refactor: change the name of the USPTO input format

Change the name of the patent USPTO input format to show the typical format (XML). Signed-off-by: Cesar Berrospi Ramis <75900930+ceberam@users.noreply.github.com>
DS4SD · Dec 16, 2024 · 264ef14 · 264ef14
1 parent c0a086c
commit 264ef14
Show file tree

Hide file tree

Showing 4 changed files with 8 additions and 8 deletions.
diff --git a/docling/backend/patent_uspto_backend.py → docling/backend/xml_uspto_backend.py b/docling/backend/patent_uspto_backend.py → docling/backend/xml_uspto_backend.py
@@ -110,7 +110,7 @@ def unload(self) -> None:
     @classmethod
     @override
     def supported_formats(cls) -> set[InputFormat]:
-        return {InputFormat.PATENT_USPTO}
+        return {InputFormat.XML_USPTO}
 
     @override
     def convert(self) -> DoclingDocument:

diff --git a/docling/datamodel/base_models.py b/docling/datamodel/base_models.py
@@ -36,7 +36,7 @@ class InputFormat(str, Enum):
     ASCIIDOC = "asciidoc"
     MD = "md"
     XLSX = "xlsx"
-    PATENT_USPTO = "uspto"
+    XML_USPTO = "uspto"
 
 
 class OutputFormat(str, Enum):
@@ -56,7 +56,7 @@ class OutputFormat(str, Enum):
     InputFormat.IMAGE: ["jpg", "jpeg", "png", "tif", "tiff", "bmp"],
     InputFormat.ASCIIDOC: ["adoc", "asciidoc", "asc"],
     InputFormat.XLSX: ["xlsx"],
-    InputFormat.PATENT_USPTO: ["xml", "txt"],
+    InputFormat.XML_USPTO: ["xml", "txt"],
 }
 
 FormatToMimeType: Dict[InputFormat, List[str]] = {
@@ -83,7 +83,7 @@ class OutputFormat(str, Enum):
     InputFormat.XLSX: [
         "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
     ],
-    InputFormat.PATENT_USPTO: ["application/xml", "text/plain"],
+    InputFormat.XML_USPTO: ["application/xml", "text/plain"],
 }
 
 MimeTypeToFormat = {

diff --git a/docling/document_converter.py b/docling/document_converter.py
@@ -15,7 +15,7 @@
 from docling.backend.msexcel_backend import MsExcelDocumentBackend
 from docling.backend.mspowerpoint_backend import MsPowerpointDocumentBackend
 from docling.backend.msword_backend import MsWordDocumentBackend
-from docling.backend.patent_uspto_backend import PatentUsptoDocumentBackend
+from docling.backend.xml_uspto_backend import PatentUsptoDocumentBackend
 from docling.datamodel.base_models import (
     ConversionStatus,
     DoclingComponentType,
@@ -118,7 +118,7 @@ def _get_default_option(format: InputFormat) -> FormatOption:
         InputFormat.HTML: FormatOption(
             pipeline_cls=SimplePipeline, backend=HTMLDocumentBackend
         ),
-        InputFormat.PATENT_USPTO: FormatOption(
+        InputFormat.XML_USPTO: FormatOption(
             pipeline_cls=SimplePipeline, backend=PatentUsptoDocumentBackend
         ),
         InputFormat.IMAGE: FormatOption(

diff --git a/tests/test_backend_patent_uspto.py b/tests/test_backend_patent_uspto.py
@@ -12,7 +12,7 @@
 from docling_core.types import DoclingDocument
 from docling_core.types.doc import DocItemLabel, TableData, TextItem
 
-from docling.backend.patent_uspto_backend import PatentUsptoDocumentBackend, XmlTable
+from docling.backend.xml_uspto_backend import PatentUsptoDocumentBackend, XmlTable
 from docling.datamodel.base_models import InputFormat
 from docling.datamodel.document import (
     ConversionResult,
@@ -45,7 +45,7 @@ def patents() -> list[tuple[Path, DoclingDocument]]:
     for in_path in patent_paths:
         in_doc = InputDocument(
             path_or_stream=in_path,
-            format=InputFormat.PATENT_USPTO,
+            format=InputFormat.XML_USPTO,
             backend=PatentUsptoDocumentBackend,
         )
         backend = PatentUsptoDocumentBackend(in_doc=in_doc, path_or_stream=in_path)