![]() So if you grabbed the latest, edited one file, but it shows 100 files were changed, that's OK, as you'd only be adding your one file: It won't accidentally add any files you don't tell it to add. It will only add the files that you explicitly tell it to add. Well, that's another argument for the command line. I have to be very careful to revert and remove anything I haven't personally changed every time I check in People who aren't programmers aren't programmers.īut the command line isn't programming. I'd argue that's way easier than trying to learn sourcetree. If that's all you do most of the time, that's 4 commands you need to know. pushing the branch with your committed changes Understanding how a visual UI is translating your button clicks into a command line is, IMHO, way more complicated to understand than just learning the handful of command line commands most people will need to know.įor the work you're doing, I imagine, 90% of the time, you'll be: The command line isn't very technical at all as it's you interacting with git directly. not sure how sound software program will manage this aspect.Īs someone that has used both command line and sourcetree, and as someone that really hates git-even after using it for several years now, I can say that the command line is WAY easier to get the hang of than Sourcetree is.Īnd sourcetree is often see as one of the more usable visual clients.Īt our company, when a branch goes awry, 9 times out of 10 it was due to a developer (yes, a technical person) screwing it up by clicking the wrong things in sourcetree. doesn't work as well with sound waves files. Lastly, git works best when using text files. sublime also has a nice merge tool with a generous trial tier. i haven't used visual studio's graph extension. ![]() Then: gitk and git kraken are what i would recommend. ![]() Otherwise, what's the point of using git?īut ok you don't want to use the command line, nor github's desktop client (which is super sleek btw). When you do, you will need to be somewhat comfortable with digging into the reflogs to undo these issues. secondly, chances are you will eventually have a mishap. i would even venture to say, that if one has a good conceptual understanding of git, it is highly unlikely that a UI client would be preferred for 95% of the use cases out there. The only times I would argue that a git UI client is better, is when you are trying to handle complex merge conflicts or if you want to visualize complex branching, or to visualize other changes over time. I would extrapolate further: for most cases, it's infinitely easier to use the command line, rather than a UI. you don't need to get super technical: just bare-basics. I don't think git can be used without a good conceptual understanding/model of how it works - simply using a nice UI client will not really solve the problem. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |