run-llama/llama_cloud_services
1
0
Fork 0
mirror of https://github.com/run-llama/llama_cloud_services.git synced 2026-07-18 15:54:38 -04:00
Code Issues 352 Packages Projects Releases 10 Wiki Activity

LlamaParse feature take_screenshot does not work with AzStorageBlobReader #254

New Issue
Open
opened 2026-02-16 00:17:16 -05:00 by yindo · 0 comments
yindo commented 2026-02-16 00:17:16 -05:00
Owner
Copy Link
Copy Source

Originally created by @galvangoh on GitHub (Sep 2, 2024).

Describe the bug
It is not possible to parse document with AzStorageBlobReader with the take_screenshot=True featurefrom LlamaParse. Also, theAzStorageBlobReader` class does not provide any interface to download screenshots of the document.

Reproducible example:

from llama_parse import LlamaParse
from llama_index.readers.azstorage_blob import AzStorageBlobReader

import os
from dotenv import load_dotenv
load_dotenv()
LLAMA_CLOUD_API_KEY = os.getenv("LLAMA_CLOUD_API_KEY")

instructions = """The provided document is an invoice. Please extract all basic
               information of the document, customer & supplier. The document
               also contains table of line items that needs to be extracted as
               well."""

container = 'CONTAINER_NAME'
folder_path = 'SUBDIR_1/SUBDIR_2'
connection_string = 'my_connection_string'
blob_name = 'MultiPageInvoice.pdf'

# parameters for LlamaParse
parser_params = {
    'api_key': LLAMA_CLOUD_API_KEY,
    'result_type': 'markdown',
    'parsing_instruction': instructions,
    'invalidate_cache': True,
    'do_not_cache': True,
    'skip_diagonal_text': True,
    'num_workers': 5,
    'ignore_errors': False,
    'use_vendor_multimodal_model': True,
    'vendor_multimodal_model_name': 'openai-gpt4o',
    'take_screenshot': True
}

# instantiate the parser
parser = LlamaParse(**parser_params)

file_extractor = {'.pdf': parser}

azure_loader = AzStorageBlobReader(
    container_name=f'{container_name}/{folder_path}', 
    connection_string=connection_string,
    blob=blob_name,
    file_extractor=file_extractor,
)

# begin parsing
document = azure_loader.load_data() # error out here

Error message:

Started parsing the file under job_id 2e1f4eb4-2025-4a23-9297-a71fd979de62
Error while parsing the file '<bytes/buffer>': Failed to parse the file: 2e1f4eb4-2025-4a23-9297-a71fd979de62, status: ERROR
Failed to load file file:///C:/####/####/####/####/####/####/MultiPageInvoice.pdf with error: Failed to parse the file: 2e1f4eb4-2025-4a23-9297-a71fd979de62, status: ERROR. Skipping...

Files
MultiPageInvoice.pdf

Job ID
2e1f4eb4-2025-4a23-9297-a71fd979de62

Screenshots
tempsnip

Client:

  • Python Library

Additional context
llama-parse==0.5.1
llama-index-readers-azstorage-blob==0.2.0

Originally created by @galvangoh on GitHub (Sep 2, 2024). **Describe the bug** It is not possible to parse document with `AzStorageBlobReader` with the `take_screenshot=True` feature` from LlamaParse. Also, the `AzStorageBlobReader` class does not provide any interface to download screenshots of the document. Reproducible example: ```python from llama_parse import LlamaParse from llama_index.readers.azstorage_blob import AzStorageBlobReader import os from dotenv import load_dotenv load_dotenv() LLAMA_CLOUD_API_KEY = os.getenv("LLAMA_CLOUD_API_KEY") instructions = """The provided document is an invoice. Please extract all basic information of the document, customer & supplier. The document also contains table of line items that needs to be extracted as well.""" container = 'CONTAINER_NAME' folder_path = 'SUBDIR_1/SUBDIR_2' connection_string = 'my_connection_string' blob_name = 'MultiPageInvoice.pdf' # parameters for LlamaParse parser_params = { 'api_key': LLAMA_CLOUD_API_KEY, 'result_type': 'markdown', 'parsing_instruction': instructions, 'invalidate_cache': True, 'do_not_cache': True, 'skip_diagonal_text': True, 'num_workers': 5, 'ignore_errors': False, 'use_vendor_multimodal_model': True, 'vendor_multimodal_model_name': 'openai-gpt4o', 'take_screenshot': True } # instantiate the parser parser = LlamaParse(**parser_params) file_extractor = {'.pdf': parser} azure_loader = AzStorageBlobReader( container_name=f'{container_name}/{folder_path}', connection_string=connection_string, blob=blob_name, file_extractor=file_extractor, ) # begin parsing document = azure_loader.load_data() # error out here ``` Error message: ``` Started parsing the file under job_id 2e1f4eb4-2025-4a23-9297-a71fd979de62 Error while parsing the file '<bytes/buffer>': Failed to parse the file: 2e1f4eb4-2025-4a23-9297-a71fd979de62, status: ERROR Failed to load file file:///C:/####/####/####/####/####/####/MultiPageInvoice.pdf with error: Failed to parse the file: 2e1f4eb4-2025-4a23-9297-a71fd979de62, status: ERROR. Skipping... ``` **Files** [MultiPageInvoice.pdf](https://github.com/user-attachments/files/16842721/MultiPageInvoice.pdf) **Job ID** 2e1f4eb4-2025-4a23-9297-a71fd979de62 **Screenshots** ![tempsnip](https://github.com/user-attachments/assets/4a13cfbc-bed9-457d-bf08-dd46e5c81fba) **Client:** - Python Library **Additional context** `llama-parse==0.5.1` `llama-index-readers-azstorage-blob==0.2.0`
yindo added the bug label 2026-02-16 00:17:16 -05:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
LlamaCloud
LlamaExtract
LlamaParse
bug
client
complex layout
documentation
duplicate
enhancement
figure
font
header footer
help wanted
ocr
parser
pdf_debug
pdf_debug
pull-request
Mirrored from GitHub Pull Request
 python-lib
question
server_issue
server_issue
ui
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: run-llama/llama_cloud_services#254
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?