I was getting an issue installing brunch today, and this was some of the output

npm http GET https://registry.npmjs.org/node-gyp
npm ERR! Error: SSL Error: CERT_UNTRUSTED
npm ERR! at ClientRequest.<anonymous> (/usr/local/lib/node_modules/npm/node_modules/request/main.js:440:26)
npm ERR! at ClientRequest.g (events.js:185:14)

if you get this try this command

npm config set ca ""

Code: https://www.dropbox.com/s/6yuwnn9rzem28y4/stairs.ino

Arduino Stair Lighting – Final step.

Parts List: http://www.axelscript.com/2014/02/06/arduino-uno-led-stair-lighting/

Video: http://youtu.be/YSPgSEaD4-s

Code: https://www.dropbox.com/s/6yuwnn9rzem28y4/stairs.ino

The goal of this is to use an Arduino UNO to light my stairs when there is motion. This project is to show a concept that I don’t have in place yet, so you’ll only see the concept in the video, after I get everything in place, I’ll make another video to show the finished solution.

This project uses the following:

• Arduino UNO
• Tri Color LED strip
• bread board, and wires (get an arduino starter kit if your just starting, and it will have what you need)
• PIR Motion Sensor

I used a few different mixes of tutorials to throw it together, and if you follow them both individually, you will see the point. You have to mix and match the two to really get it to work though.

I used the following tutorials:

• Radio Shack TriColor LED Lighting Tutorial
  • Support Files
• Using the PIR Motion Sensor
  • Support File

  Video:

  Finished Code:

  Arduino project.

  https://www.dropbox.com/s/6yuwnn9rzem28y4/stairs.ino

Using a potentiometer to turn a servo via an arduino uno, and some code.

Here is the video http://www.youtube.com/watch?v=3MyjufWWzFQ


#include <Servo.h>

http://www.onlamp.com/pub/a/onlamp/2008/03/04/step-by-step-configuring-ssl-under-apache.html

Using Brunch Banapancakes when coming from brunch eggs and bacon.

Brunch is an assembler to assemble html js bootstrap, with backbonejs apps.

There are a number of things brunch does, and one of its skeletons is to use Bootstrap.

The Problem

If you came up old school ( a whole year ago ) Node was on version 0.8.14, and brunch was on version 1.4.0, they dont have great backward compatibility, its almost little to none, so I’ve learned to document the version numbers of any package I use in order to have future use with them.

In order to get brunch’s new skeleton working I had to do a few things.

1) Install nvm – this lets me use different versions of node at the same time. https://github.com/creationix/nvm

2) Now in order to use everything the way I had it before (if you followed a post that is similar to this http://www.axelscript.com/2013/02/06/installing-brunch-io-on-windows/), you have to run this at the command line first nvm run 0.8.14 Please note, you have to install 0.8.14 first nvm install 0.8.14

3)nvm run 0.10

4)you need to see which version of brunch you have locally brunch -v you should see something like 1.4.0 on the console, IF THIS IS VALUE IS LESS THAN 1.7.0, do the following sub steps

4.1) in my system I have the following path “/usr/local/bin” > in that folder there is a symbolic link for brunch, and if you right click on it and get info, you will see it points to “/usr/local/lib/node_modules/brunch/bin/brunch”, in order to keep brunch running the way it was, you need to keep this short cut, IF YOU WANT OR HAVE LEGACY PROJECTS YOU NEED TO MANAGE “brunch”.

4.2) install brunch, by doing this

nvm use 0.10

This will install brunch under the nvm directory for node modules. npm install -g brunch

We need to get the node path for the version of node we are using with nvm which node > this will output something like “/Users/Snowleopard/.nvm/v0.10.18/bin/node” Copy this value your clipboard: note it will be different on your system, so copy what the command line gives you.

now that you know where that version of node is, you can create a sym link to the new version of brunch, so first execute this: cd /usr/local/bin

now you need to create a sym link in there for the “new” version of brunch, that is running on node 0.10.8 for me that was here ln -s /Users/Snowleopard/.nvm/v0.10.18/lib/node_modules/brunch/bin/brunch brunch7

now you should be able to execute brunch7 at the command line and it should give you some output as if the command was recognized, and how to use it.

there is also a mistake in the documentation on how to create a new project, the new version of brunch changed its param order.

build a project is something like this brunch7 new github://Anaphase/brunch-banana-pancakes pancakesTest
launch a local server is: brunch7 watch -s

Query for the facebook api to get posts related to a subject, this will get all posts that have the word “read” in it.

SELECT post_id,actor_id, target_id,message FROM stream WHERE filter_key = ‘others’
AND strpos(message, ‘read’) >= 0

Debug on facebook

After some playing, I found a lot easier way to update my xml column.

The point of this is to update certain parts of my xml, to put a new node into all xml that doesnt have this node.

begin transaction

–replace the previous contents of the NTEXT column with the newly updated XML (converting back to text)
UPDATE my_table
SET form_data.modify(‘insert <my_node_cd_label>{(/some/path/in/xml/my_node_cd/text())} after (/some/path/in/xml/my_node_cd)[1]‘)
WHERE form_number = ‘someCriteria’
AND form_data.exist(‘/some/path/in/xml/my_node_cd_label’) = 0
AND form_data.exist(‘/some/path/in/xml/my_node_cd’) = 1

SELECT form_data
FROM claim_filing
WHERE form_number = ‘ca7′
AND form_data.exist(‘/some/path/in/xml/my_node_cd_label’) = 1

– change rollback to commit to actually run this.
rollback transaction

We have Legacy xml in our SQL server xml that we need to add nodes too. This xml gets published and used by Adobe LiveCycle to produce pdf forms.

We’ve had to modify some of the xml to add new nodes to get new fields to show up, this query will update 1 of the records….

--create XML variable to hold fragment
DECLARE @xml AS XML
--convert the NTEXT to XML
SELECT @xml = CONVERT(XML, form_data)
FROM my_table
WHERE some_field = 'someData'
AND form_data.exist('/some/xquery/path/in/your/xml/yourNewField') = 0
AND form_data.exist('/some/xquery/path/in/your/xml/yourOldField') = 1
AND some_id = 104259

–perform the XML insert
SET @xml.modify(‘insert {(/some/xquery/path/in/your/xml/yourNewField/text())} after (/some/xquery/path/in/your/xml/yourOldField)[1]‘)

–replace the previous contents of the NTEXT column with the newly updated XML (converting back to text)
UPDATE my_table
SET form_data = CONVERT(VARCHAR(MAX), @xml)
WHERE some_field = ‘someData’
AND form_data.exist(‘/some/xquery/path/in/your/xml/yourNewField’) = 0
AND form_data.exist(‘/some/xquery/path/in/your/xml/yourOldField’) = 1
AND some_id = 104259

SELECT ecn, form_data
from my_table
WHERE some_field = ‘someData’
AND some_id = 104259

stay tuned on how to update all records in a table with the new values.

http://www.mssqltips.com/sqlservertip/2738/examples-of-using-xquery-to-update-xml-data-in-sql-server/

http://taytay.com/?p=55

1. Install nodejs http://nodejs.org/download/

(i’m using 0.8.14) http://blog.nodejs.org/2012/10/25/node-v0.8.14/

2. open cmd

3. reference(http://brunch.io/) execute “npm install -g brunch@1.4.0”

4. install git (git is needed for brunch-eggs-and-bacon to pull new projects)

• http://msysgit.github.com/
• http://code.google.com/p/msysgit/downloads/list?q=full+installer+official+git
• use default options next > next > next > finish

5. Open the new program “Git Bash”, and use it as your command line tool for brunch, when open verify “brunch” command is found, and “git”, if both give help on how to use, then your good.

5.1 Install ruby, (check the options that add to your PATH when installing)(which installs gem), gem is needed for the next step
http://rubyinstaller.org/downloads/ – http://rubyforge.org/frs/download.php/76706/rubyinstaller-1.9.3-p374.exe

5.2 Install sass – run “gem install sass” (this will compile your .less or scss files for css)

6. use brunch-eggs-and-bacon to start a new project

• cd /c/www
• To paste, you may have to right click the title bar of the app, and edit > paste
• brunch new brunchTest -s github://nezoomie/brunch-eggs-and-bacon
• this will run a couple lines, then should be done
• cd /c/www/brunchTest
• brunch watch –server (will create a local server to test with)
• brunch watch –o (will optimize and minify code for server when deploying)

Download Eclipse Juno:
http://www.eclipse.org/downloads/download.php?file=/technology/epp/downloads/release/juno/SR1/eclipse-java-juno-SR1-macosx-cocoa-x86_64.tar.gz

after install run this update.

http://download.eclipse.org/eclipse/updates/junoSR1Patch-tmp

Download Subclipse:
http://marketplace.eclipse.org/content/subclipse#.UN3Ta4njl6o

Install TestNG
http://beust.com/eclipse

The other day, I was trying to uninstall a program and so happened to delete my /Library folder on mac, which I then restored from another machine…

after a restart, some craziness started happening, after looking at activity monitor I found that coreaudiod was going crazy, 130% cpu usage.

Here is what i did to fix it.

terminal:

Deleted the Audio folder located at [OSX HDD Name]/Library/Preferences/ and create a new one as root in the terminal:
sudo mkdir /Library/Preferences/Audio

then set permissions:
sudo chown -R _coreaudiod:admin /Library/Preferences/Audio

https://discussions.apple.com/thread/4160772?start=0&tstart=0

I have a webserver running on my mac, and it happens to already be running on port 8080, which is jenkins default port, so i need to change it….

I found a good stackoverflow on this, and want to repost for myself, in case stack overflow ever goes away.

cd /Applications/Jenkins 
sudo vi winstone.properties 

Add httpPort=9999 to the file.

To see all the options you can put in there type java -jar jenkins.war –help

run

java -jar jenkins.war from /Applications/Jenkins. 

Your port will be changed. jenkins.war picks up config options from ./winstone.properties by default.

now go to a browser and try http://localhost:9999

stack overflow

Something like this:

sudo launchctl unload /Library/LaunchDaemons/org.jenkins-ci.plist
sudo rm -rf /Applications/Jenkins “/Library/Application Support/Jenkins” /Library/Documentation/Jenkins
and if you want to get rid of all the jobs and builds:

sudo rm -rf /Users/Shared/Jenkins
and to delete the jenkins user and group (if you chose to use them):

sudo dscl . -delete /Users/jenkins
sudo dscl . -delete /Groups/jenkins

I think that should cover it. Maybe that could be included with the Jenkins installer… But how would people find it? There is no standard place for uninstallers on Mac.

http://www.luscarpa.com/snippets/jenkins/start-stop-jenkins-daemon-on-mac-os-x/

Start Jenkins:

sudo launchctl load /Library/LaunchDaemons/org.jenkins-ci.plist

Stop Jenkins:

sudo launchctl unload /Library/LaunchDaemons/org.jenkins-ci.plist

Sometimes in integration testing you want to use automated testing of variables. In my case Im loading a complex flex application that takes 5 – 10 seconds to finish its loading process, and when its done it does a ExternalInterface.call() out to javascript and sets a public variable on the window.

Whether good or bad practice I don’t care, that is not the point of this post.

In the tests below were testing the url, we have security in place to always see if a deepLink that the user put in, is a real link, or if it requires login or not, and we need to verify that the ones publicly accessible work to our liking… This is nice because when we work on the navigation and security of those links later, it’s fairly complex, and we don’t want to have to worry about that breaking, so now all we have to do is run this test and we’ll know whats broke.

Here is an excerpt of the Selenium WebDriver 2.5 testing code.

BaseTest.java

package tests;

import java.util.concurrent.TimeUnit;

import junit.framework.Assert;

import org.openqa.selenium.By;
import org.openqa.selenium.WebDriver;
import org.openqa.selenium.WebElement;
import org.openqa.selenium.firefox.FirefoxDriver;
import org.openqa.selenium.ie.InternetExplorerDriver;
import org.testng.annotations.BeforeClass;
import org.testng.annotations.BeforeMethod;
import org.testng.annotations.BeforeTest;
import org.testng.annotations.Test;

import com.gargoylesoftware.htmlunit.WebClient;
import org.openqa.selenium.JavascriptExecutor;

public class BaseTest {
   public WebDriver driver;
   public JavascriptExecutor js;
   public String baseUrl;
   public Long defaultTimeout;
   public Long defaultFlexTimeout;
   public StringBuffer verificationErrors = new StringBuffer();
   
   @BeforeMethod 
   public void oneTimeSetUp() {
       driver = new FirefoxDriver();
       js = (JavascriptExecutor) driver;
       baseUrl = "http://someWebSite/";
       defaultTimeout = new Long(5);
       defaultFlexTimeout = new Long(10);
       driver.manage().timeouts().implicitlyWait(defaultTimeout, TimeUnit.SECONDS);
   }

}

Then NavigationPublicTest.java

package tests.navigation;
import java.util.NoSuchElementException;
import java.util.concurrent.TimeUnit;

import junit.framework.Assert;

import org.apache.http.conn.ssl.BrowserCompatHostnameVerifier;
import org.openqa.selenium.By;
import org.openqa.selenium.WebDriver;
import org.openqa.selenium.WebElement;
import org.openqa.selenium.support.ui.ExpectedCondition;
import org.openqa.selenium.support.ui.WebDriverWait;
import org.testng.annotations.Test;

import com.gargoylesoftware.htmlunit.javascript.background.JavaScriptExecutor;

import tests.BaseTest;

//@Test(groups="publicNavigation")
public class NavigationPublicTest extends BaseTest {
	
	@Test(groups="publicNavigation")
	public void testAllDeepLinksAllowAll() {
		deepLinkTest("#Test_One");
		deepLinkTest("#Test_Two");
		deepLinkTest("#Test_Three");
			
		//needs administration login, do later.
		//deepLinkTest("#Administration");
	}
	
	/**
	 * Test deep link of #Locate_Case, this should load the flex app, and not prompt login, and keep #Locate_Case on the url.
	 */
	public void deepLinkTest(String deepLink) {
		driver.get(this.baseUrl + deepLink);
		
		WebDriverWait wait = new WebDriverWait(driver, defaultFlexTimeout);
		
		// Wait for redirects to happen, look for text on page in order to move on.
		wait.until(new ExpectedCondition() {
			public Boolean apply(WebDriver webDriver) {
				System.out.println("waiting for flex app to load ...");
				
				Object isApplicationLoaded = js.executeScript("return window.isApplicationLoaded");
				
				return isApplicationLoaded != null;
			}
		});
		
		String url = driver.getCurrentUrl();
		String destinationUrl = this.baseUrl + deepLink;
		
		// wait 
		if( url.toString().equals( destinationUrl.toString() ) ) {
			System.out.println("The url is correct " + url);
		} else { 
			Assert.fail("The url is not correct, it should be: " + destinationUrl + " but is: " + url);
		}
		
	}
}

Create a symbolic link on mac to sublime text 2, so you can use it at command prompt without editing your bash file…

ln -s /Applications/Sublime\ Text\ 2.app/Contents/SharedSupport/bin/subl /usr/local/bin/subl

now at terminal you can open folders right up.

subl /www/myproject

Automate your testing with Selenium. Set it up in eclipse by following the following steps.

NOTE: This is for setting up selenium 2.5 Java.

1) Download Selenium HERE (or http://seleniumhq.org/download/, and search for Java)

2) start eclipse, and install testNG, testNG is like a beefed version of JUnit, and can be ran via ant ( so can JUnit, but I think testNG comes with some additional things that junit doesnt ) See this comparison

2.1) Install TestNG, in eclipse (on mac) go to Help > Install New Software > use the address: http://beust.com/eclipse

2.2) Select TestNG and complete the install of the plugin and restart.

3) after restart, select File > new > Java Project.

3.1) create a libs folder in the project

3.2) extract the selenium zip file jars into the libs folder (even the ones that are in the libs folder inside the extracted folder, you can just put them straight into your new libs folder)

3.3) right click your project > properties > java build path > add jars > select the libs folder in your project and highlight all the jars and add them, click ok and be done with that.

4) After project is created > Right click “src” create a folder called “tests” (this name is not important, its just a folder name)

5) right click on the new “tests” folder and click New > Other > When window comes up filter and search for “TestNG” > select “TestNG class” > and create the class, When the next window comes up choose the Source folder by clicking “browse” > select your project > then its src folder. Click browse next to the package, and select the folder you just created called “tests”.

6) when the class comes up, i usually rename the function to “main” instead of “f”.

7) paste code

package tests;

import org.openqa.selenium.By;
import org.openqa.selenium.WebDriver;
import org.openqa.selenium.WebElement;
import org.openqa.selenium.firefox.FirefoxDriver;
import org.openqa.selenium.support.ui.ExpectedCondition;
import org.openqa.selenium.support.ui.WebDriverWait;
import org.testng.annotations.Test;

public class NewTest {
  @Test
  public void main() {

		WebDriver driver = new FirefoxDriver();
		driver.get("http://www.google.com/");
		WebElement searchBox = driver.findElement(By.id("gbqfq"));
		
		if( searchBox != null ) {
			System.out.println("found search box...");
			searchBox.sendKeys("speedtest.net");
			
			//wait for google to asynchronously return.
			

			WebDriverWait wait = new WebDriverWait(driver, 3);
			
			// Wait for search to complete
		      wait.until(new ExpectedCondition() {
		          public Boolean apply(WebDriver webDriver) {
		              System.out.println("finding speedtest ...");
		              return webDriver.findElement(By.partialLinkText("Speedtest.net - The Global Broadband Speed Test")) != null;
		          }
		      });
		      
		      
		}
		

		
  }
}

8) right click project > TestNG > Convert to TestNG

9) right click project > run as TestNG Test

this should pass and launch firefox and type into the search and verify the link shows up.

Sometimes when you want to run reports on users, maybe that have sales orders associated to them, you maybe just want to put a total into the row.

Of course you could do this by looping and run a query for each user, but thats pretty inefficient.

This is a MySQL stored function that would do something like that for you.

DELIMITER $$
DROP FUNCTION IF EXISTS `getColumnTotalByDriver` $$

CREATE DEFINER=`root`@`localhost` FUNCTION `getColumnTotalByDriver`(`driver_id` INT, `columnName` VARCHAR(64), `start_date` DATETIME, `end_date` DATETIME) RETURNS FLOAT(18,2) DETERMINISTIC CONTAINS SQL SQL SECURITY DEFINER BEGIN
DECLARE total FLOAT(18,2);
SELECT SUM(receipt_amount)
from orders
WHERE accepted_by = driver_id
AND charge_amount > 0
AND date_delivered >= start_date 
AND date_delivered <= end_date 
INTO total;
RETURN total;
END $$

DELIMITER ;

To use that function you would do something like this.

SELECT * , getColumnTotalByDriver( u.user_id, '2012-11-08 00:00:00', '2012-12-08 23:59:59' ) AS total
FROM user u
WHERE user_level =10

Thanks to this post for getting me started.