HowdenParser 2.0.1__tar.gz → 2.0.3__tar.gz

{howdenparser-2.0.1 → howdenparser-2.0.3}/HowdenParser/parameter/llamaparser.py

@@ -2,12 +2,8 @@
 from typing import Literal
 from pydantic import BaseModel
 
-class Hest(BaseModel):
-    provider_and_model: str = "huggingface:trocr-large-printed"
-
 
 class Parameter(BaseModel):
-    model1: Hest = Hest()
     provider_and_model: str = "llamaparser:"
     result_type: Literal["md"] = "md"
     mode: bool = False

{howdenparser-2.0.1 → howdenparser-2.0.3}/HowdenParser/parser.py

@@ -126,6 +126,10 @@ class MistralOCRParser(BaseParser, name="mistralocr"):
 
         return "\n".join(doc.markdown for doc in ocr_response.pages)
 
+    def __call__(self, file_path: Path) -> str:
+        # When instance is called, it delegates to parse
+        return self.parse(file_path)
+
     @staticmethod
     def _count_pages(file_path: Path) -> int:
         reader = PdfReader(str(file_path))
@@ -142,26 +146,38 @@ class LangChainParser(BaseParser, name="langchain"):
         response = self.model(text)
         return {"source": "LangChain", "output": response}
 
+    def __call__(self, file_path: Path) -> str:
+        # When instance is called, it delegates to parse
+        return self.parse(file_path)
+
 
 class LlamaParser(BaseParser, name="llamaparser"):
-    def __init__(self, result_type: str, mode: bool) -> None:
+    def __init__(self, result_type: str, mode: bool, provider_and_model) -> None:
         logging.info("Initializing LlamaParser...")
 
+        self.result_type = result_type
+        self.mode = mode
+        self.provider_and_model = provider_and_model
+
         from llama_parse import LlamaParse, ResultType
 
         if result_type.lower() in ("md", "markdown"):
-            self.result_type = ResultType.MD
+            result_type = ResultType.MD
 
         api_key = os.getenv("LLAMA-PARSER-API-TOKEN")
         if not api_key:
             raise EnvironmentError("Missing LLAMA-PARSER-API-TOKEN in .env file.")
 
-        self.parser = LlamaParse(api_key=api_key, result_type=self.result_type, premium_mode=mode)
+        self._parser = LlamaParse(api_key=api_key, result_type=result_type, premium_mode=self.mode)
 
     def parse(self, file_path: Path) -> str:
-        documents = self.parser.load_data(str(file_path))
+        documents = self._parser.load_data(str(file_path))
         return "\n".join(doc.text for doc in documents)
 
+    def __call__(self, file_path: Path) -> str:
+        # When instance is called, it delegates to parse
+        return self.parse(file_path)
+
 
 class HuggingFaceParser(BaseParser, name="huggingface"):
     def __init__(self, provider_and_model: str) -> None:

{howdenparser-2.0.1 → howdenparser-2.0.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: HowdenParser
-Version: 2.0.1
+Version: 2.0.3
 Summary: A simple configuration manager with Pydantic and JSON export.
 License: MIT
 Keywords: config,configuration,pydantic,json

{howdenparser-2.0.1 → howdenparser-2.0.3}/pyproject.toml

@@ -14,7 +14,7 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "HowdenParser"
-version = "2.0.1"
+version = "2.0.3"
 description = "A simple configuration manager with Pydantic and JSON export."
 authors = [ "JesperThoftIllemannJ <jesper.jaeger@howdendanmark.dk>",]
 readme = "README.md"