;;;; Programmer: Kevin M. Rosenberg
;;;; Date Started: Apr 2000
;;;;
-;;;; $Id: strings.lisp,v 1.8 2003/04/28 16:07:43 kevin Exp $
+;;;; $Id: strings.lisp,v 1.41 2003/06/15 07:48:30 kevin Exp $
;;;;
;;;; This file, part of KMRCL, is Copyright (c) 2002 by Kevin M. Rosenberg
;;;;
;;;; *************************************************************************
-(in-package :kmrcl)
-(declaim (optimize (speed 3) (safety 1) (compilation-speed 0) (debug 3)))
+(in-package #:kmrcl)
;;; Strings
(defun count-string-words (str)
(declare (simple-string str)
- (optimize (speed 3) (safety 0)))
+ (optimize (speed 3) (safety 0) (space 0)))
(let ((n-words 0)
(in-word nil))
(declare (fixnum n-words))
- (dotimes (i (length str))
- (let ((ch (char str i)))
- (declare (character ch))
- (if (alphanumericp ch)
- (unless in-word
- (incf n-words)
- (setq in-word t))
- (setq in-word nil))))
- n-words))
-
-#+excl
-(defun delimited-string-to-list (string &optional (separator #\space))
- (excl:delimited-string-to-list string separator))
-
-#-excl
-(defun delimited-string-to-list (sequence &optional (separator #\space))
- "Split a string by a delimitor"
- (loop
- with start = 0
- for end = (position separator sequence :start start)
- collect (subseq sequence start end)
- until (null end)
- do
- (setf start (1+ end))))
-
-#+excl
-(defun list-to-delimited-string (list &optional (separator #\space))
- (excl:list-to-delimited-string list separator))
-
-#-excl
-(defun list-to-delimited-string (list &optional (separator #\space))
- (if (consp list)
- (let ((fmt (format nil "~~A~~{~A~~A~~}" separator)))
- (format nil fmt (first list) (rest list)))
- ""))
+ (do* ((len (length str))
+ (i 0 (1+ i)))
+ ((= i len) n-words)
+ (declare (fixnum i))
+ (if (alphanumericp (schar str i))
+ (unless in-word
+ (incf n-words)
+ (setq in-word t))
+ (setq in-word nil)))))
+
+;; From Larry Hunter with modifications
+(defun position-char (char string start max)
+ (declare (optimize (speed 3) (safety 0) (space 0))
+ (fixnum start max) (simple-string string))
+ (do* ((i start (1+ i)))
+ ((= i max) nil)
+ (declare (fixnum i))
+ (when (char= char (schar string i)) (return i))))
+
+(defun position-not-char (char string start max)
+ (declare (optimize (speed 3) (safety 0) (space 0))
+ (fixnum start max) (simple-string string))
+ (do* ((i start (1+ i)))
+ ((= i max) nil)
+ (declare (fixnum i))
+ (when (char/= char (schar string i)) (return i))))
+
+(defun delimited-string-to-list (string &optional (separator #\space)
+ skip-terminal)
+ "split a string with delimiter"
+ (declare (optimize (speed 3) (safety 0) (space 0) (compilation-speed 0))
+ (type string string)
+ (type character separator))
+ (do* ((len (length string))
+ (output '())
+ (pos 0)
+ (end (position-char separator string pos len)
+ (position-char separator string pos len)))
+ ((null end)
+ (if (< pos len)
+ (push (subseq string pos) output)
+ (when (or (not skip-terminal) (zerop len))
+ (push "" output)))
+ (nreverse output))
+ (declare (type fixnum pos len)
+ (type (or null fixnum) end))
+ (push (subseq string pos end) output)
+ (setq pos (1+ end))))
+
+
+(defun list-to-delimited-string (list &optional (separator " "))
+ (format nil (concatenate 'string "~{~A~^" (string separator) "~}") list))
(defun string-invert (str)
"Invert case of a string"
(let ((up nil) (down nil))
(block skip
(loop for char of-type character across str do
- (cond ((upper-case-p char) (if down (return-from skip str) (setf up t)))
- ((lower-case-p char) (if up (return-from skip str) (setf down t)))))
+ (cond ((upper-case-p char)
+ (if down (return-from skip str) (setf up t)))
+ ((lower-case-p char)
+ (if up (return-from skip str) (setf down t)))))
(if up (string-downcase str) (string-upcase str)))))
(defun add-sql-quotes (s)
(defun substitute-string-for-char (procstr match-char subst-str)
"Substitutes a string for a single matching character of a string"
- (let ((pos (position match-char procstr)))
- (if pos
- (concatenate 'string
- (subseq procstr 0 pos) subst-str
- (substitute-string-for-char (subseq procstr (1+ pos)) match-char subst-str))
- procstr)))
+ (substitute-chars-strings procstr (list (cons match-char subst-str))))
(defun string-substitute (string substring replacement-string)
"String substitute by Larry Hunter. Obtained from Google"
replacement-string))
(setq last-end (+ next-start substring-length)))))
-
(defun string-trim-last-character (s)
-"Return the string less the last character"
- (subseq s 0 (1- (length s))))
+ "Return the string less the last character"
+ (let ((len (length s)))
+ (if (plusp len)
+ (subseq s 0 (1- len))
+ s)))
+
+(defun nstring-trim-last-character (s)
+ "Return the string less the last character"
+ (let ((len (length s)))
+ (if (plusp len)
+ (nsubseq s 0 (1- len))
+ s)))
(defun string-hash (str &optional (bitmask 65535))
(let ((hash 0))
(defun whitespace? (c)
(declare (character c))
- (declare (optimize (speed 3) (safety 0)))
- (or (char= c #\Space) (char= c #\Tab) (char= c #\Return) (char= c #\Linefeed)))
+ (locally (declare (optimize (speed 3) (safety 0)))
+ (or (char= c #\Space) (char= c #\Tab) (char= c #\Return)
+ (char= c #\Linefeed))))
(defun not-whitespace? (c)
(not (whitespace? c)))
(when (stringp str)
(null (find-if #'not-whitespace? str))))
-(defun string-replace-chars-strings (str repl-alist)
- "Replace all instances of a chars with a string. repl-alist is an assoc
-list of characters and replacement strings."
- (declare (string str))
- (let* ((orig-len (length str))
- (new-len orig-len))
- (declare (fixnum orig-len new-len))
- (dotimes (i orig-len)
- (declare (fixnum i))
+(defun replaced-string-length (str repl-alist)
+ (declare (simple-string str)
+ (fixnum orig-len new-len)
+ (optimize (speed 3) (safety 0) (space 0)))
+ (do* ((i 0 (1+ i))
+ (orig-len (length str))
+ (new-len orig-len))
+ ((= i orig-len) new-len)
+ (declare (fixnum i orig-len new-len))
(let* ((c (char str i))
(match (assoc c repl-alist :test #'char=)))
(declare (character c))
(when match
- (incf new-len (length (cdr match))))))
- (let ((new-string (make-string new-len))
- (i 0))
- (declare (string new-string)
- (fixnum i))
- (dotimes (i orig-len)
- (declare (fixnum i))
- (let* ((c (char str i))
- (match (assoc c repl-alist :test #'char=)))
- (declare (character c))
- (if match
- (let ((subst (cdr match)))
- (dotimes (j (length subst))
- (setf (char new-string i) (char subst j))
- (incf i)))
- (progn
- (setf (char new-string i) c)))))
- new-string)))
+ (incf new-len (1- (length (cdr match))))))))
+
+(defun substitute-chars-strings (str repl-alist)
+ "Replace all instances of a chars with a string. repl-alist is an assoc
+list of characters and replacement strings."
+ (declare (simple-string str)
+ (optimize (speed 3) (safety 0) (space 0)))
+ (do* ((orig-len (length str))
+ (new-string (make-string (replaced-string-length str repl-alist)))
+ (spos 0 (1+ spos))
+ (dpos 0))
+ ((>= spos orig-len)
+ new-string)
+ (declare (fixnum spos dpos) (simple-string new-string))
+ (let* ((c (char str spos))
+ (match (assoc c repl-alist :test #'char=)))
+ (declare (character c))
+ (if match
+ (let* ((subst (cdr match))
+ (len (length subst)))
+ (declare (fixnum len)
+ (simple-string subst))
+ (dotimes (j len)
+ (declare (fixnum j))
+ (setf (char new-string dpos) (char subst j))
+ (incf dpos)))
+ (progn
+ (setf (char new-string dpos) c)
+ (incf dpos))))))
(defun escape-xml-string (string)
"Escape invalid XML characters"
- (string-replace-char-string
- (string-replace-char-string string #\& "&")
- #\< "<"))
-
-(defun string-replace-char-string (string repl-char repl-str)
- "Replace all occurances of repl-char with repl-str"
- (declare (simple-string string))
- (let ((count (count repl-char string)))
- (declare (fixnum count))
- (if (zerop count)
- string
- (locally (declare (optimize (speed 3) (safety 0)))
- (let* ((old-length (length string))
- (repl-length (length repl-str))
- (new-string (make-string (the fixnum
- (+ old-length
- (the fixnum
- (* count
- (the fixnum (1- repl-length)))))))))
- (declare (fixnum old-length repl-length)
- (simple-string new-string))
- (let ((newpos 0))
- (declare (fixnum newpos))
- (dotimes (oldpos (length string))
- (declare (fixnum oldpos))
- (if (char= repl-char (schar string oldpos))
- (dotimes (repl-pos repl-length)
- (declare (fixnum repl-pos))
- (setf (schar new-string newpos) (schar repl-str repl-pos))
- (incf newpos))
- (progn
- (setf (schar new-string newpos) (schar string oldpos))
- (incf newpos)))))
- new-string)))))
-
-
+ (substitute-chars-strings string '((#\& . "&") (#\< . "<"))))
(defun make-usb8-array (len)
- (make-array len :adjustable nil
- :fill-pointer nil
- :element-type '(unsigned-byte 8)))
+ (make-array len :element-type '(unsigned-byte 8)))
(defun usb8-array-to-string (vec)
+ (declare (type (simple-array (unsigned-byte 8) (*)) vec))
(let* ((len (length vec))
(str (make-string len)))
(declare (fixnum len)
(simple-string str)
(optimize (speed 3)))
- (dotimes (i len)
+ (do ((i 0 (1+ i)))
+ ((= i len) str)
(declare (fixnum i))
- (setf (schar str i) (code-char (aref vec i))))
- str))
+ (setf (schar str i) (code-char (aref vec i))))))
(defun string-to-usb8-array (str)
+ (declare (simple-string str))
(let* ((len (length str))
(vec (make-usb8-array len)))
(declare (fixnum len)
- (type (array fixnum (*)) vec)
+ (type (simple-array (unsigned-byte 8) (*)) vec)
(optimize (speed 3)))
- (dotimes (i len)
+ (do ((i 0 (1+ i)))
+ ((= i len) vec)
+ (declare (fixnum i))
+ (setf (aref vec i) (char-code (schar str i))))))
+
+(defun concat-separated-strings (separator &rest lists)
+ (format nil (concatenate 'string "~{~A~^" (string separator) "~}")
+ (append-sublists lists)))
+
+(defun only-null-list-elements-p (lst)
+ (or (null lst) (every #'null lst)))
+
+(defun print-separated-strings (strm separator &rest lists)
+ (declare (optimize (speed 3) (safety 0) (space 0) (debug 0)
+ (compilation-speed 0)))
+ (do* ((rest-lists lists (cdr rest-lists))
+ (list (car rest-lists) (car rest-lists))
+ (last-list (only-null-list-elements-p (cdr rest-lists))
+ (only-null-list-elements-p (cdr rest-lists))))
+ ((null rest-lists) strm)
+ (do* ((lst list (cdr lst))
+ (elem (car lst) (car lst))
+ (last-elem (null (cdr lst)) (null (cdr lst))))
+ ((null lst))
+ (write-string elem strm)
+ (unless (and last-elem last-list)
+ (write-string separator strm)))))
+
+(defun prefixed-fixnum-string (num pchar len)
+ "Outputs a string of LEN digit with an optional initial character PCHAR.
+Leading zeros are present."
+ (declare (optimize (speed 3) (safety 0) (space 0))
+ (type fixnum num len))
+ (when pchar
+ (incf len))
+ (do* ((zero-code (char-code #\0))
+ (result (make-string len :initial-element #\0))
+ (minus? (minusp num))
+ (val (if minus? (- 0 num) num) (floor (/ val 10)))
+ (pos (1- len) (1- pos))
+ (mod (mod val 10) (mod val 10)))
+ ((or (zerop val) (minusp pos))
+ (when pchar
+ (setf (schar result 0) pchar))
+ (when minus? (setf (schar result (if pchar 1 0)) #\-))
+ result)
+ (declare (fixnum val mod zero-code pos) (simple-string result))
+ (setf (schar result pos) (code-char (+ zero-code mod)))))
+
+(defun integer-string (num len)
+ "Outputs a string of LEN digit with an optional initial character PCHAR.
+Leading zeros are present."
+ (declare (optimize (speed 3) (safety 0) (space 0))
+ (type fixnum len) (type integer num))
+ (do* ((zero-code (char-code #\0))
+ (result (make-string len :initial-element #\0))
+ (minus? (minusp num))
+ (val (if minus? (- 0 num) num) (floor (/ val 10)))
+ (pos (1- len) (1- pos))
+ (mod (mod val 10) (mod val 10)))
+ ((or (zerop val) (minusp pos))
+ (when minus? (setf (schar result 0) #\-))
+ result)
+ (declare (fixnum mod zero-code pos) (simple-string result) (integer val))
+ (setf (schar result pos) (code-char (+ zero-code mod)))))
+
+(defun fast-string-search (substr str substr-length startpos endpos)
+ "Optimized search for a substring in a simple-string"
+ (declare (simple-string substr str)
+ (fixnum substr-length startpos endpos)
+ (optimize (speed 3) (space 0) (safety 0)))
+ (do* ((pos startpos (1+ pos))
+ (lastpos (- endpos substr-length)))
+ ((> pos lastpos) nil)
+ (declare (fixnum pos lastpos))
+ (do ((i 0 (1+ i)))
+ ((= i substr-length)
+ (return-from fast-string-search pos))
(declare (fixnum i))
- (setf (aref vec i) (char-code (schar str i))))
- vec))
+ (unless (char= (schar str (+ i pos)) (schar substr i))
+ (return nil)))))
+
+(defun string-delimited-string-to-list (str substr)
+ "splits a string delimited by substr into a list of strings"
+ #+ignore
+ (declare (simple-string str substr)
+ (optimize (speed 3) (safety 0) (space 0) (compilation-speed 0)))
+ (do* ((substr-len (length substr))
+ (strlen (length str))
+ (output '())
+ (pos 0)
+ (end (fast-string-search substr str substr-len pos strlen)
+ (fast-string-search substr str substr-len pos strlen)))
+ ((null end)
+ (when (< pos strlen)
+ (push (subseq str pos) output))
+ (nreverse output))
+ (declare (fixnum strlen substr-len pos)
+ (type (or fixnum null) end))
+ (push (subseq str pos end) output)
+ (setq pos (+ end substr-len))))
+
+(defun string-to-list-skip-delimiter (str &optional (delim #\space))
+ "Return a list of strings, delimited by spaces, skipping spaces."
+ (declare (simple-string str)
+ (optimize (speed 0) (space 0) (safety 0)))
+ (do* ((results '())
+ (end (length str))
+ (i (position-not-char delim str 0 end)
+ (position-not-char delim str j end))
+ (j (when i (position-char delim str i end))
+ (when i (position-char delim str i end))))
+ ((or (null i) (null j))
+ (when (and i (< i end))
+ (push (subseq str i end) results))
+ (nreverse results))
+ (declare (fixnum i j end))
+ (push (subseq str i j) results)))
+
+(defun string-starts-with (start str)
+ (and (>= (length str) (length start))
+ (string-equal start str :end2 (length start))))
+
+(defun count-string-char (s c)
+ "Return a count of the number of times a character appears in a string"
+ (declare (simple-string s)
+ (character c)
+ (optimize (speed 3) (safety 0)))
+ (do ((len (length s))
+ (i 0 (1+ i))
+ (count 0))
+ ((= i len) count)
+ (declare (fixnum i len count))
+ (when (char= (schar s i) c)
+ (incf count))))
+
+(defun count-string-char-if (pred s)
+ "Return a count of the number of times a predicate is true
+for characters in a string"
+ (declare (simple-string s)
+ (optimize (speed 3) (safety 0) (space 0)))
+ (do ((len (length s))
+ (i 0 (1+ i))
+ (count 0))
+ ((= i len) count)
+ (declare (fixnum i len count))
+ (when (funcall pred (schar s i))
+ (incf count))))
+
+
+;;; URL Encoding
+
+(defun non-alphanumericp (ch)
+ (not (alphanumericp ch)))
+
+(defvar +hex-chars+ "0123456789ABCDEF")
+(declaim (type (simple-array character 16) +hex-chars+))
+
+(defun hexchar (n)
+ (declare (type (integer 0 15) n))
+ (aref +hex-chars+ n))
+
+(defun escape-uri-field (query)
+ "Escape non-alphanumeric characters for URI fields"
+ (declare (simple-string query)
+ (optimize (speed 3) (safety 0) (space 0)))
+ (do* ((count (count-string-char-if #'non-alphanumericp query))
+ (len (length query))
+ (new-len (+ len (* 2 count)))
+ (str (make-string new-len))
+ (spos 0 (1+ spos))
+ (dpos 0 (1+ dpos)))
+ ((= spos len) str)
+ (declare (fixnum count len new-len spos dpos)
+ (simple-string str))
+ (let ((ch (schar query spos)))
+ (if (non-alphanumericp ch)
+ (let ((c (char-code ch)))
+ (setf (schar str dpos) #\%)
+ (incf dpos)
+ (setf (schar str dpos) (hexchar (logand (ash c -4) 15)))
+ (incf dpos)
+ (setf (schar str dpos) (hexchar (logand c 15))))
+ (setf (schar str dpos) ch)))))