mirror of
https://github.com/jbranchaud/til
synced 2026-01-19 15:08:02 +00:00
Compare commits
24 Commits
d980514bff
...
master
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
f48adc0f05 | ||
|
|
9773f10b84 | ||
|
|
bd58be8fda | ||
|
|
35f1f0b807 | ||
|
|
81afd44913 | ||
|
|
6fdadfa1fb | ||
|
|
aaf7da413c | ||
|
|
6a2a0a8ac1 | ||
|
|
712fc66aae | ||
|
|
e95477607e | ||
|
|
087766a792 | ||
|
|
4801e730f9 | ||
|
|
bd021f7eab | ||
|
|
8d8cfd56ce | ||
|
|
f4faa06258 | ||
|
|
8ccbd82320 | ||
|
|
5ea4165893 | ||
|
|
73476a8d16 | ||
|
|
c5ce81f918 | ||
|
|
32be787998 | ||
|
|
1d835d3553 | ||
|
|
0d4959046d | ||
|
|
b1198d2488 | ||
|
|
6c3805e7cd |
21
README.md
21
README.md
@@ -6,11 +6,11 @@ A collection of concise write-ups on small things I learn day to day across a
|
||||
variety of languages and technologies. These are things that don't really
|
||||
warrant a full blog post. These are things I've picked up by [Learning In
|
||||
Public™](https://dev.to/jbranchaud/how-i-built-a-learning-machine-45k9) and
|
||||
pairing with smart people at Hashrocket.
|
||||
working across different projects via [VisualMode](https://www.visualmode.dev/).
|
||||
|
||||
For a steady stream of TILs, [sign up for my newsletter](https://visualmode.kit.com/newsletter).
|
||||
|
||||
_1715 TILs and counting..._
|
||||
_1730 TILs and counting..._
|
||||
|
||||
See some of the other learning resources I work on:
|
||||
|
||||
@@ -131,6 +131,7 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
- [Clean Up Your Brew Installations](brew/clean-up-your-brew-installations.md)
|
||||
- [Configure Brew Environment Variables](brew/configure-brew-environment-variables.md)
|
||||
- [Export List Of Everything Installed By Brew](brew/export-list-of-everything-installed-by-brew.md)
|
||||
- [Install From Nonstandard Brewfile](brew/install-from-nonstandard-brewfile.md)
|
||||
- [Install Go Packages In Brewfile](brew/install-go-packages-in-brewfile.md)
|
||||
- [List All Services Managed By Brew](brew/list-all-services-managed-by-brew.md)
|
||||
|
||||
@@ -157,8 +158,10 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
|
||||
### Claude Code
|
||||
|
||||
- [Allow Edits From The Start](claude-code/allow-edits-from-the-start.md)
|
||||
- [Monitor Usage Limits From CLI](claude-code/monitor-usage-limits-from-cli.md)
|
||||
- [Open Current Prompt In Default Editor](claude-code/open-current-prompt-in-default-editor.md)
|
||||
- [Resume Specific Session](claude-code/resume-specific-session.md)
|
||||
|
||||
### Clojure
|
||||
|
||||
@@ -348,6 +351,7 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
- [Count Number Of Commits On A Branch](git/count-number-of-commits-on-a-branch.md)
|
||||
- [Create A New Branch With Git Switch](git/create-a-new-branch-with-git-switch.md)
|
||||
- [Delete All Untracked Files](git/delete-all-untracked-files.md)
|
||||
- [Determine Absolute Path Of Top-Level Project Directory](git/determine-absolute-path-of-top-level-project-directory.md)
|
||||
- [Determine The Hash Id For A Blob](git/determine-the-hash-id-for-a-blob.md)
|
||||
- [Diffing With Patience](git/diffing-with-patience.md)
|
||||
- [Dropping Commits With Git Rebase](git/dropping-commits-with-git-rebase.md)
|
||||
@@ -423,6 +427,7 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
- [Show What Is In A Stash](git/show-what-is-in-a-stash.md)
|
||||
- [Single Key Presses in Interactive Mode](git/single-key-presses-in-interactive-mode.md)
|
||||
- [Skip A Bad Commit When Bisecting](git/skip-a-bad-commit-when-bisecting.md)
|
||||
- [Skip Git Hooks As Needed](git/skip-git-hooks-as-needed.md)
|
||||
- [Skip Pre-Commit Hooks](git/skip-pre-commit-hooks.md)
|
||||
- [Staging Changes Within Vim](git/staging-changes-within-vim.md)
|
||||
- [Staging Stashes Interactively](git/staging-stashes-interactively.md)
|
||||
@@ -500,6 +505,7 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
|
||||
### Heroku
|
||||
|
||||
- [Check Ruby Version For Production App](heroku/check-ruby-version-for-production-app.md)
|
||||
- [Connect To A Database By Color](heroku/connect-to-a-database-by-color.md)
|
||||
- [Deploy A Review App To A Different Stack](heroku/deploy-a-review-app-to-a-different-stack.md)
|
||||
- [Diagnose Problems In A Heroku Postgres Database](heroku/diagnose-problems-in-a-heroku-postgres-database.md)
|
||||
@@ -545,6 +551,7 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
- [Focus The URL Bar](internet/focus-the-url-bar.md)
|
||||
- [Get Random Images From Unsplash](internet/get-random-images-from-unsplash.md)
|
||||
- [Grab The RSS Feed For A Substack Blog](internet/grab-the-rss-feed-for-a-substack-blog.md)
|
||||
- [Hide Overflowing Text For Google Sheets Column](internet/hide-overflowing-text-for-google-sheets-column.md)
|
||||
- [Search Tweets By Author](internet/search-tweets-by-author.md)
|
||||
- [Show All Pivotal Stories With Blockers](internet/show-all-pivotal-stories-with-blockers.md)
|
||||
- [Verify Site Ownership With DNS Record](internet/verify-site-ownership-with-dns-record.md)
|
||||
@@ -746,6 +753,7 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
|
||||
- [Create Umbrella Task For All Test Tasks](mise/create-umbrella-task-for-all-test-tasks.md)
|
||||
- [List The Files Being Loaded By Mise](mise/list-the-files-being-loaded-by-mise.md)
|
||||
- [Look In Ruby Version Dotfile](mise/look-in-ruby-version-dotfile.md)
|
||||
- [Preserve Color Output For Task Command](mise/preserve-color-output-for-task-command.md)
|
||||
- [Read Existing Dot Env File Into Env Vars](mise/read-existing-dot-env-file-into-env-vars.md)
|
||||
- [Run A Command With Specific Tool Version](mise/run-a-command-with-specific-tool-version.md)
|
||||
@@ -1065,6 +1073,7 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
- [Check If ActiveRecord Update Fails](rails/check-if-activerecord-update-fails.md)
|
||||
- [Check If Any Records Have A Null Value](rails/check-if-any-records-have-a-null-value.md)
|
||||
- [Check Specific Attributes On ActiveRecord Array](rails/check-specific-attributes-on-activerecord-array.md)
|
||||
- [Clean Up Memory Hungry Rails Console Processes](rails/clean-up-memory-hungry-rails-console-processes.md)
|
||||
- [Code Statistics For An Application](rails/code-statistics-for-an-application.md)
|
||||
- [Columns With Default Values Are Nil On Create](rails/columns-with-default-values-are-nil-on-create.md)
|
||||
- [Comparing DateTimes Down To Second Precision](rails/comparing-datetimes-down-to-second-precision.md)
|
||||
@@ -1558,6 +1567,7 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
- [Kill Other Connections To A Session](tmux/kill-other-connections-to-a-session.md)
|
||||
- [Kill The Current Session](tmux/kill-the-current-session.md)
|
||||
- [List All Key Bindings](tmux/list-all-key-bindings.md)
|
||||
- [List Processes Running Across All Session](tmux/list-processes-running-across-all-sessions.md)
|
||||
- [List Sessions](tmux/list-sessions.md)
|
||||
- [Open New Splits To The Current Directory](tmux/open-new-splits-to-the-current-directory.md)
|
||||
- [Open New Window With A Specific Directory](tmux/open-new-window-with-a-specific-directory.md)
|
||||
@@ -1604,6 +1614,7 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
### Unix
|
||||
|
||||
- [All The Environment Variables](unix/all-the-environment-variables.md)
|
||||
- [Apply Successive Filters To Lines In Less](unix/apply-successive-filters-to-lines-in-less.md)
|
||||
- [Authorize A cURL Request](unix/authorize-a-curl-request.md)
|
||||
- [Cat A File With Line Numbers](unix/cat-a-file-with-line-numbers.md)
|
||||
- [Cat Files With Color Using Bat](unix/cat-files-with-color-using-bat.md)
|
||||
@@ -1634,9 +1645,11 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
- [Curling For Headers](unix/curling-for-headers.md)
|
||||
- [Curling With Basic Auth Credentials](unix/curling-with-basic-auth-credentials.md)
|
||||
- [Determine ipv4 And ipv6 Public IP Addresses](unix/determine-ipv4-and-ipv6-public-ip-addresses.md)
|
||||
- [Diff Two Files In Unified Format](unix/diff-two-files-in-unified-format.md)
|
||||
- [Different Ways To Generate A v4 UUID](unix/different-ways-to-generate-a-v4-uuid.md)
|
||||
- [Display All The Terminal Colors](unix/display-all-the-terminal-colors.md)
|
||||
- [Display Free Disk Space](unix/display-free-disk-space.md)
|
||||
- [Display Line Numbers While Using Less](unix/display-line-numbers-while-using-less.md)
|
||||
- [Display The Contents Of A Directory As A Tree](unix/display-the-contents-of-a-directory-as-a-tree.md)
|
||||
- [Do A Dry Run Of An rsync](unix/do-a-dry-run-of-an-rsync.md)
|
||||
- [Do Not Overwrite Existing Files](unix/do-not-overwrite-existing-files.md)
|
||||
@@ -1745,6 +1758,7 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
- [Show A File Preview When Searching With FZF](unix/show-a-file-preview-when-searching-with-fzf.md)
|
||||
- [Show Disk Usage For The Current Directory](unix/show-disk-usage-for-the-current-directory.md)
|
||||
- [Show The Size Of Everything In A Directory](unix/show-the-size-of-everything-in-a-directory.md)
|
||||
- [Show Tree View Of Processes And Subprocesses](unix/show-tree-view-of-processes-and-subprocesses.md)
|
||||
- [Skip Paging If Output Fits On Screen With Less](unix/skip-paging-if-output-fits-on-screen-with-less.md)
|
||||
- [SSH Escape Sequences](unix/ssh-escape-sequences.md)
|
||||
- [SSH With Port Forwarding](unix/ssh-with-port-forwarding.md)
|
||||
@@ -1760,6 +1774,7 @@ If you've learned something here, support my efforts writing daily TILs by
|
||||
- [Unrestrict Where ripgrep Searches](unix/unrestrict-where-ripgrep-searches.md)
|
||||
- [Update Package Versions Known By asdf Plugin](unix/update-package-versions-known-by-asdf-plugin.md)
|
||||
- [Use fzf To Change Directories](unix/use-fzf-to-change-directories.md)
|
||||
- [Use Negative Lookbehind Matching With ripgrep](unix/use-negative-lookbehind-matching-with-ripgrep.md)
|
||||
- [Use Regex Pattern Matching With Grep](unix/use-regex-pattern-matching-with-grep.md)
|
||||
- [View A Web Page In The Terminal](unix/view-a-web-page-in-the-terminal.md)
|
||||
- [View The Source For A Brew Formula](unix/view-the-source-for-a-brew-formula.md)
|
||||
@@ -2056,7 +2071,7 @@ I shamelessly stole this idea from
|
||||
|
||||
## License
|
||||
|
||||
© 2015-2025 Josh Branchaud
|
||||
© 2015-2026 Josh Branchaud
|
||||
|
||||
This repository is licensed under the MIT license. See `LICENSE` for
|
||||
details.
|
||||
|
||||
@@ -32,8 +32,7 @@ tasks:
|
||||
notes:sync:
|
||||
desc: Sync latest changes from the notes submodule
|
||||
cmds:
|
||||
- git submodule update --remote {{.NOTES_DIR}}
|
||||
- cd {{.NOTES_DIR}} && git checkout main
|
||||
- cd {{.NOTES_DIR}} && git checkout main && git pull
|
||||
silent: false
|
||||
|
||||
notes:open:
|
||||
@@ -49,6 +48,7 @@ tasks:
|
||||
cmds:
|
||||
- git add NOTES.md
|
||||
- git commit -m "Update notes - $(date '+%Y-%m-%d %H:%M')"
|
||||
- git pull --rebase
|
||||
- git push
|
||||
status:
|
||||
- git diff --exit-code NOTES.md
|
||||
@@ -60,11 +60,6 @@ tasks:
|
||||
cmds:
|
||||
- git status
|
||||
|
||||
notes:pull:
|
||||
desc: Pull latest changes (alias for sync)
|
||||
cmds:
|
||||
- task notes:sync
|
||||
|
||||
notes:diff:
|
||||
desc: Show uncommitted changes in notes
|
||||
dir: '{{.NOTES_DIR}}'
|
||||
|
||||
25
brew/install-from-nonstandard-brewfile.md
Normal file
25
brew/install-from-nonstandard-brewfile.md
Normal file
@@ -0,0 +1,25 @@
|
||||
# Install From Nonstandard Brewfile
|
||||
|
||||
When you want to install the packages listed in the `Brewfile` for your current
|
||||
project (or dotfiles), you can run:
|
||||
|
||||
```bash
|
||||
$ brew bundle
|
||||
```
|
||||
|
||||
And `brew` knows to look for and use the `Brewfile` in the current directory.
|
||||
|
||||
If, however, you are trying to run `brew bundle` for a `Brewfile` located
|
||||
somewhere besides the current directory *OR* you want to target a file with a
|
||||
non-standard name (like
|
||||
[`Brewfile.personal`](https://github.com/jbranchaud/dotfiles/blob/main/Brewfile.personal)),
|
||||
then you can use the `--file` flag.
|
||||
|
||||
```bash
|
||||
$ brew bundle --file Brewfile.personal
|
||||
```
|
||||
|
||||
This is what I do [here in my `dotfiles`
|
||||
repo](https://github.com/jbranchaud/dotfiles/blob/b053f6251cae7ed52f698fc2a2c40ba82c5881b0/installer/mac-setup.sh#L42-L48).
|
||||
|
||||
See `man brew` and find the section on `brew bundle` for more details.
|
||||
35
claude-code/allow-edits-from-the-start.md
Normal file
35
claude-code/allow-edits-from-the-start.md
Normal file
@@ -0,0 +1,35 @@
|
||||
# Allow Edits From The Start
|
||||
|
||||
A common pattern for me when using Claude Code is that I start it up in a
|
||||
project, I prompt it with a question or feature spec, it either comes up with a
|
||||
plan or just starts working, and as soon as it is ready to make its first edits
|
||||
to a file, it prompts me something like:
|
||||
|
||||
```
|
||||
Do you want to make this edit to Taskfile.yml?
|
||||
❯ 1. Yes
|
||||
2. Yes, allow all edits during this session (shift+tab)
|
||||
3. Type here to tell Claude what to do differently
|
||||
```
|
||||
|
||||
That's a nice default so that I don't get surprised by Claude Code editing a
|
||||
bunch of files.
|
||||
|
||||
However, if I'm in a git-backed project and I'm going into a session intending
|
||||
to make edits, then I can skip the formalities. I can tell Claude Code when
|
||||
starting up the session that edits are allowed.
|
||||
|
||||
```sh
|
||||
$ claude --permission-mode acceptEdits
|
||||
```
|
||||
|
||||
When I do this, I'll see the following indicator below the prompt input field:
|
||||
|
||||
```
|
||||
⏵⏵ accept edits on (shift+tab to cycle)
|
||||
```
|
||||
|
||||
If I've already started `claude` but I forgot to specify that permission mode, I
|
||||
can also toggle right into _accept edits_ by hitting `Shift+Tab`.
|
||||
|
||||
[source](https://www.youtube.com/watch?v=_IK18goX4X8)
|
||||
22
claude-code/resume-specific-session.md
Normal file
22
claude-code/resume-specific-session.md
Normal file
@@ -0,0 +1,22 @@
|
||||
# Resume Specific Session
|
||||
|
||||
There are a few different ways to resume a [Claude
|
||||
Code](https://code.claude.com/docs/en/overview) session.
|
||||
|
||||
First, if I have exited a session for the current project and I want to pick
|
||||
back up with that most recent one, then I can use `claude --continue`.
|
||||
|
||||
If I have had a few recent sessions for the current project and I want to
|
||||
remember what they were and pick up where I left off with one of them, then I
|
||||
can use `claude --resume` (with no argument). That will open a picker where I
|
||||
can browser through a summary of the recent sessions based on their starting
|
||||
prompt. The one I pick is the session that will be resumed.
|
||||
|
||||
Finally, if I have grabbed a specific session ID (UUID) during the session from
|
||||
the `/status` output, then I can reference that value directly.
|
||||
|
||||
```sh
|
||||
$ claude --resume 92170532-be31-4a91-b2a9-025b8fa78232
|
||||
```
|
||||
|
||||
See `claude --help` for more details.
|
||||
@@ -0,0 +1,39 @@
|
||||
# Determine Absolute Path Of Top-Level Project Directory
|
||||
|
||||
The `git rev-parse` command is a git plumbing command for parsing different
|
||||
kinds of things in git into a canonical form that can be used in a deterministic
|
||||
way by scripts. I would typically think of using it to work with branch names,
|
||||
tags, and other kinds of refs.
|
||||
|
||||
There is a handy, sorta off-label use for it in determining the absolute path of
|
||||
the root directory for the current git repository. Use the `--show-toplevel`
|
||||
flag with no other arguments.
|
||||
|
||||
```bash
|
||||
❯ git rev-parse --show-toplevel
|
||||
/Users/lastword/dev/jbranchaud/til
|
||||
```
|
||||
|
||||
Here, I am in the local copy of [my TIL repo](https://github.com/jbranchaud/til). This command gives me the absolute
|
||||
path of the top-level directory where that `.git` directory resides.
|
||||
|
||||
This is useful for scripts that need to orient themselves to the current
|
||||
project's top-level directory regardless of what directory they are being
|
||||
executed from. This is useful for things like a git hook script or monorepos
|
||||
with scripts located in a specific sub-project directory.
|
||||
|
||||
Also worth mentioning is the `--show-superproject-working-tree` flag. In my TIL
|
||||
repo, I have a private repository included as a submodule. Within that directory
|
||||
`--show-toplevel` will produce the absolute path to the submodule. If I instead
|
||||
want the absolute path of the _super project_ (in this case TIL), then I can use
|
||||
this other flag.
|
||||
|
||||
```bash
|
||||
❯ git rev-parse --show-toplevel
|
||||
/Users/lastword/dev/jbranchaud/til/notes
|
||||
|
||||
❯ git rev-parse --show-superproject-working-tree
|
||||
/Users/lastword/dev/jbranchaud/til
|
||||
```
|
||||
|
||||
See `man git-rev-parse` for more details.
|
||||
33
git/skip-git-hooks-as-needed.md
Normal file
33
git/skip-git-hooks-as-needed.md
Normal file
@@ -0,0 +1,33 @@
|
||||
# Skip Git Hooks As Needed
|
||||
|
||||
Projects have Git hooks configured for all sorts of reasons. Most common are
|
||||
`pre-commit` hooks which verify certain aspects of the contents of a commit.
|
||||
A `pre-commit` hook could check that the tests all pass, that the changes don't
|
||||
include any debugging statements, and so forth. There are all kinds of hooks
|
||||
though, like `pre-rebase` and `post-checkout`.
|
||||
|
||||
These hooks can sometimes get in the way and we may need to skip or disable them
|
||||
on a one-off basis.
|
||||
|
||||
Several Git commands offer a `--no-verify` flag which can skip running the hook
|
||||
associated with that command.
|
||||
|
||||
- `git commit --no-verify` (skips `pre-commit` and `commit-msg` hooks)
|
||||
- `git push --no-verify` (skips `pre-push` hook)
|
||||
- `git merge --no-verify` (skips `pre-merge-commit` hook)
|
||||
- `git am --no-verify` (skips `applypatch-msg` and `pre-applypatch` hooks)
|
||||
|
||||
If you look in the `.git/hooks` directory, there are several other hooks not
|
||||
covered by the above. So, what if I am doing an action like `git checkout` and I
|
||||
want to skip the `post-checkout` hook?
|
||||
|
||||
I can override the `hooksPath` config for that one command with the `-c` flag.
|
||||
|
||||
```sh
|
||||
$ git -c core.hooksPath=/dev/null checkout ...
|
||||
```
|
||||
|
||||
By setting it to `/dev/null`, it will find *no* hooks available, so none will be
|
||||
executed for this command.
|
||||
|
||||
See `man git-config` for more details on `core.hooksPath`.
|
||||
32
heroku/check-ruby-version-for-production-app.md
Normal file
32
heroku/check-ruby-version-for-production-app.md
Normal file
@@ -0,0 +1,32 @@
|
||||
# Check Ruby Version For Production App
|
||||
|
||||
While deploying a fresh Rails app to Heroku recently, I ran into an issue. The
|
||||
`it` block argument wasn't working despite being on Ruby 4.0. Or so I thought.
|
||||
|
||||
Running the following command reported the Ruby version of that Heroku server
|
||||
instance:
|
||||
|
||||
```bash
|
||||
❯ heroku run -- ruby --version
|
||||
Running ruby --version on ⬢ my-app... up, run.3090
|
||||
ruby 3.3.9 (2025-07-24 revision f5c772fc7c) [x86_64-linux]
|
||||
```
|
||||
|
||||
I was on `3.3.9` which must have been the fallback default at the time.
|
||||
|
||||
Though I had set the Ruby version in my `.ruby-version` file, I had neglected to
|
||||
specify it in the `Gemfile` as well. Once I added it to the `Gemfile` and
|
||||
redeployed, my Heroku server instance was running the expected version of Ruby.
|
||||
|
||||
```bash
|
||||
❯ heroku run -- ruby --version
|
||||
Running ruby --version on ⬢ my-app... up, run.5353
|
||||
ruby 4.0.0 (2025-12-25 revision 553f1675f3) +PRISM [x86_64-linux]
|
||||
```
|
||||
|
||||
Note: because [I have set `HEROKU_ORGANIZATION` and
|
||||
`HEROKU_APP`](set-default-team-and-app-for-project.md) in my environment
|
||||
(`.envrc`) for the local copy of the app, I don't need to specify those when
|
||||
running the `heroku run` command above.
|
||||
|
||||
See `heroku run --help` for more details.
|
||||
12
internet/hide-overflowing-text-for-google-sheets-column.md
Normal file
12
internet/hide-overflowing-text-for-google-sheets-column.md
Normal file
@@ -0,0 +1,12 @@
|
||||
# Hide Overflowing Text For Google Sheets Column
|
||||
|
||||
I imported a big CSV into a new Google Sheets document. This included a
|
||||
"Description" column with many of the descriptions varying between 50 and 80
|
||||
characters. The bottom line is that the description column was flowing over the
|
||||
top of the columns next to it. Instead of expanding the width of that column as
|
||||
far as the largest description, I wanted to hide the _overflow_.
|
||||
|
||||
The way to do this in Google Sheets is to highlight the entire column by
|
||||
clicking on the column grouping. Then under the _Format_ menu item is a
|
||||
_Wrapping_ submenu. The _Clip_ option is what I was looking for because it clips
|
||||
the text that gets shown at the edge of the column.
|
||||
27
mise/look-in-ruby-version-dotfile.md
Normal file
27
mise/look-in-ruby-version-dotfile.md
Normal file
@@ -0,0 +1,27 @@
|
||||
# Look In Ruby Version Dotfile
|
||||
|
||||
Newer versions of [`mise`](https://mise.jdx.dev/dev-tools/) specifically only
|
||||
look for tool versions in `mise.toml` as well as the asdf `.tool-versions` file.
|
||||
A lot of Ruby projects use the `.ruby-version` file to indicate the Ruby version
|
||||
of a project. To continue to use the `.ruby-version` file instead of migrating
|
||||
to `mise.toml`, you need to tell `mise` that you prefer to use the idiomatic
|
||||
version file.
|
||||
|
||||
I added the following line to my
|
||||
[`~/.config/mise/config.toml`](https://github.com/jbranchaud/dotfiles/commit/8edeb7a9c53500e89e88b4079cbd1859ebebcbda)
|
||||
file:
|
||||
|
||||
```toml
|
||||
idiomatic_version_file_enable_tools = ["ruby"]
|
||||
```
|
||||
|
||||
Now, whenever `mise` is looking for the specified Ruby version of a project, it
|
||||
will also look for `.ruby-version`.
|
||||
|
||||
Here is a [full list of idomatic version files supported by
|
||||
`mise`](https://mise.jdx.dev/configuration.html#idiomatic-version-files).
|
||||
|
||||
See
|
||||
[`idiomatic_version_file_enable_tools`](https://mise.jdx.dev/configuration/settings.html#idiomatic_version_file_enable_tools)
|
||||
as well as the [Ruby-specific documentation](https://mise.jdx.dev/lang/ruby.html#ruby-version-and-gemfile-support)
|
||||
for more details.
|
||||
43
rails/clean-up-memory-hungry-rails-console-processes.md
Normal file
43
rails/clean-up-memory-hungry-rails-console-processes.md
Normal file
@@ -0,0 +1,43 @@
|
||||
# Clean Up Memory Hungry Rails Console Processes
|
||||
|
||||
I noticed (using `htop`) that a remote server hosting a Rails app had most of
|
||||
its RAM being actively consumed. This was hindering my ability to run a fresh
|
||||
deploy because the deploy processes had to do a ton of memory swapping which
|
||||
drastically slowed the whole thing down.
|
||||
|
||||
With some investigation, I discovered that most of the memory was being consumed
|
||||
by a handful of `rails console` processes. I didn't have any known active `rails console` processes that I was using. That combined with the dates of these
|
||||
processes starting way in the past suggested to me that these were abandoned
|
||||
processes that hadn't been properly cleaned up.
|
||||
|
||||
```bash
|
||||
server:~# ps aux | grep rails
|
||||
32767 878915 0.0 0.0 1227160 936 pts/0 Ssl+ 2025 0:03 /exec rails console
|
||||
32767 878942 0.9 6.5 830996 261748 pts/0 Rl+ 2025 249:51 ruby /app/bin/rails console
|
||||
32767 3004097 0.0 0.0 1227160 692 pts/0 Ssl+ 2025 0:04 /exec rails console
|
||||
32767 3004129 0.9 6.4 834672 257228 pts/0 Dl+ 2025 406:31 ruby /app/bin/rails console
|
||||
32767 3048582 0.0 0.0 1227160 940 pts/0 Ssl+ Jan09 0:00 /exec rails console
|
||||
32767 3048611 1.1 6.3 829936 253484 pts/0 Dl+ Jan09 60:50 ruby /app/bin/rails console
|
||||
32767 3060033 0.0 0.0 1227160 944 pts/0 Ssl+ 2025 0:04 /exec rails console
|
||||
32767 3060063 0.9 6.5 838084 260812 pts/0 Rl+ 2025 405:37 ruby /app/bin/rails console
|
||||
root 3699372 0.0 0.0 7008 1300 pts/0 S+ 15:51 0:00 grep --color=auto rails
|
||||
server:~# ps aux | grep 'rails console' | awk '{sum+=$6} END {print sum/1024 " MB"}'
|
||||
1014.64 MB
|
||||
```
|
||||
|
||||
As we can see by tacking on this `awk` command, these processes are consuming
|
||||
1GB of memory.
|
||||
|
||||
Each of these is a pair of processes. A parent process (`/exec rails console`)
|
||||
that kicks off and supervises the memory-hungry child process (`ruby /app/bin/rails console`).
|
||||
|
||||
To free up this memory, I targeted each of the parent processes with a `kill`
|
||||
command one by one. For example:
|
||||
|
||||
```bash
|
||||
server:~# kill 878915
|
||||
```
|
||||
|
||||
I suspect that I may have left the occasional terminal tab open with one of
|
||||
these `rails console` processes running and the SSH connection was getting
|
||||
killed without the `rails console` getting killed with it.
|
||||
46
tmux/list-processes-running-across-all-sessions.md
Normal file
46
tmux/list-processes-running-across-all-sessions.md
Normal file
@@ -0,0 +1,46 @@
|
||||
# List Processes Running Across All Session
|
||||
|
||||
I wanted an overview of all the processes running across all the tmux sessions
|
||||
that I have running on my machine right now. The `list-panes` command (with the
|
||||
`-a` flag) gives me a listing of all the panes across all session of the current
|
||||
tmux server.
|
||||
|
||||
That output on its own isn't giving me quite the info I'm looking for though.
|
||||
With the `-f` (_format_) flag, I can use variables available in that context
|
||||
like `session_name`, `pane_pid`, and `pane_current_command`.
|
||||
|
||||
I can assemble the details I want into a command like this:
|
||||
|
||||
```bash
|
||||
❯ tmux list-panes -a -F "#{session_name}:#{window_index}.#{pane_index} #{pane_pid} #{pane_current_command}"
|
||||
PLP:1.1 62364 zsh
|
||||
TIL:1.1 62345 nvim
|
||||
TIL:1.2 65838 task
|
||||
TIL:2.1 11428 tmux
|
||||
client-app:1.1 62373 ssh
|
||||
client-app:1.2 10796 zsh
|
||||
client-app:1.3 63081 zsh
|
||||
client-app:2.1 61115 overmind
|
||||
client-app:3.1 82608 zsh
|
||||
visualmode-dev:1.1 52237 zsh
|
||||
```
|
||||
|
||||
This gives me the details I want, but I can take it a step further by piping it
|
||||
to the `column` command to improve the formatting a little:
|
||||
|
||||
```bash
|
||||
❯ tmux list-panes -a -F "#{session_name}:#{window_index}.#{pane_index} #{pane_pid} #{pane_current_command}" \
|
||||
| column -t
|
||||
PLP:1.1 62364 zsh
|
||||
TIL:1.1 62345 nvim
|
||||
TIL:1.2 65838 task
|
||||
TIL:2.1 11428 tmux
|
||||
client-app:1.1 62373 ssh
|
||||
client-app:1.2 10796 zsh
|
||||
client-app:1.3 63081 zsh
|
||||
client-app:2.1 61115 overmind
|
||||
client-app:3.1 82608 zsh
|
||||
visualmode-dev:1.1 52237 zsh
|
||||
```
|
||||
|
||||
See `man tmux` and, in particular, the `FORMATS` section for more details.
|
||||
33
unix/apply-successive-filters-to-lines-in-less.md
Normal file
33
unix/apply-successive-filters-to-lines-in-less.md
Normal file
@@ -0,0 +1,33 @@
|
||||
# Apply Successive Filters To Lines In Less
|
||||
|
||||
Let's say I've opened a large Rails log file with `less`:
|
||||
|
||||
```bash
|
||||
$ less logs/development.log
|
||||
```
|
||||
|
||||
I have an idea of what I'm looking for, but there is way more noise than signal.
|
||||
I can start to filter out some of the noise. The `&` command starts a filter
|
||||
prompt. If I start to filter by something like `INSERT INTO`, then a ton of
|
||||
lines disappear leaving just those matching that pattern.
|
||||
|
||||
Scrolling through the current set of lines, I start to have a better idea of
|
||||
what I'm looking for, but there is still too much noise. I can apply an
|
||||
additional successive filter on the remaining lines by hitting `&` again and
|
||||
entering in another pattern -- e.g. `GoodJob`.
|
||||
|
||||
Now I only see lines that contain both `INSERT INTO` and `GoodJob` somewhere in
|
||||
them.
|
||||
|
||||
As `less` puts it:
|
||||
|
||||
> Multiple & commands may be entered, in which case only lines which match all
|
||||
> of the patterns will be displayed.
|
||||
|
||||
If I want to undo all the filtering, I just need to enter an empty `&` filter
|
||||
prompt and it will reset things back to displaying all lines.
|
||||
|
||||
> If pattern is empty (if you type & immediately followed by ENTER), any
|
||||
> filtering is turned off, and all lines are displayed.
|
||||
|
||||
See `man less` for more details.
|
||||
78
unix/diff-two-files-in-unified-format.md
Normal file
78
unix/diff-two-files-in-unified-format.md
Normal file
@@ -0,0 +1,78 @@
|
||||
# Diff Two Files In Unified Format
|
||||
|
||||
The `diff` command is a standalone utility that can be used to get the
|
||||
difference between two files. It is similar to what you might expect when
|
||||
running `git diff` which compares two different versions of the same file. The
|
||||
`diff` command predates `git` and its unified format is what became the standard
|
||||
that `git` uses for its own diff implementation.
|
||||
|
||||
Running `diff` with two files as is gives output like the following:
|
||||
|
||||
```bash
|
||||
❯ diff startup.sh startup2.sh
|
||||
10,13c10,14
|
||||
< declare -A SESSIONS=(
|
||||
< ["TIL"]="$HOME/dev/jbranchaud/til:setup_til"
|
||||
< ["PLP"]="$HOME/dev/jbranchaud/pool-league-pro:"
|
||||
< ["client-app"]="$HOME/dev/client/client-app:"
|
||||
---
|
||||
> # Sessions will be created in the order listed here
|
||||
> SESSIONS=(
|
||||
> "TIL:$HOME/dev/jbranchaud/til:setup_til"
|
||||
> "PLP:$HOME/dev/jbranchaud/pool-league-pro:"
|
||||
> "client-app:$HOME/dev/client/client-app:"
|
||||
73,74c74,75
|
||||
< for session_name in TIL PLP client-app; do
|
||||
< IFS=':' read -r directory setup_function <<<"${SESSIONS[$session_name]}"
|
||||
---
|
||||
> for session_config in "${SESSIONS[@]}"; do
|
||||
> IFS=':' read -r session_name directory setup_function <<<"$session_config"
|
||||
```
|
||||
|
||||
That's readable at a glance, but the unified format (with the `-u` flag) can
|
||||
provide more context:
|
||||
|
||||
```bash
|
||||
❯ diff -u startup.sh startup2.sh
|
||||
--- startup.sh 2026-01-10 12:46:52
|
||||
+++ startup2.sh 2026-01-10 12:48:00
|
||||
@@ -7,10 +7,11 @@
|
||||
|
||||
# Session configurations
|
||||
# Format: "session_name:directory:setup_function"
|
||||
-declare -A SESSIONS=(
|
||||
- ["TIL"]="$HOME/dev/jbranchaud/til:setup_til"
|
||||
- ["PLP"]="$HOME/dev/jbranchaud/pool-league-pro:"
|
||||
- ["client-app"]="$HOME/dev/client/client-app:"
|
||||
+# Sessions will be created in the order listed here
|
||||
+SESSIONS=(
|
||||
+ "TIL:$HOME/dev/jbranchaud/til:setup_til"
|
||||
+ "PLP:$HOME/dev/jbranchaud/pool-league-pro:"
|
||||
+ "client-app:$HOME/dev/client/client-app:"
|
||||
)
|
||||
|
||||
# Setup function for TIL session
|
||||
@@ -70,8 +71,8 @@
|
||||
echo ""
|
||||
|
||||
# Create sessions in order
|
||||
- for session_name in TIL PLP client-app; do
|
||||
- IFS=':' read -r directory setup_function <<<"${SESSIONS[$session_name]}"
|
||||
+ for session_config in "${SESSIONS[@]}"; do
|
||||
+ IFS=':' read -r session_name directory setup_function <<<"$session_config"
|
||||
create_session "$session_name" "$directory" "$setup_function"
|
||||
done
|
||||
```
|
||||
|
||||
Here we get additional context like surrounding lines and file name details.
|
||||
|
||||
While this is useful on its own, it also has the added benefit of making the
|
||||
output compatible with other tools we may already be using. For instance, I'm
|
||||
already using [delta](https://github.com/dandavison/delta) as my [git pager](https://github.com/jbranchaud/dotfiles/blob/main/gitconfig#L51) and [git differ](https://github.com/jbranchaud/dotfiles/blob/main/gitconfig#L139).
|
||||
|
||||
With the unified format, I can pipe the output directly to `delta` to get a
|
||||
better view of the diff that is colorized and includes syntax highlighting.
|
||||
|
||||
```bash
|
||||
❯ diff -u startup.sh startup2.sh | delta
|
||||
```
|
||||
27
unix/display-line-numbers-while-using-less.md
Normal file
27
unix/display-line-numbers-while-using-less.md
Normal file
@@ -0,0 +1,27 @@
|
||||
# Display Line Numbers While Using Less
|
||||
|
||||
Including line numbers while viewing files with `less` can provide useful
|
||||
context for understanding where you are within the file. This is especially true
|
||||
if you've used `&` to filter down to lines that match a pattern.
|
||||
|
||||
You can start `less` with line numbers with the `-N` flag (or `--LINE-NUMBERS`
|
||||
if you really want to spell it out).
|
||||
|
||||
```bash
|
||||
$ less -N log/development.log
|
||||
```
|
||||
|
||||
If you've already started up `less` and wish you had included line numbers,
|
||||
there is no reason to restart it with the flag. Instead, toggle the line numbers
|
||||
option on within the `less` process. To do this, type `-N`. It will prompt you
|
||||
with `Constantly display line numbers (press RETURN)`. Hit enter and line
|
||||
numbers will appear to the left of each line in the file.
|
||||
|
||||
Similarly, to toggle line numbers back off within `less`, hit `-n` (lower-case
|
||||
`n`), accept the prompt, and back off they go.
|
||||
|
||||
Both of these (`-N`/`-n`) are options being set (toggled) via the `-` command.
|
||||
There are many other options like these that can be configured within a `less`
|
||||
session in the same way.
|
||||
|
||||
See `man less` and find the `-` command and the available `OPTIONS`.
|
||||
36
unix/show-tree-view-of-processes-and-subprocesses.md
Normal file
36
unix/show-tree-view-of-processes-and-subprocesses.md
Normal file
@@ -0,0 +1,36 @@
|
||||
# Show Tree View Of Processes And Subprocesses
|
||||
|
||||
Though you can cobble together a command on a MacOS Unix system to output a
|
||||
hierarchical tree view of a parent process and its descendent subprocesses, it
|
||||
is easier to [`brew install pstree`](https://github.com/FredHucht/pstree) and
|
||||
use that.
|
||||
|
||||
Here is what I see when I run it for a _pid_ that corresponds to a `tmux`
|
||||
session that I have running locally:
|
||||
|
||||
```bash
|
||||
❯ pstree 61690
|
||||
-+= 61690 lastword tmux new-session -d -s TIL -c /Users/lastword/dev/jbranchaud/til
|
||||
|--= 63081 lastword /bin/zsh
|
||||
|-+= 11428 lastword zsh
|
||||
| \-+= 48511 lastword pstree 61690
|
||||
| \--- 48512 root ps -axwwo user,pid,ppid,pgid,command
|
||||
|-+= 62345 lastword zsh
|
||||
| \--= 06031 lastword claude
|
||||
|--= 62364 lastword /bin/zsh
|
||||
|-+= 62373 lastword zsh
|
||||
| \--= 64407 lastword ssh my-app-staging
|
||||
|-+= 61115 lastword /bin/zsh
|
||||
| \-+= 61579 lastword overmind start -f Procfile.dev
|
||||
| \--- 61586 lastword tmux -C -L overmind-my-app-abc123 new -n web -s my-app -P -F %overmind-process #{pane_id} web #{pane_pid} /var/folders/zc/abc123/T/overmin
|
||||
|--= 52237 lastword /bin/zsh
|
||||
|--= 82608 lastword /bin/zsh
|
||||
\--= 10796 lastword /bin/zsh
|
||||
```
|
||||
|
||||
I was looking for a frozen `claude` process that was part of this session. And I
|
||||
found it about halfway down that list -- `06031`. Now I can run `kill` on that
|
||||
process as needed.
|
||||
|
||||
For some additional context, I initially found the _pid_ for the `tmux` session
|
||||
by running `ps aux | grep tmux` and looking through those results.
|
||||
42
unix/use-negative-lookbehind-matching-with-ripgrep.md
Normal file
42
unix/use-negative-lookbehind-matching-with-ripgrep.md
Normal file
@@ -0,0 +1,42 @@
|
||||
# Use Negative Lookbehind Matching With ripgrep
|
||||
|
||||
The most straightforward way to use `ripgrep` is to hand it a pattern. It will
|
||||
take that pattern and move forward through each file trying to find matches.
|
||||
|
||||
```bash
|
||||
$ rg 'TwilioClient\.new'
|
||||
```
|
||||
|
||||
That will find all occurrences of `TwilioClient.new` in available project files.
|
||||
|
||||
What if that pattern is too permissive though? That is going to match on
|
||||
occurrences of `TwilioClient.new` as well as things like
|
||||
`LoggingTwilioClient.new`. If we want to exclude the latter, there are a few
|
||||
ways to do that. One of them being the use of [the _negative lookbehind_ regex
|
||||
feature](https://www.pcre.org/current/doc/html/pcre2syntax.html#SEC23) that is
|
||||
available with PCRE2 (Perl-Compatible Regular Expressions).
|
||||
|
||||
A _negative lookbehind_ is like a standard pattern. We look forward through the
|
||||
document for the base pattern (like `TwilioClient\.new`). However, once we find
|
||||
that match, we then look back at the previous characters and if they match our
|
||||
negative lookbehind pattern, then it is no longer a positive match.
|
||||
|
||||
We can use one of the following to forms to achieve this:
|
||||
|
||||
```
|
||||
(?<!...) )
|
||||
(*nlb:...) ) negative lookbehind
|
||||
(*negative_lookbehind:...) )
|
||||
```
|
||||
|
||||
For instance, here is what this looks like for our example:
|
||||
|
||||
```bash
|
||||
$ rg -P '(?<!Logging)TwilioClient\.new'
|
||||
```
|
||||
|
||||
Note: we have to use the `-P` flag to tell `ripgrep` that we are using PCRE2
|
||||
syntax. Otherwise, it assumes a simpler regex syntax that doesn't support
|
||||
_negative lookbehind_.
|
||||
|
||||
See `man rg` for more details.
|
||||
Reference in New Issue
Block a user