* [PATCH v4 0/8] gdb: split array and string limiting options
@ 2022-03-09 19:17 Maciej W. Rozycki
2022-03-09 19:17 ` [PATCH v4 1/8] GDB: Remove extraneous full stops from `set' command error messages Maciej W. Rozycki
` (8 more replies)
0 siblings, 9 replies; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-09 19:17 UTC (permalink / raw)
To: gdb-patches; +Cc: Andrew Burgess, Simon Marchi, Tom Tromey, Simon Sobisch
Hi,
I have retained the original subject for the cover letter for threading
purposes, however this has now otherwise become a small patch series and
Andrew's original change has a different heading now.
This is v4 for Andrew's original proposal to split the `set/show print
elements' options into two, so that there are separate `set/show print
characters' options to control how many characters of a string are
printed. Following earlier discussion for compatibility with existing
setups I have implemented a mechanism to make the `set print characters'
setting follow that of `set print elements' by default. Consequently what
was a single change has now grown into a patch series.
In this series I fix some preexisting issues in the area first, implement
the general framework for extra literals used with integer settings second
and then add the original change for `set print characters' from Andrew,
adjusted accordingly, followed by a testsuite update as a separate change.
This is so as to ensure the semantics of the existing `set print elements'
knob hasn't changed in the default `set print characters' state (a small
update for the test suite is nevertheless included with the `set print
characters' command proper, because inevitably the relevant commands that
list commands now print additional entries, plus there's been a minimal
test case update included for the verification of setting defaults).
I think there's little more to say here except that changes in the series
have been individually regression-tested with a native `x86_64-linux-gnu'
configuration, and that I have visually examined the result of changes to
the GDB manual in the info and PDF formats.
See individual change descriptions for details.
OK to apply?
Maciej
^ permalink raw reply [flat|nested] 22+ messages in thread
* [PATCH v4 1/8] GDB: Remove extraneous full stops from `set' command error messages
2022-03-09 19:17 [PATCH v4 0/8] gdb: split array and string limiting options Maciej W. Rozycki
@ 2022-03-09 19:17 ` Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 2/8] GDB/Python: Use None for `var_zuinteger_unlimited' value set to `unlimited' Maciej W. Rozycki
` (7 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-09 19:17 UTC (permalink / raw)
To: gdb-patches; +Cc: Andrew Burgess, Simon Marchi, Tom Tromey, Simon Sobisch
With errors given for bad commands such as `set annotate' or `set width'
we produce an extraneous full stop within parentheses:
(gdb) set annotate
Argument required (integer to set it to.).
(gdb) set width
Argument required (integer to set it to, or "unlimited".).
(gdb)
This is grammatically incorrect, so remove the full stop and adjust the
testsuite accordingly.
---
New change in v4.
---
gdb/cli/cli-setshow.c | 10 +++++-----
gdb/testsuite/gdb.base/with.exp | 8 ++++----
2 files changed, 9 insertions(+), 9 deletions(-)
gdb-parse-cli-var-integer-msg-stop.diff
Index: src/gdb/cli/cli-setshow.c
===================================================================
--- src.orig/gdb/cli/cli-setshow.c
+++ src/gdb/cli/cli-setshow.c
@@ -195,9 +195,9 @@ parse_cli_var_uinteger (var_types var_ty
if (*arg == nullptr || **arg == '\0')
{
if (var_type == var_uinteger)
- error_no_arg (_("integer to set it to, or \"unlimited\"."));
+ error_no_arg (_("integer to set it to, or \"unlimited\""));
else
- error_no_arg (_("integer to set it to."));
+ error_no_arg (_("integer to set it to"));
}
if (var_type == var_uinteger && is_unlimited_literal (arg, expression))
@@ -228,7 +228,7 @@ parse_cli_var_zuinteger_unlimited (const
LONGEST val;
if (*arg == nullptr || **arg == '\0')
- error_no_arg (_("integer to set it to, or \"unlimited\"."));
+ error_no_arg (_("integer to set it to, or \"unlimited\""));
if (is_unlimited_literal (arg, expression))
val = -1;
@@ -418,9 +418,9 @@ do_set_command (const char *arg, int fro
if (*arg == '\0')
{
if (c->var->type () == var_integer)
- error_no_arg (_("integer to set it to, or \"unlimited\"."));
+ error_no_arg (_("integer to set it to, or \"unlimited\""));
else
- error_no_arg (_("integer to set it to."));
+ error_no_arg (_("integer to set it to"));
}
if (c->var->type () == var_integer && is_unlimited_literal (&arg, true))
Index: src/gdb/testsuite/gdb.base/with.exp
===================================================================
--- src.orig/gdb/testsuite/gdb.base/with.exp
+++ src/gdb/testsuite/gdb.base/with.exp
@@ -99,14 +99,14 @@ with_test_prefix "maint" {
test_with_error "boolean" "2" "\"on\" or \"off\" expected\\."
test_with_error "uinteger" "-1" "integer -1 out of range"
test_with_error "uinteger" "" \
- "Argument required \\(integer to set it to, or \"unlimited\"\\.\\)\\."
+ "Argument required \\(integer to set it to, or \"unlimited\"\\)\\."
test_with_error "zuinteger" "-1" "integer -1 out of range"
test_with_error "zuinteger" "" \
- "Argument required \\(integer to set it to\\.\\)\\."
+ "Argument required \\(integer to set it to\\)\\."
test_with_error "zuinteger-unlimited" "-2" \
"only -1 is allowed to set as unlimited"
test_with_error "zuinteger-unlimited" "" \
- "Argument required \\(integer to set it to, or \"unlimited\"\\.\\)\\."
+ "Argument required \\(integer to set it to, or \"unlimited\"\\)\\."
test_with_error "filename" "" \
"Argument required \\(filename to set it to\\.\\)\\."
test_with_error "enum" "" \
@@ -243,7 +243,7 @@ with_test_prefix "errors" {
"Cannot use this setting with the \"with\" command"
gdb_test "with print elements -- p 1" \
- "Argument required \\(integer to set it to, or \"unlimited\"\\.\\)\\."
+ "Argument required \\(integer to set it to, or \"unlimited\"\\)\\."
gdb_test "with -- p 1" \
"Missing setting before '--' delimiter"
^ permalink raw reply [flat|nested] 22+ messages in thread
* [PATCH v4 2/8] GDB/Python: Use None for `var_zuinteger_unlimited' value set to `unlimited'
2022-03-09 19:17 [PATCH v4 0/8] gdb: split array and string limiting options Maciej W. Rozycki
2022-03-09 19:17 ` [PATCH v4 1/8] GDB: Remove extraneous full stops from `set' command error messages Maciej W. Rozycki
@ 2022-03-09 19:18 ` Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 3/8] GDB: Add `NUMBER' completion to `set' integer commands Maciej W. Rozycki
` (6 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-09 19:18 UTC (permalink / raw)
To: gdb-patches; +Cc: Andrew Burgess, Simon Marchi, Tom Tromey, Simon Sobisch
Consistently with the `var_integer' and `var_uinteger' parameters return
the special value of None for a `var_zuinteger_unlimited' parameter set
to `unlimited' by using the Py_RETURN_NONE macro in this case, fixing
commit 0489430a0e1a ("Handle var_zuinteger and var_zuinteger_unlimited
from Python"); cf. PR python/20084. Adjust the testsuite accordingly.
---
New change in v4.
---
gdb/python/python.c | 10 +++++++++-
gdb/testsuite/gdb.python/py-parameter.exp | 4 ++--
2 files changed, 11 insertions(+), 3 deletions(-)
gdb-python-var-zuinteger-unlimited-none.diff
Index: src/gdb/python/python.c
===================================================================
--- src.orig/gdb/python/python.c
+++ src/gdb/python/python.c
@@ -512,9 +512,17 @@ gdbpy_parameter_value (const setting &va
Py_RETURN_NONE;
/* Fall through. */
case var_zinteger:
- case var_zuinteger_unlimited:
return gdb_py_object_from_longest (var.get<int> ()).release ();
+ case var_zuinteger_unlimited:
+ {
+ int val = var.get<int> ();
+
+ if (val == -1)
+ Py_RETURN_NONE;
+ return gdb_py_object_from_longest (val).release ();
+ }
+
case var_uinteger:
{
unsigned int val = var.get<unsigned int> ();
Index: src/gdb/testsuite/gdb.python/py-parameter.exp
===================================================================
--- src.orig/gdb/testsuite/gdb.python/py-parameter.exp
+++ src/gdb/testsuite/gdb.python/py-parameter.exp
@@ -343,8 +343,8 @@ proc_with_prefix test_integer_parameter
} elseif {$kind == "PARAM_ZUINTEGER_UNLIMITED"} {
gdb_test_no_output "python test_param_$kind.value = -1" \
"check that PARAM_ZUINTEGER value can be set to -1"
- gdb_test "python print(gdb.parameter('test-$kind'))" "-1" \
- "check that PARAM_ZUINTEGER value is -1 after setting"
+ gdb_test "python print(gdb.parameter('test-$kind'))" "None" \
+ "check that PARAM_ZUINTEGER value is None after setting"
} else {
error "invalid kind: $kind"
}
^ permalink raw reply [flat|nested] 22+ messages in thread
* [PATCH v4 3/8] GDB: Add `NUMBER' completion to `set' integer commands
2022-03-09 19:17 [PATCH v4 0/8] gdb: split array and string limiting options Maciej W. Rozycki
2022-03-09 19:17 ` [PATCH v4 1/8] GDB: Remove extraneous full stops from `set' command error messages Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 2/8] GDB/Python: Use None for `var_zuinteger_unlimited' value set to `unlimited' Maciej W. Rozycki
@ 2022-03-09 19:18 ` Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 4/8] GDB/testsuite: Tighten `set print elements' error check Maciej W. Rozycki
` (5 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-09 19:18 UTC (permalink / raw)
To: gdb-patches; +Cc: Andrew Burgess, Simon Marchi, Tom Tromey, Simon Sobisch
Fix a completion consistency issue with `set' commands accepting integer
values and the special `unlimited' keyword:
(gdb) complete print -elements
print -elements NUMBER
print -elements unlimited
(gdb)
vs:
(gdb) complete set print elements
set print elements unlimited
(gdb)
(there is a space entered at the end of both commands, not shown here)
which also means if you strike <Tab> with `set print elements ' input,
it will, annoyingly, complete to `set print elements unlimited' right
away rather than showing a choice between `NUMBER' and `unlimited'.
Add `NUMBER' then as an available completion for such `set' commands:
(gdb) complete set print elements
set print elements NUMBER
set print elements unlimited
(gdb)
Adjust the testsuite accordingly.
---
New change in v4.
---
gdb/cli/cli-decode.c | 2 ++
gdb/testsuite/gdb.base/settings.exp | 8 +++++++-
2 files changed, 9 insertions(+), 1 deletion(-)
gdb-integer-complete-number.diff
Index: src/gdb/cli/cli-decode.c
===================================================================
--- src.orig/gdb/cli/cli-decode.c
+++ src/gdb/cli/cli-decode.c
@@ -989,6 +989,8 @@ integer_unlimited_completer (struct cmd_
NULL,
};
+ if (*text == '\0')
+ tracker.add_completion (make_unique_xstrdup ("NUMBER"));
complete_on_enum (tracker, keywords, text, word);
}
Index: src/gdb/testsuite/gdb.base/settings.exp
===================================================================
--- src.orig/gdb/testsuite/gdb.base/settings.exp
+++ src/gdb/testsuite/gdb.base/settings.exp
@@ -215,8 +215,14 @@ proc test-integer {variant} {
test_gdb_complete_none \
"$set_cmd "
} else {
+ test_gdb_complete_multiple "$set_cmd " "" "" {
+ "NUMBER"
+ "unlimited"
+ }
+ test_gdb_complete_none \
+ "$set_cmd 1"
test_gdb_complete_unique \
- "$set_cmd " \
+ "$set_cmd u" \
"$set_cmd unlimited"
}
^ permalink raw reply [flat|nested] 22+ messages in thread
* [PATCH v4 4/8] GDB/testsuite: Tighten `set print elements' error check
2022-03-09 19:17 [PATCH v4 0/8] gdb: split array and string limiting options Maciej W. Rozycki
` (2 preceding siblings ...)
2022-03-09 19:18 ` [PATCH v4 3/8] GDB: Add `NUMBER' completion to `set' integer commands Maciej W. Rozycki
@ 2022-03-09 19:18 ` Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 5/8] GDB/testsuite: Add coverage for `print -elements' command Maciej W. Rozycki
` (4 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-09 19:18 UTC (permalink / raw)
To: gdb-patches; +Cc: Andrew Burgess, Simon Marchi, Tom Tromey, Simon Sobisch
Match the whole error message expected to be given rather than omitting
the part about the "unlimited" keyword. There's no point in omitting
the missing part first, and second with an upcoming change the part in
parentheses will no longer be a fixed string, so doing a full match will
ensure the algorithm correctly builds the message expected here. Also
avoid any wildcard matches.
---
New change in v4.
---
gdb/testsuite/gdb.base/default.exp | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
gdb-test-set-print-elements-tightening.diff
Index: src/gdb/testsuite/gdb.base/default.exp
===================================================================
--- src.orig/gdb/testsuite/gdb.base/default.exp
+++ src/gdb/testsuite/gdb.base/default.exp
@@ -525,7 +525,8 @@ gdb_test_no_output "set print asm-demang
#test set print demangle
gdb_test_no_output "set print demangle" "set print demangle"
#test set print elements
-gdb_test "set print elements" "Argument required .integer to set it to.*"
+gdb_test "set print elements" \
+ "Argument required \\(integer to set it to, or \"unlimited\"\\)\\."
#test set print object
gdb_test_no_output "set print object" "set print object"
#test set print pretty
^ permalink raw reply [flat|nested] 22+ messages in thread
* [PATCH v4 5/8] GDB/testsuite: Add coverage for `print -elements' command
2022-03-09 19:17 [PATCH v4 0/8] gdb: split array and string limiting options Maciej W. Rozycki
` (3 preceding siblings ...)
2022-03-09 19:18 ` [PATCH v4 4/8] GDB/testsuite: Tighten `set print elements' error check Maciej W. Rozycki
@ 2022-03-09 19:18 ` Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 6/8] GDB: Allow arbitrary keywords in integer set commands Maciej W. Rozycki
` (3 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-09 19:18 UTC (permalink / raw)
To: gdb-patches; +Cc: Andrew Burgess, Simon Marchi, Tom Tromey, Simon Sobisch
We currently have no coverage for the `print -elements ...' command (or
`p -elements ...' in the shortened form), so add a couple of test cases
mimicking ones using corresponding `set print elements ...' values.
---
New change in v4.
---
gdb/testsuite/gdb.base/printcmds.exp | 9 +++++++++
1 file changed, 9 insertions(+)
gdb-test-printcmd-options.diff
Index: src/gdb/testsuite/gdb.base/printcmds.exp
===================================================================
--- src.orig/gdb/testsuite/gdb.base/printcmds.exp
+++ src/gdb/testsuite/gdb.base/printcmds.exp
@@ -519,6 +519,15 @@ proc test_print_strings {} {
gdb_test "p teststring" \
" = (.unsigned char .. )?\"teststring contents\"" "p teststring with elements set to 20"
+ gdb_test "p -elements 1 -- teststring" \
+ " = (.unsigned char .. )?\"t\"\\.\\.\\."
+ gdb_test "p -elements 5 -- teststring" \
+ " = (.unsigned char .. )?\"tests\"\\.\\.\\."
+ gdb_test "p -elements 19 -- teststring" \
+ " = (.unsigned char .. )?\"teststring contents\""
+ gdb_test "p -elements 20 -- teststring" \
+ " = (.unsigned char .. )?\"teststring contents\""
+
gdb_test "print teststring2" \
" = \\(charptr\\) \"more contents\""
^ permalink raw reply [flat|nested] 22+ messages in thread
* [PATCH v4 6/8] GDB: Allow arbitrary keywords in integer set commands
2022-03-09 19:17 [PATCH v4 0/8] gdb: split array and string limiting options Maciej W. Rozycki
` (4 preceding siblings ...)
2022-03-09 19:18 ` [PATCH v4 5/8] GDB/testsuite: Add coverage for `print -elements' command Maciej W. Rozycki
@ 2022-03-09 19:18 ` Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 7/8] GDB: Add a character string limiting option Maciej W. Rozycki
` (2 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-09 19:18 UTC (permalink / raw)
To: gdb-patches; +Cc: Andrew Burgess, Simon Marchi, Tom Tromey, Simon Sobisch
Rather than just `unlimited' allow the integer set commands (or command
options) to define arbitrary keywords for the user to use, removing
hardcoded arrangements for the `unlimited' keyword.
Make obsolete the confusingly named `var_zinteger', `var_zuinteger' and
`var_zuinteger_unlimited' `set'/`show' command variable types redefining
them in terms of `var_uinteger', `var_integer' and `var_pinteger', which
have the range of [0;UINT_MAX], [INT_MIN;INT_MAX], and [0;INT_MAX] each.
Following existing practice `var_pinteger' allows extra negative values
to be used, however unlike `var_zuinteger_unlimited' any number of such
values can be defined rather than just `-1'.
The "p" in `var_pinteger' stands for "positive", for the lack of a more
appropriate unambiguous letter, even though 0 obviously is not positive;
"n" would be confusing as to whether it stands for "non-negative" or
"negative".
Add a new structure, `literal_def', the entries of which define extra
keywords allowed for a command and numerical values they correspond to.
Those values are not verified against the basic range supported by the
underlying variable type, allowing extra values to be allowed outside
that range, which may or may not be individually made visible to the
user. An optional value translation is possible with the structure to
follow the existing practice for some commands where user-entered 0 is
internally translated to UINT_MAX or INT_MAX. Such translation can now
be arbitrary. Literals defined by this structure are automatically used
for completion as necessary.
So for example:
const literal_def integer_unlimited_literals[] =
{
{ "unlimited", 0, INT_MAX, true },
{ nullptr }
};
defines an extra `unlimited' keyword and a user-visible 0 value, both of
which get translated to INT_MAX for the setting to be used with.
Similarly:
const literal_def zuinteger_unlimited_literals[] =
{
{ "unlimited", -1, -1, true },
{ nullptr }
};
defines the same keyword and a corresponding user-visible -1 value that is
used for the requested setting. If the last member was set to `false'
here, then only the keyword would be allowed for the user to enter and
while -1 would still be used internally trying to enter it as a part of a
command would result in an "integer -1 out of range" error.
The obsolete command variable types still remain for Guile and Python
use. Eventually they need to be removed, however that would require a
further rework of our Guile and Python support code. Given that these
variable types have been exported as keywords for those languages, those
definitions have to stay, however it would be better to redefine them in
terms of the modern variable types and overall export a better interface
that reflects the flexibility of our core code now.
There is no user-visible change from this rework, so no documentation or
testsuite updates.
---
Hi,
NB I have observed that the value of INT_MIN is not correctly handled:
internally -2147483648 is interpreted as 2147483648 and consequently such
a value is rejected, e.g:
(gdb) show heuristic-fence-post
The distance searched for the start of a function is 0.
(gdb) set heuristic-fence-post -2147483647
(gdb) show heuristic-fence-post
The distance searched for the start of a function is -2147483647.
(gdb) set heuristic-fence-post -2147483648
integer 2147483648 out of range
(gdb)
umm... (not that it makes sense to set negative values for the heuristic
fence post, but that's irrelevant here). I haven't tried to track the
root cause for this.
Maciej
New change in v4.
---
gdb/cli/cli-cmds.c | 56 ++++----
gdb/cli/cli-decode.c | 302 ++++++++++++++++++++++++++++++++++++++---------
gdb/cli/cli-option.c | 116 +++++++++++-------
gdb/cli/cli-option.h | 34 +++--
gdb/cli/cli-setshow.c | 249 ++++++++++++++++++--------------------
gdb/cli/cli-setshow.h | 20 +--
gdb/command.h | 108 ++++++++++++++--
gdb/guile/scm-param.c | 102 +++++++++------
gdb/maint-test-options.c | 4
gdb/python/py-param.c | 39 ++++--
gdb/python/python.c | 49 +++----
gdb/valprint.c | 9 -
12 files changed, 716 insertions(+), 372 deletions(-)
gdb-setshow-cmd-extra-literals.diff
Index: src/gdb/cli/cli-cmds.c
===================================================================
--- src.orig/gdb/cli/cli-cmds.c
+++ src/gdb/cli/cli-cmds.c
@@ -2200,22 +2200,37 @@ value_from_setting (const setting &var,
{
switch (var.type ())
{
+ case var_uinteger:
case var_integer:
- if (var.get<int> () == INT_MAX)
- return value_from_longest (builtin_type (gdbarch)->builtin_int,
- 0);
- else
- return value_from_longest (builtin_type (gdbarch)->builtin_int,
- var.get<int> ());
- case var_zinteger:
- return value_from_longest (builtin_type (gdbarch)->builtin_int,
- var.get<int> ());
+ case var_pinteger:
+ {
+ LONGEST value
+ = (var.type () == var_uinteger
+ ? static_cast<LONGEST> (var.get<unsigned int> ())
+ : static_cast<LONGEST> (var.get<int> ()));
+
+ if (var.extra_literals () != nullptr)
+ for (const literal_def *l = var.extra_literals ();
+ l->literal != nullptr;
+ l++)
+ if (value == l->use)
+ {
+ value = l->val;
+ break;
+ }
+
+ if (var.type () == var_uinteger)
+ return
+ value_from_ulongest (builtin_type (gdbarch)->builtin_unsigned_int,
+ static_cast<const unsigned int> (value));
+ else
+ return
+ value_from_longest (builtin_type (gdbarch)->builtin_int,
+ static_cast<const int> (value));
+ }
case var_boolean:
return value_from_longest (builtin_type (gdbarch)->builtin_int,
var.get<bool> () ? 1 : 0);
- case var_zuinteger_unlimited:
- return value_from_longest (builtin_type (gdbarch)->builtin_int,
- var.get<int> ());
case var_auto_boolean:
{
int val;
@@ -2237,17 +2252,6 @@ value_from_setting (const setting &var,
return value_from_longest (builtin_type (gdbarch)->builtin_int,
val);
}
- case var_uinteger:
- if (var.get<unsigned int> () == UINT_MAX)
- return value_from_ulongest
- (builtin_type (gdbarch)->builtin_unsigned_int, 0);
- else
- return value_from_ulongest
- (builtin_type (gdbarch)->builtin_unsigned_int,
- var.get<unsigned int> ());
- case var_zuinteger:
- return value_from_ulongest (builtin_type (gdbarch)->builtin_unsigned_int,
- var.get<unsigned int> ());
case var_string:
case var_string_noescape:
case var_optional_filename:
@@ -2317,13 +2321,11 @@ str_value_from_setting (const setting &v
{
switch (var.type ())
{
+ case var_uinteger:
case var_integer:
- case var_zinteger:
+ case var_pinteger:
case var_boolean:
- case var_zuinteger_unlimited:
case var_auto_boolean:
- case var_uinteger:
- case var_zuinteger:
{
std::string cmd_val = get_setshow_command_value_string (var);
Index: src/gdb/cli/cli-decode.c
===================================================================
--- src.orig/gdb/cli/cli-decode.c
+++ src/gdb/cli/cli-decode.c
@@ -580,11 +580,51 @@ add_setshow_cmd_full_erased (const char
return {set, show};
}
+/* Completes on integer commands that support extra literals. */
+
+static void
+integer_literals_completer (struct cmd_list_element *c,
+ completion_tracker &tracker,
+ const char *text, const char *word)
+{
+ const literal_def *extra_literals = c->var->extra_literals ();
+
+ if (*text == '\0')
+ {
+ tracker.add_completion (make_unique_xstrdup ("NUMBER"));
+ for (const literal_def *l = extra_literals;
+ l->literal != nullptr;
+ l++)
+ tracker.add_completion (make_unique_xstrdup (l->literal));
+ }
+ else
+ for (const literal_def *l = extra_literals;
+ l->literal != nullptr;
+ l++)
+ if (startswith (l->literal, text))
+ tracker.add_completion (make_unique_xstrdup (l->literal));
+}
+
+/* Add element named NAME to both the command SET_LIST and SHOW_LIST.
+ THECLASS is as in add_cmd. VAR_TYPE is the kind of thing we are
+ setting. VAR is address of the variable being controlled by this
+ command. EXTRA_LITERALS if non-NULL define extra literals to be
+ accepted in lieu of a number for integer variables. If nullptr is
+ given as VAR, then both SET_SETTING_FUNC and GET_SETTING_FUNC must
+ be provided. SET_SETTING_FUNC and GET_SETTING_FUNC are callbacks
+ used to access and modify the underlying property, whatever its
+ storage is. SET_FUNC and SHOW_FUNC are the callback functions
+ (if non-NULL). SET_DOC, SHOW_DOC and HELP_DOC are the
+ documentation strings.
+
+ Return the newly created set and show commands. */
+
template<typename T>
static set_show_commands
add_setshow_cmd_full (const char *name,
enum command_class theclass,
var_types var_type, T *var,
+ const literal_def *extra_literals,
const char *set_doc, const char *show_doc,
const char *help_doc,
typename setting_func_types<T>::set set_setting_func,
@@ -595,18 +635,44 @@ add_setshow_cmd_full (const char *name,
struct cmd_list_element **show_list)
{
auto erased_args
- = setting::erase_args (var_type, var,
+ = setting::erase_args (var_type, var, extra_literals,
set_setting_func, get_setting_func);
+ auto cmds = add_setshow_cmd_full_erased (name,
+ theclass,
+ var_type, erased_args,
+ set_doc, show_doc,
+ help_doc,
+ set_func,
+ show_func,
+ set_list,
+ show_list);
- return add_setshow_cmd_full_erased (name,
- theclass,
- var_type, erased_args,
- set_doc, show_doc,
- help_doc,
- set_func,
- show_func,
- set_list,
- show_list);
+ if (extra_literals != nullptr)
+ set_cmd_completer (cmds.set, integer_literals_completer);
+
+ return cmds;
+}
+
+/* Same as above but omitting EXTRA_LITERALS. */
+
+template<typename T>
+static set_show_commands
+add_setshow_cmd_full (const char *name,
+ enum command_class theclass,
+ var_types var_type, T *var,
+ const char *set_doc, const char *show_doc,
+ const char *help_doc,
+ typename setting_func_types<T>::set set_setting_func,
+ typename setting_func_types<T>::get get_setting_func,
+ cmd_func_ftype *set_func,
+ show_value_ftype *show_func,
+ struct cmd_list_element **set_list,
+ struct cmd_list_element **show_list)
+{
+ return add_setshow_cmd_full (name, theclass, var_type, var, nullptr,
+ set_doc, show_doc, help_doc,
+ set_setting_func, get_setting_func,
+ set_func, show_func, set_list, show_list);
}
/* Add element named NAME to command list LIST (the list for set or
@@ -975,25 +1041,6 @@ add_setshow_optional_filename_cmd (const
return cmds;
}
-/* Completes on literal "unlimited". Used by integer commands that
- support a special "unlimited" value. */
-
-static void
-integer_unlimited_completer (struct cmd_list_element *ignore,
- completion_tracker &tracker,
- const char *text, const char *word)
-{
- static const char * const keywords[] =
- {
- "unlimited",
- NULL,
- };
-
- if (*text == '\0')
- tracker.add_completion (make_unique_xstrdup ("NUMBER"));
- complete_on_enum (tracker, keywords, text, word);
-}
-
/* Add element named NAME to both the set and show command LISTs (the
list for set/show or some sublist thereof). CLASS is as in
add_cmd. VAR is address of the variable which will contain the
@@ -1002,6 +1049,55 @@ integer_unlimited_completer (struct cmd_
set_show_commands
add_setshow_integer_cmd (const char *name, enum command_class theclass,
+ int *var, const literal_def *extra_literals,
+ const char *set_doc, const char *show_doc,
+ const char *help_doc,
+ cmd_func_ftype *set_func,
+ show_value_ftype *show_func,
+ struct cmd_list_element **set_list,
+ struct cmd_list_element **show_list)
+{
+ set_show_commands commands
+ = add_setshow_cmd_full<int> (name, theclass, var_integer, var,
+ extra_literals, set_doc, show_doc,
+ help_doc, nullptr, nullptr, set_func,
+ show_func, set_list, show_list);
+ return commands;
+}
+
+/* Same as above but using a getter and a setter function instead of a pointer
+ to a global storage buffer. */
+
+set_show_commands
+add_setshow_integer_cmd (const char *name, command_class theclass,
+ const literal_def *extra_literals,
+ const char *set_doc, const char *show_doc,
+ const char *help_doc,
+ setting_func_types<int>::set set_func,
+ setting_func_types<int>::get get_func,
+ show_value_ftype *show_func,
+ cmd_list_element **set_list,
+ cmd_list_element **show_list)
+{
+ auto cmds = add_setshow_cmd_full<int> (name, theclass, var_integer, nullptr,
+ extra_literals, set_doc, show_doc,
+ help_doc, set_func, get_func, nullptr,
+ show_func, set_list, show_list);
+ return cmds;
+}
+
+/* Accept `unlimited' or 0, translated internally to INT_MAX. */
+const literal_def integer_unlimited_literals[] =
+ {
+ { "unlimited", 0, INT_MAX, true },
+ { nullptr }
+ };
+
+/* Same as above but using `integer_unlimited_literals', with a pointer
+ to a global storage buffer. */
+
+set_show_commands
+add_setshow_integer_cmd (const char *name, enum command_class theclass,
int *var,
const char *set_doc, const char *show_doc,
const char *help_doc,
@@ -1012,12 +1108,10 @@ add_setshow_integer_cmd (const char *nam
{
set_show_commands commands
= add_setshow_cmd_full<int> (name, theclass, var_integer, var,
+ integer_unlimited_literals,
set_doc, show_doc, help_doc,
nullptr, nullptr, set_func,
show_func, set_list, show_list);
-
- set_cmd_completer (commands.set, integer_unlimited_completer);
-
return commands;
}
@@ -1035,12 +1129,54 @@ add_setshow_integer_cmd (const char *nam
cmd_list_element **show_list)
{
auto cmds = add_setshow_cmd_full<int> (name, theclass, var_integer, nullptr,
+ integer_unlimited_literals,
set_doc, show_doc, help_doc, set_func,
get_func, nullptr, show_func, set_list,
show_list);
+ return cmds;
+}
- set_cmd_completer (cmds.set, integer_unlimited_completer);
+/* Add element named NAME to both the set and show command LISTs (the
+ list for set/show or some sublist thereof). CLASS is as in
+ add_cmd. VAR is address of the variable which will contain the
+ value. SET_DOC and SHOW_DOC are the documentation strings. */
+
+set_show_commands
+add_setshow_pinteger_cmd (const char *name, enum command_class theclass,
+ int *var, const literal_def *extra_literals,
+ const char *set_doc, const char *show_doc,
+ const char *help_doc,
+ cmd_func_ftype *set_func,
+ show_value_ftype *show_func,
+ struct cmd_list_element **set_list,
+ struct cmd_list_element **show_list)
+{
+ set_show_commands commands
+ = add_setshow_cmd_full<int> (name, theclass, var_pinteger, var,
+ extra_literals, set_doc, show_doc,
+ help_doc, nullptr, nullptr, set_func,
+ show_func, set_list, show_list);
+ return commands;
+}
+/* Same as above but using a getter and a setter function instead of a pointer
+ to a global storage buffer. */
+
+set_show_commands
+add_setshow_pinteger_cmd (const char *name, command_class theclass,
+ const literal_def *extra_literals,
+ const char *set_doc, const char *show_doc,
+ const char *help_doc,
+ setting_func_types<int>::set set_func,
+ setting_func_types<int>::get get_func,
+ show_value_ftype *show_func,
+ cmd_list_element **set_list,
+ cmd_list_element **show_list)
+{
+ auto cmds = add_setshow_cmd_full<int> (name, theclass, var_pinteger, nullptr,
+ extra_literals, set_doc, show_doc,
+ help_doc, set_func, get_func, nullptr,
+ show_func, set_list, show_list);
return cmds;
}
@@ -1051,7 +1187,7 @@ add_setshow_integer_cmd (const char *nam
set_show_commands
add_setshow_uinteger_cmd (const char *name, enum command_class theclass,
- unsigned int *var,
+ unsigned int *var, const literal_def *extra_literals,
const char *set_doc, const char *show_doc,
const char *help_doc,
cmd_func_ftype *set_func,
@@ -1061,12 +1197,9 @@ add_setshow_uinteger_cmd (const char *na
{
set_show_commands commands
= add_setshow_cmd_full<unsigned int> (name, theclass, var_uinteger, var,
- set_doc, show_doc, help_doc,
- nullptr, nullptr, set_func,
+ extra_literals, set_doc, show_doc,
+ help_doc, nullptr, nullptr, set_func,
show_func, set_list, show_list);
-
- set_cmd_completer (commands.set, integer_unlimited_completer);
-
return commands;
}
@@ -1075,6 +1208,7 @@ add_setshow_uinteger_cmd (const char *na
set_show_commands
add_setshow_uinteger_cmd (const char *name, command_class theclass,
+ const literal_def *extra_literals,
const char *set_doc, const char *show_doc,
const char *help_doc,
setting_func_types<unsigned int>::set set_func,
@@ -1084,13 +1218,63 @@ add_setshow_uinteger_cmd (const char *na
cmd_list_element **show_list)
{
auto cmds = add_setshow_cmd_full<unsigned int> (name, theclass, var_uinteger,
- nullptr, set_doc, show_doc,
- help_doc, set_func, get_func,
- nullptr, show_func, set_list,
+ nullptr, extra_literals,
+ set_doc, show_doc, help_doc,
+ set_func, get_func, nullptr,
+ show_func, set_list,
show_list);
+ return cmds;
+}
- set_cmd_completer (cmds.set, integer_unlimited_completer);
+/* Accept `unlimited' or 0, translated internally to UINT_MAX. */
+const literal_def uinteger_unlimited_literals[] =
+ {
+ { "unlimited", 0, UINT_MAX, true },
+ { nullptr }
+ };
+
+/* Same as above but using `uinteger_unlimited_literals', with a pointer
+ to a global storage buffer. */
+
+set_show_commands
+add_setshow_uinteger_cmd (const char *name, enum command_class theclass,
+ unsigned int *var,
+ const char *set_doc, const char *show_doc,
+ const char *help_doc,
+ cmd_func_ftype *set_func,
+ show_value_ftype *show_func,
+ struct cmd_list_element **set_list,
+ struct cmd_list_element **show_list)
+{
+ set_show_commands commands
+ = add_setshow_cmd_full<unsigned int> (name, theclass, var_uinteger, var,
+ uinteger_unlimited_literals,
+ set_doc, show_doc, help_doc, nullptr,
+ nullptr, set_func, show_func,
+ set_list, show_list);
+ return commands;
+}
+
+/* Same as above but using a getter and a setter function instead of a pointer
+ to a global storage buffer. */
+set_show_commands
+add_setshow_uinteger_cmd (const char *name, command_class theclass,
+ const char *set_doc, const char *show_doc,
+ const char *help_doc,
+ setting_func_types<unsigned int>::set set_func,
+ setting_func_types<unsigned int>::get get_func,
+ show_value_ftype *show_func,
+ cmd_list_element **set_list,
+ cmd_list_element **show_list)
+{
+ auto cmds = add_setshow_cmd_full<unsigned int> (name, theclass, var_uinteger,
+ nullptr,
+ uinteger_unlimited_literals,
+ set_doc, show_doc, help_doc,
+ set_func, get_func, nullptr,
+ show_func, set_list,
+ show_list);
return cmds;
}
@@ -1109,7 +1293,7 @@ add_setshow_zinteger_cmd (const char *na
struct cmd_list_element **set_list,
struct cmd_list_element **show_list)
{
- return add_setshow_cmd_full<int> (name, theclass, var_zinteger, var,
+ return add_setshow_cmd_full<int> (name, theclass, var_integer, var,
set_doc, show_doc, help_doc,
nullptr, nullptr, set_func,
show_func, set_list, show_list);
@@ -1128,12 +1312,22 @@ add_setshow_zinteger_cmd (const char *na
cmd_list_element **set_list,
cmd_list_element **show_list)
{
- return add_setshow_cmd_full<int> (name, theclass, var_zinteger, nullptr,
+ return add_setshow_cmd_full<int> (name, theclass, var_integer, nullptr,
set_doc, show_doc, help_doc, set_func,
get_func, nullptr, show_func, set_list,
show_list);
}
+/* Accept `unlimited' or -1, using -1 internally. */
+const literal_def zuinteger_unlimited_literals[] =
+ {
+ { "unlimited", -1, -1, true },
+ { nullptr }
+ };
+
+/* Same as above but using `zuinteger_unlimited_literals', with a pointer
+ to a global storage buffer. */
+
set_show_commands
add_setshow_zuinteger_unlimited_cmd (const char *name,
enum command_class theclass,
@@ -1147,13 +1341,11 @@ add_setshow_zuinteger_unlimited_cmd (con
struct cmd_list_element **show_list)
{
set_show_commands commands
- = add_setshow_cmd_full<int> (name, theclass, var_zuinteger_unlimited, var,
+ = add_setshow_cmd_full<int> (name, theclass, var_pinteger, var,
+ zuinteger_unlimited_literals,
set_doc, show_doc, help_doc, nullptr,
nullptr, set_func, show_func, set_list,
show_list);
-
- set_cmd_completer (commands.set, integer_unlimited_completer);
-
return commands;
}
@@ -1171,13 +1363,11 @@ add_setshow_zuinteger_unlimited_cmd (con
cmd_list_element **show_list)
{
auto cmds
- = add_setshow_cmd_full<int> (name, theclass, var_zuinteger_unlimited,
- nullptr, set_doc, show_doc, help_doc, set_func,
+ = add_setshow_cmd_full<int> (name, theclass, var_pinteger, nullptr,
+ zuinteger_unlimited_literals,
+ set_doc, show_doc, help_doc, set_func,
get_func, nullptr, show_func, set_list,
show_list);
-
- set_cmd_completer (cmds.set, integer_unlimited_completer);
-
return cmds;
}
@@ -1196,7 +1386,7 @@ add_setshow_zuinteger_cmd (const char *n
struct cmd_list_element **set_list,
struct cmd_list_element **show_list)
{
- return add_setshow_cmd_full<unsigned int> (name, theclass, var_zuinteger,
+ return add_setshow_cmd_full<unsigned int> (name, theclass, var_uinteger,
var, set_doc, show_doc, help_doc,
nullptr, nullptr, set_func,
show_func, set_list, show_list);
@@ -1215,7 +1405,7 @@ add_setshow_zuinteger_cmd (const char *n
cmd_list_element **set_list,
cmd_list_element **show_list)
{
- return add_setshow_cmd_full<unsigned int> (name, theclass, var_zuinteger,
+ return add_setshow_cmd_full<unsigned int> (name, theclass, var_uinteger,
nullptr, set_doc, show_doc,
help_doc, set_func, get_func,
nullptr, show_func, set_list,
Index: src/gdb/cli/cli-option.c
===================================================================
--- src.orig/gdb/cli/cli-option.c
+++ src/gdb/cli/cli-option.c
@@ -38,7 +38,7 @@ union option_value
/* For var_uinteger options. */
unsigned int uinteger;
- /* For var_zuinteger_unlimited options. */
+ /* For var_integer and var_pinteger options. */
int integer;
/* For var_enum options. */
@@ -356,42 +356,55 @@ parse_option (gdb::array_view<const opti
return option_def_and_value {*match, match_ctx, val};
}
case var_uinteger:
- case var_zuinteger_unlimited:
+ case var_integer:
+ case var_pinteger:
{
if (completion != nullptr)
{
- if (**args == '\0')
- {
- /* Convenience to let the user know what the option
- can accept. Note there's no common prefix between
- the strings on purpose, so that readline doesn't do
- a partial match. */
- completion->tracker.add_completion
- (make_unique_xstrdup ("NUMBER"));
- completion->tracker.add_completion
- (make_unique_xstrdup ("unlimited"));
- return {};
- }
- else if (startswith ("unlimited", *args))
+ if (match->extra_literals != nullptr)
{
- completion->tracker.add_completion
- (make_unique_xstrdup ("unlimited"));
- return {};
+ /* Convenience to let the user know what the option can
+ accept. Make sure there's no common prefix between
+ "NUMBER" and all the strings when adding new ones,
+ so that readline doesn't do a partial match. */
+ if (**args == '\0')
+ {
+ completion->tracker.add_completion
+ (make_unique_xstrdup ("NUMBER"));
+ for (const literal_def *l = match->extra_literals;
+ l->literal != nullptr;
+ l++)
+ completion->tracker.add_completion
+ (make_unique_xstrdup (l->literal));
+ return {};
+ }
+ else
+ {
+ bool completions = false;
+ for (const literal_def *l = match->extra_literals;
+ l->literal != nullptr;
+ l++)
+ if (startswith (l->literal, *args))
+ {
+ completion->tracker.add_completion
+ (make_unique_xstrdup (l->literal));
+ completions = true;
+ }
+ if (completions)
+ return {};
+ }
}
}
- if (match->type == var_zuinteger_unlimited)
- {
- option_value val;
- val.integer = parse_cli_var_zuinteger_unlimited (args, false);
- return option_def_and_value {*match, match_ctx, val};
- }
+ LONGEST v = parse_cli_var_integer (match->type,
+ match->extra_literals,
+ args, false);
+ option_value val;
+ if (match->type == var_uinteger)
+ val.uinteger = v;
else
- {
- option_value val;
- val.uinteger = parse_cli_var_uinteger (match->type, args, false);
- return option_def_and_value {*match, match_ctx, val};
- }
+ val.integer = v;
+ return option_def_and_value {*match, match_ctx, val};
}
case var_enum:
{
@@ -593,7 +606,7 @@ save_option_value_in_ctx (gdb::optional<
*ov->option.var_address.uinteger (ov->option, ov->ctx)
= ov->value->uinteger;
break;
- case var_zuinteger_unlimited:
+ case var_integer:
*ov->option.var_address.integer (ov->option, ov->ctx)
= ov->value->integer;
break;
@@ -664,8 +677,20 @@ get_val_type_str (const option_def &opt,
case var_boolean:
return "[on|off]";
case var_uinteger:
- case var_zuinteger_unlimited:
- return "NUMBER|unlimited";
+ case var_integer:
+ case var_pinteger:
+ {
+ buffer = "NUMBER";
+ if (opt.extra_literals != nullptr)
+ for (const literal_def *l = opt.extra_literals;
+ l->literal != nullptr;
+ l++)
+ {
+ buffer += '|';
+ buffer += l->literal;
+ }
+ return buffer.c_str ();
+ }
case var_enum:
{
buffer = "";
@@ -789,20 +814,31 @@ add_setshow_cmds_for_options (command_cl
{
add_setshow_uinteger_cmd (option.name, cmd_class,
option.var_address.uinteger (option, data),
+ option.extra_literals,
option.set_doc, option.show_doc,
option.help_doc,
nullptr, option.show_cmd_cb,
set_list, show_list);
}
- else if (option.type == var_zuinteger_unlimited)
+ else if (option.type == var_integer)
{
- add_setshow_zuinteger_unlimited_cmd
- (option.name, cmd_class,
- option.var_address.integer (option, data),
- option.set_doc, option.show_doc,
- option.help_doc,
- nullptr, option.show_cmd_cb,
- set_list, show_list);
+ add_setshow_integer_cmd (option.name, cmd_class,
+ option.var_address.integer (option, data),
+ option.extra_literals,
+ option.set_doc, option.show_doc,
+ option.help_doc,
+ nullptr, option.show_cmd_cb,
+ set_list, show_list);
+ }
+ else if (option.type == var_pinteger)
+ {
+ add_setshow_pinteger_cmd (option.name, cmd_class,
+ option.var_address.integer (option, data),
+ option.extra_literals,
+ option.set_doc, option.show_doc,
+ option.help_doc,
+ nullptr, option.show_cmd_cb,
+ set_list, show_list);
}
else if (option.type == var_enum)
{
Index: src/gdb/cli/cli-option.h
===================================================================
--- src.orig/gdb/cli/cli-option.h
+++ src/gdb/cli/cli-option.h
@@ -49,12 +49,13 @@ struct option_def
used to create the option's "set/show" commands. */
constexpr option_def (const char *name_,
var_types var_type_,
+ const literal_def *extra_literals_,
erased_get_var_address_ftype *erased_get_var_address_,
show_value_ftype *show_cmd_cb_,
const char *set_doc_,
const char *show_doc_,
const char *help_doc_)
- : name (name_), type (var_type_),
+ : name (name_), type (var_type_), extra_literals (extra_literals_),
erased_get_var_address (erased_get_var_address_),
var_address {},
show_cmd_cb (show_cmd_cb_),
@@ -68,6 +69,9 @@ struct option_def
/* The option's type. */
var_types type;
+ /* Extra literals, such as `unlimited', accepted in lieu of a number. */
+ const literal_def *extra_literals;
+
/* A function that gets the controlling variable's address, type
erased. */
erased_get_var_address_ftype *erased_get_var_address;
@@ -160,7 +164,7 @@ struct boolean_option_def : option_def
const char *set_doc_,
const char *show_doc_ = nullptr,
const char *help_doc_ = nullptr)
- : option_def (long_option_, var_boolean,
+ : option_def (long_option_, var_boolean, nullptr,
(erased_get_var_address_ftype *) get_var_address_cb_,
show_cmd_cb_,
set_doc_, show_doc_, help_doc_)
@@ -207,11 +211,12 @@ struct uinteger_option_def : option_def
{
uinteger_option_def (const char *long_option_,
unsigned int *(*get_var_address_cb_) (Context *),
+ const literal_def *extra_literals_,
show_value_ftype *show_cmd_cb_,
const char *set_doc_,
const char *show_doc_ = nullptr,
const char *help_doc_ = nullptr)
- : option_def (long_option_, var_uinteger,
+ : option_def (long_option_, var_uinteger, extra_literals_,
(erased_get_var_address_ftype *) get_var_address_cb_,
show_cmd_cb_,
set_doc_, show_doc_, help_doc_)
@@ -220,18 +225,19 @@ struct uinteger_option_def : option_def
}
};
-/* A var_zuinteger_unlimited command line option. */
+/* A var_integer command line option. */
template<typename Context>
-struct zuinteger_unlimited_option_def : option_def
+struct integer_option_def : option_def
{
- zuinteger_unlimited_option_def (const char *long_option_,
- int *(*get_var_address_cb_) (Context *),
- show_value_ftype *show_cmd_cb_,
- const char *set_doc_,
- const char *show_doc_ = nullptr,
- const char *help_doc_ = nullptr)
- : option_def (long_option_, var_zuinteger_unlimited,
+ integer_option_def (const char *long_option_,
+ int *(*get_var_address_cb_) (Context *),
+ const literal_def *extra_literals_,
+ show_value_ftype *show_cmd_cb_,
+ const char *set_doc_,
+ const char *show_doc_ = nullptr,
+ const char *help_doc_ = nullptr)
+ : option_def (long_option_, var_integer, extra_literals_,
(erased_get_var_address_ftype *) get_var_address_cb_,
show_cmd_cb_,
set_doc_, show_doc_, help_doc_)
@@ -252,7 +258,7 @@ struct enum_option_def : option_def
const char *set_doc_,
const char *show_doc_ = nullptr,
const char *help_doc_ = nullptr)
- : option_def (long_option_, var_enum,
+ : option_def (long_option_, var_enum, nullptr,
(erased_get_var_address_ftype *) get_var_address_cb_,
show_cmd_cb_,
set_doc_, show_doc_, help_doc_)
@@ -273,7 +279,7 @@ struct string_option_def : option_def
const char *set_doc_,
const char *show_doc_ = nullptr,
const char *help_doc_ = nullptr)
- : option_def (long_option_, var_string,
+ : option_def (long_option_, var_string, nullptr,
(erased_get_var_address_ftype *) get_var_address_cb_,
show_cmd_cb_,
set_doc_, show_doc_, help_doc_)
Index: src/gdb/cli/cli-setshow.c
===================================================================
--- src.orig/gdb/cli/cli-setshow.c
+++ src/gdb/cli/cli-setshow.c
@@ -149,10 +149,11 @@ deprecated_show_value_hack (struct ui_fi
}
}
-/* Returns true if ARG is "unlimited". */
+/* Returns true and the value in VAL if ARG is an accepted literal. */
static bool
-is_unlimited_literal (const char **arg, bool expression)
+get_literal_val (LONGEST &val, const literal_def *extra_literals,
+ const char **arg, bool expression)
{
*arg = skip_spaces (*arg);
@@ -162,85 +163,108 @@ is_unlimited_literal (const char **arg,
size_t len = p - *arg;
- if (len > 0 && strncmp ("unlimited", *arg, len) == 0)
- {
- *arg += len;
-
- /* If parsing an expression (i.e., parsing for a "set" command),
- anything after "unlimited" is junk. For options, anything
- after "unlimited" might be a command argument or another
- option. */
- if (expression)
+ if (len > 0 && extra_literals != nullptr)
+ for (const literal_def *l = extra_literals;
+ l->literal != nullptr;
+ l++)
+ if (strncmp (l->literal, *arg, len) == 0)
{
- const char *after = skip_spaces (*arg);
- if (*after != '\0')
- error (_("Junk after \"%.*s\": %s"),
- (int) len, unl_start, after);
- }
+ *arg += len;
- return true;
- }
+ /* If parsing an expression (i.e., parsing for a "set" command),
+ anything after the literal is junk. For options, anything
+ after the literal might be a command argument or another
+ option. */
+ if (expression)
+ {
+ const char *after = skip_spaces (*arg);
+ if (*after != '\0')
+ error (_("Junk after \"%.*s\": %s"),
+ (int) len, unl_start, after);
+ }
+
+ val = l->use;
+ return true;
+ }
return false;
}
/* See cli-setshow.h. */
-unsigned int
-parse_cli_var_uinteger (var_types var_type, const char **arg,
- bool expression)
+LONGEST
+parse_cli_var_integer (var_types var_type, const literal_def *extra_literals,
+ const char **arg, bool expression)
{
LONGEST val;
if (*arg == nullptr || **arg == '\0')
{
- if (var_type == var_uinteger)
- error_no_arg (_("integer to set it to, or \"unlimited\""));
- else
+ if (extra_literals == nullptr)
error_no_arg (_("integer to set it to"));
- }
-
- if (var_type == var_uinteger && is_unlimited_literal (arg, expression))
- val = 0;
- else if (expression)
- val = parse_and_eval_long (*arg);
- else
- val = get_ulongest (arg);
-
- if (var_type == var_uinteger && val == 0)
- val = UINT_MAX;
- else if (val < 0
- /* For var_uinteger, don't let the user set the value
- to UINT_MAX directly, as that exposes an
- implementation detail to the user interface. */
- || (var_type == var_uinteger && val >= UINT_MAX)
- || (var_type == var_zuinteger && val > UINT_MAX))
- error (_("integer %s out of range"), plongest (val));
-
- return val;
-}
-
-/* See cli-setshow.h. */
+ else
+ {
+ std::string buffer = "";
+ size_t count = 0;
-int
-parse_cli_var_zuinteger_unlimited (const char **arg, bool expression)
-{
- LONGEST val;
+ for (const literal_def *l = extra_literals;
+ l->literal != nullptr;
+ l++, count++)
+ {
+ if (count != 0)
+ buffer += ", ";
+ buffer = buffer + '"' + l->literal + '"';
+ }
+ if (count > 1)
+ error_no_arg
+ (string_printf (_("integer to set it to, or one of: %s"),
+ buffer.c_str ()).c_str ());
+ else
+ error_no_arg
+ (string_printf (_("integer to set it to, or %s"),
+ buffer.c_str ()).c_str ());
+ }
+ }
- if (*arg == nullptr || **arg == '\0')
- error_no_arg (_("integer to set it to, or \"unlimited\""));
+ if (!get_literal_val (val, extra_literals, arg, expression))
+ {
+ if (expression)
+ val = parse_and_eval_long (*arg);
+ else
+ val = get_ulongest (arg);
- if (is_unlimited_literal (arg, expression))
- val = -1;
- else if (expression)
- val = parse_and_eval_long (*arg);
- else
- val = get_ulongest (arg);
+ enum tribool allowed = TRIBOOL_UNKNOWN;
+ if (extra_literals != nullptr)
+ {
+ for (const literal_def *l = extra_literals;
+ l->literal != nullptr;
+ l++)
+ if (val == l->val)
+ {
+ allowed = l->allow ? TRIBOOL_TRUE : TRIBOOL_FALSE;
+ val = l->use;
+ break;
+ }
+ else if (val == l->use)
+ {
+ allowed = TRIBOOL_FALSE;
+ break;
+ }
+ }
- if (val > INT_MAX)
- error (_("integer %s out of range"), plongest (val));
- else if (val < -1)
- error (_("only -1 is allowed to set as unlimited"));
+ if (allowed == TRIBOOL_UNKNOWN)
+ {
+ if (var_type == var_pinteger && val < 0)
+ error (_("only -1 is allowed to set as unlimited"));
+ else if (val > UINT_MAX || val < INT_MIN
+ || (var_type == var_uinteger && val < 0)
+ || (var_type == var_integer && val > INT_MAX)
+ || (var_type == var_pinteger && val > INT_MAX))
+ allowed = TRIBOOL_FALSE;
+ }
+ if (allowed == TRIBOOL_FALSE)
+ error (_("integer %s out of range"), plongest (val));
+ }
return val;
}
@@ -405,41 +429,18 @@ do_set_command (const char *arg, int fro
option_changed = c->var->set<enum auto_boolean> (parse_auto_binary_operation (arg));
break;
case var_uinteger:
- case var_zuinteger:
option_changed
- = c->var->set<unsigned int> (parse_cli_var_uinteger (c->var->type (),
- &arg, true));
+ = c->var->set<unsigned int> (parse_cli_var_integer (c->var->type (),
+ c->var->
+ extra_literals (),
+ &arg, true));
break;
case var_integer:
- case var_zinteger:
- {
- LONGEST val;
-
- if (*arg == '\0')
- {
- if (c->var->type () == var_integer)
- error_no_arg (_("integer to set it to, or \"unlimited\""));
- else
- error_no_arg (_("integer to set it to"));
- }
-
- if (c->var->type () == var_integer && is_unlimited_literal (&arg, true))
- val = 0;
- else
- val = parse_and_eval_long (arg);
-
- if (val == 0 && c->var->type () == var_integer)
- val = INT_MAX;
- else if (val < INT_MIN
- /* For var_integer, don't let the user set the value
- to INT_MAX directly, as that exposes an
- implementation detail to the user interface. */
- || (c->var->type () == var_integer && val >= INT_MAX)
- || (c->var->type () == var_zinteger && val > INT_MAX))
- error (_("integer %s out of range"), plongest (val));
-
- option_changed = c->var->set<int> (val);
- }
+ case var_pinteger:
+ option_changed
+ = c->var->set<int> (parse_cli_var_integer (c->var->type (),
+ c->var->extra_literals (),
+ &arg, true));
break;
case var_enum:
{
@@ -454,10 +455,6 @@ do_set_command (const char *arg, int fro
option_changed = c->var->set<const char *> (match);
}
break;
- case var_zuinteger_unlimited:
- option_changed = c->var->set<int>
- (parse_cli_var_zuinteger_unlimited (&arg, true));
- break;
default:
error (_("gdb internal error: bad var_type in do_setshow_command"));
}
@@ -551,7 +548,6 @@ do_set_command (const char *arg, int fro
}
break;
case var_uinteger:
- case var_zuinteger:
{
char s[64];
@@ -560,8 +556,7 @@ do_set_command (const char *arg, int fro
}
break;
case var_integer:
- case var_zinteger:
- case var_zuinteger_unlimited:
+ case var_pinteger:
{
char s[64];
@@ -623,36 +618,32 @@ get_setshow_command_value_string (const
}
break;
case var_uinteger:
- case var_zuinteger:
- {
- const unsigned int value = var.get<unsigned int> ();
-
- if (var.type () == var_uinteger
- && value == UINT_MAX)
- stb.puts ("unlimited");
- else
- stb.printf ("%u", value);
- }
- break;
case var_integer:
- case var_zinteger:
+ case var_pinteger:
{
- const int value = var.get<int> ();
+ bool printed = false;
+ const LONGEST value
+ = (var.type () == var_uinteger
+ ? static_cast<LONGEST> (var.get<unsigned int> ())
+ : static_cast<LONGEST> (var.get<int> ()));
- if (var.type () == var_integer
- && value == INT_MAX)
- stb.puts ("unlimited");
- else
- stb.printf ("%d", value);
- }
- break;
- case var_zuinteger_unlimited:
- {
- const int value = var.get<int> ();
- if (value == -1)
- stb.puts ("unlimited");
- else
- stb.printf ("%d", value);
+ if (var.extra_literals () != nullptr)
+ for (const literal_def *l = var.extra_literals ();
+ l->literal != nullptr;
+ l++)
+ if (value == l->use)
+ {
+ stb.puts (l->literal);
+ printed = true;
+ break;
+ }
+ if (!printed)
+ {
+ if (var.type () == var_uinteger)
+ stb.printf ("%u", static_cast<const unsigned int> (value));
+ else
+ stb.printf ("%d", static_cast<const int> (value));
+ }
}
break;
default:
Index: src/gdb/cli/cli-setshow.h
===================================================================
--- src.orig/gdb/cli/cli-setshow.h
+++ src/gdb/cli/cli-setshow.h
@@ -29,21 +29,19 @@ extern int parse_cli_boolean_value (cons
past a successfully parsed value. */
extern int parse_cli_boolean_value (const char **arg);
-/* Parse ARG, an option to a var_uinteger or var_zuinteger variable.
- Either returns the parsed value on success or throws an error. If
- EXPRESSION is true, *ARG is parsed as an expression; otherwise, it
- is parsed with get_ulongest. It's not possible to parse the
+/* Parse ARG, an option to a var_uinteger, var_integer or var_pinteger
+ variable. Return the parsed value on success or throw an error. If
+ EXTRA_LITERALS is non-null, then interpret those literals accordingly.
+ If EXPRESSION is true, *ARG is parsed as an expression; otherwise,
+ it is parsed with get_ulongest. It's not possible to parse the
integer as an expression when there may be valid input after the
integer, such as when parsing command options. E.g., "print
-elements NUMBER -obj --". In such case, parsing as an expression
would parse "-obj --" as part of the expression as well. */
-extern unsigned int parse_cli_var_uinteger (var_types var_type,
- const char **arg,
- bool expression);
-
-/* Like parse_cli_var_uinteger, for var_zuinteger_unlimited. */
-extern int parse_cli_var_zuinteger_unlimited (const char **arg,
- bool expression);
+extern LONGEST parse_cli_var_integer (var_types var_type,
+ const literal_def *extra_literals,
+ const char **arg,
+ bool expression);
/* Parse ARG, an option to a var_enum variable. ENUM is a
null-terminated array of possible values. Either returns the parsed
Index: src/gdb/command.h
===================================================================
--- src.orig/gdb/command.h
+++ src/gdb/command.h
@@ -84,16 +84,18 @@ typedef enum var_types
value. */
var_auto_boolean,
- /* Unsigned Integer. *VAR is an unsigned int. The user can type
- 0 to mean "unlimited", which is stored in *VAR as UINT_MAX. */
+ /* Unsigned Integer. *VAR is an unsigned int. In the Guile and Python
+ APIs 0 means unlimited, which is stored in *VAR as UINT_MAX. */
var_uinteger,
- /* Like var_uinteger but signed. *VAR is an int. The user can
- type 0 to mean "unlimited", which is stored in *VAR as
- INT_MAX. The only remaining use of it is the Python API.
- Don't use it elsewhere. */
+ /* Like var_uinteger but signed. *VAR is an int. In the Guile and
+ Python APIs 0 means unlimited, which is stored in *VAR as INT_MAX. */
var_integer,
+ /* Like var_integer but negative numbers are not allowed,
+ except for special values. *VAR is an int. */
+ var_pinteger,
+
/* String which the user enters with escapes (e.g. the user types
\n and it is a real newline in the stored string).
*VAR is a std::string, "" if the string is empty. */
@@ -106,15 +108,16 @@ typedef enum var_types
var_optional_filename,
/* String which stores a filename. (*VAR) is a std::string. */
var_filename,
- /* ZeroableInteger. *VAR is an int. Like var_integer except
- that zero really means zero. */
+ /* Like var_integer. *VAR is an int. The only remaining uses
+ of it are the Guile and Python APIs. Don't use it elsewhere. */
var_zinteger,
- /* ZeroableUnsignedInteger. *VAR is an unsigned int. Zero really
- means zero. */
+ /* Like var_uinteger. *VAR is an unsigned int. The only remaining
+ uses of it are the Guile and Python APIs. Don't use it elsewhere. */
var_zuinteger,
- /* ZeroableUnsignedInteger with unlimited value. *VAR is an int,
- but its range is [0, INT_MAX]. -1 stands for unlimited and
- other negative numbers are not allowed. */
+ /* Like var_pinteger. *VAR is an int, but its range is [-1, INT_MAX].
+ -1 stands for unlimited and other negative numbers are not allowed.
+ The only remaining uses of it are the Guile and Python APIs. Don't
+ use it elsewhere. */
var_zuinteger_unlimited,
/* Enumerated type. Can only have one of the specified values.
*VAR is a char pointer to the name of the element that we
@@ -123,6 +126,22 @@ typedef enum var_types
}
var_types;
+/* A structure describing an extra literal accepted and shown in place
+ of a number. */
+typedef struct literal_defs
+ {
+ /* The literal to define, e.g. "unlimited". */
+ const char *literal;
+ /* A number optionally accepted that stands for the literal. */
+ LONGEST val;
+ /* The number to substitute internally for LITERAL or VAL;
+ the use of this number is not allowed (unless the same as VAL). */
+ LONGEST use;
+ /* True if the use of VAL is allowed; otherwise only the literal is. */
+ bool allow;
+ }
+literal_def;
+
/* Return true if a setting of type VAR_TYPE is backed with type T.
This function is left without definition intentionally. This template is
@@ -153,15 +172,14 @@ inline bool var_type_uses<enum auto_bool
template<>
inline bool var_type_uses<unsigned int> (var_types t)
{
- return (t == var_uinteger || t == var_zinteger || t == var_zuinteger);
+ return t == var_uinteger;
}
/* Return true if a setting of type T is backed by an int variable. */
template<>
inline bool var_type_uses<int> (var_types t)
{
- return (t == var_integer || t == var_zinteger
- || t == var_zuinteger_unlimited);
+ return t == var_integer || t == var_pinteger;
}
/* Return true if a setting of type T is backed by a std::string variable. */
@@ -233,6 +251,7 @@ struct setting
struct erased_args
{
void *var;
+ const void *extra_literals;
erased_func setter;
erased_func getter;
};
@@ -240,6 +259,7 @@ struct setting
template<typename T>
static erased_args erase_args (var_types var_type,
T *var,
+ const literal_def *extra_literals,
typename setting_func_types<T>::set set_setting_func,
typename setting_func_types<T>::get get_setting_func)
{
@@ -254,6 +274,7 @@ struct setting
return {
var,
+ extra_literals,
reinterpret_cast<erased_func> (set_setting_func),
reinterpret_cast<erased_func> (get_setting_func)
};
@@ -265,6 +286,7 @@ struct setting
setting (var_types var_type, const erased_args &args)
: m_var_type (var_type),
m_var (args.var),
+ m_extra_literals (args.extra_literals),
m_getter (args.getter),
m_setter (args.setter)
{
@@ -295,6 +317,10 @@ struct setting
var_types type () const
{ return m_var_type; }
+ /* Access any extra literals accepted. */
+ const literal_def *extra_literals () const
+ { return static_cast<const literal_def *> (m_extra_literals); }
+
/* Return the current value.
The template parameter T is the type of the variable used to store the
@@ -357,6 +383,9 @@ struct setting
non-nullptr. */
void *m_var = nullptr;
+ /* Any extra literals accepted. */
+ const void *m_extra_literals = nullptr;
+
/* Pointer to a user provided getter. */
erased_func m_getter = nullptr;
@@ -652,6 +681,11 @@ typedef void (show_value_ftype) (struct
instead print the value out directly. */
extern show_value_ftype deprecated_show_value_hack;
+/* Various sets of extra literals accepted. */
+extern const literal_def integer_unlimited_literals[];
+extern const literal_def uinteger_unlimited_literals[];
+extern const literal_def zuinteger_unlimited_literals[];
+
extern set_show_commands add_setshow_enum_cmd
(const char *name, command_class theclass, const char *const *enumlist,
const char **var, const char *set_doc, const char *show_doc,
@@ -748,6 +782,20 @@ extern set_show_commands add_setshow_opt
cmd_list_element **show_list);
extern set_show_commands add_setshow_integer_cmd
+ (const char *name, command_class theclass, int *var,
+ const literal_def *extra_literals, const char *set_doc,
+ const char *show_doc, const char *help_doc, cmd_func_ftype *set_func,
+ show_value_ftype *show_func, cmd_list_element **set_list,
+ cmd_list_element **show_list);
+
+extern set_show_commands add_setshow_integer_cmd
+ (const char *name, command_class theclass, const literal_def *extra_literals,
+ const char *set_doc, const char *show_doc, const char *help_doc,
+ setting_func_types<int>::set set_func,
+ setting_func_types<int>::get get_func, show_value_ftype *show_func,
+ cmd_list_element **set_list, cmd_list_element **show_list);
+
+extern set_show_commands add_setshow_integer_cmd
(const char *name, command_class theclass, int *var, const char *set_doc,
const char *show_doc, const char *help_doc, cmd_func_ftype *set_func,
show_value_ftype *show_func, cmd_list_element **set_list,
@@ -760,6 +808,34 @@ extern set_show_commands add_setshow_int
setting_func_types<int>::get get_func, show_value_ftype *show_func,
cmd_list_element **set_list, cmd_list_element **show_list);
+extern set_show_commands add_setshow_pinteger_cmd
+ (const char *name, command_class theclass, int *var,
+ const literal_def *extra_literals, const char *set_doc,
+ const char *show_doc, const char *help_doc, cmd_func_ftype *set_func,
+ show_value_ftype *show_func, cmd_list_element **set_list,
+ cmd_list_element **show_list);
+
+extern set_show_commands add_setshow_pinteger_cmd
+ (const char *name, command_class theclass, const literal_def *extra_literals,
+ const char *set_doc, const char *show_doc, const char *help_doc,
+ setting_func_types<int>::set set_func,
+ setting_func_types<int>::get get_func, show_value_ftype *show_func,
+ cmd_list_element **set_list, cmd_list_element **show_list);
+
+extern set_show_commands add_setshow_uinteger_cmd
+ (const char *name, command_class theclass, unsigned int *var,
+ const literal_def *extra_literals,
+ const char *set_doc, const char *show_doc, const char *help_doc,
+ cmd_func_ftype *set_func, show_value_ftype *show_func,
+ cmd_list_element **set_list, cmd_list_element **show_list);
+
+extern set_show_commands add_setshow_uinteger_cmd
+ (const char *name, command_class theclass, const literal_def *extra_literals,
+ const char *set_doc, const char *show_doc, const char *help_doc,
+ setting_func_types<unsigned int>::set set_func,
+ setting_func_types<unsigned int>::get get_func, show_value_ftype *show_func,
+ cmd_list_element **set_list, cmd_list_element **show_list);
+
extern set_show_commands add_setshow_uinteger_cmd
(const char *name, command_class theclass, unsigned int *var,
const char *set_doc, const char *show_doc, const char *help_doc,
Index: src/gdb/guile/scm-param.c
===================================================================
--- src.orig/gdb/guile/scm-param.c
+++ src/gdb/guile/scm-param.c
@@ -117,18 +117,33 @@ struct param_smob
static setting
make_setting (param_smob *s)
{
- if (var_type_uses<bool> (s->type))
- return setting (s->type, &s->value.boolval);
- else if (var_type_uses<int> (s->type))
- return setting (s->type, &s->value.intval);
- else if (var_type_uses<auto_boolean> (s->type))
- return setting (s->type, &s->value.autoboolval);
- else if (var_type_uses<unsigned int> (s->type))
- return setting (s->type, &s->value.uintval);
- else if (var_type_uses<std::string> (s->type))
- return setting (s->type, s->value.stringval);
- else if (var_type_uses<const char *> (s->type))
- return setting (s->type, &s->value.cstringval);
+ enum var_types type = s->type;
+
+ switch (type)
+ {
+ case var_zinteger:
+ type = var_integer;
+ break;
+ case var_zuinteger:
+ type = var_uinteger;
+ break;
+ case var_zuinteger_unlimited:
+ type = var_pinteger;
+ break;
+ }
+
+ if (var_type_uses<bool> (type))
+ return setting (type, &s->value.boolval);
+ else if (var_type_uses<int> (type))
+ return setting (type, &s->value.intval);
+ else if (var_type_uses<auto_boolean> (type))
+ return setting (type, &s->value.autoboolval);
+ else if (var_type_uses<unsigned int> (type))
+ return setting (type, &s->value.uintval);
+ else if (var_type_uses<std::string> (type))
+ return setting (type, s->value.stringval);
+ else if (var_type_uses<const char *> (type))
+ return setting (type, &s->value.cstringval);
else
gdb_assert_not_reached ("unhandled var type");
}
@@ -588,10 +603,6 @@ pascm_param_type_name (enum var_types pa
static SCM
pascm_param_value (const setting &var, int arg_pos, const char *func_name)
{
- /* Note: We *could* support var_integer here in case someone is trying to get
- the value of a Python-created parameter (which is the only place that
- still supports var_integer). To further discourage its use we do not. */
-
switch (var.type ())
{
case var_string:
@@ -631,20 +642,35 @@ pascm_param_value (const setting &var, i
return auto_keyword;
}
- case var_zuinteger_unlimited:
- if (var.get<int> () == -1)
- return unlimited_keyword;
- gdb_assert (var.get<int> () >= 0);
- /* Fall through. */
- case var_zinteger:
- return scm_from_int (var.get<int> ());
-
case var_uinteger:
- if (var.get<unsigned int> ()== UINT_MAX)
- return unlimited_keyword;
- /* Fall through. */
- case var_zuinteger:
- return scm_from_uint (var.get<unsigned int> ());
+ case var_integer:
+ case var_pinteger:
+ {
+ LONGEST value
+ = (var.type () == var_uinteger
+ ? static_cast<LONGEST> (var.get<unsigned int> ())
+ : static_cast<LONGEST> (var.get<int> ()));
+
+ bool literal = false;
+ if (var.extra_literals () != nullptr)
+ for (const literal_def *l = var.extra_literals ();
+ l->literal != nullptr;
+ l++)
+ if (value == l->use)
+ {
+ if (strcmp (l->literal, "unlimited") == 0)
+ return unlimited_keyword;
+ value = l->val;
+ literal = true;
+ }
+ if (var.type () == var_pinteger && !literal)
+ gdb_assert (value >= 0);
+
+ if (var.type () == var_uinteger)
+ return scm_from_uint (static_cast<const unsigned int> (value));
+ else
+ return scm_from_int (static_cast<const int> (value));
+ }
default:
break;
@@ -666,14 +692,14 @@ pascm_set_param_value_x (param_smob *p_s
{
setting var = make_setting (p_smob);
- switch (var.type ())
+ switch (p_smob->type)
{
case var_string:
case var_string_noescape:
case var_optional_filename:
case var_filename:
SCM_ASSERT_TYPE (scm_is_string (value)
- || (var.type () != var_filename
+ || (p_smob->type != var_filename
&& gdbscm_is_false (value)),
value, arg_pos, func_name,
_("string or #f for non-PARAM_FILENAME parameters"));
@@ -739,8 +765,8 @@ pascm_set_param_value_x (param_smob *p_s
case var_uinteger:
case var_zuinteger:
case var_zuinteger_unlimited:
- if (var.type () == var_uinteger
- || var.type () == var_zuinteger_unlimited)
+ if (p_smob->type == var_uinteger
+ || p_smob->type == var_zuinteger_unlimited)
{
SCM_ASSERT_TYPE (gdbscm_is_bool (value)
|| scm_is_eq (value, unlimited_keyword),
@@ -748,7 +774,7 @@ pascm_set_param_value_x (param_smob *p_s
_("integer or #:unlimited"));
if (scm_is_eq (value, unlimited_keyword))
{
- if (var.type () == var_uinteger)
+ if (p_smob->type == var_uinteger)
var.set<unsigned int> (UINT_MAX);
else
var.set<int> (-1);
@@ -761,12 +787,12 @@ pascm_set_param_value_x (param_smob *p_s
_("integer"));
}
- if (var.type () == var_uinteger
- || var.type () == var_zuinteger)
+ if (p_smob->type == var_uinteger
+ || p_smob->type == var_zuinteger)
{
unsigned int u = scm_to_uint (value);
- if (var.type () == var_uinteger && u == 0)
+ if (p_smob->type == var_uinteger && u == 0)
u = UINT_MAX;
var.set<unsigned int> (u);
}
@@ -774,7 +800,7 @@ pascm_set_param_value_x (param_smob *p_s
{
int i = scm_to_int (value);
- if (var.type () == var_zuinteger_unlimited && i < -1)
+ if (p_smob->type == var_zuinteger_unlimited && i < -1)
{
gdbscm_out_of_range_error (func_name, arg_pos, value,
_("must be >= -1"));
Index: src/gdb/maint-test-options.c
===================================================================
--- src.orig/gdb/maint-test-options.c
+++ src/gdb/maint-test-options.c
@@ -207,6 +207,7 @@ static const gdb::option::option_def tes
gdb::option::uinteger_option_def<test_options_opts> {
"uinteger",
[] (test_options_opts *opts) { return &opts->uint_opt; },
+ uinteger_unlimited_literals,
nullptr, /* show_cmd_cb */
N_("A uinteger option."),
nullptr, /* show_doc */
@@ -214,9 +215,10 @@ static const gdb::option::option_def tes
},
/* A zuinteger_unlimited option. */
- gdb::option::zuinteger_unlimited_option_def<test_options_opts> {
+ gdb::option::integer_option_def<test_options_opts> {
"zuinteger-unlimited",
[] (test_options_opts *opts) { return &opts->zuint_unl_opt; },
+ zuinteger_unlimited_literals,
nullptr, /* show_cmd_cb */
N_("A zuinteger-unlimited option."),
nullptr, /* show_doc */
Index: src/gdb/python/py-param.c
===================================================================
--- src.orig/gdb/python/py-param.c
+++ src/gdb/python/py-param.c
@@ -96,18 +96,33 @@ struct parmpy_object
static setting
make_setting (parmpy_object *s)
{
- if (var_type_uses<bool> (s->type))
- return setting (s->type, &s->value.boolval);
- else if (var_type_uses<int> (s->type))
- return setting (s->type, &s->value.intval);
- else if (var_type_uses<auto_boolean> (s->type))
- return setting (s->type, &s->value.autoboolval);
- else if (var_type_uses<unsigned int> (s->type))
- return setting (s->type, &s->value.uintval);
- else if (var_type_uses<std::string> (s->type))
- return setting (s->type, s->value.stringval);
- else if (var_type_uses<const char *> (s->type))
- return setting (s->type, &s->value.cstringval);
+ enum var_types type = s->type;
+
+ switch (type)
+ {
+ case var_zinteger:
+ type = var_integer;
+ break;
+ case var_zuinteger:
+ type = var_uinteger;
+ break;
+ case var_zuinteger_unlimited:
+ type = var_pinteger;
+ break;
+ }
+
+ if (var_type_uses<bool> (type))
+ return setting (type, &s->value.boolval);
+ else if (var_type_uses<int> (type))
+ return setting (type, &s->value.intval);
+ else if (var_type_uses<auto_boolean> (type))
+ return setting (type, &s->value.autoboolval);
+ else if (var_type_uses<unsigned int> (type))
+ return setting (type, &s->value.uintval);
+ else if (var_type_uses<std::string> (type))
+ return setting (type, s->value.stringval);
+ else if (var_type_uses<const char *> (type))
+ return setting (type, &s->value.cstringval);
else
gdb_assert_not_reached ("unhandled var type");
}
Index: src/gdb/python/python.c
===================================================================
--- src.orig/gdb/python/python.c
+++ src/gdb/python/python.c
@@ -507,35 +507,34 @@ gdbpy_parameter_value (const setting &va
Py_RETURN_NONE;
}
- case var_integer:
- if (var.get<int> () == INT_MAX)
- Py_RETURN_NONE;
- /* Fall through. */
- case var_zinteger:
- return gdb_py_object_from_longest (var.get<int> ()).release ();
-
- case var_zuinteger_unlimited:
- {
- int val = var.get<int> ();
-
- if (val == -1)
- Py_RETURN_NONE;
- return gdb_py_object_from_longest (val).release ();
- }
-
case var_uinteger:
+ case var_integer:
+ case var_pinteger:
{
- unsigned int val = var.get<unsigned int> ();
+ LONGEST value
+ = (var.type () == var_uinteger
+ ? static_cast<LONGEST> (var.get<unsigned int> ())
+ : static_cast<LONGEST> (var.get<int> ()));
- if (val == UINT_MAX)
- Py_RETURN_NONE;
- return gdb_py_object_from_ulongest (val).release ();
- }
+ if (var.extra_literals () != nullptr)
+ for (const literal_def *l = var.extra_literals ();
+ l->literal != nullptr;
+ l++)
+ if (value == l->use)
+ {
+ if (strcmp (l->literal, "unlimited") == 0)
+ Py_RETURN_NONE;
+ value = l->val;
+ }
- case var_zuinteger:
- {
- unsigned int val = var.get<unsigned int> ();
- return gdb_py_object_from_ulongest (val).release ();
+ if (var.type () == var_uinteger)
+ return
+ gdb_py_object_from_ulongest
+ (static_cast<const unsigned int> (value)).release ();
+ else
+ return
+ gdb_py_object_from_longest
+ (static_cast<const int> (value)).release ();
}
}
Index: src/gdb/valprint.c
===================================================================
--- src.orig/gdb/valprint.c
+++ src/gdb/valprint.c
@@ -2967,8 +2967,8 @@ using boolean_option_def
= gdb::option::boolean_option_def<value_print_options>;
using uinteger_option_def
= gdb::option::uinteger_option_def<value_print_options>;
-using zuinteger_unlimited_option_def
- = gdb::option::zuinteger_unlimited_option_def<value_print_options>;
+using integer_option_def
+ = gdb::option::integer_option_def<value_print_options>;
/* Definitions of options for the "print" and "compile print"
commands. */
@@ -3004,15 +3004,17 @@ static const gdb::option::option_def val
uinteger_option_def {
"elements",
[] (value_print_options *opt) { return &opt->print_max; },
+ uinteger_unlimited_literals,
show_print_max, /* show_cmd_cb */
N_("Set limit on string chars or array elements to print."),
N_("Show limit on string chars or array elements to print."),
N_("\"unlimited\" causes there to be no limit."),
},
- zuinteger_unlimited_option_def {
+ integer_option_def {
"max-depth",
[] (value_print_options *opt) { return &opt->max_depth; },
+ zuinteger_unlimited_literals,
show_print_max_depth, /* show_cmd_cb */
N_("Set maximum print depth for nested structures, unions and arrays."),
N_("Show maximum print depth for nested structures, unions, and arrays."),
@@ -3072,6 +3074,7 @@ pretty-printers for that value.")
uinteger_option_def {
"repeats",
[] (value_print_options *opt) { return &opt->repeat_count_threshold; },
+ uinteger_unlimited_literals,
show_repeat_count_threshold, /* show_cmd_cb */
N_("Set threshold for repeated print elements."),
N_("Show threshold for repeated print elements."),
^ permalink raw reply [flat|nested] 22+ messages in thread
* [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-09 19:17 [PATCH v4 0/8] gdb: split array and string limiting options Maciej W. Rozycki
` (5 preceding siblings ...)
2022-03-09 19:18 ` [PATCH v4 6/8] GDB: Allow arbitrary keywords in integer set commands Maciej W. Rozycki
@ 2022-03-09 19:18 ` Maciej W. Rozycki
2022-03-09 19:51 ` Eli Zaretskii
2022-03-09 19:18 ` [PATCH v4 8/8] GDB/testsuite: Expand for character string limiting options Maciej W. Rozycki
2022-03-09 19:47 ` [PATCH v4 0/8] gdb: split array and " Simon Sobisch
8 siblings, 1 reply; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-09 19:18 UTC (permalink / raw)
To: gdb-patches; +Cc: Andrew Burgess, Simon Marchi, Tom Tromey, Simon Sobisch
From: Andrew Burgess <andrew.burgess@embecosm.com>
This commit splits the `set/show print elements' option into two. We
retain `set/show print elements' for controlling how many elements of an
array we print, but a new `set/show print characters' setting is added
which is used for controlling how many characters of a string are
printed.
The motivation behind this change is to allow users a finer level of
control over how data is printed, reflecting that, although strings can
be thought of as arrays of characters, users often want to treat these
two things differently.
For compatibility reasons by default the `set/show print characters'
option is set to `elements', which makes the limit for character strings
follow the setting of the `set/show print elements' option, as it used
to. Using `set print characters' with any other value makes the limit
independent from the `set/show print elements' setting, however it can
be restored to the default with the `set print characters elements'
command at any time.
A corresponding `-characters' option for the `print' command is added,
with the same semantics, i.e. one can use `elements' to make a given
`print' invocation follow the limit of elements, be it set with the
`-elements' option also given with the same invocation or taken from the
`set/show print elements' setting, for characters as well regardless of
the current setting of the `set/show print characters' option.
The GDB changes are all pretty straightforward, just changing references
to the old 'print_max' to use a new `get_print_max_chars' helper which
figures out which of the two of `print_max' and `print_max_chars' values
to use.
Likewise, the documentation is just updated to reference the new setting
where appropriate.
To make people's life easier the message shown by `show print elements'
now indicates if the setting also applies to character strings:
(gdb) set print characters elements
(gdb) show print elements
Limit on string chars or array elements to print is 200.
(gdb) set print characters unlimited
(gdb) show print elements
Limit on array elements to print is 200.
(gdb)
which keeps it the same as it used to be, although in other contexts the
setting is described as applying to array elements only, e.g.:
(gdb) help set print elements
Set limit on array elements to print.
"unlimited" causes there to be no limit.
(gdb)
In the testsuite there are two minor updates, one to add `-characters'
to the list of completions now shown for the `print' command, and a bare
minimum pair of checks for the right handling of `set print characters'
and `show print characters', copied from the corresponding checks for
`set print elements' and `show print elements' respectively.
Co-Authored-By: Maciej W. Rozycki <macro@embecosm.com>
---
Changes from v3:
- rewrite in terms of flexible literal keyword handling added with 6/8,
- add a `set print characters elements' setting, which is also the
default, to make the string character limit follow that for array
elements in backwards-compatible manner,
- update documentation accordingly,
- remove testsuite updates delegated to 8/8, guaranteeing that the new
command does not influence any preexisting environment (except for
obvious cases where it is expected, such as command completion or help
messages),
- amend gdb.base/options.exp for the added completion for `print'
command's `-characters' option,
- expand gdb.base/default.exp with minimal test cases for `set/show print
characters'.
Changes from v2:
- place the new and the changed command at the right places each in NEWS.
Changes from v1:
- rename `print_smax' setting throughout to `print_max_chars', and
likewise `show_print_smax' function to `show_print_max_chars',
- document the Python part in the manual,
- update comments for `print_max' and `print_max_chars' in
`value_print_options',
- fix some typos.
---
gdb/NEWS | 11 ++++
gdb/ada-valprint.c | 6 +-
gdb/c-lang.c | 4 -
gdb/c-valprint.c | 5 +-
gdb/doc/gdb.texinfo | 39 ++++++++++++++---
gdb/doc/python.texi | 5 ++
gdb/language.h | 2
gdb/m2-lang.c | 3 -
gdb/m2-valprint.c | 4 +
gdb/p-lang.c | 3 -
gdb/p-valprint.c | 9 ++-
gdb/printcmd.c | 9 ++-
gdb/python/py-value.c | 4 +
gdb/testsuite/gdb.base/default.exp | 7 +++
gdb/testsuite/gdb.base/options.exp | 1
gdb/tracepoint.c | 4 -
gdb/valprint.c | 84 ++++++++++++++++++++++++++++---------
gdb/valprint.h | 20 +++++++-
18 files changed, 171 insertions(+), 49 deletions(-)
gdb-aburgess-print-elements-characters.diff
Index: src/gdb/NEWS
===================================================================
--- src.orig/gdb/NEWS
+++ src/gdb/NEWS
@@ -119,6 +119,17 @@ show ada source-charset
values for this follow the values that can be passed to the GNAT
compiler via the '-gnati' option. The default is ISO-8859-1.
+set print characters LIMIT
+show print characters
+print -characters LIMIT
+ This new setting is like 'set print elements', but controls how many
+ characters of a string are printed. This functionality used to be
+ covered by 'set print elements', but it can be controlled separately
+ now. LIMIT can be set to a numerical value to request that particular
+ character count, to 'unlimited' to print all characters of a string,
+ or to 'elements', which is also the default, to follow the setting of
+ 'set print elements' as it used to be.
+
* Changed commands
maint packet
Index: src/gdb/ada-valprint.c
===================================================================
--- src.orig/gdb/ada-valprint.c
+++ src/gdb/ada-valprint.c
@@ -470,7 +470,8 @@ printstr (struct ui_file *stream, struct
return;
}
- for (i = 0; i < length && things_printed < options->print_max; i += 1)
+ unsigned int print_max_chars = get_print_max_chars (options);
+ for (i = 0; i < length && things_printed < print_max_chars; i += 1)
{
/* Position of the character we are examining
to see whether it is repeated. */
@@ -706,12 +707,13 @@ ada_val_print_string (struct type *type,
elements up to it. */
if (options->stop_print_at_null)
{
+ unsigned int print_max_chars = get_print_max_chars (options);
int temp_len;
/* Look for a NULL char. */
for (temp_len = 0;
(temp_len < len
- && temp_len < options->print_max
+ && temp_len < print_max_chars
&& char_at (valaddr + offset_aligned,
temp_len, eltlen, byte_order) != 0);
temp_len += 1);
Index: src/gdb/c-lang.c
===================================================================
--- src.orig/gdb/c-lang.c
+++ src/gdb/c-lang.c
@@ -185,8 +185,8 @@ language_defn::printchar (int c, struct
/* Print the character string STRING, printing at most LENGTH
characters. LENGTH is -1 if the string is nul terminated. Each
character is WIDTH bytes long. Printing stops early if the number
- hits print_max; repeat counts are printed as appropriate. Print
- ellipses at the end if we had to stop before printing LENGTH
+ hits print_max_chars; repeat counts are printed as appropriate.
+ Print ellipses at the end if we had to stop before printing LENGTH
characters, or if FORCE_ELLIPSES. */
void
Index: src/gdb/c-valprint.c
===================================================================
--- src.orig/gdb/c-valprint.c
+++ src/gdb/c-valprint.c
@@ -267,11 +267,12 @@ c_value_print_array (struct value *val,
print elements up to it. */
if (options->stop_print_at_null)
{
+ unsigned int print_max_chars = get_print_max_chars (options);
unsigned int temp_len;
for (temp_len = 0;
(temp_len < len
- && temp_len < options->print_max
+ && temp_len < print_max_chars
&& extract_unsigned_integer (valaddr + temp_len * eltlen,
eltlen, byte_order) != 0);
++temp_len)
@@ -280,7 +281,7 @@ c_value_print_array (struct value *val,
/* Force printstr to print ellipses if
we've printed the maximum characters and
the next character is not \000. */
- if (temp_len == options->print_max && temp_len < len)
+ if (temp_len == print_max_chars && temp_len < len)
{
ULONGEST ival
= extract_unsigned_integer (valaddr + temp_len * eltlen,
Index: src/gdb/doc/gdb.texinfo
===================================================================
--- src.orig/gdb/doc/gdb.texinfo
+++ src/gdb/doc/gdb.texinfo
@@ -10012,10 +10012,17 @@ Related setting: @ref{set print array}.
Set printing of array indexes.
Related setting: @ref{set print array-indexes}.
-@item -elements @var{number-of-elements}|@code{unlimited}
-Set limit on string chars or array elements to print. The value
+@item -characters @var{number-of-characters}|@code{elements}|@code{unlimited}
+Set limit on string characters to print. The value @code{elements}
+causes the limit on array elements to print to be used. The value
@code{unlimited} causes there to be no limit. Related setting:
-@ref{set print elements}.
+@ref{set print characters}.
+
+@item -elements @var{number-of-elements}|@code{unlimited}
+Set limit on array elements and optionally string characters to print.
+See @ref{set print characters} and the @code{-characters} option above
+for when this option applies to strings. The value @code{unlimited}
+causes there to be no limit. Related setting: @ref{set print elements}.
@item -max-depth @var{depth}|@code{unlimited}
Set the threshold after which nested structures are replaced with
@@ -11383,6 +11390,27 @@ Stop printing element indexes when displ
Show whether the index of each element is printed when displaying
arrays.
+@anchor{set print characters}
+@item set print characters @var{number-of-characters}
+@itemx set print characters elements
+@itemx set print characters unlimited
+@cindex number of string characters to print
+@cindex limit on number of printed string characters
+Set a limit on how many characters of a string @value{GDBN} will
+print. If @value{GDBN} is printing a large string, it stops printing
+after it has printed the number of characters set by the @code{set
+print characters} command.
+Setting @var{number-of-characters} to @code{elements} means
+that the limit on the number of characters to print follows
+one for array elements; see @ref{set print elements}.
+Setting @var{number-of-characters} to @code{unlimited} means
+that the number of characters to print is unlimited.
+When @value{GDBN} starts, this limit is set to @code{elements}.
+
+@item show print characters
+Display the number of characters of a large string that @value{GDBN}
+will print.
+
@anchor{set print elements}
@item set print elements @var{number-of-elements}
@itemx set print elements unlimited
@@ -11391,7 +11419,8 @@ arrays.
Set a limit on how many elements of an array @value{GDBN} will print.
If @value{GDBN} is printing a large array, it stops printing after it has
printed the number of elements set by the @code{set print elements} command.
-This limit also applies to the display of strings.
+By default this limit also applies to the display of strings; see
+@ref{set print characters}.
When @value{GDBN} starts, this limit is set to 200.
Setting @var{number-of-elements} to @code{unlimited} or zero means
that the number of elements to print is unlimited.
@@ -14894,7 +14923,7 @@ The optional @var{mods} changes the usua
@code{s} requests that pointers to chars be handled as strings, in
particular collecting the contents of the memory being pointed at, up
to the first zero. The upper bound is by default the value of the
-@code{print elements} variable; if @code{s} is followed by a decimal
+@code{print characters} variable; if @code{s} is followed by a decimal
number, that is the upper bound instead. So for instance
@samp{collect/s25 mystr} collects as many as 25 characters at
@samp{mystr}.
Index: src/gdb/doc/python.texi
===================================================================
--- src.orig/gdb/doc/python.texi
+++ src/gdb/doc/python.texi
@@ -1059,6 +1059,11 @@ the @emph{declared} type should be used.
representation of a C@t{++} object, @code{False} if they shouldn't (see
@code{set print static-members} in @ref{Print Settings}).
+@item max_characters
+Number of string characters to print, @code{0} to follow
+@code{max_elements}, or @code{UINT_MAX} to print an unlimited number
+of characters (see @code{set print characters} in @ref{Print Settings}).
+
@item max_elements
Number of array elements to print, or @code{0} to print an unlimited
number of elements (see @code{set print elements} in @ref{Print
Index: src/gdb/language.h
===================================================================
--- src.orig/gdb/language.h
+++ src/gdb/language.h
@@ -534,7 +534,7 @@ struct language_defn
struct ui_file * stream) const;
/* Print the character string STRING, printing at most LENGTH characters.
- Printing stops early if the number hits print_max; repeat counts
+ Printing stops early if the number hits print_max_chars; repeat counts
are printed as appropriate. Print ellipses at the end if we
had to stop before printing LENGTH characters, or if FORCE_ELLIPSES. */
Index: src/gdb/m2-lang.c
===================================================================
--- src.orig/gdb/m2-lang.c
+++ src/gdb/m2-lang.c
@@ -169,7 +169,8 @@ m2_language::printstr (struct ui_file *s
return;
}
- for (i = 0; i < length && things_printed < options->print_max; ++i)
+ unsigned int print_max_chars = get_print_max_chars (options);
+ for (i = 0; i < length && things_printed < print_max_chars; ++i)
{
/* Position of the character we are examining
to see whether it is repeated. */
Index: src/gdb/m2-valprint.c
===================================================================
--- src.orig/gdb/m2-valprint.c
+++ src/gdb/m2-valprint.c
@@ -327,12 +327,14 @@ m2_language::value_print_inner (struct v
elements up to it. */
if (options->stop_print_at_null)
{
+ unsigned int print_max_chars = get_print_max_chars (options);
unsigned int temp_len;
/* Look for a NULL char. */
for (temp_len = 0;
(valaddr[temp_len]
- && temp_len < len && temp_len < options->print_max);
+ && temp_len < len
+ && temp_len < print_max_chars);
temp_len++);
len = temp_len;
}
Index: src/gdb/p-lang.c
===================================================================
--- src.orig/gdb/p-lang.c
+++ src/gdb/p-lang.c
@@ -253,7 +253,8 @@ pascal_language::printstr (struct ui_fil
return;
}
- for (i = 0; i < length && things_printed < options->print_max; ++i)
+ unsigned int print_max_chars = get_print_max_chars (options);
+ for (i = 0; i < length && things_printed < print_max_chars; ++i)
{
/* Position of the character we are examining
to see whether it is repeated. */
Index: src/gdb/p-valprint.c
===================================================================
--- src.orig/gdb/p-valprint.c
+++ src/gdb/p-valprint.c
@@ -105,13 +105,16 @@ pascal_language::value_print_inner (stru
elements up to it. */
if (options->stop_print_at_null)
{
+ unsigned int print_max_chars
+ = get_print_max_chars (options);
unsigned int temp_len;
/* Look for a NULL char. */
for (temp_len = 0;
- extract_unsigned_integer (valaddr + temp_len * eltlen,
- eltlen, byte_order)
- && temp_len < len && temp_len < options->print_max;
+ (extract_unsigned_integer
+ (valaddr + temp_len * eltlen, eltlen, byte_order)
+ && temp_len < len
+ && temp_len < print_max_chars);
temp_len++);
len = temp_len;
}
Index: src/gdb/printcmd.c
===================================================================
--- src.orig/gdb/printcmd.c
+++ src/gdb/printcmd.c
@@ -962,17 +962,18 @@ find_string_backward (struct gdbarch *gd
chars_to_read * char_size);
chars_read /= char_size;
read_error = (chars_read == chars_to_read) ? 0 : 1;
+ unsigned int print_max_chars = get_print_max_chars (options);
/* Searching for '\0' from the end of buffer in backward direction. */
for (i = 0; i < chars_read && count > 0 ; ++i, ++chars_counted)
{
int offset = (chars_to_read - i - 1) * char_size;
if (integer_is_zero (&buffer[offset], char_size)
- || chars_counted == options->print_max)
+ || chars_counted == print_max_chars)
{
- /* Found '\0' or reached print_max. As OFFSET is the offset to
- '\0', we add CHAR_SIZE to return the start address of
- a string. */
+ /* Found '\0' or reached `print_max_chars'. As OFFSET
+ is the offset to '\0', we add CHAR_SIZE to return
+ the start address of a string. */
--count;
string_start_addr = addr + offset + char_size;
chars_counted = 0;
Index: src/gdb/python/py-value.c
===================================================================
--- src.orig/gdb/python/py-value.c
+++ src/gdb/python/py-value.c
@@ -645,6 +645,7 @@ valpy_format_string (PyObject *self, PyO
"actual_objects", /* See set print object on|off. */
"static_members", /* See set print static-members on|off. */
/* C non-bool options. */
+ "max_characters", /* See set print characters N. */
"max_elements", /* See set print elements N. */
"max_depth", /* See set print max-depth N. */
"repeat_threshold", /* See set print repeats. */
@@ -691,7 +692,7 @@ valpy_format_string (PyObject *self, PyO
char *format = NULL;
if (!gdb_PyArg_ParseTupleAndKeywords (args,
kw,
- "|O!O!O!O!O!O!O!O!O!O!O!IIIs",
+ "|O!O!O!O!O!O!O!O!O!O!O!IIIIs",
keywords,
&PyBool_Type, &raw_obj,
&PyBool_Type, &pretty_arrays_obj,
@@ -704,6 +705,7 @@ valpy_format_string (PyObject *self, PyO
&PyBool_Type, &deref_refs_obj,
&PyBool_Type, &actual_objects_obj,
&PyBool_Type, &static_members_obj,
+ &opts.print_max_chars,
&opts.print_max,
&opts.max_depth,
&opts.repeat_count_threshold,
Index: src/gdb/testsuite/gdb.base/default.exp
===================================================================
--- src.orig/gdb/testsuite/gdb.base/default.exp
+++ src/gdb/testsuite/gdb.base/default.exp
@@ -522,6 +522,10 @@ gdb_test_no_output "set print address" "
gdb_test_no_output "set print array" "set print array"
#test set print asm-demangle
gdb_test_no_output "set print asm-demangle" "set print asm-demangle"
+#test set print characters
+gdb_test "set print characters" \
+ "Argument required \\(integer to set it to, or one of:\
+ \"elements\", \"unlimited\"\\)\\."
#test set print demangle
gdb_test_no_output "set print demangle" "set print demangle"
#test set print elements
@@ -665,6 +669,9 @@ gdb_test "show print address" "Printing
gdb_test "show print array" "Pretty formatting of arrays is on."
#test show print asm-demangle
gdb_test "show print asm-demangle" "Demangling of C\[+\]+/ObjC names in disassembly listings is on."
+#test show print characters
+gdb_test "show print characters" \
+ "Limit on string characters to print is elements\\."
#test show print demangle
gdb_test "show print demangle" "Demangling of encoded C\[+\]+/ObjC names when displaying symbols is on."
#test show print elements
Index: src/gdb/testsuite/gdb.base/options.exp
===================================================================
--- src.orig/gdb/testsuite/gdb.base/options.exp
+++ src/gdb/testsuite/gdb.base/options.exp
@@ -165,6 +165,7 @@ proc_with_prefix test-print {{prefix ""}
"-address"
"-array"
"-array-indexes"
+ "-characters"
"-elements"
"-max-depth"
"-memory-tag-violations"
Index: src/gdb/tracepoint.c
===================================================================
--- src.orig/gdb/tracepoint.c
+++ src/gdb/tracepoint.c
@@ -546,9 +546,9 @@ decode_agent_options (const char *exp, i
if (target_supports_string_tracing ())
{
/* Allow an optional decimal number giving an explicit maximum
- string length, defaulting it to the "print elements" value;
+ string length, defaulting it to the "print characters" value;
so "collect/s80 mystr" gets at most 80 bytes of string. */
- *trace_string = opts.print_max;
+ *trace_string = get_print_max_chars (&opts);
exp++;
if (*exp >= '0' && *exp <= '9')
*trace_string = atoi (exp);
Index: src/gdb/valprint.c
===================================================================
--- src.orig/gdb/valprint.c
+++ src/gdb/valprint.c
@@ -97,8 +97,14 @@ static void val_print_type_code_flags (s
int embedded_offset,
struct ui_file *stream);
-#define PRINT_MAX_DEFAULT 200 /* Start print_max off at this value. */
-#define PRINT_MAX_DEPTH_DEFAULT 20 /* Start print_max_depth off at this value. */
+/* Start print_max at this value. */
+#define PRINT_MAX_DEFAULT 200
+
+/* Start print_max_chars at this value (meaning follow print_max). */
+#define PRINT_MAX_CHARS_DEFAULT 0
+
+/* Start print_max_depth at this value. */
+#define PRINT_MAX_DEPTH_DEFAULT 20
struct value_print_options user_print_options =
{
@@ -110,6 +116,7 @@ struct value_print_options user_print_op
1, /* addressprint */
0, /* objectprint */
PRINT_MAX_DEFAULT, /* print_max */
+ PRINT_MAX_CHARS_DEFAULT, /* print_max_chars */
10, /* repeat_count_threshold */
0, /* output_format */
0, /* format */
@@ -152,17 +159,31 @@ get_formatted_print_options (struct valu
opts->format = format;
}
+/* Implement 'show print elements'. */
+
static void
show_print_max (struct ui_file *file, int from_tty,
struct cmd_list_element *c, const char *value)
{
+ fprintf_filtered
+ (file,
+ (user_print_options.print_max_chars != 0
+ ? _("Limit on array elements to print is %s.\n")
+ : _("Limit on string chars or array elements to print is %s.\n")),
+ value);
+}
+
+/* Implement 'show print characters'. */
+
+static void
+show_print_max_chars (struct ui_file *file, int from_tty,
+ struct cmd_list_element *c, const char *value)
+{
fprintf_filtered (file,
- _("Limit on string chars or array "
- "elements to print is %s.\n"),
+ _("Limit on string characters to print is %s.\n"),
value);
}
-
/* Default input and output radixes, and output format letter. */
unsigned input_radix = 10;
@@ -2587,9 +2608,9 @@ print_converted_chars_to_obstack (struct
/* Print the character string STRING, printing at most LENGTH
characters. LENGTH is -1 if the string is nul terminated. TYPE is
the type of each character. OPTIONS holds the printing options;
- printing stops early if the number hits print_max; repeat counts
- are printed as appropriate. Print ellipses at the end if we had to
- stop before printing LENGTH characters, or if FORCE_ELLIPSES.
+ printing stops early if the number hits print_max_chars; repeat
+ counts are printed as appropriate. Print ellipses at the end if we
+ had to stop before printing LENGTH characters, or if FORCE_ELLIPSES.
QUOTE_CHAR is the character to print at each end of the string. If
C_STYLE_TERMINATOR is true, and the last character is 0, then it is
omitted. */
@@ -2643,7 +2664,8 @@ generic_printstr (struct ui_file *stream
/* Convert characters until the string is over or the maximum
number of printed characters has been reached. */
i = 0;
- while (i < options->print_max)
+ unsigned int print_max_chars = get_print_max_chars (options);
+ while (i < print_max_chars)
{
int r;
@@ -2695,7 +2717,7 @@ generic_printstr (struct ui_file *stream
/* Print a string from the inferior, starting at ADDR and printing up to LEN
characters, of WIDTH bytes a piece, to STREAM. If LEN is -1, printing
stops at the first null byte, otherwise printing proceeds (including null
- bytes) until either print_max or LEN characters have been printed,
+ bytes) until either print_max_chars or LEN characters have been printed,
whichever is smaller. ENCODING is the name of the string's
encoding. It can be NULL, in which case the target encoding is
assumed. */
@@ -2717,15 +2739,17 @@ val_print_string (struct type *elttype,
int width = TYPE_LENGTH (elttype);
/* First we need to figure out the limit on the number of characters we are
- going to attempt to fetch and print. This is actually pretty simple. If
- LEN >= zero, then the limit is the minimum of LEN and print_max. If
- LEN is -1, then the limit is print_max. This is true regardless of
- whether print_max is zero, UINT_MAX (unlimited), or something in between,
- because finding the null byte (or available memory) is what actually
- limits the fetch. */
+ going to attempt to fetch and print. This is actually pretty simple.
+ If LEN >= zero, then the limit is the minimum of LEN and print_max_chars.
+ If LEN is -1, then the limit is print_max_chars. This is true regardless
+ of whether print_max_chars is zero, UINT_MAX (unlimited), or something in
+ between, because finding the null byte (or available memory) is what
+ actually limits the fetch. */
- fetchlimit = (len == -1 ? options->print_max : std::min ((unsigned) len,
- options->print_max));
+ unsigned int print_max_chars = get_print_max_chars (options);
+ fetchlimit = (len == -1
+ ? print_max_chars
+ : std::min ((unsigned) len, print_max_chars));
err = read_string (addr, len, width, fetchlimit, byte_order,
&buffer, &bytes_read);
@@ -2970,6 +2994,15 @@ using uinteger_option_def
using integer_option_def
= gdb::option::integer_option_def<value_print_options>;
+/* Extra literals supported with the `set print characters' and
+ `print -characters' commands. */
+static const literal_def print_characters_literals[] =
+ {
+ { "elements", 0, 0, false },
+ { "unlimited", UINT_MAX, UINT_MAX, false },
+ { nullptr }
+ };
+
/* Definitions of options for the "print" and "compile print"
commands. */
static const gdb::option::option_def value_print_option_defs[] = {
@@ -3002,12 +3035,23 @@ static const gdb::option::option_def val
},
uinteger_option_def {
+ "characters",
+ [] (value_print_options *opt) { return &opt->print_max_chars; },
+ print_characters_literals,
+ show_print_max_chars, /* show_cmd_cb */
+ N_("Set limit on string chars to print."),
+ N_("Show limit on string chars to print."),
+ N_("\"elements\" causes the array element limit to be used.\n"
+ "\"unlimited\" causes there to be no limit."),
+ },
+
+ uinteger_option_def {
"elements",
[] (value_print_options *opt) { return &opt->print_max; },
uinteger_unlimited_literals,
show_print_max, /* show_cmd_cb */
- N_("Set limit on string chars or array elements to print."),
- N_("Show limit on string chars or array elements to print."),
+ N_("Set limit on array elements to print."),
+ N_("Show limit on array elements to print."),
N_("\"unlimited\" causes there to be no limit."),
},
Index: src/gdb/valprint.h
===================================================================
--- src.orig/gdb/valprint.h
+++ src/gdb/valprint.h
@@ -48,12 +48,15 @@ struct value_print_options
in its vtables. */
bool objectprint;
- /* Maximum number of chars to print for a string pointer value or vector
- contents, or UINT_MAX for no limit. Note that "set print elements 0"
- stores UINT_MAX in print_max, which displays in a show command as
- "unlimited". */
+ /* Maximum number of elements to print for vector contents, or UINT_MAX
+ for no limit. Note that "set print elements 0" stores UINT_MAX in
+ print_max, which displays in a show command as "unlimited". */
unsigned int print_max;
+ /* Maximum number of string chars to print for a string pointer value,
+ zero if to follow the value of print_max, or UINT_MAX for no limit. */
+ unsigned int print_max_chars;
+
/* Print repeat counts if there are more than this many repetitions
of an element in an array. */
unsigned int repeat_count_threshold;
@@ -105,6 +108,15 @@ struct value_print_options
bool finish_print;
};
+/* Return the character count limit for printing strings. */
+
+static inline unsigned int
+get_print_max_chars (const struct value_print_options *options)
+{
+ return (options->print_max_chars != 0
+ ? options->print_max_chars : options->print_max);
+}
+
/* Create an option_def_group for the value_print options, with OPTS
as context. */
extern gdb::option::option_def_group make_value_print_options_def_group
^ permalink raw reply [flat|nested] 22+ messages in thread
* [PATCH v4 8/8] GDB/testsuite: Expand for character string limiting options
2022-03-09 19:17 [PATCH v4 0/8] gdb: split array and string limiting options Maciej W. Rozycki
` (6 preceding siblings ...)
2022-03-09 19:18 ` [PATCH v4 7/8] GDB: Add a character string limiting option Maciej W. Rozycki
@ 2022-03-09 19:18 ` Maciej W. Rozycki
2022-03-09 19:47 ` [PATCH v4 0/8] gdb: split array and " Simon Sobisch
8 siblings, 0 replies; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-09 19:18 UTC (permalink / raw)
To: gdb-patches; +Cc: Andrew Burgess, Simon Marchi, Tom Tromey, Simon Sobisch
From: Andrew Burgess <andrew.burgess@embecosm.com>
Modify test cases that verify the operation of the array element limit
with character strings such that they are executed twice, once with the
`set print characters' option set to `elements' and the limit controlled
with the `set print elements' option, and then again with the limit
controlled with the `set print characters' option instead. Similarly
with the `-elements' and `-characters' options for the `print' command.
Additionally verify that said `print' command options combined yield the
expected result.
There are new tests for Ada and Pascal, as the string printing code for
these languages is different than the generic string printing code used
by other languages. Modula2 also has different string printing code,
but (a) this is similar to Pascal, and (b) there are no existing modula2
tests written in Modula2, so I'm not sure how I'd even test the Modula2
string printing.
Co-Authored-By: Maciej W. Rozycki <macro@embecosm.com>
---
Changes from v3:
- split off from what is now 7/8; see the change log there for earlier
changes,
- remove test case modifications to switch from the `set print elements'
command to `set print characters'; instead run them twice with each of
the two commands verified,
- likewise with the `print -elements' and `print -characters' commands,
- also cover `print -elements ... -characters ...', i.e. both options
combined,
- expand the Ada and Pascal test cases to cover `set print characters
elements' too.
---
gdb/testsuite/gdb.ada/str_chars.exp | 70 +++++++++++++++++++++++
gdb/testsuite/gdb.ada/str_chars/foo.adb | 26 ++++++++
gdb/testsuite/gdb.base/printcmds.exp | 77 ++++++++++++++++++--------
gdb/testsuite/gdb.pascal/str-chars.exp | 56 ++++++++++++++++++
gdb/testsuite/gdb.pascal/str-chars.pas | 28 +++++++++
gdb/testsuite/gdb.python/py-format-string.exp | 47 ++++++++++-----
6 files changed, 266 insertions(+), 38 deletions(-)
gdb-aburgess-print-elements-characters-test.diff
Index: src/gdb/testsuite/gdb.ada/str_chars.exp
===================================================================
--- /dev/null
+++ src/gdb/testsuite/gdb.ada/str_chars.exp
@@ -0,0 +1,70 @@
+# Copyright 2022 Free Software Foundation, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, 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 the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+# Test GDB's 'set print characters' setting works for Ada strings.
+
+load_lib "ada.exp"
+
+if { [skip_ada_tests] } { return -1 }
+
+standard_ada_testfile foo
+
+if {[gdb_compile_ada "${srcfile}" "${binfile}" executable [list debug ]] != "" } {
+ return -1
+}
+
+clean_restart ${testfile}
+
+set bp_location [gdb_get_line_number "STOP" ${testdir}/foo.adb]
+if ![runto "foo.adb:$bp_location" ] then {
+ return -1
+}
+
+gdb_test "print Arg" \
+ " = \"abcdefghijklmnopqrstuvwxyz\"" \
+ "print with default settings"
+
+gdb_test_no_output "set print characters 26"
+gdb_test "print Arg" \
+ " = \"abcdefghijklmnopqrstuvwxyz\"" \
+ "print with character limit of 26"
+
+gdb_test "print -characters 11 -- Arg" \
+ " = \"abcdefghijk\"\\.\\.\\." \
+ "print with character limit of 11"
+
+gdb_test_no_output "set print characters 10"
+gdb_test "print Arg" \
+ " = \"abcdefghij\"\\.\\.\\." \
+ "print with character limit of 10"
+
+gdb_test_no_output "set print characters unlimited"
+gdb_test "print Arg" \
+ " = \"abcdefghijklmnopqrstuvwxyz\"" \
+ "print with unlimited character limit"
+
+# The 'set print elements' command used to control printing of characters
+# in a string, before we created 'set print characters'. This test makes
+# sure that 'set print elements' doens't effect string printing any more.
+gdb_test_no_output "set print elements 12"
+gdb_test "print Arg" \
+ " = \"abcdefghijklmnopqrstuvwxyz\"" \
+ "print with unlimited character limit, but lower element limit"
+
+# Except when 'set print characters elements' has been used.
+gdb_test_no_output "set print characters elements"
+gdb_test "print Arg" \
+ " = \"abcdefghijkl\"\\.\\.\\." \
+ "print with character limit of elements"
Index: src/gdb/testsuite/gdb.ada/str_chars/foo.adb
===================================================================
--- /dev/null
+++ src/gdb/testsuite/gdb.ada/str_chars/foo.adb
@@ -0,0 +1,26 @@
+-- Copyright 2022 Free Software Foundation, Inc.
+--
+-- This program is free software; you can redistribute it and/or modify
+-- it under the terms of the GNU General Public License as published by
+-- the Free Software Foundation; either version 3 of the License, 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 the
+-- GNU General Public License for more details.
+--
+-- You should have received a copy of the GNU General Public License
+-- along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+procedure Foo is
+
+ procedure Blah (Arg : String) is
+ begin
+ null; -- STOP
+ end;
+
+begin
+
+ Blah ("abcdefghijklmnopqrstuvwxyz");
+end Foo;
Index: src/gdb/testsuite/gdb.base/printcmds.exp
===================================================================
--- src.orig/gdb/testsuite/gdb.base/printcmds.exp
+++ src/gdb/testsuite/gdb.base/printcmds.exp
@@ -434,11 +434,12 @@ proc test_print_all_chars {} {
# Test interaction of the number of print elements to print and the
# repeat count, set to the default of 10.
-proc test_print_repeats_10 {} {
+proc test_print_repeats_10_one { setting } {
global gdb_prompt decimal
for { set x 1 } { $x <= 16 } { incr x } {
- gdb_test_no_output "set print elements $x" "elements $x repeats"
+ gdb_test_no_output "set print $setting $x" \
+ "set print $setting $x repeats"
for { set e 1 } { $e <= 16 } {incr e } {
set v [expr $e - 1]
set command "p &ctable2\[${v}*16\]"
@@ -478,11 +479,19 @@ proc test_print_repeats_10 {} {
set xstr "${xstr}\[.\]\[.\]\[.\]"
}
set string " = \[(\]unsigned char \[*\]\[)\] <ctable2(\\+$decimal)?> ${a}${xstr}"
- gdb_test "$command" "$string" "$command with print elements set to $x"
+ gdb_test "$command" "$string" \
+ "$command with print $setting set to $x"
}
}
}
+proc test_print_repeats_10 {} {
+ foreach_with_prefix setting { "elements" "characters" } {
+ test_print_repeats_10_one $setting
+ }
+ gdb_test_no_output "set print characters elements"
+}
+
# This tests whether GDB uses the correct element content offsets
# (relative to the complete `some_struct' value) when counting value
# repetitions.
@@ -495,43 +504,38 @@ proc test_print_repeats_embedded_array {
"correct element repeats in array embedded at offset > 0"
}
-proc test_print_strings {} {
+proc test_print_strings_one { setting } {
global gdb_prompt decimal
# We accept "(unsigned char *) " before the string. char vs. unsigned char
# is already tested elsewhere.
- # Test that setting print elements unlimited doesn't completely suppress
- # printing; this was a bug in older gdb's.
- gdb_test_no_output "set print elements 0"
- gdb_test "p teststring" \
- " = (.unsigned char .. )?\"teststring contents\"" "p teststring with elements set to 0"
- gdb_test_no_output "set print elements 1"
+ gdb_test_no_output "set print $setting 1"
gdb_test "p teststring" \
- " = (.unsigned char .. )?\"t\"\\.\\.\\." "p teststring with elements set to 1"
- gdb_test_no_output "set print elements 5"
+ " = (.unsigned char .. )?\"t\"\\.\\.\\." "p teststring with $setting set to 1"
+ gdb_test_no_output "set print $setting 5"
gdb_test "p teststring" \
- " = (.unsigned char .. )?\"tests\"\\.\\.\\." "p teststring with elements set to 5"
- gdb_test_no_output "set print elements 19"
+ " = (.unsigned char .. )?\"tests\"\\.\\.\\." "p teststring with $setting set to 5"
+ gdb_test_no_output "set print $setting 19"
gdb_test "p teststring" \
- " = (.unsigned char .. )?\"teststring contents\"" "p teststring with elements set to 19"
- gdb_test_no_output "set print elements 20"
+ " = (.unsigned char .. )?\"teststring contents\"" "p teststring with $setting set to 19"
+ gdb_test_no_output "set print $setting 20"
gdb_test "p teststring" \
- " = (.unsigned char .. )?\"teststring contents\"" "p teststring with elements set to 20"
+ " = (.unsigned char .. )?\"teststring contents\"" "p teststring with $setting set to 20"
- gdb_test "p -elements 1 -- teststring" \
+ gdb_test "p -$setting 1 -- teststring" \
" = (.unsigned char .. )?\"t\"\\.\\.\\."
- gdb_test "p -elements 5 -- teststring" \
+ gdb_test "p -$setting 5 -- teststring" \
" = (.unsigned char .. )?\"tests\"\\.\\.\\."
- gdb_test "p -elements 19 -- teststring" \
+ gdb_test "p -$setting 19 -- teststring" \
" = (.unsigned char .. )?\"teststring contents\""
- gdb_test "p -elements 20 -- teststring" \
+ gdb_test "p -$setting 20 -- teststring" \
" = (.unsigned char .. )?\"teststring contents\""
gdb_test "print teststring2" \
" = \\(charptr\\) \"more contents\""
- gdb_test_no_output "set print elements 8"
+ gdb_test_no_output "set print $setting 8"
# Set the target-charset to ASCII, because the output varies from
# different charset.
@@ -603,6 +607,35 @@ proc test_print_strings {} {
gdb_test "p &ctable1\[31*8\]" \
" = \\(unsigned char \\*\\) <ctable1\\+$decimal> \"\\\\370\\\\371\\\\372\\\\373\\\\374\\\\375\\\\376\\\\377\"..."
}
+
+ gdb_test_no_output "set print $setting unlimited"
+}
+
+proc test_print_strings {} {
+
+ # Test that setting print elements unlimited doesn't completely suppress
+ # printing; this was a bug in older gdb's.
+ gdb_test_no_output "set print elements 0"
+ gdb_test "p teststring" \
+ " = (.unsigned char .. )?\"teststring contents\"" \
+ "p teststring with elements set to 0"
+
+ gdb_test "set print characters 0" "integer 0 out of range"
+
+ foreach_with_prefix setting { "elements" "characters" } {
+ test_print_strings_one $setting
+ }
+
+ gdb_test "p -elements 8 -- teststring" \
+ " = (.unsigned char .. )?\"teststring contents\""
+ gdb_test "p -characters 8 -- teststring" \
+ " = (.unsigned char .. )?\"teststri\"\\.\\.\\."
+ gdb_test "p -elements 8 -characters elements -- teststring" \
+ " = (.unsigned char .. )?\"teststri\"\\.\\.\\."
+
+ with_test_prefix strings {
+ gdb_test_no_output "set print characters elements"
+ }
}
proc test_print_int_arrays {} {
Index: src/gdb/testsuite/gdb.pascal/str-chars.exp
===================================================================
--- /dev/null
+++ src/gdb/testsuite/gdb.pascal/str-chars.exp
@@ -0,0 +1,56 @@
+# Copyright 2022 Free Software Foundation, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+load_lib "pascal.exp"
+
+standard_testfile .pas
+
+if {[gdb_compile_pascal "${srcdir}/${subdir}/${srcfile}" "${binfile}" executable [list debug ]] != "" } {
+ return -1
+}
+
+clean_restart ${testfile}
+gdb_breakpoint ${srcfile}:[gdb_get_line_number "set breakpoint 1 here"]
+
+# Verify that "start" lands inside the right procedure.
+if { [gdb_start_cmd] < 0 } {
+ untested start
+ return -1
+}
+
+gdb_continue_to_breakpoint "continue to breakpoint"
+
+gdb_test "print message" " = 'abcdefghijklmnopqrstuvwxyz'" \
+ "print message with the default settings"
+
+gdb_test_no_output "set print elements 10"
+gdb_test "print message" " = 'abcdefghij'\\.\\.\\." \
+ "print message with 'print elements' set to 10"
+
+gdb_test_no_output "set print characters 20"
+gdb_test "print message" " = 'abcdefghijklmnopqrst'\\.\\.\\." \
+ "print message with 'print characters' set to 20"
+
+gdb_test_no_output "set print elements 15"
+gdb_test "print message" " = 'abcdefghijklmnopqrst'\\.\\.\\." \
+ "print message with 'print elements' set to 15"
+
+gdb_test_no_output "set print characters unlimited"
+gdb_test "print message" " = 'abcdefghijklmnopqrstuvwxyz'" \
+ "print message with 'print characters' set to unlimited"
+
+gdb_test_no_output "set print characters elements"
+gdb_test "print message" " = 'abcdefghijklmno'\\.\\.\\." \
+ "print message with 'print characters' set to elements"
Index: src/gdb/testsuite/gdb.pascal/str-chars.pas
===================================================================
--- /dev/null
+++ src/gdb/testsuite/gdb.pascal/str-chars.pas
@@ -0,0 +1,28 @@
+{
+ Copyright 2022 Free Software Foundation, Inc.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 3 of the License, 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 the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>.
+}
+
+program str_char;
+
+var
+ message : string;
+
+begin
+
+ message := 'abcdefghijklmnopqrstuvwxyz';
+ writeln (message) { set breakpoint 1 here }
+
+end.
Index: src/gdb/testsuite/gdb.python/py-format-string.exp
===================================================================
--- src.orig/gdb/testsuite/gdb.python/py-format-string.exp
+++ src/gdb/testsuite/gdb.python/py-format-string.exp
@@ -632,13 +632,13 @@ proc_with_prefix test_static_members {}
}
# Test the max_elements option for gdb.Value.format_string.
-proc_with_prefix test_max_elements {} {
+proc test_max_string_one { setting unlimited } {
global current_lang
global default_pointer_regexp
# 200 is the default maximum number of elements, so setting it should
# not change the output.
- set opts "max_elements=200"
+ set opts "max_$setting=200"
with_test_prefix $opts {
check_format_string "a_point_t" $opts
check_format_string "a_point_t_pointer" $opts
@@ -649,8 +649,10 @@ proc_with_prefix test_max_elements {} {
check_format_string "a_binary_string" $opts
check_format_string "a_binary_string_array" $opts
check_format_string "a_big_string" $opts
- check_format_string "an_array" $opts
- check_format_string "an_array_with_repetition" $opts
+ if { $setting == "elements"} {
+ check_format_string "an_array" $opts
+ check_format_string "an_array_with_repetition" $opts
+ }
check_format_string "a_symbol_pointer" $opts
if { $current_lang == "c++" } {
@@ -659,7 +661,7 @@ proc_with_prefix test_max_elements {} {
}
}
- set opts "max_elements=3"
+ set opts "max_$setting=3"
with_test_prefix $opts {
check_format_string "a_point_t" $opts
check_format_string "a_point_t_pointer" $opts
@@ -676,9 +678,11 @@ proc_with_prefix test_max_elements {} {
"\"hell\"..."
check_format_string "a_big_string" $opts \
[get_cut_big_string 3]
- check_format_string "an_array" $opts
- check_format_string "an_array_with_repetition" $opts \
- "\\{1, 3 <repeats 12 times>...\\}"
+ if { $setting == "elements"} {
+ check_format_string "an_array" $opts
+ check_format_string "an_array_with_repetition" $opts \
+ "\\{1, 3 <repeats 12 times>...\\}"
+ }
check_format_string "a_symbol_pointer" $opts
if { $current_lang == "c++" } {
@@ -687,9 +691,9 @@ proc_with_prefix test_max_elements {} {
}
}
- # Both 1,000 (we don't have that many elements) and 0 (unlimited) should
+ # Both 1,000 (we don't have that many elements) and unlimited should
# mean no truncation.
- foreach opts { "max_elements=1000" "max_elements=0" } {
+ foreach opts [list "max_$setting=1000" "max_$setting=$unlimited"] {
with_test_prefix $opts {
check_format_string "a_point_t" $opts
check_format_string "a_point_t_pointer" $opts
@@ -701,8 +705,10 @@ proc_with_prefix test_max_elements {} {
check_format_string "a_binary_string_array" $opts
check_format_string "a_big_string" $opts \
[get_cut_big_string 1000]
- check_format_string "an_array" $opts
- check_format_string "an_array_with_repetition" $opts
+ if { $setting == "elements"} {
+ check_format_string "an_array" $opts
+ check_format_string "an_array_with_repetition" $opts
+ }
check_format_string "a_symbol_pointer" $opts
if { $current_lang == "c++" } {
@@ -712,15 +718,24 @@ proc_with_prefix test_max_elements {} {
}
}
- with_temp_option "set print elements 4" "set print elements 200" {
+ with_temp_option "set print $setting 4" "set print $setting 200" {
check_format_string "a_string" "" \
"${default_pointer_regexp} \"hell\"..."
check_format_string "a_binary_string" "" \
"${default_pointer_regexp} \"hell\"..."
check_format_string "a_binary_string_array" "" \
"\"hell\"..."
- check_format_string "an_array_with_repetition" "" \
- "\\{1, 3 <repeats 12 times>...\\}"
+ if { $setting == "elements"} {
+ check_format_string "an_array_with_repetition" "" \
+ "\\{1, 3 <repeats 12 times>...\\}"
+ }
+ }
+}
+
+proc_with_prefix test_max_string {} {
+ foreach_with_prefix setting { "elements" "characters" } {
+ test_max_string_one $setting \
+ [string map {elements 0 characters 4294967295} $setting]
}
}
@@ -1013,7 +1028,7 @@ proc_with_prefix test_all_common {} {
test_deref_refs
test_actual_objects
test_static_members
- test_max_elements
+ test_max_string
test_max_depth
test_repeat_threshold
test_format
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 0/8] gdb: split array and string limiting options
2022-03-09 19:17 [PATCH v4 0/8] gdb: split array and string limiting options Maciej W. Rozycki
` (7 preceding siblings ...)
2022-03-09 19:18 ` [PATCH v4 8/8] GDB/testsuite: Expand for character string limiting options Maciej W. Rozycki
@ 2022-03-09 19:47 ` Simon Sobisch
8 siblings, 0 replies; 22+ messages in thread
From: Simon Sobisch @ 2022-03-09 19:47 UTC (permalink / raw)
To: Maciej W. Rozycki, gdb-patches; +Cc: Andrew Burgess, Simon Marchi, Tom Tromey
Thank you for your work on this.
I like all patches but can't really comment on "6 of 8" (with the reason
that I don't understand it fully, the "theory" does sound good).
I'm really looking forward to see this (in GDB12 I assume).
Simon
Am 09.03.2022 um 20:17 schrieb Maciej W. Rozycki:
> Hi,
>
> I have retained the original subject for the cover letter for threading
> purposes, however this has now otherwise become a small patch series and
> Andrew's original change has a different heading now.
>
> This is v4 for Andrew's original proposal to split the `set/show print
> elements' options into two, so that there are separate `set/show print
> characters' options to control how many characters of a string are
> printed. Following earlier discussion for compatibility with existing
> setups I have implemented a mechanism to make the `set print characters'
> setting follow that of `set print elements' by default. Consequently what
> was a single change has now grown into a patch series.
>
> In this series I fix some preexisting issues in the area first, implement
> the general framework for extra literals used with integer settings second
> and then add the original change for `set print characters' from Andrew,
> adjusted accordingly, followed by a testsuite update as a separate change.
> This is so as to ensure the semantics of the existing `set print elements'
> knob hasn't changed in the default `set print characters' state (a small
> update for the test suite is nevertheless included with the `set print
> characters' command proper, because inevitably the relevant commands that
> list commands now print additional entries, plus there's been a minimal
> test case update included for the verification of setting defaults).
>
> I think there's little more to say here except that changes in the series
> have been individually regression-tested with a native `x86_64-linux-gnu'
> configuration, and that I have visually examined the result of changes to
> the GDB manual in the info and PDF formats.
>
> See individual change descriptions for details.
>
> OK to apply?
>
> Maciej
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-09 19:18 ` [PATCH v4 7/8] GDB: Add a character string limiting option Maciej W. Rozycki
@ 2022-03-09 19:51 ` Eli Zaretskii
2022-03-09 20:40 ` Maciej W. Rozycki
0 siblings, 1 reply; 22+ messages in thread
From: Eli Zaretskii @ 2022-03-09 19:51 UTC (permalink / raw)
To: Maciej W. Rozycki; +Cc: gdb-patches, simonsobisch, tom, aburgess
> Date: Wed, 9 Mar 2022 19:18:46 +0000 (GMT)
> From: "Maciej W. Rozycki" <macro@embecosm.com>
> Cc: Simon Sobisch <simonsobisch@web.de>, Tom Tromey <tom@tromey.com>,
> Andrew Burgess <aburgess@redhat.com>
>
> --- src.orig/gdb/NEWS
> +++ src/gdb/NEWS
> @@ -119,6 +119,17 @@ show ada source-charset
> values for this follow the values that can be passed to the GNAT
> compiler via the '-gnati' option. The default is ISO-8859-1.
>
> +set print characters LIMIT
> +show print characters
> +print -characters LIMIT
> + This new setting is like 'set print elements', but controls how many
"This new setting" is okay for "set print characters", but not for
"print -characters". I suggest to move "print -characters" to after
the description, and then to say that its effect is the same as a
temporary use of "set print characters".
> +@item -characters @var{number-of-characters}|@code{elements}|@code{unlimited}
> +Set limit on string characters to print. The value @code{elements}
> +causes the limit on array elements to print to be used. The value
^^^^^^^^^^^^^^
"characters", not "array elements", right?
And, btw, we actually count bytes, not characters, right? If the
string contains multibyte sequences, we don't count characters. I
think this should be mentioned.
> +@item -elements @var{number-of-elements}|@code{unlimited}
> +Set limit on array elements and optionally string characters to print.
> +See @ref{set print characters} and the @code{-characters} option above
^
Please insert a comma after the closing brace. Some relatively old
versions of makeinfo require that.
> +for when this option applies to strings. The value @code{unlimited}
> +causes there to be no limit. Related setting: @ref{set print elements}.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
This is not a very good idea. Texinfo is not HTML, so the results
produced for this in some formats will look awkward or even downright
incorrect English. It is better to use something like:
@xref{set print elements}, for a related CLI command.
Thanks.
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-09 19:51 ` Eli Zaretskii
@ 2022-03-09 20:40 ` Maciej W. Rozycki
2022-03-10 0:03 ` Maciej W. Rozycki
2022-03-10 6:28 ` Eli Zaretskii
0 siblings, 2 replies; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-09 20:40 UTC (permalink / raw)
To: Eli Zaretskii; +Cc: gdb-patches, simonsobisch, Tom Tromey, Andrew Burgess
On Wed, 9 Mar 2022, Eli Zaretskii wrote:
> > --- src.orig/gdb/NEWS
> > +++ src/gdb/NEWS
> > @@ -119,6 +119,17 @@ show ada source-charset
> > values for this follow the values that can be passed to the GNAT
> > compiler via the '-gnati' option. The default is ISO-8859-1.
> >
> > +set print characters LIMIT
> > +show print characters
> > +print -characters LIMIT
> > + This new setting is like 'set print elements', but controls how many
>
> "This new setting" is okay for "set print characters", but not for
> "print -characters". I suggest to move "print -characters" to after
> the description, and then to say that its effect is the same as a
> temporary use of "set print characters".
OK. I will update it in v5.
> > +@item -characters @var{number-of-characters}|@code{elements}|@code{unlimited}
> > +Set limit on string characters to print. The value @code{elements}
> > +causes the limit on array elements to print to be used. The value
> ^^^^^^^^^^^^^^
> "characters", not "array elements", right?
No, this is as intended. With a character string `foo' and this command:
(gdb) print -characters elements foo
the limit previously set by `set print elements' will be used (which is
usually the setting for array elements). Likewise:
(gdb) print -characters elements -elements 5 foo
will print up to 5 characters regardless of the current limit set by `set
print characters', and the `-elements' option usually applies to array
elements.
> And, btw, we actually count bytes, not characters, right? If the
> string contains multibyte sequences, we don't count characters. I
> think this should be mentioned.
We do count characters not bytes, as previously discussed; cf.:
<https://sourceware.org/pipermail/gdb-patches/2021-December/184466.html>.
Multibyte sequences comprising a single character are treated as such.
This change does not affect it.
> > +@item -elements @var{number-of-elements}|@code{unlimited}
> > +Set limit on array elements and optionally string characters to print.
> > +See @ref{set print characters} and the @code{-characters} option above
> ^
> Please insert a comma after the closing brace. Some relatively old
> versions of makeinfo require that.
OK. I will update it in v5. (Now that you mention it I remember it from
the old days.)
> > +for when this option applies to strings. The value @code{unlimited}
> > +causes there to be no limit. Related setting: @ref{set print elements}.
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> This is not a very good idea. Texinfo is not HTML, so the results
> produced for this in some formats will look awkward or even downright
> incorrect English. It is better to use something like:
>
> @xref{set print elements}, for a related CLI command.
Likewise. Thank you for your review.
Maciej
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-09 20:40 ` Maciej W. Rozycki
@ 2022-03-10 0:03 ` Maciej W. Rozycki
2022-03-10 6:35 ` Eli Zaretskii
2022-03-10 6:28 ` Eli Zaretskii
1 sibling, 1 reply; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-10 0:03 UTC (permalink / raw)
To: Eli Zaretskii; +Cc: gdb-patches, simonsobisch, Tom Tromey, Andrew Burgess
On Wed, 9 Mar 2022, Maciej W. Rozycki wrote:
> > > +for when this option applies to strings. The value @code{unlimited}
> > > +causes there to be no limit. Related setting: @ref{set print elements}.
> > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> > This is not a very good idea. Texinfo is not HTML, so the results
> > produced for this in some formats will look awkward or even downright
> > incorrect English. It is better to use something like:
> >
> > @xref{set print elements}, for a related CLI command.
>
> Likewise. Thank you for your review.
NB there are 21 preexisting cases of such phrasing across the GDB manual,
which you may want to (have people) clean up (yeah, a thought has struck
me I've seen this before, which is probably why Andrew copied it).
Maciej
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-09 20:40 ` Maciej W. Rozycki
2022-03-10 0:03 ` Maciej W. Rozycki
@ 2022-03-10 6:28 ` Eli Zaretskii
2022-03-10 10:08 ` Maciej W. Rozycki
1 sibling, 1 reply; 22+ messages in thread
From: Eli Zaretskii @ 2022-03-10 6:28 UTC (permalink / raw)
To: Maciej W. Rozycki; +Cc: gdb-patches, simonsobisch, tom, aburgess
> Date: Wed, 9 Mar 2022 20:40:20 +0000 (GMT)
> From: "Maciej W. Rozycki" <macro@embecosm.com>
> cc: gdb-patches@sourceware.org, simonsobisch@web.de,
> Tom Tromey <tom@tromey.com>, Andrew Burgess <aburgess@redhat.com>
>
> > And, btw, we actually count bytes, not characters, right? If the
> > string contains multibyte sequences, we don't count characters. I
> > think this should be mentioned.
>
> We do count characters not bytes, as previously discussed; cf.:
> <https://sourceware.org/pipermail/gdb-patches/2021-December/184466.html>.
> Multibyte sequences comprising a single character are treated as such.
> This change does not affect it.
OK, but I still think this should be mentioned in the manual.
Previously, the setting was for "array elements", and was not specific
to character strings (though I would think that "array elements" in
case of a 'char' strings are bytes, not multibyte character
sequences), but now we are specifically addressing character strings,
so this issue begs a question that IMO the manual should answer.
Thanks.
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-10 0:03 ` Maciej W. Rozycki
@ 2022-03-10 6:35 ` Eli Zaretskii
0 siblings, 0 replies; 22+ messages in thread
From: Eli Zaretskii @ 2022-03-10 6:35 UTC (permalink / raw)
To: Maciej W. Rozycki; +Cc: gdb-patches, simonsobisch, tom, aburgess
> Date: Thu, 10 Mar 2022 00:03:47 +0000 (GMT)
> From: "Maciej W. Rozycki" <macro@embecosm.com>
> cc: gdb-patches@sourceware.org, simonsobisch@web.de,
> Tom Tromey <tom@tromey.com>, Andrew Burgess <aburgess@redhat.com>
>
> On Wed, 9 Mar 2022, Maciej W. Rozycki wrote:
>
> > > > +for when this option applies to strings. The value @code{unlimited}
> > > > +causes there to be no limit. Related setting: @ref{set print elements}.
> > > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> > > This is not a very good idea. Texinfo is not HTML, so the results
> > > produced for this in some formats will look awkward or even downright
> > > incorrect English. It is better to use something like:
> > >
> > > @xref{set print elements}, for a related CLI command.
> >
> > Likewise. Thank you for your review.
>
> NB there are 21 preexisting cases of such phrasing across the GDB manual,
> which you may want to (have people) clean up (yeah, a thought has struck
> me I've seen this before, which is probably why Andrew copied it).
Yes, I know. Thanks for making that point, I will add that to my
todo.
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-10 6:28 ` Eli Zaretskii
@ 2022-03-10 10:08 ` Maciej W. Rozycki
2022-03-10 11:55 ` Eli Zaretskii
2022-03-10 17:15 ` Tom Tromey
0 siblings, 2 replies; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-10 10:08 UTC (permalink / raw)
To: Eli Zaretskii; +Cc: gdb-patches, simonsobisch, Tom Tromey, Andrew Burgess
On Thu, 10 Mar 2022, Eli Zaretskii wrote:
> > We do count characters not bytes, as previously discussed; cf.:
> > <https://sourceware.org/pipermail/gdb-patches/2021-December/184466.html>.
> > Multibyte sequences comprising a single character are treated as such.
> > This change does not affect it.
>
> OK, but I still think this should be mentioned in the manual.
> Previously, the setting was for "array elements", and was not specific
> to character strings (though I would think that "array elements" in
> case of a 'char' strings are bytes, not multibyte character
> sequences), but now we are specifically addressing character strings,
> so this issue begs a question that IMO the manual should answer.
I can see what I can do here, but I think this matter should be more
thoroughly investigated and the manual updated accordingly, e.g. what's
the effect of `set print sevenbit-strings on' on wide characters?
NB is there a way to orchestrate a wide character string in a way that
wouldn't show gibberish on a plain-ASCII terminal (both in the source form
and in output) so as to get this feature covered in the test suite? I'm
not an expert on character encodings beyond ASCII/8-bit. I can see we
have a bunch of settings related to this:
(gdb) show charset
The host character set is "auto; currently ANSI_X3.4-1968".
The target character set is "auto; currently ANSI_X3.4-1968".
The target wide character set is "auto; currently UTF-32".
(gdb)
Maciej
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-10 10:08 ` Maciej W. Rozycki
@ 2022-03-10 11:55 ` Eli Zaretskii
2022-03-10 12:25 ` Maciej W. Rozycki
2022-03-10 17:15 ` Tom Tromey
1 sibling, 1 reply; 22+ messages in thread
From: Eli Zaretskii @ 2022-03-10 11:55 UTC (permalink / raw)
To: Maciej W. Rozycki; +Cc: gdb-patches, simonsobisch, tom, aburgess
> Date: Thu, 10 Mar 2022 10:08:03 +0000 (GMT)
> From: "Maciej W. Rozycki" <macro@embecosm.com>
> cc: gdb-patches@sourceware.org, simonsobisch@web.de,
> Tom Tromey <tom@tromey.com>, Andrew Burgess <aburgess@redhat.com>
>
> NB is there a way to orchestrate a wide character string in a way that
> wouldn't show gibberish on a plain-ASCII terminal (both in the source form
> and in output) so as to get this feature covered in the test suite? I'm
> not an expert on character encodings beyond ASCII/8-bit. I can see we
> have a bunch of settings related to this:
>
> (gdb) show charset
> The host character set is "auto; currently ANSI_X3.4-1968".
> The target character set is "auto; currently ANSI_X3.4-1968".
> The target wide character set is "auto; currently UTF-32".
Does this affect how we count characters for the purpose of the patch
we are discussing?
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-10 11:55 ` Eli Zaretskii
@ 2022-03-10 12:25 ` Maciej W. Rozycki
0 siblings, 0 replies; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-10 12:25 UTC (permalink / raw)
To: Eli Zaretskii; +Cc: gdb-patches, simonsobisch, Tom Tromey, Andrew Burgess
On Thu, 10 Mar 2022, Eli Zaretskii wrote:
> > NB is there a way to orchestrate a wide character string in a way that
> > wouldn't show gibberish on a plain-ASCII terminal (both in the source form
> > and in output) so as to get this feature covered in the test suite? I'm
> > not an expert on character encodings beyond ASCII/8-bit. I can see we
> > have a bunch of settings related to this:
> >
> > (gdb) show charset
> > The host character set is "auto; currently ANSI_X3.4-1968".
> > The target character set is "auto; currently ANSI_X3.4-1968".
> > The target wide character set is "auto; currently UTF-32".
>
> Does this affect how we count characters for the purpose of the patch
> we are discussing?
Not directly, but as I understand it it affects the interpretation of a
byte stream retrieved as a character string (null-terminated in the case
of C/C++, but possibly represented differently with other languages such
as Pascal). So the same byte stream may have a different character count
depending on the encoding chosen.
Maciej
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-10 10:08 ` Maciej W. Rozycki
2022-03-10 11:55 ` Eli Zaretskii
@ 2022-03-10 17:15 ` Tom Tromey
2022-03-30 10:38 ` Maciej W. Rozycki
1 sibling, 1 reply; 22+ messages in thread
From: Tom Tromey @ 2022-03-10 17:15 UTC (permalink / raw)
To: Maciej W. Rozycki
Cc: Eli Zaretskii, gdb-patches, simonsobisch, Tom Tromey, Andrew Burgess
Maciej> NB is there a way to orchestrate a wide character string in a way that
Maciej> wouldn't show gibberish on a plain-ASCII terminal
If the host encoding is ASCII, then I think non-ASCII characters in a
wide string should be printed as escapes. This should always happen,
though see my series about string printing -- fully generic encoding
support when printing was not implemented for all languages, and that
series aims to unify it. That is still pending some updates by me, and
some missing reviews.
Tom
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-10 17:15 ` Tom Tromey
@ 2022-03-30 10:38 ` Maciej W. Rozycki
2022-03-30 13:15 ` Pedro Alves
0 siblings, 1 reply; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-30 10:38 UTC (permalink / raw)
To: Tom Tromey; +Cc: Eli Zaretskii, gdb-patches, simonsobisch, Andrew Burgess
On Thu, 10 Mar 2022, Tom Tromey wrote:
> Maciej> NB is there a way to orchestrate a wide character string in a way that
> Maciej> wouldn't show gibberish on a plain-ASCII terminal
>
> If the host encoding is ASCII, then I think non-ASCII characters in a
> wide string should be printed as escapes. This should always happen,
> though see my series about string printing -- fully generic encoding
> support when printing was not implemented for all languages, and that
> series aims to unify it. That is still pending some updates by me, and
> some missing reviews.
That sounds good (and I also mean the sources used in the test suite).
Also while preparing changes for the docs in v5 I have noticed this old
NEWS piece:
>>GDB supports wide- and unicode-literals in C, that is, L'x', L"string",
u'x', u"string", U'x', and U"string" syntax.<<
which I think means we can do verification by hand, because the limit is
of course also respected for strings supplied from the UI (and even with
no target connected), e.g.:
(gdb) print "foobar"
$1 = "foobar"
(gdb) set print characters 4
(gdb) print "foobar"
$2 = "foob"...
(gdb)
Though I'd yet have to figure out how to use the syntax quoted in NEWS in
a meaningful manner, and an artificial program may also be required,
because:
(gdb) print L"foobar"
No type named wchar_t.
(gdb) print u"foobar"
No type named char16_t.
(gdb) print U"foobar"
No type named char32_t.
(gdb)
though I guess the data types do not actually have to be system-supplied,
and it will be enough if they are reasonably defined.
Maciej
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-30 10:38 ` Maciej W. Rozycki
@ 2022-03-30 13:15 ` Pedro Alves
2022-03-30 14:17 ` Maciej W. Rozycki
0 siblings, 1 reply; 22+ messages in thread
From: Pedro Alves @ 2022-03-30 13:15 UTC (permalink / raw)
To: Maciej W. Rozycki, Tom Tromey; +Cc: simonsobisch, Andrew Burgess, gdb-patches
On 2022-03-30 11:38, Maciej W. Rozycki wrote:
>
> Though I'd yet have to figure out how to use the syntax quoted in NEWS in
> a meaningful manner, and an artificial program may also be required,
> because:
>
> (gdb) print L"foobar"
> No type named wchar_t.
> (gdb) print u"foobar"
> No type named char16_t.
> (gdb) print U"foobar"
> No type named char32_t.
> (gdb)
>
> though I guess the data types do not actually have to be system-supplied,
> and it will be enough if they are reasonably defined.
It will work without a program if you switch to C++ beforehand, with "set language c++",
because wchar_t/char16_t/char32_t are built-in types in C++.
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v4 7/8] GDB: Add a character string limiting option
2022-03-30 13:15 ` Pedro Alves
@ 2022-03-30 14:17 ` Maciej W. Rozycki
0 siblings, 0 replies; 22+ messages in thread
From: Maciej W. Rozycki @ 2022-03-30 14:17 UTC (permalink / raw)
To: Pedro Alves; +Cc: Tom Tromey, simonsobisch, Andrew Burgess, gdb-patches
On Wed, 30 Mar 2022, Pedro Alves wrote:
> > Though I'd yet have to figure out how to use the syntax quoted in NEWS in
> > a meaningful manner, and an artificial program may also be required,
> > because:
> >
> > (gdb) print L"foobar"
> > No type named wchar_t.
> > (gdb) print u"foobar"
> > No type named char16_t.
> > (gdb) print U"foobar"
> > No type named char32_t.
> > (gdb)
> >
> > though I guess the data types do not actually have to be system-supplied,
> > and it will be enough if they are reasonably defined.
>
> It will work without a program if you switch to C++ beforehand, with "set language c++",
> because wchar_t/char16_t/char32_t are built-in types in C++.
Excellent, and indeed it does the right thing too:
(gdb) set language c++
(gdb) print L"foobar"
$1 = L"foobar"
(gdb) print u"foobar"
$2 = u"foobar"
(gdb) print U"foobar"
$3 = U"foobar"
(gdb) set print characters 4
(gdb) print L"foobar"
$4 = L"foob"...
(gdb) print u"foobar"
$5 = u"foob"...
(gdb) print U"foobar"
$6 = U"foob"...
(gdb)
exactly as documented. Thanks for the hint!
Maciej
^ permalink raw reply [flat|nested] 22+ messages in thread
end of thread, other threads:[~2022-03-30 14:17 UTC | newest]
Thread overview: 22+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-03-09 19:17 [PATCH v4 0/8] gdb: split array and string limiting options Maciej W. Rozycki
2022-03-09 19:17 ` [PATCH v4 1/8] GDB: Remove extraneous full stops from `set' command error messages Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 2/8] GDB/Python: Use None for `var_zuinteger_unlimited' value set to `unlimited' Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 3/8] GDB: Add `NUMBER' completion to `set' integer commands Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 4/8] GDB/testsuite: Tighten `set print elements' error check Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 5/8] GDB/testsuite: Add coverage for `print -elements' command Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 6/8] GDB: Allow arbitrary keywords in integer set commands Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 7/8] GDB: Add a character string limiting option Maciej W. Rozycki
2022-03-09 19:51 ` Eli Zaretskii
2022-03-09 20:40 ` Maciej W. Rozycki
2022-03-10 0:03 ` Maciej W. Rozycki
2022-03-10 6:35 ` Eli Zaretskii
2022-03-10 6:28 ` Eli Zaretskii
2022-03-10 10:08 ` Maciej W. Rozycki
2022-03-10 11:55 ` Eli Zaretskii
2022-03-10 12:25 ` Maciej W. Rozycki
2022-03-10 17:15 ` Tom Tromey
2022-03-30 10:38 ` Maciej W. Rozycki
2022-03-30 13:15 ` Pedro Alves
2022-03-30 14:17 ` Maciej W. Rozycki
2022-03-09 19:18 ` [PATCH v4 8/8] GDB/testsuite: Expand for character string limiting options Maciej W. Rozycki
2022-03-09 19:47 ` [PATCH v4 0/8] gdb: split array and " Simon Sobisch
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).