Loading time based data file into GoodData

This basic example loads a CSV data file containing stock quotes from the AMEX, NASDAQ and NYSE markets from January to August 2008.

Run gdi.sh with the quotes.txt script as follows:

Windows:

c:> bin\gdi.bat -u <username> -p <password> examples\quotes\quotes.txt
Project id = 'f5977852bfec20271d4c9bc453a263cb' created.

Unix like OS (Linux, Mac OS X and others):

$ ./bin/gdi.sh -u <username> -p <password> examples/quotes/quotes.txt
Project id = 'f5977852bfec20271d4c9bc453a263cb' created.

This creates a “Quotes” project in GoodData, sets up the data model and loads the data file into the project.

What’s Inside?

Each line of the CSV file contain fields titled Id, Company, Symbol, Sector, Industry, Market, Quote Date, Open Price, High Price, Low Price, Close Price, Volume and Adjusted Close Price.

These fields are described in the pre-created examples/quotes/config.xml configuration file that consists of records such as:

    <column>
      <name>MARKET</name>
      <title>Market</title>
      <ldmType>ATTRIBUTE</ldmType>
    </column>
    ...
    <column>
      <name>CLOSE_PRICE</name>
      <title>Close Price</title>
      <ldmType>FACT</ldmType>
    </column>

First <column> record describes the first column of the CSV file, second record the second column and so on. Most of fields is either a number to be aggregated, such as a Stock Price (its ldmType is FACT) or it can be used to break down an aggregated number, e.g Market or Industry - these fields have ldmType set to ATTRIBUTE.

Other ldmTypes include CONNECTION_POINT (a unique identifier of a record within a dataset), LABEL (an alias of an ATTRIBUTE or a CONNECTION_POINT) and DATE. For example, the Id field is marked as CONNECTION_POINT, Company is an alias of SYMBOL and the Quote Date holds a date formatted as yyyy-MM-dd (e.g. 2008-02-28 for February 28th 2008).

Note the schemaReference property tells GoodData to connect this date field to the date dimension named Quote. This date dimension is created in the model by the following lines at the begining of the quotes.txt script:

    UseDateDimension(name="Quote");
    GenerateMaql(maqlFile="examples/quotes/quote_date.maql");
    ExecuteMaql(maqlFile="examples/quotes/quote_date.maql");

The structure of the logical model built on the top of this data set is as follows:

The light green boxes depict attributes including the Id connection point, the ovals represent facts. Labels and most of the attributes of the Date (Quotes) dimension hidden for the sake of simplicity. The attributes belonging to the date dimension are shown using double border line.

What Next?

Now you can log into the GoodData user interface and select the Quotes project. When you switch to the Data section and click Model in the left menu bar you can see a data model visualization similar to what’s outlined above. Then you can switch to the Reports section and start building your first reports.

Read further comments in the script file for instructions how to setup a repeatable data load or how to create another project based your own CSV data set.