Files
MagicSetEditor2/doc/function/break_text.txt
T
twanvl 158ecf67ad Now all regex functions use ScriptRule.
The *_rule functions can now be considered deprecated
Documented this by removing mentions of the rule functions, except for a mention of backwards compatibility.

git-svn-id: svn://svn.code.sf.net/p/magicseteditor/code/trunk@997 0fc631ac-6414-0410-93d0-97cfa31319b6
2008-06-19 00:37:48 +00:00

42 lines
1.7 KiB
Plaintext

Function: break_text
--Usage--
> break_text(some_string, match: regular expression, in_context: regular expression)
Break text by only keeping the parts of the input that match the regular expression.
The function returns a [[type:list]] of parts.
If @in_context@ is given, the context must also match the string where the match is represented as <tt>&lt;match></tt>.
When the @break_text@ is used many times with the same @match@ or @in_context@ values it is more efficient to declare these as default arguments:
> my_break := break_text@(match: "something")
> my_break("input") # called many times
This way the regular expression is only compiled once.
--Filter vs. break--
The function @filter_text@ is very similar to @break_text@, instead of returning a list it concatenates the items.
So for example where @break_text@ returns @["a","b","c"]@, @filter_text@ would return @"abc"@.
In fact, @filter_text@ could be implemented as
> filter_text := { for part in break_text() do part }
--Parameters--
! Parameter Type Description
| @input@ [[type:string]] String to replace in.
| @match@ [[type:regex]] Regular expression to match.
| @in_context@ [[type:regex]] (optional) Context to match
--Examples--
> break_text(match: "a", "banana") == ["a","a","a"]
> break_text(match: "na|.", "banana") == ["b","a","na","na"]
> break_text(match: "ap", "banana") == []
> break_text(match: "/", "a/b/c") == ["/","/"]
> break_text(match: "[^/]+", "a/b/c") == ["a","b","c"]
>
> f := break_text@(match: "xx+")
> f("xyzxxxxyyzz") == ["xxxx"]
--See also--
| [[fun:filter_text]] Keep only the text matching a regular expression.