Closed toloudis closed 1 year ago
Acceptable inputs:
dataset=url_prefix/filename.json (our code will split to extract the url_prefix so that file paths inside the dataset file)
dataset=url_prefix (our code will assume manifest.json
by default)
if url_prefix does not start with http
or //
then assume it to be in our internal filesystem hardcoded location. (use the /allen... prefix)
based on #36, collection=url_prefix&dataset=dataset_name.
Scenarios:
if collection is omitted, then dataset must be a full url_prefix at least. if collection is present, then it is a url_prefix to a location with a list of datasets.
We don't have any tooling to create a collection yet. Do it manually for the time being. A json file with an array of string names [{"path":"baby_bear", "name":"Baby Bear"}, ...]
Use Case
As a data stakeholder, I have the ability to host my data on AWS and would like to use the tool to point to any cloud url.
Acceptance Criteria
nucmorph colorizer has a url query parameter for http://viewer_url/?dataset=dataset_url in which dataset_url points to the top level dataset json file. The remaining file paths referenced inside should be considered relative to this url.
Details
Take url string, chop off
manifest.json
and use the remaining string as baseUrl...We will have to figure out what to do with the dataset dropdown box. That is TBD.
Ultimately we could have dataset=url&feature=featureName&track=trackId&t=time but this issue is only for dataset.