Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Magit 3.0

emacsair.me

Post reply on HN
Print view View original

121–130 of 144 posts

Re: Magit 3.0

#121

Post by jackewiehose »

dan-robertson wrote: post #119
nonbirithm wrote: post #105

I wish Git would have some kind of command-server mode or libgit2 was supported with Magit to prevent needing to create new processes on Windows. Magit sometimes creates dozens of Git processes for a single operation, but because process creation is much slower on Windows than Linux, it makes Magit unusable for me on those systems. (Sadly, I don't always get a choice to just use Linux.) Simple things like magit-statu…

Magit already has some basic support for libgit. Browsing the issues I get the impression that if someone made a reasonable pull request to support libgit2 for a few dozen methods (should be easy to prioritise based on profiling) then it would be accepted. I don’t think the maintainer wants to review pull requests for one function at a time. I think one issue is perhaps that windows performance isn’t a particularly h…

> isn’t process creation more reasonable with WSL?

In my experience WSL is overall much slower than cygwin.

Re: Magit 3.0

#122

Post by dang »

Some past related threads:

Magit – A Git Porcelain inside Emacs - https://news.ycombinator.com/item?id=24431216 - Sept 2020 (83 comments)

A walk through the Magit interface - https://news.ycombinator.com/item?id=21729597 - Dec 2019 (81 comments)

Magit 2.13 released - https://news.ycombinator.com/item?id=17220630 - June 2018 (49 comments)

Show HN: Magit, the magical Git interface - https://news.ycombinator.com/item?id=15358723 - Sept 2017 (5 comments)

Magit Kickstarter fully funded - https://news.ycombinator.com/item?id=15312288 - Sept 2017 (71 comments)

Emacs and Magit - https://news.ycombinator.com/item?id=14819256 - July 2017 (174 comments)

Magit: a Git porcelain inside emacs - https://news.ycombinator.com/item?id=10643977 - Nov 2015 (14 comments)

What's new in Magit 2.x - https://news.ycombinator.com/item?id=9936095 - July 2015 (23 comments)

Using Emacs and Git with Magit 2.1 - https://news.ycombinator.com/item?id=9873237 - July 2015 (29 comments)

Meet Magit - Git Mode for Emacs - https://news.ycombinator.com/item?id=2543265 - May 2011 (4 comments)

Re: Magit 3.0

#123

Post by jidiculous »

aryamaan wrote: post #69

I never fail to see so many high things about Magit and I believe they are rightly placed. I am a non Emac user and wonder why no one tries to bring something similar client for non Emac population.

there's an effort to replicate Magit in VS Code: https://marketplace.visualstudio.com/items?itemName=kahole.m...

Re: Magit 3.0

#125

Post by amake »

natrys wrote: post #108
the-smug-one wrote: post #102

Earlier quoted context omitted.

It's probably too late for you :), but I use native-comp Emacs and it's blazing fast. At least lsp-mode+gopls is. I haven't gotten the chance to use this setup with Typescript's LS, but I've used gopls+clangd without slow downs. Also, dang I miss coding in TS instead of Go.

I have the same experience. Being on master and nativecomp probably helps, and I also think eglot is tad smoother than lsp-mode. I am using 10 year old laptop, no friction at all with tsserver, elixir-ls or gopls. However I think the server itself is the biggest difference maker, because all hell breaks lose when I try dart/flutter even in a simple project. But to be fair, my neovim setup struggles with that too (nvi…

I use Dart/Flutter LSP (lsp-dart) in Emacs all the time very successfully. “All hell breaks loose” sounds like maybe you have a project cloned or created but not yet `flutter pub get`’d, in which case I have seen it act like practically every token is an error.

Re: Magit 3.0

#126

Post by ajross »

Tried it once. It seemed cute. That lasted precisely up to the point where I did a "git rebase -i" and landed in some crazy custom mode where I was supposed to use... I dunno, secret keybindings or something instead of just reordering the patches like I've done for years and years.

It went right in the trash and I haven't looked back. Do. Not. Break. Workflows. Ever. If a user knows they're supposed to open a file in a specific format in their editor, and you inject your pixie dust into the guts of their editor thinking you know better what they want to do than they do, then all you've done is prove to the user that you cannot be trusted to support their work.

Seriously: I don't use magit and don't want to learn its new tooling because I don't trust it not to break itself.

Re: Magit 3.0

#127

Post by AlbinoDrought »

aryamaan wrote: post #69

I never fail to see so many high things about Magit and I believe they are rightly placed. I am a non Emac user and wonder why no one tries to bring something similar client for non Emac population.

There's `tig`, but I'm not sure how well it compares to `magit`. It's still much faster than typing the commands manually.

Here's a previous discussion: https://news.ycombinator.com/item?id=10090348

Re: Magit 3.0

#128

Post by avtar »

willtim wrote: post #114
zelphirkalt wrote: post #33

Looking forward to using the new release. "git at the speed of thought" describes magit quite well. Actually sometimes magit nudged me into the direction of looking up some more special git commands, to learn about them and then making use of magit's interface to them, often simply pressing one button more. Magit has not made me forget how command line git works, because it often shows me right there, what the argume…

> "git at the speed of thought" describes magit quite well. Not under Windows, unfortunately.

Is that magit or emacs on Windows in general?

Re: Magit 3.0

#129

Post by g8oz »

divs1210 wrote: post #79

Magit must be one of the best software tools ever written. I sorely missed it when I had to use IntelliJ for a few projects, so I wrote a TUI tool like Magit that can be used inside the console window of most IDEs like IntelliJ etc. Still haven't gotten around to releasing it properly, but it's easy to setup and works well enough to be a working MVP. https://github.com/hugit-project/hugit

Brilliant!

Re: Magit 3.0

#130

Post by pcr910303 »

dpbriggs wrote: post #77
darthrupert wrote: post #38

Earlier quoted context omitted.

Perhaps this is about MacOS vs Linux. I'm also suffering from slow emacs on mac, and no combination of flags seems to help. Use case: open a yaml file with syntax highlightning, scroll it with the mouse. The latency should be as low as possible.

Emacs on MacOS was also laggy for me before I started using the native comp feature. It takes a while to compile but it's seriously one of the best QoL changes I've had in years.

Does Emacs native-comp work outside X (i.e. works with the Cocoa port)? I thought it only worked through XQuartz b.c. gcc can’t parse the Objective-C required for the Cocoa port…

If the Cocoa port runs well, well that changes everything to me

Post reply on HN