path: root/ps01_grep/ps.txt

                MASSACHVSETTS INSTITVTE OF TECHNOLOGY
      Department of Electrical Engineering and Computer Science

                          6.945 Spring 2009
                            Problem Set 1

  Issued: Wed. 4 Feb. 2009                    Due: Wed. 11 Feb. 2009


General instructions for problem sets:
   A problem set for this class generally asks you to do some
programming.  We usually give you a program to read and extend.  You
should turn in your extensions, in the form of clearly annotated code
and executions that demonstrate its effectiveness.  We may also ask
for short essays explaining an idea.  Your answers to these questions
must be clear and concise: good logic expressed in good English is
required.  Thank you.

Readings: 
    Review SICP chapter 1 (especially section 1.3)
    MIT/GNU Scheme documentation, section 5.6 (character sets)
    Debian GNU/Linux info on regular expressions 
      from the grep man page (attached).  This is sane.

Code: 
    regexp.scm, tests.txt (both attached)
    Windows grep: http://gnuwin32.sourceforge.net/packages/grep.htm

Documentation:
 The MIT/GNU Scheme installation and documentation can
    be found online at http://www.gnu.org/software/mit-scheme/
    The reference manual is in:
 http://www.gnu.org/software/mit-scheme/documentation/mit-scheme-ref/

    The (Insane) POSIX manual page for regular expressions:
 http://www.opengroup.org/onlinepubs/009695399/basedefs/xbd_chap09.html


                         Regular Expressions

Regular expressions are ubiquitous.  But the syntax of the regular-
expression language is awful.  There are various incompatable forms of
the language and the quotation conventions are baroquen [sic].
Nevertheless, there is a great deal of useful software, for example
grep, that uses regular expressions to specify the desired behavior.

Part of the value of this problem set is to experience how bad things
can be.  Although regular expression systems are derived from a
perfectly good mathematical formalism, the particular choices made by
implementers to expand the formalism into useful software systems are
often disastrous: the quotation conventions adopted are highly
irregular; the particularly egregious misuse of parenthesis, both for
grouping and for backward reference, is a miracle to behold.  In
addition, attempts to increase the expressive power and address
shortcomings of earlier designs have led to a proliferation of
incompatible derivative languages.

In this problem set we will invent both a Lisp-friendly language for
specifying regular expressions and a means of translating this
language to conventional regular-expression syntax.  The application
is to be able to use the capabilities of systems like grep from inside
the Scheme environment.

As with any language there are primitives, means of combination, and
means of abstraction.  Our language allows the construction of
patterns that utilities like grep can match against character-string
data.  Because this language is embedded in Scheme we inherit all of
the power of Scheme: we can use Scheme constructs to combine patterns
and Scheme procedures to abstract them.

Patterns are built out of primitive patterns.
The primitive patterns are:

 (r:dot)
     matches any character except newline

 (r:bol)
     matches only the beginning of a line

 (r:eol)
     matches only the end of a line

 (r:quote <string>)
     matches the given string

 (r:char-from <char-set>)
     matches one character from the specified 
     MIT/GNU Scheme character set

 (r:char-not-from <char-set>)
     matches one character not from the specified 
     MIT/GNU Scheme character set

     MIT/GNU Scheme provides a variety of character sets,
     but you can make your own with the Scheme procedures
     char-set and string->char-set.

Patterns can be combined to make compound patterns:

 (r:seq <pattern> ...)
     This pattern matches each of the argument patterns in sequence,
     from left to right

 (r:alt <pattern> ...)  
     This pattern tries each argument pattern from left to right,
     until one of these alternatives matches.  If none matches then
     this pattern does not match.

 (r:repeat <min> <max> <pattern>)
     This pattern tries to match the given argument pattern a minimum
     of min times but no more than a maximum of max times.  If max is
     given as #f then there is no maximum specified.  Note that if
     max=min the given pattern must be matched exactly that many
     times.

Because these are all Scheme procedures (in the file regexp.scm) you
can freely mix these with any Scheme code.

Here are some examples:

  Pattern:  (r:seq (r:quote "a") (r:dot) (r:quote "c"))
  Matches:  "abc" and "aac" and "acc" 

  Pattern:  (r:alt (r:quote "foo") (r:quote "bar") (r:quote "baz"))
  Matches:  "foo" and "bar" and "baz" 

  Pattern:  (r:repeat 3 5 (r:alt (r:quote "cat") (r:quote "dog")))
  Matches:  "catdogcat" and "catcatdogdog" and "dogdogcatdogdog"
            but not "catcatcatdogdogdog"

  Pattern:  (let ((digit 
                   (r:char-from (string->char-set "0123456789"))))
              (r:seq (r:bol)
                     (r:quote "[")
                     digit
                     digit
                     (r:quote "]")
                     (r:quote ".")
                     (r:quote " ")
                     (r:char-from (char-set #\a #\b))
                     (r:repeat 3 5 (r:alt (r:quote "cat") (r:quote "dog")))
                     (r:char-not-from (char-set #\d #\e #\f))
                     (r:eol)))
  Matches:  "[09]. acatdogdogcats" but not
            "[10]. ifacatdogdogs" nor
            "[11]. acatdogdogsme"

In the file regexp.scm we define an interface to the grep utility,
which allows a Scheme program to call grep with a regular expression
(constructed from the given pattern) on a given file name.  The grep
utility extracts lines that contain a substring that can match the
given pattern.

So, for example: given the test file test.txt (supplied) we can find 
the lines in the file that contain a match to the given pattern.

(pp
 (r:grep (r:seq " "
                (r:repeat 3 5 (r:alt (r:quote "cat") (r:quote "dog")))
                (r:eol)) 
         "tests.txt"))
("[09]. catdogcat" "[10]. catcatdogdog" "[11]. dogdogcatdogdog")
;Unspecified return value

Note that the pretty-printer (pp) returns an unspecified value, after
printing the list of lines that grep found matching the pattern
specified.

-------------
Problem 1.1: Warmup

In the traditional regular expression language the asterisk (*)
operator following a subpattern means zero-or-more copies of the
subpattern and the plus-sign (+) operator following a subpattern means
one-or-more copies of the subpattern.  Define Scheme procedures r:*
and r:+ to take a pattern and iterate it as necessary.  This can be
done in terms of r:repeat.

Demonstrate your procedures on real data in complex patterns.
-------------



     A Subtle Bug, One Bad Joke, Two Tweaks, and a Revelation

Ben Bitdiddle has noticed a problem with our implementation of
(r:repeat <min> <max> <pattern>).

The use of (r:seq expr "\\|") at the end of the r:repeat procedure is
a bit dodgy.  This code fragment compiles to an ERE Alternation
regular expression of the form (expr|).  (See 9.4.7 of the POSIX
regular expression document.)

This relies on the fact that alternation with something and nothing is
the equivalent of saying "one or none".  That is: (expr|) denotes one
or no instances of expr.  Unfortunately, this depends on an
undocumented GNU extension to the formal POSIX standard for REs.

Specifically, section 9.4.3 states that a vertical line appearing
immediately before a close parenthesis (or immediately after an open
parenthesis) produces undefined behavior.  In essence, an RE must not
be a null sequence.

GNU grep just happens to Do The Right Thing (tm) when presented with
(x|).  Not all grep implementations are as tolerant.

Therefore, Ben asks his team of three code hackers (Louis, Alyssa and
his niece Bonnie) to propose alternative workarounds.  Ultimately, he
proposes his own patch, which you will implement.


 * Louis Reasoner suggests that a simple, elegant fix would be to
   replace the code fragment (r:seq expr "\\|") with a straight-
   forward call to (r:repeat 0 1 expr).


 * Alyssa P. Hacker proposes that an alternative fix would be to
   rewrite the else clause of r:repeat to compile (r:repeat 3 5 <x>)
   into the equivalent of (xxx|xxxx|xxxxx) instead of the naughty
   xxx(x|)(x|) non-POSIX-compliant undefined regular expression.
   She refers to section 9.4.7 of the POSIX regular expression doc.


 * Bonnie Bitdiddle points to the question mark (?) operator in
   section 9.4.6.4 and proposes that a better fix would be to
   implement an r:? operator then replace (r:seq expr "\\|")
   with (r:? expr).


 * Meanwhile, Ben looks closely at the RE spec and has a revelation.
   He proposes that r:repeat be re-implemented to emit Interval
   expressions.  See section 9.3.6.5 of the POSIX documentation.
   Please try not to get sick.

-------------
Problem 1.2: The Proposals

Let's very briefly consider each proposal:

a. Everyone in the room immediately giggles at Louis' silly joke.
   What's so funny about it?  That is, what's wrong with this idea?

   A one-sentence punchline will do.

b. What advantages does Bonnie's proposal have over Alyssa's
   in terms of both code and data?

   A brief, concise yet convincing few sentences suffices.

c. What advantage does Ben's proposal have over all the others?
   Specifically, ponder which section of the POSIX document he cites
   versus which sections the others cite, then take a quick peek at
   Problem 1.5 below and consider the implications.  Also, consider
   the size of the output strings in this new code as well as the
   overall clarity of the code.

   Again, a brief sentence or two is sufficient.

d. Following Ben's proposal, re-implement r:repeat to emit Interval
   expressions.  Hint: Scheme's number->string procedure should be
   handy.  Caveat: beware the backslashes.

   Show the output it generates on a few well-chosen sample inputs.
   Demonstrate your procedure on real data in complex patterns.
-------------


                           Too Much Nesting

Our program produces excessively nested regular expressions: it makes
groups even when they are not necessary.  For example, the following
simple pattern leads to an overly complex regular expression:  

   (display (r:seq (r:quote "a") (r:dot) (r:quote "c")))
   \(\(a\).\(c\)\)

Another problem is that BREs may involve back-references.  (See
9.3.6.3 of the POSIX regular expression documentation.)  A
back-reference refers to a previously parenthesized subexpression.  So
it is important that the parenthesized subexpressions be ones
explicitly placed by the author of the pattern.  (Aargh! This is one
of the worst ideas I (GJS) have ever heard of -- grouping, which is
necessary for iteration, was confused with naming for later reference.
What a crock!)

-------------
Problem 1.3: Optimization

Edit our program to eliminate as much of the unnecessary nesting as
you can.  Caution: there are subtle cases here that you have to watch
out for.  What is such a case?  Demonstrate your better version of our
program and show how it handles the subtleties.

Hint: Our program uses strings as its intermediate representation as
well as its result.  You might consider using a different intermediate
representation.
-------------

-------------
Problem 1.4: Back-references

Add in a procedure for constructing back-references.
Have fun getting confused about BREs.
-------------

                              Standards?

                The best thing about standards is that 
                there are so many to choose from.
                              Tom Knight

There are also Extended Regular Expressions (EREs) defined in the
POSIX regular expression documentation.  Some software, such as egrep,
uses this version of regular expressions.  Unfortunately EREs are not
a conservative extension of BREs: ERE syntax is actually inconsistent
with BRE syntax!  It is an interesting project to extend our Scheme
pattern language so that the target can be either BREs or EREs.

-------------
Problem 1.5: Ugh!

  a. What are the significant differences between BREs and EREs that
  make this a pain?  List the differences that must be addressed.

  b. How can the back end be factored so that our language can compile
  into either kind of regular expression, depending on what is needed?
  How can we maintain the abstract layer that is independent of the 
  target regular expression language?  Explain your strategy.

  c. Extend our implementation to have both back ends.

Demonstrate your work by making sure that you can run egrep as well as
grep, with equivalent results in cases that test the differences you
found in part a.
-------------

           End of Problem Set.  Reference Material Follows.
-----------------------------------------------------------------------

The following is an excerpt from the Debian GNU/Linux man page on grep.

REGULAR EXPRESSIONS
    A  regular  expression  is a pattern that describes a set of strings.
    Regular expressions are constructed analogously to arithmetic expres-
    sions, by using various operators to combine smaller expressions.

    Grep  understands three different versions of regular expression syn-
    tax: "basic," "extended," and "perl."  In GNU grep, there is no  dif-
    ference in available functionality using either of the first two syn-
    taxes.  In other implementations, basic regular expressions are  less
    powerful.   The  following  description  applies  to extended regular
    expressions; differences for basic regular expressions are summarized
    afterwards.   Perl  regular expressions add additional functionality,
    but the implementation used here is undocumented and is not  compati-
    ble with other grep implementations.

    The  fundamental  building  blocks  are  the regular expressions that
    match a single character.  Most characters, including all letters and
    digits, are regular expressions that match themselves.  Any metachar-
    acter with special meaning may be quoted by preceding it with a back-
    slash.

    A bracket expression is a list of characters enclosed by [ and ].  It
    matches any single character in that list; if the first character  of
    the  list  is  the  caret  ^ then it matches any character not in the
    list.  For example, the regular expression [0123456789]  matches  any
    single digit.

    Within a bracket expression, a range expression consists of two char-
    acters separated by a hyphen.  It matches any single  character  that
    sorts  between the two characters, inclusive, using the locale's col-
    lating sequence and character set.  For example,  in  the  default  C
    locale,  [a-d] is equivalent to [abcd].  Many locales sort characters
    in dictionary order, and in these  locales  [a-d]  is  typically  not
    equivalent  to [abcd]; it might be equivalent to [aBbCcDd], for exam-
    ple.  To obtain the traditional  interpretation  of  bracket  expres-
    sions,  you  can  use  the C locale by setting the LC_ALL environment
    variable to the value C.

    Finally, certain named classes of characters  are  predefined  within
    bracket  expressions,  as follows.  Their names are self explanatory,
    and they are [:alnum:], [:alpha:], [:cntrl:],  [:digit:],  [:graph:],
    [:lower:],    [:print:],   [:punct:],   [:space:],   [:upper:],   and
    [:xdigit:].  For example, [[:alnum:]] means [0-9A-Za-z],  except  the
    latter  form depends upon the C locale and the ASCII character encod-
    ing, whereas the former is independent of locale and  character  set.
    (Note that the brackets in these class names are part of the symbolic
    names, and must be included in addition to  the  brackets  delimiting
    the  bracket  list.)   Most metacharacters lose their special meaning
    inside lists.  To include a literal ] place it  first  in  the  list.
    Similarly,  to  include  a  literal  ^  place  it anywhere but first.
    Finally, to include a literal - place it last.

     The period .  matches any single character.  The symbol \w is a  syn-
     onym for [[:alnum:]] and \W is a synonym for [^[:alnum]].

     The  caret  ^  and  the dollar sign $ are metacharacters that respec-
     tively match the empty string at the beginning and  end  of  a  line.
     The  symbols  \<  and  \>  respectively match the empty string at the
     beginning and end of a word.  The symbol \b matches the empty  string
     at  the edge of a word, and \B matches the empty string provided it's
     not at the edge of a word.

     A regular expression may be followed by  one  of  several  repetition
     operators:
     ?      The preceding item is optional and matched at most once.
     *      The preceding item will be matched zero or more times.
     +      The preceding item will be matched one or more times.
     {n}    The preceding item is matched exactly n times.
     {n,}   The preceding item is matched n or more times.
     {n,m}  The  preceding  item is matched at least n times, but not more
            than m times.

     Two regular expressions may be concatenated;  the  resulting  regular
     expression  matches any string formed by concatenating two substrings
     that respectively match the concatenated subexpressions.

     Two regular expressions may be joined by the infix  operator  |;  the
     resulting  regular  expression  matches  any  string  matching either
     subexpression.

     Repetition takes precedence over concatenation, which in  turn  takes
     precedence  over  alternation.  A whole subexpression may be enclosed
     in parentheses to override these precedence rules.

     The backreference \n, where n is a single  digit,  matches  the  sub-
     string  previously  matched by the nth parenthesized subexpression of
     the regular expression.

     In basic regular expressions the metacharacters ?, +, {, |, (, and  )
     lose  their special meaning; instead use the backslashed versions \?,
     \+, \{, \|, \(, and \).

     Traditional egrep did not support the { metacharacter, and some egrep
     implementations  support \{ instead, so portable scripts should avoid
     { in egrep patterns and should use [{] to match a literal {.

     GNU egrep attempts to support traditional usage by assuming that { is
     not  special if it would be the start of an invalid interval specifi-
     cation.  For example, the shell command egrep '{1' searches  for  the
     two-character  string  {1  instead of reporting a syntax error in the
     regular expression.  POSIX.2 allows this behavior  as  an  extension,
     but portable scripts should avoid it.

GNU Project                      2002/01/22                          GREP(1)

;;;; Scheme Regular Expression Language Implementation -- regexp.scm

(define (r:dot) ".")

(define (r:bol) "^")

(define (r:eol) "$")

(define (r:quote string)
  (r:seq
   (call-with-output-string		; see RefMan section 14.3
    (lambda (port)
      (let ((end (string-length string)))
        (do ((i 0 (+ i 1))) ((not (< i end))) ; see RefMan 2.9
          (let ((c (string-ref string i)))
            (if (or (char=? c #\.)
                    (char=? c #\[)
                    (char=? c #\\)
                    (char=? c #\^)
                    (char=? c #\$)
		    (char=? c #\*))
                (write-char #\\ port))
            (write-char c port))))))))

(define (r:char-from char-set)		; see RefMan section 5.6
  (let ((members (char-set-members char-set)))
    (cond ((not (pair? members))
	   (r:seq))
          ((not (pair? (cdr members)))
	   (r:quote (string (car members))))
          (else
           (%char-from #f members)))))

(define (r:char-not-from char-set)
  (%char-from #t (char-set-members char-set)))

(define (%char-from negate? members)
  (let ((right? (memv #\] members))
        (caret? (memv #\^ members))
        (hyphen? (memv #\- members))
        (others
         (delete-matching-items members
                                (lambda (c)
                                  (or (char=? c #\])
                                      (char=? c #\^)
                                      (char=? c #\-))))))
    (if (and caret?
             hyphen?
             (not right?)
             (not negate?)
             (null? others))
        "[-^]"
        (string-append "["
                       (if negate? "^" "")
                       (if right? "]" "")
                       (list->string others)
                       (if caret? "^" "")
                       (if hyphen? "-" "")
                       "]"))))

;;; Means of combination for patterns

(define (r:seq . exprs)
  (string-append "\\(" (apply string-append exprs) "\\)"))

(define (r:alt . exprs)
  (if (pair? exprs)
      (apply r:seq
             (cons (car exprs)
                   (append-map (lambda (expr)
                                 (list "\\|" expr))
                               (cdr exprs))))
      (r:seq)))

(define (r:repeat min max expr)
  (if (not (exact-nonnegative-integer? min))
      (error "Min must be non-negative integer:" min))
  (if max
      (begin
        (if (not (exact-nonnegative-integer? max))
            (error "Max must be non-negative integer:" max))
        (if (not (<= min max))
            (error "Min not less than max:" min max))))
  (cond ((not max)
	 (apply r:seq
		(append (make-list min expr)
			(list expr "*"))))
	((= max min)
	 (apply r:seq (make-list min expr)))
	(else
	 (apply r:seq
		(append (make-list min expr)
			(make-list (- max min)
				   (r:seq expr "\\|")))))))

;;; The following magic allows a program in MIT/GNU Scheme to call the
;;; grep system utility, returning the list of grep output lines to
;;; the caller.  You can make similar mechanisms to call other system
;;; utilities.

(load-option 'synchronous-subprocess)

(define (r:grep expr filename)
  (r:grep-like "grep" '() expr filename))

(define (r:egrep expr filename)
  (if (eq? microcode-id/operating-system 'nt)
      (r:grep-like "grep" '("-E") expr filename)
      (r:grep-like "egrep" '() expr filename)))

(define (r:grep-like program options expr filename)
  (let ((port (open-output-string)))
    (and (= (run-synchronous-subprocess program
	      (append options
		      (list "-e" expr (->namestring filename)))
	      'output port)
            0)
	 (r:split-lines (get-output-string port)))))

(define (r:split-lines string)
  (reverse
   (let ((end (string-length string)))
     (let loop ((i 0) (lines '()))
       (if (< i end)
	   (let ((j
		  (substring-find-next-char string i end #\newline)))
	     (if j
		 (loop (+ j 1)
		       (cons (substring string i j) lines))
		 (cons (substring string i end) lines)))
	   lines)))))

#|
;;; An alternate implementation using MIT/GNU Scheme's internal
;;; regular-expression interpreter.

(define (r:grep expr filename)
  (call-with-input-file filename
    (lambda (port)
      (let loop ((lines '()))
	(let ((line (read-line port)))
	  (if (eof-object? line)
	      (reverse lines)
	      (loop (if (re-string-search-forward expr line #f)
			(cons line lines)
			lines))))))))
|#

#|
;;; For example...

;;; Note, the result of the next two requests were not in this file
;;; when the requests were made!

(pp (r:grep (r:quote "r:sex") "regexp.scm"))
("(pp (r:grep (r:quote \"r:sex\") \"regexp.scm\"))")
;Unspecified return value

(pp (r:grep (r:quote "r:seq") "regexp.scm"))
("  (r:seq"
 "\t   (r:seq))"
 "(define (r:seq . exprs)"
 "      (apply r:seq"
 "      (r:seq)))"
 "\t (apply r:seq"
 "\t (apply r:seq (make-list min expr)))"
 "\t (apply r:seq"
 "\t\t\t\t   (r:seq expr \"\\\\|\")))))))"
 "(pp (r:grep (r:quote \"r:seq\") \"regexp.scm\"))"
 "(pp (r:grep (r:seq (r:quote \"a\") (r:dot) (r:quote \"c\")) \"tests.txt\"))"
 " (r:grep (r:seq \" \""
 "    (r:seq (r:bol)")
;Unspecified return value

(pp (r:grep (r:seq (r:quote "a") (r:dot) (r:quote "c")) "tests.txt"))
("[00]. abc"
 "[01]. aac"
 "[02]. acc"
 "[03]. zzzaxcqqq"
 "[10]. catcatdogdog"
 "[12]. catcatcatdogdogdog")
;Unspecified return value

;;; And...

(pp (r:grep (r:alt (r:quote "foo") (r:quote "bar") (r:quote "baz"))
	    "tests.txt"))
("[05]. foo" "[06]. bar" "[07]. foo bar baz quux")
;Unspecified return value


(pp (r:grep (r:repeat 3 5 (r:alt (r:quote "cat") (r:quote "dog")))
	    "tests.txt"))
("[09]. catdogcat"
 "[10]. catcatdogdog"
 "[11]. dogdogcatdogdog"
 "[12]. catcatcatdogdogdog"
 "[13]. acatdogdogcats"
 "[14]. ifacatdogdogs"
 "[15]. acatdogdogsme")
;Unspecified return value

(pp
 (r:grep (r:seq " "
		(r:repeat 3 5 (r:alt (r:quote "cat") (r:quote "dog")))
		(r:eol)) 
         "tests.txt"))
("[09]. catdogcat" "[10]. catcatdogdog" "[11]. dogdogcatdogdog")
;Unspecified return value

(pp
 (r:grep
  (let ((digit 
	 (r:char-from (string->char-set "0123456789"))))
    (r:seq (r:bol)
	   (r:quote "[")
	   digit
	   digit
	   (r:quote "]")
	   (r:quote ".")
	   (r:quote " ")
	   (r:char-from (char-set #\a #\b))
	   (r:repeat 3 5 (r:alt "cat" "dog"))
	   (r:char-not-from (char-set #\d #\e #\f))
	   (r:eol)))
  "tests.txt"))
("[13]. acatdogdogcats")
;Unspecified return value
|#

;;; This is the file tests.txt

[00]. abc
[01]. aac
[02]. acc
[03]. zzzaxcqqq
[04]. abdabec

[05]. foo
[06]. bar
[07]. foo bar baz quux
[08]. anything containing them

[09]. catdogcat
[10]. catcatdogdog
[11]. dogdogcatdogdog
[12]. catcatcatdogdogdog

[13]. acatdogdogcats
[14]. ifacatdogdogs
[15]. acatdogdogsme