summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorMichael Gran <spk121@yahoo.com>2009-08-10 22:55:29 -0700
committerMichael Gran <spk121@yahoo.com>2009-08-10 23:05:52 -0700
commitf5d7662fc86462fef68477fbfed994d2cf228e3e (patch)
treeb1797cb3b6c022cb922b4dd3079c647fac184a0f
parent88ed5759cd257f412aa1955c10c3fcea49ccade5 (diff)
downloadguile-f5d7662fc86462fef68477fbfed994d2cf228e3e.tar.gz
More string and symbol tests
* test-suite/tests/strings.test: more tests * test-suite/tests/symbols.test: more tests
-rw-r--r--test-suite/tests/strings.test244
-rw-r--r--test-suite/tests/symbols.test80
2 files changed, 317 insertions, 7 deletions
diff --git a/test-suite/tests/strings.test b/test-suite/tests/strings.test
index ffc6955ca..d82a4723d 100644
--- a/test-suite/tests/strings.test
+++ b/test-suite/tests/strings.test
@@ -1,7 +1,7 @@
;;;; strings.test --- test suite for Guile's string functions -*- scheme -*-
;;;; Jim Blandy <jimb@red-bean.com> --- August 1999
;;;;
-;;;; Copyright (C) 1999, 2001, 2004, 2005, 2006, 2008 Free Software Foundation, Inc.
+;;;; Copyright (C) 1999, 2001, 2004, 2005, 2006, 2008, 2009 Free Software Foundation, Inc.
;;;;
;;;; This library is free software; you can redistribute it and/or
;;;; modify it under the terms of the GNU Lesser General Public
@@ -20,14 +20,219 @@
(define-module (test-strings)
#:use-module (test-suite lib))
-
(define exception:read-only-string
(cons 'misc-error "^string is read-only"))
+(define exception:illegal-escape
+ (cons 'read-error "illegal character in escape sequence"))
;; Create a string from integer char values, eg. (string-ints 65) => "A"
(define (string-ints . args)
(apply string (map integer->char args)))
+;;
+;; string internals
+;;
+
+;; Some abbreviations
+;; BMP - Basic Multilingual Plane (codepoints below U+FFFF)
+;; SMP - Suplementary Multilingual Plane (codebpoints from U+10000 to U+1FFFF)
+
+(with-test-prefix "string internals"
+
+ (pass-if "new string starts at 1st char in stringbuf"
+ (let ((s "abc"))
+ (= 0 (assq-ref (%string-dump s) 'start))))
+
+ (pass-if "length of new string same as stringbuf"
+ (let ((s "def"))
+ (= (string-length s) (assq-ref (%string-dump s) 'stringbuf-length))))
+
+ (pass-if "contents of new string same as stringbuf"
+ (let ((s "ghi"))
+ (string=? s (assq-ref (%string-dump s) 'stringbuf-chars))))
+
+ (pass-if "writable strings are not read-only"
+ (let ((s "zyx"))
+ (not (assq-ref (%string-dump s) 'read-only))))
+
+ (pass-if "read-only strings are read-only"
+ (let ((s (substring/read-only "zyx" 0)))
+ (assq-ref (%string-dump s) 'read-only)))
+
+ (pass-if "null strings are inlined"
+ (let ((s ""))
+ (assq-ref (%string-dump s) 'stringbuf-inline)))
+
+ (pass-if "short Latin-1 encoded strings are inlined"
+ (let ((s "m"))
+ (assq-ref (%string-dump s) 'stringbuf-inline)))
+
+ (pass-if "long Latin-1 encoded strings are not inlined"
+ (let ((s "0123456789012345678901234567890123456789"))
+ (not (assq-ref (%string-dump s) 'stringbuf-inline))))
+
+ (pass-if "short UCS-4 encoded strings are not inlined"
+ (let ((s "\u0100"))
+ (not (assq-ref (%string-dump s) 'stringbuf-inline))))
+
+ (pass-if "long UCS-4 encoded strings are not inlined"
+ (let ((s "\u010012345678901234567890123456789"))
+ (not (assq-ref (%string-dump s) 'stringbuf-inline))))
+
+ (pass-if "new Latin-1 encoded strings are not shared"
+ (let ((s "abc"))
+ (not (assq-ref (%string-dump s) 'stringbuf-shared))))
+
+ (pass-if "new UCS-4 encoded strings are not shared"
+ (let ((s "\u0100bc"))
+ (not (assq-ref (%string-dump s) 'stringbuf-shared))))
+
+ ;; Should this be true? It isn't currently true.
+ (pass-if "null shared substrings are shared"
+ (let* ((s1 "")
+ (s2 (substring/shared s1 0 0)))
+ (throw 'untested)
+ (eq? (assq-ref (%string-dump s2) 'shared)
+ s1)))
+
+ (pass-if "ASCII shared substrings are shared"
+ (let* ((s1 "foobar")
+ (s2 (substring/shared s1 0 3)))
+ (eq? (assq-ref (%string-dump s2) 'shared)
+ s1)))
+
+ (pass-if "BMP shared substrings are shared"
+ (let* ((s1 "\u0100\u0101\u0102\u0103\u0104\u0105")
+ (s2 (substring/shared s1 0 3)))
+ (eq? (assq-ref (%string-dump s2) 'shared)
+ s1)))
+
+ (pass-if "null substrings are not shared"
+ (let* ((s1 "")
+ (s2 (substring s1 0 0)))
+ (not (eq? (assq-ref (%string-dump s2) 'shared)
+ s1))))
+
+ (pass-if "ASCII substrings are not shared"
+ (let* ((s1 "foobar")
+ (s2 (substring s1 0 3)))
+ (not (eq? (assq-ref (%string-dump s2) 'shared)
+ s1))))
+
+ (pass-if "BMP substrings are not shared"
+ (let* ((s1 "\u0100\u0101\u0102\u0103\u0104\u0105")
+ (s2 (substring s1 0 3)))
+ (not (eq? (assq-ref (%string-dump s2) 'shared)
+ s1))))
+
+ (pass-if "ASCII substrings share stringbufs before copy-on-write"
+ (let* ((s1 "foobar")
+ (s2 (substring s1 0 3)))
+ (assq-ref (%string-dump s1) 'stringbuf-shared)))
+
+ (pass-if "BMP substrings share stringbufs before copy-on-write"
+ (let* ((s1 "\u0100\u0101\u0102\u0103\u0104\u0105")
+ (s2 (substring s1 0 3)))
+ (assq-ref (%string-dump s1) 'stringbuf-shared)))
+
+ (pass-if "ASCII substrings don't share stringbufs after copy-on-write"
+ (let* ((s1 "foobar")
+ (s2 (substring s1 0 3)))
+ (string-set! s2 0 #\F)
+ (not (assq-ref (%string-dump s2) 'stringbuf-shared))))
+
+ (pass-if "BMP substrings don't share stringbufs after copy-on-write"
+ (let* ((s1 "\u0100\u0101\u0102\u0103\u0104\u0105")
+ (s2 (substring s1 0 3)))
+ (string-set! s2 0 #\F)
+ (not (assq-ref (%string-dump s2) 'stringbuf-shared))))
+
+ (with-test-prefix "encodings"
+
+ (pass-if "null strings are Latin-1 encoded"
+ (let ((s ""))
+ (not (assq-ref (%string-dump s) 'stringbuf-wide))))
+
+ (pass-if "ASCII strings are Latin-1 encoded"
+ (let ((s "jkl"))
+ (not (assq-ref (%string-dump s) 'stringbuf-wide))))
+
+ (pass-if "Latin-1 strings are Latin-1 encoded"
+ (let ((s "\xC0\xC1\xC2"))
+ (not (assq-ref (%string-dump s) 'stringbuf-wide))))
+
+ (pass-if "BMP strings are UCS-4 encoded"
+ (let ((s "\u0100\u0101\x0102"))
+ (assq-ref (%string-dump s) 'stringbuf-wide)))
+
+ (pass-if "SMP strings are UCS-4 encoded"
+ (let ((s "\U010300\u010301\x010302"))
+ (assq-ref (%string-dump s) 'stringbuf-wide)))
+
+ (pass-if "null list->string is Latin-1 encoded"
+ (let ((s (string-ints)))
+ (not (assq-ref (%string-dump s) 'stringbuf-wide))))
+
+ (pass-if "ASCII list->string is Latin-1 encoded"
+ (let ((s (string-ints 65 66 67)))
+ (not (assq-ref (%string-dump s) 'stringbuf-wide))))
+
+ (pass-if "Latin-1 list->string is Latin-1 encoded"
+ (let ((s (string-ints #xc0 #xc1 #xc2)))
+ (not (assq-ref (%string-dump s) 'stringbuf-wide))))
+
+ (pass-if "BMP list->string is UCS-4 encoded"
+ (let ((s (string-ints #x0100 #x0101 #x0102)))
+ (assq-ref (%string-dump s) 'stringbuf-wide)))
+
+ (pass-if "SMP list->string is UCS-4 encoded"
+ (let ((s (string-ints #x010300 #x010301 #x010302)))
+ (assq-ref (%string-dump s) 'stringbuf-wide)))
+
+ (pass-if "encoding of string not based on escape style"
+ (let ((s "\U000040"))
+ (not (assq-ref (%string-dump s) 'stringbuf-wide))))))
+
+(with-test-prefix "hex escapes"
+
+ (pass-if-exception "non-hex char in two-digit hex-escape"
+ exception:illegal-escape
+ (with-input-from-string "\"\\x0g\"" read))
+
+ (pass-if-exception "non-hex char in four-digit hex-escape"
+ exception:illegal-escape
+ (with-input-from-string "\"\\u000g\"" read))
+
+ (pass-if-exception "non-hex char in six-digit hex-escape"
+ exception:illegal-escape
+ (with-input-from-string "\"\\U00000g\"" read))
+
+ (pass-if-exception "premature termination of two-digit hex-escape"
+ exception:illegal-escape
+ (with-input-from-string "\"\\x0\"" read))
+
+ (pass-if-exception "premature termination of four-digit hex-escape"
+ exception:illegal-escape
+ (with-input-from-string "\"\\u000\"" read))
+
+ (pass-if-exception "premature termination of six-digit hex-escape"
+ exception:illegal-escape
+ (with-input-from-string "\"\\U00000\"" read))
+
+ (pass-if "extra hex digits ignored for two-digit hex escape"
+ (eqv? (string-ref "--\xfff--" 2)
+ (integer->char #xff)))
+
+ (pass-if "extra hex digits ignored for four-digit hex escape"
+ (eqv? (string-ref "--\u0100f--" 2)
+ (integer->char #x0100)))
+
+ (pass-if "extra hex digits ignored for six-digit hex escape"
+ (eqv? (string-ref "--\U010300f--" 2)
+ (integer->char #x010300)))
+
+ (pass-if "escaped characters match non-escaped ASCII characters"
+ (string=? "ABC" "\x41\u0042\U000043")))
;;
;; string=?
@@ -181,8 +386,20 @@
exception:out-of-range
(string-ref "hello" -1))
- (pass-if "regular string"
- (char=? (string-ref "GNU Guile" 4) #\G)))
+ (pass-if "regular string, ASCII char"
+ (char=? (string-ref "GNU Guile" 4) #\G))
+
+ (pass-if "regular string, hex escaped Latin-1 char"
+ (char=? (string-ref "--\xff--" 2)
+ (integer->char #xff)))
+
+ (pass-if "regular string, hex escaped BMP char"
+ (char=? (string-ref "--\u0100--" 2)
+ (integer->char #x0100)))
+
+ (pass-if "regular string, hex escaped SMP char"
+ (char=? (string-ref "--\U010300--" 2)
+ (integer->char #x010300))))
;;
;; string-set!
@@ -210,10 +427,25 @@
exception:read-only-string
(string-set! (substring/read-only "abc" 0) 1 #\space))
- (pass-if "regular string"
+ (pass-if "regular string, ASCII char"
(let ((s (string-copy "GNU guile")))
(string-set! s 4 #\G)
- (char=? (string-ref s 4) #\G))))
+ (char=? (string-ref s 4) #\G)))
+
+ (pass-if "regular string, Latin-1 char"
+ (let ((s (string-copy "GNU guile")))
+ (string-set! s 4 (integer->char #xfe))
+ (char=? (string-ref s 4) (integer->char #xfe))))
+
+ (pass-if "regular string, BMP char"
+ (let ((s (string-copy "GNU guile")))
+ (string-set! s 4 (integer->char #x0100))
+ (char=? (string-ref s 4) (integer->char #x0100))))
+
+ (pass-if "regular string, SMP char"
+ (let ((s (string-copy "GNU guile")))
+ (string-set! s 4 (integer->char #x010300))
+ (char=? (string-ref s 4) (integer->char #x010300)))))
(with-test-prefix "string-split"
diff --git a/test-suite/tests/symbols.test b/test-suite/tests/symbols.test
index 5be2743b2..3b1abe1e9 100644
--- a/test-suite/tests/symbols.test
+++ b/test-suite/tests/symbols.test
@@ -1,6 +1,6 @@
;;;; symbols.test --- test suite for Guile's symbols -*- scheme -*-
;;;;
-;;;; Copyright (C) 2001, 2006, 2008 Free Software Foundation, Inc.
+;;;; Copyright (C) 2001, 2006, 2008, 2009 Free Software Foundation, Inc.
;;;;
;;;; This library is free software; you can redistribute it and/or
;;;; modify it under the terms of the GNU Lesser General Public
@@ -31,6 +31,84 @@
(define (documented? object)
(not (not (object-documentation object))))
+(define (symbol-length s)
+ (string-length (symbol->string s)))
+
+;;
+;; symbol internals
+;;
+
+(with-test-prefix "symbol internals"
+
+ (pass-if "length of new symbol same as stringbuf"
+ (let ((s 'def))
+ (= (symbol-length s) (assq-ref (%symbol-dump s) 'stringbuf-length))))
+
+ (pass-if "contents of new symbol same as stringbuf"
+ (let ((s 'ghi))
+ (string=? (symbol->string s)
+ (assq-ref (%symbol-dump s) 'stringbuf-chars))))
+
+ (pass-if "the null symbol is inlined"
+ (let ((s '#{}#))
+ (assq-ref (%symbol-dump s) 'stringbuf-inline)))
+
+ (pass-if "short Latin-1-encoded symbols are inlined"
+ (let ((s 'm))
+ (assq-ref (%symbol-dump s) 'stringbuf-inline)))
+
+ (pass-if "long Latin-1-encoded symbols are not inlined"
+ (let ((s 'x0123456789012345678901234567890123456789))
+ (not (assq-ref (%symbol-dump s) 'stringbuf-inline))))
+
+ ;; symbol->string isn't ready for UCS-4 yet
+
+ ;;(pass-if "short UCS-4-encoded symbols are not inlined"
+ ;; (let ((s (string->symbol "\u0100")))
+ ;; (not (assq-ref (%symbol-dump s) 'stringbuf-inline))))
+
+ ;;(pass-if "long UCS-4-encoded symbols are not inlined"
+ ;; (let ((s (string->symbol "\u010012345678901234567890123456789")))
+ ;; (not (assq-ref (%symbol-dump s) 'stringbuf-inline))))
+
+ (with-test-prefix "hashes"
+
+ (pass-if "equal symbols have equal hashes"
+ (let ((s1 'mux)
+ (s2 'mux))
+ (= (assq-ref (%symbol-dump s1) 'hash)
+ (assq-ref (%symbol-dump s2) 'hash))))
+
+ (pass-if "different symbols have different hashes"
+ (let ((s1 'mux)
+ (s2 'muy))
+ (not (= (assq-ref (%symbol-dump s1) 'hash)
+ (assq-ref (%symbol-dump s2) 'hash))))))
+
+ (with-test-prefix "encodings"
+
+ (pass-if "the null symbol is Latin-1 encoded"
+ (let ((s '#{}#))
+ (not (assq-ref (%symbol-dump s) 'stringbuf-wide))))
+
+ (pass-if "ASCII symbols are Latin-1 encoded"
+ (let ((s 'jkl))
+ (not (assq-ref (%symbol-dump s) 'stringbuf-wide))))
+
+ (pass-if "Latin-1 symbols are Latin-1 encoded"
+ (let ((s (string->symbol "\xC0\xC1\xC2")))
+ (not (assq-ref (%symbol-dump s) 'stringbuf-wide))))
+
+ ;; symbol->string isn't ready for UCS-4 yet
+
+ ;;(pass-if "BMP symbols are UCS-4 encoded"
+ ;; (let ((s (string->symbol "\u0100\u0101\x0102")))
+ ;; (assq-ref (%symbol-dump s) 'stringbuf-wide)))
+
+ ;;(pass-if "SMP symbols are UCS-4 encoded"
+ ;; (let ((s (string->symbol "\U010300\u010301\x010302")))
+ ;; (assq-ref (%symbol-dump s) 'stringbuf-wide)))
+ ))
;;;
;;; symbol?