In the programming world there are various types of numbers. In some programming languages, numbers need to be identified by type in the program in order for the program to recognize them properly. For example, the number 2 is an integer, but the number 2.222 is of a type called floating point due to its fraction in decimal format. However, for the general purpose of using the Visualization API, all types of numbers are treated simply as one type. In the previous example, 2 and 2.222 are both simply called numbers in the Visualization API.

The concept of a Boolean value is fairly straightforward. A Boolean value can only hold the values "True" or "False". Using a Boolean variable is similar to asking a yes or no question without the ability to accept "maybe" as an answer. In programming practice, Boolean variables are very useful when a condition needs to be tested to be true or false before taking a certain action in the program. How to propose a condition to be tested is covered in the Conditional logic section of this chapter.

A string is exactly as it sounds: it is a collection of characters saved in one variable. Note that characters are not simply A to Z and 0 to 9, but also include punctuation and often other special characters. Similar to number variables, strings can be used to hold information for the application, or store static information to be used by the application. For example, a string containing the characters that spell the word Google would look like this:

string_name = 'Google'

It is important that characters designated to be a single string must be enclosed in quotes. Depending on the specific program method being used, the quotes may be required to be single or double quotes. An example of using a string as a constant label is as follows:

Varname_label = 'Name: '

Here, 'Name: ' could be a field label appearing in an address form application. Note the blank space to the right of the colon. For strings, whitespace is considered a character and may change the expected output of an application if not taken into account.

Also, a variable may be used to store user input as a string. For example, to capture the name entered by a user in an address form application, a variable could be defined as follows:

Var name_field = ''

Notice, in the previous statement there is nothing, not even a whitespace, being stored in the name_field variable. When nothing is assigned to a variable, the variable is considered to have a value of null , or nothing. Giving a variable a value of null is commonly used to set a clean default state. When the application runs, the name_field variable can then be used with confidence to capture user input accurately. Given the variable name of name_field, it would seem that this particular variable is intended to capture name data from a form application.

Arrays look a lot like tables, but they are more versatile than a simple table as they can have various dimensions and hold several layers of information. Yet for purposes of the Visualization API, most arrays are generally two-dimensional, thus having the familiar format of columns and rows with single values in the cells. The underlying structure of the Visualization API DataTable function is essentially an array. The DataTable function is used to create a dataset within the API script. For example, the 2010 Chicago Census data from previous chapters can be created completely within the script, and then manipulated by the Visualization API. The following example constructs an array of data using a DataTable helper function called arrayToDataTable(), and consists of several lines of data from the 2010 Chicago Census data.

To reference a specific cell in the data table, a coordinate system is used. For example, to find the number of females that are 20 to 24 years old, matrix coordinates are attached to the data variable.

data[5,2]

It is important to observe that the column and row ranges begin at zero and not with the number one. Also note the entries 'Under 5 years' is in quotes while the other entries are numbers. Arrays can hold a variety of data types; in this specific case, both strings and numbers are contained within a single array.

Just as in mathematics, equations in Google API applications are a method of performing an operation and equating it to something else. Common mathematical operators (+, -, *, /), as well as Boolean operators (AND, OR, NOT, or alternatively notated as &&, ||, !) can be used.

var data = google.visualization.arrayToDataTable([['Age', 'Male', 'Female', 'Both Sexes'],
    ['Under 5 years',    94100,       91787,         185887],
    ['5 to 9 years',     84122,       81955,         166077]
]);

If (sunny AND warm) {
airconditioning = true
}

The purpose of good commenting practice is to ensure that another developer will be able to determine the purpose of the application without significant forensic efforts. In general, single-line comments are marked by a double forward slash // and blocks of comments are indicated by a forward slash plus asterisks combination /*…*/. It is generally appropriate to provide a block comment at the beginning of a script, application, or even a function. There are a handful of topics that can be included in the comment block:

/* Description: This application draws a bar chart of some data.
Author: Traci Ruthkoski
Date Revised: 9/1/2012
*/

The purpose of single-line commenting is to give short descriptions of a specific line or section of code in order to aid in its readability. However, a potential pitfall when commenting is the overuse of comments, ultimately reducing the readability of the code.

var year='2012';  // Define the year.

Of course, lack of comments does not prevent code from running. The developer must determine the types of comments that are appropriate to their code. Ultimately, commenting is not a requirement but a strongly suggested element of an application, and is often a requirement when publishing applications to Google galleries and forums.

When writing good code, it is also important to address its readability from a formatting perspective. Just as writing prose requires paragraphs and margins to be read easily, code is more readable when its logical functions are visually segmented. The collection of proper spacing and formatting for a given code language is known as code conventions. The following illustrates a simple framework for detecting an error in an application. Indentation indicates that the error checking statement is encapsulated within the computation of the DrawChart function.

function DrawChart {
     if (something == error) {
          message = 'This app encountered an error!'
     }
}