search

run-llama / llama_parse

Parse files for optimal RAG
https://www.llamaindex.ai
MIT License
3.28k stars 320 forks source link

page separator not working on current llama parser version on python. #395

Open bilalayaz2 opened 2 months ago

bilalayaz2 commented 2 months ago

Note: we're aware of some missing content in the output and layout issues on tables. Please refrain from opening new issues on this topic unless if you think it's different from what has already been reported.

Describe the bug Llama parse current version's page separator parameter is not functional while accessing through the api.

Client: Please remove untested options:

Options parser = LlamaParse( api_key=self.api_key, parsing_instruction=self.pi, language="en", page_separator="|======================|", result_type=self.result_type, do_not_cache=True, invalidate_cache=True, )

hexapode commented 2 months ago

Could you share a job_id with us where you envcounter the issue, we were not able to reproduce when using the api? You can found them in the history tab of cloud.llamaindex.ai/parse for past jobs.

bilalayaz2 commented 2 months ago

I just tried it again for you to give you a job_id but the history tab did not update so i currently have no job_id at hand. And the same code works in llama-parse==0.4.9 on python.

hexapode commented 2 months ago

We just update the history tab if you log now, it should allow you to see: Job id What went wrong during the parsing (page by page granularity) The output

hexapode commented 2 months ago

What version of the python package are you using we are currently at 0.5.5 on python

bilalayaz2 commented 2 months ago

I'm facing this issue on llama-parser 0.5.5 and tried on both python 3.12.0 and 3.12.6 Job-id = 26064475-e158-4eb1-8ade-49ed0dc25528 The result in job details on the history tab does contain the page separator but the response from the api thats saved as markdown does not have this page separator, but when I use llama-parse 0.4.9, page separator is present in the file saved.