The default value of filename should be returned by the server either from header, or from url instead of an empty one

[opsdep]

Downloaded dataset should be named based on its entry_id for all cases.

For copernicus harvested file the header of the response of the download link provide the associated part named 'filename=' for find the name of the dataset based on it. But for the osisaf and other future download links, the header of the response whould be something like:

So no information can be gained about the filename from such a header. Thus, this line

django-geo-spaas-processing/geospaas_processing/downloaders.py

Line 65 in 7793104

return ''

should be corrected for solving this issue.

[akorosov]

Rather the filename should be returned by the server either from header, or from url.