iupMask

IUP - Portable User Interface

iupMask - Pattern Specification

The pattern to be searched in the text can be defined by the rules given below. Note that such rules are very similar to the ones used by Lua, even though they are not the same. For further information on these patterns, please refer to the Lua Manual.

Notes

  • "Function" codes (such as /l, /D, /w) cannot be used inside a class ([...]).
  • If the character following a / does not mean a special case (such as /w or /n), it is matched with no / - that means that /x will match only x, and not /x. If you want to match /x, use //x.
  • The caret (^) character has different meanings when used inside or outside a class - inside a class it means negative, and outside a class it is an anchor to the beginning of a line.
  • The boundary function (/b) anchors the pattern to a word boundary - it does not match anything. A word boundary is a point between a /w and a /W character.
  • Capture operators (f and g) group patterns and are also used to keep matched sections of texts.
  • A word on precedence: concatenation has precedence over the alternation (j) operator - that is, faj fej fi will match fa OR fe OR fi.
  • The @ character is used to determine that, instead of searching the text until the first match is made, the function should try to match the pattern only with the first character. If present, it must be the first character of the pattern.
  • The % character is used to determine that the text should be searched to its end, independently of the number of matches found. If present, it must be the first character of the pattern. This is only useful when combined with the capture feature.

Allowed pattern characters

c Matches a c (non-special) character
. Matches any single character
[abc] Matches an a, b or c
[a-d] Matches any character between a and d, including them (just like [abcd])
[^a-dg] Matches any character which is neither between a and d nor a g
/d Matches any digit (just like [0-9])
/D Matches any non-digit (just like [^0-9])
/l Matches any letter (just like [a-zA-Z])
/L Matches any non-letter (just like [^a-zA-Z])
/w Matches any alphanumeric character (just like [0-9a-zA-Z ])
/W Matches any non-alphanumeric character (just like [^0-9a-zA-Z ])
/s Matches any "blank" character (TAB, SPACE, CR)
/S Matches ant non-blank character
/n Matches a newline character
/t Matches a tabulation character
/nnn Matches an ASCII character with a nnn value (decimal)
/xnn Matches an ASCII character with a nn value (hexadecimal)
/special Matches the special character literally (/[, //, /.)
abc Matches a sequence of a, b and c patterns in order
aj bj c Matches a pattern a, b or c
a* Matches 0 or more characters a
a+ Matches 1 or more characters a
a? Matches 1 or no characters a
(pattern) Considers pattern as one character for the above
fpatterng Captures pattern for later reference
/b Anchors to a word boundary
/B Anchors to a non-boundary
^pattern Anchors pattern to the beginning of a line
pattern$ Anchors pattern to the end of a line
@pattern Returns the match found only in the beginning of the text
%pattern Returns the firstmatch found, but searches all the text

Examples

(my|his) Matches both my pattern and his pattern.
/d/d:/d/d(:/d/d)? Matches time with seconds (01:25:32) or without seconds (02:30).
[A-D]/l+ Matches names such as Australia, Bolivia, Canada or Denmark, but not England, Spain or single letters such as A.
/l/w* my variable = 23 * width;
^Subject:[^/n]*/n Subject: How to match a subject line.1
/b[ABab]/w* Matches any word that begins with A or B
from:/s*/w+ Captures "sender" in a message from sender