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

Issue in getting data from tables spanning across pages with footer text #11

New Issue
Open
opened 2026-02-16 00:16:36 -05:00 by yindo · 2 comments
yindo commented 2026-02-16 00:16:36 -05:00
Owner
Copy Link
Copy Source

Originally created by @rit-github on GitHub (Feb 27, 2024).

If tables are spanning across pages - data gets split across multiple tables with inconsistent results:
Instance 1. Started a new table structure with different order of column names in the second part of the table (rows carrying over into the next page)
Instance 2. Assume first row in the next page as the header (although the header in the previous page was formatted as Bold Centered header row - different from rest of the table).
Instance 3. Completely omitted rows in the next page and started with text after the table - so erroneously dropped rows that carried into next page.

If there is a footer on each page, at times the footer is ignored and table is continued. In most instances, table gets split and text from footer is parsed as text between 2 tables. Again the table often assumes a different order or even different column header names.

Instance 4. Complex columns with Nested Structure (e.g. Merged Header field with 2 sub headers) does not produce consistent output
e.g.
| | | Merged Col C |
|Col-A|Col-A|Col-C1 | Col-C2|
|A1 |B1 |C1-1 |C2-1 |
|A2 |B2 |C1-2 |C2-2 |
|A3 |B3 |C1-3 |C2-3 |

The output from tables like these (Merged Col-C with split sub-columns) results in inconsistent output and headers getting confused.

Originally created by @rit-github on GitHub (Feb 27, 2024). If tables are spanning across pages - data gets split across multiple tables with inconsistent results: Instance 1. Started a new table structure with different order of column names in the second part of the table (rows carrying over into the next page) Instance 2. Assume first row in the next page as the header (although the header in the previous page was formatted as Bold Centered header row - different from rest of the table). Instance 3. Completely omitted rows in the next page and started with text after the table - so erroneously dropped rows that carried into next page. If there is a footer on each page, at times the footer is ignored and table is continued. In most instances, table gets split and text from footer is parsed as text between 2 tables. Again the table often assumes a different order or even different column header names. Instance 4. Complex columns with Nested Structure (e.g. Merged Header field with 2 sub headers) does not produce consistent output e.g. | | | Merged Col C | |Col-A|Col-A|Col-C1 | Col-C2| |A1 |B1 |C1-1 |C2-1 | |A2 |B2 |C1-2 |C2-2 | |A3 |B3 |C1-3 |C2-3 | The output from tables like these (Merged Col-C with split sub-columns) results in inconsistent output and headers getting confused.
yindo added the enhancement label 2026-02-16 00:16:36 -05:00
yindo commented 2026-02-16 00:16:36 -05:00
Author
Owner
Copy Link
Copy Source

@hexapode commented on GitHub (Feb 27, 2024):

Thanks, we do not handle very well multiple following page yet, as the parser look at each page individually.

In a future release (it's currently in development), we will have a continuous mode that tell the parser to look at the document in it's totality, removing header/footer and merging content across pages for when it is needed.

@hexapode commented on GitHub (Feb 27, 2024): Thanks, we do not handle very well multiple following page yet, as the parser look at each page individually. In a future release (it's currently in development), we will have a continuous mode that tell the parser to look at the document in it's totality, removing header/footer and merging content across pages for when it is needed.
yindo commented 2026-02-16 00:16:36 -05:00
Author
Owner
Copy Link
Copy Source

@dajagbaZ commented on GitHub (Apr 25, 2024):

@hexapode -- Any updates on this issue? My team is also facing this same problem when we try to work across multiple pages.

@dajagbaZ commented on GitHub (Apr 25, 2024): @hexapode -- Any updates on this issue? My team is also facing this same problem when we try to work across multiple pages.
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 enhancement
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#11
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?