I haven't completely grokked the entire FSGit experience, however the piece that has intrigued me with the filetree repository that Otto has done is the fusion between monticello and "git" (technically filetree doesn't depend upon git, it just stores monticello package information in a structured directory tree). So filetree can be used as the basis for integrating any file-based SCM into Smalltalk...personally I will focus on git...
Basically, I have taken myself up on the challenge to provide a git workflow experience for Smalltalk:) and my approach is to create a fusion of Monticello, Metacello, and git... For an alternate SCM to get traction within the community, I think that there needs to be easy migration between this alternate approach and Monticello and for me filetree was the missing piece ... If you want to see what I am currently up to, take a look at the metacello project on github[1]. Take a look at the network view[2] of the project. I've pushed the Metacello code for 1.0-beta.31[6] (the latest release) into the repository using filetree[3], essentially a fork of Otto's smallsource project[4]. I've also pushed the code for my current work-in-progress for 1.0-beta.32[5]. Finally I created the MetacelloGitProject[7] branch for the project, where I have started doing development (one must eat their own dog food!) using github as the sole repository for my code (other than local clones on my work desktop machine and my home laptop). You can look at one of my commits[8] from last night where you can see a nice difference view that actually involves changes to 4 monticello packages ... I do a fair amount of the actual loading manually as I'm still trying to get a feel for using git in development...FSGit looks very interesting as well, if we are going to use git, then having a Smalltalk interface to git is a requirement ... filetree is SCM neutral, but I am building an extension of filetree that is aimed at git (in git, the repository is versioned not the individual files, so it is an interesting challenge to integrate into Monticello). I am very encouraged that the fusion of Monticello, Metacello and git is coming together pretty nicely (i.e., without tons of work)...this is very early work, so I don't recommend anyone actually try to use the project (it is only 2 days old:) as there is a fair amount of hand holding that I have to do right now ... However, if you are interested in contributing then I encourage you to fork the metacello project on github[1] and get busy. I will report progress and solicit feedback in general on the metacello mailing list[9], but if there is interest in collaborating we'll just use github for the nitty gritty communication. Dale [1] https://github.com/dalehenrich/metacello [2] https://github.com/dalehenrich/metacello/network [3] https://github.com/dalehenrich/filetree [4] https://github.com/finworks/smallsource [5] https://github.com/dalehenrich/metacello/tree/1.0-beta.32 [6] https://github.com/dalehenrich/metacello/tree/1.0-beta.31 [7] https://github.com/dalehenrich/metacello/tree/MetacelloGitProject [8] https://github.com/dalehenrich/metacello/commit/b79df4a2cd63200a8a1712fe629dbcc8c6d0a3a7 [9] http://groups.google.com/group/metacello ----- Original Message ----- | From: "Damien Pollet" <[hidden email]> | To: [hidden email], "Bruni camillo" <[hidden email]> | Sent: Wednesday, February 1, 2012 6:44:37 AM | Subject: Re: [Pharo-project] [squeak-dev] Re: [Smalltalk for small projects only? | | On 31 January 2012 15:59, Otto Behrens <[hidden email]> wrote: | > I was wondering more if it is possible to do without having a file | > system checkout of the class definition. In other words, if the | > class | > definition in the image could serve as the working copy of the | > object | > in the git repository. I must read more on what the other guys did | > with the git interfaces (such as gitfs and gitosis) to understand | > what | > it comes down to. | > | > If this is possible, it would make merging, loading and committing | > essentially one step. | | Have you seen what Camillo did? | http://www.squeaksource.com/FSGit | https://github.com/dh83/fs-git-test | | Note the layout with class comment in a README file, etc :) | | There is a also nearly working implementation of the git network | protocol (stubbed using the git binaries ATM). | | | > The disadvantage could be a tight coupling with git. | | is that really a disadvantage, compared to a tight coupling with | monticello? | | | -- | Damien Pollet | type less, do more [ | ] http://people.untyped.org/damien.pollet | | |
Getting deeper in FSGit and it looks like it has a Monticello mapping ... now to work:)
Dale ----- Original Message ----- | From: "Stefan Marr" <[hidden email]> | To: [hidden email] | Cc: [hidden email], "Bruni camillo" <[hidden email]> | Sent: Wednesday, February 1, 2012 11:02:16 AM | Subject: Re: [Pharo-project] [squeak-dev] Re: [Smalltalk for small projects only? | | | On 01 Feb 2012, at 19:45, Dale Henrichs wrote: | | > I haven't completely grokked the entire FSGit experience, however | > the piece that has intrigued me with the filetree repository that | > Otto has done is the fusion between monticello and "git" | > (technically filetree doesn't depend upon git, it just stores | > monticello package information in a structured directory tree). So | > filetree can be used as the basis for integrating any file-based | > SCM into Smalltalk...personally I will focus on git... | > | > Basically, I have taken myself up on the challenge to provide a git | > workflow experience for Smalltalk:) and my approach is to create a | > fusion of Monticello, Metacello, and git... | | Cool! | | Camillo, did you plan be in Brussels on Saturday for the sprint? | Would that be something to pair on for a few hours? | | Best regards | Stefan | | | | -- | Stefan Marr | Software Languages Lab | Vrije Universiteit Brussel | Pleinlaan 2 / B-1050 Brussels / Belgium | http://soft.vub.ac.be/~smarr | Phone: +32 2 629 2974 | Fax: +32 2 629 3525 | | | |
Free forum by Nabble | Edit this page |