fix: add_logs

megaparse/api/app.py

@@ -80,15 +80,17 @@ async def parse_file(
         language=language,
         parsing_instruction=parsing_instruction,
     )
-
-    parser = parser_builder.build(parser_config)
-    with tempfile.NamedTemporaryFile(
-        delete=False, suffix=f".{str(file.filename).split('.')[-1]}"
-    ) as temp_file:
-        temp_file.write(file.file.read())
-        megaparse = MegaParse(parser=parser)
-        result = await megaparse.aload(file_path=temp_file.name)
-        return {"message": "File parsed successfully", "result": result}
+    try:
+        parser = parser_builder.build(parser_config)
+        with tempfile.NamedTemporaryFile(
+            delete=False, suffix=f".{str(file.filename).split('.')[-1]}"
+        ) as temp_file:
+            temp_file.write(file.file.read())
+            megaparse = MegaParse(parser=parser)
+            result = await megaparse.aload(file_path=temp_file.name)
+            return {"message": "File parsed successfully", "result": result}
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
 
 
 @app.post("/v1/url")
@@ -107,9 +109,14 @@ async def upload_url(
 
         with tempfile.NamedTemporaryFile(delete=False, suffix="pdf") as temp_file:
             temp_file.write(response.content)
-            megaparse = MegaParse(parser=UnstructuredParser(strategy=StrategyEnum.AUTO))
-            result = megaparse.load(temp_file.name)
-            return {"message": "File parsed successfully", "result": result}
+            try:
+                megaparse = MegaParse(
+                    parser=UnstructuredParser(strategy=StrategyEnum.AUTO)
+                )
+                result = megaparse.load(temp_file.name)
+                return {"message": "File parsed successfully", "result": result}
+            except Exception as e:
+                raise HTTPException(status_code=500, detail=str(e))
     else:
         data = await playwright_loader.aload()
         # Now turn the data into a string

megaparse/core/parser/unstructured_parser.py

@@ -1,13 +1,11 @@
-import asyncio
-from unstructured.partition.auto import partition
+import re
+
 from dotenv import load_dotenv
-from megaparse.core.parser import MegaParser
 from langchain_core.language_models.chat_models import BaseChatModel
 from langchain_core.prompts import ChatPromptTemplate
-import re
-from langchain_openai import ChatOpenAI
-import os
+from unstructured.partition.auto import partition
 
+from megaparse.core.parser import MegaParser
 from megaparse.core.parser.type import StrategyEnum
 
 

megaparse/sdk/examples/usage_example.py

@@ -1,7 +1,7 @@
 import asyncio
 import os
 
-from megaparse.sdk import MegaParseSDK
+from megaparse_sdk import MegaParseSDK
 
 
 async def main():

megaparse/sdk/megaparse_sdk/client.py

@@ -5,7 +5,7 @@
 
 class MegaParseClient:
     def __init__(self, api_key: str | None = None):
-        self.base_url = "https://megaparse.tooling.quivr.app"  # to define once in production  # to define once in production
+        self.base_url = "https://megaparse.tooling.quivr.app"
 
         self.api_key = api_key
         if self.api_key:

pyproject.toml

@@ -26,7 +26,7 @@ all = [
     "playwright>=1.47.0",
     "langchain-anthropic>=0.2.3",
     "python-magic>=0.4.27",
-    "unstructured[all-docs]>=0.15.0",
+    "unstructured[all-docs]==0.15.0",
     "langchain>=0.2.0",
     "langchain-community>=0.2.0",
     "langchain-openai>=0.1.0",

requirements-dev.lock

@@ -168,8 +168,6 @@ grpcio-status==1.67.1
 h11==0.14.0
     # via httpcore
     # via uvicorn
-html5lib==1.1
-    # via unstructured
 httpcore==1.0.6
     # via httpx
 httpx==0.27.2
@@ -360,7 +358,6 @@ numpy==1.26.4
     # via torchvision
     # via transformers
     # via unstructured
-    # via unstructured-inference
 nvidia-cublas-cu12==12.4.5.8 ; platform_machine == 'x86_64' and platform_system == 'Linux'
     # via nvidia-cudnn-cu12
     # via nvidia-cusolver-cu12
@@ -420,6 +417,7 @@ packaging==24.1
     # via matplotlib
     # via onnxruntime
     # via pikepdf
+    # via pytesseract
     # via pytest
     # via transformers
     # via unstructured-pytesseract
@@ -443,8 +441,6 @@ pdfplumber==0.11.4
     # via megaparse
 pexpect==4.9.0 ; sys_platform != 'emscripten' and sys_platform != 'win32'
     # via ipython
-pi-heif==0.20.0
-    # via unstructured
 pikepdf==9.4.0
     # via unstructured
 pillow==11.0.0
@@ -453,11 +449,14 @@ pillow==11.0.0
     # via matplotlib
     # via pdf2image
     # via pdfplumber
-    # via pi-heif
     # via pikepdf
+    # via pillow-heif
+    # via pytesseract
     # via python-pptx
     # via torchvision
     # via unstructured-pytesseract
+pillow-heif==0.20.0
+    # via unstructured
 platformdirs==4.3.6
     # via black
     # via jupyter-core
@@ -543,6 +542,8 @@ pypdfium2==4.30.0
     # via pdfplumber
 pyreadline3==3.5.4 ; sys_platform == 'win32'
     # via humanfriendly
+pytesseract==0.3.13
+    # via unstructured
 pytest==8.3.3
     # via pytest-asyncio
     # via pytest-cov
@@ -570,7 +571,7 @@ python-multipart==0.0.17
     # via unstructured-inference
 python-oxmsg==0.0.1
     # via unstructured
-python-pptx==1.0.2
+python-pptx==0.6.23
     # via unstructured
 pytz==2024.2
     # via pandas
@@ -628,7 +629,6 @@ setuptools==75.3.0
     # via torch
 six==1.16.0
     # via asttokens
-    # via html5lib
     # via langdetect
     # via python-dateutil
 sniffio==1.3.1
@@ -650,6 +650,8 @@ starlette==0.41.2
 sympy==1.13.1
     # via onnxruntime
     # via torch
+tabulate==0.9.0
+    # via unstructured
 tenacity==8.5.0
     # via langchain
     # via langchain-community
@@ -711,7 +713,6 @@ typing-extensions==4.12.2
     # via pyee
     # via python-docx
     # via python-oxmsg
-    # via python-pptx
     # via sqlalchemy
     # via torch
     # via typing-inspect
@@ -723,11 +724,11 @@ typing-inspect==0.9.0
     # via unstructured-client
 tzdata==2024.2
     # via pandas
-unstructured==0.16.4
+unstructured==0.15.0
     # via megaparse
 unstructured-client==0.27.0
     # via unstructured
-unstructured-inference==0.8.1
+unstructured-inference==0.7.36
     # via unstructured
 unstructured-pytesseract==0.3.13
     # via unstructured
@@ -739,8 +740,6 @@ virtualenv==20.27.1
     # via pre-commit
 wcwidth==0.2.13
     # via prompt-toolkit
-webencodings==0.5.1
-    # via html5lib
 wrapt==1.16.0
     # via deprecated
     # via llama-index-core

requirements.lock

@@ -138,8 +138,6 @@ grpcio-status==1.67.1
 h11==0.14.0
     # via httpcore
     # via uvicorn
-html5lib==1.1
-    # via unstructured
 httpcore==1.0.6
     # via httpx
 httpx==0.27.2
@@ -305,7 +303,6 @@ numpy==1.26.4
     # via torchvision
     # via transformers
     # via unstructured
-    # via unstructured-inference
 nvidia-cublas-cu12==12.4.5.8 ; platform_machine == 'x86_64' and platform_system == 'Linux'
     # via nvidia-cudnn-cu12
     # via nvidia-cusolver-cu12
@@ -363,6 +360,7 @@ packaging==24.1
     # via matplotlib
     # via onnxruntime
     # via pikepdf
+    # via pytesseract
     # via transformers
     # via unstructured-pytesseract
 pandas==2.2.3
@@ -379,8 +377,6 @@ pdfminer-six==20231228
 pdfplumber==0.11.4
     # via layoutparser
     # via megaparse
-pi-heif==0.20.0
-    # via unstructured
 pikepdf==9.4.0
     # via unstructured
 pillow==11.0.0
@@ -389,11 +385,14 @@ pillow==11.0.0
     # via matplotlib
     # via pdf2image
     # via pdfplumber
-    # via pi-heif
     # via pikepdf
+    # via pillow-heif
+    # via pytesseract
     # via python-pptx
     # via torchvision
     # via unstructured-pytesseract
+pillow-heif==0.20.0
+    # via unstructured
 playwright==1.48.0
     # via megaparse
 portalocker==2.10.1
@@ -459,6 +458,8 @@ pypdfium2==4.30.0
     # via pdfplumber
 pyreadline3==3.5.4 ; sys_platform == 'win32'
     # via humanfriendly
+pytesseract==0.3.13
+    # via unstructured
 python-dateutil==2.8.2
     # via matplotlib
     # via pandas
@@ -478,7 +479,7 @@ python-multipart==0.0.17
     # via unstructured-inference
 python-oxmsg==0.0.1
     # via unstructured
-python-pptx==1.0.2
+python-pptx==0.6.23
     # via unstructured
 pytz==2024.2
     # via pandas
@@ -529,7 +530,6 @@ scipy==1.14.1
 setuptools==75.3.0
     # via torch
 six==1.16.0
-    # via html5lib
     # via langdetect
     # via python-dateutil
 sniffio==1.3.1
@@ -549,6 +549,8 @@ starlette==0.41.2
 sympy==1.13.1
     # via onnxruntime
     # via torch
+tabulate==0.9.0
+    # via unstructured
 tenacity==8.5.0
     # via langchain
     # via langchain-community
@@ -598,7 +600,6 @@ typing-extensions==4.12.2
     # via pyee
     # via python-docx
     # via python-oxmsg
-    # via python-pptx
     # via sqlalchemy
     # via torch
     # via typing-inspect
@@ -610,20 +611,18 @@ typing-inspect==0.9.0
     # via unstructured-client
 tzdata==2024.2
     # via pandas
-unstructured==0.16.4
+unstructured==0.15.0
     # via megaparse
 unstructured-client==0.27.0
     # via unstructured
-unstructured-inference==0.8.1
+unstructured-inference==0.7.36
     # via unstructured
 unstructured-pytesseract==0.3.13
     # via unstructured
 urllib3==2.2.3
     # via requests
 uvicorn==0.32.0
     # via megaparse
-webencodings==0.5.1
-    # via html5lib
 wrapt==1.16.0
     # via deprecated
     # via llama-index-core