handle long requests with POST #42
Conversation
| "usehistory": "y", | ||
| "format": "json", | ||
| } | ||
| if len(term) <= 200: |
There was a problem hiding this comment.
Please don't use "magic numbers" in the code.
Define it as a constant.
If the number's source's external documentation, please add a link to the relevant document.
There was a problem hiding this comment.
The documentation is very cryptic, literally says more than several hundred characters long for eSearch, and if more than about 200 UIDs are to be provided, the request should be made using the HTTP POST method for EFetch. That's why I chose 200, the terms field can probably take more characters but I'd need to make tests to see how many hundred.
The documentation is here https://www.ncbi.nlm.nih.gov/books/NBK25499/#chapter4.ESearch for Esearch and https://www.ncbi.nlm.nih.gov/books/NBK25499/#_chapter4_EFetch_ for EFetch
There was a problem hiding this comment.
I will add it as a constant, though!
| if efetch_response.status_code == STATUS_ERROR_CODE: | ||
| raise handle_errors.NotFoundSRA(efetch_response, accessions) | ||
| return efetch_response | ||
| elif mode == 'prepare': | ||
| return Request(method='GET', | ||
| if len(accessions_string) <= 200: |
There was a problem hiding this comment.
My earlier comment about the '200' magic number is amplified here.
Using something more then once is even a stringer reason to have it as a constant.
amnonkhen
changed the title
As explained in the documentation for Esearch and EFetch, when dealing with long lists of terms/ids, they need to be POSTed rather than included in the url as a parameter.
This fixes an issue with
GSE196830, which has too many runs.Also, minor fix on the requests library, which was imported twice