Next: , Previous: Introduction, Up: Top   [Contents][Index]

5 Literals in source code

To push an integer number on the data stack, you write the number in source code, e.g., 123. You can prefix the digits with - to indicate a negative number, e.g. -123. This works both inside colon definitions and outside. The number is interpreted according to the value of base (see Number Conversion). The digits are 0 to 9 and a (decimal 10) to z (decimal 35), but only digits smaller than the base are recognized. The conversion is case-insensitive, so A and a are the same digit.

You can make the base explicit for the number by using a prefix:

For combinations including base-prefix and sign, the standard order is to have the base-prefix first (e.g., #-123); Gforth supports both orders.

You can put a decimal point . at the end of a number (or, non-standardly, anywhere else except before a prefix) to get a double-cell integer (e.g., #-123. or #-.123 (the same number)). If users experienced in another programming language see or write such a number without base prefix (e.g., -123.), they may expect that the number represents a floating-point value. To clear up the confusion early, Gforth warns of such usage; to avoid the warnings, the best approach is to always write double numbers with a base prefix (e.g., #-123.)

Here are some examples, with the equivalent decimal number shown after in braces:

$-41 (-65), %1001101 (205), %1001.0001 (145 - a double-precision number), #905 (905), $abc (2478), $ABC (2478).

You can get the numeric value of a (character) code point by surrounding the character with ' (e.g., 'a'). The trailing ' is required by the standard, but you can leave it away in Gforth. Note that this also works for non-ASCII characters. For many uses, it is more useful to have the character as a string rather than as a cell; see below for the string syntax.

For floating-point numbers in Forth, you recognize them due to their exponent. I.e. 1. is a double-cell integer, and 1e0 is a floating-point number; the latter can be (and usually is) shortened to 1e. Both the significand (the part before the e or E) and the exponent may have signs (including +); the significand must contain at least one digit and may contain a decimal point, the exponent can be empty. Floating-point numbers always use decimal base for both significand and exponent, and are only recognized when the base is decimal. Examples are: 1e 1e0 1.e 1.e0 +1e+0 (which all represent the same number) +12.E-4.

A Gforth extension (since 1.0) is to write a floating-point number in scaled notation: It can optionally have a sign, then one or more digits, then use one of the mostly SI-defined scaling symbols (aka metric prefixes) or %, and then optionally more digits. Here’s the full list of scaling symbols that Gforth accepts:

Unlike most of the rest of Gforth, scaling symbols are treated case-sensitively. Using the scaled notation is equivalent to using a decimal point instead of the scaling symbol and appending the exponential notation at the end. Examples of scaled notation: 6k5 (6500e) 23% (0.23e).

You can input a string by surrounding it with " (e.g. "abc", "a b"). The result is the starting address and byte (=char) count of the string on the data stack.

You have to escape any " inside the string with \ (e.g., "double-quote->\"<-"). In addition, this string syntax supports all the ways to write control characters that are supported by s\" (see String and character literals). A disadvantage of this string syntax is that it is non-standard; for standard programs, use s\" instead.

You can input an environment variable by first loading rec-env.fs and then prefixing the environment variable with $, e.g., $HOME; the result is a string descriptor on the data stack in the format described above. This is equivalent to "HOME" getenv, i.e., the environment variable is resolved at run-time.

You can input an execution token (xt) of a word by prefixing the name of the word with ` (e.g., `dup). An advantage over using ' or ['] is you do not need to switch between them when copying and pasting code from inside to outside a colon definition or vice versa. A disadvantage is that this syntax is non-standard.

You can input a name token (nt) of a word by prefixing the name of the word with `` (e.g., ``dup). This syntax is also non-standard.

You can input a body address of a word by surrounding it with < and > (e.g., <spaces>). You can also input an address that is at a positive offset from the body address (typically an address in that body), by putting + and a number (see syntax above) between the word name and the closing > (e.g., <spaces+$15>, spaces+-3). You will get the body address plus the number. This non-standard feature exists to allow copying and pasting the output of ... (see Examining data).

Next: , Previous: Introduction, Up: Top   [Contents][Index]