rails app_name --database=db_name

I’ve created a script written in Ruby to do the initial import. Now the project just needs to be checked out and the tmp & log files need to be ignored. See below.

Create the repository

svnadmin create path_to_repository

Run this script:

# == Description
# Creates a new Rails project and imports it into a Subversion repository.
# Both Rails and Subversion must be in the command line path.
#
# == Usage
# ruby create-rails-project.rb name_of_project path_to_repository
#
# name_of_project: A name for the project you wisth to create.
# Cannot contain spaces.
#
# path_to_repository: A path or URL to an existing Subversion repository.
#
# == Author
# Bob Nadler, Jr.
# Last Updated: 2007-02-20

require ‘rdoc/usage’

RDoc::usage unless ARGV.length == 2

#Assign arguments to variables
name_of_project = ARGV[0]
path_to_repository = ARGV[1]
temp_folder = “tmp-#{name_of_project}-tmp”

puts ‘Creating temporary folders…’
`mkdir #{temp_folder}`

#Create repository folder structure
`mkdir #{temp_folder}/#{name_of_project}`
`mkdir #{temp_folder}/#{name_of_project}/trunk`
`mkdir #{temp_folder}/#{name_of_project}/tags`
`mkdir #{temp_folder}/#{name_of_project}/branches`

puts ‘Creating rails app…’
`rails #{temp_folder}/#{name_of_project}_temp`

#Remove files that should not be in repository
`rm -r #{temp_folder}/#{name_of_project}_temp/log`
`rm -r #{temp_folder}/#{name_of_project}_temp/tmp`

#Copy rails app into trunk
`cp -r #{temp_folder}/#{name_of_project}_temp/* #{temp_folder}/#{name_of_project}/trunk`
`rm -r #{temp_folder}/#{name_of_project}_temp`

puts ‘Importing project into repository…’
`svn import #{temp_folder} #{path_to_repository} -m “Initial import”`

puts ‘Deleting temporary files…’
`rm -r #{temp_folder}`

puts
puts “Script Complete. Project successfully created in: #{path_to_repository}/#{name_of_project}”

Create folder structure of repository

svn mkdir file:///path_to_repository/trunk file:///path_to_repository/branches file:///path_to_repository/tags -m "Creating repository folder structure"

Create the rails application

rails name_of_project

Import rails application

svn import . file:///path_to_repository/trunk -m "Initial import"

Delete the rails application folder

rmdir /s name_of_project

Checkout the application

svn checkout file:///path_to_repository/trunk name_of_project

Ignore log files

svn propset svn:ignore "*.log" log/
svn ci -m "Ignoring all log files"
svn up

Ignore all tmp files

svn propset svn:ignore "*" tmp/
svn ci -m "Ignoring all files in tmp folder"
svn up

Museum of Natural History

Caitlin, Kristen, Stephen and I took a trip to the Museum of Natural History today in NYC. Look for photos in the near future.

Lego my Nintendo

I knew I should have saved my old Legos for something.

I created a new VB.NET Windows Project with a single form. On it I added two file folder dialog boxes that are used to specify the folder where the pictures are located and another for the target folder where the results will be saved. I then searched the source folder for files with a “.jpg” extension, made copies of each one and made a thumbnail of each in the target folder. Then I created an HTML file with a table, and looped through the thumbnail folder adding the thumbnail image and hyperlink.

It took me about half an hour to write the code and it does precisely what I need it to do. I’ve includes a link to the project file (I wrote it in VB.NET) below. You can change it to suite your needs. Some ideas for enhancements might be: supporting other image format besides JPEG’s, the ability to pick and choose images from the source folder, and an input to specify the thumbnail size (I hard-coded it to 164×164.)

Here is a link to the project source code. (Zip File 33KB)

The connection string should be set as follows:

• Provider=Microsoft.Jet.OLEDB.4.0
• Data Source=InsertWorkbookNameHere;
• Extended Properties=”Excel 8.0;HDR=YES;”

Under Extended Properties, if your spreadsheet does not contain a header row, use HDR=NO. If you are wrapping this functionality into a class, I would create properties to set these values.

Care must also be taken when constructing your SQL statement. If you have set the HDR property to YES, then you use the row heading names in your statement. If it is set to NO, the columns are assigned field names as F1, F2, F3, etc. In the FROM clause you can use a named range (e.g. SELECT * FROM [MyRange]) or the name of the worksheet (e.g. SELECT * FROM [Sheet1$]). In both cases, the source be enclosed in brackets, and when refering to a worksheet, the worksheet name must be followed by a $. I decided to reference the data in my project using the worksheet name instead of a named range. The reason is because I do not know how many rows of data there are until runtime. If you know exactly how many rows you will import every time, then I would use a named range.

I guess I should also mention the format of the data in your spreadsheet. The data should not contain any blank rows or subtotals, otherwise your dataset will include them as data rows (unless that’s what you want for some reason). If you’re not sure if your data will contain any of this, you can write a macro that copies the worksheet, removes the subtotals, etc., performs any data scrubbing and use that worksheet as your data source. You can then delete the worksheet you created after importing the data into the ADO.NET dataset. I did not run into any problems with converting data types, but all of my data was in either text or currency format. If using other types you should test the resulting dataset thoroughly.

I’ve referenced a link to the full KB article below which provides greater detail if you need it. The KB also describes how to update the data in a spreadsheet via ADO.NET, but that functionality was not required for this project so I will not discuss it here.

Link to KB article: http://support.microsoft.com/default.aspx?scid=kb;EN-US;316934