Hi Disclamier: I have never used forest! Please don't hesitate to correct me (rudely if needed) if I'm wrong or don't make sense below.
johncch wrote: > > Hey guys, > > I'm trying out the forest options in ME and I realised they don't > really do much other than copy the files over. And after fcloning the > forest, I was unable to "hack" each individual projects. > > I have a proposal to flesh out the forest concept on ME as I feel it > is pretty important for projects that reach a certain scale. > Unfortunately I'm not very proficient in Eclipse plug-in development > so I would like to have the idea evaluated on feasibility and > conplexity. > > Thinking in the workflow sense, I would imagine the things that are to > be added: > 1. In the new project part, in addition to the "clone repository with > Mercurial", we can have another one which is called "clone forest with > mercurial". What this does is that it does a clone to mercurial and > instead adds all the subprojects as projects in eclipse. I think that we shall select a name that indicates to a new mercurial user that this is probably no the item he wants, e.g. it should not scare new user. Could this consedering the UI be an option in the current "clone" wizard? There are drawbacks with this approach also as it will hide this nice feature. > 2. While each subproject appears as a project in eclipse, they retain > metadata on which forest they belong to, so under the team menu, there > would be fpush, fpull and fupdate options available that is basically > just a recursive push or recursive pull/update on all projects. Should the normal team->"update" becom a team->"fupdate" in this case or are both "update" and "fupdate" intersting to do? > 3. Actually in short, what it really is is just to let the top level > project be a container for all related projects so that when we need > to pass projects to other developers it becomes simpler. We can just > tell them please clone project A instead of please clone project A, B, > C, D, E, F, G.. and so on. > I think this sounds much simpler than it actually is. I can already > imagine a few limitations of what it is, first being that this forest > pull destroys the "forest" layer of the forest. I was thinking of a > symbolic project as the forest head but I'm not too sure how it works > out. I don't know how forest deals with nested projects right now and > it would probably be a nightmare. And last but not least there is > probably issues with keeping consistency within the "forest" metadata. If I understand you you want to visualize it something like A (forest empty) B (part of forest A) C (part of forest A) D (part of forest A) When one use your new "clone forest with mercurial" on A then B,C,D is also added? Could it be visualized as A/B /C /D e.g. A is one project and B,C,D is subdirs in that project? (This is how made "forest" look in my mind, but as stated I never used it an almost know nothing about it) I would not sat that either way is the prefered way Im just "tossing my mind out" to try to understand this better. You approach is how we did it in project I worked at with a humongous big project (it was in clearcase and was done my hand) the nice thing there is that you might not want/care of all of the B,C,D project and you can treat them differently. Maybe you even close "C" in eclipse to limit number of files. > Either way, please let me know what you think about it. As this is > probably more of a tool of convenience, and would probably improve my > happiness level, if it is feasible I am willing to spend some time to > hack on it, to make everybody's life better. Great, Just hack away! -- Zingo Andersen (zingo.org and vectrace.com) --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "MercurialEclipse" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/mercurialeclipse?hl=en -~----------~----~----~----~------~----~------~--~---
