string truth = "2² = 4"; auto m = utfMatcher!char(unicode.Number); assert(m.match(truth)); // '2' is a number all right assert(truth == "² = 4"); // skips on match assert(m.match(truth)); // so is the superscript '2' assert(!m.match(truth)); // space is not a number assert(truth == " = 4"); // unaffected on no match assert(!m.skip(truth)); // same test ... assert(truth == "= 4"); // but skips a codepoint regardless assert(!m.test(truth)); // '=' is not a number assert(truth == "= 4"); // test never affects argument
Perform a semantic equivalent 2 operations: decoding a $(CODEPOINT) at front of inp and testing if it belongs to the set of $(CODEPOINTS) of this matcher.
The effect on inp depends on the kind of function called:
Match. If the codepoint is found in the set then range inp is advanced by its size in code units, otherwise the range is not modifed.
Skip. The range is always advanced by the size of the tested $(CODEPOINT) regardless of the result of test.
Test. The range is left unaffected regardless of the result of test.