alita-sdk 0.3.347__py3-none-any.whl → 0.3.348__py3-none-any.whl

alita_sdk/runtime/clients/artifact.py

@@ -42,7 +42,17 @@ class Artifact:
             return f"{data['error']}. {data['content'] if data['content'] else ''}"
         detected = chardet.detect(data)
         if detected['encoding'] is not None:
-            return data.decode(detected['encoding'])
+            try:
+                return data.decode(detected['encoding'])
+            except Exception:
+                logger.error("Error while default encoding")
+                return parse_file_content(file_name=artifact_name,
+                                          file_content=data,
+                                          is_capture_image=is_capture_image,
+                                          page_number=page_number,
+                                          sheet_name=sheet_name,
+                                          excel_by_sheets=excel_by_sheets,
+                                          llm=llm)
         else:
             return parse_file_content(file_name=artifact_name,
                                   file_content=data,

alita_sdk/runtime/clients/client.py

@@ -69,6 +69,7 @@ class AlitaClient:
         self.configurations_url = f'{self.base_url}{self.api_path}/integrations/integrations/default/{self.project_id}?section=configurations&unsecret=true'
         self.ai_section_url = f'{self.base_url}{self.api_path}/integrations/integrations/default/{self.project_id}?section=ai'
         self.configurations: list = configurations or []
+        self.model_timeout = kwargs.get('model_timeout', 120)
 
     def get_mcp_toolkits(self):
         if user_id := self._get_real_user_id():
@@ -184,6 +185,7 @@ class AlitaClient:
             model=embedding_model,
             api_key=self.auth_token,
             openai_organization=str(self.project_id),
+            request_timeout=self.model_timeout
         )
 
     def get_llm(self, model_name: str, model_config: dict) -> ChatOpenAI:

alita_sdk/runtime/langchain/document_loaders/AlitaExcelLoader.py

@@ -12,27 +12,32 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import io
+import os
 from typing import Iterator
 import pandas as pd
 from json import loads
 
 from openpyxl import load_workbook
+from xlrd import open_workbook
 from langchain_core.documents import Document
 from .AlitaTableLoader import AlitaTableLoader
-    
-cell_delimeter = " | "
 
-class AlitaExcelLoader(AlitaTableLoader):
+cell_delimiter = " | "
 
+class AlitaExcelLoader(AlitaTableLoader):
     excel_by_sheets: bool = False
     sheet_name: str = None
     return_type: str = 'str'
+    file_name: str = None
 
     def __init__(self, **kwargs):
         if not kwargs.get('file_path'):
             file_content = kwargs.get('file_content')
             if file_content:
+                self.file_name = kwargs.get('file_name')
                 kwargs['file_path'] = io.BytesIO(file_content)
+        else:
+            self.file_name = kwargs.get('file_path')
         super().__init__(**kwargs)
         self.excel_by_sheets = kwargs.get('excel_by_sheets')
         self.return_type = kwargs.get('return_type')
@@ -40,36 +45,82 @@ class AlitaExcelLoader(AlitaTableLoader):
 
     def get_content(self):
         try:
-            # Load the workbook
-            workbook = load_workbook(self.file_path, data_only=True)  # `data_only=True` ensures we get cell values, not formulas
-
-            if self.sheet_name:
-                # If a specific sheet name is provided, parse only that sheet
-                if self.sheet_name in workbook.sheetnames:
-                    sheet_content = self.parse_sheet(workbook[self.sheet_name])
-                    return sheet_content
-                else:
-                    raise ValueError(f"Sheet '{self.sheet_name}' does not exist in the workbook.")
-            elif self.excel_by_sheets:
-                # Parse each sheet individually and return as a dictionary
-                result = {}
-                for sheet_name in workbook.sheetnames:
-                    sheet_content = self.parse_sheet(workbook[sheet_name])
-                    result[sheet_name] = sheet_content
-                return result
+            # Determine file extension
+            file_extension = os.path.splitext(self.file_name)[-1].lower()
+
+            if file_extension == '.xlsx':
+                # Use openpyxl for .xlsx files
+                return self._read_xlsx()
+            elif file_extension == '.xls':
+                # Use xlrd for .xls files
+                return self._read_xls()
             else:
-                # Combine all sheets into a single string result
-                result = []
-                for sheet_name in workbook.sheetnames:
-                    sheet_content = self.parse_sheet(workbook[sheet_name])
-                    result.append(f"====== Sheet name: {sheet_name} ======\n{sheet_content}")
-                return "\n\n".join(result)
+                raise ValueError(f"Unsupported file format: {file_extension}")
         except Exception as e:
             return f"Error reading Excel file: {e}"
 
+    def _read_xlsx(self):
+        """
+        Reads .xlsx files using openpyxl.
+        """
+        workbook = load_workbook(self.file_path, data_only=True)  # `data_only=True` ensures we get cell values, not formulas
+
+        if self.sheet_name:
+            # If a specific sheet name is provided, parse only that sheet
+            if self.sheet_name in workbook.sheetnames:
+                sheet_content = self.parse_sheet(workbook[self.sheet_name])
+                return sheet_content
+            else:
+                raise ValueError(f"Sheet '{self.sheet_name}' does not exist in the workbook.")
+        elif self.excel_by_sheets:
+            # Parse each sheet individually and return as a dictionary
+            result = {}
+            for sheet_name in workbook.sheetnames:
+                sheet_content = self.parse_sheet(workbook[sheet_name])
+                result[sheet_name] = sheet_content
+            return result
+        else:
+            # Combine all sheets into a single string result
+            result = []
+            for sheet_name in workbook.sheetnames:
+                sheet_content = self.parse_sheet(workbook[sheet_name])
+                result.append(f"====== Sheet name: {sheet_name} ======\n{sheet_content}")
+            return "\n\n".join(result)
+
+    def _read_xls(self):
+        """
+        Reads .xls files using xlrd.
+        """
+        workbook = open_workbook(filename=self.file_name, file_contents=self.file_content)
+
+        if self.sheet_name:
+            # If a specific sheet name is provided, parse only that sheet
+            if self.sheet_name in workbook.sheet_names():
+                sheet = workbook.sheet_by_name(self.sheet_name)
+                sheet_content = self.parse_sheet_xls(sheet)
+                return sheet_content
+            else:
+                raise ValueError(f"Sheet '{self.sheet_name}' does not exist in the workbook.")
+        elif self.excel_by_sheets:
+            # Parse each sheet individually and return as a dictionary
+            result = {}
+            for sheet_name in workbook.sheet_names():
+                sheet = workbook.sheet_by_name(sheet_name)
+                sheet_content = self.parse_sheet_xls(sheet)
+                result[sheet_name] = sheet_content
+            return result
+        else:
+            # Combine all sheets into a single string result
+            result = []
+            for sheet_name in workbook.sheet_names():
+                sheet = workbook.sheet_by_name(sheet_name)
+                sheet_content = self.parse_sheet_xls(sheet)
+                result.append(f"====== Sheet name: {sheet_name} ======\n{sheet_content}")
+            return "\n\n".join(result)
+
     def parse_sheet(self, sheet):
         """
-        Parses a single sheet, extracting text and hyperlinks, and formats them.
+        Parses a single .xlsx sheet, extracting text and hyperlinks, and formats them.
         """
         sheet_content = []
 
@@ -85,17 +136,52 @@ class AlitaExcelLoader(AlitaTableLoader):
                     # If no hyperlink, use the cell value (computed value if formula)
                     row_content.append(str(cell.value) if cell.value is not None else "")
             # Join the row content into a single line using `|` as the delimiter
-            sheet_content.append(cell_delimeter.join(row_content))
+            sheet_content.append(cell_delimiter.join(row_content))
+
+        # Format the sheet content based on the return type
+        return self._format_sheet_content(sheet_content)
+
+    def parse_sheet_xls(self, sheet):
+        """
+        Parses a single .xls sheet using xlrd, extracting text and hyperlinks, and formats them.
+        """
+        sheet_content = []
+
+        # Extract hyperlink map (if available)
+        hyperlink_map = getattr(sheet, 'hyperlink_map', {})
+
+        for row_idx in range(sheet.nrows):
+            row_content = []
+            for col_idx in range(sheet.ncols):
+                cell = sheet.cell(row_idx, col_idx)
+                cell_value = cell.value
+
+                # Check if the cell has a hyperlink
+                cell_address = (row_idx, col_idx)
+                if cell_address in hyperlink_map:
+                    hyperlink = hyperlink_map[cell_address].url_or_path
+                    if cell_value:
+                        row_content.append(f"[{cell_value}]({hyperlink})")
+                else:
+                    row_content.append(str(cell_value) if cell_value is not None else "")
+            # Join the row content into a single line using `|` as the delimiter
+            sheet_content.append(cell_delimiter.join(row_content))
 
         # Format the sheet content based on the return type
+        return self._format_sheet_content(sheet_content)
+
+    def _format_sheet_content(self, sheet_content):
+        """
+        Formats the sheet content based on the return type.
+        """
         if self.return_type == 'dict':
             # Convert to a list of dictionaries (each row is a dictionary)
-            headers = sheet_content[0].split(cell_delimeter) if sheet_content else []
+            headers = sheet_content[0].split(cell_delimiter) if sheet_content else []
             data_rows = sheet_content[1:] if len(sheet_content) > 1 else []
-            return [dict(zip(headers, row.split(cell_delimeter))) for row in data_rows]
+            return [dict(zip(headers, row.split(cell_delimiter))) for row in data_rows]
         elif self.return_type == 'csv':
             # Return as CSV (newline-separated rows, comma-separated values)
-            return "\n".join([",".join(row.split(cell_delimeter)) for row in sheet_content])
+            return "\n".join([",".join(row.split(cell_delimiter)) for row in sheet_content])
         else:
             # Default: Return as plain text (newline-separated rows, pipe-separated values)
             return "\n".join(sheet_content)

alita_sdk/runtime/tools/vectorstore.py

@@ -137,7 +137,7 @@ class VectorStoreWrapper(BaseToolApiWrapper):
     embedding_model_params: dict
     vectorstore_type: str
     vectorstore_params: dict
-    max_docs_per_add: int = 100
+    max_docs_per_add: int = 20
     dataset: str = None
     embedding: Any = None
     vectorstore: Any = None

alita_sdk/runtime/tools/vectorstore_base.py

@@ -135,7 +135,7 @@ class VectorStoreWrapperBase(BaseToolApiWrapper):
     embedding_model: Optional[str] = None
     vectorstore_type: Optional[str]  = None
     vectorstore_params: Optional[dict]  = None
-    max_docs_per_add: int = 100
+    max_docs_per_add: int = 20
     dataset: Optional[str] = None
     vectorstore: Any = None
     pg_helper: Any = None

{alita_sdk-0.3.347.dist-info → alita_sdk-0.3.348.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: alita_sdk
-Version: 0.3.347
+Version: 0.3.348
 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-0.3.347.dist-info → alita_sdk-0.3.348.dist-info}/RECORD

@@ -35,8 +35,8 @@ alita_sdk/configurations/zephyr_enterprise.py,sha256=UaBk3qWcT2-bCzko5HEPvgxArw1
 alita_sdk/configurations/zephyr_essential.py,sha256=tUIrh-PRNvdrLBj6rJXqlF-h6oaMXUQI1wgit07kFBw,752
 alita_sdk/runtime/__init__.py,sha256=4W0UF-nl3QF2bvET5lnah4o24CoTwSoKXhuN0YnwvEE,828
 alita_sdk/runtime/clients/__init__.py,sha256=BdehU5GBztN1Qi1Wul0cqlU46FxUfMnI6Vq2Zd_oq1M,296
-alita_sdk/runtime/clients/artifact.py,sha256=TPvROw1qu4IyUEGuf7x40IKRpb5eFZpYGN3-8LfQE0M,3461
-alita_sdk/runtime/clients/client.py,sha256=ZOWsv-JJl54lzQ4JzYFBKslt4DI0ExNZ3zQ_U7zA3uE,43590
+alita_sdk/runtime/clients/artifact.py,sha256=Tt3aWcxu20bVW6EX7s_iX5CTmcItKhUnkk8Q2gv2vw0,4036
+alita_sdk/runtime/clients/client.py,sha256=T3hmVnT63iLWEGeuJb8k8Httw-sSWUpy6rsrumD0P0w,43699
 alita_sdk/runtime/clients/datasource.py,sha256=HAZovoQN9jBg0_-lIlGBQzb4FJdczPhkHehAiVG3Wx0,1020
 alita_sdk/runtime/clients/prompt.py,sha256=li1RG9eBwgNK_Qf0qUaZ8QNTmsncFrAL2pv3kbxZRZg,1447
 alita_sdk/runtime/langchain/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -56,7 +56,7 @@ alita_sdk/runtime/langchain/document_loaders/AlitaCSVLoader.py,sha256=3ne-a5qIkB
 alita_sdk/runtime/langchain/document_loaders/AlitaConfluenceLoader.py,sha256=NzpoL4C7UzyzLouTSL_xTQw70MitNt-WZz3Eyl7QkTA,8294
 alita_sdk/runtime/langchain/document_loaders/AlitaDirectoryLoader.py,sha256=fKezkgvIcLG7S2PVJp1a8sZd6C4XQKNZKAFC87DbQts,7003
 alita_sdk/runtime/langchain/document_loaders/AlitaDocxMammothLoader.py,sha256=9hi5eHgDIfa9wBWqTuwMM6D6W64czrDTfZl_htooe8Y,5943
-alita_sdk/runtime/langchain/document_loaders/AlitaExcelLoader.py,sha256=P17csHx94JkXiyo1a2V-CrfP2E5XCG4uZC31ulZ_Ab4,5817
+alita_sdk/runtime/langchain/document_loaders/AlitaExcelLoader.py,sha256=h8x1Xma_IBM4NdGXVVuvHHSlFQgY0S7Xjj8oGZhdFL8,9256
 alita_sdk/runtime/langchain/document_loaders/AlitaGitRepoLoader.py,sha256=5WXGcyHraSVj3ANHj_U6X4EDikoekrIYtS0Q_QqNIng,2608
 alita_sdk/runtime/langchain/document_loaders/AlitaImageLoader.py,sha256=QwgBJE-BvOasjgT1hYHZc0MP0F_elirUjSzKixoM6fY,6610
 alita_sdk/runtime/langchain/document_loaders/AlitaJSONLoader.py,sha256=Nav2cgCQKOHQi_ZgYYn_iFdP_Os56KVlVR5nHGXecBc,3445
@@ -121,8 +121,8 @@ alita_sdk/runtime/tools/prompt.py,sha256=nJafb_e5aOM1Rr3qGFCR-SKziU9uCsiP2okIMs9
 alita_sdk/runtime/tools/router.py,sha256=wCvZjVkdXK9dMMeEerrgKf5M790RudH68pDortnHSz0,1517
 alita_sdk/runtime/tools/sandbox.py,sha256=WNz-aUMtkGCPg84dDy_0BPkyp-6YjoYB-xjIEFFrtKw,11601
 alita_sdk/runtime/tools/tool.py,sha256=lE1hGi6qOAXG7qxtqxarD_XMQqTghdywf261DZawwno,5631
-alita_sdk/runtime/tools/vectorstore.py,sha256=UFBAJ_N2F6uB0xxIy1VMx581tHco-xDl7v2Hl6u0Xzw,34468
-alita_sdk/runtime/tools/vectorstore_base.py,sha256=F2EFwq5LFwCpV6U9D5Jq1dxYrV3lxOErLfgWTXqEVRI,27293
+alita_sdk/runtime/tools/vectorstore.py,sha256=8vRhi1lGFEs3unvnflEi2p59U2MfV32lStpEizpDms0,34467
+alita_sdk/runtime/tools/vectorstore_base.py,sha256=7ZkbegFG0XTQBYGsJjtrkK-zrqKwketfx8vSJzuPCug,27292
 alita_sdk/runtime/utils/AlitaCallback.py,sha256=E4LlSBuCHWiUq6W7IZExERHZY0qcmdjzc_rJlF2iQIw,7356
 alita_sdk/runtime/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 alita_sdk/runtime/utils/constants.py,sha256=Xntx1b_uxUzT4clwqHA_U6K8y5bBqf_4lSQwXdcWrp4,13586
@@ -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-0.3.347.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-alita_sdk-0.3.347.dist-info/METADATA,sha256=ZvJklicNWTOf3Q1MrNdtVdLhUEnd4oVDokj_y4J_Ecg,19015
-alita_sdk-0.3.347.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-alita_sdk-0.3.347.dist-info/top_level.txt,sha256=0vJYy5p_jK6AwVb1aqXr7Kgqgk3WDtQ6t5C-XI9zkmg,10
-alita_sdk-0.3.347.dist-info/RECORD,,
+alita_sdk-0.3.348.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+alita_sdk-0.3.348.dist-info/METADATA,sha256=_oiAJpxGjG23s01B-P41PkJqG1oUgdAgS7QUnlyS5gc,19015
+alita_sdk-0.3.348.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+alita_sdk-0.3.348.dist-info/top_level.txt,sha256=0vJYy5p_jK6AwVb1aqXr7Kgqgk3WDtQ6t5C-XI9zkmg,10
+alita_sdk-0.3.348.dist-info/RECORD,,