Task outputs tab should now try and read files from task.taskOutputs collection

When the task output tab was initially created task outputs in storage were only using Ivan's task output convention library that persisted files to the linked storage account. Now i see that a Batch Task supports task outputs as well. Yay!


While loading this tab we should check the task for task outputs and load them from these containers instead of the default job container location. Generally speaking i would imagine that there will only be one container used for multiple outputs, but this might not always be the case. We should iterate through the task outputs and load files from these containers instead.

These are listed as SAS URL's in the task outputs collection, so we could either use these instead of automatically defaulting to the linked storage account that we have keys for, or we could use the SAS container URL to get the keys from the particular storage account as it's not guaranteed to be the linked storage account.

We could make this a little smarter so that instead of having "saved output files" AND "saved logs", we just show one tab and add a DDL filter option to show either log files (.txt|.log) and the all other files. We could list the multiple containers in here as well in the case that the user used multiple storage locations. Which would mean in the DDL filter we would have:

* Log files (.log|.txt)
* 1213-tropospheria-arnold-robot-finish-ma-nicola-20170725-212516
* 1213-tropospheria-arnold-robot-finish-ma-images

Would default to log files, but selecting another container would load those instead.

This is obviously up for debate as there might be a better way to do this.

Andrew