Merge pull request #495 from VikParuchuri/hotfix-2

Fix issues with newer converter class call

convert.py

@@ -1,4 +1,4 @@
 from marker.scripts.convert import convert_cli
 
 if __name__ == "__main__":
-    main()
+    convert_cli()

marker/scripts/convert.py

@@ -1,5 +1,7 @@
 import os
 
+from marker.converters.pdf import PdfConverter
+
 os.environ["GRPC_VERBOSITY"] = "ERROR"
 os.environ["GLOG_minloglevel"] = "2"
 os.environ["PYTORCH_ENABLE_MPS_FALLBACK"] = "1" # Transformers uses .isin for a simple op, which is not supported on MPS
@@ -44,7 +46,7 @@ def process_single_pdf(args):
     if cli_options.get('skip_existing') and output_exists(out_folder, base_name):
         return
 
-    converter_cls = config_parser.get_converter_cls()
+    converter_cls = PdfConverter
 
     try:
         converter = converter_cls(

marker/scripts/convert_single.py

@@ -1,5 +1,7 @@
 import os
 
+from marker.converters.pdf import PdfConverter
+
 os.environ["GRPC_VERBOSITY"] = "ERROR"
 os.environ["GLOG_minloglevel"] = "2"
 os.environ["PYTORCH_ENABLE_MPS_FALLBACK"] = "1"  # Transformers uses .isin for a simple op, which is not supported on MPS
@@ -24,7 +26,7 @@ def convert_single_cli(fpath: str, **kwargs):
     start = time.time()
     config_parser = ConfigParser(kwargs)
 
-    converter_cls = config_parser.get_converter_cls()
+    converter_cls = PdfConverter
     converter = converter_cls(
         config=config_parser.generate_config_dict(),
         artifact_dict=models,

marker/scripts/server.py

@@ -90,7 +90,8 @@ async def _convert_pdf(params: CommonParams):
         config_parser = ConfigParser(options)
         config_dict = config_parser.generate_config_dict()
         config_dict["pdftext_workers"] = 1
-        converter = PdfConverter(
+        converter_cls = PdfConverter
+        converter = converter_cls(
             config=config_dict,
             artifact_dict=app_data["models"],
             processor_list=config_parser.get_processors(),

marker/scripts/streamlit_app.py

@@ -28,7 +28,8 @@ def load_models():
 def convert_pdf(fname: str, config_parser: ConfigParser) -> (str, Dict[str, Any], dict):
     config_dict = config_parser.generate_config_dict()
     config_dict["pdftext_workers"] = 1
-    converter = PdfConverter(
+    converter_cls = PdfConverter
+    converter = converter_cls(
         config=config_dict,
         artifact_dict=model_dict,
         processor_list=config_parser.get_processors(),

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "marker-pdf"
-version = "1.2.6"
+version = "1.2.7"
 description = "Convert PDF to markdown with high speed and accuracy."
 authors = ["Vik Paruchuri <[email protected]>"]
 readme = "README.md"