# HG changeset patch # User Aidan Kehoe # Date 1255363275 -3600 # Node ID a357478dd457c5e9ec91eb4a5ac270628ed4468c # Parent 84f870bbd17bd363b41e40b617c29fc1e79724bb Fix some test failures, mule-tests.el. 2009-10-12 Aidan Kehoe * automated/mule-tests.el : Revert to the old Unicode mapping for scaron once we're finished testing it. Don't check the fixed-width coding systems with odd line endings for ASCII-transparency; maybe we should, but that would require that invalid sequence characters for on-disk ?\x0a be generated by Macintosh line-ending coding systems, for example. diff -r 84f870bbd17b -r a357478dd457 tests/ChangeLog --- a/tests/ChangeLog Mon Oct 12 16:42:38 2009 +0100 +++ b/tests/ChangeLog Mon Oct 12 17:01:15 2009 +0100 @@ -1,3 +1,13 @@ +2009-10-12 Aidan Kehoe + + * automated/mule-tests.el : + Revert to the old Unicode mapping for scaron once we're finished + testing it. + Don't check the fixed-width coding systems with odd line endings + for ASCII-transparency; maybe we should, but that would require + that invalid sequence characters for on-disk ?\x0a be generated by + Macintosh line-ending coding systems, for example. + 2009-10-05 Jerry James * gtk/event-stream-tests.el: Add GPL v2 or later notice with diff -r 84f870bbd17b -r a357478dd457 tests/automated/ccl-tests.el --- a/tests/automated/ccl-tests.el Mon Oct 12 16:42:38 2009 +0100 +++ b/tests/automated/ccl-tests.el Mon Oct 12 17:01:15 2009 +0100 @@ -132,7 +132,7 @@ "CCL TEST temporary coding-system." '(mnemonic "CCL-TEST" eol-type lf - safe-chars t + safe-charsets t decode ccl-test-decoder encode ccl-test-encoder)))) diff -r 84f870bbd17b -r a357478dd457 tests/automated/mule-tests.el --- a/tests/automated/mule-tests.el Mon Oct 12 16:42:38 2009 +0100 +++ b/tests/automated/mule-tests.el Mon Oct 12 17:01:15 2009 +0100 @@ -427,12 +427,15 @@ ;;--------------------------------------------------------------- (let* ((scaron (make-char 'latin-iso8859-2 57))) ;; Used to try #x0000, but you can't change ASCII or Latin-1 - (loop for code in '(#x0100 #x2222 #x4444 #xffff) do + (loop + for code in '(#x0100 #x2222 #x4444 #xffff) + with initial-unicode = (char-to-unicode scaron) + do (progn (set-unicode-conversion scaron code) (Assert (eq code (char-to-unicode scaron))) - (Assert (eq scaron (unicode-to-char code '(latin-iso8859-2)))))) - + (Assert (eq scaron (unicode-to-char code '(latin-iso8859-2))))) + finally (set-unicode-conversion scaron initial-unicode)) (Check-Error wrong-type-argument (set-unicode-conversion scaron -10000))) (dolist (utf-8-char @@ -531,17 +534,16 @@ collect i)) do (when (and (eq 'fixed-width (coding-system-type coding-system)) - ;; Don't check the coding systems with autodetect, they are - ;; not round-trip compatible for the possible line-ending - ;; characters. - (string-match #r"-\(unix\|dos\|mac\)$" - (symbol-name coding-system))) + ;; Don't check the coding systems with odd line endings + ;; (maybe we should): + (eq 'lf (coding-system-eol-type coding-system))) ;; These coding systems are round-trip compatible with themselves. (Assert (equal (encode-coding-string (decode-coding-string all-possible-octets coding-system) coding-system) - all-possible-octets)))) + all-possible-octets) + (format "checking %s is transparent" coding-system)))) ;;--------------------------------------------------------------- ;; Test charset-in-* functions