Description
The IssueTracker/JiraHost classes in Skara make the assumption that the base URL for Jira is just a hostname with no path. In order to use Skara with a Jira instance with a URL that has a path element, this needs to be fixed.
Instead of always basing REST call URLs on the configured base path and appending to it, there are many cases where the URIBuilder::setPath is used instead, which overwrites any existing path element. This is especially troublesome when combined with using IssueProject::webUrl as the base for these URLs.
I would like to fix this by adding a supported way of getting the base URI from an IssueTracker (IssueTracker::uri), replace all uses of ::webUrl for constructing other URLs with this new call, and replacing all ::setPath with ::appendPath.
This change will require all configurations to be careful not to add a trailing slash to the URL for the Jira instance.
Instead of always basing REST call URLs on the configured base path and appending to it, there are many cases where the URIBuilder::setPath is used instead, which overwrites any existing path element. This is especially troublesome when combined with using IssueProject::webUrl as the base for these URLs.
I would like to fix this by adding a supported way of getting the base URI from an IssueTracker (IssueTracker::uri), replace all uses of ::webUrl for constructing other URLs with this new call, and replacing all ::setPath with ::appendPath.
This change will require all configurations to be careful not to add a trailing slash to the URL for the Jira instance.