Hi JPatrick, 
yes, now its much more clearer to me and I see that this is definitely a 
usefull thing for ArangoDB users.

I've send you a PR with some tiny updates to the readme, feel Free to merge 
or decline ;-)

While I now see that you didn't choose yaml for the reason of limiting 
dependencies, I'd like to explain my question a bit more.
Comming from an environment where Apache JMeter was used for QA & load 
tests, you have a nice & snappy UI where you can click your http requests 
together with a nice UI. 
However, jmeters storage backend is writing XML files.
If you now want to do code reviews on change sets, thats next to impossible 
- the XML is hardly as readable as binary. 
My company back then developed YAML based testcases on top of TheGrinder, 
which overcame all these limits in a very nice way, plus developers really 
enjoyed working with it.
Since one may want to do code reviews on schema evolution descriptions, I 
would sugest that you (similar as Liquibase) at least offer a way to have 
yaml support on top of the current implementation.

One similar tool in the ArangoDB stack is swagger.io - which has a json (as 
not human readable) representation of its REST grammer; If you load it in 
the swagger editor ( http://swagger.io/swagger-editor/ ), it offers you a 
YAML representation that is more compact and offers a better overview for 
the reader.

So, maybe a derived Migrant Verde, that has the additional YAML dependency 
could offer better user experience, plus maintain the XML schema being 
easily portable to YAML?

Cheers, 

Willi

PS: maybe you could think about a little less explicit language in the 
README?

On Saturday, July 9, 2016 at 6:19:05 AM UTC+2, JPatrick Davenport wrote:
>
> Sorry for the delay. Had somethings to take care of.
>
> I've updated the README. For those watching here, Migrant Verde is a 
> schema evolution tool. It supports embedded and command line execution. 
> This means that a Java/JVM application can use it directly and a Python or 
> Ruby app can shell out.
>
> The goal is to automate the process of database lifecycle management. If 
> you have a new developer, the library enables creating a prod mirror 
> locally when the application boots the first time. It allows you to manage 
> your DB changes in source control without having to invent your own 
> migration scheme. Ultimately it aims to be Liquibase for ArangoDB.
>
> I'd be happy work with the Arango team on a blog post. At present a person 
> has to clone the repo and build the jar manually. I'm working on getting a 
> maven central account setup. I'll follow up here when that's done.
>
> To the rest of the community,
> If you don't mind the annoyance of having to build the project before 
> using it, please kick the tires. Hopefully it helps out. If you want to add 
> features like configuring a collection at creation time, please submit a 
> pull request. We've got limited resources so we focus on our immediate 
> needs.
>
> Thanks,
> JPD
>

-- 
You received this message because you are subscribed to the Google Groups 
"ArangoDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to