Docs questions about eido HOT 11 CLOSED

I guess a related follow-up question is: where should we document 'how to write a PEP schema' ? Does it go with eido or with PEP docs? Right now it's here: http://pep.databio.org/en/latest/howto_schema/

It's also incomplete

from eido.

another one: handling string or array for attributes

from eido.

is this complete?

I added two points to your list

Since eido is really only relevant in the context of PEP, do you think I should consolidate the eido docs page into the new pep spec site?

hmmm do you think we should remove eido docs altogether and just make it a part of PEP spec?

from eido.

hmmm do you think we should remove eido docs altogether and just make it a part of PEP spec?

I keep going back and forth...

option 1: No eido-specific docs, just consolidate these docs into pepspec
option 2: keep docs separate...but then what goes where? I guess I'd put the 'how to write a schema' docs into the eido page.

Now I'm leaning toward option 2

from eido.

The fact that eido docs have grown (especially the jupyter notebooks) also justifies making it separate.

from eido.

I added two points to your list

Thanks, got it. What do you think about the 'ngs_inputs' thing we used to have in the pipeline interface? is this worth adding?

from eido.

The fact that eido docs have grown (especially the jupyter notebooks) also justifies making it separate.

Alright, I'll do it that way.

from eido.

Ok I've gone through the docs. I added some details to the intro and added the section on 'writing a schema'. deployed on dev here:

http://eido.databio.org/en/dev/

I'll now remove those docs from pepspec.

let me know what you think

from eido.

What do you think about the 'ngs_inputs' thing we used to have in the pipeline interface? is this worth adding?

What exactly was the purpose of this? I can't find any docs about it. Currently there's nothing left in looper that would respect this

from eido.

Ok. let's drop it for now.

what it did was, if you told looper 'this is an NGS input', then it had some functions that could automatically determine some attributes, like "is this paired end or single end?" and "what is the length of reads". it was kind of nice to not have to include those as sample attributes because they could be learned from the input -- but only if we flagged which inputs were sequencing reads.

from eido.

Ok pepspec docs are live now as well. I think I'm happy with the eido and pepspec docs now. let me know if you have any issues.

from eido.