4 ms·

I use vim because it’s the only real option for my workflow. I work on an extremely large codebase that easily eats up several hundred GB of disk space. This c

I use vim because it’s the only real option for my workflow. I work on an extremely large codebase that easily eats up several hundred GB of disk space. This code lives on a workstation under my desk. I do my coding on a laptop. I want only one feature: Low latency for typing, find/replace, search over a wireless connection from my laptop to my workstation, possibly while I’m traveling. Nothing beats ssh’ing into my workstation and launching vim. (Emacs will also work - I just prefer vim because I’m used to it) I would love to use an IDE and the powerful features that come with it. But I wouldn’t trade off latency for those features. Additionally, there are a ton of people who use the same workflow as me - ssh + vim. Any impact from changes to a network security policy will quickly be fixed because it impacts soo many of us.

dividuum 6y ago
Genuinely curious: How do you efficiently navigate such an extremely large code base using vim? ctags? Any plugin recommendation?
- dilippkumar 6y ago
  I use ripgrep[0] to look for things. It’s incredibly fast - I have found that it’s usually faster than what an IDE takes on a much smaller project. Once ripgrep has found something, jumping to that file is super easy (and fast!) I use ctags on smaller subdirectories when I’m not familiar with the codebase and I need to jump around a lot to find definitions within a small number of files. [0] https://manpages.debian.org/testing/ripgrep/rg.1.en.html https://manpages.debian.org/testing/ripgrep/rg.1.en.html
TeMPOraL 6y ago
Similarly here. The codebases I work on are smaller, but then, I sometimes work on a go from a sidearm 2-in-1 computer. It's lovely to be able to just SSH (well, mosh) into my beefy desktop and run Emacs in a terminal, and have it both look and work identically to the GUI version I use at home.
x86_64Ubuntu 6y ago
What is the language of your codebase? And which industry are you in so I can avoid it. My goodness, several hundred GB of code sounds like pure misery.
commandlinefan 6y ago
> I use vim ... I work on an extremely large codebase That's surprising to me - it's usually once I get up into the triple-digit file count that I give up on command-line tools and start letting an IDE index things for me.

dividuum 6y ago

Genuinely curious: How do you efficiently navigate such an extremely large code base using vim? ctags? Any plugin recommendation?

dilippkumar 6y ago
I use ripgrep[0] to look for things. It’s incredibly fast - I have found that it’s usually faster than what an IDE takes on a much smaller project. Once ripgrep has found something, jumping to that file is super easy (and fast!) I use ctags on smaller subdirectories when I’m not familiar with the codebase and I need to jump around a lot to find definitions within a small number of files. [0] https://manpages.debian.org/testing/ripgrep/rg.1.en.html https://manpages.debian.org/testing/ripgrep/rg.1.en.html

TeMPOraL 6y ago

Similarly here. The codebases I work on are smaller, but then, I sometimes work on a go from a sidearm 2-in-1 computer. It's lovely to be able to just SSH (well, mosh) into my beefy desktop and run Emacs in a terminal, and have it both look and work identically to the GUI version I use at home.

x86_64Ubuntu 6y ago

What is the language of your codebase? And which industry are you in so I can avoid it. My goodness, several hundred GB of code sounds like pure misery.

commandlinefan 6y ago

> I use vim ... I work on an extremely large codebase That's surprising to me - it's usually once I get up into the triple-digit file count that I give up on command-line tools and start letting an IDE index things for me.