[Feature/Enhancement] support for integration with document readers

[jfferson]

Enhancement / feature description:
I am considering the possibility of referencing the notes and bookmarks done through document readers. For enabling it the user could define a program for searching those metadata on the mindforger settings; other than that for this application what would be required is that it signals the search key to be used on the program

for it to work the developer would also need to show a relation of index/content, so that the user knows how to make the refference, but the developer could provide that interface independently, or this application could provide a field for the listing; the really essential part from here is the signaling

then what would happen is that mindforger would transparently request the search as needed to the program (say if the user clicks on the link name) and open the content of the search as a view of a plain document or a section

another option is that mindforger could just have its own file format as an index store for enabling such integration

particularly I would be using it for providing integration with okular
...

Screenshots (with sketch), drawings, mocks, wireframes or diagrams: (if available)

...

[dvorka]

@jfferson This is a very nice and useful idea. I also use Okular - both bookmarks and highlighting, notes only rarely. Anyway, I would like to understand the idea more in depth...

In the new version of MindForger (not released yet) you can configure a document directory (a library) and MindForger indexes all files from subdirectories so that they can be searched, linked from notes and opened. The metadata (PDF) of the documents is not indexed in any way yet.

Could you please describe in more detail how MindForger should retrieve information about notes, bookmarks and highlighted sections? Subsequent saving inside MindForger in a note representing the document will already be easy with what I have done.

Thanks again for the idea and your interest in MindForger!

[jfferson]

The option I explored for this feature is using a specialized library for executing the parsing and a while ago I found out that libokular provides those representations for multiple types of documents, hence the idea of integration, though only pdf has metadata written directly on the document, for the others okular creates a .okular file. Anyway at least for getting it from a page in a document there is a solution that can be get from https://superuser.com/questions/673917/modify-okular-highlight-to-automatically-copy-highlighted-text-into-comment and https://github.com/jsqliu/okular/commit/ee9a8f5272c1ea4b20648fe57b52c726751073aa ; that is for the work of getting the information. Then that execution would be called with also the index of the relevant annotation when it is requested for that annotation to be shown and would output the textual content to the consumer or handler. For it to work the user needs to have beforehand an option to have a listing of the pairs of [index,annotation] on the whole documents so that the user can do the refferences on the notes

[dvorka]

Nice! It looks doable. Thank you for the clarification!