1 # This testcase is part of GDB, the GNU debugger.
3 # Copyright 2001-2019 Free Software Foundation, Inc.
5 # This program is free software; you can redistribute it and/or modify
6 # it under the terms of the GNU General Public License as published by
7 # the Free Software Foundation; either version 3 of the License, or
8 # (at your option) any later version.
10 # This program is distributed in the hope that it will be useful,
11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 # GNU General Public License for more details.
15 # You should have received a copy of the GNU General Public License
16 # along with this program. If not, see <http://www.gnu.org/licenses/>.
18 # Please email any bugs, comments, and/or additions to this file to:
21 # Test GDB's character set support.
24 standard_testfile .c charset-malloc.c
26 if { [prepare_for_testing "failed to prepare" ${testfile} [list $srcfile $srcfile2]] } {
30 # Parse the output from a `show charset' command. Return the host
31 # and target charset as a two-element list.
32 proc parse_show_charset_output {testname} {
36 -re "The host character set is \"(.*)\"\\.\[\r\n\]+The target character set is \"(.*)\"\\.\[\r\n\]+The target wide character set is \"(.*)\"\\.\[\r\n\]+$gdb_prompt $" {
37 set host_charset $expect_out(1,string)
38 set target_charset $expect_out(2,string)
39 set retlist [list $host_charset $target_charset]
42 -re "The host character set is \"(.*)\"\\.\[\r\n\]+$gdb_prompt $" {
43 set host_charset $expect_out(1,string)
44 set retlist [list $host_charset]
47 -re "The target character set is \"(.*)\"\\.\[\r\n\]+$gdb_prompt $" {
48 set target_charset $expect_out(1,string)
49 set retlist [list $target_charset]
52 -re ".*$gdb_prompt $" {
56 fail "$testname (timeout)"
64 # Try the various `show charset' commands.
66 send_gdb "show charset\n"
67 set show_charset [parse_show_charset_output "show charset"]
69 send_gdb "show target-charset\n"
70 set show_target_charset \
71 [lindex [parse_show_charset_output "show target-charset"] 0]
73 if {[lsearch -exact $show_charset $show_target_charset] >= 0} {
74 pass "check `show target-charset' against `show charset'"
76 fail "check `show target-charset' against `show charset'"
79 send_gdb "show host-charset\n"
80 set show_host_charset \
81 [lindex [parse_show_charset_output "show host-charset"] 0]
83 if {[lsearch -exact $show_charset $show_host_charset] >= 0} {
84 pass "check `show host-charset' against `show charset'"
86 fail "check `show host-charset' against `show charset'"
89 # Try a malformed `set charset'.
90 gdb_test "set charset" \
91 "Requires an argument. Valid arguments are.*" \
92 "try malformed `set charset'"
94 # Try using `set host-charset' on an invalid character set.
95 gdb_test "set host-charset my_grandma_bonnie" \
96 "Undefined item: \"my_grandma_bonnie\"." \
97 "try `set host-charset' with invalid charset"
99 # Try using `set target-charset' on an invalid character set.
100 gdb_test "set target-charset my_grandma_bonnie" \
101 "Undefined item: \"my_grandma_bonnie\"." \
102 "try `set target-charset' with invalid charset"
104 # A Tcl array mapping the names of all the character sets we've seen
105 # to "1" if the character set can be used as a host character set, or
106 # "0" otherwise. We can use `array names charsets' just to get a list
107 # of all character sets.
108 array set charsets {}
110 proc all_charset_names {} {
112 return [array names charsets]
115 proc valid_host_charset {charset} {
117 return [expr {[info exists charsets($charset)] && $charsets($charset)}]
120 proc valid_target_charset {charset} {
122 return [info exists charsets($charset)]
125 send_gdb "set host-charset\n"
127 -re "Requires an argument. Valid arguments are (.*)\\.\r\n$gdb_prompt $" {
128 set host_charset_list $expect_out(1,string)
129 regsub -all {, } $host_charset_list {,} host_charset_list
130 foreach host_charset [split $host_charset_list ","] {
131 set charsets($host_charset) 1
133 pass "capture valid host charsets"
136 -re ".*$gdb_prompt $" {
137 fail "capture valid host charsets"
141 fail "(timeout) capture valid host charsets"
145 # If gdb was built with a phony iconv, it will only have two character
146 # sets: "auto" and the default. In this situation, this set of tests
148 if {[llength [array names charsets]] < 3} {
149 untested "fewer than 3 charsets"
153 send_gdb "set target-charset\n"
155 -re "Requires an argument. Valid arguments are (.*)\\.\r\n$gdb_prompt $" {
156 set target_charset_list $expect_out(1,string)
157 regsub -all {, } $target_charset_list {,} target_charset_list
158 foreach target_charset [split $target_charset_list ","] {
159 if {! [info exists charsets($target_charset)]} {
160 set charsets($target_charset) 0
163 pass "capture valid target charsets"
166 -re ".*$gdb_prompt $" {
167 fail "capture valid target charsets"
171 fail "(timeout) capture valid target charsets"
175 # We don't want to test all the charset names here, since that would
176 # be too many combinations. We we pick a subset.
177 set charset_subset {ASCII ISO-8859-1 EBCDIC-US IBM1047}
178 foreach host_charset $charset_subset {
179 if {[valid_host_charset $host_charset]} {
181 set testname "try `set host-charset $host_charset'"
182 send_gdb "set host-charset $host_charset\n"
184 -re "GDB doesn't know of any character set named.*\[\r\n]+${gdb_prompt} $" {
185 # How did it get into `charsets' then?
186 fail "$testname (didn't recognize name)"
188 -re "GDB can't use `.*' as its host character set\\.\[\r\n]+${gdb_prompt} $" {
189 # Well, then why does its `charsets' entry say it can?
192 -re "${gdb_prompt} $" {
196 fail "$testname (timeout)"
200 # Check that the command actually had its intended effect:
201 # $host_charset should now be the host character set.
202 send_gdb "show charset\n"
203 set result [parse_show_charset_output "parse `show charset' after `set host-charset $host_charset'"]
204 if {! [string compare [lindex $result 0] $host_charset]} {
205 pass "check effect of `set host-charset $host_charset'"
207 fail "check effect of `set host-charset $host_charset'"
210 # Now try setting every possible target character set,
211 # given that host charset.
212 foreach target_charset $charset_subset {
213 if {![valid_target_charset $target_charset]} {
216 set testname "try `set target-charset $target_charset'"
217 send_gdb "set target-charset $target_charset\n"
219 -re "GDB doesn't know of any character set named.*\[\r\n]+${gdb_prompt} $" {
220 fail "$testname (didn't recognize name)"
222 -re "GDB can't convert from the .* character set to .*\\.\[\r\n\]+${gdb_prompt} $" {
223 # This is a serious problem. GDB should be able to convert
224 # between any arbitrary pair of character sets.
225 fail "$testname (can't convert)"
227 -re "${gdb_prompt} $" {
231 fail "$testname (timeout)"
235 # Check that the command actually had its intended effect:
236 # $target_charset should now be the target charset.
237 send_gdb "show charset\n"
238 set result [parse_show_charset_output "parse `show charset' after `set target-charset $target_charset'"]
239 if {! [string compare $result [list $host_charset $target_charset]]} {
240 pass "check effect of `set target-charset $target_charset'"
242 fail "check effect of `set target-charset $target_charset'"
245 # Test handling of characters in the host charset which
246 # can't be translated into the target charset. \xA2 is
247 # `cent' in ISO-8859-1, which has no equivalent in ASCII.
249 # On some systems, the pseudo-tty through which we
250 # communicate with GDB insists on stripping the high bit
251 # from input characters, meaning that `cent' turns into
252 # `"'. Since ISO-8859-1 and ASCII are identical in the
253 # lower 128 characters, it's tough to see how we can test
254 # this behavior on such systems, so we just xfail it.
256 # Note: the \x16 (Control-V) is an escape to allow \xA2 to
258 if {! [string compare $host_charset iso-8859-1] && ! [string compare $target_charset ascii]} {
260 set testname "untranslatable character in character literal"
261 send_gdb "print '\x16\xA2'\n"
263 -re "There is no character corresponding to .* in the target character set .*\\.\[\r\n\]+$gdb_prompt $" {
266 -re " = 34 '\"'\[\r\n\]+$gdb_prompt $" {
267 xfail "$testname (DejaGNU's pseudo-tty strips eighth bit)"
269 -re "$gdb_prompt $" {
273 fail "$testname (timeout)"
277 set testname "untranslatable character in string literal"
278 # If the PTTY zeros bit seven, then this turns into
280 # which gets us a syntax error. We don't care.
281 send_gdb "print \"\x16\xA2\"\n"
283 -re "There is no character corresponding to .* in the target character set .*\\.\[\r\n\]+$gdb_prompt $" {
286 -re "Unterminated string in expression.\[\r\n\]+$gdb_prompt $" {
287 xfail "$testname (DejaGNU's pseudo-tty strips eighth bit)"
289 -re "$gdb_prompt $" {
293 fail "$testname (timeout)"
297 set testname "untranslatable characters in backslash escape"
298 send_gdb "print '\\\x16\xA2'\n"
300 -re "The escape sequence .* is equivalent to plain .*, which has no equivalent\[\r\n\]+in the .* character set\\.\[\r\n\]+$gdb_prompt $" {
303 -re " = 34 '\"'\[\r\n\]+$gdb_prompt $" {
304 xfail "$testname (DejaGNU's pseudo-tty strips eighth bit)"
306 -re "$gdb_prompt $" {
310 fail "$testname (timeout)"
319 # Set the host character set to plain ASCII, and try actually printing
320 # some strings in various target character sets. We need to run the
321 # test program to the point at which the strings have been
323 gdb_test "break ${srcfile}:[gdb_get_line_number "all strings initialized"]" \
324 ".*Breakpoint.* at .*" \
325 "set breakpoint after all strings have been initialized"
327 gdb_test "" "Breakpoint.*all strings initialized.*" "run until all strings have been initialized"
329 # We only try the wide character tests on machines where the wchar_t
330 # typedef in the test case has the right size.
331 set wchar_size [get_sizeof wchar_t 99]
333 if {$wchar_size == 2} {
334 lappend charset_subset UTF-16
336 } elseif {$wchar_size == 4} {
337 lappend charset_subset UTF-32
341 gdb_test_no_output "set host-charset ASCII"
342 foreach target_charset $charset_subset {
343 if {![valid_target_charset $target_charset]} {
347 if {$target_charset == "UTF-32" || $target_charset == "UTF-16"} {
348 set param target-wide-charset
351 set param target-charset
354 gdb_test_no_output "set $param $target_charset"
356 # Try printing the null character. There seems to be a bug in
357 # gdb_test that requires us to use gdb_expect here.
358 send_gdb "print $L'\\0'\n"
360 -re "\\\$${decimal} = 0 $L'\\\\000'\[\r\n\]+$gdb_prompt $" {
361 pass "print the null character in ${target_charset}"
363 -re "$gdb_prompt $" {
364 fail "print the null character in ${target_charset}"
367 fail "print the null character in ${target_charset} (timeout)"
371 # Compute the name of the variable in the test program that holds
372 # a string in $target_charset. The variable's name is the
373 # character set's name, in lower-case, with all non-identifier
374 # characters replaced with '_', with "_string" stuck on the end.
375 if {$target_charset == "UTF-16"} {
376 # We still use the utf_32_string variable -- but the size is
377 # correct for UTF-16.
378 set var_name utf_32_string
380 set var_name [string tolower "${target_charset}_string"]
381 regsub -all -- "\[^a-z0-9_\]" $var_name "_" var_name
384 # Compute a regexp matching the results we expect. This is static,
385 # but it's easier than writing it out.
386 regsub -all "." "abfnrtv" "(\\\\&|x)" escapes
387 set uppercase "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
388 set lowercase "abcdefghijklmnopqrstuvwxyz"
389 set digits "0123456789"
390 set octal_escape "\\\\\[0-9\]+"
392 send_gdb "print $var_name\n"
393 # ${escapes}${uppercase}${lowercase}${digits}${octal}${octal}
395 -re ".* = $L\"(\\\\a|x)(\\\\b|x)(\\\\f|x)(\\\\n|x)(\\\\r|x)(\\\\t|x)(\\\\v|x)${uppercase}${lowercase}${digits}(${octal_escape}|x)+\"\[\r\n\]+$gdb_prompt $" {
396 pass "print string in $target_charset"
398 -re "$gdb_prompt $" {
399 fail "print string in $target_charset"
402 fail "print string in $target_charset (timeout)"
406 # Try entering a character literal, and see if it comes back unchanged.
407 gdb_test "print $L'A'" \
408 " = \[0-9-\]+ $L'A'" \
409 "parse character literal in ${target_charset}"
411 # Check that the character literal was encoded correctly.
412 gdb_test "print $L'A' == $var_name\[7\]" \
414 "check value of parsed character literal in ${target_charset}"
416 # Try entering a string literal, and see if it comes back unchanged.
417 gdb_test "print $L\"abcdefABCDEF012345\"" \
418 " = $L\"abcdefABCDEF012345\"" \
419 "parse string literal in ${target_charset}"
421 # Check that the string literal was encoded correctly.
422 gdb_test "print $L\"q\"\[0\] == $var_name\[49\]" \
424 "check value of parsed string literal in ${target_charset}"
426 # Test handling of characters in the target charset which
427 # can't be translated into the host charset.
428 if {! [string compare $target_charset iso-8859-1]} {
429 gdb_test "print iso_8859_1_string\[69\]" \
430 " = \[0-9-\]+ '\\\\242'" \
431 "print character with no equivalent in host character set"
432 gdb_test "print iso_8859_1_string + 70" \
433 " = ${hex} \"\\\\242.*\"" \
434 "print string with no equivalent in host character set"
437 # Make sure that we don't apply the ISO-8859-1 `print_literally'
439 if {! [string compare $target_charset ascii]} {
440 gdb_test "print iso_8859_1_string\[69\]" \
441 " = \[0-9-\]+ '\\\\242'" \
442 "print ASCII unprintable character"
443 gdb_test "print iso_8859_1_string + 70" \
444 " = ${hex} \"\\\\242.*\"" \
445 "print ASCII unprintable string"
448 # Try printing characters with backslash escape equivalents.
449 set escapees {a b f n r t v}
450 for {set i 0} {$i < [llength $escapees]} {incr i} {
451 set escape [lindex $escapees $i]
452 send_gdb "print $var_name\[$i\]\n"
455 -re "= \[0-9-\]+ $L'\\\\${escape}'\[\r\n\]+$gdb_prompt $" {
456 pass "try printing '\\${escape}' in ${target_charset}"
458 -re "= \[0-9-\]+ 'x'\[\r\n\]+$gdb_prompt $" {
459 xfail "try printing '\\${escape}' in ${target_charset} (no such escape)"
462 -re "$gdb_prompt $" {
463 fail "try printing '\\${escape}' in ${target_charset}"
466 fail "try printing '\\${escape}' in ${target_charset} (timeout)"
472 # Try parsing a backslash escape in a character literal.
473 gdb_test "print $L'\\${escape}' == $var_name\[$i\]" \
475 "check value of '\\${escape}' in ${target_charset}"
477 # Try parsing a backslash escape in a string literal.
478 gdb_test "print $L\"\\${escape}\"\[0\] == $var_name\[$i\]" \
480 "check value of \"\\${escape}\" in ${target_charset}"
484 # Try printing a character escape that doesn't exist. We should
485 # get the unescaped character, in the target character set.
486 gdb_test "print $L'\\q'" " = \[0-9-\]+ $L'q'" \
487 "print escape that doesn't exist in $target_charset"
488 gdb_test "print $L'\\q' == $var_name\[49\]" " = 1" \
489 "check value of escape that doesn't exist in $target_charset"
492 # Reset the target charset.
493 gdb_test_no_output "set target-charset UTF-8"
495 # \242 is not a valid UTF-8 character.
496 gdb_test "print \"\\242\"" " = \"\\\\242\"" \
497 "non-representable target character"
499 gdb_test "print '\\x'" "\\\\x escape without a following hex digit."
500 gdb_test "print '\\u'" "\\\\u escape without a following hex digit."
501 gdb_test "print '\\9'" " = \[0-9\]+ '9'"
503 # An octal escape can only be 3 digits.
504 gdb_test "print \"\\1011\"" " = \"A1\""
506 # Tests for wide- or unicode- strings. L is the prefix letter to use,
507 # either "L" (for wide strings), "u" (for UTF-16), or "U" (for UTF-32).
508 # NAME is used in the test names and should be related to the prefix
509 # letter in some easy-to-undestand way.
510 proc test_wide_or_unicode {L name} {
511 gdb_test "print $L\"ab\" $L\"c\"" " = $L\"abc\"" \
512 "basic $name string concatenation"
513 gdb_test "print $L\"ab\" \"c\"" " = $L\"abc\"" \
514 "narrow and $name string concatenation"
515 gdb_test "print \"ab\" $L\"c\"" " = $L\"abc\"" \
516 "$name and narrow string concatenation"
517 gdb_test "print $L\"\\xe\" $L\"c\"" " = $L\"\\\\016c\"" \
518 "$name string concatenation with escape"
519 gdb_test "print $L\"\" \"abcdef\" \"g\"" \
521 "concatenate three strings with empty $name string"
523 gdb_test "print $L'a'" "= \[0-9\]+ $L'a'" \
524 "basic $name character"
528 test_wide_or_unicode L wide
531 set ucs2_ok [expr {[get_sizeof char16_t 99] == 2}]
533 if ![valid_host_charset "UTF-16"] {
534 verbose -log "Disabling UTF-16 tests."
539 test_wide_or_unicode u UTF-16
542 set ucs4_ok [expr {[get_sizeof char32_t 99] == 4}]
544 test_wide_or_unicode U UTF-32
547 # Test an invalid string combination.
548 proc test_combination {L1 name1 L2 name2} {
549 gdb_test "print $L1\"abc\" $L2\"def\"" \
550 "Undefined string concatenation." \
551 "undefined concatenation of $name1 and $name2"
554 if {$wchar_ok && $ucs2_ok} {
555 test_combination L wide u UTF-16
557 if {$wchar_ok && $ucs4_ok} {
558 test_combination L wide U UTF-32
559 # Regression test for a typedef to a typedef.
560 gdb_test "print myvar" "= \[0-9\]+ L'A'" \
563 if {$ucs2_ok && $ucs4_ok} {
564 test_combination u UTF-16 U UTF-32
569 gdb_test_multiple "python print ('hello, world!')" \
570 "verify python support for charset tests" {
571 -re "not supported.*$gdb_prompt $" {
572 unsupported "python support is disabled"
575 -re "$gdb_prompt $" {}
579 gdb_test "print u\"abcdef\"" " = u\"abcdef\"" \
580 "set up for python printing of utf-16 string"
582 gdb_test "python print (gdb.history(0).string())" "abcdef" \
583 "extract utf-16 string using python"
587 # Regression test for a cleanup bug in the charset code.
588 gdb_test "print 'a' == 'a' || 'b' == 'b'" \
590 "EVAL_SKIP cleanup handling regression test"
593 proc string_display { var_name set_prefix x_size x_type} {
594 gdb_test_no_output "set ${var_name} = ${set_prefix}\"Test String\\0with zeroes\"" "assign ${var_name} with prefix ${set_prefix}"
595 gdb_test "x /2${x_size}s ${var_name}" ".*\t${x_type}\"Test String\"\[\r\n\]+.*\t${x_type}\"with zeroes\"" "display String ${var_name} with x/${x_size}s"
599 string_display String16 u h u
600 if {$wchar_size == 2} {
601 string_display String16 L h u
605 string_display String32 U w U
606 if {$wchar_size == 4} {
607 string_display String32 L w U
611 foreach name {short int long} {
612 # We're really just checking to make sure this doesn't give an
614 gdb_test "print ${name}_array = \"hi\"" \
616 "assign string to $name array"