper(NonCodeIndexerToolkit):
             field_dicts = case.get('work_item', {}).get('work_item_fields', [])
             data = {k: v for d in field_dicts for k, v in d.items()}
             if chunking_tool:
+                steps = data.get('Microsoft.VSTS.TCM.Steps', '')
+                # Remove XML declaration if present (like <?xml version="1.0" encoding="utf-16"?>) to avoid encoding issues
+                steps_no_decl = re.sub(r'<\?xml[^>]*\?>', '', steps, count=1).lstrip()
+
                 yield Document(
                     page_content='',
                     metadata={
@@ -394,7 +399,7 @@ class TestPlanApiWrapper(NonCodeIndexerToolkit):
                         'description': data.get('System.Description', ''),
                         'updated_on': data.get('System.Rev', ''),
                         # content is in metadata for chunking tool post-processing
-                        IndexerKeywords.CONTENT_IN_BYTES.value: data.get('Microsoft.VSTS.TCM.Steps', '').encode("utf-8")
+                        IndexerKeywords.CONTENT_IN_BYTES.value: steps_no_decl.encode("utf-8")
                     })
             else:
                 yield Document(
@@ -411,7 +416,7 @@ class TestPlanApiWrapper(NonCodeIndexerToolkit):
     def _index_tool_params(self):
         """Return the parameters for indexing data."""
         return {
-            'chunking_tool': (Literal['html', ''], Field(description="Name of chunking tool", default='html')),
+            'chunking_tool': (Literal['xml', ''], Field(description="Name of chunking tool", default='xml')),
             "plan_id": (int, Field(description="ID of the test plan for which test cases are requested")),
             "suite_ids": (Optional[List[int]], Field(description='List of test suite IDs for which test cases are requested '
                                                                  '(can be empty for all suites indexing from the plan). '

{alita_sdk-0.3.342 → alita_sdk-0.3.344}/alita_sdk/tools/jira/api_wrapper.py

@@ -1043,27 +1043,39 @@ class JiraApiWrapper(NonCodeIndexerToolkit):
 
     def _extract_image_data(self, field_data):
         """
-        Extracts image data from general JSON response
+        Extracts image data from general JSON response.
+        Handles lists, dicts with image info, and plain strings.
         """
-        if isinstance(field_data, dict) and 'filename' in field_data and 'content' in field_data:
-            return f"!{field_data['filename']}|alt={field_data['filename']}!"
-        if 'content' in field_data and isinstance(field_data['content'], list):
-            result = []
-            for content_item in field_data['content']:
-                if 'content' in content_item and isinstance(content_item['content'], list) and len(content_item['content']) > 0:
-                    if content_item.get('type') == 'mediaSingle':
-                        media = content_item['content'][0]
-                        attrs = media.get('attrs', {})
-                        if attrs.get('type') == 'file':
-                            alt = attrs.get('alt', '')
-                            image_str = f'!{alt}|alt="{alt}"!'
-                            result.append(image_str)
-                    elif content_item.get('type') == 'paragraph':
-                        result.append(content_item['content'][0].get('text', ''))
-                    else:
-                        result.append(self._extract_image_data(content_item))
-            return '\n'.join(result)
-        return str(field_data)
+        if isinstance(field_data, list):
+            return ' '.join(self._extract_image_data(item) for item in field_data)
+        if isinstance(field_data, dict):
+            if 'filename' in field_data and 'content' in field_data:
+                return f"!{field_data['filename']}|alt={field_data['filename']}!"
+            if 'content' in field_data and isinstance(field_data['content'], list):
+                result = []
+                for content_item in field_data['content']:
+                    if (
+                        isinstance(content_item, dict)
+                        and 'content' in content_item
+                        and isinstance(content_item['content'], list)
+                        and content_item['content']
+                    ):
+                        if content_item.get('type') == 'mediaSingle':
+                            media = content_item['content'][0]
+                            attrs = media.get('attrs', {})
+                            if attrs.get('type') == 'file':
+                                alt = attrs.get('alt', '')
+                                image_str = f'!{alt}|alt="{alt}"!'
+                                result.append(image_str)
+                        elif content_item.get('type') == 'paragraph':
+                            result.append(content_item['content'][0].get('text', ''))
+                        else:
+                            result.append(self._extract_image_data(content_item))
+                return '\n'.join(result)
+            return f"Unsupported format of field content."
+        if isinstance(field_data, str):
+            return field_data
+        return f"Unsupported field content type: {type(field_data)}. Expected a string, list, or dict."
 
     def get_field_with_image_descriptions(self, jira_issue_key: str, field_name: str, prompt: Optional[str] = None,
                                           context_radius: int = 500):
@@ -1098,12 +1110,7 @@ class JiraApiWrapper(NonCodeIndexerToolkit):
                 return f"Unable to find field '{field_name}' or it's empty. Available fields are: {existing_fields_str}"
 
             # Handle multiple images or non-string content
-            if isinstance(field_content, list):
-                field_content = ' '.join(map(self._extract_image_data, field_content))
-            elif isinstance(field_content, dict):
-                field_content = self._extract_image_data(field_content)
-            elif not isinstance(field_content, str):
-                return f"Unsupported field content type: {type(field_content)}. Expected a string, list, or dict."
+            field_content = self._extract_image_data(field_content)
 
             # Regular expression to find image references in Jira markup
             image_pattern = r'!([^!|]+)(?:\|[^!]*)?!'
@@ -1237,24 +1244,22 @@ class JiraApiWrapper(NonCodeIndexerToolkit):
 
             comment_author = comment.get('author', {}).get('displayName', 'Unknown')
             comment_created = comment.get('created', 'Unknown date')
+            comment_body = self._extract_image_data(comment_body)
 
             # Process the comment body by replacing image references with descriptions
-            try:
-                processed_body = re.sub(image_pattern,
-                                        lambda match: self.process_image_match(match, comment_body, attachment_resolver, context_radius, prompt),
-                                        comment_body)
-
-                # Add the processed comment to our results
-                processed_comments.append({
-                    "author": comment_author,
-                    "created": comment_created,
-                    "id": comment.get('id'),
-                    "original_content": comment_body,
-                    "processed_content": processed_body
-                })
-            except Exception as e:
-                logger.error(f"Error processing image references in comment: {str(e)}")
-                # TODO process comment in api 3 format
+            processed_body = re.sub(image_pattern,
+                                    lambda match: self.process_image_match(match, comment_body, attachment_resolver, context_radius, prompt),
+                                    comment_body)
+
+            # Add the processed comment to our results
+            processed_comments.append({
+                "author": comment_author,
+                "created": comment_created,
+                "id": comment.get('id'),
+                "original_content": comment_body,
+                "processed_content": processed_body
+            })
+        
         return processed_comments
 
     def get_comments_with_image_descriptions(self, jira_issue_key: str, prompt: Optional[str] = None, context_radius: int = 500):

{alita_sdk-0.3.342 → alita_sdk-0.3.344}/alita_sdk/tools/utils/content_parser.py

@@ -306,6 +306,8 @@ def file_extension_by_chunker(chunker_name: str) -> str:
         return ".txt"
     if name == "html":
         return ".html"
+    if name == "xml":
+        return ".xml"
     if name == "csv":
         return ".csv"
     return None

{alita_sdk-0.3.342 → alita_sdk-0.3.344}/alita_sdk.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.342
+Version: 0.3.344
 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-0.3.342 → alita_sdk-0.3.344}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "alita_sdk"
-version = "0.3.342"
+version = "0.3.344"
 description = "SDK for building langchain agents using resources from Alita"
 readme = "README.md"
 requires-python = ">=3.10"