Expose OnDeployExecutor Interface and store exception message on result node

[geertsmichael]

Expose OnDeployExecutor Interface: so scripts can be triggered externally
Store exception message on result node: so it can be read

[update-changelog]

Thanks for opening this pull request! The maintainers of this repository would appreciate it if you would update the CHANGELOG.md file with this pull request.

[coveralls]

Pull Request Test Coverage Report for Build 4748

• 5 of 5 (100.0%) changed or added relevant lines in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.007%) to 53.699%

---

Totals
Change from base Build 4744:	0.007%
Covered Lines:	13929
Relevant Lines:	25939

---

💛 - Coveralls

[HitmanInWis]

Will try to get to this next week.

[HitmanInWis]

Can you please help me understand the use case for Expose OnDeployExecutor Interface: so scripts can be triggered externally?

[davidjgonzalez]

@HitmanInWis I believe @geertsmichael discussed the use-case here: #1956

[HitmanInWis]

I'm not really seeing the use case, just the request to be able to call the OnDeployExecutor from other code. i.e. I see "what" is being requested (ability to re-trigger a script from code), but don't understand "why" it's being requested especially since the JMX console already allows re-running a script I believe AND a script can always be re-run by re-deploying the bundle (simply need to delete the status node if you are re-running a previously successful script).

OnDeployScripts were always kinda meant to be one-time things, but I imagine folks have come up with other use cases that I'm not thinking of. Seeking to understand.

[geertsmichael]

The use-case is about a tool that I wrote, which will visualize the script execution (success, running, fail). This tool allows to do a re-run through my Touch UI tool.

[HitmanInWis]

One quick question remains about CHANGELOG.md - other than that this looks good and I'll test it out later today.

[HitmanInWis]

Everything tests out, merged. Thanks @michael-geerts for the contribution and @shsteimer for the recommended approach.