Automate browser based workflows with AI
Python aiapiautomationbrowserbrowser-automationcomputergptllmplaywrightpowerautomatepuppeteerpythonrpaseleniumvisionworkflow
feat(SKY-9160): default file parser type to Auto detect (backend) (#5576)
Aaron Perez8efd3308ba23
diff --git a/fern/openapi/skyvern_openapi.json b/fern/openapi/skyvern_openapi.jsonindex e4df50773..aadd2e5b8 100644--- a/fern/openapi/skyvern_openapi.json+++ b/fern/openapi/skyvern_openapi.json@@ -9685,7 +9685,8 @@"title": "File Url"},"file_type": {- "$ref": "#/components/schemas/FileType"+ "$ref": "#/components/schemas/FileType",+ "default": "auto_detect"},"json_schema": {"anyOf": [@@ -9704,8 +9705,7 @@"required": ["label","output_parameter",- "file_url",- "file_type"+ "file_url"],"title": "FileParserBlock"},@@ -9761,7 +9761,8 @@"title": "File Url"},"file_type": {- "$ref": "#/components/schemas/FileType"+ "$ref": "#/components/schemas/FileType",+ "default": "auto_detect"},"json_schema": {"anyOf": [@@ -9779,8 +9780,7 @@"type": "object","required": ["label",- "file_url",- "file_type"+ "file_url"],"title": "FileParserBlockYAML"},@@ -9795,6 +9795,7 @@"FileType": {"type": "string","enum": [+ "auto_detect","csv","excel","pdf",diff --git a/skyvern/client/types/file_parser_block.py b/skyvern/client/types/file_parser_block.pyindex c6cbdb2f0..e43bf501f 100644--- a/skyvern/client/types/file_parser_block.py+++ b/skyvern/client/types/file_parser_block.py@@ -25,7 +25,7 @@ class FileParserBlock(UniversalBaseModel):disable_cache: typing.Optional[bool] = Nonenext_loop_on_failure: typing.Optional[bool] = Nonefile_url: str- file_type: FileType+ file_type: typing.Optional[FileType] = Nonejson_schema: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = Noneif IS_PYDANTIC_V2:diff --git a/skyvern/client/types/file_type.py b/skyvern/client/types/file_type.pyindex 7e754261e..619ce5455 100644--- a/skyvern/client/types/file_type.py+++ b/skyvern/client/types/file_type.py@@ -2,4 +2,4 @@import typing-FileType = typing.Union[typing.Literal["csv", "excel", "pdf", "image", "docx"], typing.Any]+FileType = typing.Union[typing.Literal["auto_detect", "csv", "excel", "pdf", "image", "docx"], typing.Any]diff --git a/skyvern/client/types/for_loop_block_loop_blocks_item.py b/skyvern/client/types/for_loop_block_loop_blocks_item.pyindex ad0e488f6..52c1d7325 100644--- a/skyvern/client/types/for_loop_block_loop_blocks_item.py+++ b/skyvern/client/types/for_loop_block_loop_blocks_item.py@@ -265,7 +265,7 @@ class ForLoopBlockLoopBlocksItem_FileUrlParser(UniversalBaseModel):disable_cache: typing.Optional[bool] = Nonenext_loop_on_failure: typing.Optional[bool] = Nonefile_url: str- file_type: FileType+ file_type: typing.Optional[FileType] = Nonejson_schema: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = Noneif IS_PYDANTIC_V2:diff --git a/skyvern/client/types/for_loop_block_yaml_loop_blocks_item.py b/skyvern/client/types/for_loop_block_yaml_loop_blocks_item.pyindex cc4e1159e..08f92f9fe 100644--- a/skyvern/client/types/for_loop_block_yaml_loop_blocks_item.py+++ b/skyvern/client/types/for_loop_block_yaml_loop_blocks_item.py@@ -224,7 +224,7 @@ class ForLoopBlockYamlLoopBlocksItem_FileUrlParser(UniversalBaseModel):model: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = Nonenext_loop_on_failure: typing.Optional[bool] = Nonefile_url: str- file_type: FileType+ file_type: typing.Optional[FileType] = Nonejson_schema: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = Noneif IS_PYDANTIC_V2:diff --git a/skyvern/client/types/workflow_definition_blocks_item.py b/skyvern/client/types/workflow_definition_blocks_item.pyindex 6162fd81d..82e84d1e2 100644--- a/skyvern/client/types/workflow_definition_blocks_item.py+++ b/skyvern/client/types/workflow_definition_blocks_item.py@@ -265,7 +265,7 @@ class WorkflowDefinitionBlocksItem_FileUrlParser(UniversalBaseModel):disable_cache: typing.Optional[bool] = Nonenext_loop_on_failure: typing.Optional[bool] = Nonefile_url: str- file_type: FileType+ file_type: typing.Optional[FileType] = Nonejson_schema: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = Noneif IS_PYDANTIC_V2:diff --git a/skyvern/client/types/workflow_definition_yaml_blocks_item.py b/skyvern/client/types/workflow_definition_yaml_blocks_item.pyindex 6edab9847..1d48bda50 100644--- a/skyvern/client/types/workflow_definition_yaml_blocks_item.py+++ b/skyvern/client/types/workflow_definition_yaml_blocks_item.py@@ -200,7 +200,7 @@ class WorkflowDefinitionYamlBlocksItem_FileUrlParser(UniversalBaseModel):model: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = Nonenext_loop_on_failure: typing.Optional[bool] = Nonefile_url: str- file_type: FileType+ file_type: typing.Optional[FileType] = Nonejson_schema: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] = Noneif IS_PYDANTIC_V2:diff --git a/skyvern/forge/prompts/skyvern/workflow_knowledge_base.txt b/skyvern/forge/prompts/skyvern/workflow_knowledge_base.txtindex 44f2aed3f..4010033ef 100644--- a/skyvern/forge/prompts/skyvern/workflow_knowledge_base.txt+++ b/skyvern/forge/prompts/skyvern/workflow_knowledge_base.txt@@ -532,13 +532,13 @@ blocks:** FILE PARSER BLOCK (file_url_parser) **-Purpose: Parse PDFs, CSVs, and Excel files. This is the "File Parser Block" in the UI.+Purpose: Parse PDFs, CSVs, Excel files, images, and DOCX files. This is the "File Parser Block" in the UI.Structure:block_type: file_url_parserlabel: <unique_label>file_url: <https_url_or_file_url> # Required: URL to the file-file_type: <pdf|csv|excel> # Required: File type+file_type: <auto_detect|csv|excel|pdf|image|docx> # Optional: defaults to auto_detect (infer from URL/content)json_schema: <json_schema> # Optional: Structure of parsed outputUse Cases:diff --git a/skyvern/forge/sdk/workflow/models/block.py b/skyvern/forge/sdk/workflow/models/block.pyindex ff11fbe2a..273809da7 100644--- a/skyvern/forge/sdk/workflow/models/block.py+++ b/skyvern/forge/sdk/workflow/models/block.py@@ -3747,7 +3747,7 @@ class FileParserBlock(Block):_MAX_CSV_FIELD_SIZE_BYTES = 10 * 1024 * 1024file_url: str- file_type: FileType+ file_type: FileType = FileType.AUTO_DETECTjson_schema: dict[str, Any] | None = Nonedef get_failure_error_codes(self) -> list[str]:@@ -4180,7 +4180,8 @@ class FileParserBlock(Block):# Download the file.file_path = await download_file(self.file_url, organization_id=organization_id)- # Auto-detect file type if not explicitly set (IMAGE/EXCEL/PDF/DOCX are explicit choices)+ # Resolve AUTO_DETECT (and legacy CSV-as-default) via URL/magic-byte detection;+ # IMAGE/EXCEL/PDF/DOCX are honored as user overrides.if self.file_type not in (FileType.IMAGE, FileType.EXCEL, FileType.PDF, FileType.DOCX):self.file_type = self._detect_file_type_from_url(self.file_url, file_path=file_path)diff --git a/skyvern/schemas/workflows.py b/skyvern/schemas/workflows.pyindex 043a5e146..941fa7d12 100644--- a/skyvern/schemas/workflows.py+++ b/skyvern/schemas/workflows.py@@ -442,6 +442,7 @@ class BlockResult:class FileType(StrEnum):+ AUTO_DETECT = "auto_detect"CSV = "csv"EXCEL = "excel"PDF = "pdf"@@ -831,7 +832,7 @@ class FileParserBlockYAML(BlockYAML):block_type: Literal[BlockType.FILE_URL_PARSER] = BlockType.FILE_URL_PARSER # type: ignorefile_url: str- file_type: FileType+ file_type: FileType = FileType.AUTO_DETECTjson_schema: dict[str, Any] | None = None