[Bioc-devel] Confusion with how to maintain release/devel files on local computer.
Thank you for the detailed description, it is much easier to understand what is going on. I have made some comments below
On 11/02/2017 11:14 AM, Arman Shahrisa wrote:
> There is no branch releas_3_6.
I don't know what 'it' is -- I guess your local repository that you cloned, but where did you clone it from?
I repeated all the process again. I opened git bash as an administrator and ran commands described here:
it is not necessary or recommended to run these commands as an administrator; generally all activities should be done as a regular user. Likely you need, as a regular user, to regenerate your ssh key pair, and to re-submit the public key to Bioconductor. Alternatively, perhaps as a regular user you already have an ssh key pair, and you can submit the public key part of the key pair. Remember that this will take at least 24 hours to process.
This workflow means that you want to maintain only the Bioconductor version of your repository, you do not want to maintain the github version of your repository that you used during package submission.
> Arman at Arman-VAIO MINGW64 ~
> $ cd /z/cbaf/Source
> Arman at Arman-VAIO MINGW64 /z/cbaf/Source
> $ git clone git at git.bioconductor.org:packages/cbaf
> Cloning into 'cbaf'...
> Enter passphrase for key '/c/Users/Arman/.ssh/id_rsa':
> remote: Counting objects: 732, done.
> remote: Compressing objects: 100% (218/218), done.
> remote: Total 732 (delta 503), reused 725 (delta 499)
> Receiving objects: 100% (732/732), 749.19 KiB | 182.00 KiB/s, done.
> Resolving deltas: 100% (503/503), done.
> Arman at Arman-VAIO MINGW64 /z/cbaf/Source
> $ git remote -v
> fatal: Not a git repository (or any of the parent directories): .git
> Arman at Arman-VAIO MINGW64 /z/cbaf/Source
> $ cd /z/cbaf/Source/cbaf
> Arman at Arman-VAIO MINGW64 /z/cbaf/Source/cbaf (master)
> $ git remote -v
> origin? git at git.bioconductor.org:packages/cbaf (fetch)
> origin? git at git.bioconductor.org:packages/cbaf (push)
ok, all of the above looks good. You cloned the repository, and your ssh key was recognized.
> Arman at Arman-VAIO MINGW64 /z/cbaf/Source/cbaf (master)
> $ git pull
> Enter passphrase for key '/c/Users/Arman/.ssh/id_rsa':
> Already up-to-date.
The I ran this one:
> git remote add upstream git at git.bioconductor.org:packages/cbaf.git
This step is not required. It is not part of the workflow you cited. Earlier, when you issued the command 'git remote -v' git indicated that you already had a remote named 'origin' and referencing the git server git at git.bioconductor.org. What the command above does is to add a second remote pointing to git.bioconductor.org, but this one named 'upstream'. I can see that the documentation in this work flow needs to be adjusted, because in other It would be appropriate to rename the remote, e.g., git remote rename origin upstream nonetheless, issuing the command should not cause a problem. It seems like at this stage you should continue with the workflow that you started, i.e., - 'Commit changes to your local repository' to develop a new feature or fix a bug, including checking that your code changes are correct by running R CMD build and R CMD check using the 'devel' version of Bioconductor, and - 'Push your local changes to the Bioconductor repository' to make your code changes available to the Bioconductor build server. Remember that builds run once a day, and that you need to confirm that the build has been successful by visiting http://bioconductor.org/checkResults/. The steps below are not necessary for you at the moment. Martin
Then these commands from c
> Arman at Arman-VAIO MINGW64 /z/cbaf/Source/cbaf (master)
> $ git checkout master
> Already on 'master'
> Your branch is up-to-date with 'origin/master'.
> Arman at Arman-VAIO MINGW64 /z/cbaf/Source/cbaf (master)
> $ git fetch upstream
> Enter passphrase for key '/c/Users/Arman/.ssh/id_rsa':
> From git.bioconductor.org:packages/cbaf
> ?* [new branch]????? RELEASE_3_6 -> upstream/RELEASE_3_6
> ?* [new branch]????? master? ????-> upstream/master
> Arman at Arman-VAIO MINGW64 /z/cbaf/Source/cbaf (master)
> $ git merge upstream/master
> Already up-to-date.
I have attached an screenshot of what I have now. The following is the Command-line output:
> Arman at Arman-VAIO MINGW64 /z/cbaf/Source/cbaf (master)
> $ git branch -a
> * master
> ??remotes/origin/HEAD -> origin/master
> ??remotes/origin/RELEASE_3_6
> ??remotes/origin/master
> ??remotes/upstream/RELEASE_3_6
> ??remotes/upstream/master
Thank you very much for your help. Best regards, Arman *From: *Martin Morgan <mailto:martin.morgan at roswellpark.org> *Sent: *Thursday, November 2, 2017 17:58 *To: *Arman Shahrisa <mailto:shahrisa.arman at hotmail.com>; bioc-devel <mailto:bioc-devel at r-project.org> *Subject: *Re: [Bioc-devel] Confusion with how to maintain release/devel files on local computer. On 11/02/2017 10:11 AM, Arman Shahrisa wrote:
I can see two branches master and releas_3_6. I created GMaster to
Again, you need to be very precise here. There is no branch releas_3_6. I don't know what 'it' is -- I guess your local repository that you cloned, but where did you clone it from? I don't know what you mean by 'created GMaster'
connect it with GitHub. But as far as it looks like, it?s only a local directory.
the cloned repository is a local directory, yes.
Also, I don?t know weather I can remove it or not. Should upsteam appear here?
If you remove it, then any changes not 'pushed' to an online repository will be lost. 'upstream' would appear with ?? git remote -v if you had added an 'upstream' remote ?? git remote add upstream git at git.bioconductor.org:packages/your-package or if you had renamed an existing remote ?? git rename origin upstream 'origin' and 'upstream' are just names; you can call remotes whatever you want. It would be helpful to report as carefully as possible what your configure and commands are. For instance instead of your message, you could say I can see two branches $ git branch? -a ?? RELEASE_3_6 * master I have created a remote connecting to GitHub $ git remote -v origin? git at github.com:Bioconductor/your-package (fetch) origin? git at github.com:Bioconductor/your-package (push) upstream??????? git at git.bioconductor.org:packages/your-package (fetch) upstream??????? git at git.bioconductor.org:packages/your-package (push) and so on. Martin
Best regards, Arman *From: *Martin Morgan <mailto:martin.morgan at roswellpark.org> *Sent: *Thursday, November 2, 2017 05:00 *To: *Arman Shahrisa <mailto:shahrisa.arman at hotmail.com>; Stephanie M. Gogarten <mailto:sdmorris at uw.edu>; bioc-devel <mailto:bioc-devel at r-project.org> *Subject: *Re: [Bioc-devel] Confusion with how to maintain release/devel files on local computer. On 11/01/2017 07:53 PM, Arman Shahrisa wrote:
I cloned the package into a clean directory.
It helps to be more precise here. Were did you clone the package from? In the simplest case it would be ? ?? git clone git at git.bioconductor.org:packages/your-package and you'd have a single remote named 'origin' ? ?? git remote -v You can add or rename remotes, e.g., ? ?? git remote rename origin upstream ? ?? git remote add origin git at github.com:your-name/your-package and you can pull (or fetch and then merge) from each ? ?? git pull upstream? # pull changes from 'upstream' remote ? ?? git pull origin??? # pull changes from 'origin' Study the how-to documents for different scenarios, and be precise as possible with what you are trying to do. Also, when learning, it can be comforting to know that removing your local repository from your disk is not the end of the world, you can just start over with a new clone from git.bioconductor.org or from github.
Then I followed ?New package workflow? strategy. By checking Description file while switching between branches, I can easily see the version change which is correct. If I understand it correctly, GitHub can only contain changes I make to master branch. If I want to push changes to GitHub as well, what should I do?
You can do whatever you want with your local clone, and with github. The Bioconductor git repository allows you to push only to 'master' and the current release, RELEASE_3_6. Suppose you'd made changes to master, and you named the remotes 'origin' (for github) and 'upstream' (for bioconductor). Then after after adding (git add ...) and committing (git commit ...) to the local repository, you'd push to bioconductor with ? ?? git push upstream master and to github with ? ?? git push origin master Martin
Best regards, Arman From: Stephanie M. Gogarten<mailto:sdmorris at uw.edu> Sent: Thursday, November 2, 2017 02:19 To: Arman Shahrisa<mailto:shahrisa.arman at hotmail.com> Cc: bioc-devel<mailto:bioc-devel at r-project.org> Subject: Re: [Bioc-devel] Confusion with how to maintain release/devel files on local computer. One possible point of confusion: Laurent's workflow includes maintaining separate branches "master" and "devel", which he syncs to his own Github repo and Bioconductor's git repo respectively. However, the documentation on the bioc website (https://bioconductor.org/developers/how-to/git/) assumes that you have only one "master" branch that you push to both remotes. On 11/1/17 2:33 PM, Laurent Gatto wrote:
On? 1 November 2017 20:36, Arman Shahrisa wrote:
I'm confused with development process. At first, I need to have a folder with accepted packaged. Then I need to pull origion RELEASE_3_6? Then in another folder, I need to pull origion master?
No, it all happens in the same folder, but switching between branches using git. Here's an example of one of my own packages. The first command list all available branches (all, using -a, means also remote-only branches). My current branch is noted with an *, and I also have a feature branch called writeMSData, which also lives on GitHub (https://github.com/lgatto/MSnbase/, but that's optional). $ git branch -a ???? devel * master ???? writeMSData ???? remotes/origin/HEAD -> origin/master ???? remotes/origin/centroiding ???? remotes/origin/fixBracketSubset ???? remotes/origin/issue82 ???? remotes/origin/master ???? remotes/origin/orbifilter ???? remotes/origin/processingData ???? remotes/origin/removePrecMz ???? remotes/origin/writeMSData ???? remotes/upstream/RELEASE_2_10 ???? remotes/upstream/RELEASE_2_11 ???? remotes/upstream/RELEASE_2_12 ???? remotes/upstream/RELEASE_2_13 ???? remotes/upstream/RELEASE_2_14 ???? remotes/upstream/RELEASE_2_8 ???? remotes/upstream/RELEASE_2_9 ???? remotes/upstream/RELEASE_3_0 ???? remotes/upstream/RELEASE_3_1 ???? remotes/upstream/RELEASE_3_2 ???? remotes/upstream/RELEASE_3_3 ???? remotes/upstream/RELEASE_3_4 ???? remotes/upstream/RELEASE_3_5 ???? remotes/upstream/master As you can see (and as specified by Gabe in his earlier reply), I haven't have pulled all Bioconductor releases. master points to GitHub's origin/master branch, and devel points to Bioconductor's upstream/master. As you can see above, I haven't got the latest release references yet. I can do this with $ git fetch --all Fetching origin Fetching upstream remote: Counting objects: 6, done. remote: Compressing objects: 100% (6/6), done. remote: Total 6 (delta 4), reused 0 (delta 0) Unpacking objects: 100% (6/6), done. ?? From git.bioconductor.org:packages/MSnbase ??? * [new branch]????? RELEASE_3_6 -> upstream/RELEASE_3_6 ????? b680678..a98138c? master???? -> upstream/master And now $ git branch -a ???? devel * master ???? writeMSData ???? remotes/origin/HEAD -> origin/master ???? remotes/origin/centroiding ???? remotes/origin/fixBracketSubset ???? remotes/origin/issue82 ???? remotes/origin/master ???? remotes/origin/orbifilter ???? remotes/origin/processingData ???? remotes/origin/removePrecMz ???? remotes/origin/writeMSData ???? remotes/upstream/RELEASE_2_10 ???? remotes/upstream/RELEASE_2_11 ???? remotes/upstream/RELEASE_2_12 ???? remotes/upstream/RELEASE_2_13 ???? remotes/upstream/RELEASE_2_14 ???? remotes/upstream/RELEASE_2_8 ???? remotes/upstream/RELEASE_2_9 ???? remotes/upstream/RELEASE_3_0 ???? remotes/upstream/RELEASE_3_1 ???? remotes/upstream/RELEASE_3_2 ???? remotes/upstream/RELEASE_3_3 ???? remotes/upstream/RELEASE_3_4 ???? remotes/upstream/RELEASE_3_5 ???? remotes/upstream/RELEASE_3_6 ???? remotes/upstream/master If I want to modify the development branch (i.e. Bioconductor's upstreams/master), then I checkout devel (that's how I named it locally), do changes and push. $ git checkout devel ## do stuff $ git push Same principle for other branches.
So that by opening each folder, I know what I'm editing. Also during push, I need to be careful about where I'm pushing changes. Origion is bioc's git address of my package whereas master is the package directory in GitHub?
No - I suggest you read a bit about git (GitHub is a web interface using git) to familiarise yourself with the concepts and vocabulary.
Am I getting it correct? Is there anywhere that contains whole the process and codes in steps?
All the setup and more details are provided in
https://github.com/bioconductor/bioc_git_transition/
in particular the FAQ and all the scenarios at the bottom
https://github.com/Bioconductor/bioc_git_transition/blob/master/doc/faq.md
Best wishes,
Laurent
Best regards, Arman ??????? [[alternative HTML version deleted]]
_______________________________________________ Bioc-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/bioc-devel
??????? [[alternative HTML version deleted]]
_______________________________________________ Bioc-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/bioc-devel
This email message may contain legally privileged and/or confidential information.? If you are not the intended recipient(s), or the employee or agent responsible for the delivery of this message to the intended recipient(s), you are hereby notified that any disclosure, copying, distribution, or use of this email message is prohibited.? If you have received this message in error, please notify the sender immediately by e-mail and delete this email message from your computer. Thank you.
This email message may contain legally privileged and/or confidential information.? If you are not the intended recipient(s), or the employee or agent responsible for the delivery of this message to the intended recipient(s), you are hereby notified that any disclosure, copying, distribution, or use of this email message is prohibited.? If you have received this message in error, please notify the sender immediately by e-mail and delete this email message from your computer. Thank you.
This email message may contain legally privileged and/or...{{dropped:2}}