Use of default value for for properties using the Key Expansion mechanism

[andreipet]

We have to get default values even for properties using the Key Expansion mechanism.
I see 2 alternatives:

1. We have to try get the value for unexpanded key when the expanded key is not defined in source.
2. Fill the properties manager with default values for all expanded keys, then add values found in source.

I had chosen the first solution.

[coveralls]

Coverage increased (+0.01%) when pulling 9ee32ee on andreipet:master into ddfdfc0 on lviggiano:master.

[lviggiano]

Thanks. I'll look into it asap.

Luigi.

[lviggiano]

The unit test looks correct, so the use case you spotted should work as you expect. But I'll try to get a more linear implementation, based on your commit, before merging. Thanks.

[lviggiano]

Hi @andreipet,

I've checked your code, but I need some more time to check the consistency of other features. Give me some more time, in this period I'm busy with work and personal life.

Thank you again for your contribution.

L.

[andreipet]

Sure man. Take your time. Thank you for your reply.
On Jul 15, 2014 4:01 PM, "Luigi R. Viggiano" notifications@github.com
wrote:

Hi @andreipet,

I've checked your code, but I need some more time to check the
consistency of other features. Give me some more time, in this period I'm
busy with work and personal life.

Thank you again for your contribution.

L.

—
Reply to this email directly or view it on GitHub.

[andreipet]

By the way. When you will have time, we could talk about performance. ☺
I've started recently to develop in Java. But looking in owner code, I
found a way to deliver properties 10x times faster at least. Yeap , cache
works for read only, or even writable properties. CU
On 15 Jul 2014 17:21, "Andrei Petrovici" andrei.petrovici@gmail.com wrote:

Sure man. Take your time. Thank you for your reply.
On Jul 15, 2014 4:01 PM, "Luigi R. Viggiano" notifications@github.com
wrote:

Hi @andreipet,

I've checked your code, but I need some more time to check the
consistency of other features. Give me some more time, in this period I'm
busy with work and personal life.

Thank you again for your contribution.

L.

—
Reply to this email directly or view it on GitHub.

[lviggiano]

I invite you to open an improvement request and explain the code spot where you found the performance bottlenecks.

What I learned about performance tuning is: first you profile, then you optimize: http://c2.com/cgi/wiki?ProfileBeforeOptimizing

Understanding performance hot spot just by looking at the code, is 99% of the times deceiving.