Workflows - Access OCR content for condition for adding tags

[ZiZoTeX]

Hello there,

I have trouble accessing the content of a Document to determine if a add a tag.
I tried it with the document.latest_version.content variable, but I guess this is no accessible from a workflow.
I have a workflow which enters a state when OCRing is finished. Then I would like to look for certain words an add a tag if they are in the content.
Unfortunatley I have no clue how I can access the OCR Content from within the workflow state action.


Any help is greatly appreciated.

Best regards

[ZiZoTeX]

just realized that I posted the question in the wrong forum. Sorry bout that!

[ZiZoTeX]

And now I feel pretty stupid.
Of course I can access everything from workflow_instance.document.latest_version.ocr_content.

I didn't realize I was not using the workflow_instance entrypoint there.

[lonestar]

Hello,
I'm new here and I'm starting to use Mayan EDMS just now so please forgive me if I'm asking something that sounds obvious, but I am trying to do this same thing: a workflow for adding tags to documents depending on the presence of certain words in their OCR content.

I really don't understand how the action condition should be written

I've tried things like:

Code: Select all

{% if "my word" in workflow_instance.document.latest_version.ocr_content %}True{% endif %}

and many variations of this but I don't seem to find a way. The action is not performed.

Could anyone please give me a hint of what I'm doing wrong?

Thanks
Luigi

[Gubert]

Hello,
I'm new here and I'm starting to use Mayan EDMS just now so please forgive me if I'm asking something that sounds obvious, but I am trying to do this same thing: a workflow for adding tags to documents depending on the presence of certain words in their OCR content.

I really don't understand how the action condition should be written

I've tried things like:

Code: Select all

{% if "my word" in workflow_instance.document.latest_version.ocr_content %}True{% endif %}

and many variations of this but I don't seem to find a way. The action is not performed.

Could anyone please give me a hint of what I'm doing wrong?

Thanks
Luigi

Stuck here too. Can anyone help please?
Tried things like:

Code: Select all

{% if "try" in workflow_instance.document.latest_version.ocr_content %}
{% endif %}

Code: Select all

{% if "try" in workflow_instance.document.latest_version.ocr_content %}True{% endif %}

Without condition the workflow works fine. The documentation is really incomplete. No examples for that.

[fgdutoit]

I found that aplying the Django filter join to workflow_instance.document.latest_version.ocr_content worked for me

{% if "my word" in workflow_instance.document.latest_version.ocr_content|join:" " %}True{% endif %}

P.S:The Template Sandbox (in document view) really help with figuring out templates

[michael]

Expanding on fgdutoit's solution:

I found that aplying the Django filter join to workflow_instance.document.latest_version.ocr_content worked for me

{% if "my word" in workflow_instance.document.latest_version.ocr_content|join:" " %}True{% endif %}

P.S:The Template Sandbox (in document view) really help with figuring out templates

Since "ocr_content" is a generator that returns the OCR content of each page, you can save a bit of memory and get a potential speed boost by using:

{% for page_ocr in workflow_instance.document.latest_version.ocr_content %}
{% if "my word" in page_ocr %}True{% endif %}
{% endfor %}

This iterates over the OCR content of each page instead of joining all the OCR content in a single string, which can be more efficient for documents with a large number of pages.

[Gubert]

2 years later, fresh install, new shot. Same problem.
Workflow exist and works until tagging.
Action add Tag wont be add tag with any condition that includes ocr.
Document OCR-Content result looks fine in the UI.

Tryed

{% if "word" in document.latest_version.ocr_content|join:" " %}True{% endif %}

{% for page_ocr in workflow_instance.document.latest_version.ocr_content %}
{% if "my word" in page_ocr %}True{% endif %}
{% endfor %}

{% if "my word" in workflow_instance.document.latest_version.ocr_content|join:" " %}True{% endif %}

Whats the point i'm missing?

Sandboxresult for all these conditions is empty.

EDIT: Shouldnt the "{{ document.versions__version_pages__ocr_content__content }}" give any result in the sandbox? Result is empty.

[cktwo]

Hello,

what worked for me was:

Code: Select all

{% if "word" in workflow_instance.document.ocr_content|join:" " %}True{% endif %}"

You can even filter with "and" for multiple strings:

Code: Select all

{% if "word1" and "word2" in workflow_instance.document.ocr_content|join:" " %}True{% endif %}"

To test these in the sandbox of a document just remove the "workflow_instance." part, you only need this for the workflow.

But it was important to add a "Transition" in the workflow config for this to work.
So try creating a transition with the same origin and destination state (if you only have one like me).
In the transition you can enable a trigger, e.g. "Document version OCR finished".

Hope this helps.

Kind regards