JRSETI

HTML/CSS class versus id

2011-11-04T10:43:00.000-07:00

CSS Alabama (1862 - 1864)

HTML/CSS "class" versus id. I always get confused. I recently found a great explanation at StackOverflow. I fully attribute this to StackOverflow and I am just repeating the answer here so I can always remember where to get it.

Think of University.

<student id="JonathanSampson" class="Biology" />
<student id="MarySmith" class="Biology" />

Student ID cards are distinct. No two students on campus will have the same student ID card. However, many students can and will share at least one Class with each other.

It's okay to put multiple students under one Class title, Biology 101. But it's never acceptable to put multiple students under one student ID.

When giving Rules over the school intercom system, you can give Rules to a Class:

"Would the Biology class please wear Red Shirts tomorrow?"

.BiologyClass {
  shirt-color:red;
}

Or you can give rules to a Specific Student, by calling his unique ID:

"Would Jonathan Sampson please wear a Green Shirt tomorrow?"

#JonathanSampson {
  shirt-color:green;
}

One interesting comment to note - this can cause problems sometimes, so watch out:

Can be confusing. You can't have on the same page <student id="JonathanSampson" class="Physics" /> – Luc M Aug 15 at 16:36

ATA Un-hibernation - Report 3

2011-10-10T03:32:00.000-07:00

We are continuing the re-opening effort of the ATA. No exact date is set yet, but it should be several weeks at the most. Don't worry - we will make a big announcement on the front page of setiQuest.org when a date is set.

On September 29th, 2011, last week, I drove a van filled with equipment from the SETI headquarters in Mountain View, CA, to the ATA. It took me about 6 hours, I didn't speed on the highway (much).

When I arrived I was met by the sign in the picture informing the public to please be patient and not disturb the staff. So things are really busy there at the moment.We are in the process of continuing antenna maintenance, reinstalling computers that were removed and taken to Mountain View when the hibernation, and generally ramping things up again.

The work is going well and we are all pleased with the progress.I arrived late on Wednesday night and slept in the dorm. Early in the morning I drove the 1/2 mile to the main lab to unload the computers.

The ATA just as the sun was rising on Sept 29, 2011

All the antennas were standing at attention pointed south towards Mount Lassen.

I zoomed in on one of the antennas and saw
there was a row of birds on top. Cute!
I tried to get closer, but that spooked the birds.

I unloaded the van with the computers. This was 6 Dell servers and each was heavy. We rented a van for this because we wanted the servers to have a smooth ride to avoid dislodging anything.

Nice van. It has satellite radio, so the ride seemed short.

We strapped the servers into the back seat.

The glow of the sever rack in the signal processing room.
The sun was not quite up yet, so the signal processing room was dark.

On this trip I was just a delivery boy. I unloaded the computers and drove back down to Mountain View.

Plants at the Allen Telescope Array

2011-09-16T11:11:00.000-07:00

Last week while at the ATA I learned that there is an endangered plant that we need to pay attention to. It is the Baker's globe-mallow, aka Iliamna bakeri. My picture of it is to the left. See the Wikipedia article about this plant:

It is native to northern California and southern Oregon, where it grows in mountain forests and woodland on volcanic soils.

"woodland on volcanic soils" exactly describes the ground the ATA is built upon.

Here is a picture someone took of the same plants one month ago as they were starting to dry out:

This pic taken in August 2011 at the ATA - Baker's globe-mallow

I have seen this plant around, but I never knew what it was until someone pointed it out this week. To be honest, it looks like a weed, one of the first I'l pull up or mow over if I didn't know any better. Here is a picture from Wikipedia (this picture is in the public domain):

Picture from Wikipedia...

When the ATA was built, several of these plants were marked with wooden stakes in the ground so we would not accidentally disturb it, either by weeding, mowing, or building an antenna on top of it. Soon the marked plants died, but more grew up in other spots. As you can see from the picture I took, the plant is dying for the winter and it is ready to drop seeds.

Another interesting plant, I assume not endangered, has been growing this year in abundance because the ATA grounds have not been maintained and mowed as usual. These plants were given the chance to flourish:

What is this plant?

Maybe someone can tell me what this plant is? It is about 6 feet (2 meters) high. Some have gorgeous deformed tops that are about the size of oven mitts. I assume this signifies the sex of the plant?

Update: It is Mullein.

See the tops of the stalks?

Holding hands?

Creating a database in Ruby-On-Rails

2011-09-15T15:18:00.000-07:00

Actually, I should name this post "Creating a database in Ruby-on-Rails - That works!"

Today I had to create a MySQL database on a new server. And make it available to my Ruby-on-Rails application. Here are the basic steps I followed.

Note; I am using Ubuntu, rails 3.0.6, and I had to start the mysql server:

sudo /etc/init.d/mysqld start

I created a config/database.yml file with the following contents:

production:
   adapter: mysql2
   encoding: utf8
   database: my_dbase
   pool: 5
   username: tom
   password: x1234
   socket: /var/run/mysqld/mysqld.sock

Note: The 'socket' definition must be the same one defined in /etc/my.cnf.

Then I ran:

rvmsudo rake db:migrate

This gave me an error:

** Invoke db:migrate (first_time)
** Invoke environment (first_time)
** Execute environment
** Execute db:migrate
rake aborted!
Please install the mysql2 adapter: `gem install activerecord-mysql2-adapter` (no such file to load -- active_record/connection_adapters/mysql2_adapter)

Doing what it told me to do gave me an error:

rvmsudo gem install activerecord-mysql2-adapter
ERROR:  Could not find a valid gem 'activerecord-mysql2-adapter' (>= 0) in any repository

So I had to get clever.

In my Gemfile I added the gem definition for the mysql adapter:

Added this to Gemfile:
  gem 'mysql2', '~> 0.2.7'

Then I updated with bundle:

rvmsudo bundle update

That fixed that. Now running rake gave me a new error:

rvmsudo rake db:migrate --trace
** Invoke db:migrate (first_time)
** Invoke environment (first_time)
** Execute environment
** Execute db:migrate
rake aborted!
Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2)

This error was because I did not yet set up the mysql database. I logged into the database

sudo mysql root

Then I created the database and set the privileges:

mysql> create database my_dbase;
mysql> grant usage on *.* to tom@localhost identified by 'x1234';
mysql> grant all privileges on my_dbase.* to tom@localhost ;

Finally, I re-ran rake, and the database created and was accessible to my Ruby-on-Rails location:

rvmsudo rake db:migrate --trace
** Invoke db:migrate (first_time)
** Invoke environment (first_time)
** Execute environment
** Execute db:migrate
** Invoke db:schema:dump (first_time)
** Invoke environment 
** Execute db:schema:dump

Web Scraping With Ruby

2011-09-14T10:33:00.000-07:00

Have you ever wanted to "scrape" information from a website and use the information in your own program? That is what "web scraping" is for. I recently wrote a program to do just that using the programming language Ruby, and a software package called Nokogiri.

Nokogiri is an HTML, XML, SAX, & Reader parser. You can "suck in" an HTML page from the internet and easily parse through all the data, picking out the information you need. Nokogiri is a free open source package.

I initially tied to use a package called scrubyt (great name). But I could not get scrubyt to work on Mac OSX, Fedora, SUSE, or Debian! Scrubyt seems to not be maintained and there was too much incompatibility with other packages. So, I gave up and tried Nokogiri.

Nokogiri worked immediately after I followed the simple install instructions on my Mac OSX. I was immediately off and running.

For my first try at web scraping with Nokogiri I tried something pretty simple. This program obtains the schedule information for NFL football games. The source code for my simple exercise is included below.

The site I picked to get the schedules from seemed to be fairly complex, as far as HTML goes. I brought up the schedule page in my browser, then selected "View Source" to see all the HTML. By looking at the HTML source I experimented with the program till I got it parsing just right.

The syntax for Nokogiri code in Ruby and get a bit complex, but that us understandable. After a bit of experimentation I got it.

I ask that anyone reviewing this code to please critique it. Is there a better or more concise way to parse through the data than the way I did it? Please comment about this if you have any suggestions.



#!/usr/bin/env ruby

#######################################################################
#
# Read in NLF schedules
# Simple example of how to scrape a page using ruby.
# This program uses Nokogiri to retrieve or parse a website for NFL
# schedules, then prints out the information. This is meant to be
# used as an example of how to use Nokogiri.
# See http://nokogiri.org/ for information baout Nokogiri and how to
# install/use it.
#
# Author: Jon Richards, jrseti.blogspot.com
#
# You may use this code for anything you like. No restrictions, just
# note I am not liable for any problems caused by this code.
#
#######################################################################

require 'rubygems'
require 'nokogiri'
require 'open-uri'

# Print the welcome
puts
puts "***************************************";
puts "* Welcome to the NFL Schedule Reader! *";
puts "***************************************";

# Query the user what week number
print  "\nEnter the week number (1 to 17): "
weekToPrint = gets.chomp;

# Check for a valid week number
if(weekToPrint.to_i <1 || weekToPrint.to_i > 17)

  puts "Invalid week number - should be a number 1 to 17\n"  Process.exit;

end
# Get the html data
# Catch the exception if there is no internet access, or nfl.com is down
begin
  html = Nokogiri::HTML(open("http://www.nfl.com/schedules")) ;
rescue
  puts "Error: could not access http://www.nfl.com/schedules.";
  puts "Are you connected to the internet?\n\n";
  Process.exit;
end

# We will be counting the weeks, so initialize to 0
week = 0;


# Get the tables with class="data-table1". These are the table that contain
# the schedule data. There should be 17 of these tables, one for each
# NFL game week.
tables = html.xpath('//table[@class  = "data-table1" ]');

# Print out the week number
puts "\nWeek " + weekToPrint.to_s + "\n\n";

# For each table retrieved, loop trough. For te week selected by the user,
# parse the data and report.
tables.each do |table|

  week = week + 1;
  if(week.to_i == weekToPrint.to_i)

    # Get the table rows

    rows = table.css('tr');

    # Loop through each row
    rows.each  do |row|

        #Get the text of the away and home teams
        linkTexts = row.css('a/text()'); # Ger the text enclosed by the link

        #there must be > 2 links for a valid game line
        if(linkTexts.count > 3)

          # Parse out the data
          away = linkTexts[0].to_s;
          home = linkTexts[1].to_s;

          # Get the Game time, if not played yet, or the text Final, if game finished
          tdTexts = row.css('td/text()');
          isFinal = linkTexts[2].to_s.strip;
          gameTime = tdTexts[3].to_s.strip;

          if(isFinal == "Final")

              # Get the scores
              awayScore = tdTexts[1].to_s.strip.gsub('\t', '').gsub(' ', '').gsub('@', '').to_i;
              homeScore = tdTexts[2].to_s.strip.gsub('\t', '').gsub(' ', '').to_i;

              # Print out
              puts away.to_s + "(" + awayScore.to_s + ") @ " + home.to_s + "(" + homeScore.to_s + ")";
          else
             #Get the channels and stations for broadcast
            network = tdTexts[5].to_s.gsub(' ', '');
            directTVChannel = tdTexts[6].to_s.gsub(' ', '').to_s.strip.gsub('\t', '').gsub(' ', '');
            sirius = tdTexts[7].to_s.gsub(' ', '').strip.gsub('\t', '').gsub(' ', '').gsub('
', '') + ", " +  tdTexts[8].to_s.gsub(' ', '').strip.gsub('\t', '').gsub(' ', '');


            # Print out
            puts away.to_s + "@" + home.to_s + " Game Time: " + gameTime.to_s + ", Network: " + network.to_s + ", DirectTV: " + directTVChannel.to_s + ", Sirius: " + sirius.to_s;
          end

        end
      end
    end

end

Sample output:

For Week 1, games already played:


***************************************
* Welcome to the NFL Schedule Reader! *
***************************************

Enter the week number (1 to 17): 1

Week 1

NO(34) @ GB(42)
CIN(27) @ CLE(17)
ATL(12) @ CHI(30)
DET(27) @ TB(20)
PHI(31) @ STL(13)
IND(7) @ HOU(34)
TEN(14) @ JAC(16)
BUF(41) @ KC(7)
PIT(7) @ BAL(35)
NYG(14) @ WAS(28)
MIN(17) @ SD(24)
CAR(21) @ ARI(28)
SEA(17) @ SF(33)
DAL(24) @ NYJ(27)
NE(38) @ MIA(24)
OAK(23) @ DEN(20)

For Week 2 (games not played yet, as of this writing)


***************************************
* Welcome to the NFL Schedule Reader! *
***************************************

Enter the week number (1 to 17): 2

Week 2

SEA@PIT Game Time: 1:00 PM, Network: FOX, DirectTV: 713, Sirius: 139(SEA), 93(PIT)
OAK@BUF Game Time: 1:00 PM, Network: CBS, DirectTV: 705, Sirius: 148(OAK), 104(BUF)
ARI@WAS Game Time: 1:00 PM, Network: FOX, DirectTV: 709, Sirius: 91(ARI), 112(WAS)
TB@MIN Game Time: 1:00 PM, Network: FOX, DirectTV: 710, Sirius: 104(TB), 113(MIN)
JAC@NYJ Game Time: 1:00 PM, Network: CBS, DirectTV: 707, Sirius: 136(JAC), 85(NYJ)
CHI@NO Game Time: 1:00 PM, Network: FOX, DirectTV: 712, Sirius: 128(CHI), 86(NO)
GB@CAR Game Time: 1:00 PM, Network: FOX, DirectTV: 711, Sirius: 94(GB), 117(CAR)
BAL@TEN Game Time: 1:00 PM, Network: CBS, DirectTV: 706, Sirius: 134(BAL), 149(TEN)
KC@DET Game Time: 1:00 PM, Network: CBS, DirectTV: 704, Sirius: 137(KC), 135(DET)
CLE@IND Game Time: 1:00 PM, Network: CBS, DirectTV: 708, Sirius: 106(CLE), 92(IND)
DAL@SF Game Time: 4:05 PM, Network: FOX, DirectTV: 714, Sirius: 92(DAL), 139(SF)
CIN@DEN Game Time: 4:15 PM, Network: CBS, DirectTV: 715, Sirius: 91(CIN), 94(DEN)
SD@NE Game Time: 4:15 PM, Network: CBS, DirectTV: 717, Sirius: 93(SD), 86(NE)
HOU@MIA Game Time: 4:15 PM, Network: CBS, DirectTV: 716, Sirius: 112(HOU), 85(MIA)
PHI@ATL Game Time: 8:20 PM, Network: NBC, DirectTV: , Sirius: 92(PHI), 93(ATL)
STL@NYG Game Time: 8:30 PM, Network: ESPN, DirectTV: , Sirius: 92(STL), 93(NYG)

Installing MySQL instead of MariaDB

2011-09-12T16:17:00.000-07:00

Today I got a request to "get MySQL working on the new server." This was on a server with a fresh install of OpenSUSE 11.4. It appeared that MySQL was installed, but it turned out to be MariaDB.

What is MariaDB? It is a drop-in replacement for MySQL, supported by Michael "Monty" Widenius, the original author of MySQL. I can only think that the MySQL open-source project was branched moments before Oracle took it over, and this is the continuation of that branch. Am I correct about this? Let me know if not.

I just want the standard MySQL client and server installed. So, I have to uninstall MariaDB and install MySQL. this is how I did it:

zypper remove mariadb
zypper remove mariadb-client
zypper install mysql-community-server

If zypper does not work (it did not for me!), see my post about how I got it working:
http://jrseti.blogspot.com/2011/09/zypper-does-not-work-on-opensuse-fixed.html

Zypper does not work on OpenSUSE (Fixed)

2011-09-12T16:04:00.000-07:00

Today I got a request to "get MySQL working on the new server." This was on a server with a fresh install of OpenSUSE 11.4.

The standard install, at least I assume this was standard, comes with MariaDB masquerading as MySQL. So first, I had to uninstall MariaDB. I executed the following command:

zypper remove mariadb

Zypper returned the error:

zypper: symbol lookup error: /usr/local/lib/libreadline.so.6: undefined symbol: PC

Needless to say, this is irritating. Zypper, the number one program you use to configure your server, does not work! It turns out that the library /usr/local/lib/libreadline.so.6 is not any good. It must be corrupted. Lucily there is a good one in /lib64, so I amended the value of the LD_LIBRARY_PATH to point to the correct one:

export LD_LIBRARY_PATH=/lib64:$LD_LIBRARY_PATH

Now, zypper works as expected! Fixed!

ATA Un-Hibernation - Day 2

2011-09-08T16:23:00.000-07:00

September 08, 2011
The picture to the left is of Mount Lassen. It still has snow!

Today was the second day I was at the ATA to help with the ATA un-hibernation. Today we concentrated on antenna maintenance. Every 6 months each antenna is scheduled to be inspected, greased, and repaired if necessary.

Because the ATA has been in hibernation since April 15, 2011, none of the antennas have has their maintenance on time. We will begin getting the maintenance up to date in the following weeks.

The maintenance process requires the antenna dish to be moved to several locations. To facilitate this in the field (literally "in the field") we open up the node house that provides all the network and air cooling to the dish antenna.

One of the 5 node houses

Inside the node house we open up the network router cabinet and plug in the laptop. With the laptop connected we can TCP into the antenna and control the antenna dish position during maintenance:

We position the antenna so that the alidade, the bulge at the top of the mast which contains the computers and gears, is shaded from the sun.

It is at this time my camera batteries died! Someone else has picture and I'll post them when I get them.

We performed the following actions:

Remove alidade cover. Inspect cable wear from dish rotation
Inspect the AZ drive belt and oil
Inspect various items for wear
Clean anything too dirty
Apply oil to various locations
Remove the cover on the elevation drive shaft. Clean out old oil and dirt. Re-oil and replace cover
Inspect inside the radome. Inspect all supports, radome cover for tears
Move the antenna focus in and out. Does it work?

Finally, rotate the antenna dish 180 degrees. As it is moving, hold your ear to one of the mast support poles. There should not be any strange grinding or knocking noises.

Weird ssh problem

2011-09-08T10:43:00.000-07:00

Today I came across a weird one (aren't they all?).

On a computer running OpenSUSE 11.3 Linux OS.

I could "ping github.com"

But! "ssh github.com" would not work! "ssh: Could not resolve hostname" was the error. Why would ping work and not ssh?

Through some fiddling around I determined that ssh only queries the first entry in the DNS file /etc/resolv.conf!!! The first entry in my /etc/resolv.conf was a local DNS for internal purposes.

I solved the problem by moving the first entry to the end of the list in /etc/resolv.conf. Now ssh can find the address of github.com.

ATA Hibernation

2011-09-07T17:20:00.000-07:00

The Allen Telescope Array - Mount Shasta still has snow!

September 07, 2011

As you may know the Allen Telescope Array (the ATA) went into hibernation on April 15, 2011 due to lack of funds. A sad day indeed.

But, thanks to the kind donations from supporters of SETIStars and the space surveillance desires of the United States Air Force Space Command (AFSPC) , the ATA is now in the process of exiting its hibernation. We started the process just this week.

I arrived this week with my colleagues Gerry Harp and John Ross, both from The SETI Institute to begin the work.

I saw a mower amongst the dishes

When I arrived this morning I first saw something that I haven't seen in almost 5 months; the grass being mowed! I saw the mower snaking around the far antennas. Since the ATA went into hibernation the grass has not been mowed. Actually, it is not grass, it is weeds. The stickery kind that stick to your socks and shoes and take forever to pick out. I'm glad to see the mower.

We need to "bring the ATA out of hibernation" because the ATA is a complex and sophisticated instrument that is not

The mower - or rather a tractor - doing what needs to be done

meant to be in hibernation. It needs to be constantly tuned, or things stop working. 5 months is a long time for such a place to go without maintenance.

Each antenna needs to be checked. Gears need to be greased. Each antenna feed contains a dewar which cools the internal low noise amplifier down to 62 degrees Kelvin. That is -348 degrees Fahrenheit, or -211 Celsius. Occasionally the dewar leaks or fails, and and the feed has to be removed and sent back to the factory for repair. Today 4 antennas had their feeds removed and they should be back from the shop sometime in October.

I observed the removing of some of the antenna feeds. This is how it went:

Power was removed from the feed by unplugging the RIM Box

The case structure was mounted around the feed

The case enclosed the feed. Bolts are then removed.

The feed - with PAX box - pulled out and ready for the factory

And the rest of the day was meetings! Tomorrow we will be performing more maintenance and checking out the computer systems.