Archive

just has been by go to tool for saving commands in a way that I can replay them and have team members replay them without relying on the shell history of any given machine. This is my go to default step, it lets you pick a just command to run with a fuzzy picker.

default:
  @just --list

TIL how to display the list of nfs mounts on your network.

showmount -e

You can even look for mounts of other machines on your network.

showmount -e <hostname>

To allow access only to the , you can pass add the Resource field to the User Policy when you create a new token.

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "admin:*"
      ]
    },
    {
      "Effect": "Allow",
      "Action": [
        "kms:*"
      ]
    },
    {
      "Effect": "Allow",
      "Action": [
        "s3:*"
      ],
      "Resource": [
        "arn:aws:s3:::<bucket>",
        "arn:aws:s3:::<bucket>/*"
      ]
    }
  ]
}

You can inspect sqlite tables with the sqlite shell.

note that you get into the shell with sqlite3 database.db

.tables

I also learned that .tables, .index and .schema are helper functions that query the sqlite_master table on the main database.

Here is an output from my redka database. The sqlite_master table contains all the sqlite objects type, name, tbl_name, rootpage, and sql to create them.

❯ sqlite3 database.db "SELECT * from sqlite_master;"
table|rkey|rkey|2|CREATE TABLE rkey (
    id       integer primary key,
    key      text not null,
    type     integer not null,
        version  integer not null,
    etime    integer,
        mtime    integer not null
)
index|rkey_key_idx|rkey|3|CREATE UNIQUE INDEX rkey_key_idx on rkey (key)
index|rkey_etime_idx|rkey|4|CREATE INDEX rkey_etime_idx on rkey (etime)
where etime is not null
trigger|rkey_on_type_update|rkey|0|CREATE TRIGGER rkey_on_type_update
before update of type on rkey
for each row
when old.type is not new.type
begin
    select raise(abort, 'key type mismatch');
end
table|rstring|rstring|5|CREATE TABLE rstring (
    key_id integer not null,
    value  blob not null,

    foreign key (key_id) references rkey (id)
          on delete cascade
)
index|rstring_pk_idx|rstring|6|CREATE UNIQUE INDEX rstring_pk_idx on rstring (key_id)
view|vstring|vstring|0|CREATE VIEW vstring as
  select
    rkey.id as key_id, rkey.key, rstring.value,
        datetime(etime/1000, 'unixepoch') as etime,
        datetime(mtime/1000, 'unixepoch') as mtime
  from rkey join rstring on rkey.id = rstring.key_id
  where rkey.type = 1
    and (rkey.etime is null or rkey.etime > unixepoch('subsec'))
table|rhash|rhash|7|CREATE TABLE rhash (
    key_id integer not null,
    field text not null,
    value blob not null,

    foreign key (key_id) references rkey (id)
      on delete cascade
)
index|rhash_pk_idx|rhash|8|CREATE UNIQUE INDEX rhash_pk_idx on rhash (key_id, field)
index|rhash_key_id_idx|rhash|9|CREATE INDEX rhash_key_id_idx on rhash (key_id)
view|vhash|vhash|0|CREATE VIEW vhash as
  select
    rkey.id as key_id, rkey.key, rhash.field, rhash.value,
        datetime(etime/1000, 'unixepoch') as etime,
        datetime(mtime/1000, 'unixepoch') as mtime
  from rkey join rhash on rkey.id = rhash.key_id
  where rkey.type = 4
    and (rkey.etime is null or rkey.etime > unixepoch('subsec'))

With the liscense changes to redis there are several new forks out there. One that I am particularly interested in is redka.

curl https://i.jpillora.com/nalgeon/redka | bash
chmod +x redka
./redka database.db

We now have redis running on port 6379 that we can connect to with a redis client. And we have a sqlite database that we can inspect.

❯ sqlite3 database.db "SELECT name FROM sqlite_master;"
rkey
rkey_key_idx
rkey_etime_idx
rkey_on_type_update
rstring
rstring_pk_idx
vstring
rhash
rhash_pk_idx
rhash_key_id_idx
vhash

We can look at the values in the vstring table.

sqlite3 database.db "SELECT * from vstring;"
1|hi|hello there you||2024-04-17 01:46:26

The main system that I am concerned about is my arch BTW machine. I found a great article from the official archlinux site covering it.

For my machine I am concerned with this line.

The xz packages prior to version 5.6.1-2 (specifically 5.6.0-1 and 5.6.1-1) contain this backdoor.

I checked my xz package with AUR.">paru, and I am good.

paru -Qii zx

AUR.">paru has some nice features that I rarely use, and hav to look up when I need them. Here are two commands to help with dependency management.

❯ paru -Qii nodejs
Name            : nodejs
Version         : 21.7.2-1
Description     : Evented I/O for V8 javascript
Architecture    : x86_64
URL             : https://nodejs.org/
Licenses        : MIT
Groups          : None
Provides        : None
Depends On      : icu  libuv  libnghttp2  libnghttp3  libngtcp2  openssl  zlib  brotli  c-ares
Optional Deps   : npm: nodejs package manager [installed]
Required By     : node-gyp  nodejs-nopt  npm  semver
Optional For    : None
Conflicts With  : None
Replaces        : None
Installed Size  : 46.86 MiB
Packager        : Felix Yan <[email protected]>
Build Date      : Thu 04 Apr 2024 05:11:09 AM CDT
Install Date    : Mon 15 Apr 2024 07:27:02 AM CDT
Install Reason  : Installed as a dependency for another package
Install Script  : No
Validated By    : Signature
Backup Files    : None
Extended Data   : pkgtype=pkg

You can check all the packages depended on by nodejs by running the following. This is everything from all of the repos you have configured, not what you have installed.

❯ pactree --reverse --sync --depth 1 nodejs

nodejs
├─acorn
├─ansible-language-server
├─asar
├─babel-cli
├─babel-core
├─bash-language-server
├─blinksocks
├─bower
├─browserify
├─coffeescript
├─dot-language-server
├─emscripten
├─eslint
├─eslint-language-server
├─eslint_d
├─gitlab
├─gnomon
├─grunt-cli
├─gtop
├─gulp
├─hedgedoc
├─jake
├─markdownlint-cli2
├─marked
├─marked-man
├─matrix-appservice-irc
├─modclean
├─node-gyp
├─nodejs-emojione
├─nodejs-material-design-icons
├─nodejs-nopt
├─nodejs-source-map
├─nodejs-yaml
├─npm
├─openui5
├─pm2
├─prettier
├─pyright
├─rapydscript-ng
├─s3rver
├─semver
├─serverless
├─stylelint
├─stylus
├─svelte-language-server
├─tailwindcss-language-server
├─ts-node
├─typescript
├─typescript-svelte-plugin
├─uglify-js
├─vscode-css-languageserver
├─vscode-html-languageserver
├─vscode-json-languageserver
├─vue-language-server
├─vue-typescript-plugin
├─wasm-bindgen
├─web-ext
├─wrangler
├─yaml-language-server
├─yarn

I recently had to update my copier-gallery command to trust my own templates because some of them have shell scripts that run afterwards. Be warned that this could be a dangerous feature to run on random templates you get off the internet, but these are all mine, so if I wreck it its my own fault.

copier copy --trust <template> <destination>

All the the copier copy api can be found with help.

❯ copier copy --help
copier copy 8.3.0

Copy from a template source to a destination.

Usage:
    copier copy [SWITCHES] template_src destination_path

Hidden-switches:
    -h, --help                         Prints this help message and quits
    --help-all                         Prints help messages of all sub-commands and quits
    -v, --version                      Prints the program's version and quits

Switches:
    -C, --no-cleanup                   On error, do not delete destination if it was
                                       created by Copier.
    --UNSAFE, --trust                  Allow templates with unsafe features (Jinja
                                       extensions, migrations, tasks)
    -a, --answers-file VALUE:str       Update using this path (relative to
                                       `destination_path`) to find the answers file
    -d, --data VARIABLE=VALUE:str      Make VARIABLE available as VALUE when rendering the
                                       template; may be given multiple times
    -f, --force                        Same as `--defaults --overwrite`.
    -g, --prereleases                  Use prereleases to compare template VCS tags.
    -l, --defaults                     Use default answers to questions, which might be
                                       null if not specified.
    -n, --pretend                      Run but do not make any changes
    -q, --quiet                        Suppress status output
    -r, --vcs-ref VALUE:str            Git reference to checkout in `template_src`. If you
                                       do not specify it, it will try to checkout the
                                       latest git tag, as sorted using the PEP 440
                                       algorithm. If you want to checkout always the
                                       latest version, use `--vcs-ref=HEAD`.
    -s, --skip VALUE:str               Skip specified files if they exist already; may be
                                       given multiple times
    -w, --overwrite                    Overwrite files that already exist, without asking.
    -x, --exclude VALUE:str            A name or shell-style pattern matching files or
                                       folders that must not be copied; may be given
                                       multiple times

Today I accidentally ran f2 in ipython to discover that it opens your $EDITOR! I use this feature quite often in zsh, it is bound to <c-e> for me, and since I have my environment variable EDITOR set to nvim it opens nvim when I hit <c-e>. Today I discovered that Ipython has this bound to F2. If you know how to set it to <c-e> let me know I’ve tried, a lot.

export EDITOR=nvim
ipython
<F2>

better yet add export EDITOR=nvim to your .zshrc

# ~/.zshrc
export EDITOR=nvim

I’ve really been enjoying using sqlmodel for my projects that need a database. One thing that I definitely lacked on for too long was indexing my database. I hit a point with one database where it was taking 7s for pretty simple paginated queries to return 10 records.

For every field that you will be querying on, you can create an index, by setting it equal to Field(index=True)

class Hero(SQLModel, table=True):
    id: int | None = Field(default=None, primary_key=True)
    name: str = Field(index=True)
    secret_name: str
    age: int | None = Field(default=None, index=True)

example courtesy of the docs

 primary keys are indexed by default.

The docs cover this pretty well, and in quite depth - Optimizing Queries

inline-snapshot is a new tool that I am trying out for python testing. It takes snapshots of your outputs and places them inline with the test.

Here is the most basic starter.

import inline_snapshot

def test_one():
    assert 1 == snapshot()

Now when I run pytest my tests will fail because my assert has no value, but if I run pytest --inline-snapshot=create it will fill out my snapshot values and the file will then look like this.

import inline_snapshot

def test_one():
    assert 1 == snapshot(1)

It also works with pydantic models.

class MyModel(BaseModel):
    name: str
    age: int
    nickname: str | None = None


def test_my_model_instance():
    assert MyModel(name="Waylon", age=1) == snapshot(MyModel(name="Waylon", age=1))


def test_my_model_fields():
    me = MyModel(name="Waylon", age=1, nickname='Waylon')
    assert me.name == snapshot("Waylon")
    assert me.age == snapshot(1)
    assert me.nickname == snapshot("Waylon")