Phil Hadviger datfinesoul

There is extra code in here, but the key takeaway is:

Use jq to convert an object with key/value pairs, into a key=value string list
Feed that listinto a while loop, which then exports the environment variables

While this is done in a script, it needs to live in a function, if the environment variables are supposed to persist in the current shell.

`elaborate-example.js`

Shows the use of generators to control concurrency. This example illustrates the following.

3 workers running through the data
Each piece of "work" takes 200-2000 milliseconds to run
The 5th, 10th and 17th item will throw an exception
Workers will exit because the exception is unhandled

worker='1' itemnum='0' item='F'

A teaching example to show what happens when you don't have an await on an async method invocation.

Without the away, the program will print:

1612325635779 slowMessage start
1612325635781 leaving program
1612325637784 slowMessage end

Tag an `<untagged>` ECR image in China

export AWS_PROFILE="<profile>"
aws ecr get-login-password --region cn-north-1 | \
  docker login --username AWS --password-stdin <account_id>.dkr.ecr.cn-north-1.amazonaws.com.cn
MANIFEST=$( \
  aws ecr batch-get-image --repository-name <ecr_repo_name> \
  --image-ids imageDigest=sha256:<sha> \
  --query 'images[].imageManifest' --output text \
 )

MSSQL naming conventions

The best take on this I've seen so far.

Based on a post by Dan Guzman, SQL Server MVP, http://weblogs.sqlteam.com/dang/.

https://social.msdn.microsoft.com/Forums/sqlserver/en-US/76db4119-d040-446e-9ea7-6ba90e017338/naming-conventions-for-constraints?forum=transactsql

Picture a database with a Schema called "In" and another schema called "Out", representing input data and output data respectively. These schema have table names that're the same - for example: each schema has it's own IncomeExpense

BITs vs DATETIMEs for flags

My background has been more in the MSSQL (Microsoft SQL Server) side of relational databases in the past, so I'll use DATETIME and BIT as the data types of the discussion, but it should work the same regardless of DBMS.

Take for example this table:

CREATE TABLE blog.article
(
 article_id INT NOT NULL,

SPLUNK & Logging

Tips

SET TIMEZONE IN PROFILE

creating a basic saved search

sourcetype=syslog (host=east OR host=europe OR host=asia OR host=jobs) marklar

	SHELL=/bin/bash

	scan:
	@driftctl scan \
	--from "tfstate+tfcloud://$(shell <".tfcloud/.wsid" cat)" \
	--tfc-token "$(shell <"${HOME}/.terraform.d/credentials.tfrc.json" jq -r '.credentials."app.terraform.io".token')"

	ignore:
	@driftctl scan \
	--from "tfstate+tfcloud://$(shell <".tfcloud/.wsid" cat)" \

	[*.bash]
	indent_style = tab
	indent_size = 2

	[*.{txt,json}]
	indent_style = space
	indent_size = 2

	const partition = "aws-cn";

	const multilineRegex = (...rest) => new RegExp(rest.map(x => (x instanceof RegExp) ? x.source : x).join(''));

	const re = multilineRegex(
	/aws:/, // I can COMMENT!!!!
	partition,
	/:secretsmanager:cn-north-1:111111111111:/
	);