fix #49 regular expressions, operator match and docs

docs/Regex.md

@@ -0,0 +1,91 @@
+# Regular expressions
+
+In lyng, you create regular expressions using class `Regex` or `String.re` methods:
+
+    assert( "\d*".re is Regex )
+    assert( Regex("\d*") is Regex )
+    >>> void
+
+We plan to add slash syntax at some point.
+
+To check that some string matches as whole to some regex:
+
+    assert( "123".matches("\d{3}".re) )
+    assert( !"123".matches("\d{4}".re) )
+    assert( !"1234".matches("\d".re) )
+    >>> void
+
+To check that _part of the string_ matches some regular expession, use _match operator_ `=~` just like in Ruby, and its
+counterpart, _not match_ operator `!~`:
+
+    assert( "abc123def" =~ "\d\d\d".re )
+    assert( "abc" !~ "\d\d\d".re )
+    >>> void
+
+When you need to find groups, and more detailed match information, use `Regex.find`:
+
+    val result = Regex("abc(\d)(\d)(\d)").find( "bad456 good abc123")
+    assert( result  != null )
+    assertEquals( 12 .. 17, result.range )
+    assertEquals( "abc123", result[0] )
+    assertEquals( "1", result[1] )
+    assertEquals( "2", result[2] )
+    assertEquals( "3", result[3] )
+    >>> void
+
+Note that the object `RegexMatch`, returned by [Regex.find], behaves much like in many other languages: it provides the
+index range and groups matches as indexes.
+
+Match operator actually also provides `RegexMatch` in `$~` reserved variable (borrowed from Ruby too):
+
+    assert( "bad456 good abc123" =~ "abc(\d)(\d)(\d)".re )
+    assertEquals( 12 .. 17, $~.range )
+    assertEquals( "abc123", $~[0] )
+    assertEquals( "1", $~[1] )
+    assertEquals( "2", $~[2] )
+    assertEquals( "3", $~[3] )
+    >>> void
+
+This is often more readable than calling `find`.
+
+Note that `=~` and `!~` operators against strings and regular expressions are commutative, e.g. regular expression and a
+string can be either left or right operator, but not both:
+
+    assert( "abc" =~ "\wc".re )
+    assert( "abc" !~ "\w1c".re )
+    assert( "a\wc".re =~ "abcd" )
+    assert( "a[a-z]c".re !~ "a2cd" )
+    >>> void
+
+Also, string indexing is Regex-aware, and works like `Regex.find` (_not findall!_):
+
+    assert( "cd" == "abcdef"[ "c.".re ].value )
+    >>> void
+
+
+# Regex class reference
+
+| name         | description                         | notes |
+|--------------|-------------------------------------|-------|
+| matches(str) | true if the whole `str` matches     |       |
+| find(str)    | find first match in `str` or null   | (1)   |
+| findAll(str) | find all matches in `str` as [List] | (1)   |
+
+(1)
+:: See `RegexMatch` class description below
+
+# RegexMatch
+
+| name  | description                               | notes |
+|-------|-------------------------------------------|-------|
+| range | the [Range] of the match in source string |       |
+| value | the value that matches                    |       |
+| [n]   | [List] of group matches                   | (1)   |
+
+(1)
+:: the [0] element is always value, [1] is group 1 match of any, etc.
+
+[List]: List.md
+
+[Range]: Range.md
+

docs/tutorial.md

@@ -275,6 +275,8 @@ Logical operation could be used the same
 |   ===    |     | Any               | (2)      |
 |   !==    |     | Any               | (2)      |
 |    !=    |     | Any               | (1)      |
+|    =~    |     |                   | (3)      |
+|    !~    |     |                   | (3)      |
 | ++a, a++ |     | Int               |          |
 | --a, a-- |     | Int               |          |
 
@@ -286,6 +288,9 @@ Logical operation could be used the same
 singleton object, like `null`, are referentially equal too, while string different literals even being equal are most
 likely referentially not equal
 
+(3)
+: Implemented now in String and Regex as regular expression match and not match, see [Regex].
+
 Reference quality and object equality example:
 
     assert( null == null)  // singletons
@@ -1285,9 +1290,26 @@ Open-ended ranges could be used to get start and end too:
     assertEquals( "pult", "catapult"[ 4.. ])
     >>> void
 
+
 ### String operations
 
-Concatenation is a `+`: `"hello " + name` works as expected. No confusion.
+Concatenation is a `+`: `"hello " + name` works as expected. No confusion. There is also
+[Regex] support for strings, see the link, for example, whole string match:
+
+    assert( !"123".matches( "\d\d".re ) )
+    assert( "123".matches( "\d\d\d".re ) )
+    >>> void
+
+Extraction:
+
+    "abcd42def"[ "\d+".re ].value
+    >>> "42"
+
+Part match:
+
+    assert( "abc foo def" =~ "f[oO]+".re )
+    assert( "foo" == $~.value )
+    >>> void
 
 Typical set of String functions includes:
 
@@ -1305,17 +1327,24 @@ Typical set of String functions includes:
 | size               | size in characters like `length` because String is [Array] |
 | (args...)          | sprintf-like formatting, see [string formatting]           |
 | [index]            | character at index                                         | 
-| [Range]            | substring at range                                         |
+| [Range]            | substring at range (2)                                     |
+| [Regex]            | find first match of regex, like [Regex.find] (2)           |
 | s1 + s2            | concatenation                                              |
 | s1 += s2           | self-modifying concatenation                               |
 | toReal()           | attempts to parse string as a Real value                   |
 | toInt()            | parse string to Int value                                  |
 | characters()       | create [List] of characters (1)                            |
 | encodeUtf8()       | returns [Buffer] with characters encoded to utf8           |
+| matches(re)        | matches the regular expression (2)                         |
+|                    |                                                            |
+
 
 (1)
 : List is mutable therefore a new copy is created on each call.
 
+(2)
+: See [Regex]
+
 ### Literals
 
 String literal could be multiline:
@@ -1390,4 +1419,6 @@ See [math functions](math.md). Other general purpose functions are:
 
 [Collection]: Collection.md
 
-[Array]: Array.md
+[Array]: Array.md
+
+[Regex]: Regex.md