My production environment has to use PHP 5.6.40 because of some legacy plugins that do not exist, or were not installed with a package manager (e.g. composer) because it didn’t exist at the time, and the business has not decided to invest the time to update all the features to use newer versions…and in some cases there are no newer versions.

So after updating my environment, I had to make sure I didn’t get put back on PHP 7.4.x, and was still on PHP 5.6. I was, but it was now broken.

Error Messages

The first error message I got when trying to check the PHP version

tsnMac:~ neotsn$ php -v
dyld: Library not loaded: /usr/local/opt/openssl/lib/libcrypto.1.0.0.dylib
  Referenced from: /usr/local/bin/php
  Reason: image not found
Abort trap: 6

This is because macOS has its own SSL library installed, and PHP 5.6 requires OpenSSL…and isn’t compatible with OpenSSL 1.1, which is what got installed in the update.

After reinstalling OpenSSL 1.0 from a direct package link, I encountered the next message, which was much more difficult to solve, since icu4c is covered by macOS Data Integrity Protection.

dyld: Library not loaded: /usr/local/opt/icu4c/lib/libicui18n.64.dylib
  Referenced from: /usr/local/Cellar/php@5.6/5.6.40/bin/php
  Reason: image not found

The Fixes

Thanks to the instructions over in this Github Issue from JParkinson1991, and various other articles and blogs around the internet I’ve pieced together this solution, but his updated comment provided the majority of the answer.

Starting over?

Obviously you’re going to need Homebrew, which you probably already have if you’re here… but in case you wiped your hard disk (as I did) and have to start over…

# Install Homebrew
/usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"

# Bring in deprecated php versions
brew tap exolnet/homebrew-deprecated

Now you’ll have Homebrew installed and the deprecated PHP versions available to you.

Update these packages

Before you start down the path, make sure the problematic packages (openssl and icu4c) are at the latest versions (and indeed installed).

# Make sure these are at the latest versions
brew install openssl && brew upgrade openssl
brew upgrade icu4c

Patching PHP 5.6

Note: This will soon be fixed

At the time of this writing, there exists a merge request for the php@5.6 package where the following patch will no longer be required. I do not have instructions on what to do once it has been, but presumably this article will never reach your eyeballs if you have the latest version of these eXolnet repositories.

The patch…

If the above merge request is still outstanding, then let’s edit the php@5.6 package…

# Edit php@5.6 to add a patch 
brew edit php@5.6

This will open up the edited file in vim (or at least it did for me). So jump to line 45-48. Look for the def install around there. Right before that line, insert these lines…

# At line 45, before the `def install` command, add these lines
patch do
  url "https://raw.githubusercontent.com/opencomputeproject/Rack-Manager/master/Contrib-Inspur/openbmc/meta-openembedded/meta-oe/recipes-devtools/php/php/0001-PHP-5.6-LibSSL-1.1-compatibility.patch"
  sha256 "c9715b544ae249c0e76136dfadd9d282237233459694b9e75d0e3e094ab0c993"
end

Save your changes and exit with an esc press and a :wq.

Reinstall php@5.6 from source

Now that our package is different from what was downloaded, we need to reinstall it from that source…

# Reinstall php@5.6 from the edited source
brew reinstall --build-from-source php@5.6

This will take a minute or two. When it’s done you should be able to do a php -v command to see that there are no more errors and you’re on the correct version.

Run the checkup

Afterward, be sure to run the doctor and cleanup commands to make sure everything is both good, and clean.

# Get a checkup
brew doctor
brew cleanup

Minor Troubleshooting Suggestions

If you happen to see you’re not on the right version of PHP after this, you may have to do a forced overwrite link.

# Only if php -v is not 5.6.x
brew link --overwrite --force php@5.6

If you need to have multiple versions of PHP installed, then you’re going to want sphp installed.

# Install sphp to switch versions 
curl -L https://gist.githubusercontent.com/rhukster/f4c04f1bf59e0b74e335ee5d186a98e2/raw > /usr/local/bin/sphp
# Make it executable
chmod +x /usr/local/bin/sphp

Then you’ll be able to switch between them with sphp 5.6 or any other installed versions using their major/minor version numbers.

If you have trouble, you can try dropping a comment here, or onto the linked Github pages where there are better experts than I who may be able to help.

It turns out that the existing topics on the phpBB forums are not entirely helpful for doing something as advanced as this project has turned out to be, and most topics are trivially solved with clearing cache or reading the documentation.

Occasionally, I need to do something that requires multiple help forum threads and documentation pages, so as I come across those solutions, I’ll combine them into a post here, because that’s the purpose of this blog, after all.

Rendering Template Partials

The first super complicated issue to solve was rendering what amounts to “part” of a template (hereafter referred to as “Partials”), without the html shell. So, let’s get right to it then…you didn’t come here for my life’s story.

Goal: AJAX HTML Response

To start, I have an infinite scroll that delivers a set of Topics, instead of using pagination. To avoid duplicating twig templating engine in JS, I’ll just deliver HTML, and use JS to update any existing elements, or append new elements.

Step 1: Create template class

First, create a template.php file to add a method for rendering Partials. I’ve placed mine for the extension “tsn/tsn” at tsn/tsn/framework/logic/template.php. Here’s a look at the basics of the file.

<?php

namespace tsn\tsn\framework\logic;

use phpbb\template\twig\twig;
use Twig\Error\LoaderError;
use Twig\Error\RuntimeError;
use Twig\Error\SyntaxError;

/**
 * Class template
 * Adds functionality for rendering template partials
 * @package tsn\tsn\framework\logic
 */
class template extends twig
{

    // Directories
    const TEMPLATE_DIR = '@tsn_tsn/';
    const PARTIALS_DIR = self::TEMPLATE_DIR . 'partials/';

    // Partials
    const P_MYSPOT_FEED = self::PARTIALS_DIR . 'tsn_myspot_feed.html';
    const P_TOPIC_CARD = self::PARTIALS_DIR . 'tsn_topic_card.html';

    /**
     * @param string $templateConstant A template path name or constant
     *
     * @return false|string|null
     */
    public function renderPartial($templateConstant)
    {
        try {

            $output = $this->twig->render($templateConstant, $this->get_template_vars());
            // Usually this is via AJAX, so compress the whitespace
            $output = preg_replace('/\s+/', ' ', $output);

        } catch (LoaderError $e) {
            $output = false;
            error_log('Template Loader error: ' . $e->getMessage() . ' :: ' . $e->getCode());
        } catch (RuntimeError $e) {
            $output = false;
            error_log('Template Runtime error: ' . $e->getMessage() . ' :: ' . $e->getCode());
        } catch (SyntaxError $e) {
            $output = false;
            error_log('Template Syntax error: ' . $e->getMessage() . ' :: ' . $e->getCode());
        }

        return $output;
    }
}

It inherits phpbb\template\twig\twig, and thus its constructor. It has constants to help combine the various directories and file names should any of them need to be changed along the way. And it adds a single method to allow the render of a partial template.

This new method is a wrapper for the call to the inherited twig property and its render() method. This new method receives a single parameter for the Partial template name to render. The template name (or constant) and template vars are then passed into the twig’s render method, and the HTML is returned. If the render method fails, it throws exceptions, so catch them and handle them as appropriate for your needs.

Optional: In my case, I am using this for AJAX right now, so I am reducing all sequential whitespace to a single space character.

Step 2: Service Decoration

Now that we have the class, let’s put it in play. This is done with Service Decoration in phpBB 3.2.x, which tells the phpBB code to use this class instead of the original class. In this case, anywhere we would have used $this->template, it will invoke our class instead of the original class.

This block goes into your services.yml file

tsn.tsn.framework.logic.template:
    class: tsn\tsn\framework\logic\template
    decorates: 'template'
    arguments:
        - '@path_helper'
        - '@config'
        - '@template_context'
        - '@template.twig.environment'
        - '%core.template.cache_path%'
        - '@user'
        - '@template.twig.extensions.collection'
        - '@ext.manager'

The key line is the decorates: 'template' to tell the extension manager which service to replace.

Step 3: Method Usage

As mentioned above, I am using this implementation in an AJAX controller to return HTML strings for further processing. When standing up my AJAX controller, I hint that this template variable is of my new template class.

For context, my AJAX controller route looks like this in services.yml:

tsn.tsn.controller.ajax:
    class: tsn\tsn\controller\ajax_controller
    arguments:
        - '@auth'
        - '@auth.provider_collection'
        - '@captcha.factory'
        - '@config'
        - '@content.visibility'
        - '@dbal.conn'
        - '@controller.helper'
        - '@language'
        - '@path_helper'
        - '@request'
        - '@template'
        - '@user'

And for now, it looks like this in routing.yml:

tsn_tsn_ajax:
    pattern: /tsn/ajax/{route}
    defaults: { _controller: tsn.tsn.controller.ajax:doIndex, route: "index" }

So the actual controller looks like this (with the extra stuff commented away)…

<?php

// ... use statements

/**
 * Class ajax_controller
 * @package tsn\tsn\controller
 */
class ajax_controller extends AbstractBase
{
    /** @var stdClass the JSON response */
    private $response = null;

    /**
     * ajax_controller constructor.
     *
     * // Other Params
     * @param \tsn\tsn\framework\logic\template $template
     * // Other params
     */
    public function __construct(/* other params... , */ template $template /* , ... other params */)
    {
        parent::__construct(/* other params... , */ $template /* , ... other params */);

        $this->response = new stdClass();
        $this->response->status = 0; // 0: error; 1: success, 2: info/warning
        $this->response->data = []; // whatever is necessary
        $this->response->message = null; // message for 0/2 status
    }

    /**
     * @param $route
     *
     * @return \Symfony\Component\HttpFoundation\Response
     * @uses url::AJAX_MYSPOT_FEED_PAGE
     */
    public function doIndex($route)
    {
        // ... Setup User Access

        $statusCode = Response::HTTP_OK;

        switch ($route) {
            case url::AJAX_MYSPOT_FEED_PAGE:

                // ...do work to set template variables

                $this->response->status = 1;
                $this->response->data['html'] = $this->template->renderPartial(template::P_MYSPOT_FEED);
                break;
            default:
                $statusCode = Response::HTTP_NOT_FOUND;
                break;
        }

        $headers = ['Content-Type' => 'application/json; charset=UTF-8'];

        if (!empty($this->user->data['is_bot'])) {
            $headers['X-PHPBB-IS-BOT'] = 'yes';
        }

        return new Response(json_encode($this->response), $statusCode, $headers);
    }
}

The key line for usage is $this->template->renderPartial(template::P_MYSPOT_FEED); with the constant for the template to be rendered as the only parameter.

Step 4: The Partial

And finally, the partial for this example looks like any other Twig partial:

{#
    This template spools up a page of Topic Cards for the MySpot Feed
    This template requires a `topics` blockvar to iterate over
#}
{% if topics %}
    {% for topic in topics %}
        {% include '@tsn_tsn/partials/tsn_topic_card.html' %}
    {% endfor %}
{% endif %}

Key Takeaway

Avoid duplicating code where necessary.

• Use the existing template engine, and avoid recreating it in either PHP or JS.
• Use Partials to render the same Template everywhere, in as generic a way as possible. Using wrapping elements with modifier classes to affect minor visual changes to the inner elements.
• Add your own PHP classes & traits to handle shared logic between Controllers.
• Use Service Decoration to replace/extend existing phpBB Services with your own classes.

However, I’ve spent the last several days of the holiday break considering all the things I would like to do in my “moonlighting” hours of the evenings, and now that I’ve settled on a few items, I figure I’d write about them before they all slip away.

[[Oracle]] Assistant

For the past several months, I’ve been kicking around the idea of building my own virtual assistant to aid in various work-related tasks that remain unfulfilled by existing digital assistants.

I am heavily invested into the Google Ecosystem at this point (beta Chromebook, beta Google Glass, beta everything Google), but as good as the Google Assistant is, I would like even more specific tasks to be handled with voice commands than are currently available.

I’ve been working with the DialogFlow interface to create a set of voice interactions that will understand what I want to do, and interpret them to the correct variables for executing. I just haven’t pulled the trigger on the Fulfillment part of the setup yet, given that it is necessary to make requests into my work environment to execute the commands at this time.

While this approach is working well for what I’ve done thus far, I decided to go bigger, and hooked up [[Oracle]] as a Discord Bot, where Voice Input is replaced with an @-message in my Discord Server (over on https://guild.the-spot.net, if interested). After initially standing up my own npm-server bot for Discord, I decided to allow all commands to flow into DialogFlow, but I have yet to finish setting up the Fulfillment side, so nothing actually does anything yet.

This is one of the many projects. I’d like to get the bot a presence via my tsn Slack Channel, as well as figure out a way to more tightly integrate it into the phpBB forum architecture for the-spot.net.

The-Spot.net & #tsn9

I recently posted my “decade in review” article, but due to the level of personal details and the audience it would make most sense to, I posted it over in my Journals thread on the-spot.net forums.

While doing so, I encountered all kinds of bugs from the upgrade to phpBB 3.2.8 that were not addressed in the tsn8 templates due to the sheer number of differences between it and the default proSilver theme. Those errors, and the 5-year-old #tsnSpecialReport thread about a new design idea got me in the mood to actually do something about it.

To be fair, every year, around this time, I have nothing but time, and always think about what it would be like to finally get around to updating the site again to fulfill that design. However, this time, I am 5-years smarter than I was when I originally posted it…and I have a pretty good understanding of what it takes to get the pieces in place.

So while #tsn9 is going to be the main project for 2020, it comes with a couple sub-projects…

tsn phpBB Extension

When tsn8 was developed, I tried my best to rely as much as possible on the phpBB Event architecture, but there were just some places that the data I needed was not exposed via event emission, and I had to write custom code into the core files.

The first project, thus, is going to be extracting all custom code into the tsn phpBB Extension, and confining my work to that repository instead of allowing myself to edit “vendor” files.

#tsn9 & Material Design Components

The main driver of the updated design concept was to treat Forums like Directories & their Topics like email threads. Then to cleanup the layout of the old-style forums to be cleaner and more representative of single-threaded communications: basically, emails.

Usually when I redo the website, there is an educational factor behind it all. Last time was tsn8 and I had just learned about front-end frameworks and the responsive design / grid system. I barely knew what I was even looking at when I did it, and was especially surprised to see that phpBB had anything at all for responsive design, since that didn’t exist until then.

This time around, it’s a double-whammy learning curve. I’ve got the responsive stuff, and I’ve got (at least a super basic understanding of) design principles… so this time I’m learning to use the Material Design Component library for Web, and going deeper into whatever it is phpBB wants developers to do for their extensions.

It’ll be far easier than building an “API Theme” as I originally started out doing, with the intent to put a front-end layer on top of that. For this, I’ll focus only on rebuilding the templates that make tsn unique, and then I’ll move to reworking everything else that is left in proSilver.

[[Oracle]] & #tsn9

The last bit of project that I’d like to work into the decade is a deeper integration of the [[Oracle]] Assistant and the #tsn9 forums.

This idea is a little more fluid at this time, since the Assistant is still largely undefined. However, there are a couple ideas that I come across randomly which have broad third-party support or libraries that would make better sense if they were consolidated into the same bot.

Some examples:

• Asking [[Oracle]] if there is anything new to see on the site
• Getting a notification for Private Messages, or mentions
• Sharing outbound from the site…or maybe inbound to the site.
• [[Oracle]] as a pass-through from some chat service to the forums
• Better forum search functionality with questions to [[Oracle]]

I’m open to any other ideas for what forum users would like to do, or even what you may wish some social media platforms would let you do. Leave those in the comments if you have some.

In the end, it likely has to do with something on Google’s end, like:

• Enabling “Letting Less Secure Apps access your account“
• Using the correct Gmail IMAP/SMTP settings
• Or a combination of “Turning on 2-Step Verification“, so that you can then generate and “Sign in using App Passwords“

My approach to solving the issue took into account all of these on the way to the final solution:

• An /etc/hosts entry for smtp.gmail.com

# This is just 1 of 2 IPv4 servers for the domain
173.194.208.108 smtp.gmail.com

Diagnosis

In phpBB’s “Admin Control Panel,” there is an Errors page that would constantly spit out a “timeout” error no matter what settings I used.

Email error
» EMAIL/SMTP
/phorums/memberlist.php

Could not connect to smtp host : 110 : Connection timed out

Errno 2: stream_socket_client(): unable to connect to smtp.gmail.com:465 (Connection timed out) at [ROOT]/includes/functions_messenger.php line 1188

The email settings I have in place are related to a couple things:

• I own several domains, the one for this purpose is: the-spot.net
• I am grandfathered into the original Google Apps Suite, and I have that attached to the domain.
• All email addresses are routed to a catch-all address with an long-random-alphanumeric account name, so when I need to send from a specific account, I have to create a new User in Google Apps Admin Console.
• As the Admin User, I also had to:
  • Enable users to set up 2-Step Verification
  • Enable users to set their own “Less Secure Apps” access setting
• Then as the Newly Created User, I had to:
  • Log into the account for the first time, and create a[nother] password
  • Turn on “Less Secure Apps” access
  • Set up 2-Step Verification (I used my Google Voice number, as I have with all my Google Accounts)
  • Generate an App Password.

Once I had followed all those account-configuration steps, I then referenced the IMAP/SMTP Settings documentation, and started entering the values into the fields, as shown.

Testing

Obviously, the settings in the image above did not work. They continually provided the same “Connection Timeout” error mentioned earlier.

In research and trying to test whether or not my settings were correct, I saw some mention that…

…the “Send a test email” button doesn’t work, and you have to send either an email from the User’s Profile, or a Private Message.

Some random reply on one of the many support threads

A little foreshadowing here… that is not true. The button works fine.

If you do have command-line access, and/or can make changes to the /etc/hosts file that will stick – i.e. you have CLI root access, or you have a cPanel interface to set your own local dns entries, then you’re in luck.

To see what is going on, get to your web server, and run the following command. You’ll see what IP is initially connected to when trying to telnet into the email server at the tls port:

root@web:/home/web# telnet smtp.gmail.com 587
Trying 2607:f8b0:400d:c0a::6d...

It’s an IPv6. That’s unexpected…but I recall Google announcing they are making as many of their services prefer IPv6 where supported. (The link is the best I could find…the original announcement is gone.)

So there’s the problem… the PHP socket stuff is having a hard time with the IPv6 connection.

Now, if you run the following command from the web server’s terminal, you’ll see all the things that are related to the smtp.gmail.com domain:

root@web:/home/web# dig +short smtp.gmail.com
gmail-smtp-msa.l.google.com.
173.194.208.108
173.194.208.109

Two of those are IP addresses. One of them is a domain, and is of little use to us: we already have the smtp.gmail.com domain, and we can’t put a domain into the left-side of the hosts file.

Take either of the IP addresses, and insert them into your /etc/hosts file…

root@web:/home/web# sudo vi /etc/hosts
### 
173.194.208.108 smtp.gmail.com
:wq

Save the file, go back to the phpBB interface and make sure:

• Use STMP server for email: yes
• Your password is an App Password
• Your username, from, reply-to, and other email addresses are the Google Account for the App Password
• The domain is: ssl://smtp.gmail.com (you might be able to use tls://)
• The port is: 465 (you might be able to use 587 for TLS)
• Verify SSL Cert: yes
• Verify SMTP Peer: yes

Then hit the “Send a test email” button in the top section. Once you do, you should get: a fairly quick green success message page, no errors in the log, and an email that arrives to the user account you are currently logged into.

Alternatives

I haven’t tried some of these alternatives, because I was able to get it working via the method above. However, if you can’t make changes to the /etc/hosts file, perhaps try some of the following:

• Create a DNS entry that points a subdomain you can control to the IP (or if possible, both IPs) for the smtp.gmail.com server.
• Create an email account via the webhost/cpanel interface, or maybe even the domain registrar, and go the self-hosted route. At least it’ll be fully under your control, and you can IMAP the account into another Gmail account, and tag those messages via Filters, if you need to receive them.
  • This is the solution I originally used, until I moved my domain over to Google Apps, causing this problem in the first place.
• Or as a last resort, change the code making the request to use a PHP library that will support making calls directly to the original domain. There are many support threads about this problem that end with simply changing the properties sent to the mailer class.

Let me know in the comments if you have found a different solution, or see problems with my approach (other than the obvious and expected “Unexpected IP Changes” or “Single IP, no redundancy”).

A little over a year ago, I wrote about my creation of a service to check Hazardous Weather Outlooks (HWOs) that were posted to the National Weather Service (NWS) website. Everything worked fine, for about a year – until March 31st, 2015.

Oops…they noticed

Alerts suddenly stopped, without warning, just silence. After some investigation and debugging, it appeared to be that my domain was blacklisted from scraping the HWOs off of the NWS website.

When I developed the service, I looked at several APIs and feeds (RSS, XML, ATOM) to get the information I needed, but none of them were for HWOs, and none of them included the Spotter Information Statement. I searched for months before the decision to scrape their website, and without having any luck, I decided they wouldn’t mind – being a government website with lots of visitors – and probably wouldn’t even notice.

I guess they did.

What do I do now?

For months, I tried different aspects of the www.weather.gov website, documentation, Google searches and third-party weather APIs. It wasn’t until June that I was able to finally stumble upon the AERIS weather API.

I used this new API documentation in conjunction with some documents found during my research:

• NWS Count-Public Forecast Zones Correlation file [Website w/ download link]
  • I used the newer link on that page to create the locations database, allowing me to offer county/area name, with state breakdown, and association to Weather Zone and FIPS code.
  • The Zone+FIPS row is given a unique ID in the database (since neither of those values are unique).
• FIPS to County-code [PDF]
  • This did not make it into the database but was valuable in determining what information was available to make an informed decision related to API queries

I have fully recoded the Storm Spotter Status service to make use of the AERIS weather API “advisory” endpoint, with a filter on “outlooks.” Currently, it is setup to use a Developer API, as that is free and the service is used only by a few people. However, there are some throttling limits in place, being a developer account to only allow 10 requests per minute, and only 750,000 requests per day.

That may sound like a lot, but if you do the math for a user base as small as 15 people, you get something like this:

• 1 request x 15 users x 1 hour = 15 requests per hour
• 15 requests x 24 hours = 360 requests per day

If you allow multiple people to set multiple locations as version 1 of the service did, that 360 jumps up pretty quickly. So a couple things were put in place with version 2 to help keep things under control…and free.

Keeping it free

Limiting locations

The main thing to keep things on the free side of the service is to limit every user to only 1 location. Previously, it didn’t matter how many locations you wanted to watch, because there was no cost to gathering the data. Now there is a potential cost.

Limiting everyone to a single location allows for a maximum of 31,250 users selecting a single location. There are less than 4,400 distinct locations in the database

Caching Results

The second thing I had to come up with was how to cache results and make sure I didn’t waste another request to get data I already retrieved. This was the source of many evenings and iterations of code to decide what “supported place” would return the most relevant and useful information.

Originally I did everything by FIPS code because a request would return something like this:

{
 "details": {
 "type": "HWO",
 "name": "HAZARDOUS WEATHER OUTLOOK",
 "loc": "TXZ012",
 "body": "THIS HAZARDOUS WEATHER OUTLOOK IS FOR THE TEXAS AND OKLAHOMA\nPANHANDLES.\n\n.DAY ONE...TONIGHT.\n\nISOLATED THUNDERSTORMS ARE POSSIBLE THROUGH TONIGHT. SEVERE WEATHER\nIS NOT EXPECTED...BUT A FEW STRONGER STORMS COULD PRODUCE WIND GUSTS\nTO AROUND 50 MPH AND SMALL HAIL. \n\n.DAYS TWO THROUGH SEVEN...WEDNESDAY THROUGH MONDAY.\n\nISOLATED TO SCATTERED THUNDERSTORMS ARE POSSIBLE WEDNESDAY THROUGH\nTHURSDAY NIGHT. THUNDERSTORM CHANCES WILL INCREASE FRIDAY INTO THE\nWEEKEND. SEVERE THUNDERSTORMS ARE POSSIBLE ON FRIDAY AND THE\nPOTENTIAL FOR HEAVY RAIN INCREASES FRIDAY THROUGH THE\nWEEKEND...WHICH COULD LEAD TO SOME FLOODING OR FLASH FLOODING.\n\n.SPOTTER INFORMATION STATEMENT...\n\nSPOTTER ACTIVATION IS NOT ANTICIPATED AT THIS TIME.",
 "bodyFull": "FLUS44 KAMA 091926\nHWOAMA\n\nHAZARDOUS WEATHER OUTLOOK\nNATIONAL WEATHER SERVICE AMARILLO TX\n226 PM CDT TUE JUN 9 2015\n\n\n\nOKZ001>003-TXZ001>020-101100-\nCIMARRON-TEXAS-BEAVER-DALLAM-SHERMAN-HANSFORD-OCHILTREE-LIPSCOMB-\nHARTLEY-MOORE-HUTCHINSON-ROBERTS-HEMPHILL-OLDHAM-POTTER-CARSON-\nGRAY-WHEELER-DEAF SMITH-RANDALL-ARMSTRONG-DONLEY-COLLINGSWORTH-\n226 PM CDT TUE JUN 9 2015\n\n\n\nTHIS HAZARDOUS WEATHER OUTLOOK IS FOR THE TEXAS AND OKLAHOMA\nPANHANDLES.\n\n.DAY ONE...TONIGHT.\n\nISOLATED THUNDERSTORMS ARE POSSIBLE THROUGH TONIGHT. SEVERE WEATHER\nIS NOT EXPECTED...BUT A FEW STRONGER STORMS COULD PRODUCE WIND GUSTS\nTO AROUND 50 MPH AND SMALL HAIL. \n\n.DAYS TWO THROUGH SEVEN...WEDNESDAY THROUGH MONDAY.\n\nISOLATED TO SCATTERED THUNDERSTORMS ARE POSSIBLE WEDNESDAY THROUGH\nTHURSDAY NIGHT. THUNDERSTORM CHANCES WILL INCREASE FRIDAY INTO THE\nWEEKEND. SEVERE THUNDERSTORMS ARE POSSIBLE ON FRIDAY AND THE\nPOTENTIAL FOR HEAVY RAIN INCREASES FRIDAY THROUGH THE\nWEEKEND...WHICH COULD LEAD TO SOME FLOODING OR FLASH FLOODING.\n\n.SPOTTER INFORMATION STATEMENT...\n\nSPOTTER ACTIVATION IS NOT ANTICIPATED AT THIS TIME."
 },
 "timestamps": {
 "issued": 1433877960,
 "issuedISO": "2015-06-09T14:26:00-05:00",
 "begins": 1433877960,
 "beginsISO": "2015-06-09T14:26:00-05:00",
 "expires": 1433934000,
 "expiresISO": "2015-06-10T06:00:00-05:00",
 "added": 1433878037,
 "addedISO": "2015-06-09T14:27:17-05:00"
 },
 "poly": "",
 "includes": {
 "counties": [ "OKC007", "OKC025", "OKC139", "TXC011", "TXC065", "TXC087", "TXC111", "TXC117", "TXC129", "TXC179", "TXC195", "TXC205", "TXC211", "TXC233", "TXC295", "TXC341", "TXC357", "TXC359", "TXC375", "TXC381", "TXC393", "TXC421", "TXC483" ],
 "fips": [ "40007", "40025", "40139", "48011", "48065", "48087", "48111", "48117", "48129", "48179", "48195", "48205", "48211", "48233", "48295", "48341", "48357", "48359", "48375", "48381", "48393", "48421", "48483" ],
 "wxzones": [ "OKZ001", "OKZ002", "OKZ003", "TXZ001", "TXZ002", "TXZ003", "TXZ004", "TXZ005", "TXZ006", "TXZ007", "TXZ008", "TXZ009", "TXZ010", "TXZ011", "TXZ012", "TXZ013", "TXZ014", "TXZ015", "TXZ016", "TXZ017", "TXZ018", "TXZ019", "TXZ020" ],
 "zipcodes": [ 73844, 73901, 73931, 73932, 73933, 73937, 73938, 73939, 73942, 73944, 73945, 73946, 73947, 73949, 73950, 73951, 73960, 79001, 79002, 79003, 79005, 79007, 79008, 79010, 79011, 79012, 79013, 79014, 79015, 79016, 79018, 79019, 79022, 79024, 79025, 79029, 79033, 79034, 79036, 79039, 79040, 79044, 79045, 79046, 79051, 79054, 79056, 79057, 79058, 79059, 79061, 79062, 79065, 79066, 79068, 79070, 79077, 79078, 79079, 79080, 79081, 79083, 79084, 79086, 79087, 79091, 79092, 79093, 79094, 79095, 79096, 79097, 79098, 79101, 79102, 79103, 79104, 79105, 79106, 79107, 79108, 79109, 79110, 79111, 79114, 79116, 79117, 79118, 79119, 79120, 79121, 79124, 79159, 79166, 79168, 79172, 79174, 79178, 79185, 79189, 79226, 79230, 79237, 79240, 79251 ]
 },
 "place": {
 "name": "potter",
 "state": "tx",
 "country": "us"
 },
 "profile": {
 "tz": "America/Chicago"
 }
}

There is an array of FIPS codes included in the response, as well as counties, zip codes, and weather zones. But there were also 9 other sets of these pieces of data that had the same content, with the exception of the details.loc value and the place information. This made the association with a given FIPS code nearly impossible without several queries using the data already provided in these two locations…while the rest of the dataset was duplicated in the 9 other datasets returned.

So, a rewrite to use the “wxzone” and “details.loc” values was decided on, as well as a limit of 1 dataset in the response.

Now, all the State Zones (or “wxzones”) included in the response will receive a cache of this report in the database. This means that a single request will cover everyone near the first person to request the area. With this in mind, the request throttling now had to be updated to prevent duplicate requests.

Throttling Requests

With a single request returning information for multiple locations, and a limit on 10 requests every 60 seconds in place, I had to build a loop that would keep track of three three things in a strange way:

• What State Zones have been seen
• How many requests have been made
• Time

The algorithm works like this:

1. Get all the unique locations requested by users (and sort them into various arrays with varying keys…user to location, location to user, location to data, location to preformatted data)
2. Start the loop – while less than 15 minutes have elapsed total, or until we’re done
   1. If the counter is less than 10 and timer is less than 60, continue; else go to 4**
   2. Get the next item (State Zone) in thelocation list
      1. If there are no more items, break out of the loop
   3. Check if thislocationhas been added to the “already checked” list, and continue if not
      1. Request the data from the API
      2. Process the data for storage and alert (create database row and hash spotter information statement for comparison)
      3. Add the State Zone to the “already checked” list
      4. Iterate over the State Zones in the response
      5. If there are any users requesting an included State Zone
         1. Calculate the statement issued time vs user’s last alert issue time and statement hash vs last alert hash (don’t double alert on time or hash)
         2. if time are different, add them to an “Alert List”
      6. Write the advisory data to the database (prior to alerting users)
      7. If there are any users in the “Alert List”
         1. Pull the corresponding Location Row for the user’s location selection and cache it (by db Location ID)
         2. If we can find the Location Row, and the user can receiveDMs (twitter settings)
            1. Construct the message from the Location Row data (county, state, NWS Office [cwa]) and Response data (spotter statement, URL construction)
            2. Attempt to alert the user
            3. Record any errors
   4. ** If we skipped right to here, don’t increment the Request Counter, and continue the loop
      1. Else: Check the time from the start of the loop
         1. If time elapsed is < 60 seconds, increment the counter (keep going until 10)
         2. If time elapsed is > 60, reset the timer and reset the counter to 0 (time ran out, reset it all)
3. Handle any Developer Errors

It took a while to put all that together, and was a departure from how I normally program. My normal work is intended to be efficient and streamlined. This required me to be willing to do things I wouldn’t normally do in a loop so that I could throttle processing time and avoid “dead loops” where nothing is done because 10 iterations had elapsed in very short time.

Conclusions and Future Updates

After writing all of this up, some thoughts occurred while looking at the data. Chief of which is that there are only 4389 specific locations. But the limit of one location per user and 750,000 requests in a 24 hour period means comes out to 31,250 requests every hour can be made before that 750k limit is reached. That means there are enough endpoint requests available to make a call for every specific location every hour.

However, there is the limit of 10 requests per minute. This means 439 batches of 10 requests must be made to get every location. This would take 7.5 hours to complete due to the 10-request throttling. When the caching is taken into account, though, a single request can cover around 23 State Zones (like in the example above), which brings the total time to complete all requests to somewhere around 19 minutes.

With this understanding of just how big or small the dataset is and the requirements to get it in its entirety, everything seems doable while maintaining a free status. However, in an effort to avoid the throttling and possible pricing plan changes – and if the service becomes popular, I will be researching adding a tiered pricing plan to allow multiple locations.

This would take a while to design how a free and paid service would work in parallel, but it’s good to know that it can operate on the free tier as-is for now.

Since it served me so well, I decided I’d try and use it for the Nexus, since I had nothing else on-hand. Unfortunately, the USB connector is facing the opposite direction from the port on the Nexus – meaning my phone would be docked facing backward.

Obviously that will not do. So I tried to reverse it. The TL;DR is: I could not reverse it, the part is a solid mold designed specifically for the direction it is facing. I ended up getting the Nexus Wireless Charger instead. So this blog is about how to disassemble the device, since I could not find one myself.

What you’ll need

You’ll need some tiny screwdrivers – a phillips-head #0 size, to be specific, and some needle-nose pliers. I managed to get by with my trusty Leatherman, but something with a smaller nose-point would be better.

Exposing the screws

This works best when the device is a little warm…like having sat on a Macbook Power brick, or equally radiant device. This will loosen up the glue and soften the rubber. Flip the device and slowly peel the rubber bottom off at a 45-degree angle. This will preserve as much of the glue as possible without tearing the glue or the rubber.

There will be several screw holes on the bottom of the device (more than is necessary, IMO). Unscrew those, and set them aside in a cup or dish so you don’t lose them. Then comes the hard part…

Separating the top & bottom

Starting with a corner of the device, where it has the most give, using only your index and middle fingers on both hands, place them on the corner of the device in the crack between the top/bottom where the pieces are connected and slowly pull them apart. You’re dealing with several small plastic connectors that look like this…

Once you have one corner, the adjacent front-corner will be easier to pry apart. Then move to the back corners, with ever so slightly more force to pry those apart. You’ll now have 3 pieces: rubber, top and bottom.

What’s inside?

Inside, you’ll find a weight (wtf?), a circuit board with a normal USB connection, a 3.5mm connection, and a ribbon USB connection running to the charger USB port.

There are tiny springs holding the USB charger port in place, bringing it back to normal, and those springs sit on tiny hooks, held in with 4 tiny screws. The screws hold down the hook assembly that keeps the rounded-rectangle piece in place.

If you remove the hook assemblies, you can then remove the rounded-rectangle piece with the USB charger port.

From there, your options are limited with the tools I’ve listed. You’ll need a star-like screw to remove the spring-loaded rounded rectangle covering from the piece holding the USB charger port. The circuit board could also be removed, and the whole circuitry could be transplanted into another housing.

Sorry for the bad picture quality. This was an impromptu post and Google Glass has an infinite focal point, and the further from the lens the object is the better the quality. Hope this helps someone looking to take apart their desktop dock. The most important thing for me is knowing what kind of connectors are holding it together, so I don’t break them.

The gist of the concept is to create a “Friends List” containing all of your friends (from whom you’d like to see status updates) and use that list in lieu of your news feed. As mentioned in the video above, there are some drawbacks to this, which I will get into later, but they are not show stoppers.

Now, lets get the thing set up.

Create the list

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2033.png”][/lightbox]

First off, you’ll need to create the list in order to add people to it. To do this:

1. Go to your Profile page
2. Click the Friends tab
3. Mouse over any friend’s name
4. Click the Friends button
5. Click the “Add to another list…” option in the drop down menu

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2038.png”][/lightbox]

In the drop down menu click New List… and you’ll get an input box where you’ll type a new list name.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2040.png”][/lightbox]

Give it a name like “Most Recent” or “Facebook” and hit Enter. What ever name you type it will show as the Page/Browser Tab title (which you’ll see later when I bookmark the page). It will also be what you will look for in your left sidebar on the Facebook homepage.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2042.png”][/lightbox]

Find your new list on the homepage

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2044.png”][/lightbox]

Visit the Facebook homepage and find the new list you just created.

It will be located in the “Friends” section of your left sidebar. Mine was called “Most Recent”.

When you hover the list item, you’ll notice a gear icon next to it on the left, and probably a number on the right.

The number indicates how many new posts have been made by people in the list since the last time you looked at the list. It counts to 20, then simply shows “20+”.

The gear icon on friend lists in this particular “Friends” section is used for 2 things: Add to Favorites, and Delete List.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2046.png”][/lightbox]

Click the gear icon and then click “Add to favorites”.

Once you have done that, your new friends list will have moved to the top section of your left sidebar…

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2049.png”][/lightbox]

Put your mouse over the friends list in its new location and you’ll find another gear icon. Click the gear icon again, and click “Rearrange” from the menu. This will let you move it to a more prominent location in that top section…like right under “News Feed”.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2051.png”][/lightbox]

Click and Drag the friend list to your desired position, and then click “Done” to save it.

Add more friends to the new list

Now that you’ve got the list in a prominent position, click the friend list and view its page.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2053.png”][/lightbox]

Notice on the right-side of the page there is a “List Suggestions” section containing several of your friends with an Add button next to their name and photo. This is where you’ll easily add (by clicking the “Add” button), or skip adding (by clicking the X when you mouse over their suggestion) to the list the people that you actually do or do not want to see updates from in your new timeline.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2057.png”][/lightbox]

Start adding friends…

As you click “Add” or skip adding friends, eventually you’ll run out of suggestions – at least from the initial batch of suggestions. When you do, you’ll either need to click “See More Suggestions” link or just simply refresh the page.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2058.png”][/lightbox]

Continue adding friends to your list until you get the message “There are no more suggestions”.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2059.png”][/lightbox]

If at this point you haven’t added all your friends, try refreshing the page. If that doesn’t get you any more suggestions, but you know you still actually have friends to add, try again later, or type the names of the ones you really want into the “Add friends to this list” input box.

Choosing the updates you see

With this method, one of the pros is that you can select what type of updates you actually see. The possible downside is that this only pertains to your friends you’ve added – this news feed doesn’t include pages you’ve liked, or communities you’re in. (If you’re like me, that’s not a problem. However if you are interested in seeing it, it is available in the “Pages” section of your left sidebar, as “Pages Feed”.)

Just above the List Suggestions, there is a “Manage List” button. Click that and you’ll be presented with a drop down menu of options.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2102.png”][/lightbox]

Click “Choose Update Types…” from that drop down menu and it will expand to show you the various types of status updates you can show/hide.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2103.png”][/lightbox]

Your options are:

• Status Updates: This is why you’re here after all. Leave checked.
• Photos: When your friends post photos. Probably leave checked.
• Games: No one likes Facebook Game updates, wtf. Uncheck.
• Comments and Likes: This is when “Susie commented on Chuck’s post” and you don’t know who Chuck is. It’s also when “Susie likes Spamalama’s page”. Who cares. Uncheck.
• Music and Videos: Susie is listening to Simple Plan on Pandora. Uncheck.
• Other Activity: Susie became friends with Chuck and John and 345 other people. Unless you care, uncheck.

Now that you’ve got the list cleaned up, refresh the page and see what your new News Feed looks like. You’ve probably already noticed some posts from people you haven’t seen in years.

Suddenly, you remember why you blocked them…

Chances are, you have that 1 or 2 (or 2 dozen) people on your friends list that you remember blocking because everything they posted was a survey or Farmville update, and now you regret adding them to your pristine new News Feed list.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2107.png”][/lightbox]

Well, removing them is as easy as adding the first person…

1. Mouse over their name to get the friend card
2. Click the “Friend” button
3. Click “Add to another list…” from the drop down menu
4. Click on your news feed list to remove the check mark.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2109.png”][/lightbox]

Just like that, no more annoying posts from people with annoying views, or that like to overshare everything that they’re doing.

The Caveats

Remembering to Bookmark the list

Since there is no way to default your Facebook feed to use this list, the next best thing is to change out your Facebook bookmark with a link pointed at this new list.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_21141.png”][/lightbox]

Posting from this News Feed

One very important thing to keep in mind is about posting from this “friend list” and other lists like it:

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/2014-07-21_2110.png”][/lightbox]

Anything you post while on this page will ONLY be visible to people you have put into this list. To post updates visible with your normal privacy settings, make them from your profile page or the Facebook homepage.

If you don’t intend for your posts to be limited to only the people you have added to this list (i.e. invisible to new friends you haven’t added to the list, or invisible to people you have removed from the list) then you should continue to post on the homepage, your profile, or through your third-party service/syndication tool of choice.

Most of my posts are syndicated from Google+ to Twitter, and then finally to Facebook. However, there are times when what I want to post is only relevant to people on Facebook, and for those I will have to go to the Facebook Homepage or my profile to make the post.

Adding new friends

Occasionally you’ll probably make new friend-connections with your Facebook profile, and when you do, you’ll have to remember to also add them to this friends list. Often times, they will show up in the List Suggestions section and you can add them from there.

For me, this will be just another list I add when adding new friends, as I always classify friends based on the organization and/or city where I first met them. This helps when I need to direct posts at people where the content is most applicable (i.e. everyone I went to high school with, vs everyone who lived in my high school’s town…there is some overlap in that group, but certain things only pertain to those from school, not their parents and other townsfolk).

If you have any other ideas that might make Facebook easier to use for others, or if I left something out of this article, let me know in the comments below.

I decided I should cut a USB pass-through slot in the cover of my Raspberry Pi case, since the USB draped over the circuit board in a simple arc, and laid down very nicely by the HDMI port and USB ports.

So here’s how I did it…

The Plan

The plan is to cut an arched slot to the left of the HDMI port (since that’s how the Atrix Laptop Dock is configured) for the cable’s entrance, and another arched slot to the right of the USB ports for the cable to exit the case and loop back on itself to plug into the adjacent USB ports.

There are some things you’re going to need for this:

• A Dremel or similar rotary tool for precision grinding/plastic cutting
• A well-lit work space with a mat for plastic shavings to fall
• A circular bit for cutting straight edges into plastic
• A conic bit for smoothing/melting plastic (preferably metal, not sand)
• A clear, or otherwise feature-marked Raspberry Pi Case to see where the internal components/supports are
• A Micro USB Female to Male extension cable (UK, US) from the previous blog post
• Some paper for marking distance measurements
• A fine point marker/pen for writing on plastic & paper
• A straight-edge capable of making accurate 90-degree lines (T-ruler, L-ruler, etc)
• Be willing and able to modify your Raspberry Pi case in a permanent way (for good or bad, as your skills allow)… I can’t be held responsible for any personal mistakes you might make.

The Measurements

First things first! Read this whole section before doing any actual measurements. You will need to get an understanding of what we’re trying to do with the case before you go marking it up needlessly. Then come back and follow the instructions to do the actual measurements.

Unfortunately I don’t have pictures for this (sorry v_v) since I eyeballed it, and wasn’t sure if it would work. So I’ll try to describe it as best I can. What you’ll need is your Pi plugged into the Laptop Dock as an initial frame of reference, like below.
[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/20140706_133746_LLS1.jpg”]

[/lightbox]

To start, you’re going to need a piece of paper/ruler, and your case (with the cover removed from the Pi initially).

The HDMI-side Slot

First we need to measure from the right edge of the Ethernet Jack’s housing (large silver cube on the bottom right of the board above) on the Pi to where the USB extension cable’s overlap begins (just to the left of the HDMI port on the Pi). You’ll take note/make marks where the right edge of the Ethernet housing is, and both sides of the cable’s circuit-board overlap are.

Your measurement paper should have 3 marks now:

• two marks close together on the left (the width of the USB cable’s support cover)
• and one mark further away to the right (the rightmost edge of the Ethernet Jack housing).

Now, remove the Pi from the laptop dock and put the case cover put back on the Pi. Remember this process I’m about to describe – you’re going to need it for the USB-side slot later.

Place your measurement paper over HDMI-port side of the case cover and line up the mark for the Ethernet housing with the right-side of the Ethernet housing. Once you have that lined up, at the other end of your paper, lightly mark the bottom edge of the case to match the marks for your USB extension cable. This is going to be the starting point for the slot cut.

Your Case Cover should have only 2 marks: one for each side of the slot you’re going to cut.

Using a straight-edge, go up from each of those case-edge marks, and trace a line to be perpendicular with the visible circuit board inside the case, and stop at the circuit board.

Using the same measurement paper, place it along that perpendicular line, so that one USB slot mark is at the end of the line, and the other continues up past the circuit board.

Mark the case at the location of the top-most USB slot mark, and extend your perpendicular line to that mark. This will provide enough room for the cable’s diameter to sit comfortably on the circuit board.

Do the same for the other marked side of the slot. You should now have two roughly equal parallel lines that your USB extension cable’s wire should fit nicely between. Then, using a punched hole in a notebook paper, your imagination, or some other similarly sized arc too, connect the two lines with a nice arch.

You should now have 2 lines drawn on your case: one for each side of the slot that will be cut next to the HDMI port. They should be connected with an arch.

What you should end up with is something like the image below. In my build, I only estimated the height, I didn’t measure it, so it extends up a little bit more than necessary.

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/20140706_154601_LLS.jpg”]

[/lightbox]

The USB-side Slot

For this side you’ll need that same measurement paper, but we’re going to use it differently.

Since the need for a Starting/Ending point is less important here, all we really need is the width of the USB cable. If you wanted, you could take new measurements of the width of the actual cable itself (not it’s joint support), but that’s really going to limit the USB cables you can use in the future if you have to replace it.

Take a look at the case cover’s USB port slot for the stacked USB ports. Just to the right of it (above the board’s LED area) there is a “support beam” of sorts. You don’t want to cut into that: it’s extra plastic to have to melt away, and it’s probably there for a reason. Move to the right of the “support beam” about 2-3mm and place your measurement paper’s marks for the USB extension cable slot width.

Align the measurement paper with the bottom edge of the case, and make marks for the slot’s initial cuts.

You should have your HDMI-side slot marked, and now the two starting points for your USB-side marks on the edge of your case.

Now, using the same technique you used for the HDMI-side slot, draw your perpendicular lines up from both marks, up to the circuit board. Place one of your measurement paper’s USB extension cable mark at the end of those lines, and continue the lines up to the second USB extension cable mark. Then arc the lines the same way you did before.

You should have your HDMI-side slot marked, and a matching USB-side slot marked

Again, I estimated my lines, and hand-drew my arcs in the image below, but yours should look similar (aside from the mistake line that got away from me).

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/20140706_154629_LLS.jpg”]

[/lightbox]

Doing the Mod

First things first! Read this section through before doing any actual cutting. You will need to get an understanding of what we’re trying to do with the case before you go cutting it up needlessly. Then come back and follow the instructions to do the actual cuts.

Now for the fun part – the Dremel. For this project, I used two pieces: a circular pad and a conic metal bit.

The circular bit is for cutting the straight edges of your slots. The conic bit is for melting away the plastic in the arch as well as smoothing the interior/exterior edges of the slots. There will be melted plastic on the inside wall of the case cover at the edges of your slots. However, the plastic cools after a few seconds and can easily be chipped away manually using the conic bit in your hand.

But, before we get started…

Warning: Always use Eye Protection. You’ll notice in my photos the blinds were open – and every light was turned on; I was wearing Google Glass with the sunglasses attached to get the videos and images.

Helpful Tip: As with any power puncture tool, make sure you put your bits on straight and tight. Before you put it to the plastic, give the bit a spin in the air and make sure it’s not wobbly or off center. If it is, you run a pretty good risk of your circular bit shattering, or your case getting all scratched up.

The Straight Edges – Circular Bit

Start with the circular pad, on a low speed and approach the bottom edge, between your marks, on your side of choice to get a feel for what it’s going to take to cut through this plastic. Once you are comfortable with the speed of the drill and understand how the plastic will react, turn up the speed a click, and move to the edges and make your straight cuts. Cut along the lines up to the bottom of the arc.

In the images above, I started out on the HDMI side, since it was broader, and gave me a better feel for cutting for the first time. While making the cuts, melted plastic would seep around the edge of the slot to the interior of the case cover, so I used my conic bit in my hand to carefully scrape it off the corners of the cut. Finer smoothing comes after the arch is cut.

The Arches – Conic Bit

Before you put it to the plastic, test the Conic Bit when placed in the drill to make sure the tip doesn’t wobble. If your conic bit is like mine pictured above, it is machined and rough, for smoothing.

The goal here is to get rid of the plastic in the slot. I used my finger to press down on the plastic protrusion I had made with my circular bit until it started to turn white at the base of the arc where it was still connected to the case. Eventually it snapped right off and left only the plastic in the arch.

Then I turned my conic bit on and bumped up the speed one click and set it against that plastic to try and melt it out of the arch, being careful not to get too close to the actual line I drew until I was ready to smooth out the arch.

Once the majority of the plastic is removed from the archway, you should start smoothing down the excess plastic to make the arch cleaner and more defined, eventually melting through your drawn line. During this process, run the conic bit down the straight edges of the slot to remove any stray/excess roughness, and ease the side of the slot into the arch smoothly.

The Cleanup

After you’re done with smoothing the edges, brush the plastic dust off with your hand and run some water over the case cover to get rid of the hard to reach dust. Be sure you dry off the case cover thoroughly before placing back on your Raspberry Pi.

Now, just place your USB Extension Cable over the circuit board and snap the cover back onto the Raspberry Pi, making sure the female end is by the HDMI port and the male end is by the USB ports. Run the male end into the Micro-to-Normal adapter plugged into the Pi’s USB port, and you should get this…

[lightbox link=”http://thepizzy.net/blog/wp-content/uploads/2014/07/20140706_165143.jpg”]

[/lightbox]

Questions? Ideas? Comments?

I haven’t done any case modding in at least 7 years, and this is the first time I’ve used my Dremel since I bought it in as many years ago. So if you have any tips, or thoughts for how this process can be improved, leave them in a comment below.

If you have questions about something I didn’t cover here, let me know, and I’ll do my best to answer them.

I can’t take credit for this, as the project idea and implementation has been around for at least 2 years. However, I did make some changes to the steps I followed, and modified my Raspberry Pi case to make the circuit board a little safer.

The Source Information

Back in March, my friend [wizard] sent me a link to a blog post about how to make a Raspberry Pi Laptop out of the Atrix 4G Laptop Dock (which I had purchased, and ended up being disappointed with its limited functionality).

I ordered the parts from the blog (which I have a modified list below) and waited 2 months for everything to get in from China. When I put it all together, I found there was not enough power from the HDMI cable/coupler/connector chain to activate the laptop dock. After checking out this forum thread, I realized I could ditch the extension cable and coupler, and just go with the micro-to-normal HDMI converter directly into the Pi board.

The [actually used] Parts List

This is taken directly from the blog post mentioned above, with my formatting:

• 1 x Raspberry Pi Revision 2 ( 004 ) ( although I hear a revision 003 may work )
• 1 x Motorola Atrix LapDock (UK , US)
• 1 x Female USB Micro to Male USB A connector  (UK , US)
• 1 x Micro USB Female to Male extension cable  (UK , US)
• 1 x Female Micro HDMI to HDMI Male connector  (UK , US)
• 1 x HDMI coupler  (UK , US)
• 1 x 30cm HDMI cable  (UK , US) 
• 1 x SKPang Raspberry Pi Starter Kit    ( Optional )
• 1 x Self Adhesive Velcro for attaching the SKPang Breadboard area ( Optional )
• 1 x Edimax EW-7811UN Wireless 802.11 b/g/n USb Adaptor ( Optional )  (UK , US)

I didn’t have a need for Velcro or the breadboard components (he never described their purpose, and I didn’t bother investigating it). I also ended up not using the HDMI coupler or the HDMI cable, as mentioned before. I used a TP-Link Wireless adapter instead of the Edimax simply because I already had one. However the TP-Link Wireless adapter comes with its own headache of installing drivers. So go with the Edimax one if you have a choice.

The Build

The build process is pretty straight forward, as he describes in his post. The only change being Step 5.

In Step 5, instead of connecting all the HDMI components together, you’ve only purchased the one HDMI connector.

1. Plug in the Micro USB Female to Male extension cable to the laptop dock’s Micro USB port first.
2. Plug the Female USB Micro to Male USB A connector into the Pi’s USB port, but don’t connect the cable to the adapter yet
3. Plug in the Female Micro HDMI to HDMI Male connector into the laptop dock’s Micro HDMI port next (because of the connector’s width and shaving, it makes it easier to do it in this order).
4. Carefully slide the Pi board onto the HDMI Male connector.
5. Plug the USB cable into the USB connector.
6. Your Raspberry Pi should then start its boot sequence, flickering the ACT light after a few seconds, and the laptop dock’s screen should show your usual boot sequence.

The Caveats of the Raspberry Pi Laptop

One thing I found, by accident, was that closing the lid sends a power surge/signal to the Raspberry Pi board, and it causes it to power cycle. The same thing happens when you open it.

Now, as an owner of the Atrix 4G phone, I can tell you that this signal/surge causes the phone to come out of docked mode (where the screen functions differently on the phone) and that makes the phone go back to normal. Unfortunately, the dock doesn’t know it’s connected to the Pi, and not the phone.

So, don’t close your lid until you’re ready to power down. Once you do close it, disconnect the USB cable from the USB adapter…this saves wear & tear on the USB ports for your Pi and Dock (the adapter is cheap enough to replace).

I’ll have another post in a couple of days describing the modifications I made to my Raspberry Pi case to make the cable management better and circuitry a little more protected.

After a few months of bouncing around ideas, learning a couple new things, and venturing into subjects I have before found difficult to learn, I started asking myself “What is stopping me from learning this? Is there any real that I can’t?”

Well, I can’t come up with any prohibitive reason, only one of application. How can I apply these new things in some way to solve a problem or create something new? How can I tie all this stuff into a single project, that enables me to learn all the things I haven’t before learned? And I came up with the answer…

Earlier this year, I built a LEGO-mount for a USB Webcam that I attached to my telescope.

I was very excited to have built something from nothing, and for it to work as flawlessly as it did. The mount held the webcam perfectly centered over the eyepiece of the telescope, and brought images into my laptop with an image-to-video program, and I created a video of the moon moving across the sky. The only problem was my telescope wasn’t calibrated correctly, so while it tried to track the moon, the moon’s path diverged out of frame, and thwarted my plan to record the blood moon eclipse.

So, I bought a cable to control my telescope from my computer, but I soon lost interest before I got anywhere with that effort.

Prior to the telescope stuff, I had purchased a Raspberry Pi to mess around with, since I got a book about programming one for Christmas. So when March rolled around after messing with the telescope, I decided I’d order some parts to turn the Pi into the CPU of a laptop, using the Motorola Atrix 4G laptop dock.

It took nearly 2 months for all the various parts to arrive from different areas in China, and by then I pretty much lost interest in the Pi, and moved on to the next thing. I did learn to program Python (albeit not fluently) with the help of @snubs and @padresj on twit.tv‘s Coding 101. That should come in handy one day.

The research into what I could do with this Arduino board on the Flexbot Hexcopter and RC/RF devices in general brought me to study radio waves, light waves. After “Nerd Night” with my friend [wizard] about radio waves, light waves, the mathematics of antennas and HAM Radio, it turned out that I understand the physics of radio waves quite well, but not the terminology (2m waves = 140.000+ish MHz frequency, etc) of the stuff, nor electrical engineering and circuitry.

[Wizard] and I went to Ham-Com in Plano, TX this year on my birthday, and there I picked up the Technician’s License book, a Baofeng UV-5R hand radio, a Diamond SRJ77CA antenna, and a new understanding of what HAM radio is all about – especially the future of it: software defined radios and satellites.

The frontiers of space is what interests me the most about the HAM radio hobby, and that is what brings me full circle to the original project: my telescope.

I realized that the common theme causing me to lose interest in the tech projects I’ve played with for the past couple months was my lack of understanding when it comes to electricity. This was abundantly clear when I took some practice exams for the HAM License, and failed each one of them because I didn’t understand circuit diagrams, what circuit board components do, and how circuits interpret binary logic.

As a programmer, I understand binary code, low-level languages, and high level languages. I can interpret how a closed-system handles information to get the desired output from an input, and can turn that process into software without a problem. But if I had to turn that logic into a piece of hardware: forget it.

The Road map

However, if I sit down and learn how circuits work, I could use my GPIO pins in my Raspberry Pi to communicate with my telescope. If I can tell my telescope where to adjust and track, I can affix a small satellite dish to it, and have it follow radio satellites or the International Space Station, or even things like mythical Black Knight satellite. Then I can pick up any radio transmissions, or make transmissions myself, with my HAM Radio.

I have a Software Defined Radio (SDR) coming from Amazon next week, which will allow me to visually see what my radio is picking up. This will make what is going on a little clearer to me, as well as open up some future possibilities I don’t have with the limited radio equipment I own.

I’m not sure what all this will actually end up becoming, but that’s the fun part. The only thing I know is that I’m going to learn something doing it, and eventually I’ll be able to apply it to something else, just as I have applied creating an ETBU Students Yahoo eGroup to creating the-spot.net, and applied the needs of the-spot.net’s online community to learning to program as a full-time job and this project.

And as I have always done – if I find something technical and interesting, you can bet I’ll blog about how to do it yourself here.