
Problem/Motivation
In the AI: summarize document module I use a native method to parse PDF documents to text. It would be nice to have a generic "Document to Text" operation type that could be further extended with different methods of parsing documents by other providers.
Proposed resolution
Create the DocumentToText operation type.
Remaining tasks
User interface changes
API changes
Data model changes
Issue fork ai-3482004
Show commands
Start within a Git clone of the project using the version control instructions.
Or, if you do not have SSH keys set up on git.drupalcode.org:
- 3482004-document-to-text
changes, plain diff MR !209
Comments
Comment #2
joevagyok CreditAttribution: joevagyok at European Commission and European Union Institutions, Agencies and Bodies for European Commission and European Union Institutions, Agencies and Bodies commentedComment #4
prashant.cThanks for the great work done @joevagyok on this but there is no activity from last 6 months changing the assignee to unassigned.
Comment #5
prashant.cOn the page
/admin/config/ai/settings
getting the following error:Fatal error: Class Drupal\ai\OperationType\DocumentToText\DocumentToTextInput contains 3 abstract methods and must therefore be declared abstract or implement the remaining methods (Drupal\ai\OperationType\InputInterface::getDebugData, Drupal\ai\OperationType\InputInterface::setDebugData, Drupal\ai\OperationType\InputInterface::setDebugDataValue) in /var/www/html/modules/contrib/ai/src/OperationType/DocumentToText/DocumentToTextInput.php on line 11
Comment #6
prashant.cComment #7
kristen polWe are doing some issue management housekeeping and adding/removing components.
We are removing the "Code" component and want people to categorize issues with the best module/submodule component.
Moving this issue to "AI Core module".
See #3533272: Update AI module project components for more details.
Comment #8
marcus_johansson CreditAttribution: marcus_johansson at FreelyGive commentedWe are moving to having document loaders as its own section, so will not fix this.