Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [jakartaee-platform-dev] Scanning specification

Hi,



On Thu, Mar 18, 2021 at 7:46 PM Thomas Bitonti <bitonti@xxxxxxxxxx> wrote:

+1

Jandex provides an approximate solution, but needs more lightweight ways to access index data.

At Piranha we created somewhat of what I think is a simpler API on top of Jandex. It's not super well thought out by any means and the javadocs are not really there yet, but we could use it as an additional source for inspiration.


Requirements become more complicated if the scope of caching becomes larger than a single JAR.

I think at least for a 1.0 we should not really extend beyond what existing solutions like Jandex do today; i.e. the prime goal should be an index per jar, and an API to work with that index.

We probably also want to standardize the (binary?) file format of the index, so users don't have to create an index per vendor in their archives.

Kind regards,
Arjan Tijms




HK2 and Jandex are both inspirations to start from. In HK2, build-time scanning has an important role, and the so-called hk2-inhabitant-generator is often used. See this article from Jonathan that shows this: https://blog.payara.fish/hk2-the-hundred-kilobyte-kernel




On Wed, Mar 17, 2021 at 4:01 PM Thiago Henrique Hupner <thihup@xxxxxxxxx> wrote:

_______________________________________________
jakartaee-platform-dev mailing list
jakartaee-platform-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/jakartaee-platform-dev

Back to the top