StanGirard/quivr
1
1
Fork 0
mirror of https://github.com/StanGirard/quivr.git synced 2024-09-11 13:15:41 +03:00
Code Issues Packages Projects Releases Wiki Activity

fix: Update parsing instruction in common.py (#2531)

Browse Source 
This pull request fixes the parsing instruction in the common.py file.
The result_type has been corrected to "markdown" and the
parsing_instruction has been updated to handle checkboxes, tables, and
other elements that are hard to parse in a meaningful way.
This commit is contained in:
Stan Girard 2024-05-02 10:12:24 +02:00 committed by GitHub
parent efe51659a3
commit 76f21bc2f5
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
backend/packages/files/parsers/common.py
View File

@ -42,8 +42,8 @@ async def process_file(
document_tmp.write(doc.file.read())
parser = LlamaParse(
result_type="mardown", # "markdown" and "text" are available
parsing_instruction="Extract all the information as possible in a way a human can understand by being as verbose as possible.",
result_type="markdown", # "markdown" and "text" are available
parsing_instruction="The document might contain checkboxes, tables and others elements hard to parse. Find a way to parse them in a meaningful way. ",
)
document_llama_parsed = parser.load_data(document_tmp.name)