FEED Validator

for Atom and RSS and KML

Sorry

This feed does not validate.

line 1, column 38: Missing feed element: title [help]

<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/At ...
                                      ^

In addition, interoperability with the widest range of feed readers could be improved by implementing the following recommendations.

line 1, column 239: Self reference doesn't match document location [help]

... el="self" type="application/atom+xml" /><link href="http://localhost:400 ...
                                             ^

line 54, column 83: name should not be blank (10 occurrences) [help]

... means.&lt;/p&gt;</content><author><name></name></author><category term=" ...
                                             ^

line 273, column 0: content should not contain onclick attribute (13 occurrences) [help]
```
  &lt;td class=&quot;line_numbers&quot; title=&quot;click to toggle&quot; on ...
```
line 301, column 0: content should not contain ondblclick attribute (13 occurrences) [help]
```
  &lt;td class=&quot;code&quot;&gt;&lt;pre ondblclick=&quot;with (this.style ...
```

Source: https://rhnh.net/feed.xml

<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="3.3.1">Jekyll</generator><link href="http://localhost:4000/feed.xml" rel="self" type="application/atom+xml" /><link href="http://localhost:4000/" rel="alternate" type="text/html" /><updated>2021-04-02T00:15:12+00:00</updated><id>http://localhost:4000/feed.xml</id><entry><title type="html">Using Size to Debug D3 Selections</title><link href="http://localhost:4000/2018/10/21/debugging-d3-selections/" rel="alternate" type="text/html" title="Using Size to Debug D3 Selections" /><published>2018-10-21T00:00:00+00:00</published><updated>2018-10-21T00:00:00+00:00</updated><id>http://localhost:4000/2018/10/21/debugging-d3-selections</id><content type="html" xml:base="http://localhost:4000/2018/10/21/debugging-d3-selections/">Yesterday I was learning about the relatively new <a href="https://bl.ocks.org/mbostock/3808218">General Update
Pattern</a> in D3, but I couldn’t get it
working. I knew I was missing something obvious, but how to figure out what?
Because I was dealing with nested attributes, I was assigning the selections
to variables and then merging them later, so ended up with this heavily
simplified (and broken) code to display a list of tiles:
<div class="language-javascript highlighter-coderay"><div class="CodeRay">
<div class="code"><pre> 1 let tiles = container.selectAll('.tile').data(data, d =&gt; d.id)
 2
 3 let tilesEnter = tiles.enter()
 4 .append('div')
5 .attr(&quot;class&quot;, &quot;tiles&quot;)
 6
 7 tiles.select('.tile').merge(tilesEnter)
 8 .attr('color', d =&gt; d.color)
 9
10 let contentEnter = tilesEnter
11 .append('span')
12 .attr('class', 'content')
13
14 tiles.select('.content').merge(contentEnter)
15 html(d =&gt; d.content)
</pre></div>
</div>
</div>
When I updated the data, the content of the tile in the child element updated,
but the color at the root level did not!
I tried a number of debugging approaches, but the one that I found easiest to
wrap my head around, and that eventually led me to a solution, was using the
<a href="https://github.com/d3/d3-selection#selection_size"><code>size()</code></a> to verify how
many elements where in each selection.
<div class="language-javascript highlighter-coderay"><div class="CodeRay">
<div class="code"><pre>1 console.log(&quot;tiles entering&quot;, tilesEnter.size())
2 console.log(&quot;tiles updating&quot;, tiles.select('.tile').size())
3 console.log(&quot;content entering&quot;, contentEnter.size())
4 console.log(&quot;content updating&quot;, tiles.select('.content').size())
</pre></div>
</div>
</div>
This allowed me to verify that for the second working case (for data with
four elements) that the enter/update selections went from 4 and 0
respectively to 0 and 4 when data was updated. For the first case, the update
selection was always zero, and this led me to notice the extra
<code>select('.tile')</code> shouldn’t be there for the root case, since we’re already
on that selection from the <code>selectAll</code> in the initial setup!
I found logging the entire selection to not be as useful, because it’s
confusing what its internal state actually means.</content><author><name></name></author><category term="code" /><category term="javascript" /><category term="d3" /><summary type="html">Yesterday I was learning about the relatively new General Update Pattern in D3, but I couldn’t get it working. I knew I was missing something obvious, but how to figure out what?</summary></entry><entry><title type="html">Adding Last-Modified response header to Haskell Servant API</title><link href="http://localhost:4000/2018/09/30/haskell-servant-last-modified/" rel="alternate" type="text/html" title="Adding Last-Modified response header to Haskell Servant API" /><published>2018-09-30T00:00:00+00:00</published><updated>2018-09-30T00:00:00+00:00</updated><id>http://localhost:4000/2018/09/30/haskell-servant-last-modified</id><content type="html" xml:base="http://localhost:4000/2018/09/30/haskell-servant-last-modified/">Given the following <a href="https://haskell-servant.github.io/">Servant API</a>
(boilerplate redacted for brevity):
<div class="language-haskell highlighter-coderay"><div class="CodeRay">
<div class="code"><pre>1 type MyAPI = &quot;some-api&quot; :&gt; Get '[JSON] NoContent
2
3 someApi = return NoContent
</pre></div>
</div>
</div>
How do you add a <code>Last-Modified</code> header? As a first attempt, we can use the
<code>Header</code> type with <a href="http://hackage.haskell.org/package/servant-0.14.1/docs/Servant-API-ResponseHeaders.html#v:addHeader"><code>addHeader</code></a> and a <code>UTCTime</code>:
<div class="language-haskell highlighter-coderay"><div class="CodeRay">
<div class="code"><pre> 1 import Data.Time.Clock (UTCTime, getCurrentTime)
 2
 3 type LastModifiedHeader = Header &quot;Last-Modified&quot; UTCTime
 4 type MyAPI = &quot;some-api&quot; :&gt; Get '[JSON] (Headers '[LastModifiedHeader] NoContent)
 5
 6 someApi = do
 7 now &lt;- getCurrentTime
 8 addHeader now
 9 return NoContent
</pre></div>
</div>
</div>
Unfortunately, this returns the time in the wrong format!
<pre><code>&gt; curl -I localhost/some-api | grep Last-Modified
Last-Modified: 2018-09-30T19:56:39Z
</code></pre>
It <a href="https://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html#sec3.3.1">should be RFC
1123</a>. We can
fix this with a <code>newtype</code> that wraps the
formatting functions available in <code>Data.Time.Format</code>:
<div class="language-haskell highlighter-coderay"><div class="CodeRay">
<div class="code"><pre> 1 {-# LANGUAGE GeneralizedNewtypeDeriving #-}
 2
 3 import Data.ByteString (pack)
 4 import Data.Time.Clock (UTCTime, getCurrentTime)
 5 import Data.Time.Format (formatTime, defaultTimeLocale, rfc1123DateFormat)
 6
 7 newtype RFC1123Time = RFC1123Time UTCTime
 8 deriving (Show, FormatTime)
 9
10 instance ToHttpApiData RFC1123Time where
11 toUrlPiece = error &quot;Not intended to be used in URLs&quot;
12 toHeader =
13 let rfc1123DateFormat = &quot;%a, %_d %b %Y %H:%M:%S GMT&quot; in
14 pack . formatTime defaultTimeLocale rfc1123DateFormat
15
16 type LastModifiedHeader = Header &quot;Last-Modified&quot; RFC1123Time
17 type MyAPI = &quot;some-api&quot; :&gt; Get '[JSON] (Headers '[LastModifiedHeader] NoContent)
18
19 someApi = do
20 now &lt;- getCurrentTime
21 addHeader $ RFC1123Time now
22 return NoContent
</pre></div>
</div>
</div>
<pre><code>&gt; curl -I localhost/some-api | grep Last-Modified
Last-Modified: Sun, 30 Sep 2018 20:44:16 GMT
</code></pre>
If anyone knows a simpler way, please let me know!
<h3 id="irreverant-technical-asides">Irreverant technical asides</h3>
Many implementations reference RFC822 for <code>Last-Modified</code> format. What gives?
RFC822 was updated by RFC1123, which only adds a few clauses to tighten up the
definition. Most importantly, it updates the year format from 2 digits to 4!
Note that
<a href="http://hackage.haskell.org/package/time-1.9.2/docs/Data-Time-Format.html#v:rfc822DateFormat"><code>Date.Time.Format.rfc882DateFormat</code></a>
is technically incorrect here, specifying a four digit year.
<a href="http://hackage.haskell.org/package/time-http-0.5/docs/Data-Time-Format-RFC822.html"><code>Data.Time.Format.RFC822</code></a>
gets it right.
<code>rfc822DateFormat</code> is also technically incorrect in another way: it uses the
<code>%Z</code> format specifier for timezone, which produces <code>UTC</code> on a <code>UTCTime</code>. This
is not an allowed value! However,
<a href="https://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html#sec3.3.1">RFC 2616</a> says
“for the purposes of HTTP, GMT is exactly equal to UTC” so GMT can safely be
hardcoded here since we know we always have a UTC time.</content><author><name></name></author><category term="code" /><category term="haskell" /><category term="servant" /><summary type="html">Given the following Servant API (boilerplate redacted for brevity):</summary></entry><entry><title type="html">Using Haskell Servant to Power a Decoupled React Single Page Application</title><link href="http://localhost:4000/2018/05/10/haskell-servant-react-auth-example/" rel="alternate" type="text/html" title="Using Haskell Servant to Power a Decoupled React Single Page Application" /><published>2018-05-10T00:00:00+00:00</published><updated>2018-05-10T00:00:00+00:00</updated><id>http://localhost:4000/2018/05/10/haskell-servant-react-auth-example</id><content type="html" xml:base="http://localhost:4000/2018/05/10/haskell-servant-react-auth-example/">Recently I’ve been experimenting with different ways of building web
applications. In particular, I’m interested to what extent it is feasible to
start an application with a “pure” API, as distinct from a typical Ruby on
Rails application. This approach would limit the backend server to only API
endpoints, and restrict it from any kind of HTML generation. All HTML concerns
would be pushed to a frontend using something like React.
I published an <a href="https://github.com/xaviershay/haskell-servant-react-auth-example">example
application</a>
that demostrates this architecture using
<a href="http://haskell-servant.readthedocs.io/en/stable/">Servant</a> and
<a href="https://reactjs.org/">React</a>. In this post, I’ll detail some of the issues I
came across getting this working.
<h2 id="authentication">Authentication</h2>
One difficultly I came across was how to handle third-party authentication (via
Google OAuth) in this scenario when running the backend and frontend as
completely separate services. A typical OAuth flow requires server side calls
and interactions that don’t work when the flow is split over two different
services.
Google provides an <a href="https://developers.google.com/identity/protocols/OAuth2UserAgent">OAuth flow for Web
Applications</a>
that addresses the first issue. The hard part is how to verify that
authentication in the backend.
This OAuth flow provides the client with a <a href="https://jwt.io/">JWT</a> containing
information about the user, such as their email and granted scopes. This can be
<a href="https://github.com/xaviershay/haskell-servant-react-auth-example/blob/master/api/src/Auth.hs">verified and trusted on the
server</a>
using Google’s public key, which needs to be <a href="https://github.com/xaviershay/haskell-servant-react-auth-example/blob/master/api/src/KeyFetcher.hs">continually fetched from their
endpoint</a>
to keep it current.
This verification can be done in Servant using a <a href="http://haskell-servant.readthedocs.io/en/stable/tutorial/Authentication.html#generalized-authentication-in-action">Generalized Authentication handler</a>.
<h2 id="cors">CORS</h2>
Requests between applications on different hosts have to negotiate CORS
correctly. This could be mitigated by running a reverse proxy in front of both
services and presenting them at a single domain, but I wanted to see if I could
make it work without this.
A few things are required for correct <a href="https://developer.mozilla.org/en-US/docs/Web/HTTP/CORS">CORS handling</a>. First, appropriate <code>Access-Control-Allow-Origin</code> headers need to be set on requests. This is best handled with a <a href="https://github.com/xaviershay/haskell-servant-react-auth-example/blob/master/api/app/Main.hs#L31">middleware from the <code>wai-cors</code> package</a>.
That would be sufficient for “simple” requests, but for since our API uses both
a non-simple content type (<code>application/json</code>) and the <code>Authorization</code> header,
they need to be added to the default policy:
<div class="language-haskell highlighter-coderay"><div class="CodeRay">
<div class="code"><pre>1 corsPolicy = simpleCorsResourcePolicy
2 { corsRequestHeaders = [ &quot;authorization&quot;, &quot;content-type&quot; ]
3 }
</pre></div>
</div>
</div>
Also, these non-simple API requests will trigger a <a href="https://developer.mozilla.org/en-US/docs/Web/HTTP/CORS#Preflighted_requests">CORS
preflight</a>,
which sends an <code>OPTIONS</code> request to our API. The API needs to be extended to
handle these requests.
<a href="https://github.com/sordina/servant-options"><code>servant-options</code></a> provides a
middleware to do this automatically from an API definition. Unfortunately, <code>servant-options</code> didn’t work <a href="https://github.com/sordina/servant-options/issues/2">out of the box</a> with <code>servant-auth</code>. I needed to provide an instance of <code>HasForeign</code> for <code>AuthProtect</code>. A simple pass-through implementation looks like this:
<div class="language-haskell highlighter-coderay"><div class="CodeRay">
<div class="code"><pre>1 instance (HasForeign lang ftype api) =&gt;
2 HasForeign lang ftype (AuthProtect k :&gt; api) where
3
4 type Foreign ftype (AuthProtect k :&gt; api) = Foreign ftype api
5
6 foreignFor lang Proxy Proxy subR =
7 foreignFor lang Proxy (Proxy :: Proxy api) subR
</pre></div>
</div>
</div>
I later <a href="https://github.com/xaviershay/haskell-servant-react-auth-example/blob/master/api/src/Api.hs#L29">extended
this</a>
to include appropriate metadata so that I could use it to generate clients
correctly.
<h2 id="js-clients">JS Clients</h2>
A nice thing about Servant is the ability to auto-generate client wrappers for
your API. <code>servant-js</code> provides a number of formats for this, though they
weren’t as ergonomic as I was hoping. It doesn’t currently have <a href="https://github.com/haskell-servant/servant-auth/issues/8">support for
<code>servant-auth</code></a> nor
support for ES6-style <code>exports</code>. Rather than solve this generically, I wrote a
<a href="https://github.com/xaviershay/haskell-servant-react-auth-example/blob/master/api/src/JsGeneration.hs">custom
generator</a>.
For fun, it outputs an API class that allows an authorization token to be
supplied in the constructor, rather than as an argument to every function:
<div class="language-javascript highlighter-coderay"><div class="CodeRay">
<div class="code"><pre>1 let api = new Api(jwt);
2 api.getEmail();
</pre></div>
</div>
</div>
I’m not sure what the best way to distribute this API is. Currently, the
example writes out a file in the frontend’s source tree. This works great for
development, but for production I would consider either a dedicated build step
in packaging, or serving the JS up directly from the API server.
Aside from this generated client, I didn’t do anything particularly interesting
on the React front. The app included in the example is very simple.
<h2 id="conclusion">Conclusion</h2>
This wasn’t a big enough project to draw any serious conclusions about the
approach. It is evident however that Servant still has a couple of rough edges
when you get outside of the common cases. It took a while to wrap my head around how Servant uses the type system. I found <a href="https://www.well-typed.com/blog/2015/11/implementing-a-minimal-version-of-haskell-servant/">this post and exercise</a> very helpful.
I hadn’t used JWTs before, and they strike me as a pretty neat way to thread authentication through a distributed application.</content><author><name></name></author><category term="code" /><category term="haskell" /><category term="servant" /><summary type="html">Recently I’ve been experimenting with different ways of building web applications. In particular, I’m interested to what extent it is feasible to start an application with a “pure” API, as distinct from a typical Ruby on Rails application. This approach would limit the backend server to only API endpoints, and restrict it from any kind of HTML generation. All HTML concerns would be pushed to a frontend using something like React.</summary></entry><entry><title type="html">2017 Contributions</title><link href="http://localhost:4000/2017/12/31/2017-contributions/" rel="alternate" type="text/html" title="2017 Contributions" /><published>2017-12-31T00:00:00+00:00</published><updated>2017-12-31T00:00:00+00:00</updated><id>http://localhost:4000/2017/12/31/2017-contributions</id><content type="html" xml:base="http://localhost:4000/2017/12/31/2017-contributions/">This was a light year for open source contribution, 168 by Github’s count (and
some uncounted on Bitbucket). I’ve continued on the RSpec core team, though the
project is mostly stable and I haven’t been particularly active. I did find
time to experiment with some new things however:
<ul>
<li>Windows development! A <a href="https://bitbucket.org/Nicksaurus/foreman/pull-requests/17">rather major PR</a> to Foreman to switch out the core solver and fix some UI lockups (among others) was the first time I’ve written C# in a decade. It, and the relevant UI libraries, are quite nice. Unfortunately the maintainer hasn’t had time for the project so it’s basically stalled. A fork is probably due. I also made a trivial patch to the <a href="https://bitbucket.org/Nicksaurus/foreman/pull-requests/17">Ori Livesplitter</a>.</li>
<li>Small forays into frontend development with a <a href="https://github.com/xaviershay/kssu-timer">KSSU timer</a> and a <a href="https://github.com/xaviershay/runnersonly">glorious troll site</a> both in React, and some <a href="https://github.com/yupferris/assistant/pull/3">assistant</a> <a href="https://github.com/yupferris/assistant/pull/4">contributions</a> with VueJS.</li>
<li>Taking advantage of my new Pixel phone to try my hand at Android development, with <a href="https://github.com/aricneto/TwistyTimer/pull/113">contributions to Twisty Timer</a> and a new <a href="https://github.com/yupferris/assistant/pull/4">system keyboard for input of Rubix cube algorithms</a>.</li>
<li>Continued leveling up in Haskell with a <a href="https://github.com/xaviershay/factorio-hs">Factorio data parser</a> (likely broken on recent versions) and a woefully incomplete <a href="https://github.com/xaviershay/mopus-gagnum">clone of Opus Magnum</a>. I’m feeling reasonably confident with it now, and it’s currently my go-to language for new projects.</li>
</ul>
And as always, a <a href="https://github.com/looker/looker-sdk-ruby/issues/55">smattering</a> <a href="https://github.com/stripe-contrib/pagerbot/issues/46">of</a> <a href="https://github.com/seattlerb/minitest/pull/672">issues</a> to help improve documentation. Leave things better than you found them!
Next year, I’m looking forward to playing around with <a href="https://firebase.google.com/">Firebase</a>, and checking back in on <a href="https://www.rust-lang.org/en-US/">Rust</a> to see how it’s progressing.</content><author><name></name></author><category term="code" /><summary type="html">This was a light year for open source contribution, 168 by Github’s count (and some uncounted on Bitbucket). I’ve continued on the RSpec core team, though the project is mostly stable and I haven’t been particularly active. I did find time to experiment with some new things however:</summary></entry><entry><title type="html">Migrating Enki to Jekyll</title><link href="http://localhost:4000/2017/01/01/migrating-enki-to-jekyll/" rel="alternate" type="text/html" title="Migrating Enki to Jekyll" /><published>2017-01-01T00:00:00+00:00</published><updated>2017-01-01T00:00:00+00:00</updated><id>http://localhost:4000/2017/01/01/migrating-enki-to-jekyll</id><content type="html" xml:base="http://localhost:4000/2017/01/01/migrating-enki-to-jekyll/">I just converted this blog from a dynamic <a href="https://github.com/xaviershay/enki">Enki</a> site to a static <a href="https://jekyllrb.com/">Jekyll</a> one. I wanted to get rid of the comments, add SSL, and not have to upgrade Rails so often. I prefer composing locally also.
First, I exported all of the posts to lesstile templates using a rake task.<table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt>3<tt>
</tt>4<tt>
</tt>5<tt>
</tt>6<tt>
</tt>7<tt>
</tt>8<tt>
</tt>9<tt>
</tt>10<tt>
</tt>11<tt>
</tt>12<tt>
</tt>13<tt>
</tt>14<tt>
</tt>15<tt>
</tt>16<tt>
</tt>17<tt>
</tt>18<tt>
</tt>19<tt>
</tt>20<tt>
</tt>21<tt>
</tt>22<tt>
</tt>23<tt>
</tt>24<tt>
</tt>25<tt>
</tt>26<tt>
</tt>27<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }">task :export_posts =&gt; :environment do<tt>
</tt> Post.find_each do |post|<tt>
</tt> filename = &quot;%s-%s.lesstile&quot; % [<tt>
</tt> post.published_at.strftime(&quot;%Y-%m-%d&quot;),<tt>
</tt> post.slug<tt>
</tt> ]<tt>
</tt><tt>
</tt> dir = &quot;_posts&quot;<tt>
</tt> yaml_sep = &quot;---&quot;<tt>
</tt><tt>
</tt> puts filename<tt>
</tt><tt>
</tt> body = &lt;&lt;-EOS<tt>
</tt>#{yaml_sep}<tt>
</tt>layout: post<tt>
</tt>title: #{post.title.inspect}<tt>
</tt>date: #{post.published_at.strftime(&quot;%F %T %:z&quot;)}<tt>
</tt>tags: #{post.tags.map {|x| x.name.downcase }.sort.inspect}<tt>
</tt>#{yaml_sep}<tt>
</tt>{% raw %}<tt>
</tt>#{post.body}<tt>
</tt>{% endraw %}<tt>
</tt> EOS<tt>
</tt><tt>
</tt> File.write(File.join(dir, filename), body)<tt>
</tt> end<tt>
</tt>end<tt>
</tt></pre></td>
</tr></table>
<a href="https://github.com/xaviershay/lesstile">Lesstile</a> is a wrapper around Textile that provides some extra functionality, so a custom converter is also needed. Put the following in <code>_plugins/lesstile.rb</code> (with associated additions to your <code>Gemfile</code>):<table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt>3<tt>
</tt>4<tt>
</tt>5<tt>
</tt>6<tt>
</tt>7<tt>
</tt>8<tt>
</tt>9<tt>
</tt>10<tt>
</tt>11<tt>
</tt>12<tt>
</tt>13<tt>
</tt>14<tt>
</tt>15<tt>
</tt>16<tt>
</tt>17<tt>
</tt>18<tt>
</tt>19<tt>
</tt>20<tt>
</tt>21<tt>
</tt>22<tt>
</tt>23<tt>
</tt>24<tt>
</tt>25<tt>
</tt>26<tt>
</tt>27<tt>
</tt>28<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }">require 'lesstile'<tt>
</tt>require 'coderay'<tt>
</tt>require 'RedCloth'<tt>
</tt><tt>
</tt>module Jekyll<tt>
</tt> class LesstileConverter &lt; Converter<tt>
</tt> safe true<tt>
</tt> priority :low<tt>
</tt><tt>
</tt> def matches(ext)<tt>
</tt> ext =~ /^\.lesstile$/i<tt>
</tt> end<tt>
</tt><tt>
</tt> def output_ext(ext)<tt>
</tt> &quot;.html&quot;<tt>
</tt> end<tt>
</tt><tt>
</tt> def convert(content)<tt>
</tt> Lesstile.format_as_xhtml(<tt>
</tt> content,<tt>
</tt> :text_formatter =&gt; lambda {|text|<tt>
</tt> RedCloth.new(CGI::unescapeHTML(text)).to_html<tt>
</tt> },<tt>
</tt> :code_formatter =&gt; Lesstile::CodeRayFormatter<tt>
</tt> )<tt>
</tt> end<tt>
</tt> end<tt>
</tt>end<tt>
</tt></pre></td>
</tr></table>
The <code>permalink</code> configuration option needs to be set to match existing URLs, and to create the tag pages, use the <code>jekyll-archives</code> plugin.<table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt>3<tt>
</tt>4<tt>
</tt>5<tt>
</tt>6<tt>
</tt>7<tt>
</tt>8<tt>
</tt>9<tt>
</tt>10<tt>
</tt>11<tt>
</tt>12<tt>
</tt>13<tt>
</tt>14<tt>
</tt>15<tt>
</tt>16<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }">permalink: &quot;/:year/:month/:day/:title/&quot;<tt>
</tt><tt>
</tt>assets:<tt>
</tt> digest: true<tt>
</tt><tt>
</tt>&quot;jekyll-archives&quot;:<tt>
</tt> enabled:<tt>
</tt> - tags<tt>
</tt> layout: 'tag'<tt>
</tt> permalinks:<tt>
</tt> tag: '/:name/'<tt>
</tt><tt>
</tt>gems:<tt>
</tt> - jekyll-feed<tt>
</tt> - jekyll-assets<tt>
</tt> - jekyll-archives<tt>
</tt></pre></td>
</tr></table>
For the archives page, use an empty <code>archives.md</code> in the root directory with a custom layout:<table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt>3<tt>
</tt>4<tt>
</tt>5<tt>
</tt>6<tt>
</tt>7<tt>
</tt>8<tt>
</tt>9<tt>
</tt>10<tt>
</tt>11<tt>
</tt>12<tt>
</tt>13<tt>
</tt>14<tt>
</tt>15<tt>
</tt>16<tt>
</tt>17<tt>
</tt>18<tt>
</tt>19<tt>
</tt>20<tt>
</tt>21<tt>
</tt>22<tt>
</tt>23<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }">{% include head.html %}<tt>
</tt>{% assign last_month = nil %}<tt>
</tt>&lt;ul&gt;<tt>
</tt>{% for post in site.posts %}<tt>
</tt> {% assign current_month = post.date | date: '%B %Y' %}<tt>
</tt> {% if current_month != last_month %}<tt>
</tt> &lt;/ul&gt;<tt>
</tt> &lt;h3&gt;{{ current_month }}&lt;/h3&gt;<tt>
</tt> &lt;ul&gt;<tt>
</tt> {% endif %}<tt>
</tt><tt>
</tt> &lt;li&gt;<tt>
</tt> &lt;a href=&quot;{{ post.url }}&quot;&gt;{{ post.title }}&lt;/a&gt;<tt>
</tt><tt>
</tt> {% if post.tags != empty %}<tt>
</tt> ({% for tag in post.tags %}&lt;a href='/{{ tag }}'&gt;{{ tag }}&lt;/a&gt;{% if forloop.last %}{% else %}, {% endif %}{% endfor %})<tt>
</tt> {% endif %}<tt>
</tt> &lt;/li&gt;<tt>
</tt><tt>
</tt> {% assign last_month = current_month %}<tt>
</tt>{% endfor %}<tt>
</tt>&lt;/ul&gt;<tt>
</tt>{% include footer.html %}<tt>
</tt></pre></td>
</tr></table>
For a full example, including a recommended set of layouts and includes, see the <a href="https://github.com/xaviershay/rhnh-static">new sources</a> for this site.</content><author><name></name></author><category term="code" /><category term="enki" /><category term="jekyll" /><summary type="html">I just converted this blog from a dynamic Enki site to a static Jekyll one. I wanted to get rid of the comments, add SSL, and not have to upgrade Rails so often. I prefer composing locally also.</summary></entry><entry><title type="html">New non-tech blog</title><link href="http://localhost:4000/2014/10/12/new-non-tech-blog/" rel="alternate" type="text/html" title="New non-tech blog" /><published>2014-10-12T22:30:00+00:00</published><updated>2014-10-12T22:30:00+00:00</updated><id>http://localhost:4000/2014/10/12/new-non-tech-blog</id><content type="html" xml:base="http://localhost:4000/2014/10/12/new-non-tech-blog/">Have been writing a bit recently, just not here. More politics and book reviews. I&#8217;ve separated it out over on <a href="https://xaviershay.github.io/blog/">Github pages</a>.</content><author><name></name></author><summary type="html">Have been writing a bit recently, just not here. More politics and book reviews. I&#8217;ve separated it out over on Github pages.</summary></entry><entry><title type="html">Dropwizard logger for Ruby and WEBrick</title><link href="http://localhost:4000/2014/08/17/dropwizard-logger-for-ruby-and-webrick/" rel="alternate" type="text/html" title="Dropwizard logger for Ruby and WEBrick" /><published>2014-08-17T22:30:19+00:00</published><updated>2014-08-17T22:30:19+00:00</updated><id>http://localhost:4000/2014/08/17/dropwizard-logger-for-ruby-and-webrick</id><content type="html" xml:base="http://localhost:4000/2014/08/17/dropwizard-logger-for-ruby-and-webrick/">Wouldn&#8217;t it be great if instead of webrick logs looking like:<table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt>3<tt>
</tt>4<tt>
</tt>5<tt>
</tt>6<tt>
</tt>7<tt>
</tt>8<tt>
</tt>9<tt>
</tt>10<tt>
</tt>11<tt>
</tt>12<tt>
</tt>13<tt>
</tt>14<tt>
</tt>15<tt>
</tt>16<tt>
</tt>17<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }">&gt; ruby server.rb<tt>
</tt>[2014-08-17 15:29:10] INFO WEBrick 1.3.1<tt>
</tt>[2014-08-17 15:29:10] INFO ruby 2.1.1 (2014-02-24) [x86_64-darwin13.0]<tt>
</tt>[2014-08-17 15:29:10] INFO WEBrick::HTTPServer#start: pid=17304 port=8000<tt>
</tt>D, [2014-08-17T15:29:11.452223 #17304] DEBUG -- : hello from in the request<tt>
</tt>localhost - - [17/Aug/2014:15:29:11 PDT] &quot;GET / HTTP/1.1&quot; 200 13<tt>
</tt>- -&gt; /<tt>
</tt>E, [2014-08-17T15:29:12.787505 #17304] ERROR -- : fail (RuntimeError)<tt>
</tt>server.rb:57:in `block in &lt;main&gt;'<tt>
</tt>/Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpservlet/prochandler.rb:38:in `call'<tt>
</tt>/Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpservlet/prochandler.rb:38:in `do_GET'<tt>
</tt>/Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpservlet/abstract.rb:106:in `service'<tt>
</tt>/Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpserver.rb:138:in `service'<tt>
</tt>/Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpserver.rb:94:in `run'<tt>
</tt>/Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/server.rb:295:in `block in start_thread'<tt>
</tt>localhost - - [17/Aug/2014:15:29:12 PDT] &quot;GET /fail HTTP/1.1&quot; 500 6<tt>
</tt>- -&gt; /fail<tt>
</tt></pre></td>
</tr></table>
They looked like:<table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt>3<tt>
</tt>4<tt>
</tt>5<tt>
</tt>6<tt>
</tt>7<tt>
</tt>8<tt>
</tt>9<tt>
</tt>10<tt>
</tt>11<tt>
</tt>12<tt>
</tt>13<tt>
</tt>14<tt>
</tt>15<tt>
</tt>16<tt>
</tt>17<tt>
</tt>18<tt>
</tt>19<tt>
</tt>20<tt>
</tt>21<tt>
</tt>22<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }">&gt; ruby server.rb<tt>
</tt><tt>
</tt> ,~~.,''&quot;'`'.~~.<tt>
</tt> : {` .- _ -. '} ;<tt>
</tt> `: O(_)O ;'<tt>
</tt> '; ._|_, ;` i am starting the server<tt>
</tt> '`-.\_/,.'`<tt>
</tt><tt>
</tt>INFO [2014-08-17 22:28:13,186] webrick: WEBrick 1.3.1<tt>
</tt>INFO [2014-08-17 22:28:13,186] webrick: ruby 2.1.1 (2014-02-24) [x86_64-darwin13.0]<tt>
</tt>INFO [2014-08-17 22:28:13,187] webrick: WEBrick::HTTPServer#start: pid=17253 port=8000<tt>
</tt>DEBUG [2014-08-17 22:28:14,738] app: hello from in the request<tt>
</tt>INFO [2014-08-17 15:28:14,736] webrick: GET / 200<tt>
</tt>ERROR [2014-08-17 22:28:15,603] app: RuntimeError: fail<tt>
</tt>! server.rb:57:in `block in &lt;main&gt;'<tt>
</tt>! /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpservlet/prochandler.rb:38:in `call'<tt>
</tt>! /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpservlet/prochandler.rb:38:in `do_GET'<tt>
</tt>! /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpservlet/abstract.rb:106:in `service'<tt>
</tt>! /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpserver.rb:138:in `service'<tt>
</tt>! /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpserver.rb:94:in `run'<tt>
</tt>! /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/server.rb:295:in `block in start_thread'<tt>
</tt>INFO [2014-08-17 15:28:15,602] webrick: GET /fail 500<tt>
</tt></pre></td>
</tr></table>
I thought so, hence:<table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt>3<tt>
</tt>4<tt>
</tt>5<tt>
</tt>6<tt>
</tt>7<tt>
</tt>8<tt>
</tt>9<tt>
</tt>10<tt>
</tt>11<tt>
</tt>12<tt>
</tt>13<tt>
</tt>14<tt>
</tt>15<tt>
</tt>16<tt>
</tt>17<tt>
</tt>18<tt>
</tt>19<tt>
</tt>20<tt>
</tt>21<tt>
</tt>22<tt>
</tt>23<tt>
</tt>24<tt>
</tt>25<tt>
</tt>26<tt>
</tt>27<tt>
</tt>28<tt>
</tt>29<tt>
</tt>30<tt>
</tt>31<tt>
</tt>32<tt>
</tt>33<tt>
</tt>34<tt>
</tt>35<tt>
</tt>36<tt>
</tt>37<tt>
</tt>38<tt>
</tt>39<tt>
</tt>40<tt>
</tt>41<tt>
</tt>42<tt>
</tt>43<tt>
</tt>44<tt>
</tt>45<tt>
</tt>46<tt>
</tt>47<tt>
</tt>48<tt>
</tt>49<tt>
</tt>50<tt>
</tt>51<tt>
</tt>52<tt>
</tt>53<tt>
</tt>54<tt>
</tt>55<tt>
</tt>56<tt>
</tt>57<tt>
</tt>58<tt>
</tt>59<tt>
</tt>60<tt>
</tt>61<tt>
</tt>62<tt>
</tt>63<tt>
</tt>64<tt>
</tt>65<tt>
</tt>66<tt>
</tt>67<tt>
</tt>68<tt>
</tt>69<tt>
</tt>70<tt>
</tt>71<tt>
</tt>72<tt>
</tt>73<tt>
</tt>74<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }">require 'webrick'<tt>
</tt>require 'logger'<tt>
</tt><tt>
</tt>puts &lt;&lt;-BANNER<tt>
</tt><tt>
</tt> ,~~.,''&quot;'`'.~~.<tt>
</tt> : {` .- _ -. '} ;<tt>
</tt> `: O(_)O ;'<tt>
</tt> '; ._|_, ;` i am starting the server<tt>
</tt> '`-.\\_/,.'`<tt>
</tt><tt>
</tt>BANNER<tt>
</tt><tt>
</tt>class DropwizardLogger &lt; Logger<tt>
</tt> def initialize(label, *args)<tt>
</tt> super(*args)<tt>
</tt> @label = label<tt>
</tt> end<tt>
</tt><tt>
</tt> def format_message(severity, timestamp, progname, msg)<tt>
</tt> &quot;%-5s [%s] %s: %s\n&quot; % [<tt>
</tt> severity,<tt>
</tt> timestamp.utc.strftime(&quot;%Y-%m-%d %H:%M:%S,%3N&quot;),<tt>
</tt> @label,<tt>
</tt> msg2str(msg),<tt>
</tt> ]<tt>
</tt> end<tt>
</tt><tt>
</tt> def msg2str(msg)<tt>
</tt> case msg<tt>
</tt> when String<tt>
</tt> msg<tt>
</tt> when Exception<tt>
</tt> (&quot;%s: %s&quot; % [msg.class, msg.message]) +<tt>
</tt> (msg.backtrace ? msg.backtrace.map {|x| &quot;\n! #{x}&quot; }.join : &quot;&quot;)<tt>
</tt> else<tt>
</tt> msg.inspect<tt>
</tt> end<tt>
</tt> end<tt>
</tt><tt>
</tt> def self.webrick_format(label)<tt>
</tt> &quot;INFO [%{%Y-%m-%d %H:%M:%S,%3N}t] #{label}: %m %U %s&quot;<tt>
</tt> end<tt>
</tt>end<tt>
</tt><tt>
</tt>server = WEBrick::HTTPServer.new \<tt>
</tt> :Port =&gt; 8000,<tt>
</tt> :Logger =&gt; DropwizardLogger.new(&quot;webrick&quot;, $stdout).tap {|x|<tt>
</tt> x.level = Logger::INFO<tt>
</tt> },<tt>
</tt> :AccessLog =&gt; [[$stdout, DropwizardLogger.webrick_format(&quot;webrick&quot;)]]<tt>
</tt><tt>
</tt>$logger = DropwizardLogger.new(&quot;app&quot;, $stdout)<tt>
</tt><tt>
</tt>server.mount_proc '/fail' do |req, res|<tt>
</tt> begin<tt>
</tt> raise 'fail'<tt>
</tt> rescue =&gt; e<tt>
</tt> $logger.error(e)<tt>
</tt> end<tt>
</tt> res.body = &quot;failed&quot;<tt>
</tt> res.status = 500<tt>
</tt>end<tt>
</tt><tt>
</tt>server.mount_proc '/' do |req, res|<tt>
</tt> $logger.debug(&quot;hello from in the request&quot;)<tt>
</tt> res.body = 'Hello, world!'<tt>
</tt>end<tt>
</tt><tt>
</tt>trap 'INT' do<tt>
</tt> server.shutdown<tt>
</tt>end<tt>
</tt><tt>
</tt>server.start<tt>
</tt></pre></td>
</tr></table></content><author><name></name></author><category term="code" /><category term="dropwizard" /><category term="ruby" /><category term="webrick" /><summary type="html">Wouldn&#8217;t it be great if instead of webrick logs looking like: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 &gt; ruby server.rb [2014-08-17 15:29:10] INFO WEBrick 1.3.1 [2014-08-17 15:29:10] INFO ruby 2.1.1 (2014-02-24) [x86_64-darwin13.0] [2014-08-17 15:29:10] INFO WEBrick::HTTPServer#start: pid=17304 port=8000 D, [2014-08-17T15:29:11.452223 #17304] DEBUG -- : hello from in the request localhost - - [17/Aug/2014:15:29:11 PDT] &quot;GET / HTTP/1.1&quot; 200 13 - -&gt; / E, [2014-08-17T15:29:12.787505 #17304] ERROR -- : fail (RuntimeError) server.rb:57:in `block in &lt;main&gt;' /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpservlet/prochandler.rb:38:in `call' /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpservlet/prochandler.rb:38:in `do_GET' /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpservlet/abstract.rb:106:in `service' /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpserver.rb:138:in `service' /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpserver.rb:94:in `run' /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/server.rb:295:in `block in start_thread' localhost - - [17/Aug/2014:15:29:12 PDT] &quot;GET /fail HTTP/1.1&quot; 500 6 - -&gt; /fail They looked like: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 &gt; ruby server.rb ,~~.,''&quot;'`'.~~. : {` .- _ -. '} ; `: O(_)O ;' '; ._|_, ;` i am starting the server '`-.\_/,.'` INFO [2014-08-17 22:28:13,186] webrick: WEBrick 1.3.1 INFO [2014-08-17 22:28:13,186] webrick: ruby 2.1.1 (2014-02-24) [x86_64-darwin13.0] INFO [2014-08-17 22:28:13,187] webrick: WEBrick::HTTPServer#start: pid=17253 port=8000 DEBUG [2014-08-17 22:28:14,738] app: hello from in the request INFO [2014-08-17 15:28:14,736] webrick: GET / 200 ERROR [2014-08-17 22:28:15,603] app: RuntimeError: fail ! server.rb:57:in `block in &lt;main&gt;' ! /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpservlet/prochandler.rb:38:in `call' ! /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpservlet/prochandler.rb:38:in `do_GET' ! /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpservlet/abstract.rb:106:in `service' ! /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpserver.rb:138:in `service' ! /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/httpserver.rb:94:in `run' ! /Users/xavier/.rubies/cruby-2.1.1/lib/ruby/2.1.0/webrick/server.rb:295:in `block in start_thread' INFO [2014-08-17 15:28:15,602] webrick: GET /fail 500 I thought so, hence: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 require 'webrick' require 'logger' puts &lt;&lt;-BANNER ,~~.,''&quot;'`'.~~. : {` .- _ -. '} ; `: O(_)O ;' '; ._|_, ;` i am starting the server '`-.\\_/,.'` BANNER class DropwizardLogger &lt; Logger def initialize(label, *args) super(*args) @label = label end def format_message(severity, timestamp, progname, msg) &quot;%-5s [%s] %s: %s\n&quot; % [ severity, timestamp.utc.strftime(&quot;%Y-%m-%d %H:%M:%S,%3N&quot;), @label, msg2str(msg), ] end def msg2str(msg) case msg when String msg when Exception (&quot;%s: %s&quot; % [msg.class, msg.message]) + (msg.backtrace ? msg.backtrace.map {|x| &quot;\n! #{x}&quot; }.join : &quot;&quot;) else msg.inspect end end def self.webrick_format(label) &quot;INFO [%{%Y-%m-%d %H:%M:%S,%3N}t] #{label}: %m %U %s&quot; end end server = WEBrick::HTTPServer.new \ :Port =&gt; 8000, :Logger =&gt; DropwizardLogger.new(&quot;webrick&quot;, $stdout).tap {|x| x.level = Logger::INFO }, :AccessLog =&gt; [[$stdout, DropwizardLogger.webrick_format(&quot;webrick&quot;)]] $logger = DropwizardLogger.new(&quot;app&quot;, $stdout) server.mount_proc '/fail' do |req, res| begin raise 'fail' rescue =&gt; e $logger.error(e) end res.body = &quot;failed&quot; res.status = 500 end server.mount_proc '/' do |req, res| $logger.debug(&quot;hello from in the request&quot;) res.body = 'Hello, world!' end trap 'INT' do server.shutdown end server.start</summary></entry><entry><title type="html">Querying consul with range</title><link href="http://localhost:4000/2014/08/16/querying-consul-with-range/" rel="alternate" type="text/html" title="Querying consul with range" /><published>2014-08-16T23:32:00+00:00</published><updated>2014-08-16T23:32:00+00:00</updated><id>http://localhost:4000/2014/08/16/querying-consul-with-range</id><content type="html" xml:base="http://localhost:4000/2014/08/16/querying-consul-with-range/">Disclaimer: this has not been tried in a production environment. It is a weekend hack.
<a href="http://www.consul.io/">Consul</a> is a highly available, datacenter aware, service discovery mechanism. <a href="http://godoc.org/github.com/xaviershay/grange">Range</a> is a query language for selecting information out of arbitrary, self-referential metadata. I combined the two!
Start by firing up a two node consul cluster, per the <a href="http://www.consul.io/intro/getting-started/join.html">getting started guide</a>. On the master node, grab the <a href="https://github.com/xaviershay/grange-server/pull/4"><code>consul</code> branch of grange-server</a> and run it with the following config:<table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt>3<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }">[rangeserver]<tt>
</tt>loglevel=DEBUG<tt>
</tt>consul=true<tt>
</tt></pre></td>
</tr></table>
(It could run against any consul agent, but it&#8217;s easier to demo on the master node.)
Querying range, we already see a consul cluster, <code>cluster</code>. This is a default service containing the consul servers.<table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt>3<tt>
</tt>4<tt>
</tt>5<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }">&gt; export RANGE_HOST=172.20.20.10<tt>
</tt>&gt; erg &quot;allclusters()&quot;<tt>
</tt>consul<tt>
</tt>&gt; erg &quot;%consul&quot;<tt>
</tt>agent-one<tt>
</tt></pre></td>
</tr></table>
Add a new service to the agents, and it shows up in range!<table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt>3<tt>
</tt>4<tt>
</tt>5<tt>
</tt>6<tt>
</tt>7<tt>
</tt>8<tt>
</tt>9<tt>
</tt>10<tt>
</tt>11<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }">n2&gt; curl -v -X PUT --data '{&quot;name&quot;: &quot;web&quot;, &quot;port&quot;: 80}' http://localhost:8500/v1/agent/service/register<tt>
</tt><tt>
</tt>&gt; erg &quot;allclusters()&quot;<tt>
</tt>consul,web<tt>
</tt>&gt; erg &quot;%web&quot;<tt>
</tt>agent-two<tt>
</tt><tt>
</tt>n1&gt; curl -v -X PUT --data '{&quot;name&quot;: &quot;web&quot;, &quot;port&quot;: 80}' http://localhost:8500/v1/agent/service/register<tt>
</tt><tt>
</tt>&gt; erg &quot;%web&quot;<tt>
</tt>agent-one,agent-two<tt>
</tt></pre></td>
</tr></table>
Though eventually consistent, range is a big improvement over the consul HTTP API for quick ad-hoc queries against your production layout, particularly when combined with other metadata. How many nodes are running redis? What services are running on a particular rack?
This is just a proof of concept for now, but I&#8217;m excited about the potential. To be useable it needs to be tested against production sized clusters, better handling of error conditions, and some code review (in particular around handling cluster state changes).</content><author><name></name></author><category term="code" /><category term="consul" /><category term="range" /><summary type="html">Disclaimer: this has not been tried in a production environment. It is a weekend hack. Consul is a highly available, datacenter aware, service discovery mechanism. Range is a query language for selecting information out of arbitrary, self-referential metadata. I combined the two! Start by firing up a two node consul cluster, per the getting started guide. On the master node, grab the consul branch of grange-server and run it with the following config: 1 2 3 [rangeserver] loglevel=DEBUG consul=true (It could run against any consul agent, but it&#8217;s easier to demo on the master node.) Querying range, we already see a consul cluster, cluster. This is a default service containing the consul servers. 1 2 3 4 5 &gt; export RANGE_HOST=172.20.20.10 &gt; erg &quot;allclusters()&quot; consul &gt; erg &quot;%consul&quot; agent-one Add a new service to the agents, and it shows up in range! 1 2 3 4 5 6 7 8 9 10 11 n2&gt; curl -v -X PUT --data '{&quot;name&quot;: &quot;web&quot;, &quot;port&quot;: 80}' http://localhost:8500/v1/agent/service/register &gt; erg &quot;allclusters()&quot; consul,web &gt; erg &quot;%web&quot; agent-two n1&gt; curl -v -X PUT --data '{&quot;name&quot;: &quot;web&quot;, &quot;port&quot;: 80}' http://localhost:8500/v1/agent/service/register &gt; erg &quot;%web&quot; agent-one,agent-two Though eventually consistent, range is a big improvement over the consul HTTP API for quick ad-hoc queries against your production layout, particularly when combined with other metadata. How many nodes are running redis? What services are running on a particular rack? This is just a proof of concept for now, but I&#8217;m excited about the potential. To be useable it needs to be tested against production sized clusters, better handling of error conditions, and some code review (in particular around handling cluster state changes).</summary></entry><entry><title type="html">Bash script to keep a git clone synced with a remote</title><link href="http://localhost:4000/2014/04/26/bash-script-to-keep-a-git-clone-synced-with-a-remote/" rel="alternate" type="text/html" title="Bash script to keep a git clone synced with a remote" /><published>2014-04-26T18:20:00+00:00</published><updated>2014-04-26T18:20:00+00:00</updated><id>http://localhost:4000/2014/04/26/bash-script-to-keep-a-git-clone-synced-with-a-remote</id><content type="html" xml:base="http://localhost:4000/2014/04/26/bash-script-to-keep-a-git-clone-synced-with-a-remote/">Use the following under a process manager (such as runit) to keep a local git clone in sync with a remote, when a push based solution isn&#8217;t an option. Most other versions either neglect to verify remote is correct, or use <code>git pull</code> which can fail if someone has been monkeying with the local version.<table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt>3<tt>
</tt>4<tt>
</tt>5<tt>
</tt>6<tt>
</tt>7<tt>
</tt>8<tt>
</tt>9<tt>
</tt>10<tt>
</tt>11<tt>
</tt>12<tt>
</tt>13<tt>
</tt>14<tt>
</tt>15<tt>
</tt>16<tt>
</tt>17<tt>
</tt>18<tt>
</tt>19<tt>
</tt>20<tt>
</tt>21<tt>
</tt>22<tt>
</tt>23<tt>
</tt>24<tt>
</tt>25<tt>
</tt>26<tt>
</tt>27<tt>
</tt>28<tt>
</tt>29<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }">function update_git_repo() {<tt>
</tt> GIT_DIR=$1<tt>
</tt> GIT_REMOTE=$2<tt>
</tt> GIT_BRANCH=${3:-master}<tt>
</tt><tt>
</tt> if [ ! -d $GIT_DIR ]; then<tt>
</tt> CURRENT_SHA=&quot;&quot;<tt>
</tt> git clone --depth 1 $GIT_REMOTE $GIT_DIR -b $GIT_BRANCH<tt>
</tt> else<tt>
</tt> CURRENT_REMOTE=$(cd $GIT_DIR &amp;&amp; git config --get remote.origin.url || true)<tt>
</tt><tt>
</tt> if [ &quot;$GIT_REMOTE&quot; == &quot;$CURRENT_REMOTE&quot; ]; then<tt>
</tt> CURRENT_SHA=$(cat $GIT_DIR/.git/refs/heads/$GIT_BRANCH)<tt>
</tt> else<tt>
</tt> rm -Rf $GIT_DIR<tt>
</tt> exit 0 # Process manager should restart this script<tt>
</tt> fi<tt>
</tt> fi<tt>
</tt><tt>
</tt> cd $GIT_DIR &amp;&amp; \<tt>
</tt> git fetch &amp;&amp; \<tt>
</tt> git reset --hard origin/$GIT_BRANCH<tt>
</tt><tt>
</tt> NEW_SHA=$(cat $GIT_DIR/.git/refs/heads/$GIT_BRANCH)<tt>
</tt>}<tt>
</tt><tt>
</tt>update_git_repo &quot;/tmp/myrepo&quot; &quot;git://example.com/my/repo.git&quot;<tt>
</tt><tt>
</tt>sleep 60 # No need for a tight loop<tt>
</tt></pre></td>
</tr></table></content><author><name></name></author><category term="bash" /><category term="code" /><category term="git" /><summary type="html">Use the following under a process manager (such as runit) to keep a local git clone in sync with a remote, when a push based solution isn&#8217;t an option. Most other versions either neglect to verify remote is correct, or use git pull which can fail if someone has been monkeying with the local version. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 function update_git_repo() { GIT_DIR=$1 GIT_REMOTE=$2 GIT_BRANCH=${3:-master} if [ ! -d $GIT_DIR ]; then CURRENT_SHA=&quot;&quot; git clone --depth 1 $GIT_REMOTE $GIT_DIR -b $GIT_BRANCH else CURRENT_REMOTE=$(cd $GIT_DIR &amp;&amp; git config --get remote.origin.url || true) if [ &quot;$GIT_REMOTE&quot; == &quot;$CURRENT_REMOTE&quot; ]; then CURRENT_SHA=$(cat $GIT_DIR/.git/refs/heads/$GIT_BRANCH) else rm -Rf $GIT_DIR exit 0 # Process manager should restart this script fi fi cd $GIT_DIR &amp;&amp; \ git fetch &amp;&amp; \ git reset --hard origin/$GIT_BRANCH NEW_SHA=$(cat $GIT_DIR/.git/refs/heads/$GIT_BRANCH) } update_git_repo &quot;/tmp/myrepo&quot; &quot;git://example.com/my/repo.git&quot; sleep 60 # No need for a tight loop</summary></entry><entry><title type="html">Ruby progress bar, no gems</title><link href="http://localhost:4000/2014/03/29/ruby-progress-bar-no-gems/" rel="alternate" type="text/html" title="Ruby progress bar, no gems" /><published>2014-03-29T22:17:00+00:00</published><updated>2014-03-29T22:17:00+00:00</updated><id>http://localhost:4000/2014/03/29/ruby-progress-bar-no-gems</id><content type="html" xml:base="http://localhost:4000/2014/03/29/ruby-progress-bar-no-gems/"><table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt>3<tt>
</tt>4<tt>
</tt>5<tt>
</tt>6<tt>
</tt>7<tt>
</tt>8<tt>
</tt>9<tt>
</tt>10<tt>
</tt>11<tt>
</tt>12<tt>
</tt>13<tt>
</tt>14<tt>
</tt>15<tt>
</tt>16<tt>
</tt>17<tt>
</tt>18<tt>
</tt>19<tt>
</tt>20<tt>
</tt>21<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }">def import(filename, out = $stdout, &amp;block)<tt>
</tt> # Yes, there are gems that do progress bars.<tt>
</tt> # No, I'm not about to add another dependency for something this simple.<tt>
</tt> width = 50<tt>
</tt> processed = 0<tt>
</tt> printed = 0<tt>
</tt> total = File.read(filename).lines.length.to_f<tt>
</tt> label = File.basename(filename, '.csv')<tt>
</tt><tt>
</tt> out.print &quot;%11s: |&quot; % label<tt>
</tt><tt>
</tt> CSV.foreach(filename, headers: true) do |row|<tt>
</tt> yield row<tt>
</tt><tt>
</tt> processed += 1<tt>
</tt> wanted = (processed / total * width).to_i<tt>
</tt> out.print &quot;-&quot; * (wanted - printed)<tt>
</tt> printed = wanted<tt>
</tt> end<tt>
</tt> out.puts &quot;|&quot;<tt>
</tt>end<tt>
</tt></pre></td>
</tr></table>
<table class="CodeRay"><tr>
<td class="line_numbers" title="click to toggle" onclick="with (this.firstChild.style) { display = (display == '') ? 'none' : '' }"><pre>1<tt>
</tt>2<tt>
</tt></pre></td>
<td class="code"><pre ondblclick="with (this.style) { overflow = (overflow == 'auto' || overflow == '') ? 'visible' : 'auto' }"> file_1: |--------------------------------------------------|<tt>
</tt> file_2: |--------------------------------------------------|<tt>
</tt></pre></td>
</tr></table></content><author><name></name></author><category term="code" /><category term="ruby" /><summary type="html">1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 def import(filename, out = $stdout, &amp;block) # Yes, there are gems that do progress bars. # No, I'm not about to add another dependency for something this simple. width = 50 processed = 0 printed = 0 total = File.read(filename).lines.length.to_f label = File.basename(filename, '.csv') out.print &quot;%11s: |&quot; % label CSV.foreach(filename, headers: true) do |row| yield row processed += 1 wanted = (processed / total * width).to_i out.print &quot;-&quot; * (wanted - printed) printed = wanted end out.puts &quot;|&quot; end 1 2 file_1: |--------------------------------------------------| file_2: |--------------------------------------------------|</summary></entry></feed>

Home · About · News · Docs · Terms