Alberto Islas

@mlissner I'm back into this topic: > So I think the solution is: > > pacer_case_id — keep blank if we can (Alberto to research a bit) > pacer_doc_id —...

@mlissner I've done some changes to this. Seems that it's almost done. - Now we get the `pacer_case_id` from NDAs. - Since we don't have a document number in NDAs,...

@mlissner I've done some research about document numbers for NDAs. For some documents is possible to get the document accessing the document URL in NDAs without the "magic_number": So here...

Great, I'll check with the examples we have if there is a court-by-court or a case-by-case thing about using pacer_doc_id instead of the document number and I'll be back with...

@mlissner I've reviewed all the NDAs examples we have so far. I think we have at least one NDA for each appellate court except for ca8 and ca10. - **ca1**...

@mlissner I've checked some cases in Courtlistener for the courts that we don't have NDAs in order to determine if they use `document_number` or `pacer_case_id` also I did an additional...

Great! Yes, we'll have the PDF unless it is sealed or for some other reason we are unable to download the PDF. In those cases the docket entry will remain...

@mlissner After reviewing a bigger sample of appellate documents and trying to get their document number directly from the PDF using `pdfplumber` (thanks to @flooie sample code) I notice some...

Great, I think that's a good approach, working on it. Thanks!

This is updated to get the document number for appellate documents either from the PDF or the download confirmation page. In a brief, this PR contains: - Get the metatada...