The Trunk: Regex-Help-pre.1.mcz

Patrick Rein uploaded a new version of Regex-Help to project The Trunk:
http://source.squeak.org/trunk/Regex-Help-pre.1.mcz

==================== Summary ====================

Name: Regex-Help-pre.1
Author: pre
Time: 6 July 2018, 5:14:56.533269 pm
UUID: 476b203d-1709-a54e-9a96-f0dfc3a93dfd
Ancestors:

Converts the regex documentation from class methods to a full help topic.

==================== Snapshot ====================

SystemOrganization addCategory: #'Regex-Help'!

CustomHelp subclass: #RegexHelp
        instanceVariableNames: ''
        classVariableNames: ''
        poolDictionaries: ''
        category: 'Regex-Help'!

----- Method: RegexHelp class>>bookName (in category 'as yet unclassified') -----
bookName

        ^ 'Regex'!

----- Method: RegexHelp class>>changelog (in category 'as yet unclassified') -----
changelog
        "This method was automatically generated. Edit it using:"
        "RegexHelp edit: #changelog"
        ^(HelpTopic
                title: 'Changelog'
                contents:
'VERSION 1.3.1 (September 2008)
1. Updated documentation of character classes, making clear the problems of locale - an area for future improvement

VERSION 1.3 (September 2008)
1. \w now matches underscore as well as alphanumerics, in line with most other regex libraries (and our documentation!!!!).  
2. \W rejects underscore as well as alphanumerics
3. added tests for this at end of testSuite
4. updated documentation and added note to old incorrect comments in version 1.1 below

VERSION 1.2.3 (November 2007)

1. Regexs with ^ or $ applied to copy empty strings caused infinite loops, e.g. ('''' copyWithRegex: ''^.*$'' matchesReplacedWith: ''foo''). Applied a similar correction to that from version 1.1c, to #copyStream:to:(replacingMatchesWith:|translatingMatchesUsing:).
2. Extended RxParser testing to run each test for #copy:translatingMatchesUsing: as well as #search:.
3. Corrected #testSuite test that a dot does not match a null, which was passing by luck with Smalltalk code in a literal array.
4. Added test to end of test suite for fix 1 above.

VERSION 1.2.2 (November 2006)

There was no way to specify a backslash in a character set. Now [\\] is accepted.

VERSION 1.2.1 (August 2006)

1. Support for returning all ranges (startIndex to: stopIndex) matching a regex - #allRangesOfRegexMatches:, #matchingRangesIn:
2. Added hint to usage documentation on how to get more information about matches when enumerating
3. Syntax description of dot corrected: matches anything but NUL since 1.1a

VERSION 1.2 (May 2006)

Fixed case-insensitive search for character sets.

VERSION 1.1c (December 2004)

Fixed the issue with #matchesOnStream:do: which caused infinite loops for matches
that matched empty strings.

VERSION 1.1b (November 2001)

Changes valueNowOrOnUnwindDo: to ensure:, plus incorporates some earlier fixes.

VERSION 1.1a (May 2001)

1. Support for keeping track of multiple subexpressions.
2. Dot (.) matches anything but NUL character, as it should per POSIX spec.
3. Some bug fixes.

VERSION 1.1 (October 1999)

Regular expression syntax corrections and enhancements:

1. Backslash escapes similar to those in Perl are allowed in patterns:

        \w any word constituent character (equivalent to [a-zA-Z0-9_]) *** underscore only since 1.3 ***
        \W any character but a word constituent (equivalent to [^a-xA-Z0-9_] *** underscore only since 1.3 ***
        \d a digit (same as [0-9])
        \D anything but a digit
        \s a whitespace character
        \S anything but a whitespace character
        \b an empty string at a word boundary
        \B an empty string not at a word boundary
        \< an empty string at the beginning of a word
        \> an empty string at the end of a word

For example, ''\w+'' is now a valid expression matching any word.

2. The following backslash escapes are also allowed in character sets
(between square brackets):

        \w, \W, \d, \D, \s, and \S.

3. The following grep(1)-compatible named character classes are
recognized in character sets as well:

        [:alnum:]
        [:alpha:]
        [:cntrl:]
        [:digit:]
        [:graph:]
        [:lower:]
        [:print:]
        [:punct:]
        [:space:]
        [:upper:]
        [:xdigit:]

For example, the following patterns are equivalent:

        ''[[:alnum:]_]+'' ''\w+''  ''[\w]+'' ''[a-zA-Z0-9_]+'' *** underscore only since 1.3 ***

4. Some non-printable characters can be represented in regular
expressions using a common backslash notation:

        \t tab (Character tab)
        \n newline (Character lf)
        \r carriage return (Character cr)
        \f form feed (Character newPage)
        \e escape (Character esc)

5. A dot is corectly interpreted as ''any character but a newline''
instead of ''anything but whitespace''.

6. Case-insensitive matching.  The easiest access to it are new
messages CharacterArray understands: #asRegexIgnoringCase,
#matchesRegexIgnoringCase:, #prefixMatchesRegexIgnoringCase:.

7. The matcher (an instance of RxMatcher, the result of
String>>asRegex) now provides a collection-like interface to matches
in a particular string or on a particular stream, as well as
substitution protocol. The interface includes the following messages:

        matchesIn: aString
        matchesIn: aString collect: aBlock
        matchesIn: aString do: aBlock

        matchesOnStream: aStream
        matchesOnStream: aStream collect: aBlock
        matchesOnStream: aStream do: aBlock

        copy: aString translatingMatchesUsing: aBlock
        copy: aString replacingMatchesWith: replacementString

        copyStream: aStream to: writeStream translatingMatchesUsing: aBlock
        copyStream: aStream to: writeStream replacingMatchesWith: aString

Examples:

        ''\w+'' asRegex matchesIn: ''now is the time''

returns an OrderedCollection containing four strings: ''now'', ''is'',
''the'', and ''time''.

        ''\<t\w+'' asRegexIgnoringCase
                copy: ''now is the Time''
                translatingMatchesUsing: [:match | match asUppercase]

returns ''now is THE TIME'' (the regular expression matches words
beginning with either an uppercase or a lowercase T).

ACKNOWLEDGEMENTS

Since the first release of the matcher, thanks to the input from
several fellow Smalltalkers, I became convinced a native Smalltalk
regular expression matcher was worth the effort to keep it alive. For
the contributions, suggestions, and bug reports that made this release
possible, I want to thank:

        Felix Hack
        Peter Hatch
        Alan Knight
        Eliot Miranda
        Thomas Muhr
        Robb Shecter
        David N. Smith
        Francis Wolinski

and anyone whom I haven''t yet met or heard from, but who agrees this
has not been a complete waste of time.!!' readStream nextChunkText)
                        key: #changelog!

----- Method: RegexHelp class>>examples (in category 'as yet unclassified') -----
examples
        "This method was automatically generated. Edit it using:"
        "RegexHelp edit: #examples"
        ^(HelpTopic
                title: 'Examples'
                contents:
'As the introductions said, a great use for regular expressions is userinput validation. Following are a few examples of regular expressionsthat might be handy in checking input entered by the user in an inputfield. Try them out by entering something between the quotes andprint-iting. (Also, try to imagine Smalltalk code that each validationwould require if coded by hand).  Most example expressions could havebeen written in alternative ways.

Checking if aString may represent a nonnegative integer number:

        '''' matchesRegex: '':isDigit:+''or '''' matchesRegex: ''[0-9]+''or '''' matchesRegex: ''\d+''

Checking if aString may represent an integer number with an optionalsign in front:

        '''' matchesRegex: ''(\+|-)?\d+''

Checking if aString is a fixed-point number, with at least one digitis required after a dot:

        '''' matchesRegex: ''(\+|-)?\d+(\.\d+)?''

The same, but allow notation like `123.'':

        '''' matchesRegex: ''(\+|-)?\d+(\.\d*)?''

Recognizer for a string that might be a name: one word with firstcapital letter, no blanks, no digits.  More traditional:

        '''' matchesRegex: ''[A-Z][A-Za-z]*''

more Smalltalkish:

        '''' matchesRegex: '':isUppercase::isAlphabetic:*''

A date in format MMM DD, YYYY with any number of spaces in between, inXX century:

        '''' matchesRegex: ''(Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec)[ ]+(\d\d?)[ ]*,[ ]*19(\d\d)''

Note parentheses around some components of the expression above. As`Usage'' section shows, they will allow us to obtain the actual stringsthat have matched them (i.e. month name, day number, and year number).

For dessert, coming back to numbers: here is a recognizer for ageneral number format: anything like 999, or 999.999, or -999.999e+21.

        '''' matchesRegex: ''(\+|-)?\d+(\.\d*)?((e|E)(\+|-)?\d+)?''!!!!!!' readStream nextChunkText)
                        key: #examples!

----- Method: RegexHelp class>>implementationNotes (in category 'as yet unclassified') -----
implementationNotes
        "This method was automatically generated. Edit it using:"
        "RegexHelp edit: #implementationNotes"
        ^(HelpTopic
                title: 'Implementation Notes'
                contents:
'WHAT TO LOOK AT FIRST

String>>matchesRegex: -- in 90% cases this method is all you need to access the package.

RxParser -- accepts a string or a stream of characters with a regular expression, and produces a syntax tree corresponding to the expression. The tree is made of instances of Rxs<whatever> classes.

RxMatcher -- accepts a syntax tree of a regular expression built by the parser and compiles it into a matcher: a structure made of instances of Rxm<whatever> classes. The RxMatcher instance can test whether a string or a positionable stream of characters matches the original regular expression, or search a string or a stream for substrings matching the expression. After a match is found, the matcher can report a specific string that matched the whole expression, or any parenthesized subexpression of it.

All other classes support the above functionality and are used by RxParser, RxMatcher, or both.


        CAVEATS

The matcher is similar in spirit, but NOT in the design--let alone the code--to the original Henry Spencer''s regular expression implementation in C.  The focus is on simplicity, not on efficiency. I didn''t optimize or profile anything.  I may in future--or I may not: I do this in my spare time and I don''t promise anything.

The matcher passes H. Spencer''s test suite (see ''test suite'' protocol), with quite a few extra tests added, so chances are good there are not too many bugs.  But watch out anyway.


        EXTENSIONS, FUTURE, ETC.

With the existing separation between the parser, the syntax tree, and the matcher, it is easy to extend the system with other matchers based on other algorithms. In fact, I have a DFA-based matcher right now, but I don''t feel it is good enough to include it here.  I might add automata-based matchers later, but again I don''t promise anything.

        HOW TO REACH ME

As of today (December 20, 2000), you can contact me at <[hidden email]>. If this doesn''t work, look around comp.lang.smalltalk or comp.lang.lisp.  !!' readStream nextChunkText)
                        key: #implementationNotes!

----- Method: RegexHelp class>>introduction (in category 'as yet unclassified') -----
introduction
        "This method was automatically generated. Edit it using:"
        "RegexHelp edit: #introduction"
        ^(HelpTopic
                title: 'Introduction'
                contents:
'A regular expression is a template specifying a class of strings. A regular expression matcher is an tool that determines whether a string belongs to a class specified by a regular expression.  This is a common task of a user input validation code, and the use of regular expressions can GREATLY simplify and speed up development of such code.  As an example, here is how to verify that a string is a valid hexadecimal number in Smalltalk notation, using this matcher package:

        aString matchesRegex: ''16r[[:xdigit:]]+''

(Coding the same "the hard way'''' is an exercise to a curious reader).

This matcher is offered to the Smalltalk community in hope it will be useful. It is free in terms of money, and to a large extent -- in terms of rights of use. Refer to the "Boring Stuff" section for legalese.

The "Syntax" section explains the recognized syntax of regular expressions.

The "Usage" section explains matcher capabilities that go beyond what String>>matchesRegex: method offers.

The "Implementation Notes" sections says a few words about what is under the hood.

The "Changelog" section describes the functionality introduced in 1.1 release.

Happy hacking,

--Vassili Bykov
<[hidden email]> <[hidden email]>
!!
]style[(479 40 712),daString matchesRegex: ''16r[[:xdigit:]]+'';;,!!' readStream nextChunkText)
                        key: #introduction!

----- Method: RegexHelp class>>license (in category 'as yet unclassified') -----
license
        "This method was automatically generated. Edit it using:"
        "RegexHelp edit: #license"
        ^(HelpTopic
                title: 'License'
                contents:
'The Regular Expression Matcher (``The Software'''') is Copyright (C) 1996, 1999 Vassili Bykov. It is provided to the Smalltalk community in hope it will be useful.

1. This license applies to the package as a whole, as well as to any component of it. By performing any of the activities described below, you accept the terms of this agreement.

2. The software is provided free of charge, and "as is'''', in hope that it will be useful, with ABSOLUTELY NO WARRANTY. The entire risk and all responsibility for the use of the software is with you.  Under no circumstances the author may be held responsible for loss of data, loss of profit, or any other damage resulting directly or indirectly from the use of the software, even if the damage is caused by defects in the software.

3. You may use this software in any applications you build.

4. You may distribute this software provided that the software documentation and copyright notices are included and intact.

5. You may create and distribute modified versions of the software, such as ports to other Smalltalk dialects or derived work, provided that:

  a. any modified version is expressly marked as such and is not misrepresented as the original software;

  b. credit is given to the original software in the source code and documentation of the derived work;

  c. the copyright notice at the top of this document accompanies copyright notices of any modified version.!!' readStream nextChunkText)
                        key: #license!

----- Method: RegexHelp class>>pages (in category 'as yet unclassified') -----
pages

        ^ #(introduction syntax examples usage implementationNotes license changelog)!

----- Method: RegexHelp class>>syntax (in category 'as yet unclassified') -----
syntax
        "This method was automatically generated. Edit it using:"
        "RegexHelp edit: #syntax"
        ^(HelpTopic
                title: 'Syntax'
                contents:
'The simplest regular expression is a single character.  It matchesexactly that character. A sequence of characters matches a string withexactly the same sequence of characters:

        ''a'' matchesRegex: ''a'' -- true
        ''foobar'' matchesRegex: ''foobar'' -- true
        ''blorple'' matchesRegex: ''foobar'' -- false

The above paragraph introduced a primitive regular expression (acharacter), and an operator (sequencing). Operators are applied toregular expressions to produce more complex regular expressions.Sequencing (placing expressions one after another) as an operator is,in a certain sense, ''invisible''--yet it is arguably the most common.

A more ''visible'' operator is Kleene closure, more often simplyreferred to as ''a star''.  A regular expression followed by an asteriskmatches any number (including 0) of matches of the originalexpression. For example:

        ''ab'' matchesRegex: ''a*b'' -- true
        ''aaaaab'' matchesRegex: ''a*b'' -- true
        ''b'' matchesRegex: ''a*b'' -- true
        ''aac'' matchesRegex: ''a*b'' -- false: b does not match

A star''s precedence is higher than that of sequencing. A star appliesto the shortest possible subexpression that precedes it. For example,''ab*'' means ''a followed by zero or more occurrences of b'', not ''zeroor more occurrences of ab'':

        ''abbb'' matchesRegex: ''ab*'' -- true
        ''abab'' matchesRegex: ''ab*'' -- false

To actually make a regex matching ''zero or more occurrences of ab'',''ab'' is enclosed in parentheses:

        ''abab'' matchesRegex: ''(ab)*'' -- true
        ''abcab'' matchesRegex: ''(ab)*'' -- false: c spoils the fun

Two other operators similar to ''*'' are ''+'' and ''?''. ''+'' (positiveclosure, or simply ''plus'') matches one or more occurrences of theoriginal expression. ''?'' (''optional'') matches zero or one, but nevermore, occurrences.

        ''ac'' matchesRegex: ''ab*c'' -- true
        ''ac'' matchesRegex: ''ab+c'' -- false: need at least one b
        ''abbc'' matchesRegex: ''ab+c'' -- true
        ''abbc'' matchesRegex: ''ab?c'' -- false: too many b''s

As we have seen, characters ''*'', ''+'', ''?'', ''('', and '')'' have specialmeaning in regular expressions. If one of them is to be usedliterally, it should be quoted: preceded with a backslash. (Thus,backslash is also special character, and needs to be quoted for aliteral match--as well as any other special character describedfurther).

        ''ab*'' matchesRegex: ''ab*'' -- false: star in the right string is special
        ''ab*'' matchesRegex: ''ab\*'' -- true
        ''a\c'' matchesRegex: ''a\\c'' -- true

The last operator is ''|'' meaning ''or''. It is placed between tworegular expressions, and the resulting expression matches if one ofthe expressions matches. It has the lowest possible precedence (lowerthan sequencing). For example, ''ab*|ba*'' means ''a followed by anynumber of b''s, or b followed by any number of a''s'':

        ''abb'' matchesRegex: ''ab*|ba*'' -- true
        ''baa'' matchesRegex: ''ab*|ba*'' -- true
        ''baab'' matchesRegex: ''ab*|ba*'' -- false

A bit more complex example is the following expression, matching thename of any of the Lisp-style ''car'', ''cdr'', ''caar'', ''cadr'',... functions:

        c(a|d)+r

It is possible to write an expression matching an empty string, forexample: ''a|''.  However, it is an error to apply ''*'', ''+'', or ''?'' tosuch expression: ''(a|)*'' is an invalid expression.

So far, we have used only characters as the ''smallest'' components ofregular expressions. There are other, more ''interesting'', components.

A character set is a string of characters enclosed in squarebrackets. It matches any single character if it appears between thebrackets. For example, ''[01]'' matches either ''0'' or ''1'':

        ''0'' matchesRegex: ''[01]'' -- true
        ''3'' matchesRegex: ''[01]'' -- false
        ''11'' matchesRegex: ''[01]'' -- false: a set matches only one character

Using plus operator, we can build the following binary numberrecognizer:

        ''10010100'' matchesRegex: ''[01]+'' -- true ''10001210'' matchesRegex: ''[01]+'' -- false

If the first character after the opening bracket is ''^'', the set isinverted: it matches any single character *not* appearing between thebrackets:

        ''0'' matchesRegex: ''[^01]''   -- false
        ''3'' matchesRegex: ''[^01]'' -- true

For convenience, a set may include ranges: pairs of charactersseparated with ''-''. This is equivalent to listing all charactersbetween them: ''[0-9]'' is the same as ''[0123456789]''.

Special characters within a set are ''^'', ''-'', and '']'' that closes theset. Below are the examples of how to literally use them in a set:

        [01^] -- put the caret anywhere except the beginning
        [01-] -- put the dash as the last character
        []01] -- put the closing bracket as the first character
        [^]01] (thus, empty and universal sets cannot be specified)

Regular expressions can also include the following backquote escapesto refer to popular classes of characters:

        \w any word constituent character (same as [a-zA-Z0-9_])
        \W any character but a word constituent
        \d a digit (same as [0-9])
        \D anything but a digit
        \s a whitespace character (same as [:space:] below)
        \S anything but a whitespace character

These escapes are also allowed in character classes: ''[\w+-]'' means''any character that is either a word constituent, or a plus, or aminus''.

Character classes can also include the following grep(1)-compatibleelements to refer to:

        [:alnum:] any alphanumeric character (same as [a-zA-Z0-9])
        [:alpha:] any alphabetic character (same as [a-zA-Z])
        [:cntrl:] any control character. (any character with code < 32)
        [:digit:] any decimal digit (same as [0-9])
        [:graph:] any graphical character. (any character with code >= 32).
        [:lower:] any lowercase character (including non-ASCII lowercase characters)
        [:print:] any printable character. In this version, this is the same as [:graph:]
        [:punct:] any punctuation character:  . , !!!!!!!! ? ; : '' - ( ) '' and double quotes
        [:space:] any whitespace character (space, tab, CR, LF, null, form feed, Ctrl-Z, 16r2000-16r200B, 16r3000)
        [:upper:] any uppercase character (including non-ASCII uppercase characters)
        [:xdigit:] any hexadecimal character (same as [a-fA-F0-9]).

Note that many of these are only as consistent or inconsistent on issuesof locale as the underlying Smalltalk implementation. Values shown hereare for VisualWorks 7.6.

Note that these elements are components of the character classes,i.e. they have to be enclosed in an extra set of square brackets toform a valid regular expression.  For example, a non-empty string ofdigits would be represented as ''[[:digit:]]+''.

The above primitive expressions and operators are common to manyimplementations of regular expressions. The next primitive expressionis unique to this Smalltalk implementation.

A sequence of characters between colons is treated as a unary selectorwhich is supposed to be understood by Characters. A character matchessuch an expression if it answers true to a message with thatselector. This allows a more readable and efficient way of specifyingcharacter classes. For example, ''[0-9]'' is equivalent to '':isDigit:'',but the latter is more efficient. Analogously to character sets,character classes can be negated: '':^isDigit:'' matches a Characterthat answers false to #isDigit, and is therefore equivalent to''[^0-9]''.

As an example, so far we have seen the following equivalent ways towrite a regular expression that matches a non-empty string of digits:

        ''[0-9]+''
        ''\d+''
        ''[\d]+''
        ''[[:digit:]]+''
        '':isDigit:+''

The last group of special primitive expressions includes:

        . matching any character except a NULL;
        ^ matching an empty string at the beginning of a line;
        $ matching an empty string at the end of a line.
        \b an empty string at a word boundary
        \B an empty string not at a word boundary
        \< an empty string at the beginning of a word
        \> an empty string at the end of a word

        ''axyzb'' matchesRegex: ''a.+b'' -- true
        ''ax zb'' matchesRegex: ''a.+b'' -- true (space is matched by ''.'')
        ''axzb'' matchesRegex: ''a.+b'' -- true (carriage return is matched by ''.'')

Again, the dot ., caret ^ and dollar $ characters are special and should be quotedto be matched literally.!!
]style[(179 21 7851),f5,!!' readStream nextChunkText)
                        key: #syntax!

----- Method: RegexHelp class>>usage (in category 'as yet unclassified') -----
usage
        "This method was automatically generated. Edit it using:"
        "RegexHelp edit: #usage"
        ^(HelpTopic
                title: 'Usage'
                contents:
'The preceding section covered the syntax of regular expressions. It used the simplest possible interface to the matcher: sending #matchesRegex: message to the sample string, with regular expression string as the argument.  This section explains hairier ways of using the matcher.

       
        PREFIX MATCHING AND CASE-INSENSITIVE MATCHING

A CharacterArray (an EsString in VA) also understands these messages:

        #prefixMatchesRegex: regexString
        #matchesRegexIgnoringCase: regexString
        #prefixMatchesRegexIgnoringCase: regexString

#prefixMatchesRegex: is just like #matchesRegex, except that the whole receiver is not expected to match the regular expression passed as the argument; matching just a prefix of it is enough.  For example:

        ''abcde'' matchesRegex: ''(a|b)+'' -- false
        ''abcde'' prefixMatchesRegex: ''(a|b)+'' -- true

The last two messages are case-insensitive versions of matching.

       
        ENUMERATION INTERFACE

An application can be interested in all matches of a certain regular expression within a String.  The matches are accessible using a protocol modelled after the familiar Collection-like enumeration protocol:

        #regex: regexString matchesDo: aBlock

Evaluates a one-argument <aBlock> for every match of the regular expression within the receiver string.

        #regex: regexString matchesCollect: aBlock

Evaluates a one-argument <aBlock> for every match of the regular expression within the receiver string. Collects results of evaluations and anwers them as a SequenceableCollection.

        #allRegexMatches: regexString

Returns a collection of all matches (substrings of the receiver string) of the regular expression.  It is an equivalent of <aString regex: regexString matchesCollect: [:each | each]>.

        #allRangesOfRegexMatches: regexString

Returns a collection of all character ranges (startIndex to: stopIndex) that match the regular expression.

       
        REPLACEMENT AND TRANSLATION

It is possible to replace all matches of a regular expression with a
certain string using the message:

        #copyWithRegex: regexString matchesReplacedWith: aString

For example:

        ''ab cd ab'' copyWithRegex: ''(a|b)+'' matchesReplacedWith: ''foo''

A more general substitution is match translation:

        #copyWithRegex: regexString matchesTranslatedUsing: aBlock

This message evaluates a block passing it each match of the regular expression in the receiver string and answers a copy of the receiver with the block results spliced into it in place of the respective matches.  For example:

        ''ab cd ab'' copyWithRegex: ''(a|b)+'' matchesTranslatedUsing: [:each | each asUppercase]

All messages of enumeration and replacement protocols perform a case-sensitive match.  Case-insensitive versions are not provided as part of a CharacterArray protocol.  Instead, they are accessible using the lower-level matching interface.


        LOWER-LEVEL INTERFACE

Internally, #matchesRegex: works as follows:

1. A fresh instance of RxParser is created, and the regular expression string is passed to it, yielding the expression''s syntax tree.

2. The syntax tree is passed as an initialization parameter to an instance of RxMatcher. The instance sets up some data structure that will work as a recognizer for the regular expression described by the tree.

3. The original string is passed to the matcher, and the matcher checks for a match.


        THE MATCHER

If you repeatedly match a number of strings against the same regular expression using one of the messages defined in CharacterArray, the regular expression string is parsed and a matcher is created anew for every match.  You can avoid this overhead by building a matcher for the regular expression, and then reusing the matcher over and over again. You can, for example, create a matcher at a class or instance initialization stage, and store it in a variable for future use.

You can create a matcher using one of the following methods:

        - Sending #forString:ignoreCase: message to RxMatcher class, with the regular expression string and a Boolean indicating whether case is ignored as arguments.

        - Sending #forString: message.  It is equivalent to <... forString: regexString ignoreCase: false>.

A more convenient way is using one of the two matcher-created messages understood by CharacterArray.

        - <regexString asRegex> is equivalent to <RxMatcher forString: regexString>.

        - <regexString asRegexIgnoringCase> is equivalent to <RxMatcher forString: regexString ignoreCase: true>.

Here are four examples of creating a matcher:

        hexRecognizer := RxMatcher forString: ''16r[0-9A-Fa-f]+''
        hexRecognizer := RxMatcher forString: ''16r[0-9A-Fa-f]+'' ignoreCase: false
        hexRecognizer := ''16r[0-9A-Fa-f]+'' asRegex
        hexRecognizer := ''16r[0-9A-F]+'' asRegexIgnoringCase


        MATCHING

The matcher understands these messages (all of them return true to indicate successful match or search, and false otherwise):

matches: aString

        True if the whole target string (aString) matches.

matchesPrefix: aString

        True if some prefix of the string (not necessarily the whole
        string) matches.

search: aString

        Search the string for the first occurrence of a matching
        substring. (Note that the first two methods only try matching from
        the very beginning of the string). Using the above example with a
        matcher for `a+'', this method would answer success given a string
        `baaa'', while the previous two would fail.

matchesStream: aStream
matchesStreamPrefix: aStream
searchStream: aStream

        Respective analogs of the first three methods, taking input from a
        stream instead of a string. The stream must be positionable and
        peekable.

All these methods answer a boolean indicating success. The matcher
also stores the outcome of the last match attempt and can report it:

lastResult

        Answers a Boolean -- the outcome of the most recent match
        attempt. If no matches were attempted, the answer is unspecified.


        SUBEXPRESSION MATCHES

After a successful match attempt, you can query the specifics of which part of the original string has matched which part of the whole expression.

A subexpression is a parenthesized part of a regular expression, or the whole expression. When a regular expression is compiled, its subexpressions are assigned indices starting from 1, depth-first, left-to-right. For example, `((ab)+(c|d))?ef'' includes the following subexpressions with these indices:

        1: ((ab)+(c|d))?ef
        2: (ab)+(c|d)
        3: ab
        4: c|d

After a successful match, the matcher can report what part of the original string matched what subexpression. It understandards these messages:

subexpressionCount

        Answers the total number of subexpressions: the highest value that
        can be used as a subexpression index with this matcher. This value
        is available immediately after initialization and never changes.

subexpression: anIndex

        An index must be a valid subexpression index, and this message
        must be sent only after a successful match attempt. The method
        answers a substring of the original string the corresponding
        subexpression has matched to.

subBeginning: anIndex
subEnd: anIndex

        Answer positions within the original string or stream where the
        match of a subexpression with the given index has started and
        ended, respectively.

This facility provides a convenient way of extracting parts of input strings of complex format. For example, the following piece of code uses the ''MMM DD, YYYY'' date format recognizer example from the `Syntax'' section to convert a date to a three-element array with year, month, and day strings (you can select and evaluate it right here):

        | matcher |
        matcher := RxMatcher forString: ''(Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec)[ ]+(:isDigit::isDigit:?)[ ]*,[ ]*(19|20)(:isDigit::isDigit:)''.
        (matcher matches: ''Aug 6, 1996'')
                ifTrue:
                        [Array
                                with: (matcher subexpression: 5)
                                with: (matcher subexpression: 2)
                                with: (matcher subexpression: 3)]
                ifFalse: [''no match'']

(should answer ` #(''96'' ''Aug'' ''6'')'').


        ENUMERATION AND REPLACEMENT

The enumeration and replacement protocols exposed in CharacterArray are actually implemented by the matcher.  The following messages are understood:

        #matchesIn: aString
        #matchesIn: aString do: aBlock
        #matchesIn: aString collect: aBlock
        #copy: aString replacingMatchesWith: replacementString
        #copy: aString translatingMatchesUsing: aBlock
        #matchingRangesIn: aString

        #matchesOnStream: aStream
        #matchesOnStream: aStream do: aBlock
        #matchesOnStream: aStream collect: aBlock
        #copy: sourceStream to: targetStream replacingMatchesWith: replacementString
        #copy: sourceStream to: targetStream translatingMatchesWith: aBlock

Note that in those methods that take a block, the block may refer to the rxMatcher itself, e.g. to collect information about the position the match occurred at, or the subexpressions of the match. An example can be seen in #matchingRangesIn:


        ERROR HANDLING

Exception signaling objects (Signals in VisualWorks, Exceptions in VisualAge) are accessible through RxParser class protocol. To handle possible errors, use the protocol described below to obtain the exception objects and use the protocol of the native Smalltalk implementation to handle them.

If a syntax error is detected while parsing expression, RxParser>>syntaxErrorSignal is raised/signaled.

If an error is detected while building a matcher, RxParser>>compilationErrorSignal is raised/signaled.

If an error is detected while matching (for example, if a bad selector was specified using `:<selector>:'' syntax, or because of the matcher''s internal error), RxParser>>matchErrorSignal is raised

RxParser>>regexErrorSignal is the parent of all three.  Since any of the three signals can be raised within a call to #matchesRegex:, it is handy if you want to catch them all.  For example:

VisualWorks:

        RxParser regexErrorSignal
                handle: [:ex | ex returnWith: nil]
                do: [''abc'' matchesRegex: ''))garbage['']

VisualAge:

        [''abc'' matchesRegex: ''))garbage['']
                when: RxParser regexErrorSignal
                do: [:signal | signal exitWith: nil]!!' readStream nextChunkText)
                        key: #usage!