]
Max Rydahl Andersen edited comment on JBIDE-20375 at 7/24/15 9:30 AM:
----------------------------------------------------------------------
So we will move from having a stacks.yml with runtimes listed, keyed to the archetypes
and examples and url to still have stacks.yml with all this info but now the url will have
to be calculated in a more complex manner via a rest service ?
I don't really follow why that is an advantage. the new proposal for me seem much more
fragile and won't reduce the need for us to maintain extra metadata.
was (Author: maxandersen):
So we will move from having a stacks.yml with runtimes listed, keyed to the archetypes
and examples and url to still have stacks.yml with all this info but now the url will have
to be calculated in a more complex manner via a rest service ?
I don't really follow why that is an advantage. it is for sure much more fragile.
Investigate possibility of reading Download Manager data from a REST
service instead of YAML
--------------------------------------------------------------------------------------------
Key: JBIDE-20375
URL:
https://issues.jboss.org/browse/JBIDE-20375
Project: Tools (JBoss Tools)
Issue Type: Feature Request
Reporter: David Hladky
Download Manager has a rest service, that contains information about all downloads on
developers.redhat.com web page (and also many projects on
www.jboss.org product pages). In
the next year it will start also showing content from ITs unified Download Manager so the
number of downloads it describes will become even higher.
Because our productization team (I think [~pmuir], [~cobrien_redhat]), has to fill the
information about existing downloads to many places (from what I know they must upload it
to access.cdn, they must enter information about it to Download Manager, they must mention
it in Yaml files, they must upload it to customer portal, some of those to Unified
download manager as well...). In my opinion it is quite tedious process that we have a
chance to make less complicated.
Download manager will be a source for the product web pages and as it already contains
most of the necessary information [~wesbos] is working on it an he is fairly close to this
goal.
And if we have enough information about that why not use it and replicate the information
in a different process?