Abstract: An improved method for text searching has application to application programs in computers, where the application program handles files written in a language. Advantageously a character within the first string is selected. In the general case the character selected is a relatively rarely used character in the language, and in simplified cases it is selected as non-alphabetic or as found within a set of characters known to be relatively rarely used.The file is then scanned, starting not at the beginning but rather from a position offset from the beginning of the file according to the position of the selected character within the first string. In the event of a match for that character, the remainder of the character positions of the first string are compared for a string-length match. The scanning terminates at a position short of the end of the file according to the position of the selected character within the first string, or terminates on a string-length match.