| ;--------------------------------------------------------------------- | |
| ; Read (using bsread) a pair of ASCII chars representing a printable | |
| ; byte value, and convert it to that byte value, returning it in A. | |
| asc_0 .equ 30 | |
| asc_9 .equ 39 | |
| asc_A .equ 41 | |
| asc_F .equ 46 | |
| .area ZP |
This article gives a simple purely functional sorted, unbalanced tree structure in 67 lines of java. The direct Haskell translation below is 13 lines; Java's verbosity ratio is 5.15.
data (Ordered a) => SortedSet a = Empty | Node (SortedSet a) a (SortedSet a)
| #!/usr/bin/env bash | |
| set -o pipefail | |
| . "$(git --exec-path)/git-sh-setup" | |
| # | |
| # Also see "$(git --exec-path)/git-filter-branch" for, e.g., a list of | |
| # functions available in the commit filter. | |
| # set_reflog_action |
| #!/usr/bin/env bash | |
| # | |
| # describe-pr - print a Pull Request description | |
| # | |
| type xsel >/dev/null 2>&1 || { | |
| echo 1>&2 "$(basename $0): xsel command not found; please install it." | |
| exit 1 | |
| } | |
| # (This doesn't currently work on a Mac, where we'd use the `pbpaste` |
| # XXX This worked in 2009 with Ruby 1.8; I've not tested on anything | |
| # more recent. But you get the general idea. | |
| class Class | |
| def initialize_vars(*vars, &initialize_block) | |
| vars.each { |arg| attr_reader arg.to_sym } | |
| varsyms = vars.collect { |arg| ('@' + arg.to_s).to_sym } | |
| define_method(:initialize) { |*init_args| | |
| raise(ArgumentError, "wrong number of arguments " + |
| #!/usr/bin/env bash | |
| # If no commits specified, we default to HEAD@{u}..HEAD. However if HEAD is uptodate, | |
| # then `rev-parse` returns nothing. In that case we fallback to HEAD^..HEAD. | |
| # XXX: This breaks if `some-other-brach@{u}..some-other-brach` is specified. | |
| # adding a `--boundary` should have solved this issue, but there seems to be | |
| # a bug in `git rev-parse` and `git rev-list`. | |
| parsed_revs=($(git rev-parse --revs-only "$@" --default HEAD@{u}..HEAD)) | |
| [ ${#parsed_revs[@]} = 0 ] && parsed_revs=(HEAD ^HEAD^) |
| set_basedir() { # This is portable; don't touch it! | |
| local dir path0="$0" | |
| while [ -h "$path0" ]; do | |
| dir="$(cd "$(dirname "$path0")" && pwd -P)" | |
| path0="$(readlink "$path0")"; | |
| [ "$(echo "$path0" | cut -b 1)" = / ] || path0="$dir/$path0" | |
| done | |
| basedir="$(cd "$(dirname "$path0")/.." && pwd -P)" | |
| }; set_basedir |
| #!/bin/bash | |
| # | |
| # mdpreview - preview a markdown file | |
| # | |
| # Idea from: | |
| # https://ubuntuincident.wordpress.com/2011/05/05/readme-markdown-on-github/ | |
| # | |
| # This does not handle the Github variations documented at: | |
| # https://help.github.com/articles/github-flavored-markdown | |
| # You may want to try using a gist to get a more accurate preview. |
Hello there! This is a sample post for gist.io, a super-lightweight writing soapbox for hackers.
Now look up. Further. Above the post title. See that grey text with the gist ID?
Now back to me. That grey text is a link! Open that sucker in a new tab to see the source for this post. Also, I'm on a horse.
If you peek at it with a web inspector, you'll see that it is a second-level heading. You can use first level headings, but they'll look just like the second level ones, and the gods of the HTML5 outlining algorithm will frown upon you.