HowdenParser 5.2.3__tar.gz → 5.2.5__tar.gz

{howdenparser-5.2.3 → howdenparser-5.2.5}/HowdenParser/parsers/llama_parser.py

@@ -4,6 +4,7 @@ import logging
 from pathlib import Path
 from ..parser import BaseParser
 from typing import Any
+from llama_parse import ResultType
 
 class LlamaParser(BaseParser, name="llamaparser"):
     def __init__(self, result_type: str, model: str, parse_mode: str, provider_and_model: str,
@@ -45,13 +46,15 @@ class LlamaParser(BaseParser, name="llamaparser"):
         if not api_key:
             raise EnvironmentError("Missing LLAMA-PARSER-API-TOKEN in .env file.")
 
-        rt = self.result_type
-
-        rt_lower = str(rt).lower()
-        if rt_lower in ("md", "markdown"):
+        rt_lower = str(self.result_type).lower()
+        if rt_lower in ("md","markdown"):
             rt = ResultType.MD
-        elif rt_lower in ("json", "layout_json", "jsonl"):
+        elif rt_lower in ("txt","text"):
+            rt = ResultType.TXT
+        elif rt_lower in ("json"):
             rt = ResultType.JSON
+        else:
+            raise NotImplementedError(f"Result type {self.result_type} is not supported.")
 
         # Construct the parser INSIDE the worker thread
         self._parser = LlamaParse(
@@ -70,28 +73,20 @@ class LlamaParser(BaseParser, name="llamaparser"):
     def parse(self, file_path: Path, include_pagenumbers: bool = False) -> Any:
         # Ensure parser is created inside the thread event loop
         self._lazy_init()
+        def page_break(text: str, idx: int) -> str:
+            return f"<PAGE_NUMBER {idx}>{text}</PAGE_NUMBER {idx}>"
 
-        # Standard calls
-        if not include_pagenumbers:
-            documents = self._parser.load_data(str(file_path))
-            return "\n".join(doc.text for doc in documents)
-
-        # With page numbers
         documents = self._parser.parse(file_path)
-
-        if self.result_type.lower() in ("md", "markdown"):
-            return "\n".join(
-                f"<PAGE_NUMBER {idx}>{page.md}</PAGE_NUMBER {idx}>"
-                for idx, page in enumerate(documents.pages, start=1)
-            )
-
-        if self.result_type.lower() in ("json", "layout_json", "jsonl"):
-            return documents.model_dump() if hasattr(documents, "model_dump") else documents
-
-        return "\n".join(
-            f"<PAGE_NUMBER {idx}>{page.text}</PAGE_NUMBER {idx}>"
-            for idx, page in enumerate(documents.pages, start=1)
-        )
+        if self.result_type in ['md','markdown']:
+            if include_pagenumbers: return "\n".join(page_break(page.md, idx) for idx, page in enumerate(documents.pages, start=1))
+            else: return "\n".join(page.md for page in documents.pages)
+        elif self.result_type in ['txt','text']:
+            if include_pagenumbers: return "\n".join(page_break(page.text, idx) for idx, page in enumerate(documents.pages, start=1))
+            else: return "\n".join(page.text for page in documents.pages)
+        elif self.result_type.lower() == 'json':
+            return documents.model_dump()
+        else:
+            raise NotImplementedError(f"Result type {self.result_type} is not supported.")
 
     def __call__(self, file_path: Path) -> str:
         return self.parse(file_path)

{howdenparser-5.2.3 → howdenparser-5.2.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: HowdenParser
-Version: 5.2.3
+Version: 5.2.5
 Summary: A simple configuration manager with Pydantic and JSON export.
 License: MIT
 Keywords: config,configuration,pydantic,json
@@ -16,6 +16,8 @@ Requires-Dist: llama-parse (>=0.6.58,<0.7.0)
 Requires-Dist: mistralai (>=1.9.3,<2.0.0)
 Requires-Dist: pdf2image (>=1.17.0,<2.0.0)
 Requires-Dist: pypdf2 (>=3.0.1,<4.0.0)
+Requires-Dist: pytest (>=9.0.2,<10.0.0)
+Requires-Dist: tomli-w (>=1.2.0,<2.0.0)
 Description-Content-Type: text/markdown
 
 # OCR & LLM Parser

{howdenparser-5.2.3 → howdenparser-5.2.5}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "HowdenParser"
-version = "5.2.3"
+version = "5.2.5"
 description = "A simple configuration manager with Pydantic and JSON export."
 readme = "README.md"
 requires-python = ">=3.12,<3.14"
@@ -22,6 +22,8 @@ dependencies = [
     "pypdf2 (>=3.0.1,<4.0.0)",
     "pdf2image (>=1.17.0,<2.0.0)",
     "langchain (>=1.1.3,<2.0.0)",
+    "pytest (>=9.0.2,<10.0.0)",
+    "tomli-w (>=1.2.0,<2.0.0)",
 ]
 
 [project.license]