YOU MUST:
- Parse what exactly is being asked
- Identify if this is a feature (use feature.md) or fix (use fix.md)
- If neither, continue with this workflow
- Break down into specific, measurable subtasks
THIS IS A MANDATORY WORKFLOW - NO STEPS CAN BE SKIPPED
YOU MUST USE ALL AVAILABLE RESOURCES:
get_usage_rules
MCP tool or CLAUDE.md linkspackage_docs_search
for ALL potentially relevant packages#!/bin/bash | |
# This is a simple script to copy non tmp files from zed config to a repo you can use for storing your config. | |
# This is slightly modified from what I actually use, which stores a number of other config files that I share across machines. | |
# rsync is used instead of cp for flexibility | |
ZED_SOURCE_DIR="/Users/USER/.config/zed" | |
ZED_DEST_DIR="/Users/USER/PATH_TO_SYNC_REPO/.config/zed" | |
# absolute path to repo storing your config |
defmodule Example.Poller do | |
use GenServer | |
@backoff [ | |
50, | |
100, | |
150, | |
250, | |
400, | |
650 |
(defface tree-sitter-hl-face:warning | |
'((default :inherit font-lock-warning-face)) | |
"Face for parser errors" | |
:group 'tree-sitter-hl-faces) | |
(defun hook/tree-sitter-common () | |
(unless font-lock-defaults | |
(setq font-lock-defaults '(nil))) | |
(setq tree-sitter-hl-use-font-lock-keywords nil) | |
(tree-sitter-mode +1) |
# This is the proper BEAM test for https://github.com/uber/denial-by-dns. The Erlang test in that repo is sequential | |
# (https://github.com/uber/denial-by-dns/blob/b809cc561a691d9d6201d06d38d06c33c9c9f9ec/erlang-httpc/main.erl) | |
# which is not consistent with the test description (https://github.com/uber/denial-by-dns/tree/b809cc561a691d9d6201d06d38d06c33c9c9f9ec#how-it-works) | |
# and also differs from e.g. go test in the same repo which issues requests concurrently. | |
# | |
# Consequently, the conclusion in that repo as well as the original article (https://eng.uber.com/denial-by-dns/) is incorrect. | |
# A properly written Erlang test would pass, as demonstrated by this Elixir script. | |
# | |
# This code performs a slightly refined and a correct version of that tests in Elixir: | |
# |
# Lets you call a block for each item in a list, just like `each`. | |
# But instead of running serially, it runs in a limited number of parallel threads. | |
# This is useful when you don't just want one thread per item, e.g. to avoid rate limiting or network saturation. | |
class EachInThreadPool | |
def self.call(inputs, pool_size:, &block) | |
queue = Queue.new | |
inputs.each { queue << _1 } | |
pool_size.times.map { |
@spec function_name(arg_1_type, arg_2_type) :: return_type
@type new_type_name :: existing_type
@typedoc """ description """
above type definition.@type statuses :: [atom]
@type number_with_remark :: {number, String.t}