Matching rule against document metadata [Feature Request] #6690
Replies: 1 comment
-
I actually came here to see examples of writing a post-consume script that assigned the Original Filename to a custom field so I can use it in general with searching, sorting, etc. I think it would be better if Original Filename itself could be used in these processes as I see Title in Paperless and then the new filename as things with separate uses. At present, I find myself not changing the Titles of my documents because searching and sorting by them is useful in finding them again (they contain information about when they were scanned and which scanner was used), but it's less helpful in quickly identifying a document for what it is, such as linking related things together in custom fields. @shamoon, you've been helpful in the past with contributing the front-end part to the preservation of the original filename and also answering my n00b questions when struggling with developing on my M1 Mac - if I were to give this task a go, are there any resources you could point me towards to help me understand how the front-end development works? My dev experience is heavily back-end and even the front-end is limited to languages like PHP, Python, and good old HTML + CSS. I get the sense that making changes in Angular JS is very different and I just could not work it out last time. |
Beta Was this translation helpful? Give feedback.
-
Description
Hello,
Current rules to Tag / Type / Storage can match only document contents.
I have documents with bad quality or hand-written, and organized them with defined filenames. I am using special script to scan+watch+upload into Paperless-ngx new documents using REST API.
For example: I have letter from Commerzbank, it does not contain ocr-recognized text with
commerzbank
in contents, but it has metadata asOriginal filename
with value2023-01-01 - Commerzbank info letter.pdf
. My rule withAny word: Commerzbank
did not match (as word is only in metadata, but not in text).It would be nice, if Paperless-ngx would have option to match also metadata fields to assign tags and other rules as well. It may help a lot, when uploading large archive of documents.
Other
No response
Beta Was this translation helpful? Give feedback.
All reactions