hi
in the view
Version Control
tab: Log
So each time I merge master back into my "feature/blabala" branch I kinda lose oversight of what I actually did to that branch, because the commits from master are all over the place
is there a way to filter for only commits directly to thath branch? if I select the right branch in that view, I still get the merge commits all over the place
Just trying to get a sense of best practices.
Yowza, that’s a lot of files. Try again with fewer than 100 files.
anyone knows how to send my files (version control) to github?
Hello!
I know there's are packages in node.js like Husky that enable one to have a pre-commit hook.
I can set up a pre-commit hook on my personal computer, but I'd like to have one via version control so that everyone will have their commits checked.
I'd prefer a method that isn't just telling people in the README.md to set it up via a script or whatever
Any ideas on how to do this? Or is this not possible in flutter?
Hey y'all,
What alternatives would you recommend to using Splice for version tracking? After some research, it seems like there isn't anything similar to that of Github for software developers.
I'm wrestling with splice to stay in sync and it's creating nightmares for consistent versioning across a few collaborators. Splice does not seem helpful for collaborating if it doesn't doesn't make it easier or efficient to track version changes nor does it ensure you don't both:
A) lose work
B) work on old versions without getting newest pushed changes, etc. True version control like Git listens for changes on local machines and forces you to deal with merge conflicts, assigns a "main" branch, etc.
Now, nobody likes a good ol' merge conflict....but that's sorta the point of version control.
Thanks in advance for any suggestions on alternatives to version tracking with splice. Right now just sending the zip file back and forth is simple enough, but something like Git would be very useful.
Seeking updated community feedback and insights on this thanks!
Thanks in advance!
I`ve been struggling to maintain version control of Jupyter Notebooks through pure Git because of all the issues of git dffing detecting cell output changes and stuff.
Do you use any specific tools to keep up a good gitlfow-like version control scheme of your data science jupyter notebooks?
Hey,
I work as a programmer, so I version-control every project I work on, usually with git. I also used git to version-control my kicad projects. The major downside always was, that you cant really do much with the versions(for example: a diff of a .sch file is hard to read) and I had to trust my commit messages, to know what happened in a version.
Today I found a cool tool which solves this issue: https://cadlab.io/ Its basically a git remote-server, which allows you to make visual diffs of your pcbs and schematics. And I already love it and wanted to share my findings.
For me it was always clear, that I use git. But what about you, do you even version-control your projects?
I work alone, but I use a Git repo as a backup and a "super-undo button". I feel that I don't make commits often enough as sometimes I'll work for a few hours and end up committing 20+ changed files in one transaction.
Does anyone have suggestions for a rule-of-thumb for how often commits should be made, when either working alone or in a team? I want to develop good habits for when I start working with others.
Coming from a software engineer background, I've been using git to version control my files, along with a config.json that includes all model and training hyperparameters. A trainer class would read a specified configuration file, train the model, and save the best one with the highest validation score.
This pipeline worked in the beginning when I was iterating at a slow pace. Recently, however, I'm increasingly realizing that saved checkpoint files are obsolete because non-trivial changes have been made to the model architecture since the last experiment. By checkpoints, I'm really referring to torch.load(some_state_dict), but my question is framework agnostic.
I could, of course, check when the experiment was conducted and git checkout the repository to that specific point in time. However, part of me believes that this is an incredibly common thing for ML engineers, and that there must be a more elegant solution. So far, my research has brought me to
- Sacred, which I got from this post
- Keepsake
- DVC, which appears tangential since my question pertains more to models, not data
Thanks in advance for sharing your insight!
Hi everyone,
first of all, that's not a blame post, I just want to give my feedback on one simple but a key feature to me.
I use a lot Visual Studio code for frontend development and one thing that they really did right is the git view. Is simple, not distracting, the file list is on the left and there is plenty of space for checking the diff view. Personally, I think that the intellij view is a bit too cluttered and there are too many options there. Also, I find myself scrolling beetween the file view and then scrolling again the diff view and so on...especially if I'm on my 15" macbook's screen and not using an external monitor.
I would love to see a simplified version of the commit window, something really simple like VSCode where I can in a few clicks add, commit and go!
I have tried dropbox (with its version history feature) and traditional naming and file management strategies. I was wondering if there were any more efficient, user-friendly ways of keeping track of your desktop files and their respective versions? Almost a GitHub for desktop files.
I have in mind a solution where I will have a hosted repository of all my files, and once I select one file (eg 'Sales Deck') it would show me all versions (past commits) of the 'Sales Deck' file, from both myself and anyone else I share that particular repository with.
I know dropbox and (kind of) G Suite have this baked in as a feature, but find it hard to use as the UX isn't tailored towards this feature.
If I can't find anything that better suits my needs, then I may have a go developing something that does!
Would love to hear your thoughts.
I'm fairly new to Gulp (and even NEWER to Git), but today I decided I had lost my last file due to a bad upload. I decided to implement a version control system into my development.
I've got git pretty well figure out. I'm a one-man-band freelance dev, so my requirements are pretty simple. Basically, I want to use Gulp to watch my theme folder for changes. When there are changes I would like Gulp to do a git commit, and to push to my github repository.
Is that possible? I've employed all my available google-fu and haven't been able to find a working solution.
UPDATE: Thanks for all the replies. This really is a great community!
So it does look like I didn't seem to have the right understanding of the 'scope' of Git. What I was looking for initially was a very 'zoomed in' approach to file backups, where every time I saved a file, it was backed up. It's true that that would create rather LARGE backup folders, and Git isn't really supposed to be used for that. It seems to me that Git's scope is more about pushing commits at 'stopping points' in the dev process (ie, like when you're done for the day, or just finished coding a particular section and you're now 'switching gears').
Being that I don't really like the idea of having GBs of backups sitting around, documenting every time I changed 12px to 14px, I think the right way for me to go is to use Git, running commits whenever I get to a meaningful stopping point in the project, or the end of the workday (whichever comes first). That way, if I have some sort of catastrophic file loss, the most work I would ever lose is a day's worth.
If I'm still missing something, or you can suggest a better workflow for my needs, I'm all ears. Thanks again for all the thoughtful input!
Sorry if this is stupid, because this is one of those things in programming that I feel like should be easy, but I just cant seem to understand it. Can someone explain to me in simple terms how version control actually works? I know the vocab like push, pull, commit, merge, branch, etc.., it just gets so confusing without simple terms. If it helps, this is my situation.
For School, I am working on a team project and I was tasked with creating and styling the homepage. I have the completed and working code sitting in Visual Studio Code ready to be put in the repository. I now simply need to get it into the Gitlab Repository we have for the team. There is a master branch as well.
If anyone is really good at explaining stuff, and wants to tackle this I would appreciate it so much!
I have about a year of Python experience under my belt. I haven't used the language in any professional projects, nor have I worked on a team with anyone before.
I have read a few version control tutorials and they really confused me.
Does anyone have a link to a tutorial or blog that explains what GIT hub is, what the heck a repository is, what is a commit, and how version control works, what is a fork? I assume these topics are all related.
I got a ISR 4000 series router. I used to handle juniper before and they have a slick little function call "commit", where after you make changes in conf t you have to type commit <$seconds>(after x seconds it will revert the changes) or commit -a(for final commit). Does cisco router have something similar?
Thanks
Hi everyone,
I've recently been contacted by a music producer of an indie label regarding a desktop application I am building. I am creating a version control software for graphic designers and 2D/3D artists. Think of it as a time machine for graphic projects in which you can go back and forth without ever losing file versions. It basically helps to avoid project filenames like "my_project-final_FINAL-10.psd".
They told me, that the problems I am addressing with my software for the graphics and design industry are very similar to the ones you have in music production. They gave me a basic idea of how music is produced. And from what I have understood, I do agree that the similarities of file workflow problems are strikingly similar.
So I am reaching out to you to get more information on this topic. Is version control a known term in your industry? Is this something your industry is looking for or are there already available solutions?
My project is called Snowtrack www.snowtrack.io. I can't see any obstacles at the moment it not being compatible with music files and music projects, but the user experience is maybe slightly off given its initial target audience.
Let me know what you think. Any feedback or comments around version control are highly appreciated!
Best, Seb
I use git for version control. Things can get hairy when I use git branches. Especially when I'm developing an experimental feature, on a separate git branch, that involves database migrations and then trying to cherry pick certain updates back into the main branch. (It doesn't even have to be cherry picking. Can be just trying to merge in feature branches back into mainline in different order.) Then I end up with migration conflicts because of migration files that depend on other migration files that don't exist. What strategies or tools do people use to cope with this situation?
Thanks.
Do I have to buy the ultimate edition to get the ps5 upgrade ??
My 2021 resolution is to do a better job of keeping my files tidy and organized. My band's projects are full of what seems like endless audio. We record during rehearsals and end up with lots of music. Cool right?
&#x200B;
When it comes to mixing down a selection here or there, I struggle deeply with version control, and I tend to make duplicate copies of entire projects and ALL data (yes those pesky check boxes) out of fear of losing some file or piece of automation data in a separate version that i'll never be able to find again.
&#x200B;
Got a tip that could help me (and others) reach the following goals? Please number and share!
&#x200B;
Goals:
- Relocate and keep Logic Sound Library on External Drive (✓ - Done)
- Save/load all projects to/from external SSD drive #1 (X - I've read about using "alias" folders or "symbolic" folder? What's the best way?)
- Duplicate external drive #1 to external drive #2 once per week (maybe automated somehow?) (X - There must be a simple way to back up one SSD to another without re-transfering back to the computer - also open to CLOUD alternatives but depends on price for TBs of storage and ease of access**)**
- HOW TO SAVE Know when to use "project alternative" versus "save as" - what's the pro tip here?
Been working on a project for 2 weeks and have a tonne of version folders ... cumbersome.
Ideally, I'd like to archive a snapshot of my project files on github -- pushing 1-2 times a day. I'd like to be able to go back and see what changed, or re-use any code that I deleted.
I've got a github repo, downloaded git, and successfully staged and pushed my files to the repo.
But when I tried to push an update, it seems you could only overwrite the previous.
Github won't let me fork my own repo. I tried cloning it, but I don't think that solves anything. I'd have to push to a new repo.
Not sure how to just ... fork my own repo I guess?? Appreciate any help -- thanks.
A primarily creative tool for redstone development, like an advanced copy-paste timeline of an in-game area
Motivation
If you wish to develop or refactor a complex redstone circuit you'll probably destroy and rebuild some aspects of it several times, wasting a lot of time.
When I have an idea for something to add to a compact contraption, I have to remake / clone it somewhere else before making edits, to avoid the risk of destroying valuable progress.
How version control could work
An inventory tool could be used to mark off an area and save the area's state (like a github commit).
A separate tool can then quicksave (git commit) or quickload (git checkout) this area, or make "branches" (for a new timeline of possible save's and load's)
Potential
- open source redstone contraptions via pull requests
- possibility for collaboration on large projects (like an 8bit computer) without need for a server, reducing spaghetti-redstone
- steal ideas from github
I got the game and realized I need to use dualshock trackpad to play, but mine for some reason stopped working. Is any way to play this game without the trackpad?
Hello!
I'm writing a series of two posts on how I've been maintaining many Vivado projects under git over the years. Maybe that is useful for someone in our community :-)
Here goes part 1, focusing on IPs alone:
It shouldn't matter how big or small a girl's tits are in order for you to date her, what should really matter for both girls and guys is personality, personally i like big tities but i look more into personality and tity size is not even on the list of things I'm looking for in a girl
)