Inserting Characters

Typing letters, numbers and symbols from the keyboard inserts them into the current equation at the current cursor position. In addition, special "keyboard shortcuts" can be used to cut, paste, insert templates and perform other useful functions. Both kinds of keyboard input are described in more detail below.

MathFlow will accept characters from international keyboards.

In MathML equations, all characters are categorized according to type. In the MathFlow editor, the most important types are text, identifiers, numbers and operators. These types correspond to the MathML elements <mtext>, <mi>, <mn> and <mo>. In MathML, these basic character types are called tokens.

As you enter characters from the keyboard, the MathFlow editor analyzes the input and automatically inserts the characters into the appropriate MathML token element. Ordinarily, this results in the appropriate MathML markup, and makes authoring MathML equations much easier. As with any automatic algorithm there are cases where problems can arise. However, by understanding how the MathFlow editor chooses token types for characters, you can easily avoid or correct problem situations.

In the General Editing Preferences dialog, there is an option called Recognize Common Functions. How MathFlow handles characters and invisible operators such as the invisible multiplication character depends on whether or not you have selected this option.

If the "Recognize Common Functions" option is not selected and two or more characters of the same type are entered in succession, the MathFlow editor will place the characters into the same MathML token element. Thus, typing 's', 'i', 'n' will result in a single identifier token containing 'sin'. Similarly, if you place two or more characters of the same token type together by some other means, such as cut and paste, backspacing, etc. the MathFlow editor will merge them together into a single token.

However, when the "Recognize Common Functions" is selected, MathFlow will look at the characters you type in and try to recognize a common function, such as cosine. If it determines you have typed in a function, the individual characters are combined in a single token. If a function is not found, each character is placed in a separate token.

The MathFlow editor uses the following rules to decide what token type to assign to characters. A run of alphabetic characters is placed in an identifier, or <mi>, token. A run consisting of digits, commas, and/or periods is placed in a number, or <mn>, token. By default, everything else is placed in an operator, or <mo>, token. Whenever you enter a character of a type that doesn't match the preceding characters, the Editor starts a new token for it.

When problems arise, they usually result when a run of characters should be broken into several tokens, or when separate tokens merge as a result of editing. In the following sections, we outline the most common issues.