;;; @Package     csv
;;; @Subtitle    Comma-Separated Value (CSV) Utilities in Scheme
;;; @HomePage    http://www.neilvandyke.org/csv-scheme/
;;; @Author      Neil Van Dyke
;;; @Version     0.9
;;; @Date        2009-03-14
;;; @PLaneT      neil/csv:1:5

;; $Id: csv.ss,v 1.196 2009/03/14 07:28:03 neilpair Exp $

;;; @legal
;;; Copyright @copyright{} 2004--2009 Neil Van Dyke.  This program is Free
;;; Software; you can redistribute it and/or modify it under the terms of the
;;; GNU Lesser General Public License as published by the Free Software
;;; Foundation; either version 3 of the License (LGPL 3), or (at your option)
;;; any later version.  This program is distributed in the hope that it will be
;;; useful, but without any warranty; without even the implied warranty of
;;; merchantability or fitness for a particular purpose.  See
;;; @indicateurl{http://www.gnu.org/licenses/} for details.  For other licenses
;;; and consulting, please contact the author.
;;; @end legal

#lang scheme/base

;;; @section Introduction

;;; The @b{csv} Scheme library provides utilities for reading various kinds of
;;; what are commonly known as ``comma-separated value'' (CSV) files.  Since
;;; there is no standard CSV format, this library permits CSV readers to be
;;; constructed from a specification of the peculiarities of a given variant.
;;; A default reader handles the majority of formats.
;;; One of the main uses of this library is to import data from old crusty
;;; legacy applications into Scheme for data conversion and other processing.
;;; To that end, this library includes various conveniences for iterating over
;;; parsed CSV rows, and for converting CSV input to the
;;; @uref{http://pobox.com/~oleg/ftp/Scheme/SXML.html, SXML 3.0} Scheme XML
;;; format.
;;; This library requires R5RS, SRFI-6, SRFI-23, and an @code{integer->char}
;;; procedure that accepts ASCII values.
;;; Other implementations of some kind of CSV reading for Scheme include
;;; Gauche's @code{text.csv} module, and Scsh's @code{record-reader} and
;;; related procedures.  This library intends to be portable and more
;;; comprehensive.

;; TODO: Briefly introduce terms "row", "column", and "field".

(define-syntax %csv:error
  (syntax-rules () ((_ p m o)
                    (error (string-append p " : " m) o)
                    ;; Bigloo: (error p m o)

(define-syntax %csv:type-error
  (syntax-rules ()
    ((_ proc-str expected-str got-value)
     (%csv:error proc-str
                 (string-append "expected " expected-str ", received:")

(define %csv:a2c integer->char)

(define %csv:cr (%csv:a2c 13))
(define %csv:lf (%csv:a2c 10))

(define-syntax %csv:gosc
  (syntax-rules ()
    ((_ os-stx)
     (let* ((os  os-stx)
            (str (get-output-string os)))
       (close-output-port os)

(define (%csv:in-arg proc-name in)
  (cond ((input-port? in) in)
        ((string?     in) (open-input-string in))
        (else (%csv:type-error proc-name "input port or string" in))))

(define (%csv:reader-or-in-arg proc-name reader-or-in)
  (cond ((procedure?  reader-or-in) reader-or-in)
        ((input-port? reader-or-in) (make-csv-reader reader-or-in))
        ((string?     reader-or-in) (make-csv-reader (open-input-string
        (else (%csv:type-error proc-name
                               "csv reader or input port or string"

;;; @section Reader Specs

;;; CSV readers are constructed using @dfn{reader specs}, which are sets of
;;; attribute-value pairs, represented in Scheme as association lists keyed on
;;; symbols.  Each attribute has a default value if not specified otherwise.
;;; The attributes are:

;;; @table @code
;;; @item newline-type
;;; Symbol representing the newline, or record-terminator, convention.  The
;;; convention can be a fixed character sequence (@code{lf}, @code{crlf}, or
;;; @code{cr}, corresponding to combinations of line-feed and carriage-return),
;;; any string of one or more line-feed and carriage-return characters
;;; (@code{lax}), or adaptive (@code{adapt}).  @code{adapt} attempts to detect
;;; the newline convention at the start of the input and assume that convention
;;; for the remainder of the input.  Default: @code{lax}
;;; @item separator-chars
;;; Non-null list of characters that serve as field separators.  Normally, this
;;; will be a list of one character.  Default: @code{(#\,)} (list of the comma
;;; character)
;;; @item quote-char
;;; Character that should be treated as the quoted field delimiter character,
;;; or @code{#f} if fields cannot be quoted.  Note that there can be only one
;;; quote character.  Default: @code{#\"} (double-quote)
;;; @item quote-doubling-escapes?
;;; Boolean for whether or not a sequence of two @code{quote-char} quote
;;; characters within a quoted field constitute an escape sequence for
;;; including a single @code{quote-char} within the string.  Default: @code{#t}
;;; @item comment-chars
;;; List of characters, possibly null, which comment out the entire line of
;;; input when they appear as the first character in a line.  Default:
;;; @code{()} (null list)
;;; @item whitespace-chars
;;; List of characters, possibly null, that are considered @dfn{whitespace}
;;; constituents for purposes of the @code{strip-leading-whitespace?} and
;;; @code{strip-trailing-whitespace?} attributes described below.
;;; Default: @code{(#\space)} (list of the space character)
;;; @item strip-leading-whitespace?
;;; Boolean for whether or not leading whitespace in fields should be
;;; stripped.  Note that whitespace within a quoted field is never stripped.
;;; Default: @code{#f}
;;; @item strip-trailing-whitespace?
;;; Boolean for whether or not trailing whitespace in fields should be
;;; stripped.  Note that whitespace within a quoted field is never stripped.
;;; Default: @code{#f}
;;; @item newlines-in-quotes?
;;; Boolean for whether or not newline sequences are permitted within quoted
;;; fields.  If true, then the newline characters are included as part of the
;;; field value; if false, then the newline sequence is treated as a premature
;;; record termination.  Default: @code{#t}
;;; @end table

;; TODO: Do not expose this procedure for now.  We expect it to go away and be
;; replaced with two other procedures.
;; @defproc %csv:csv-spec-derive orig-spec changes
;; Yields a new CSV spec that is derived from @var{orig-spec} by applying spec
;; @var{changes} as attribute substitions and additions to the original.  For
;; example, given an original CSV reader spec:
;; @lisp
;; (define my-first-csv-spec
;;   '((newline-type            . lax)
;;     (separator-chars         . (#\,))
;;     (quote-char              . #\")
;;     (quote-doubling-escapes? . #t)
;;     (whitespace-chars        . (#\space))))
;; @end lisp
;; a derived spec with a different @code{separator-chars} attribute and an
;; added @code{comment-chars} attribute can be created like:
;; @lisp
;; (%csv:csv-spec-derive my-first-csv-spec
;;                  '((separator-chars . (#\%))
;;                    (comment-chars   . (#\#))))
;; @result{}
;; ((separator-chars         . (#\%))
;;  (comment-chars           . (#\#))
;;  (newline-type            . lax)
;;  (quote-char              . #\")
;;  (quote-doubling-escapes? . #t)
;;  (whitespace-chars        . (#\space)))
;; @end lisp
;; In that the yielded spec might share some structure with @var{orig-spec}
;; and/or @var{changes}.  Most applications will not use this procedure
;; directly.

(define (%csv:csv-spec-derive orig-spec changes)
  ;; TODO: Make this not share structure.  Error-check and normalize at the
  ;; same time we clone.
  (let ((new-spec '()))
    (let ((add-to-new-spec
           (lambda (alist)
             (for-each (lambda (cell)
                         (or (assq (car cell) new-spec)
                             (set! new-spec (cons cell new-spec))))
      (add-to-new-spec changes)
      (add-to-new-spec orig-spec)
      (reverse new-spec))))

;;; @section Making Reader Makers

;;; CSV readers are procedures that are constructed dynamically to close over a
;;; particular CSV input and yield a parsed row value each time the procedure
;;; is applied.  For efficiency reasons, the reader procedures are themselves
;;; constructed by another procedure, @code{make-csv-reader-maker}, for
;;; particular CSV reader specs.

(define (%csv:csv-error code extra)
  ;; TODO: Maybe make the CSV error handler user-specifiable, or allow user to
  ;; specify some errors that should be disregarded.
  ;; TODO: Add position information.  Keep track of character position while
  ;; reading.
   (string-append "Erroneous CSV format: "
                  (case code
                     "Junk after close of quoted field:")
                    (else (string-append "INTERNAL ERROR: Unknown code: "
                                         (symbol->string code)))))

(define (%csv:newline-check-step0 newline-type c port)
  ;; (display "*DEBUG* (equal? newline-type 'lax) = ")
  ;; (write (equal? newline-type 'lax))
  ;; (newline)
  ;; (display "*DEBUG* (eqv? newline-type 'lax) = ")
  ;; (write (eqv? newline-type 'lax))
  ;; (newline)
  (case newline-type
    ((cr)   (eqv? c %csv:cr))
    ((lf)   (eqv? c %csv:lf))
    ((crlf) (if (eqv? c %csv:cr)
                (let ((c2 (peek-char port)))
                  (cond ((eof-object? c2)
                         ;; Note: This is a CR-EOF in an input that
                         ;; uses CR-LF for terminating records.  We
                         ;; are discarding the CR, so it will not be
                         ;; added to the field string.  We might want
                         ;; to signal an error.
                        ((eqv? c2 %csv:lf)
                         (read-char port)
                        (else #f)))
    ((lax detect) (cond ((eqv? c %csv:cr)
                         (let ((c2 (peek-char port)))
                           (cond ((eof-object? c2) #t)
                                 ((eqv? c2 %csv:lf)
                                  (read-char port)
                                 (else 'cr))))
                        ((eqv? c %csv:lf) 'lf)
                        (else #f)))
    (else (%csv:error
           "unrecognized newline-type"

(define %csv:make-portreader/positional
        (syntax-rules ()
          ((_ newline-type c port detected-newline-type)
           ;; Note: "port" and "detected-newline-type" must be identifiers.
           ;; "newline-type" and "c" must be identifiers or self-evals.
           (if (eqv? newline-type 'detect)
               (begin (set! detected-newline-type
                            (%csv:newline-check-step0 newline-type c port))
               (%csv:newline-check-step0 newline-type c port)))))
        ;; Note: This is to ensure the output string is gotten and closed
        ;; before consing it with the result of a recursive call.
        (syntax-rules ()
          ((_ os b) (let ((s (%csv:gosc os))) (cons s b))))))
    (lambda (newline-type
      (lambda (port)
        (let ((dnlt #f))
          (let read-fields-or-eof ((c (read-char port)))
             ((eof-object? c) '())
             ((and strip-leading-whitespace? (memv c whitespace-chars))
              ;; It's leading whitespace char when we're ignoring leading
              ;; whitespace in fields, and there might just be whitespace and
              ;; then an EOF, which should probably be considered just an EOF
              ;; rather than a row with one empty field, so just skip this
              ;; whitespace char.
              (read-fields-or-eof (read-char port)))
             ((and (not (null? comment-chars)) (memv c comment-chars))
              ;; It's a comment char in the first column (or in the first
              ;; non-whitespace column, if "strip-leading-whitespace?" is true),
              ;; so skip to end of line.
              (let ((fake-dnlt #f))
                (let loop ((c (read-char port)))
                  (cond ((eof-object? c) '())
                        ((newline-check newline-type c port fake-dnlt)
                         (read-fields-or-eof (read-char port)))
                        (else (loop (read-char port)))))))
              ;; It's not going to be just an EOF, so try to read a row.
              (let ((row
                     (let read-fields ((c c))
                        ;; If an EOF or newline in an unquoted field,
                        ;; consider the field and row finished.  (We don't
                        ;; consider EOF before newline to be an error,
                        ;; although perhaps that would be a useful check
                        ;; for a freak premature end-of-input when dealing
                        ;; with "well-formed" CSV).
                        ((or (eof-object? c)
                             (newline-check newline-type c port dnlt))
                         (list ""))
                        ;; If a field separator, finish this field and cons
                        ;; with value of recursive call to get the next
                        ;; field.
                        ((memv c separator-chars)
                         (cons "" (read-fields (read-char port))))
                        ;; If we're ignoring leading whitespace, and it's a
                        ;; whitespace-chars character, then recurse to keep
                        ;; finding the field start.
                        ((and strip-leading-whitespace?
                              (memv c whitespace-chars))
                         (read-fields (read-char port)))
                        ;; If a quote, read a quoted field.
                        ((and quote-char (eqv? c quote-char))
                         (let ((os (open-output-string)))
                           (let loop ((c (read-char port)))
                              ((or (eof-object? c)
                                   (and (not newlines-in-quotes?)
                                        (newline-check newline-type
                                                       c port dnlt)))
                               (list (%csv:gosc os)))
                              ((and quote-char (eqv? c quote-char))
                               (if quote-doubling-escapes?
                                   (let ((c2 (read-char port)))
                                     (if (eqv? c2 quote-char)
                                         (begin (write-char c2 os)
                                                (loop (read-char port)))
                                          (let skip-after ((c c2))
                                             ((or (eof-object? c)
                                                   newline-type c port dnlt))
                                             ((memv c separator-chars)
                                              (read-fields (read-char port)))
                                             ((memv c whitespace-chars)
                                              ;; Note: We tolerate
                                              ;; whitespace after field
                                              ;; close quote even if
                                              ;; skip-trailing-whitespace?
                                              ;; is false.
                                              (skip-after (read-char port)))
                                             (else (%csv:csv-error
                                   (gosc-cons os
                                              (read-fields (read-char port)))))
                              (else (write-char c os)
                                    (loop (read-char port)))))))
                        ;; It's the start of an unquoted field.
                         (let ((os (open-output-string)))
                           (write-char c os)
                           (let loop ((c (read-char port)))
                              ((or (eof-object? c)
                                   (newline-check newline-type c port dnlt))
                               (list (get-output-string os)))
                              ((memv c separator-chars)
                               (gosc-cons os (read-fields (read-char port))))
                              ((and strip-trailing-whitespace?
                                    (memv c whitespace-chars))
                               ;; TODO: Maybe optimize to avoid creating a new
                               ;; output string every time we see whitespace.
                               ;; We could use a string collector with unwrite.
                               ;; And/or do lookahead to see whether whitespace
                               ;; is only one character.  Do this after we have
                               ;; a better regression test suite.
                               (let ((ws-os (open-output-string)))
                                 (write-char c ws-os)
                                 (let ws-loop ((c (read-char port)))
                                    ((or (eof-object? c)
                                          newline-type c port dnlt))
                                     (close-output-port ws-os)
                                     (list (%csv:gosc os)))
                                    ((memv c separator-chars)
                                     (close-output-port ws-os)
                                     (gosc-cons os (read-fields (read-char
                                    ((memv c whitespace-chars)
                                     (write-char c ws-os)
                                     (ws-loop (read-char port)))
                                     (display (%csv:gosc ws-os) os)
                                     (write-char c os)
                                     (loop (read-char port)))))))
                              (else (write-char c os)
                                    (loop (read-char port)))))))))))
                (if (null? row)
                    (if (eq? newline-type 'detect)
                        (cons dnlt row)

(define %csv:make-portreader
  ;; TODO: Make a macro for the three times we list the spec attributes.
  (letrec ((pb (lambda (x) (if x #t #f)))
           (pc (lambda (x)
                 (cond ((char?   x) x)
                       ((string? x) (case (string-length x)
                                      ((1)  (string-ref  x 0))
                                      (else (%csv:type-error
                       (else (%csv:type-error "make-csv-reader-maker"
           (pc-f (lambda (x)
                   (cond ((not     x) x)
                         ((char?   x) x)
                         ((string? x) (case (string-length x)
                                        ((0)  #f)
                                        ((1)  (string-ref  x 0))
                                        (else (%csv:type-error
                                               "character or #f"
                         (else (%csv:type-error "make-csv-reader-maker"
                                                "character or #f"
           (pe (lambda (x acceptable)
                 (if (memq x acceptable)
                      (let ((os (open-output-string)))
                        (display "symbol from the set " os)
                        (write acceptable os)
                        (%csv:gosc os))
           (plc-n (lambda (x)
                    (or (list? x)
                        (%csv:type-error "make-csv-reader-maker"
                                         "list of characters"
                    (map pc x)))
           (plc (lambda (x)
                  (let ((result (plc-n x)))
                    (if (null? result)
                        (%csv:type-error "make-csv-reader-maker"
                                         "non-null list of characters"
    (lambda (reader-spec)
      (let ((newline-type               'lax)
            (separator-chars            '(#\,))
            (quote-char                 #\")
            (quote-doubling-escapes?    #t)
            (comment-chars              '())
            (whitespace-chars           '(#\space))
            (strip-leading-whitespace?  #f)
            (strip-trailing-whitespace? #f)
            (newlines-in-quotes?        #t))
        ;; TODO: It's erroneous to have two entries for the same attribute in a
        ;; spec.  However, it would be nice if we error-detected duplicate
        ;; entries, or at least had assq semantics (first, rather than last,
        ;; wins).  Use csv-spec-derive's descendants for that.
         (lambda (item)
           (let ((v (cdr item)))
             (case (car item)
                (set! newline-type (pe v '(cr crlf detect lax lf))))
                (set! separator-chars (plc v)))
                (set! quote-char (pc-f v)))
                (set! quote-doubling-escapes? (pb v)))
                (set! comment-chars (plc-n v)))
                (set! whitespace-chars (plc-n v)))
                (set! strip-leading-whitespace?  (pb v)))
                (set! strip-trailing-whitespace? (pb v)))
                (set! newlines-in-quotes? (pb v))))))

;;; @defproc make-csv-reader-maker reader-spec
;;; Constructs a CSV reader constructor procedure from the @var{reader-spec},
;;; with unspecified attributes having their default values.
;;; For example, given the input file @code{fruits.csv} with the content:
;;; @example
;;; apples  |  2 |  0.42
;;; bananas | 20 | 13.69
;;; @end example
;;; a reader for the file's apparent format can be constructed like:
;;; @lisp
;;; (define make-food-csv-reader
;;;   (make-csv-reader-maker
;;;    '((separator-chars            . (#\|))
;;;      (strip-leading-whitespace?  . #t)
;;;      (strip-trailing-whitespace? . #t))))
;;; @end lisp
;;; The resulting @code{make-food-csv-reader} procedure accepts one argument,
;;; which is either an input port from which to read, or a string from which to
;;; read.  Our example input file then can be be read by opening an input port
;;; on a file and using our new procedure to construct a reader on it:
;;; @lisp
;;; (define next-row
;;;   (make-food-csv-reader (open-input-file "fruits.csv")))
;;; @end lisp
;;; This reader, @code{next-row}, can then be called repeatedly to yield a
;;; parsed representation of each subsequent row.  The parsed format is a list
;;; of strings, one string for each column.  The null list is yielded to
;;; indicate that all rows have already been yielded.
;;; @lisp
;;; (next-row) @result{} ("apples" "2" "0.42")
;;; (next-row) @result{} ("bananas" "20" "13.69")
;;; (next-row) @result{} ()
;;; @end lisp

(define (make-csv-reader-maker reader-spec)
  (let ((make-portread
         (if (let ((p (assq 'newline-type reader-spec))) (and p (cdr p)))
             ;; Newline-adapting portreader-maker.
                    (%csv:csv-spec-derive reader-spec
                                          '((newline-type . detect)))))
                  ;; TODO: The set of cr/crlf/lf newline-type portreaders are
                  ;; constructed optimistically right now for two reasons:
                  ;; 1. we don't yet sanitize reader-specs of shared structure
                  ;; that can be mutated behind our backs; 2. eventually, we
                  ;; want to add a "lots-o-shots?" argument that, when true,
                  ;; would do this anyway.  Consider.
                    (%csv:csv-spec-derive reader-spec
                                          '((newline-type . cr)))))
                    (%csv:csv-spec-derive reader-spec
                                          '((newline-type . crlf)))))
                    (%csv:csv-spec-derive reader-spec
                                          '((newline-type . lf))))))
               (lambda ()
                 (let ((actual-portread #f))
                   (let ((adapt-portread
                          (lambda (port)
                            (let ((dnlt-row (detect-portread port)))
                              (if (null? dnlt-row)
                                  (begin (set! actual-portread
                                               (case (car dnlt-row)
                                                 ((cr)   cr-portread)
                                                 ((crlf) crlf-portread)
                                                 ((lf)   lf-portread)
                                                 (else   actual-portread)))
                                         (cdr dnlt-row)))))))
                     (set! actual-portread adapt-portread)
                     (lambda (port) (actual-portread port))))))
             ;; Stateless portreader-maker.
             (let ((reusable-portread
                    (%csv:make-portreader reader-spec)))
               (lambda () reusable-portread)))))
    (lambda (in)
      (let ((port     (%csv:in-arg "[csv-reader]" in))
            (portread (make-portread)))
        (lambda () (portread port))))))

;;; @section Making Readers

;;; In addition to being constructed from the result of
;;; @code{make-csv-reader-maker}, CSV readers can also be constructed using
;;; @code{make-csv-reader}.

;;; @defproc make-csv-reader in [reader-spec]
;;; Construct a CSV reader on the input @var{in}, which is an input port or a
;;; string.  If @var{reader-spec} is given, and is not the null list, then a
;;; ``one-shot'' reader constructor is constructed with that spec and used.  If
;;; @var{reader-spec} is not given, or is the null list, then the default CSV
;;; reader constructor is used.  For example, the reader from the
;;; @code{make-csv-reader-maker} example could alternatively have been
;;; constructed like:
;;; @lisp
;;; (define next-row
;;;   (make-csv-reader
;;;    (open-input-file "fruits.csv")
;;;    '((separator-chars            . (#\|))
;;;      (strip-leading-whitespace?  . #t)
;;;      (strip-trailing-whitespace? . #t))))
;;; @end lisp

(define make-csv-reader
  (let ((default-maker (make-csv-reader-maker '())))
    (lambda (in . rest)
      (let ((spec (cond ((null? rest)       '())
                        ((null? (cdr rest)) (car rest))
                        (else (%csv:error "make-csv-reader"
                                          "extraneous arguments"
                                          (cdr rest))))))
        ((if (null? spec)
             (make-csv-reader-maker spec))
         (%csv:in-arg "make-csv-reader" in))))))

;;; @section High-Level Conveniences

;;; Several convenience procedures are provided for iterating over the CSV rows
;;; and for converting the CSV to a list.
;;; To the dismay of some Scheme purists, each of these procedures accepts a
;;; @var{reader-or-in} argument, which can be a CSV reader, an input port, or a
;;; string.  If not a CSV reader, then the default reader constructor is used.
;;; For example, all three of the following are equivalent:
;;; @lisp
;;; (csv->list                                     STRING  )
;;; @equiv{}
;;; (csv->list (make-csv-reader                    STRING ))
;;; @equiv{}
;;; (csv->list (make-csv-reader (open-input-string STRING )))
;;; @end lisp

;;; @defproc csv-for-each proc reader-or-in
;;; Similar to Scheme's @code{for-each}, applies @var{proc}, a procedure of one
;;; argument, to each parsed CSV row in series.  @var{reader-or-in} is the CSV
;;; reader, input port, or string.  The return value is undefined.

;; TODO: Doc an example for this.

(define (csv-for-each proc reader-or-in)
  (let ((reader (%csv:reader-or-in-arg "csv-for-each" reader-or-in)))
    (let loop ((row (reader)))
      (or (null? row)
          (begin (proc row)
                 (loop (reader)))))))

;;; @defproc csv-map proc reader-or-in
;;; Similar to Scheme's @code{map}, applies @var{proc}, a procedure of one
;;; argument, to each parsed CSV row in series, and yields a list of the values
;;; of each application of @var{proc}, in order.  @var{reader-or-in} is the CSV
;;; reader, input port, or string.

;; TODO: Doc an example for this.

;; (define (csv-map proc reader-or-in)
;;   (let ((reader (%csv:reader-or-in-arg "csv-for-each" reader-or-in)))
;;     (let ((head '()))
;;       (let ((row (reader)))
;;         (if (null? row)
;;             head
;;             (let ((pair (cons (proc row) '())))
;;               (set! head pair)
;;               (let loop ((prior pair))
;;                 (let ((row (reader)))
;;                   (if (null? row)
;;                       head
;;                       (let ((pair (cons (proc row) '())))
;;                         (set-cdr! prior pair)
;;                         (loop pair)))))))))))

(define (csv-map proc reader-or-in)
  (let ((reader (%csv:reader-or-in-arg "csv-for-each" reader-or-in)))
    (let loop ((row (reader)) (ret null))
      (if (null? row)
          (reverse ret)
          (let ((ret (cons (proc row) ret)))
            (loop (reader) ret))))))

;;; @defproc csv->list reader-or-in
;;; Yields a list of CSV row lists from input @var{reader-or-in}, which is a
;;; CSV reader, input port, or string.

;; TODO: Doc an example for this.

;; (define (csv->list reader-or-in)
;;   (let ((reader (%csv:reader-or-in-arg "csv->list" reader-or-in)))
;;     (let ((head '()))
;;       (let ((row (reader)))
;;         (if (null? row)
;;             head
;;             (let ((pair (cons row '())))
;;               (set! head pair)
;;               (let loop ((prior pair))
;;                 (let ((row (reader)))
;;                   (if (null? row)
;;                       head
;;                       (let ((pair (cons row '())))
;;                         (set-cdr! prior pair)
;;                         (loop pair)))))))))))

(define (csv->list reader-or-in)
  (csv-map values reader-or-in))

;;; @section Converting CSV to SXML

;;; The @code{csv->sxml} procedure can be used to convert CSV to SXML format,
;;; for processing with various XML tools.

;;; @defproc csv->sxml reader-or-in [row-element [col-elements]]
;;; Reads CSV from input @var{reader-or-in} (which is a CSV reader, input port,
;;; or string), and yields an SXML representation.  If given, @var{row-element}
;;; is a symbol for the XML row element.  If @var{row-element} is not given,
;;; the default is the symbol @code{row}.  If given @var{col-elements} is a
;;; list of symbols for the XML column elements.  If not given, or there are
;;; more columns in a row than given symbols, column element symbols are of the
;;; format @code{col-@var{n}}, where @var{n} is the column number (the first
;;; column being number 0, not 1).
;;; For example, given a CSV-format file @code{friends.csv} that has the
;;; contents:
;;; @example
;;; Binoche,Ste. Brune,33-1-2-3
;;; Posey,Main St.,555-5309
;;; Ryder,Cellblock 9,
;;; @end example
;;; with elements not given, the result is:
;;; @lisp
;;; (csv->sxml (open-input-file "friends.csv"))
;;; @result{}
;;; (*TOP*
;;;  (row (col-0 "Binoche") (col-1 "Ste. Brune")  (col-2 "33-1-2-3"))
;;;  (row (col-0 "Posey")   (col-1 "Main St.")    (col-2 "555-5309"))
;;;  (row (col-0 "Ryder")   (col-1 "Cellblock 9") (col-2 "")))
;;; @end lisp
;;; With elements given, the result is like:
;;; @lisp
;;; (csv->sxml (open-input-file "friends.csv")
;;;            'friend
;;;            '(name address phone))
;;; @result{}
;;; (*TOP* (friend (name    "Binoche")
;;;                (address "Ste. Brune")
;;;                (phone   "33-1-2-3"))
;;;        (friend (name    "Posey")
;;;                (address "Main St.")
;;;                (phone   "555-5309"))
;;;        (friend (name    "Ryder")
;;;                (address "Cellblock 9")
;;;                (phone   "")))
;;; @end lisp

(define csv->sxml
  (let* ((top-symbol
          (string->symbol "*TOP*"))
          (lambda (n)
            (string->symbol (string-append "col-" (number->string n)))))
           (let loop ((i 0))
             (if (= i 32) ; arbitrary magic number
                 (cons (make-col-symbol i) (loop (+ 1 i)))))))
    ;; TODO: Have option to error when columns count doesn't match provided
    ;; column name list.
    (lambda (reader-or-in . rest)
      (let ((reader       (%csv:reader-or-in-arg "csv->sxml"
            (row-element  'row)
            (col-elements #f))
        ;; TODO: Maybe use case-lambda.
        (or (null? rest)
            (begin (set! row-element (car rest))
                   (let ((rest (cdr rest)))
                     (or (null? rest)
                         (begin (set! col-elements (car rest))
                                (let ((rest (cdr rest)))
                                  (or (null? rest)
                                       "extraneous arguments"
        ;; TODO: We could clone and grow default-col-elements for the duration
        ;; of this procedure.
        (cons top-symbol
              (csv-map (lambda (row)
                         (cons row-element
                               (let loop ((vals  row)
                                          (i     0)
                                          (names (or col-elements
                                 (if (null? vals)
                                     (cons (list (if (null? names)
                                                     (make-col-symbol i)
                                                     (car names))
                                                 (car vals))
                                           (loop (cdr vals)
                                                 (+ 1 i)
                                                 (if (null? names)
                                                     (cdr names))))))))

;; TODO: Make a define-csv-reader/positional, for great constant-folding.
;; That's part of the reason some things are done the way they are.

;; TODO: Make a csv-bind, as a newbie convenience for people without advanced
;; match forms, which looks good in examples.  This is better than a
;; csv-map/bind and a csv-for-each/bind.
;; (csv-for-each/bind ((column-binding ...) body ...)
;;               { (else => closure) | (else body ...) | }
;;               input-port
;;               [ csv-reader ])
;; (csv-for-each/bind
;;  ((lastname firstname email)
;;   ...)
;;  (else => (lambda (row) (error "CSV row didn't match pattern" row)))
;;  my-input-port
;;  my-csv-reader)

;; TODO: Handle escapes, once we find an actual example or specification of any
;; flavor of escapes in CSV other than quote-doubling inside a quoted field.

;; TODO: Add a spec attribute for treating adjacent separators as one, or
;; skipping empty fields.  This would probably only be used in practice for
;; parsing whitespace-separated input.

;; TODO: Get access to MS Excel or documentation, and make this correct.
;; (define msexcel-csv-reader-spec
;;   '((newline-type               . crlf)
;;     (separator-chars            . (#\,))
;;     (quote-char                 . #\")
;;     (quote-doubling-escapes?    . #t)
;;     (comment-chars              . ())
;;     (whitespace-chars           . (#\space))
;;     (strip-leading-whitespace?  . #f)
;;     (strip-trailing-whitespace? . #f)
;;     (newlines-in-quotes?        . #t)))

;; TODO: Maybe put this back in.
;; (define default-csv-reader-spec
;;   '((newline-type               . lax)
;;     (separator-chars            . (#\,))
;;     (quote-char                 . #\")
;;     (quote-doubling-escapes?    . #t)
;;     (comment-chars              . ())
;;     (whitespace-chars           . (#\space))
;;     (strip-leading-whitespace?  . #f)
;;     (strip-trailing-whitespace? . #f)
;;     (newlines-in-quotes?        . #t)))

;; TODO: Implement CSV writing, after CSV reading is field-tested and polished.

;; TODO: Call "close-input-port" once eof-object is hit, but make sure we still
;; can return an empty list on subsequent calls to the CSV reader.

;; TODO: Consider switching back to returning eof-object at the end of input.
;; We originally changed to returning the null list because we might want to
;; synthesize the EOF, and there is no R5RS binding for the eof-object.

;; TODO: [2005-12-09] In one test, Guile has a stack overflow when parsing a
;; row with 425 columns.  Wouldn't hurt to see if we can make things more
;; tail-recursive.

;;; @unnumberedsec History

;;; @table @asis
;;; @item Version 0.9 -- 2009-03-14 -- PLaneT @code{(1 5)}
;;; Documentation fix.
;;; @item Version 0.8 -- 2009-02-23 -- PLaneT @code{(1 4)}
;;; Documentation changes.
;;; @item Version 0.7 -- 2009-02-22 -- PLaneT @code{(1 3)}
;;; License is now LGPL 3.  Moved to author's new Scheme administration system.
;;; @item Version 0.6 -- 2008-08-12 -- PLaneT @code{(1 2)}
;;; For PLT 4 compatibility, new versions of @code{csv-map} and
;;; @code{csv->list} that don't use @code{set-cdr!} (courtesy of Doug
;;; Orleans). PLT 4 @code{if} compatibility change.  Minor documentation fixes.
;;; @item Version 0.5 --- 2005-12-09
;;; Changed a non-R5RS use of @code{letrec} to @code{let*}, caught by Guile and
;;; David Pirotte.
;;; @item Version 0.4 --- 2005-06-07
;;; Converted to Testeez.  Minor documentation changes.
;;; @item Version 0.3 --- 2004-07-21
;;; Minor documentation changes.  Test suite now disabled by default.
;;; @item Version 0.2 --- 2004-06-01
;;; Work-around for @code{case}-related bug observed in Gauche 0.8 and
;;; in @code{csv-internal:make-portreader/positional}.  Thanks to Grzegorz
;;; Chrupa@l{}a for reporting.
;;; @item Version 0.1 --- 2004-05-31
;;; First release, for testing with real-world input.
;;; @end table