DBrEx going crazy :P

[cidico]

Hi guys,

Today I noticed that the DBrEx is going crazy.

I have a branch that was created and a changeset was created by the developer A.

The developer B gets the newly created branch from developer A to his machine. In developer B's machine the DBrEx shows the color of changeset created by the branch A correctly.

Then, I get the branch created by developer A accidentally from the developer B machine. My DBrEx is coloring the changeset from Developer A with the color of Developer B, even if the changeset info points to developer A.

Is it a bug or the expected behavior ?

[jgonzalez]

Hi,

I have to confess that I'm a little bit confused. Please confirm that you have followed the next steps:

1-create branch on machine 1's repository (rep1)

2-create changeset on branch A from rep1.

3- pull branch A from rep1 to rep2 (rep2 is on machine #2). ->Colors on machine 2's rep2 are OK

4- pull branch A from rep2 to rep1 -> branch A's changeset is wrongly colored on rep1 on machine 1

I've performed those steps and changeset colors are fine. Am I missing something?

I've also tried to create a new changeset on branch A on rep2 before step #4 and colors are fine too.

The DBrEx paints changesets depending on the "Replication Source" property. Please check whether the property

states that the replication source is "Not replicated" or is "rep2" for the affected changeset.

Also please check that the colors foreach replication source don't match!

Sorry for asking you some more info...

Kind regards,

-Jesús M.

[cidico]

Hi Jesús!

It's almost what you described, just a slight change in the step 4 that I change to:

4- Pull branch A from rep2(Machine #2 instead Machine #1) to rep1 (#Machine#3).

When I'm pulling updates from DBrEx (right clicking on an branch > Replication > Pull this branch) Plastic shows a new window where you set the source and then click in replicate.

Sometimes the source's value is already filled by Plastic itself using some kind of cache from the last time, I guess.

This time I tried to do this, but accidentally I choose the wrong machine to get the code. Not a problem because the developer had the latest branch, as I wanted.

I noticed a "something is wrong here" with the colors because he didn't change nothing in that branch, but the changesets were painted with his color from Replication Source.

I guess that it's not a bug, since the DBrEx uses the color from replication source which, by the way, uses a color specified to a machine instead of a color by contributor.

Would be really nice to associate a color in DBrEx with a Machine and be possible to associate that color to a contributor too.

Thanks anyway and sorry to waste your time with a not understood feature.

[jgonzalez]

You're welcome

And thanks for your suggestion!

-Jesús M.