I am using Mercurial, but I imagine that any merge tool that is aware of the version control system below it could do several things that a merge tool which is not aware of the version control system and only sees two "files" in two different folders, could never do.
I have been using KDIFF3, and recently tried BeyondCompare, and neither of them will do this, at least not that I could figure out.
What I want to do is best shown in this picture, an annotation column and perhaps even ability to open other windows from those annotation columns so I could browse specific versions of specific files to see context when trying to do a merge.
In the image here, I am showing a two way merge, but the same applies for a three way merge. To the right or to the left of the actual file content being shown, I would like a gutter or a right side annotation column showing some kind of annotation of where this change came from. Since Mercurial hex ids are relatively unfriendly and unhelpful, and since repository-local-revision-numbers are repository local, I think that a short text description based on commit comments would be most helpful. Of course, with Mercurial, 99% of these commit comments are going to say "Merge", and nothing else. (Groan.) But lets pretend for a minute that we weren't using tools and workflows that left us that crippled at merge time, and instead, that we could have a useful commit comment show up each time:
Right now the workflow for complex merges looks like this for me:
Using my distributed version control tool (mercurial), pull changes from another repository which is in effect a branch. Merge. The merge window for TortoiseHg is usually where I start all this from. This in turn lets me configure a merge tool (beyond compare or Kdiff3).
However, it does not appear that there is any merge tool (that I have seen) that can be told, "hey you're not just merging two way or three way with different versions of a file in the two completely different folders, with the names I told you, but those files are also files that have a complete edit history available to you to show your human the actual context, the commits that those line changes came from with their commit comments, often having a bug number as part of the commit which will give the person doing the merge the ability to see What in the Heck is Really Going on.
I would change from Mercurial to Git, for example, even, for a real merge experience that didn't force me to do manually what I think my tools could be doing for me automatically. I'm using Mercurial, TortoiseHG, and KDIFF3, and if I could just change from KDIFF3 to some other tool, or do ANYTHING at all to get annotations and merges together on one screen, I would like to do so.
Related
I'm wondering if there any plugins for Bitbucket that allow you to filter out certain kinds of changes from the diffs? What I have in mind would probably need to be "sensitive" to the coding language
There's a wide plethora of things I wish I could (dynamically) choose to exclude including:
Inserting a "final" keyword in Java
Changes to comments
Splitting or combining lines (nominally a "whitespace" change, but not recognized as such by most diff utilities)
Reordered Java imports
Any changes to Java imports
Moving entire methods or field declarations (without changing them)
Almost any change that where the "new" is semantically equivalent to the "old"
Now folks, I hope we aren't going to get get into a debates about whether or not ignoring such diffs is a good idea. What I would like to see is a filter where these things can be turned on and off. Not any sort of permanent and fixed suppression of such things.
I would argue that sometimes/often it is handy for the diffs to focus in on "the semantic" changes.
My specific needs is for Bitbucket pull requests, but I'm kinda interested to hear about such things for any "diff" utility, including general utilities not specific to Bitbucket, nor even pull requests.
I am using mercurial via tortoiseHg (windows) as a source control management tool.
I am used to merge using beyond a compare. Today, I have to perform a very complex merge and I just discovered a new feature (my client was updated some days ago) that is extremely annoying.
When I have a conflit and ask Mercurial to take the "other" file and keep the original in a .orig file, the .orig is added with <<<<<<< local and >>>>>>> other, but more than this, the other part is merged into the original one !!!
The two parts are then unaligned and it's impossible to guarantee that the merge is OK because you have to review it line by line with no help from the comparision tool. (see screen below).
http://s13.postimg.org/yor6gno47/Untitled.jpg
I want to disable this feature, but so far, I am unable to do it. Thanks so much for help as this is furthermore blocking my work.
Regards.
The launching of a specific merge tool isn't something Mercurial controls. It does, however, have a robust mechanism for Merge Tool Configuration that allows you to provide a preference order and it will use the first one it can find. The builders of various Mercurial installation packages (ubuntu, etc.) and tools that include Mercurial (TortoiseHG, etc.) all provide their own Merge tool configuration preference list.
Either the old merge tool configuration list you had not longer points to Beyond Compare at the right location (upgraded BC and the directory name changed, etc.) or you got a new merge tool configuration list when you updated some software that included mercurial. Either way that page on MergeToolConfiguration will help you find your preference list in your hgrc files and update or correct it.
Tl;Dr: this isn't a "new" feature it's your new installation being less tailored for your system than your old one. Maybe find who packaged that one and copy the merge tool config.
I'm curious to get people's thoughts on how to manage version control for unrelated functions in Matlab.
I keep a reasonably large set of general purpose scripts, each of which is more or less independent of the others. I've been keeping them all in a single directory, containing a single repository in Mercurial. I'm starting to collaborate much more, and I'd like my collaborators to be able modify the files, commit, branch, and merge.
The problem is that the files are independent of one another. Essentially, they're like many separate little projects. But Mercurial treats the repository as a single entity. So if a collaborator modifies file A and B, and I only want to merge in the changes from file A, things get complicated. I know that I could merge from the collaborator, then revert file B, but I'm wondering if there's a simpler way to handle this setup.
I could set up many tiny repositories to manage each file separately, but that also gets complicated.
I'm open to changing version control systems (although I like Mercurial a lot). Any suggestions?
It is considered a best practice to check in code after each bug fix/feature addition/or what not. Given your files are really independent "projects" it seems unlikely a bug or feature would span multiple files. Probably the best you can do is encourage your colleagues in best practices to commit changes only for a single file at once. Explain that better discipline about checking in leads to more manageable source control later. Hopefully you can get most to follow the practice and the few obstinate ones just stop taking their commits for.
It really depends on your typical reasons for merging one change but not the other. If you're using it to create a software configuration, i.e. sometimes you want to use version 1 of file A and version 2 of file B and sometimes it's the other way around, then you probably want to use subrepos to hold each file. If it's because you never want to accept part of a collaborator's change, then they need to be instructed how to make their changes more cohesive and submit them separately. That can sometimes be a difficult concept for people who either haven't used source control before, or who are accustomed to source control like svn that has little or no intrinsic concept of a changeset.
It depends whether you want to maintain a single 'master' version of the files, merging in changes that you like and ignoring others. If collaborators want to develop other branches, then they should perhaps clone the repository, and you can then accept the changesets that you want in the master.
If you want to veto changes by other collaborators, then the changes either need to be kept separate (via a cloned repository or branch) or you need a review process before changes are pushed back to the trunk.
I always use incoming repositories for collaborators. They match what the other person has made, but it avoids messing with my own repository. When you do this, you can then cherrypick their new changesets into your own repository with the transplant extension.
I'm learning mercurial as my solo scm software. With other management software, you can put change comments into the file header through tags. With hg you comment the change set, and that doesn't get into the source. I'm more used to central control like VSS.
Why should I put the file history into the header of the source file? Should I let mercurial manage the history with my changeset comments?
Let the source control system handle it.
If you put change details in the header it will soon become unwieldy and overwhelm the actual code.
Additionally if the scm has the concept of changelists (where many files are grouped into a single change) then you'll be able to write the comment so that it applies to the whole change and not just the edits in the one file (if that makes sense), giving you a clearer picture of why the edit was required.
Yes; let the source control system handle your changeset comments. The rationale for this is that it makes considerably more sense when you're viewing the change log later, trying to work out what's going on between two versions of a file - the source control system can present the change comment to try and enlighten the situation.
There's no reason to manually maintain a file history when SCM software is much better suited to solve this problem. All too often I see partially-completed file histories in the source, which actually hurts, because people incorrectly assume it is accurate.
The difference is not whether it's a centralized or distributed VCS, it's more about what's being changed.
When I moved to .Net, the number of files updated for any individual change seemed to skyrocket. If I had to log the change in each file, I'd never get any real work done. By commenting on the set of changes, it doesn't matter how many files I had to update.
If I ever needed to identify all of the changes for a particular change, I can diff between the two versions of the project.
The biggest difference (and advantage) I saw when switching away from SourceSafe was the switch from file based to project based commits. As soon as I got used to that, I stopped adding change-log type comments to all of my files.
(As a side effect, I've found that my process description comments have gotten better)
I'm not a big proponent of littering the code with change comments. In the event they are needed they can be looked up in the SCM (at least for the SCM variants I have used). If you do want them in the file, consider putting them at the end instead of the beginning. That way you won't have to scroll down past the (uninteresting, to me at least) comments before you get to the actual code.
Another vote for letting the SCM system handle the checkin comments, but I do have one thing to add.
Some systems allow you to use RCS tags in your source code where the SCM can insert the change history directly into the source file being committed automatically. Sounds like a nice balance because the history is then in the SCM system and then automatically put into the source code itself.
The problem is that this process changes the source file. I think that's a bad idea because the file cannot be changed on disk until after you comment is inserted. If you were a good engineer, you should have built and tested changes before the commit. If your source changes after the commit, then you've essentially got a build that could be broken - but most engineers won't build after a commit - why should they?
But it's just a comment you say! True, but I did have a case where there was code in my source file that strangely enough had reason to look like an RCS header tag and that section of the code got replaced on checkin, thereby munging my code. Easy enough to fix, but bad that a build got broken for 20+ users
Much easier to forget to maintain history in the source, as one always (imo) should comment commits to source control system that problem dissappers. Also if changing lots of files before commit, changing history in every file will be annoying work. This is really one of the points with having scm.
I have experience with this. I've had the file history in the comments, it was awful. Nothing but garbage, sometimes you would have to scroll down almost 1k lines of code changes before you finally got to what you wanted. Not to mention, you're slowing down other aspects of your build process by adding more kb to your source code tree.
It finally started to annoy me enough to ask this question: how do I do a basic diff between two revisions of a file in CVS? Usually I want to compare the latest revision and some random old one. I'm using the Eclipse CVS plugin. When I use "compare with->Another branch or version..." from the selected file's (latest revision from HEAD or another branch) context menu, I get a list of branches, tags and dates but not revisions. Usually I have just created a date which I know is far enough in the past so I can compare the needed revisions but I thought that there must be a better way.
The answer is to show the file's history using context menu->Team->Show history, then choose two revisions and context menu for the selection->compare with each other.
There seems to be two main ways:
context menu->Team->Show history
which shows a linear history and you can select and compare between them, however it can be very bloated and hard to read when your project has lots of branches / tags. Personally i have found it less useful than:
context menu->Team->Show Commit history
Which seems to show the history of what has been committed to the specific branch/tag you are on. You can do it per file or per folder. The output is very similar but i find it clearer. You can click on a commit date and it will show you all the files (that you are interested in) that were committed on that date.
If you double click the file, it will then bring up another menu so that you can compare it with another file in the commit history
EDIT
(i find if you double click the "other" file, it doesn't do anything, you need to click "OK" in the dialogue, which seems silly to me. This might be effected by the fact I have the beyond compare 3 plug in, im not sure if it behaves the same without it)
EDIT
There is also a little button in the top right of the commit history window that allows you to switch to history view (but i always find it easy to read than the normal history view if i do it this way round)
Both should show you the comment added when committed and you should try and read about the differences between the but personally I haven't and its only form personal experience that i prefer commit history.
I apologize for not giving formal descriptions of each, this is purely from my personal experience of using them, i have not actually researched them both yet myself...