1 @c Copyright (C) 2008-2018 Free Software Foundation, Inc.
2 @c Permission is granted to copy, distribute and/or modify this document
3 @c under the terms of the GNU Free Documentation License, Version 1.3 or
4 @c any later version published by the Free Software Foundation; with the
5 @c Invariant Sections being ``Free Software'' and ``Free Software Needs
6 @c Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
7 @c and with the Back-Cover Texts as in (a) below.
9 @c (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
10 @c this GNU Manual. Buying copies from GNU Press supports the FSF in
11 @c developing GNU and promoting software freedom.''
14 @section Extending @value{GDBN} using Python
15 @cindex python scripting
16 @cindex scripting with python
18 You can extend @value{GDBN} using the @uref{http://www.python.org/,
19 Python programming language}. This feature is available only if
20 @value{GDBN} was configured using @option{--with-python}.
21 @value{GDBN} can be built against either Python 2 or Python 3; which
22 one you have depends on this configure-time option.
24 @cindex python directory
25 Python scripts used by @value{GDBN} should be installed in
26 @file{@var{data-directory}/python}, where @var{data-directory} is
27 the data directory as determined at @value{GDBN} startup (@pxref{Data Files}).
28 This directory, known as the @dfn{python directory},
29 is automatically added to the Python Search Path in order to allow
30 the Python interpreter to locate all scripts installed at this location.
32 Additionally, @value{GDBN} commands and convenience functions which
33 are written in Python and are located in the
34 @file{@var{data-directory}/python/gdb/command} or
35 @file{@var{data-directory}/python/gdb/function} directories are
36 automatically imported when @value{GDBN} starts.
39 * Python Commands:: Accessing Python from @value{GDBN}.
40 * Python API:: Accessing @value{GDBN} from Python.
41 * Python Auto-loading:: Automatically loading Python code.
42 * Python modules:: Python modules provided by @value{GDBN}.
46 @subsection Python Commands
47 @cindex python commands
48 @cindex commands to access python
50 @value{GDBN} provides two commands for accessing the Python interpreter,
51 and one related setting:
54 @kindex python-interactive
56 @item python-interactive @r{[}@var{command}@r{]}
57 @itemx pi @r{[}@var{command}@r{]}
58 Without an argument, the @code{python-interactive} command can be used
59 to start an interactive Python prompt. To return to @value{GDBN},
60 type the @code{EOF} character (e.g., @kbd{Ctrl-D} on an empty prompt).
62 Alternatively, a single-line Python command can be given as an
63 argument and evaluated. If the command is an expression, the result
64 will be printed; otherwise, nothing will be printed. For example:
67 (@value{GDBP}) python-interactive 2 + 3
73 @item python @r{[}@var{command}@r{]}
74 @itemx py @r{[}@var{command}@r{]}
75 The @code{python} command can be used to evaluate Python code.
77 If given an argument, the @code{python} command will evaluate the
78 argument as a Python command. For example:
81 (@value{GDBP}) python print 23
85 If you do not provide an argument to @code{python}, it will act as a
86 multi-line command, like @code{define}. In this case, the Python
87 script is made up of subsequent command lines, given after the
88 @code{python} command. This command list is terminated using a line
89 containing @code{end}. For example:
94 End with a line saying just "end".
100 @kindex set python print-stack
101 @item set python print-stack
102 By default, @value{GDBN} will print only the message component of a
103 Python exception when an error occurs in a Python script. This can be
104 controlled using @code{set python print-stack}: if @code{full}, then
105 full Python stack printing is enabled; if @code{none}, then Python stack
106 and message printing is disabled; if @code{message}, the default, only
107 the message component of the error is printed.
110 It is also possible to execute a Python script from the @value{GDBN}
114 @item source @file{script-name}
115 The script name must end with @samp{.py} and @value{GDBN} must be configured
116 to recognize the script language based on filename extension using
117 the @code{script-extension} setting. @xref{Extending GDB, ,Extending GDB}.
121 @subsection Python API
123 @cindex programming in python
125 You can get quick online help for @value{GDBN}'s Python API by issuing
126 the command @w{@kbd{python help (gdb)}}.
128 Functions and methods which have two or more optional arguments allow
129 them to be specified using keyword syntax. This allows passing some
130 optional arguments while skipping others. Example:
131 @w{@code{gdb.some_function ('foo', bar = 1, baz = 2)}}.
134 * Basic Python:: Basic Python Functions.
135 * Exception Handling:: How Python exceptions are translated.
136 * Values From Inferior:: Python representation of values.
137 * Types In Python:: Python representation of types.
138 * Pretty Printing API:: Pretty-printing values.
139 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
140 * Writing a Pretty-Printer:: Writing a Pretty-Printer.
141 * Type Printing API:: Pretty-printing types.
142 * Frame Filter API:: Filtering Frames.
143 * Frame Decorator API:: Decorating Frames.
144 * Writing a Frame Filter:: Writing a Frame Filter.
145 * Unwinding Frames in Python:: Writing frame unwinder.
146 * Xmethods In Python:: Adding and replacing methods of C++ classes.
147 * Xmethod API:: Xmethod types.
148 * Writing an Xmethod:: Writing an xmethod.
149 * Inferiors In Python:: Python representation of inferiors (processes)
150 * Events In Python:: Listening for events from @value{GDBN}.
151 * Threads In Python:: Accessing inferior threads from Python.
152 * Recordings In Python:: Accessing recordings from Python.
153 * Commands In Python:: Implementing new commands in Python.
154 * Parameters In Python:: Adding new @value{GDBN} parameters.
155 * Functions In Python:: Writing new convenience functions.
156 * Progspaces In Python:: Program spaces.
157 * Objfiles In Python:: Object files.
158 * Frames In Python:: Accessing inferior stack frames from Python.
159 * Blocks In Python:: Accessing blocks from Python.
160 * Symbols In Python:: Python representation of symbols.
161 * Symbol Tables In Python:: Python representation of symbol tables.
162 * Line Tables In Python:: Python representation of line tables.
163 * Breakpoints In Python:: Manipulating breakpoints using Python.
164 * Finish Breakpoints in Python:: Setting Breakpoints on function return
166 * Lazy Strings In Python:: Python representation of lazy strings.
167 * Architectures In Python:: Python representation of architectures.
171 @subsubsection Basic Python
173 @cindex python stdout
174 @cindex python pagination
175 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
176 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
177 A Python program which outputs to one of these streams may have its
178 output interrupted by the user (@pxref{Screen Size}). In this
179 situation, a Python @code{KeyboardInterrupt} exception is thrown.
181 Some care must be taken when writing Python code to run in
182 @value{GDBN}. Two things worth noting in particular:
186 @value{GDBN} install handlers for @code{SIGCHLD} and @code{SIGINT}.
187 Python code must not override these, or even change the options using
188 @code{sigaction}. If your program changes the handling of these
189 signals, @value{GDBN} will most likely stop working correctly. Note
190 that it is unfortunately common for GUI toolkits to install a
191 @code{SIGCHLD} handler.
194 @value{GDBN} takes care to mark its internal file descriptors as
195 close-on-exec. However, this cannot be done in a thread-safe way on
196 all platforms. Your Python programs should be aware of this and
197 should both create new file descriptors with the close-on-exec flag
198 set and arrange to close unneeded file descriptors before starting a
202 @cindex python functions
203 @cindex python module
205 @value{GDBN} introduces a new Python module, named @code{gdb}. All
206 methods and classes added by @value{GDBN} are placed in this module.
207 @value{GDBN} automatically @code{import}s the @code{gdb} module for
208 use in all scripts evaluated by the @code{python} command.
210 Some types of the @code{gdb} module come with a textual representation
211 (accessible through the @code{repr} or @code{str} functions). These are
212 offered for debugging purposes only, expect them to change over time.
214 @findex gdb.PYTHONDIR
215 @defvar gdb.PYTHONDIR
216 A string containing the python directory (@pxref{Python}).
220 @defun gdb.execute (command @r{[}, from_tty @r{[}, to_string@r{]]})
221 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
222 If a GDB exception happens while @var{command} runs, it is
223 translated as described in @ref{Exception Handling,,Exception Handling}.
225 The @var{from_tty} flag specifies whether @value{GDBN} ought to consider this
226 command as having originated from the user invoking it interactively.
227 It must be a boolean value. If omitted, it defaults to @code{False}.
229 By default, any output produced by @var{command} is sent to
230 @value{GDBN}'s standard output (and to the log output if logging is
231 turned on). If the @var{to_string} parameter is
232 @code{True}, then output will be collected by @code{gdb.execute} and
233 returned as a string. The default is @code{False}, in which case the
234 return value is @code{None}. If @var{to_string} is @code{True}, the
235 @value{GDBN} virtual terminal will be temporarily set to unlimited width
236 and height, and its pagination will be disabled; @pxref{Screen Size}.
239 @findex gdb.breakpoints
240 @defun gdb.breakpoints ()
241 Return a sequence holding all of @value{GDBN}'s breakpoints.
242 @xref{Breakpoints In Python}, for more information. In @value{GDBN}
243 version 7.11 and earlier, this function returned @code{None} if there
244 were no breakpoints. This peculiarity was subsequently fixed, and now
245 @code{gdb.breakpoints} returns an empty sequence in this case.
248 @defun gdb.rbreak (regex @r{[}, minsyms @r{[}, throttle, @r{[}, symtabs @r{]]]})
249 Return a Python list holding a collection of newly set
250 @code{gdb.Breakpoint} objects matching function names defined by the
251 @var{regex} pattern. If the @var{minsyms} keyword is @code{True}, all
252 system functions (those not explicitly defined in the inferior) will
253 also be included in the match. The @var{throttle} keyword takes an
254 integer that defines the maximum number of pattern matches for
255 functions matched by the @var{regex} pattern. If the number of
256 matches exceeds the integer value of @var{throttle}, a
257 @code{RuntimeError} will be raised and no breakpoints will be created.
258 If @var{throttle} is not defined then there is no imposed limit on the
259 maximum number of matches and breakpoints to be created. The
260 @var{symtabs} keyword takes a Python iterable that yields a collection
261 of @code{gdb.Symtab} objects and will restrict the search to those
262 functions only contained within the @code{gdb.Symtab} objects.
265 @findex gdb.parameter
266 @defun gdb.parameter (parameter)
267 Return the value of a @value{GDBN} @var{parameter} given by its name,
268 a string; the parameter name string may contain spaces if the parameter has a
269 multi-part name. For example, @samp{print object} is a valid
272 If the named parameter does not exist, this function throws a
273 @code{gdb.error} (@pxref{Exception Handling}). Otherwise, the
274 parameter's value is converted to a Python value of the appropriate
279 @defun gdb.history (number)
280 Return a value from @value{GDBN}'s value history (@pxref{Value
281 History}). The @var{number} argument indicates which history element to return.
282 If @var{number} is negative, then @value{GDBN} will take its absolute value
283 and count backward from the last element (i.e., the most recent element) to
284 find the value to return. If @var{number} is zero, then @value{GDBN} will
285 return the most recent element. If the element specified by @var{number}
286 doesn't exist in the value history, a @code{gdb.error} exception will be
289 If no exception is raised, the return value is always an instance of
290 @code{gdb.Value} (@pxref{Values From Inferior}).
293 @findex gdb.convenience_variable
294 @defun gdb.convenience_variable (name)
295 Return the value of the convenience variable (@pxref{Convenience
296 Vars}) named @var{name}. @var{name} must be a string. The name
297 should not include the @samp{$} that is used to mark a convenience
298 variable in an expression. If the convenience variable does not
299 exist, then @code{None} is returned.
302 @findex gdb.set_convenience_variable
303 @defun gdb.set_convenience_variable (name, value)
304 Set the value of the convenience variable (@pxref{Convenience Vars})
305 named @var{name}. @var{name} must be a string. The name should not
306 include the @samp{$} that is used to mark a convenience variable in an
307 expression. If @var{value} is @code{None}, then the convenience
308 variable is removed. Otherwise, if @var{value} is not a
309 @code{gdb.Value} (@pxref{Values From Inferior}), it is is converted
310 using the @code{gdb.Value} constructor.
313 @findex gdb.parse_and_eval
314 @defun gdb.parse_and_eval (expression)
315 Parse @var{expression}, which must be a string, as an expression in
316 the current language, evaluate it, and return the result as a
319 This function can be useful when implementing a new command
320 (@pxref{Commands In Python}), as it provides a way to parse the
321 command's argument as an expression. It is also useful simply to
325 @findex gdb.find_pc_line
326 @defun gdb.find_pc_line (pc)
327 Return the @code{gdb.Symtab_and_line} object corresponding to the
328 @var{pc} value. @xref{Symbol Tables In Python}. If an invalid
329 value of @var{pc} is passed as an argument, then the @code{symtab} and
330 @code{line} attributes of the returned @code{gdb.Symtab_and_line} object
331 will be @code{None} and 0 respectively.
334 @findex gdb.post_event
335 @defun gdb.post_event (event)
336 Put @var{event}, a callable object taking no arguments, into
337 @value{GDBN}'s internal event queue. This callable will be invoked at
338 some later point, during @value{GDBN}'s event processing. Events
339 posted using @code{post_event} will be run in the order in which they
340 were posted; however, there is no way to know when they will be
341 processed relative to other events inside @value{GDBN}.
343 @value{GDBN} is not thread-safe. If your Python program uses multiple
344 threads, you must be careful to only call @value{GDBN}-specific
345 functions in the @value{GDBN} thread. @code{post_event} ensures
349 (@value{GDBP}) python
353 > def __init__(self, message):
354 > self.message = message;
355 > def __call__(self):
356 > gdb.write(self.message)
358 >class MyThread1 (threading.Thread):
360 > gdb.post_event(Writer("Hello "))
362 >class MyThread2 (threading.Thread):
364 > gdb.post_event(Writer("World\n"))
369 (@value{GDBP}) Hello World
374 @defun gdb.write (string @r{[}, stream{]})
375 Print a string to @value{GDBN}'s paginated output stream. The
376 optional @var{stream} determines the stream to print to. The default
377 stream is @value{GDBN}'s standard output stream. Possible stream
384 @value{GDBN}'s standard output stream.
389 @value{GDBN}'s standard error stream.
394 @value{GDBN}'s log stream (@pxref{Logging Output}).
397 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
398 call this function and will automatically direct the output to the
404 Flush the buffer of a @value{GDBN} paginated stream so that the
405 contents are displayed immediately. @value{GDBN} will flush the
406 contents of a stream automatically when it encounters a newline in the
407 buffer. The optional @var{stream} determines the stream to flush. The
408 default stream is @value{GDBN}'s standard output stream. Possible
415 @value{GDBN}'s standard output stream.
420 @value{GDBN}'s standard error stream.
425 @value{GDBN}'s log stream (@pxref{Logging Output}).
429 Flushing @code{sys.stdout} or @code{sys.stderr} will automatically
430 call this function for the relevant stream.
433 @findex gdb.target_charset
434 @defun gdb.target_charset ()
435 Return the name of the current target character set (@pxref{Character
436 Sets}). This differs from @code{gdb.parameter('target-charset')} in
437 that @samp{auto} is never returned.
440 @findex gdb.target_wide_charset
441 @defun gdb.target_wide_charset ()
442 Return the name of the current target wide character set
443 (@pxref{Character Sets}). This differs from
444 @code{gdb.parameter('target-wide-charset')} in that @samp{auto} is
448 @findex gdb.solib_name
449 @defun gdb.solib_name (address)
450 Return the name of the shared library holding the given @var{address}
451 as a string, or @code{None}.
454 @findex gdb.decode_line
455 @defun gdb.decode_line (@r{[}expression@r{]})
456 Return locations of the line specified by @var{expression}, or of the
457 current line if no argument was given. This function returns a Python
458 tuple containing two elements. The first element contains a string
459 holding any unparsed section of @var{expression} (or @code{None} if
460 the expression has been fully parsed). The second element contains
461 either @code{None} or another tuple that contains all the locations
462 that match the expression represented as @code{gdb.Symtab_and_line}
463 objects (@pxref{Symbol Tables In Python}). If @var{expression} is
464 provided, it is decoded the way that @value{GDBN}'s inbuilt
465 @code{break} or @code{edit} commands do (@pxref{Specify Location}).
468 @defun gdb.prompt_hook (current_prompt)
471 If @var{prompt_hook} is callable, @value{GDBN} will call the method
472 assigned to this operation before a prompt is displayed by
475 The parameter @code{current_prompt} contains the current @value{GDBN}
476 prompt. This method must return a Python string, or @code{None}. If
477 a string is returned, the @value{GDBN} prompt will be set to that
478 string. If @code{None} is returned, @value{GDBN} will continue to use
481 Some prompts cannot be substituted in @value{GDBN}. Secondary prompts
482 such as those used by readline for command input, and annotation
483 related prompts are prohibited from being changed.
486 @node Exception Handling
487 @subsubsection Exception Handling
488 @cindex python exceptions
489 @cindex exceptions, python
491 When executing the @code{python} command, Python exceptions
492 uncaught within the Python code are translated to calls to
493 @value{GDBN} error-reporting mechanism. If the command that called
494 @code{python} does not handle the error, @value{GDBN} will
495 terminate it and print an error message containing the Python
496 exception name, the associated value, and the Python call stack
497 backtrace at the point where the exception was raised. Example:
500 (@value{GDBP}) python print foo
501 Traceback (most recent call last):
502 File "<string>", line 1, in <module>
503 NameError: name 'foo' is not defined
506 @value{GDBN} errors that happen in @value{GDBN} commands invoked by
507 Python code are converted to Python exceptions. The type of the
508 Python exception depends on the error.
512 This is the base class for most exceptions generated by @value{GDBN}.
513 It is derived from @code{RuntimeError}, for compatibility with earlier
514 versions of @value{GDBN}.
516 If an error occurring in @value{GDBN} does not fit into some more
517 specific category, then the generated exception will have this type.
519 @item gdb.MemoryError
520 This is a subclass of @code{gdb.error} which is thrown when an
521 operation tried to access invalid memory in the inferior.
523 @item KeyboardInterrupt
524 User interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
525 prompt) is translated to a Python @code{KeyboardInterrupt} exception.
528 In all cases, your exception handler will see the @value{GDBN} error
529 message as its value and the Python call stack backtrace at the Python
530 statement closest to where the @value{GDBN} error occured as the
534 When implementing @value{GDBN} commands in Python via
535 @code{gdb.Command}, or functions via @code{gdb.Function}, it is useful
536 to be able to throw an exception that doesn't cause a traceback to be
537 printed. For example, the user may have invoked the command
538 incorrectly. @value{GDBN} provides a special exception class that can
539 be used for this purpose.
543 When thrown from a command or function, this exception will cause the
544 command or function to fail, but the Python stack will not be
545 displayed. @value{GDBN} does not throw this exception itself, but
546 rather recognizes it when thrown from user Python code. Example:
550 >class HelloWorld (gdb.Command):
551 > """Greet the whole world."""
552 > def __init__ (self):
553 > super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_USER)
554 > def invoke (self, args, from_tty):
555 > argv = gdb.string_to_argv (args)
556 > if len (argv) != 0:
557 > raise gdb.GdbError ("hello-world takes no arguments")
558 > print "Hello, World!"
562 hello-world takes no arguments
566 @node Values From Inferior
567 @subsubsection Values From Inferior
568 @cindex values from inferior, with Python
569 @cindex python, working with values from inferior
571 @cindex @code{gdb.Value}
572 @value{GDBN} provides values it obtains from the inferior program in
573 an object of type @code{gdb.Value}. @value{GDBN} uses this object
574 for its internal bookkeeping of the inferior's values, and for
575 fetching values when necessary.
577 Inferior values that are simple scalars can be used directly in
578 Python expressions that are valid for the value's data type. Here's
579 an example for an integer or floating-point value @code{some_val}:
586 As result of this, @code{bar} will also be a @code{gdb.Value} object
587 whose values are of the same type as those of @code{some_val}. Valid
588 Python operations can also be performed on @code{gdb.Value} objects
589 representing a @code{struct} or @code{class} object. For such cases,
590 the overloaded operator (if present), is used to perform the operation.
591 For example, if @code{val1} and @code{val2} are @code{gdb.Value} objects
592 representing instances of a @code{class} which overloads the @code{+}
593 operator, then one can use the @code{+} operator in their Python script
601 The result of the operation @code{val3} is also a @code{gdb.Value}
602 object corresponding to the value returned by the overloaded @code{+}
603 operator. In general, overloaded operators are invoked for the
604 following operations: @code{+} (binary addition), @code{-} (binary
605 subtraction), @code{*} (multiplication), @code{/}, @code{%}, @code{<<},
606 @code{>>}, @code{|}, @code{&}, @code{^}.
608 Inferior values that are structures or instances of some class can
609 be accessed using the Python @dfn{dictionary syntax}. For example, if
610 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
611 can access its @code{foo} element with:
614 bar = some_val['foo']
617 @cindex getting structure elements using gdb.Field objects as subscripts
618 Again, @code{bar} will also be a @code{gdb.Value} object. Structure
619 elements can also be accessed by using @code{gdb.Field} objects as
620 subscripts (@pxref{Types In Python}, for more information on
621 @code{gdb.Field} objects). For example, if @code{foo_field} is a
622 @code{gdb.Field} object corresponding to element @code{foo} of the above
623 structure, then @code{bar} can also be accessed as follows:
626 bar = some_val[foo_field]
629 A @code{gdb.Value} that represents a function can be executed via
630 inferior function call. Any arguments provided to the call must match
631 the function's prototype, and must be provided in the order specified
634 For example, @code{some_val} is a @code{gdb.Value} instance
635 representing a function that takes two integers as arguments. To
636 execute this function, call it like so:
639 result = some_val (10,20)
642 Any values returned from a function call will be stored as a
645 The following attributes are provided:
647 @defvar Value.address
648 If this object is addressable, this read-only attribute holds a
649 @code{gdb.Value} object representing the address. Otherwise,
650 this attribute holds @code{None}.
653 @cindex optimized out value in Python
654 @defvar Value.is_optimized_out
655 This read-only boolean attribute is true if the compiler optimized out
656 this value, thus it is not available for fetching from the inferior.
660 The type of this @code{gdb.Value}. The value of this attribute is a
661 @code{gdb.Type} object (@pxref{Types In Python}).
664 @defvar Value.dynamic_type
665 The dynamic type of this @code{gdb.Value}. This uses the object's
666 virtual table and the C@t{++} run-time type information
667 (@acronym{RTTI}) to determine the dynamic type of the value. If this
668 value is of class type, it will return the class in which the value is
669 embedded, if any. If this value is of pointer or reference to a class
670 type, it will compute the dynamic type of the referenced object, and
671 return a pointer or reference to that type, respectively. In all
672 other cases, it will return the value's static type.
674 Note that this feature will only work when debugging a C@t{++} program
675 that includes @acronym{RTTI} for the object in question. Otherwise,
676 it will just return the static type of the value as in @kbd{ptype foo}
677 (@pxref{Symbols, ptype}).
680 @defvar Value.is_lazy
681 The value of this read-only boolean attribute is @code{True} if this
682 @code{gdb.Value} has not yet been fetched from the inferior.
683 @value{GDBN} does not fetch values until necessary, for efficiency.
687 myval = gdb.parse_and_eval ('somevar')
690 The value of @code{somevar} is not fetched at this time. It will be
691 fetched when the value is needed, or when the @code{fetch_lazy}
695 The following methods are provided:
697 @defun Value.__init__ (@var{val})
698 Many Python values can be converted directly to a @code{gdb.Value} via
699 this object initializer. Specifically:
703 A Python boolean is converted to the boolean type from the current
707 A Python integer is converted to the C @code{long} type for the
708 current architecture.
711 A Python long is converted to the C @code{long long} type for the
712 current architecture.
715 A Python float is converted to the C @code{double} type for the
716 current architecture.
719 A Python string is converted to a target string in the current target
720 language using the current target encoding.
721 If a character cannot be represented in the current target encoding,
722 then an exception is thrown.
724 @item @code{gdb.Value}
725 If @code{val} is a @code{gdb.Value}, then a copy of the value is made.
727 @item @code{gdb.LazyString}
728 If @code{val} is a @code{gdb.LazyString} (@pxref{Lazy Strings In
729 Python}), then the lazy string's @code{value} method is called, and
734 @defun Value.cast (type)
735 Return a new instance of @code{gdb.Value} that is the result of
736 casting this instance to the type described by @var{type}, which must
737 be a @code{gdb.Type} object. If the cast cannot be performed for some
738 reason, this method throws an exception.
741 @defun Value.dereference ()
742 For pointer data types, this method returns a new @code{gdb.Value} object
743 whose contents is the object pointed to by the pointer. For example, if
744 @code{foo} is a C pointer to an @code{int}, declared in your C program as
751 then you can use the corresponding @code{gdb.Value} to access what
752 @code{foo} points to like this:
755 bar = foo.dereference ()
758 The result @code{bar} will be a @code{gdb.Value} object holding the
759 value pointed to by @code{foo}.
761 A similar function @code{Value.referenced_value} exists which also
762 returns @code{gdb.Value} objects corresonding to the values pointed to
763 by pointer values (and additionally, values referenced by reference
764 values). However, the behavior of @code{Value.dereference}
765 differs from @code{Value.referenced_value} by the fact that the
766 behavior of @code{Value.dereference} is identical to applying the C
767 unary operator @code{*} on a given value. For example, consider a
768 reference to a pointer @code{ptrref}, declared in your C@t{++} program
776 intptr &ptrref = ptr;
779 Though @code{ptrref} is a reference value, one can apply the method
780 @code{Value.dereference} to the @code{gdb.Value} object corresponding
781 to it and obtain a @code{gdb.Value} which is identical to that
782 corresponding to @code{val}. However, if you apply the method
783 @code{Value.referenced_value}, the result would be a @code{gdb.Value}
784 object identical to that corresponding to @code{ptr}.
787 py_ptrref = gdb.parse_and_eval ("ptrref")
788 py_val = py_ptrref.dereference ()
789 py_ptr = py_ptrref.referenced_value ()
792 The @code{gdb.Value} object @code{py_val} is identical to that
793 corresponding to @code{val}, and @code{py_ptr} is identical to that
794 corresponding to @code{ptr}. In general, @code{Value.dereference} can
795 be applied whenever the C unary operator @code{*} can be applied
796 to the corresponding C value. For those cases where applying both
797 @code{Value.dereference} and @code{Value.referenced_value} is allowed,
798 the results obtained need not be identical (as we have seen in the above
799 example). The results are however identical when applied on
800 @code{gdb.Value} objects corresponding to pointers (@code{gdb.Value}
801 objects with type code @code{TYPE_CODE_PTR}) in a C/C@t{++} program.
804 @defun Value.referenced_value ()
805 For pointer or reference data types, this method returns a new
806 @code{gdb.Value} object corresponding to the value referenced by the
807 pointer/reference value. For pointer data types,
808 @code{Value.dereference} and @code{Value.referenced_value} produce
809 identical results. The difference between these methods is that
810 @code{Value.dereference} cannot get the values referenced by reference
811 values. For example, consider a reference to an @code{int}, declared
812 in your C@t{++} program as
820 then applying @code{Value.dereference} to the @code{gdb.Value} object
821 corresponding to @code{ref} will result in an error, while applying
822 @code{Value.referenced_value} will result in a @code{gdb.Value} object
823 identical to that corresponding to @code{val}.
826 py_ref = gdb.parse_and_eval ("ref")
827 er_ref = py_ref.dereference () # Results in error
828 py_val = py_ref.referenced_value () # Returns the referenced value
831 The @code{gdb.Value} object @code{py_val} is identical to that
832 corresponding to @code{val}.
835 @defun Value.reference_value ()
836 Return a @code{gdb.Value} object which is a reference to the value
837 encapsulated by this instance.
840 @defun Value.const_value ()
841 Return a @code{gdb.Value} object which is a @code{const} version of the
842 value encapsulated by this instance.
845 @defun Value.dynamic_cast (type)
846 Like @code{Value.cast}, but works as if the C@t{++} @code{dynamic_cast}
847 operator were used. Consult a C@t{++} reference for details.
850 @defun Value.reinterpret_cast (type)
851 Like @code{Value.cast}, but works as if the C@t{++} @code{reinterpret_cast}
852 operator were used. Consult a C@t{++} reference for details.
855 @defun Value.string (@r{[}encoding@r{[}, errors@r{[}, length@r{]]]})
856 If this @code{gdb.Value} represents a string, then this method
857 converts the contents to a Python string. Otherwise, this method will
860 Values are interpreted as strings according to the rules of the
861 current language. If the optional length argument is given, the
862 string will be converted to that length, and will include any embedded
863 zeroes that the string may contain. Otherwise, for languages
864 where the string is zero-terminated, the entire string will be
867 For example, in C-like languages, a value is a string if it is a pointer
868 to or an array of characters or ints of type @code{wchar_t}, @code{char16_t},
871 If the optional @var{encoding} argument is given, it must be a string
872 naming the encoding of the string in the @code{gdb.Value}, such as
873 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
874 the same encodings as the corresponding argument to Python's
875 @code{string.decode} method, and the Python codec machinery will be used
876 to convert the string. If @var{encoding} is not given, or if
877 @var{encoding} is the empty string, then either the @code{target-charset}
878 (@pxref{Character Sets}) will be used, or a language-specific encoding
879 will be used, if the current language is able to supply one.
881 The optional @var{errors} argument is the same as the corresponding
882 argument to Python's @code{string.decode} method.
884 If the optional @var{length} argument is given, the string will be
885 fetched and converted to the given length.
888 @defun Value.lazy_string (@r{[}encoding @r{[}, length@r{]]})
889 If this @code{gdb.Value} represents a string, then this method
890 converts the contents to a @code{gdb.LazyString} (@pxref{Lazy Strings
891 In Python}). Otherwise, this method will throw an exception.
893 If the optional @var{encoding} argument is given, it must be a string
894 naming the encoding of the @code{gdb.LazyString}. Some examples are:
895 @samp{ascii}, @samp{iso-8859-6} or @samp{utf-8}. If the
896 @var{encoding} argument is an encoding that @value{GDBN} does
897 recognize, @value{GDBN} will raise an error.
899 When a lazy string is printed, the @value{GDBN} encoding machinery is
900 used to convert the string during printing. If the optional
901 @var{encoding} argument is not provided, or is an empty string,
902 @value{GDBN} will automatically select the encoding most suitable for
903 the string type. For further information on encoding in @value{GDBN}
904 please see @ref{Character Sets}.
906 If the optional @var{length} argument is given, the string will be
907 fetched and encoded to the length of characters specified. If
908 the @var{length} argument is not provided, the string will be fetched
909 and encoded until a null of appropriate width is found.
912 @defun Value.fetch_lazy ()
913 If the @code{gdb.Value} object is currently a lazy value
914 (@code{gdb.Value.is_lazy} is @code{True}), then the value is
915 fetched from the inferior. Any errors that occur in the process
916 will produce a Python exception.
918 If the @code{gdb.Value} object is not a lazy value, this method
921 This method does not return a value.
925 @node Types In Python
926 @subsubsection Types In Python
927 @cindex types in Python
928 @cindex Python, working with types
931 @value{GDBN} represents types from the inferior using the class
934 The following type-related functions are available in the @code{gdb}
937 @findex gdb.lookup_type
938 @defun gdb.lookup_type (name @r{[}, block@r{]})
939 This function looks up a type by its @var{name}, which must be a string.
941 If @var{block} is given, then @var{name} is looked up in that scope.
942 Otherwise, it is searched for globally.
944 Ordinarily, this function will return an instance of @code{gdb.Type}.
945 If the named type cannot be found, it will throw an exception.
948 If the type is a structure or class type, or an enum type, the fields
949 of that type can be accessed using the Python @dfn{dictionary syntax}.
950 For example, if @code{some_type} is a @code{gdb.Type} instance holding
951 a structure type, you can access its @code{foo} field with:
954 bar = some_type['foo']
957 @code{bar} will be a @code{gdb.Field} object; see below under the
958 description of the @code{Type.fields} method for a description of the
959 @code{gdb.Field} class.
961 An instance of @code{Type} has the following attributes:
964 The alignment of this type, in bytes. Type alignment comes from the
965 debugging information; if it was not specified, then @value{GDBN} will
966 use the relevant ABI to try to determine the alignment. In some
967 cases, even this is not possible, and zero will be returned.
971 The type code for this type. The type code will be one of the
972 @code{TYPE_CODE_} constants defined below.
976 The name of this type. If this type has no name, then @code{None}
981 The size of this type, in target @code{char} units. Usually, a
982 target's @code{char} type will be an 8-bit byte. However, on some
983 unusual platforms, this type may have a different size.
987 The tag name for this type. The tag name is the name after
988 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
989 languages have this concept. If this type has no tag name, then
990 @code{None} is returned.
993 The following methods are provided:
995 @defun Type.fields ()
996 For structure and union types, this method returns the fields. Range
997 types have two fields, the minimum and maximum values. Enum types
998 have one field per enum constant. Function and method types have one
999 field per parameter. The base types of C@t{++} classes are also
1000 represented as fields. If the type has no fields, or does not fit
1001 into one of these categories, an empty sequence will be returned.
1003 Each field is a @code{gdb.Field} object, with some pre-defined attributes:
1006 This attribute is not available for @code{enum} or @code{static}
1007 (as in C@t{++}) fields. The value is the position, counting
1008 in bits, from the start of the containing type.
1011 This attribute is only available for @code{enum} fields, and its value
1012 is the enumeration member's integer representation.
1015 The name of the field, or @code{None} for anonymous fields.
1018 This is @code{True} if the field is artificial, usually meaning that
1019 it was provided by the compiler and not the user. This attribute is
1020 always provided, and is @code{False} if the field is not artificial.
1023 This is @code{True} if the field represents a base class of a C@t{++}
1024 structure. This attribute is always provided, and is @code{False}
1025 if the field is not a base class of the type that is the argument of
1026 @code{fields}, or if that type was not a C@t{++} class.
1029 If the field is packed, or is a bitfield, then this will have a
1030 non-zero value, which is the size of the field in bits. Otherwise,
1031 this will be zero; in this case the field's size is given by its type.
1034 The type of the field. This is usually an instance of @code{Type},
1035 but it can be @code{None} in some situations.
1038 The type which contains this field. This is an instance of
1043 @defun Type.array (@var{n1} @r{[}, @var{n2}@r{]})
1044 Return a new @code{gdb.Type} object which represents an array of this
1045 type. If one argument is given, it is the inclusive upper bound of
1046 the array; in this case the lower bound is zero. If two arguments are
1047 given, the first argument is the lower bound of the array, and the
1048 second argument is the upper bound of the array. An array's length
1049 must not be negative, but the bounds can be.
1052 @defun Type.vector (@var{n1} @r{[}, @var{n2}@r{]})
1053 Return a new @code{gdb.Type} object which represents a vector of this
1054 type. If one argument is given, it is the inclusive upper bound of
1055 the vector; in this case the lower bound is zero. If two arguments are
1056 given, the first argument is the lower bound of the vector, and the
1057 second argument is the upper bound of the vector. A vector's length
1058 must not be negative, but the bounds can be.
1060 The difference between an @code{array} and a @code{vector} is that
1061 arrays behave like in C: when used in expressions they decay to a pointer
1062 to the first element whereas vectors are treated as first class values.
1065 @defun Type.const ()
1066 Return a new @code{gdb.Type} object which represents a
1067 @code{const}-qualified variant of this type.
1070 @defun Type.volatile ()
1071 Return a new @code{gdb.Type} object which represents a
1072 @code{volatile}-qualified variant of this type.
1075 @defun Type.unqualified ()
1076 Return a new @code{gdb.Type} object which represents an unqualified
1077 variant of this type. That is, the result is neither @code{const} nor
1081 @defun Type.range ()
1082 Return a Python @code{Tuple} object that contains two elements: the
1083 low bound of the argument type and the high bound of that type. If
1084 the type does not have a range, @value{GDBN} will raise a
1085 @code{gdb.error} exception (@pxref{Exception Handling}).
1088 @defun Type.reference ()
1089 Return a new @code{gdb.Type} object which represents a reference to this
1093 @defun Type.pointer ()
1094 Return a new @code{gdb.Type} object which represents a pointer to this
1098 @defun Type.strip_typedefs ()
1099 Return a new @code{gdb.Type} that represents the real type,
1100 after removing all layers of typedefs.
1103 @defun Type.target ()
1104 Return a new @code{gdb.Type} object which represents the target type
1107 For a pointer type, the target type is the type of the pointed-to
1108 object. For an array type (meaning C-like arrays), the target type is
1109 the type of the elements of the array. For a function or method type,
1110 the target type is the type of the return value. For a complex type,
1111 the target type is the type of the elements. For a typedef, the
1112 target type is the aliased type.
1114 If the type does not have a target, this method will throw an
1118 @defun Type.template_argument (n @r{[}, block@r{]})
1119 If this @code{gdb.Type} is an instantiation of a template, this will
1120 return a new @code{gdb.Value} or @code{gdb.Type} which represents the
1121 value of the @var{n}th template argument (indexed starting at 0).
1123 If this @code{gdb.Type} is not a template type, or if the type has fewer
1124 than @var{n} template arguments, this will throw an exception.
1125 Ordinarily, only C@t{++} code will have template types.
1127 If @var{block} is given, then @var{name} is looked up in that scope.
1128 Otherwise, it is searched for globally.
1131 @defun Type.optimized_out ()
1132 Return @code{gdb.Value} instance of this type whose value is optimized
1133 out. This allows a frame decorator to indicate that the value of an
1134 argument or a local variable is not known.
1137 Each type has a code, which indicates what category this type falls
1138 into. The available type categories are represented by constants
1139 defined in the @code{gdb} module:
1142 @vindex TYPE_CODE_PTR
1143 @item gdb.TYPE_CODE_PTR
1144 The type is a pointer.
1146 @vindex TYPE_CODE_ARRAY
1147 @item gdb.TYPE_CODE_ARRAY
1148 The type is an array.
1150 @vindex TYPE_CODE_STRUCT
1151 @item gdb.TYPE_CODE_STRUCT
1152 The type is a structure.
1154 @vindex TYPE_CODE_UNION
1155 @item gdb.TYPE_CODE_UNION
1156 The type is a union.
1158 @vindex TYPE_CODE_ENUM
1159 @item gdb.TYPE_CODE_ENUM
1160 The type is an enum.
1162 @vindex TYPE_CODE_FLAGS
1163 @item gdb.TYPE_CODE_FLAGS
1164 A bit flags type, used for things such as status registers.
1166 @vindex TYPE_CODE_FUNC
1167 @item gdb.TYPE_CODE_FUNC
1168 The type is a function.
1170 @vindex TYPE_CODE_INT
1171 @item gdb.TYPE_CODE_INT
1172 The type is an integer type.
1174 @vindex TYPE_CODE_FLT
1175 @item gdb.TYPE_CODE_FLT
1176 A floating point type.
1178 @vindex TYPE_CODE_VOID
1179 @item gdb.TYPE_CODE_VOID
1180 The special type @code{void}.
1182 @vindex TYPE_CODE_SET
1183 @item gdb.TYPE_CODE_SET
1186 @vindex TYPE_CODE_RANGE
1187 @item gdb.TYPE_CODE_RANGE
1188 A range type, that is, an integer type with bounds.
1190 @vindex TYPE_CODE_STRING
1191 @item gdb.TYPE_CODE_STRING
1192 A string type. Note that this is only used for certain languages with
1193 language-defined string types; C strings are not represented this way.
1195 @vindex TYPE_CODE_BITSTRING
1196 @item gdb.TYPE_CODE_BITSTRING
1197 A string of bits. It is deprecated.
1199 @vindex TYPE_CODE_ERROR
1200 @item gdb.TYPE_CODE_ERROR
1201 An unknown or erroneous type.
1203 @vindex TYPE_CODE_METHOD
1204 @item gdb.TYPE_CODE_METHOD
1205 A method type, as found in C@t{++}.
1207 @vindex TYPE_CODE_METHODPTR
1208 @item gdb.TYPE_CODE_METHODPTR
1209 A pointer-to-member-function.
1211 @vindex TYPE_CODE_MEMBERPTR
1212 @item gdb.TYPE_CODE_MEMBERPTR
1213 A pointer-to-member.
1215 @vindex TYPE_CODE_REF
1216 @item gdb.TYPE_CODE_REF
1219 @vindex TYPE_CODE_RVALUE_REF
1220 @item gdb.TYPE_CODE_RVALUE_REF
1221 A C@t{++}11 rvalue reference type.
1223 @vindex TYPE_CODE_CHAR
1224 @item gdb.TYPE_CODE_CHAR
1227 @vindex TYPE_CODE_BOOL
1228 @item gdb.TYPE_CODE_BOOL
1231 @vindex TYPE_CODE_COMPLEX
1232 @item gdb.TYPE_CODE_COMPLEX
1233 A complex float type.
1235 @vindex TYPE_CODE_TYPEDEF
1236 @item gdb.TYPE_CODE_TYPEDEF
1237 A typedef to some other type.
1239 @vindex TYPE_CODE_NAMESPACE
1240 @item gdb.TYPE_CODE_NAMESPACE
1241 A C@t{++} namespace.
1243 @vindex TYPE_CODE_DECFLOAT
1244 @item gdb.TYPE_CODE_DECFLOAT
1245 A decimal floating point type.
1247 @vindex TYPE_CODE_INTERNAL_FUNCTION
1248 @item gdb.TYPE_CODE_INTERNAL_FUNCTION
1249 A function internal to @value{GDBN}. This is the type used to represent
1250 convenience functions.
1253 Further support for types is provided in the @code{gdb.types}
1254 Python module (@pxref{gdb.types}).
1256 @node Pretty Printing API
1257 @subsubsection Pretty Printing API
1258 @cindex python pretty printing api
1260 A pretty-printer is just an object that holds a value and implements a
1261 specific interface, defined here. An example output is provided
1262 (@pxref{Pretty Printing}).
1264 @defun pretty_printer.children (self)
1265 @value{GDBN} will call this method on a pretty-printer to compute the
1266 children of the pretty-printer's value.
1268 This method must return an object conforming to the Python iterator
1269 protocol. Each item returned by the iterator must be a tuple holding
1270 two elements. The first element is the ``name'' of the child; the
1271 second element is the child's value. The value can be any Python
1272 object which is convertible to a @value{GDBN} value.
1274 This method is optional. If it does not exist, @value{GDBN} will act
1275 as though the value has no children.
1278 @defun pretty_printer.display_hint (self)
1279 The CLI may call this method and use its result to change the
1280 formatting of a value. The result will also be supplied to an MI
1281 consumer as a @samp{displayhint} attribute of the variable being
1284 This method is optional. If it does exist, this method must return a
1287 Some display hints are predefined by @value{GDBN}:
1291 Indicate that the object being printed is ``array-like''. The CLI
1292 uses this to respect parameters such as @code{set print elements} and
1293 @code{set print array}.
1296 Indicate that the object being printed is ``map-like'', and that the
1297 children of this value can be assumed to alternate between keys and
1301 Indicate that the object being printed is ``string-like''. If the
1302 printer's @code{to_string} method returns a Python string of some
1303 kind, then @value{GDBN} will call its internal language-specific
1304 string-printing function to format the string. For the CLI this means
1305 adding quotation marks, possibly escaping some characters, respecting
1306 @code{set print elements}, and the like.
1310 @defun pretty_printer.to_string (self)
1311 @value{GDBN} will call this method to display the string
1312 representation of the value passed to the object's constructor.
1314 When printing from the CLI, if the @code{to_string} method exists,
1315 then @value{GDBN} will prepend its result to the values returned by
1316 @code{children}. Exactly how this formatting is done is dependent on
1317 the display hint, and may change as more hints are added. Also,
1318 depending on the print settings (@pxref{Print Settings}), the CLI may
1319 print just the result of @code{to_string} in a stack trace, omitting
1320 the result of @code{children}.
1322 If this method returns a string, it is printed verbatim.
1324 Otherwise, if this method returns an instance of @code{gdb.Value},
1325 then @value{GDBN} prints this value. This may result in a call to
1326 another pretty-printer.
1328 If instead the method returns a Python value which is convertible to a
1329 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
1330 the resulting value. Again, this may result in a call to another
1331 pretty-printer. Python scalars (integers, floats, and booleans) and
1332 strings are convertible to @code{gdb.Value}; other types are not.
1334 Finally, if this method returns @code{None} then no further operations
1335 are peformed in this method and nothing is printed.
1337 If the result is not one of these types, an exception is raised.
1340 @value{GDBN} provides a function which can be used to look up the
1341 default pretty-printer for a @code{gdb.Value}:
1343 @findex gdb.default_visualizer
1344 @defun gdb.default_visualizer (value)
1345 This function takes a @code{gdb.Value} object as an argument. If a
1346 pretty-printer for this value exists, then it is returned. If no such
1347 printer exists, then this returns @code{None}.
1350 @node Selecting Pretty-Printers
1351 @subsubsection Selecting Pretty-Printers
1352 @cindex selecting python pretty-printers
1354 The Python list @code{gdb.pretty_printers} contains an array of
1355 functions or callable objects that have been registered via addition
1356 as a pretty-printer. Printers in this list are called @code{global}
1357 printers, they're available when debugging all inferiors.
1358 Each @code{gdb.Progspace} contains a @code{pretty_printers} attribute.
1359 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
1362 Each function on these lists is passed a single @code{gdb.Value}
1363 argument and should return a pretty-printer object conforming to the
1364 interface definition above (@pxref{Pretty Printing API}). If a function
1365 cannot create a pretty-printer for the value, it should return
1368 @value{GDBN} first checks the @code{pretty_printers} attribute of each
1369 @code{gdb.Objfile} in the current program space and iteratively calls
1370 each enabled lookup routine in the list for that @code{gdb.Objfile}
1371 until it receives a pretty-printer object.
1372 If no pretty-printer is found in the objfile lists, @value{GDBN} then
1373 searches the pretty-printer list of the current program space,
1374 calling each enabled function until an object is returned.
1375 After these lists have been exhausted, it tries the global
1376 @code{gdb.pretty_printers} list, again calling each enabled function until an
1379 The order in which the objfiles are searched is not specified. For a
1380 given list, functions are always invoked from the head of the list,
1381 and iterated over sequentially until the end of the list, or a printer
1384 For various reasons a pretty-printer may not work.
1385 For example, the underlying data structure may have changed and
1386 the pretty-printer is out of date.
1388 The consequences of a broken pretty-printer are severe enough that
1389 @value{GDBN} provides support for enabling and disabling individual
1390 printers. For example, if @code{print frame-arguments} is on,
1391 a backtrace can become highly illegible if any argument is printed
1392 with a broken printer.
1394 Pretty-printers are enabled and disabled by attaching an @code{enabled}
1395 attribute to the registered function or callable object. If this attribute
1396 is present and its value is @code{False}, the printer is disabled, otherwise
1397 the printer is enabled.
1399 @node Writing a Pretty-Printer
1400 @subsubsection Writing a Pretty-Printer
1401 @cindex writing a pretty-printer
1403 A pretty-printer consists of two parts: a lookup function to detect
1404 if the type is supported, and the printer itself.
1406 Here is an example showing how a @code{std::string} printer might be
1407 written. @xref{Pretty Printing API}, for details on the API this class
1411 class StdStringPrinter(object):
1412 "Print a std::string"
1414 def __init__(self, val):
1417 def to_string(self):
1418 return self.val['_M_dataplus']['_M_p']
1420 def display_hint(self):
1424 And here is an example showing how a lookup function for the printer
1425 example above might be written.
1428 def str_lookup_function(val):
1429 lookup_tag = val.type.tag
1430 if lookup_tag == None:
1432 regex = re.compile("^std::basic_string<char,.*>$")
1433 if regex.match(lookup_tag):
1434 return StdStringPrinter(val)
1438 The example lookup function extracts the value's type, and attempts to
1439 match it to a type that it can pretty-print. If it is a type the
1440 printer can pretty-print, it will return a printer object. If not, it
1441 returns @code{None}.
1443 We recommend that you put your core pretty-printers into a Python
1444 package. If your pretty-printers are for use with a library, we
1445 further recommend embedding a version number into the package name.
1446 This practice will enable @value{GDBN} to load multiple versions of
1447 your pretty-printers at the same time, because they will have
1450 You should write auto-loaded code (@pxref{Python Auto-loading}) such that it
1451 can be evaluated multiple times without changing its meaning. An
1452 ideal auto-load file will consist solely of @code{import}s of your
1453 printer modules, followed by a call to a register pretty-printers with
1454 the current objfile.
1456 Taken as a whole, this approach will scale nicely to multiple
1457 inferiors, each potentially using a different library version.
1458 Embedding a version number in the Python package name will ensure that
1459 @value{GDBN} is able to load both sets of printers simultaneously.
1460 Then, because the search for pretty-printers is done by objfile, and
1461 because your auto-loaded code took care to register your library's
1462 printers with a specific objfile, @value{GDBN} will find the correct
1463 printers for the specific version of the library used by each
1466 To continue the @code{std::string} example (@pxref{Pretty Printing API}),
1467 this code might appear in @code{gdb.libstdcxx.v6}:
1470 def register_printers(objfile):
1471 objfile.pretty_printers.append(str_lookup_function)
1475 And then the corresponding contents of the auto-load file would be:
1478 import gdb.libstdcxx.v6
1479 gdb.libstdcxx.v6.register_printers(gdb.current_objfile())
1482 The previous example illustrates a basic pretty-printer.
1483 There are a few things that can be improved on.
1484 The printer doesn't have a name, making it hard to identify in a
1485 list of installed printers. The lookup function has a name, but
1486 lookup functions can have arbitrary, even identical, names.
1488 Second, the printer only handles one type, whereas a library typically has
1489 several types. One could install a lookup function for each desired type
1490 in the library, but one could also have a single lookup function recognize
1491 several types. The latter is the conventional way this is handled.
1492 If a pretty-printer can handle multiple data types, then its
1493 @dfn{subprinters} are the printers for the individual data types.
1495 The @code{gdb.printing} module provides a formal way of solving these
1496 problems (@pxref{gdb.printing}).
1497 Here is another example that handles multiple types.
1499 These are the types we are going to pretty-print:
1502 struct foo @{ int a, b; @};
1503 struct bar @{ struct foo x, y; @};
1506 Here are the printers:
1510 """Print a foo object."""
1512 def __init__(self, val):
1515 def to_string(self):
1516 return ("a=<" + str(self.val["a"]) +
1517 "> b=<" + str(self.val["b"]) + ">")
1520 """Print a bar object."""
1522 def __init__(self, val):
1525 def to_string(self):
1526 return ("x=<" + str(self.val["x"]) +
1527 "> y=<" + str(self.val["y"]) + ">")
1530 This example doesn't need a lookup function, that is handled by the
1531 @code{gdb.printing} module. Instead a function is provided to build up
1532 the object that handles the lookup.
1537 def build_pretty_printer():
1538 pp = gdb.printing.RegexpCollectionPrettyPrinter(
1540 pp.add_printer('foo', '^foo$', fooPrinter)
1541 pp.add_printer('bar', '^bar$', barPrinter)
1545 And here is the autoload support:
1550 gdb.printing.register_pretty_printer(
1551 gdb.current_objfile(),
1552 my_library.build_pretty_printer())
1555 Finally, when this printer is loaded into @value{GDBN}, here is the
1556 corresponding output of @samp{info pretty-printer}:
1559 (gdb) info pretty-printer
1566 @node Type Printing API
1567 @subsubsection Type Printing API
1568 @cindex type printing API for Python
1570 @value{GDBN} provides a way for Python code to customize type display.
1571 This is mainly useful for substituting canonical typedef names for
1574 @cindex type printer
1575 A @dfn{type printer} is just a Python object conforming to a certain
1576 protocol. A simple base class implementing the protocol is provided;
1577 see @ref{gdb.types}. A type printer must supply at least:
1579 @defivar type_printer enabled
1580 A boolean which is True if the printer is enabled, and False
1581 otherwise. This is manipulated by the @code{enable type-printer}
1582 and @code{disable type-printer} commands.
1585 @defivar type_printer name
1586 The name of the type printer. This must be a string. This is used by
1587 the @code{enable type-printer} and @code{disable type-printer}
1591 @defmethod type_printer instantiate (self)
1592 This is called by @value{GDBN} at the start of type-printing. It is
1593 only called if the type printer is enabled. This method must return a
1594 new object that supplies a @code{recognize} method, as described below.
1598 When displaying a type, say via the @code{ptype} command, @value{GDBN}
1599 will compute a list of type recognizers. This is done by iterating
1600 first over the per-objfile type printers (@pxref{Objfiles In Python}),
1601 followed by the per-progspace type printers (@pxref{Progspaces In
1602 Python}), and finally the global type printers.
1604 @value{GDBN} will call the @code{instantiate} method of each enabled
1605 type printer. If this method returns @code{None}, then the result is
1606 ignored; otherwise, it is appended to the list of recognizers.
1608 Then, when @value{GDBN} is going to display a type name, it iterates
1609 over the list of recognizers. For each one, it calls the recognition
1610 function, stopping if the function returns a non-@code{None} value.
1611 The recognition function is defined as:
1613 @defmethod type_recognizer recognize (self, type)
1614 If @var{type} is not recognized, return @code{None}. Otherwise,
1615 return a string which is to be printed as the name of @var{type}.
1616 The @var{type} argument will be an instance of @code{gdb.Type}
1617 (@pxref{Types In Python}).
1620 @value{GDBN} uses this two-pass approach so that type printers can
1621 efficiently cache information without holding on to it too long. For
1622 example, it can be convenient to look up type information in a type
1623 printer and hold it for a recognizer's lifetime; if a single pass were
1624 done then type printers would have to make use of the event system in
1625 order to avoid holding information that could become stale as the
1628 @node Frame Filter API
1629 @subsubsection Filtering Frames
1630 @cindex frame filters api
1632 Frame filters are Python objects that manipulate the visibility of a
1633 frame or frames when a backtrace (@pxref{Backtrace}) is printed by
1636 Only commands that print a backtrace, or, in the case of @sc{gdb/mi}
1637 commands (@pxref{GDB/MI}), those that return a collection of frames
1638 are affected. The commands that work with frame filters are:
1640 @code{backtrace} (@pxref{backtrace-command,, The backtrace command}),
1641 @code{-stack-list-frames}
1642 (@pxref{-stack-list-frames,, The -stack-list-frames command}),
1643 @code{-stack-list-variables} (@pxref{-stack-list-variables,, The
1644 -stack-list-variables command}), @code{-stack-list-arguments}
1645 @pxref{-stack-list-arguments,, The -stack-list-arguments command}) and
1646 @code{-stack-list-locals} (@pxref{-stack-list-locals,, The
1647 -stack-list-locals command}).
1649 A frame filter works by taking an iterator as an argument, applying
1650 actions to the contents of that iterator, and returning another
1651 iterator (or, possibly, the same iterator it was provided in the case
1652 where the filter does not perform any operations). Typically, frame
1653 filters utilize tools such as the Python's @code{itertools} module to
1654 work with and create new iterators from the source iterator.
1655 Regardless of how a filter chooses to apply actions, it must not alter
1656 the underlying @value{GDBN} frame or frames, or attempt to alter the
1657 call-stack within @value{GDBN}. This preserves data integrity within
1658 @value{GDBN}. Frame filters are executed on a priority basis and care
1659 should be taken that some frame filters may have been executed before,
1660 and that some frame filters will be executed after.
1662 An important consideration when designing frame filters, and well
1663 worth reflecting upon, is that frame filters should avoid unwinding
1664 the call stack if possible. Some stacks can run very deep, into the
1665 tens of thousands in some cases. To search every frame when a frame
1666 filter executes may be too expensive at that step. The frame filter
1667 cannot know how many frames it has to iterate over, and it may have to
1668 iterate through them all. This ends up duplicating effort as
1669 @value{GDBN} performs this iteration when it prints the frames. If
1670 the filter can defer unwinding frames until frame decorators are
1671 executed, after the last filter has executed, it should. @xref{Frame
1672 Decorator API}, for more information on decorators. Also, there are
1673 examples for both frame decorators and filters in later chapters.
1674 @xref{Writing a Frame Filter}, for more information.
1676 The Python dictionary @code{gdb.frame_filters} contains key/object
1677 pairings that comprise a frame filter. Frame filters in this
1678 dictionary are called @code{global} frame filters, and they are
1679 available when debugging all inferiors. These frame filters must
1680 register with the dictionary directly. In addition to the
1681 @code{global} dictionary, there are other dictionaries that are loaded
1682 with different inferiors via auto-loading (@pxref{Python
1683 Auto-loading}). The two other areas where frame filter dictionaries
1684 can be found are: @code{gdb.Progspace} which contains a
1685 @code{frame_filters} dictionary attribute, and each @code{gdb.Objfile}
1686 object which also contains a @code{frame_filters} dictionary
1689 When a command is executed from @value{GDBN} that is compatible with
1690 frame filters, @value{GDBN} combines the @code{global},
1691 @code{gdb.Progspace} and all @code{gdb.Objfile} dictionaries currently
1692 loaded. All of the @code{gdb.Objfile} dictionaries are combined, as
1693 several frames, and thus several object files, might be in use.
1694 @value{GDBN} then prunes any frame filter whose @code{enabled}
1695 attribute is @code{False}. This pruned list is then sorted according
1696 to the @code{priority} attribute in each filter.
1698 Once the dictionaries are combined, pruned and sorted, @value{GDBN}
1699 creates an iterator which wraps each frame in the call stack in a
1700 @code{FrameDecorator} object, and calls each filter in order. The
1701 output from the previous filter will always be the input to the next
1704 Frame filters have a mandatory interface which each frame filter must
1705 implement, defined here:
1707 @defun FrameFilter.filter (iterator)
1708 @value{GDBN} will call this method on a frame filter when it has
1709 reached the order in the priority list for that filter.
1711 For example, if there are four frame filters:
1722 The order that the frame filters will be called is:
1725 Filter3 -> Filter2 -> Filter1 -> Filter4
1728 Note that the output from @code{Filter3} is passed to the input of
1729 @code{Filter2}, and so on.
1731 This @code{filter} method is passed a Python iterator. This iterator
1732 contains a sequence of frame decorators that wrap each
1733 @code{gdb.Frame}, or a frame decorator that wraps another frame
1734 decorator. The first filter that is executed in the sequence of frame
1735 filters will receive an iterator entirely comprised of default
1736 @code{FrameDecorator} objects. However, after each frame filter is
1737 executed, the previous frame filter may have wrapped some or all of
1738 the frame decorators with their own frame decorator. As frame
1739 decorators must also conform to a mandatory interface, these
1740 decorators can be assumed to act in a uniform manner (@pxref{Frame
1743 This method must return an object conforming to the Python iterator
1744 protocol. Each item in the iterator must be an object conforming to
1745 the frame decorator interface. If a frame filter does not wish to
1746 perform any operations on this iterator, it should return that
1749 This method is not optional. If it does not exist, @value{GDBN} will
1750 raise and print an error.
1753 @defvar FrameFilter.name
1754 The @code{name} attribute must be Python string which contains the
1755 name of the filter displayed by @value{GDBN} (@pxref{Frame Filter
1756 Management}). This attribute may contain any combination of letters
1757 or numbers. Care should be taken to ensure that it is unique. This
1758 attribute is mandatory.
1761 @defvar FrameFilter.enabled
1762 The @code{enabled} attribute must be Python boolean. This attribute
1763 indicates to @value{GDBN} whether the frame filter is enabled, and
1764 should be considered when frame filters are executed. If
1765 @code{enabled} is @code{True}, then the frame filter will be executed
1766 when any of the backtrace commands detailed earlier in this chapter
1767 are executed. If @code{enabled} is @code{False}, then the frame
1768 filter will not be executed. This attribute is mandatory.
1771 @defvar FrameFilter.priority
1772 The @code{priority} attribute must be Python integer. This attribute
1773 controls the order of execution in relation to other frame filters.
1774 There are no imposed limits on the range of @code{priority} other than
1775 it must be a valid integer. The higher the @code{priority} attribute,
1776 the sooner the frame filter will be executed in relation to other
1777 frame filters. Although @code{priority} can be negative, it is
1778 recommended practice to assume zero is the lowest priority that a
1779 frame filter can be assigned. Frame filters that have the same
1780 priority are executed in unsorted order in that priority slot. This
1781 attribute is mandatory. 100 is a good default priority.
1784 @node Frame Decorator API
1785 @subsubsection Decorating Frames
1786 @cindex frame decorator api
1788 Frame decorators are sister objects to frame filters (@pxref{Frame
1789 Filter API}). Frame decorators are applied by a frame filter and can
1790 only be used in conjunction with frame filters.
1792 The purpose of a frame decorator is to customize the printed content
1793 of each @code{gdb.Frame} in commands where frame filters are executed.
1794 This concept is called decorating a frame. Frame decorators decorate
1795 a @code{gdb.Frame} with Python code contained within each API call.
1796 This separates the actual data contained in a @code{gdb.Frame} from
1797 the decorated data produced by a frame decorator. This abstraction is
1798 necessary to maintain integrity of the data contained in each
1801 Frame decorators have a mandatory interface, defined below.
1803 @value{GDBN} already contains a frame decorator called
1804 @code{FrameDecorator}. This contains substantial amounts of
1805 boilerplate code to decorate the content of a @code{gdb.Frame}. It is
1806 recommended that other frame decorators inherit and extend this
1807 object, and only to override the methods needed.
1809 @tindex gdb.FrameDecorator
1810 @code{FrameDecorator} is defined in the Python module
1811 @code{gdb.FrameDecorator}, so your code can import it like:
1813 from gdb.FrameDecorator import FrameDecorator
1816 @defun FrameDecorator.elided (self)
1818 The @code{elided} method groups frames together in a hierarchical
1819 system. An example would be an interpreter, where multiple low-level
1820 frames make up a single call in the interpreted language. In this
1821 example, the frame filter would elide the low-level frames and present
1822 a single high-level frame, representing the call in the interpreted
1823 language, to the user.
1825 The @code{elided} function must return an iterable and this iterable
1826 must contain the frames that are being elided wrapped in a suitable
1827 frame decorator. If no frames are being elided this function may
1828 return an empty iterable, or @code{None}. Elided frames are indented
1829 from normal frames in a @code{CLI} backtrace, or in the case of
1830 @code{GDB/MI}, are placed in the @code{children} field of the eliding
1833 It is the frame filter's task to also filter out the elided frames from
1834 the source iterator. This will avoid printing the frame twice.
1837 @defun FrameDecorator.function (self)
1839 This method returns the name of the function in the frame that is to
1842 This method must return a Python string describing the function, or
1845 If this function returns @code{None}, @value{GDBN} will not print any
1846 data for this field.
1849 @defun FrameDecorator.address (self)
1851 This method returns the address of the frame that is to be printed.
1853 This method must return a Python numeric integer type of sufficient
1854 size to describe the address of the frame, or @code{None}.
1856 If this function returns a @code{None}, @value{GDBN} will not print
1857 any data for this field.
1860 @defun FrameDecorator.filename (self)
1862 This method returns the filename and path associated with this frame.
1864 This method must return a Python string containing the filename and
1865 the path to the object file backing the frame, or @code{None}.
1867 If this function returns a @code{None}, @value{GDBN} will not print
1868 any data for this field.
1871 @defun FrameDecorator.line (self):
1873 This method returns the line number associated with the current
1874 position within the function addressed by this frame.
1876 This method must return a Python integer type, or @code{None}.
1878 If this function returns a @code{None}, @value{GDBN} will not print
1879 any data for this field.
1882 @defun FrameDecorator.frame_args (self)
1885 This method must return an iterable, or @code{None}. Returning an
1886 empty iterable, or @code{None} means frame arguments will not be
1887 printed for this frame. This iterable must contain objects that
1888 implement two methods, described here.
1890 This object must implement a @code{argument} method which takes a
1891 single @code{self} parameter and must return a @code{gdb.Symbol}
1892 (@pxref{Symbols In Python}), or a Python string. The object must also
1893 implement a @code{value} method which takes a single @code{self}
1894 parameter and must return a @code{gdb.Value} (@pxref{Values From
1895 Inferior}), a Python value, or @code{None}. If the @code{value}
1896 method returns @code{None}, and the @code{argument} method returns a
1897 @code{gdb.Symbol}, @value{GDBN} will look-up and print the value of
1898 the @code{gdb.Symbol} automatically.
1903 class SymValueWrapper():
1905 def __init__(self, symbol, value):
1915 class SomeFrameDecorator()
1918 def frame_args(self):
1921 block = self.inferior_frame.block()
1925 # Iterate over all symbols in a block. Only add
1926 # symbols that are arguments.
1928 if not sym.is_argument:
1930 args.append(SymValueWrapper(sym,None))
1932 # Add example synthetic argument.
1933 args.append(SymValueWrapper(``foo'', 42))
1939 @defun FrameDecorator.frame_locals (self)
1941 This method must return an iterable or @code{None}. Returning an
1942 empty iterable, or @code{None} means frame local arguments will not be
1943 printed for this frame.
1945 The object interface, the description of the various strategies for
1946 reading frame locals, and the example are largely similar to those
1947 described in the @code{frame_args} function, (@pxref{frame_args,,The
1948 frame filter frame_args function}). Below is a modified example:
1951 class SomeFrameDecorator()
1954 def frame_locals(self):
1957 block = self.inferior_frame.block()
1961 # Iterate over all symbols in a block. Add all
1962 # symbols, except arguments.
1966 vars.append(SymValueWrapper(sym,None))
1968 # Add an example of a synthetic local variable.
1969 vars.append(SymValueWrapper(``bar'', 99))
1975 @defun FrameDecorator.inferior_frame (self):
1977 This method must return the underlying @code{gdb.Frame} that this
1978 frame decorator is decorating. @value{GDBN} requires the underlying
1979 frame for internal frame information to determine how to print certain
1980 values when printing a frame.
1983 @node Writing a Frame Filter
1984 @subsubsection Writing a Frame Filter
1985 @cindex writing a frame filter
1987 There are three basic elements that a frame filter must implement: it
1988 must correctly implement the documented interface (@pxref{Frame Filter
1989 API}), it must register itself with @value{GDBN}, and finally, it must
1990 decide if it is to work on the data provided by @value{GDBN}. In all
1991 cases, whether it works on the iterator or not, each frame filter must
1992 return an iterator. A bare-bones frame filter follows the pattern in
1993 the following example.
1998 class FrameFilter():
2001 # Frame filter attribute creation.
2003 # 'name' is the name of the filter that GDB will display.
2005 # 'priority' is the priority of the filter relative to other
2008 # 'enabled' is a boolean that indicates whether this filter is
2009 # enabled and should be executed.
2015 # Register this frame filter with the global frame_filters
2017 gdb.frame_filters[self.name] = self
2019 def filter(self, frame_iter):
2020 # Just return the iterator.
2024 The frame filter in the example above implements the three
2025 requirements for all frame filters. It implements the API, self
2026 registers, and makes a decision on the iterator (in this case, it just
2027 returns the iterator untouched).
2029 The first step is attribute creation and assignment, and as shown in
2030 the comments the filter assigns the following attributes: @code{name},
2031 @code{priority} and whether the filter should be enabled with the
2032 @code{enabled} attribute.
2034 The second step is registering the frame filter with the dictionary or
2035 dictionaries that the frame filter has interest in. As shown in the
2036 comments, this filter just registers itself with the global dictionary
2037 @code{gdb.frame_filters}. As noted earlier, @code{gdb.frame_filters}
2038 is a dictionary that is initialized in the @code{gdb} module when
2039 @value{GDBN} starts. What dictionary a filter registers with is an
2040 important consideration. Generally, if a filter is specific to a set
2041 of code, it should be registered either in the @code{objfile} or
2042 @code{progspace} dictionaries as they are specific to the program
2043 currently loaded in @value{GDBN}. The global dictionary is always
2044 present in @value{GDBN} and is never unloaded. Any filters registered
2045 with the global dictionary will exist until @value{GDBN} exits. To
2046 avoid filters that may conflict, it is generally better to register
2047 frame filters against the dictionaries that more closely align with
2048 the usage of the filter currently in question. @xref{Python
2049 Auto-loading}, for further information on auto-loading Python scripts.
2051 @value{GDBN} takes a hands-off approach to frame filter registration,
2052 therefore it is the frame filter's responsibility to ensure
2053 registration has occurred, and that any exceptions are handled
2054 appropriately. In particular, you may wish to handle exceptions
2055 relating to Python dictionary key uniqueness. It is mandatory that
2056 the dictionary key is the same as frame filter's @code{name}
2057 attribute. When a user manages frame filters (@pxref{Frame Filter
2058 Management}), the names @value{GDBN} will display are those contained
2059 in the @code{name} attribute.
2061 The final step of this example is the implementation of the
2062 @code{filter} method. As shown in the example comments, we define the
2063 @code{filter} method and note that the method must take an iterator,
2064 and also must return an iterator. In this bare-bones example, the
2065 frame filter is not very useful as it just returns the iterator
2066 untouched. However this is a valid operation for frame filters that
2067 have the @code{enabled} attribute set, but decide not to operate on
2070 In the next example, the frame filter operates on all frames and
2071 utilizes a frame decorator to perform some work on the frames.
2072 @xref{Frame Decorator API}, for further information on the frame
2073 decorator interface.
2075 This example works on inlined frames. It highlights frames which are
2076 inlined by tagging them with an ``[inlined]'' tag. By applying a
2077 frame decorator to all frames with the Python @code{itertools imap}
2078 method, the example defers actions to the frame decorator. Frame
2079 decorators are only processed when @value{GDBN} prints the backtrace.
2081 This introduces a new decision making topic: whether to perform
2082 decision making operations at the filtering step, or at the printing
2083 step. In this example's approach, it does not perform any filtering
2084 decisions at the filtering step beyond mapping a frame decorator to
2085 each frame. This allows the actual decision making to be performed
2086 when each frame is printed. This is an important consideration, and
2087 well worth reflecting upon when designing a frame filter. An issue
2088 that frame filters should avoid is unwinding the stack if possible.
2089 Some stacks can run very deep, into the tens of thousands in some
2090 cases. To search every frame to determine if it is inlined ahead of
2091 time may be too expensive at the filtering step. The frame filter
2092 cannot know how many frames it has to iterate over, and it would have
2093 to iterate through them all. This ends up duplicating effort as
2094 @value{GDBN} performs this iteration when it prints the frames.
2096 In this example decision making can be deferred to the printing step.
2097 As each frame is printed, the frame decorator can examine each frame
2098 in turn when @value{GDBN} iterates. From a performance viewpoint,
2099 this is the most appropriate decision to make as it avoids duplicating
2100 the effort that the printing step would undertake anyway. Also, if
2101 there are many frame filters unwinding the stack during filtering, it
2102 can substantially delay the printing of the backtrace which will
2103 result in large memory usage, and a poor user experience.
2106 class InlineFilter():
2109 self.name = "InlinedFrameFilter"
2112 gdb.frame_filters[self.name] = self
2114 def filter(self, frame_iter):
2115 frame_iter = itertools.imap(InlinedFrameDecorator,
2120 This frame filter is somewhat similar to the earlier example, except
2121 that the @code{filter} method applies a frame decorator object called
2122 @code{InlinedFrameDecorator} to each element in the iterator. The
2123 @code{imap} Python method is light-weight. It does not proactively
2124 iterate over the iterator, but rather creates a new iterator which
2125 wraps the existing one.
2127 Below is the frame decorator for this example.
2130 class InlinedFrameDecorator(FrameDecorator):
2132 def __init__(self, fobj):
2133 super(InlinedFrameDecorator, self).__init__(fobj)
2136 frame = fobj.inferior_frame()
2137 name = str(frame.name())
2139 if frame.type() == gdb.INLINE_FRAME:
2140 name = name + " [inlined]"
2145 This frame decorator only defines and overrides the @code{function}
2146 method. It lets the supplied @code{FrameDecorator}, which is shipped
2147 with @value{GDBN}, perform the other work associated with printing
2150 The combination of these two objects create this output from a
2154 #0 0x004004e0 in bar () at inline.c:11
2155 #1 0x00400566 in max [inlined] (b=6, a=12) at inline.c:21
2156 #2 0x00400566 in main () at inline.c:31
2159 So in the case of this example, a frame decorator is applied to all
2160 frames, regardless of whether they may be inlined or not. As
2161 @value{GDBN} iterates over the iterator produced by the frame filters,
2162 @value{GDBN} executes each frame decorator which then makes a decision
2163 on what to print in the @code{function} callback. Using a strategy
2164 like this is a way to defer decisions on the frame content to printing
2167 @subheading Eliding Frames
2169 It might be that the above example is not desirable for representing
2170 inlined frames, and a hierarchical approach may be preferred. If we
2171 want to hierarchically represent frames, the @code{elided} frame
2172 decorator interface might be preferable.
2174 This example approaches the issue with the @code{elided} method. This
2175 example is quite long, but very simplistic. It is out-of-scope for
2176 this section to write a complete example that comprehensively covers
2177 all approaches of finding and printing inlined frames. However, this
2178 example illustrates the approach an author might use.
2180 This example comprises of three sections.
2183 class InlineFrameFilter():
2186 self.name = "InlinedFrameFilter"
2189 gdb.frame_filters[self.name] = self
2191 def filter(self, frame_iter):
2192 return ElidingInlineIterator(frame_iter)
2195 This frame filter is very similar to the other examples. The only
2196 difference is this frame filter is wrapping the iterator provided to
2197 it (@code{frame_iter}) with a custom iterator called
2198 @code{ElidingInlineIterator}. This again defers actions to when
2199 @value{GDBN} prints the backtrace, as the iterator is not traversed
2202 The iterator for this example is as follows. It is in this section of
2203 the example where decisions are made on the content of the backtrace.
2206 class ElidingInlineIterator:
2207 def __init__(self, ii):
2208 self.input_iterator = ii
2214 frame = next(self.input_iterator)
2216 if frame.inferior_frame().type() != gdb.INLINE_FRAME:
2220 eliding_frame = next(self.input_iterator)
2221 except StopIteration:
2223 return ElidingFrameDecorator(eliding_frame, [frame])
2226 This iterator implements the Python iterator protocol. When the
2227 @code{next} function is called (when @value{GDBN} prints each frame),
2228 the iterator checks if this frame decorator, @code{frame}, is wrapping
2229 an inlined frame. If it is not, it returns the existing frame decorator
2230 untouched. If it is wrapping an inlined frame, it assumes that the
2231 inlined frame was contained within the next oldest frame,
2232 @code{eliding_frame}, which it fetches. It then creates and returns a
2233 frame decorator, @code{ElidingFrameDecorator}, which contains both the
2234 elided frame, and the eliding frame.
2237 class ElidingInlineDecorator(FrameDecorator):
2239 def __init__(self, frame, elided_frames):
2240 super(ElidingInlineDecorator, self).__init__(frame)
2242 self.elided_frames = elided_frames
2245 return iter(self.elided_frames)
2248 This frame decorator overrides one function and returns the inlined
2249 frame in the @code{elided} method. As before it lets
2250 @code{FrameDecorator} do the rest of the work involved in printing
2251 this frame. This produces the following output.
2254 #0 0x004004e0 in bar () at inline.c:11
2255 #2 0x00400529 in main () at inline.c:25
2256 #1 0x00400529 in max (b=6, a=12) at inline.c:15
2259 In that output, @code{max} which has been inlined into @code{main} is
2260 printed hierarchically. Another approach would be to combine the
2261 @code{function} method, and the @code{elided} method to both print a
2262 marker in the inlined frame, and also show the hierarchical
2265 @node Unwinding Frames in Python
2266 @subsubsection Unwinding Frames in Python
2267 @cindex unwinding frames in Python
2269 In @value{GDBN} terminology ``unwinding'' is the process of finding
2270 the previous frame (that is, caller's) from the current one. An
2271 unwinder has three methods. The first one checks if it can handle
2272 given frame (``sniff'' it). For the frames it can sniff an unwinder
2273 provides two additional methods: it can return frame's ID, and it can
2274 fetch registers from the previous frame. A running @value{GDBN}
2275 mantains a list of the unwinders and calls each unwinder's sniffer in
2276 turn until it finds the one that recognizes the current frame. There
2277 is an API to register an unwinder.
2279 The unwinders that come with @value{GDBN} handle standard frames.
2280 However, mixed language applications (for example, an application
2281 running Java Virtual Machine) sometimes use frame layouts that cannot
2282 be handled by the @value{GDBN} unwinders. You can write Python code
2283 that can handle such custom frames.
2285 You implement a frame unwinder in Python as a class with which has two
2286 attributes, @code{name} and @code{enabled}, with obvious meanings, and
2287 a single method @code{__call__}, which examines a given frame and
2288 returns an object (an instance of @code{gdb.UnwindInfo class)}
2289 describing it. If an unwinder does not recognize a frame, it should
2290 return @code{None}. The code in @value{GDBN} that enables writing
2291 unwinders in Python uses this object to return frame's ID and previous
2292 frame registers when @value{GDBN} core asks for them.
2294 An unwinder should do as little work as possible. Some otherwise
2295 innocuous operations can cause problems (even crashes, as this code is
2296 not not well-hardened yet). For example, making an inferior call from
2297 an unwinder is unadvisable, as an inferior call will reset
2298 @value{GDBN}'s stack unwinding process, potentially causing re-entrant
2301 @subheading Unwinder Input
2303 An object passed to an unwinder (a @code{gdb.PendingFrame} instance)
2304 provides a method to read frame's registers:
2306 @defun PendingFrame.read_register (reg)
2307 This method returns the contents of the register @var{reg} in the
2308 frame as a @code{gdb.Value} object. @var{reg} can be either a
2309 register number or a register name; the values are platform-specific.
2310 They are usually found in the corresponding
2311 @file{@var{platform}-tdep.h} file in the @value{GDBN} source tree. If
2312 @var{reg} does not name a register for the current architecture, this
2313 method will throw an exception.
2315 Note that this method will always return a @code{gdb.Value} for a
2316 valid register name. This does not mean that the value will be valid.
2317 For example, you may request a register that an earlier unwinder could
2318 not unwind---the value will be unavailable. Instead, the
2319 @code{gdb.Value} returned from this method will be lazy; that is, its
2320 underlying bits will not be fetched until it is first used. So,
2321 attempting to use such a value will cause an exception at the point of
2324 The type of the returned @code{gdb.Value} depends on the register and
2325 the architecture. It is common for registers to have a scalar type,
2326 like @code{long long}; but many other types are possible, such as
2327 pointer, pointer-to-function, floating point or vector types.
2330 It also provides a factory method to create a @code{gdb.UnwindInfo}
2331 instance to be returned to @value{GDBN}:
2333 @defun PendingFrame.create_unwind_info (frame_id)
2334 Returns a new @code{gdb.UnwindInfo} instance identified by given
2335 @var{frame_id}. The argument is used to build @value{GDBN}'s frame ID
2336 using one of functions provided by @value{GDBN}. @var{frame_id}'s attributes
2337 determine which function will be used, as follows:
2341 The frame is identified by the given stack address and PC. The stack
2342 address must be chosen so that it is constant throughout the lifetime
2343 of the frame, so a typical choice is the value of the stack pointer at
2344 the start of the function---in the DWARF standard, this would be the
2345 ``Call Frame Address''.
2347 This is the most common case by far. The other cases are documented
2348 for completeness but are only useful in specialized situations.
2350 @item sp, pc, special
2351 The frame is identified by the stack address, the PC, and a
2352 ``special'' address. The special address is used on architectures
2353 that can have frames that do not change the stack, but which are still
2354 distinct, for example the IA-64, which has a second stack for
2355 registers. Both @var{sp} and @var{special} must be constant
2356 throughout the lifetime of the frame.
2359 The frame is identified by the stack address only. Any other stack
2360 frame with a matching @var{sp} will be considered to match this frame.
2361 Inside gdb, this is called a ``wild frame''. You will never need
2365 Each attribute value should be an instance of @code{gdb.Value}.
2369 @subheading Unwinder Output: UnwindInfo
2371 Use @code{PendingFrame.create_unwind_info} method described above to
2372 create a @code{gdb.UnwindInfo} instance. Use the following method to
2373 specify caller registers that have been saved in this frame:
2375 @defun gdb.UnwindInfo.add_saved_register (reg, value)
2376 @var{reg} identifies the register. It can be a number or a name, just
2377 as for the @code{PendingFrame.read_register} method above.
2378 @var{value} is a register value (a @code{gdb.Value} object).
2381 @subheading Unwinder Skeleton Code
2383 @value{GDBN} comes with the module containing the base @code{Unwinder}
2384 class. Derive your unwinder class from it and structure the code as
2388 from gdb.unwinders import Unwinder
2390 class FrameId(object):
2391 def __init__(self, sp, pc):
2396 class MyUnwinder(Unwinder):
2398 supe(MyUnwinder, self).__init___(<expects unwinder name argument>)
2400 def __call__(pending_frame):
2401 if not <we recognize frame>:
2403 # Create UnwindInfo. Usually the frame is identified by the stack
2404 # pointer and the program counter.
2405 sp = pending_frame.read_register(<SP number>)
2406 pc = pending_frame.read_register(<PC number>)
2407 unwind_info = pending_frame.create_unwind_info(FrameId(sp, pc))
2409 # Find the values of the registers in the caller's frame and
2410 # save them in the result:
2411 unwind_info.add_saved_register(<register>, <value>)
2414 # Return the result:
2419 @subheading Registering a Unwinder
2421 An object file, a program space, and the @value{GDBN} proper can have
2422 unwinders registered with it.
2424 The @code{gdb.unwinders} module provides the function to register a
2427 @defun gdb.unwinder.register_unwinder (locus, unwinder, replace=False)
2428 @var{locus} is specifies an object file or a program space to which
2429 @var{unwinder} is added. Passing @code{None} or @code{gdb} adds
2430 @var{unwinder} to the @value{GDBN}'s global unwinder list. The newly
2431 added @var{unwinder} will be called before any other unwinder from the
2432 same locus. Two unwinders in the same locus cannot have the same
2433 name. An attempt to add a unwinder with already existing name raises
2434 an exception unless @var{replace} is @code{True}, in which case the
2435 old unwinder is deleted.
2438 @subheading Unwinder Precedence
2440 @value{GDBN} first calls the unwinders from all the object files in no
2441 particular order, then the unwinders from the current program space,
2442 and finally the unwinders from @value{GDBN}.
2444 @node Xmethods In Python
2445 @subsubsection Xmethods In Python
2446 @cindex xmethods in Python
2448 @dfn{Xmethods} are additional methods or replacements for existing
2449 methods of a C@t{++} class. This feature is useful for those cases
2450 where a method defined in C@t{++} source code could be inlined or
2451 optimized out by the compiler, making it unavailable to @value{GDBN}.
2452 For such cases, one can define an xmethod to serve as a replacement
2453 for the method defined in the C@t{++} source code. @value{GDBN} will
2454 then invoke the xmethod, instead of the C@t{++} method, to
2455 evaluate expressions. One can also use xmethods when debugging
2456 with core files. Moreover, when debugging live programs, invoking an
2457 xmethod need not involve running the inferior (which can potentially
2458 perturb its state). Hence, even if the C@t{++} method is available, it
2459 is better to use its replacement xmethod if one is defined.
2461 The xmethods feature in Python is available via the concepts of an
2462 @dfn{xmethod matcher} and an @dfn{xmethod worker}. To
2463 implement an xmethod, one has to implement a matcher and a
2464 corresponding worker for it (more than one worker can be
2465 implemented, each catering to a different overloaded instance of the
2466 method). Internally, @value{GDBN} invokes the @code{match} method of a
2467 matcher to match the class type and method name. On a match, the
2468 @code{match} method returns a list of matching @emph{worker} objects.
2469 Each worker object typically corresponds to an overloaded instance of
2470 the xmethod. They implement a @code{get_arg_types} method which
2471 returns a sequence of types corresponding to the arguments the xmethod
2472 requires. @value{GDBN} uses this sequence of types to perform
2473 overload resolution and picks a winning xmethod worker. A winner
2474 is also selected from among the methods @value{GDBN} finds in the
2475 C@t{++} source code. Next, the winning xmethod worker and the
2476 winning C@t{++} method are compared to select an overall winner. In
2477 case of a tie between a xmethod worker and a C@t{++} method, the
2478 xmethod worker is selected as the winner. That is, if a winning
2479 xmethod worker is found to be equivalent to the winning C@t{++}
2480 method, then the xmethod worker is treated as a replacement for
2481 the C@t{++} method. @value{GDBN} uses the overall winner to invoke the
2482 method. If the winning xmethod worker is the overall winner, then
2483 the corresponding xmethod is invoked via the @code{__call__} method
2484 of the worker object.
2486 If one wants to implement an xmethod as a replacement for an
2487 existing C@t{++} method, then they have to implement an equivalent
2488 xmethod which has exactly the same name and takes arguments of
2489 exactly the same type as the C@t{++} method. If the user wants to
2490 invoke the C@t{++} method even though a replacement xmethod is
2491 available for that method, then they can disable the xmethod.
2493 @xref{Xmethod API}, for API to implement xmethods in Python.
2494 @xref{Writing an Xmethod}, for implementing xmethods in Python.
2497 @subsubsection Xmethod API
2500 The @value{GDBN} Python API provides classes, interfaces and functions
2501 to implement, register and manipulate xmethods.
2502 @xref{Xmethods In Python}.
2504 An xmethod matcher should be an instance of a class derived from
2505 @code{XMethodMatcher} defined in the module @code{gdb.xmethod}, or an
2506 object with similar interface and attributes. An instance of
2507 @code{XMethodMatcher} has the following attributes:
2510 The name of the matcher.
2514 A boolean value indicating whether the matcher is enabled or disabled.
2518 A list of named methods managed by the matcher. Each object in the list
2519 is an instance of the class @code{XMethod} defined in the module
2520 @code{gdb.xmethod}, or any object with the following attributes:
2525 Name of the xmethod which should be unique for each xmethod
2526 managed by the matcher.
2529 A boolean value indicating whether the xmethod is enabled or
2534 The class @code{XMethod} is a convenience class with same
2535 attributes as above along with the following constructor:
2537 @defun XMethod.__init__ (self, name)
2538 Constructs an enabled xmethod with name @var{name}.
2543 The @code{XMethodMatcher} class has the following methods:
2545 @defun XMethodMatcher.__init__ (self, name)
2546 Constructs an enabled xmethod matcher with name @var{name}. The
2547 @code{methods} attribute is initialized to @code{None}.
2550 @defun XMethodMatcher.match (self, class_type, method_name)
2551 Derived classes should override this method. It should return a
2552 xmethod worker object (or a sequence of xmethod worker
2553 objects) matching the @var{class_type} and @var{method_name}.
2554 @var{class_type} is a @code{gdb.Type} object, and @var{method_name}
2555 is a string value. If the matcher manages named methods as listed in
2556 its @code{methods} attribute, then only those worker objects whose
2557 corresponding entries in the @code{methods} list are enabled should be
2561 An xmethod worker should be an instance of a class derived from
2562 @code{XMethodWorker} defined in the module @code{gdb.xmethod},
2563 or support the following interface:
2565 @defun XMethodWorker.get_arg_types (self)
2566 This method returns a sequence of @code{gdb.Type} objects corresponding
2567 to the arguments that the xmethod takes. It can return an empty
2568 sequence or @code{None} if the xmethod does not take any arguments.
2569 If the xmethod takes a single argument, then a single
2570 @code{gdb.Type} object corresponding to it can be returned.
2573 @defun XMethodWorker.get_result_type (self, *args)
2574 This method returns a @code{gdb.Type} object representing the type
2575 of the result of invoking this xmethod.
2576 The @var{args} argument is the same tuple of arguments that would be
2577 passed to the @code{__call__} method of this worker.
2580 @defun XMethodWorker.__call__ (self, *args)
2581 This is the method which does the @emph{work} of the xmethod. The
2582 @var{args} arguments is the tuple of arguments to the xmethod. Each
2583 element in this tuple is a gdb.Value object. The first element is
2584 always the @code{this} pointer value.
2587 For @value{GDBN} to lookup xmethods, the xmethod matchers
2588 should be registered using the following function defined in the module
2591 @defun register_xmethod_matcher (locus, matcher, replace=False)
2592 The @code{matcher} is registered with @code{locus}, replacing an
2593 existing matcher with the same name as @code{matcher} if
2594 @code{replace} is @code{True}. @code{locus} can be a
2595 @code{gdb.Objfile} object (@pxref{Objfiles In Python}), or a
2596 @code{gdb.Progspace} object (@pxref{Progspaces In Python}), or
2597 @code{None}. If it is @code{None}, then @code{matcher} is registered
2601 @node Writing an Xmethod
2602 @subsubsection Writing an Xmethod
2603 @cindex writing xmethods in Python
2605 Implementing xmethods in Python will require implementing xmethod
2606 matchers and xmethod workers (@pxref{Xmethods In Python}). Consider
2607 the following C@t{++} class:
2613 MyClass (int a) : a_(a) @{ @}
2615 int geta (void) @{ return a_; @}
2616 int operator+ (int b);
2623 MyClass::operator+ (int b)
2630 Let us define two xmethods for the class @code{MyClass}, one
2631 replacing the method @code{geta}, and another adding an overloaded
2632 flavor of @code{operator+} which takes a @code{MyClass} argument (the
2633 C@t{++} code above already has an overloaded @code{operator+}
2634 which takes an @code{int} argument). The xmethod matcher can be
2638 class MyClass_geta(gdb.xmethod.XMethod):
2640 gdb.xmethod.XMethod.__init__(self, 'geta')
2642 def get_worker(self, method_name):
2643 if method_name == 'geta':
2644 return MyClassWorker_geta()
2647 class MyClass_sum(gdb.xmethod.XMethod):
2649 gdb.xmethod.XMethod.__init__(self, 'sum')
2651 def get_worker(self, method_name):
2652 if method_name == 'operator+':
2653 return MyClassWorker_plus()
2656 class MyClassMatcher(gdb.xmethod.XMethodMatcher):
2658 gdb.xmethod.XMethodMatcher.__init__(self, 'MyClassMatcher')
2659 # List of methods 'managed' by this matcher
2660 self.methods = [MyClass_geta(), MyClass_sum()]
2662 def match(self, class_type, method_name):
2663 if class_type.tag != 'MyClass':
2666 for method in self.methods:
2668 worker = method.get_worker(method_name)
2670 workers.append(worker)
2676 Notice that the @code{match} method of @code{MyClassMatcher} returns
2677 a worker object of type @code{MyClassWorker_geta} for the @code{geta}
2678 method, and a worker object of type @code{MyClassWorker_plus} for the
2679 @code{operator+} method. This is done indirectly via helper classes
2680 derived from @code{gdb.xmethod.XMethod}. One does not need to use the
2681 @code{methods} attribute in a matcher as it is optional. However, if a
2682 matcher manages more than one xmethod, it is a good practice to list the
2683 xmethods in the @code{methods} attribute of the matcher. This will then
2684 facilitate enabling and disabling individual xmethods via the
2685 @code{enable/disable} commands. Notice also that a worker object is
2686 returned only if the corresponding entry in the @code{methods} attribute
2687 of the matcher is enabled.
2689 The implementation of the worker classes returned by the matcher setup
2690 above is as follows:
2693 class MyClassWorker_geta(gdb.xmethod.XMethodWorker):
2694 def get_arg_types(self):
2697 def get_result_type(self, obj):
2698 return gdb.lookup_type('int')
2700 def __call__(self, obj):
2704 class MyClassWorker_plus(gdb.xmethod.XMethodWorker):
2705 def get_arg_types(self):
2706 return gdb.lookup_type('MyClass')
2708 def get_result_type(self, obj):
2709 return gdb.lookup_type('int')
2711 def __call__(self, obj, other):
2712 return obj['a_'] + other['a_']
2715 For @value{GDBN} to actually lookup a xmethod, it has to be
2716 registered with it. The matcher defined above is registered with
2717 @value{GDBN} globally as follows:
2720 gdb.xmethod.register_xmethod_matcher(None, MyClassMatcher())
2723 If an object @code{obj} of type @code{MyClass} is initialized in C@t{++}
2731 then, after loading the Python script defining the xmethod matchers
2732 and workers into @code{GDBN}, invoking the method @code{geta} or using
2733 the operator @code{+} on @code{obj} will invoke the xmethods
2744 Consider another example with a C++ template class:
2751 MyTemplate () : dsize_(10), data_ (new T [10]) @{ @}
2752 ~MyTemplate () @{ delete [] data_; @}
2754 int footprint (void)
2756 return sizeof (T) * dsize_ + sizeof (MyTemplate<T>);
2765 Let us implement an xmethod for the above class which serves as a
2766 replacement for the @code{footprint} method. The full code listing
2767 of the xmethod workers and xmethod matchers is as follows:
2770 class MyTemplateWorker_footprint(gdb.xmethod.XMethodWorker):
2771 def __init__(self, class_type):
2772 self.class_type = class_type
2774 def get_arg_types(self):
2777 def get_result_type(self):
2778 return gdb.lookup_type('int')
2780 def __call__(self, obj):
2781 return (self.class_type.sizeof +
2783 self.class_type.template_argument(0).sizeof)
2786 class MyTemplateMatcher_footprint(gdb.xmethod.XMethodMatcher):
2788 gdb.xmethod.XMethodMatcher.__init__(self, 'MyTemplateMatcher')
2790 def match(self, class_type, method_name):
2791 if (re.match('MyTemplate<[ \t\n]*[_a-zA-Z][ _a-zA-Z0-9]*>',
2793 method_name == 'footprint'):
2794 return MyTemplateWorker_footprint(class_type)
2797 Notice that, in this example, we have not used the @code{methods}
2798 attribute of the matcher as the matcher manages only one xmethod. The
2799 user can enable/disable this xmethod by enabling/disabling the matcher
2802 @node Inferiors In Python
2803 @subsubsection Inferiors In Python
2804 @cindex inferiors in Python
2806 @findex gdb.Inferior
2807 Programs which are being run under @value{GDBN} are called inferiors
2808 (@pxref{Inferiors and Programs}). Python scripts can access
2809 information about and manipulate inferiors controlled by @value{GDBN}
2810 via objects of the @code{gdb.Inferior} class.
2812 The following inferior-related functions are available in the @code{gdb}
2815 @defun gdb.inferiors ()
2816 Return a tuple containing all inferior objects.
2819 @defun gdb.selected_inferior ()
2820 Return an object representing the current inferior.
2823 A @code{gdb.Inferior} object has the following attributes:
2825 @defvar Inferior.num
2826 ID of inferior, as assigned by GDB.
2829 @defvar Inferior.pid
2830 Process ID of the inferior, as assigned by the underlying operating
2834 @defvar Inferior.was_attached
2835 Boolean signaling whether the inferior was created using `attach', or
2836 started by @value{GDBN} itself.
2839 A @code{gdb.Inferior} object has the following methods:
2841 @defun Inferior.is_valid ()
2842 Returns @code{True} if the @code{gdb.Inferior} object is valid,
2843 @code{False} if not. A @code{gdb.Inferior} object will become invalid
2844 if the inferior no longer exists within @value{GDBN}. All other
2845 @code{gdb.Inferior} methods will throw an exception if it is invalid
2846 at the time the method is called.
2849 @defun Inferior.threads ()
2850 This method returns a tuple holding all the threads which are valid
2851 when it is called. If there are no valid threads, the method will
2852 return an empty tuple.
2855 @findex Inferior.read_memory
2856 @defun Inferior.read_memory (address, length)
2857 Read @var{length} addressable memory units from the inferior, starting at
2858 @var{address}. Returns a buffer object, which behaves much like an array
2859 or a string. It can be modified and given to the
2860 @code{Inferior.write_memory} function. In Python 3, the return
2861 value is a @code{memoryview} object.
2864 @findex Inferior.write_memory
2865 @defun Inferior.write_memory (address, buffer @r{[}, length@r{]})
2866 Write the contents of @var{buffer} to the inferior, starting at
2867 @var{address}. The @var{buffer} parameter must be a Python object
2868 which supports the buffer protocol, i.e., a string, an array or the
2869 object returned from @code{Inferior.read_memory}. If given, @var{length}
2870 determines the number of addressable memory units from @var{buffer} to be
2874 @findex gdb.search_memory
2875 @defun Inferior.search_memory (address, length, pattern)
2876 Search a region of the inferior memory starting at @var{address} with
2877 the given @var{length} using the search pattern supplied in
2878 @var{pattern}. The @var{pattern} parameter must be a Python object
2879 which supports the buffer protocol, i.e., a string, an array or the
2880 object returned from @code{gdb.read_memory}. Returns a Python @code{Long}
2881 containing the address where the pattern was found, or @code{None} if
2882 the pattern could not be found.
2885 @findex Inferior.thread_from_thread_handle
2886 @defun Inferior.thread_from_thread_handle (thread_handle)
2887 Return the thread object corresponding to @var{thread_handle}, a thread
2888 library specific data structure such as @code{pthread_t} for pthreads
2889 library implementations.
2892 @node Events In Python
2893 @subsubsection Events In Python
2894 @cindex inferior events in Python
2896 @value{GDBN} provides a general event facility so that Python code can be
2897 notified of various state changes, particularly changes that occur in
2900 An @dfn{event} is just an object that describes some state change. The
2901 type of the object and its attributes will vary depending on the details
2902 of the change. All the existing events are described below.
2904 In order to be notified of an event, you must register an event handler
2905 with an @dfn{event registry}. An event registry is an object in the
2906 @code{gdb.events} module which dispatches particular events. A registry
2907 provides methods to register and unregister event handlers:
2909 @defun EventRegistry.connect (object)
2910 Add the given callable @var{object} to the registry. This object will be
2911 called when an event corresponding to this registry occurs.
2914 @defun EventRegistry.disconnect (object)
2915 Remove the given @var{object} from the registry. Once removed, the object
2916 will no longer receive notifications of events.
2922 def exit_handler (event):
2923 print "event type: exit"
2924 print "exit code: %d" % (event.exit_code)
2926 gdb.events.exited.connect (exit_handler)
2929 In the above example we connect our handler @code{exit_handler} to the
2930 registry @code{events.exited}. Once connected, @code{exit_handler} gets
2931 called when the inferior exits. The argument @dfn{event} in this example is
2932 of type @code{gdb.ExitedEvent}. As you can see in the example the
2933 @code{ExitedEvent} object has an attribute which indicates the exit code of
2936 The following is a listing of the event registries that are available and
2937 details of the events they emit:
2942 Emits @code{gdb.ThreadEvent}.
2944 Some events can be thread specific when @value{GDBN} is running in non-stop
2945 mode. When represented in Python, these events all extend
2946 @code{gdb.ThreadEvent}. Note, this event is not emitted directly; instead,
2947 events which are emitted by this or other modules might extend this event.
2948 Examples of these events are @code{gdb.BreakpointEvent} and
2949 @code{gdb.ContinueEvent}.
2951 @defvar ThreadEvent.inferior_thread
2952 In non-stop mode this attribute will be set to the specific thread which was
2953 involved in the emitted event. Otherwise, it will be set to @code{None}.
2956 Emits @code{gdb.ContinueEvent} which extends @code{gdb.ThreadEvent}.
2958 This event indicates that the inferior has been continued after a stop. For
2959 inherited attribute refer to @code{gdb.ThreadEvent} above.
2962 Emits @code{events.ExitedEvent} which indicates that the inferior has exited.
2963 @code{events.ExitedEvent} has two attributes:
2964 @defvar ExitedEvent.exit_code
2965 An integer representing the exit code, if available, which the inferior
2966 has returned. (The exit code could be unavailable if, for example,
2967 @value{GDBN} detaches from the inferior.) If the exit code is unavailable,
2968 the attribute does not exist.
2970 @defvar ExitedEvent.inferior
2971 A reference to the inferior which triggered the @code{exited} event.
2975 Emits @code{gdb.StopEvent} which extends @code{gdb.ThreadEvent}.
2977 Indicates that the inferior has stopped. All events emitted by this registry
2978 extend StopEvent. As a child of @code{gdb.ThreadEvent}, @code{gdb.StopEvent}
2979 will indicate the stopped thread when @value{GDBN} is running in non-stop
2980 mode. Refer to @code{gdb.ThreadEvent} above for more details.
2982 Emits @code{gdb.SignalEvent} which extends @code{gdb.StopEvent}.
2984 This event indicates that the inferior or one of its threads has received as
2985 signal. @code{gdb.SignalEvent} has the following attributes:
2987 @defvar SignalEvent.stop_signal
2988 A string representing the signal received by the inferior. A list of possible
2989 signal values can be obtained by running the command @code{info signals} in
2990 the @value{GDBN} command prompt.
2993 Also emits @code{gdb.BreakpointEvent} which extends @code{gdb.StopEvent}.
2995 @code{gdb.BreakpointEvent} event indicates that one or more breakpoints have
2996 been hit, and has the following attributes:
2998 @defvar BreakpointEvent.breakpoints
2999 A sequence containing references to all the breakpoints (type
3000 @code{gdb.Breakpoint}) that were hit.
3001 @xref{Breakpoints In Python}, for details of the @code{gdb.Breakpoint} object.
3003 @defvar BreakpointEvent.breakpoint
3004 A reference to the first breakpoint that was hit.
3005 This function is maintained for backward compatibility and is now deprecated
3006 in favor of the @code{gdb.BreakpointEvent.breakpoints} attribute.
3009 @item events.new_objfile
3010 Emits @code{gdb.NewObjFileEvent} which indicates that a new object file has
3011 been loaded by @value{GDBN}. @code{gdb.NewObjFileEvent} has one attribute:
3013 @defvar NewObjFileEvent.new_objfile
3014 A reference to the object file (@code{gdb.Objfile}) which has been loaded.
3015 @xref{Objfiles In Python}, for details of the @code{gdb.Objfile} object.
3018 @item events.clear_objfiles
3019 Emits @code{gdb.ClearObjFilesEvent} which indicates that the list of object
3020 files for a program space has been reset.
3021 @code{gdb.ClearObjFilesEvent} has one attribute:
3023 @defvar ClearObjFilesEvent.progspace
3024 A reference to the program space (@code{gdb.Progspace}) whose objfile list has
3025 been cleared. @xref{Progspaces In Python}.
3028 @item events.inferior_call
3029 Emits events just before and after a function in the inferior is
3030 called by @value{GDBN}. Before an inferior call, this emits an event
3031 of type @code{gdb.InferiorCallPreEvent}, and after an inferior call,
3032 this emits an event of type @code{gdb.InferiorCallPostEvent}.
3035 @tindex gdb.InferiorCallPreEvent
3036 @item @code{gdb.InferiorCallPreEvent}
3037 Indicates that a function in the inferior is about to be called.
3039 @defvar InferiorCallPreEvent.ptid
3040 The thread in which the call will be run.
3043 @defvar InferiorCallPreEvent.address
3044 The location of the function to be called.
3047 @tindex gdb.InferiorCallPostEvent
3048 @item @code{gdb.InferiorCallPostEvent}
3049 Indicates that a function in the inferior has just been called.
3051 @defvar InferiorCallPostEvent.ptid
3052 The thread in which the call was run.
3055 @defvar InferiorCallPostEvent.address
3056 The location of the function that was called.
3060 @item events.memory_changed
3061 Emits @code{gdb.MemoryChangedEvent} which indicates that the memory of the
3062 inferior has been modified by the @value{GDBN} user, for instance via a
3063 command like @w{@code{set *addr = value}}. The event has the following
3066 @defvar MemoryChangedEvent.address
3067 The start address of the changed region.
3070 @defvar MemoryChangedEvent.length
3071 Length in bytes of the changed region.
3074 @item events.register_changed
3075 Emits @code{gdb.RegisterChangedEvent} which indicates that a register in the
3076 inferior has been modified by the @value{GDBN} user.
3078 @defvar RegisterChangedEvent.frame
3079 A gdb.Frame object representing the frame in which the register was modified.
3081 @defvar RegisterChangedEvent.regnum
3082 Denotes which register was modified.
3085 @item events.breakpoint_created
3086 This is emitted when a new breakpoint has been created. The argument
3087 that is passed is the new @code{gdb.Breakpoint} object.
3089 @item events.breakpoint_modified
3090 This is emitted when a breakpoint has been modified in some way. The
3091 argument that is passed is the new @code{gdb.Breakpoint} object.
3093 @item events.breakpoint_deleted
3094 This is emitted when a breakpoint has been deleted. The argument that
3095 is passed is the @code{gdb.Breakpoint} object. When this event is
3096 emitted, the @code{gdb.Breakpoint} object will already be in its
3097 invalid state; that is, the @code{is_valid} method will return
3100 @item events.before_prompt
3101 This event carries no payload. It is emitted each time @value{GDBN}
3102 presents a prompt to the user.
3104 @item events.new_inferior
3105 This is emitted when a new inferior is created. Note that the
3106 inferior is not necessarily running; in fact, it may not even have an
3107 associated executable.
3109 The event is of type @code{gdb.NewInferiorEvent}. This has a single
3112 @defvar NewInferiorEvent.inferior
3113 The new inferior, a @code{gdb.Inferior} object.
3116 @item events.inferior_deleted
3117 This is emitted when an inferior has been deleted. Note that this is
3118 not the same as process exit; it is notified when the inferior itself
3119 is removed, say via @code{remove-inferiors}.
3121 The event is of type @code{gdb.InferiorDeletedEvent}. This has a single
3124 @defvar NewInferiorEvent.inferior
3125 The inferior that is being removed, a @code{gdb.Inferior} object.
3128 @item events.new_thread
3129 This is emitted when @value{GDBN} notices a new thread. The event is of
3130 type @code{gdb.NewThreadEvent}, which extends @code{gdb.ThreadEvent}.
3131 This has a single attribute:
3133 @defvar NewThreadEvent.inferior_thread
3139 @node Threads In Python
3140 @subsubsection Threads In Python
3141 @cindex threads in python
3143 @findex gdb.InferiorThread
3144 Python scripts can access information about, and manipulate inferior threads
3145 controlled by @value{GDBN}, via objects of the @code{gdb.InferiorThread} class.
3147 The following thread-related functions are available in the @code{gdb}
3150 @findex gdb.selected_thread
3151 @defun gdb.selected_thread ()
3152 This function returns the thread object for the selected thread. If there
3153 is no selected thread, this will return @code{None}.
3156 A @code{gdb.InferiorThread} object has the following attributes:
3158 @defvar InferiorThread.name
3159 The name of the thread. If the user specified a name using
3160 @code{thread name}, then this returns that name. Otherwise, if an
3161 OS-supplied name is available, then it is returned. Otherwise, this
3162 returns @code{None}.
3164 This attribute can be assigned to. The new value must be a string
3165 object, which sets the new name, or @code{None}, which removes any
3166 user-specified thread name.
3169 @defvar InferiorThread.num
3170 The per-inferior number of the thread, as assigned by GDB.
3173 @defvar InferiorThread.global_num
3174 The global ID of the thread, as assigned by GDB. You can use this to
3175 make Python breakpoints thread-specific, for example
3176 (@pxref{python_breakpoint_thread,,The Breakpoint.thread attribute}).
3179 @defvar InferiorThread.ptid
3180 ID of the thread, as assigned by the operating system. This attribute is a
3181 tuple containing three integers. The first is the Process ID (PID); the second
3182 is the Lightweight Process ID (LWPID), and the third is the Thread ID (TID).
3183 Either the LWPID or TID may be 0, which indicates that the operating system
3184 does not use that identifier.
3187 @defvar InferiorThread.inferior
3188 The inferior this thread belongs to. This attribute is represented as
3189 a @code{gdb.Inferior} object. This attribute is not writable.
3192 A @code{gdb.InferiorThread} object has the following methods:
3194 @defun InferiorThread.is_valid ()
3195 Returns @code{True} if the @code{gdb.InferiorThread} object is valid,
3196 @code{False} if not. A @code{gdb.InferiorThread} object will become
3197 invalid if the thread exits, or the inferior that the thread belongs
3198 is deleted. All other @code{gdb.InferiorThread} methods will throw an
3199 exception if it is invalid at the time the method is called.
3202 @defun InferiorThread.switch ()
3203 This changes @value{GDBN}'s currently selected thread to the one represented
3207 @defun InferiorThread.is_stopped ()
3208 Return a Boolean indicating whether the thread is stopped.
3211 @defun InferiorThread.is_running ()
3212 Return a Boolean indicating whether the thread is running.
3215 @defun InferiorThread.is_exited ()
3216 Return a Boolean indicating whether the thread is exited.
3219 @node Recordings In Python
3220 @subsubsection Recordings In Python
3221 @cindex recordings in python
3223 The following recordings-related functions
3224 (@pxref{Process Record and Replay}) are available in the @code{gdb}
3227 @defun gdb.start_recording (@r{[}method@r{]}, @r{[}format@r{]})
3228 Start a recording using the given @var{method} and @var{format}. If
3229 no @var{format} is given, the default format for the recording method
3230 is used. If no @var{method} is given, the default method will be used.
3231 Returns a @code{gdb.Record} object on success. Throw an exception on
3234 The following strings can be passed as @var{method}:
3240 @code{"btrace"}: Possible values for @var{format}: @code{"pt"},
3241 @code{"bts"} or leave out for default format.
3245 @defun gdb.current_recording ()
3246 Access a currently running recording. Return a @code{gdb.Record}
3247 object on success. Return @code{None} if no recording is currently
3251 @defun gdb.stop_recording ()
3252 Stop the current recording. Throw an exception if no recording is
3253 currently active. All record objects become invalid after this call.
3256 A @code{gdb.Record} object has the following attributes:
3258 @defvar Record.method
3259 A string with the current recording method, e.g.@: @code{full} or
3263 @defvar Record.format
3264 A string with the current recording format, e.g.@: @code{bt}, @code{pts} or
3268 @defvar Record.begin
3269 A method specific instruction object representing the first instruction
3274 A method specific instruction object representing the current
3275 instruction, that is not actually part of the recording.
3278 @defvar Record.replay_position
3279 The instruction representing the current replay position. If there is
3280 no replay active, this will be @code{None}.
3283 @defvar Record.instruction_history
3284 A list with all recorded instructions.
3287 @defvar Record.function_call_history
3288 A list with all recorded function call segments.
3291 A @code{gdb.Record} object has the following methods:
3293 @defun Record.goto (instruction)
3294 Move the replay position to the given @var{instruction}.
3297 The common @code{gdb.Instruction} class that recording method specific
3298 instruction objects inherit from, has the following attributes:
3300 @defvar Instruction.pc
3301 An integer representing this instruction's address.
3304 @defvar Instruction.data
3305 A buffer with the raw instruction data. In Python 3, the return value is a
3306 @code{memoryview} object.
3309 @defvar Instruction.decoded
3310 A human readable string with the disassembled instruction.
3313 @defvar Instruction.size
3314 The size of the instruction in bytes.
3317 Additionally @code{gdb.RecordInstruction} has the following attributes:
3319 @defvar RecordInstruction.number
3320 An integer identifying this instruction. @code{number} corresponds to
3321 the numbers seen in @code{record instruction-history}
3322 (@pxref{Process Record and Replay}).
3325 @defvar RecordInstruction.sal
3326 A @code{gdb.Symtab_and_line} object representing the associated symtab
3327 and line of this instruction. May be @code{None} if no debug information is
3331 @defvar RecordInstruction.is_speculative
3332 A boolean indicating whether the instruction was executed speculatively.
3335 If an error occured during recording or decoding a recording, this error is
3336 represented by a @code{gdb.RecordGap} object in the instruction list. It has
3337 the following attributes:
3339 @defvar RecordGap.number
3340 An integer identifying this gap. @code{number} corresponds to the numbers seen
3341 in @code{record instruction-history} (@pxref{Process Record and Replay}).
3344 @defvar RecordGap.error_code
3345 A numerical representation of the reason for the gap. The value is specific to
3346 the current recording method.
3349 @defvar RecordGap.error_string
3350 A human readable string with the reason for the gap.
3353 A @code{gdb.RecordFunctionSegment} object has the following attributes:
3355 @defvar RecordFunctionSegment.number
3356 An integer identifying this function segment. @code{number} corresponds to
3357 the numbers seen in @code{record function-call-history}
3358 (@pxref{Process Record and Replay}).
3361 @defvar RecordFunctionSegment.symbol
3362 A @code{gdb.Symbol} object representing the associated symbol. May be
3363 @code{None} if no debug information is available.
3366 @defvar RecordFunctionSegment.level
3367 An integer representing the function call's stack level. May be
3368 @code{None} if the function call is a gap.
3371 @defvar RecordFunctionSegment.instructions
3372 A list of @code{gdb.RecordInstruction} or @code{gdb.RecordGap} objects
3373 associated with this function call.
3376 @defvar RecordFunctionSegment.up
3377 A @code{gdb.RecordFunctionSegment} object representing the caller's
3378 function segment. If the call has not been recorded, this will be the
3379 function segment to which control returns. If neither the call nor the
3380 return have been recorded, this will be @code{None}.
3383 @defvar RecordFunctionSegment.prev
3384 A @code{gdb.RecordFunctionSegment} object representing the previous
3385 segment of this function call. May be @code{None}.
3388 @defvar RecordFunctionSegment.next
3389 A @code{gdb.RecordFunctionSegment} object representing the next segment of
3390 this function call. May be @code{None}.
3393 The following example demonstrates the usage of these objects and
3394 functions to create a function that will rewind a record to the last
3395 time a function in a different file was executed. This would typically
3396 be used to track the execution of user provided callback functions in a
3397 library which typically are not visible in a back trace.
3401 rec = gdb.current_recording ()
3405 insn = rec.instruction_history
3410 position = insn.index (rec.replay_position)
3414 filename = insn[position].sal.symtab.fullname ()
3418 for i in reversed (insn[:position]):
3420 current = i.sal.symtab.fullname ()
3424 if filename == current:
3431 Another possible application is to write a function that counts the
3432 number of code executions in a given line range. This line range can
3433 contain parts of functions or span across several functions and is not
3434 limited to be contiguous.
3437 def countrange (filename, linerange):
3440 def filter_only (file_name):
3441 for call in gdb.current_recording ().function_call_history:
3443 if file_name in call.symbol.symtab.fullname ():
3448 for c in filter_only (filename):
3449 for i in c.instructions:
3451 if i.sal.line in linerange:
3460 @node Commands In Python
3461 @subsubsection Commands In Python
3463 @cindex commands in python
3464 @cindex python commands
3465 You can implement new @value{GDBN} CLI commands in Python. A CLI
3466 command is implemented using an instance of the @code{gdb.Command}
3467 class, most commonly using a subclass.
3469 @defun Command.__init__ (name, @var{command_class} @r{[}, @var{completer_class} @r{[}, @var{prefix}@r{]]})
3470 The object initializer for @code{Command} registers the new command
3471 with @value{GDBN}. This initializer is normally invoked from the
3472 subclass' own @code{__init__} method.
3474 @var{name} is the name of the command. If @var{name} consists of
3475 multiple words, then the initial words are looked for as prefix
3476 commands. In this case, if one of the prefix commands does not exist,
3477 an exception is raised.
3479 There is no support for multi-line commands.
3481 @var{command_class} should be one of the @samp{COMMAND_} constants
3482 defined below. This argument tells @value{GDBN} how to categorize the
3483 new command in the help system.
3485 @var{completer_class} is an optional argument. If given, it should be
3486 one of the @samp{COMPLETE_} constants defined below. This argument
3487 tells @value{GDBN} how to perform completion for this command. If not
3488 given, @value{GDBN} will attempt to complete using the object's
3489 @code{complete} method (see below); if no such method is found, an
3490 error will occur when completion is attempted.
3492 @var{prefix} is an optional argument. If @code{True}, then the new
3493 command is a prefix command; sub-commands of this command may be
3496 The help text for the new command is taken from the Python
3497 documentation string for the command's class, if there is one. If no
3498 documentation string is provided, the default value ``This command is
3499 not documented.'' is used.
3502 @cindex don't repeat Python command
3503 @defun Command.dont_repeat ()
3504 By default, a @value{GDBN} command is repeated when the user enters a
3505 blank line at the command prompt. A command can suppress this
3506 behavior by invoking the @code{dont_repeat} method. This is similar
3507 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
3510 @defun Command.invoke (argument, from_tty)
3511 This method is called by @value{GDBN} when this command is invoked.
3513 @var{argument} is a string. It is the argument to the command, after
3514 leading and trailing whitespace has been stripped.
3516 @var{from_tty} is a boolean argument. When true, this means that the
3517 command was entered by the user at the terminal; when false it means
3518 that the command came from elsewhere.
3520 If this method throws an exception, it is turned into a @value{GDBN}
3521 @code{error} call. Otherwise, the return value is ignored.
3523 @findex gdb.string_to_argv
3524 To break @var{argument} up into an argv-like string use
3525 @code{gdb.string_to_argv}. This function behaves identically to
3526 @value{GDBN}'s internal argument lexer @code{buildargv}.
3527 It is recommended to use this for consistency.
3528 Arguments are separated by spaces and may be quoted.
3532 print gdb.string_to_argv ("1 2\ \\\"3 '4 \"5' \"6 '7\"")
3533 ['1', '2 "3', '4 "5', "6 '7"]
3538 @cindex completion of Python commands
3539 @defun Command.complete (text, word)
3540 This method is called by @value{GDBN} when the user attempts
3541 completion on this command. All forms of completion are handled by
3542 this method, that is, the @key{TAB} and @key{M-?} key bindings
3543 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
3546 The arguments @var{text} and @var{word} are both strings; @var{text}
3547 holds the complete command line up to the cursor's location, while
3548 @var{word} holds the last word of the command line; this is computed
3549 using a word-breaking heuristic.
3551 The @code{complete} method can return several values:
3554 If the return value is a sequence, the contents of the sequence are
3555 used as the completions. It is up to @code{complete} to ensure that the
3556 contents actually do complete the word. A zero-length sequence is
3557 allowed, it means that there were no completions available. Only
3558 string elements of the sequence are used; other elements in the
3559 sequence are ignored.
3562 If the return value is one of the @samp{COMPLETE_} constants defined
3563 below, then the corresponding @value{GDBN}-internal completion
3564 function is invoked, and its result is used.
3567 All other results are treated as though there were no available
3572 When a new command is registered, it must be declared as a member of
3573 some general class of commands. This is used to classify top-level
3574 commands in the on-line help system; note that prefix commands are not
3575 listed under their own category but rather that of their top-level
3576 command. The available classifications are represented by constants
3577 defined in the @code{gdb} module:
3580 @findex COMMAND_NONE
3581 @findex gdb.COMMAND_NONE
3582 @item gdb.COMMAND_NONE
3583 The command does not belong to any particular class. A command in
3584 this category will not be displayed in any of the help categories.
3586 @findex COMMAND_RUNNING
3587 @findex gdb.COMMAND_RUNNING
3588 @item gdb.COMMAND_RUNNING
3589 The command is related to running the inferior. For example,
3590 @code{start}, @code{step}, and @code{continue} are in this category.
3591 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
3592 commands in this category.
3594 @findex COMMAND_DATA
3595 @findex gdb.COMMAND_DATA
3596 @item gdb.COMMAND_DATA
3597 The command is related to data or variables. For example,
3598 @code{call}, @code{find}, and @code{print} are in this category. Type
3599 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
3602 @findex COMMAND_STACK
3603 @findex gdb.COMMAND_STACK
3604 @item gdb.COMMAND_STACK
3605 The command has to do with manipulation of the stack. For example,
3606 @code{backtrace}, @code{frame}, and @code{return} are in this
3607 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
3608 list of commands in this category.
3610 @findex COMMAND_FILES
3611 @findex gdb.COMMAND_FILES
3612 @item gdb.COMMAND_FILES
3613 This class is used for file-related commands. For example,
3614 @code{file}, @code{list} and @code{section} are in this category.
3615 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
3616 commands in this category.
3618 @findex COMMAND_SUPPORT
3619 @findex gdb.COMMAND_SUPPORT
3620 @item gdb.COMMAND_SUPPORT
3621 This should be used for ``support facilities'', generally meaning
3622 things that are useful to the user when interacting with @value{GDBN},
3623 but not related to the state of the inferior. For example,
3624 @code{help}, @code{make}, and @code{shell} are in this category. Type
3625 @kbd{help support} at the @value{GDBN} prompt to see a list of
3626 commands in this category.
3628 @findex COMMAND_STATUS
3629 @findex gdb.COMMAND_STATUS
3630 @item gdb.COMMAND_STATUS
3631 The command is an @samp{info}-related command, that is, related to the
3632 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
3633 and @code{show} are in this category. Type @kbd{help status} at the
3634 @value{GDBN} prompt to see a list of commands in this category.
3636 @findex COMMAND_BREAKPOINTS
3637 @findex gdb.COMMAND_BREAKPOINTS
3638 @item gdb.COMMAND_BREAKPOINTS
3639 The command has to do with breakpoints. For example, @code{break},
3640 @code{clear}, and @code{delete} are in this category. Type @kbd{help
3641 breakpoints} at the @value{GDBN} prompt to see a list of commands in
3644 @findex COMMAND_TRACEPOINTS
3645 @findex gdb.COMMAND_TRACEPOINTS
3646 @item gdb.COMMAND_TRACEPOINTS
3647 The command has to do with tracepoints. For example, @code{trace},
3648 @code{actions}, and @code{tfind} are in this category. Type
3649 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
3650 commands in this category.
3652 @findex COMMAND_USER
3653 @findex gdb.COMMAND_USER
3654 @item gdb.COMMAND_USER
3655 The command is a general purpose command for the user, and typically
3656 does not fit in one of the other categories.
3657 Type @kbd{help user-defined} at the @value{GDBN} prompt to see
3658 a list of commands in this category, as well as the list of gdb macros
3659 (@pxref{Sequences}).
3661 @findex COMMAND_OBSCURE
3662 @findex gdb.COMMAND_OBSCURE
3663 @item gdb.COMMAND_OBSCURE
3664 The command is only used in unusual circumstances, or is not of
3665 general interest to users. For example, @code{checkpoint},
3666 @code{fork}, and @code{stop} are in this category. Type @kbd{help
3667 obscure} at the @value{GDBN} prompt to see a list of commands in this
3670 @findex COMMAND_MAINTENANCE
3671 @findex gdb.COMMAND_MAINTENANCE
3672 @item gdb.COMMAND_MAINTENANCE
3673 The command is only useful to @value{GDBN} maintainers. The
3674 @code{maintenance} and @code{flushregs} commands are in this category.
3675 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
3676 commands in this category.
3679 A new command can use a predefined completion function, either by
3680 specifying it via an argument at initialization, or by returning it
3681 from the @code{complete} method. These predefined completion
3682 constants are all defined in the @code{gdb} module:
3685 @vindex COMPLETE_NONE
3686 @item gdb.COMPLETE_NONE
3687 This constant means that no completion should be done.
3689 @vindex COMPLETE_FILENAME
3690 @item gdb.COMPLETE_FILENAME
3691 This constant means that filename completion should be performed.
3693 @vindex COMPLETE_LOCATION
3694 @item gdb.COMPLETE_LOCATION
3695 This constant means that location completion should be done.
3696 @xref{Specify Location}.
3698 @vindex COMPLETE_COMMAND
3699 @item gdb.COMPLETE_COMMAND
3700 This constant means that completion should examine @value{GDBN}
3703 @vindex COMPLETE_SYMBOL
3704 @item gdb.COMPLETE_SYMBOL
3705 This constant means that completion should be done using symbol names
3708 @vindex COMPLETE_EXPRESSION
3709 @item gdb.COMPLETE_EXPRESSION
3710 This constant means that completion should be done on expressions.
3711 Often this means completing on symbol names, but some language
3712 parsers also have support for completing on field names.
3715 The following code snippet shows how a trivial CLI command can be
3716 implemented in Python:
3719 class HelloWorld (gdb.Command):
3720 """Greet the whole world."""
3722 def __init__ (self):
3723 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_USER)
3725 def invoke (self, arg, from_tty):
3726 print "Hello, World!"
3731 The last line instantiates the class, and is necessary to trigger the
3732 registration of the command with @value{GDBN}. Depending on how the
3733 Python code is read into @value{GDBN}, you may need to import the
3734 @code{gdb} module explicitly.
3736 @node Parameters In Python
3737 @subsubsection Parameters In Python
3739 @cindex parameters in python
3740 @cindex python parameters
3741 @tindex gdb.Parameter
3743 You can implement new @value{GDBN} parameters using Python. A new
3744 parameter is implemented as an instance of the @code{gdb.Parameter}
3747 Parameters are exposed to the user via the @code{set} and
3748 @code{show} commands. @xref{Help}.
3750 There are many parameters that already exist and can be set in
3751 @value{GDBN}. Two examples are: @code{set follow fork} and
3752 @code{set charset}. Setting these parameters influences certain
3753 behavior in @value{GDBN}. Similarly, you can define parameters that
3754 can be used to influence behavior in custom Python scripts and commands.
3756 @defun Parameter.__init__ (name, @var{command-class}, @var{parameter-class} @r{[}, @var{enum-sequence}@r{]})
3757 The object initializer for @code{Parameter} registers the new
3758 parameter with @value{GDBN}. This initializer is normally invoked
3759 from the subclass' own @code{__init__} method.
3761 @var{name} is the name of the new parameter. If @var{name} consists
3762 of multiple words, then the initial words are looked for as prefix
3763 parameters. An example of this can be illustrated with the
3764 @code{set print} set of parameters. If @var{name} is
3765 @code{print foo}, then @code{print} will be searched as the prefix
3766 parameter. In this case the parameter can subsequently be accessed in
3767 @value{GDBN} as @code{set print foo}.
3769 If @var{name} consists of multiple words, and no prefix parameter group
3770 can be found, an exception is raised.
3772 @var{command-class} should be one of the @samp{COMMAND_} constants
3773 (@pxref{Commands In Python}). This argument tells @value{GDBN} how to
3774 categorize the new parameter in the help system.
3776 @var{parameter-class} should be one of the @samp{PARAM_} constants
3777 defined below. This argument tells @value{GDBN} the type of the new
3778 parameter; this information is used for input validation and
3781 If @var{parameter-class} is @code{PARAM_ENUM}, then
3782 @var{enum-sequence} must be a sequence of strings. These strings
3783 represent the possible values for the parameter.
3785 If @var{parameter-class} is not @code{PARAM_ENUM}, then the presence
3786 of a fourth argument will cause an exception to be thrown.
3788 The help text for the new parameter is taken from the Python
3789 documentation string for the parameter's class, if there is one. If
3790 there is no documentation string, a default value is used.
3793 @defvar Parameter.set_doc
3794 If this attribute exists, and is a string, then its value is used as
3795 the help text for this parameter's @code{set} command. The value is
3796 examined when @code{Parameter.__init__} is invoked; subsequent changes
3800 @defvar Parameter.show_doc
3801 If this attribute exists, and is a string, then its value is used as
3802 the help text for this parameter's @code{show} command. The value is
3803 examined when @code{Parameter.__init__} is invoked; subsequent changes
3807 @defvar Parameter.value
3808 The @code{value} attribute holds the underlying value of the
3809 parameter. It can be read and assigned to just as any other
3810 attribute. @value{GDBN} does validation when assignments are made.
3813 There are two methods that may be implemented in any @code{Parameter}
3816 @defun Parameter.get_set_string (self)
3817 If this method exists, @value{GDBN} will call it when a
3818 @var{parameter}'s value has been changed via the @code{set} API (for
3819 example, @kbd{set foo off}). The @code{value} attribute has already
3820 been populated with the new value and may be used in output. This
3821 method must return a string. If the returned string is not empty,
3822 @value{GDBN} will present it to the user.
3825 @defun Parameter.get_show_string (self, svalue)
3826 @value{GDBN} will call this method when a @var{parameter}'s
3827 @code{show} API has been invoked (for example, @kbd{show foo}). The
3828 argument @code{svalue} receives the string representation of the
3829 current value. This method must return a string.
3832 When a new parameter is defined, its type must be specified. The
3833 available types are represented by constants defined in the @code{gdb}
3837 @findex PARAM_BOOLEAN
3838 @findex gdb.PARAM_BOOLEAN
3839 @item gdb.PARAM_BOOLEAN
3840 The value is a plain boolean. The Python boolean values, @code{True}
3841 and @code{False} are the only valid values.
3843 @findex PARAM_AUTO_BOOLEAN
3844 @findex gdb.PARAM_AUTO_BOOLEAN
3845 @item gdb.PARAM_AUTO_BOOLEAN
3846 The value has three possible states: true, false, and @samp{auto}. In
3847 Python, true and false are represented using boolean constants, and
3848 @samp{auto} is represented using @code{None}.
3850 @findex PARAM_UINTEGER
3851 @findex gdb.PARAM_UINTEGER
3852 @item gdb.PARAM_UINTEGER
3853 The value is an unsigned integer. The value of 0 should be
3854 interpreted to mean ``unlimited''.
3856 @findex PARAM_INTEGER
3857 @findex gdb.PARAM_INTEGER
3858 @item gdb.PARAM_INTEGER
3859 The value is a signed integer. The value of 0 should be interpreted
3860 to mean ``unlimited''.
3862 @findex PARAM_STRING
3863 @findex gdb.PARAM_STRING
3864 @item gdb.PARAM_STRING
3865 The value is a string. When the user modifies the string, any escape
3866 sequences, such as @samp{\t}, @samp{\f}, and octal escapes, are
3867 translated into corresponding characters and encoded into the current
3870 @findex PARAM_STRING_NOESCAPE
3871 @findex gdb.PARAM_STRING_NOESCAPE
3872 @item gdb.PARAM_STRING_NOESCAPE
3873 The value is a string. When the user modifies the string, escapes are
3874 passed through untranslated.
3876 @findex PARAM_OPTIONAL_FILENAME
3877 @findex gdb.PARAM_OPTIONAL_FILENAME
3878 @item gdb.PARAM_OPTIONAL_FILENAME
3879 The value is a either a filename (a string), or @code{None}.
3881 @findex PARAM_FILENAME
3882 @findex gdb.PARAM_FILENAME
3883 @item gdb.PARAM_FILENAME
3884 The value is a filename. This is just like
3885 @code{PARAM_STRING_NOESCAPE}, but uses file names for completion.
3887 @findex PARAM_ZINTEGER
3888 @findex gdb.PARAM_ZINTEGER
3889 @item gdb.PARAM_ZINTEGER
3890 The value is an integer. This is like @code{PARAM_INTEGER}, except 0
3891 is interpreted as itself.
3893 @findex PARAM_ZUINTEGER
3894 @findex gdb.PARAM_ZUINTEGER
3895 @item gdb.PARAM_ZUINTEGER
3896 The value is an unsigned integer. This is like @code{PARAM_INTEGER},
3897 except 0 is interpreted as itself, and the value cannot be negative.
3899 @findex PARAM_ZUINTEGER_UNLIMITED
3900 @findex gdb.PARAM_ZUINTEGER_UNLIMITED
3901 @item gdb.PARAM_ZUINTEGER_UNLIMITED
3902 The value is a signed integer. This is like @code{PARAM_ZUINTEGER},
3903 except the special value -1 should be interpreted to mean
3904 ``unlimited''. Other negative values are not allowed.
3907 @findex gdb.PARAM_ENUM
3908 @item gdb.PARAM_ENUM
3909 The value is a string, which must be one of a collection string
3910 constants provided when the parameter is created.
3913 @node Functions In Python
3914 @subsubsection Writing new convenience functions
3916 @cindex writing convenience functions
3917 @cindex convenience functions in python
3918 @cindex python convenience functions
3919 @tindex gdb.Function
3921 You can implement new convenience functions (@pxref{Convenience Vars})
3922 in Python. A convenience function is an instance of a subclass of the
3923 class @code{gdb.Function}.
3925 @defun Function.__init__ (name)
3926 The initializer for @code{Function} registers the new function with
3927 @value{GDBN}. The argument @var{name} is the name of the function,
3928 a string. The function will be visible to the user as a convenience
3929 variable of type @code{internal function}, whose name is the same as
3930 the given @var{name}.
3932 The documentation for the new function is taken from the documentation
3933 string for the new class.
3936 @defun Function.invoke (@var{*args})
3937 When a convenience function is evaluated, its arguments are converted
3938 to instances of @code{gdb.Value}, and then the function's
3939 @code{invoke} method is called. Note that @value{GDBN} does not
3940 predetermine the arity of convenience functions. Instead, all
3941 available arguments are passed to @code{invoke}, following the
3942 standard Python calling convention. In particular, a convenience
3943 function can have default values for parameters without ill effect.
3945 The return value of this method is used as its value in the enclosing
3946 expression. If an ordinary Python value is returned, it is converted
3947 to a @code{gdb.Value} following the usual rules.
3950 The following code snippet shows how a trivial convenience function can
3951 be implemented in Python:
3954 class Greet (gdb.Function):
3955 """Return string to greet someone.
3956 Takes a name as argument."""
3958 def __init__ (self):
3959 super (Greet, self).__init__ ("greet")
3961 def invoke (self, name):
3962 return "Hello, %s!" % name.string ()
3967 The last line instantiates the class, and is necessary to trigger the
3968 registration of the function with @value{GDBN}. Depending on how the
3969 Python code is read into @value{GDBN}, you may need to import the
3970 @code{gdb} module explicitly.
3972 Now you can use the function in an expression:
3975 (gdb) print $greet("Bob")
3979 @node Progspaces In Python
3980 @subsubsection Program Spaces In Python
3982 @cindex progspaces in python
3983 @tindex gdb.Progspace
3985 A program space, or @dfn{progspace}, represents a symbolic view
3986 of an address space.
3987 It consists of all of the objfiles of the program.
3988 @xref{Objfiles In Python}.
3989 @xref{Inferiors and Programs, program spaces}, for more details
3990 about program spaces.
3992 The following progspace-related functions are available in the
3995 @findex gdb.current_progspace
3996 @defun gdb.current_progspace ()
3997 This function returns the program space of the currently selected inferior.
3998 @xref{Inferiors and Programs}.
4001 @findex gdb.progspaces
4002 @defun gdb.progspaces ()
4003 Return a sequence of all the progspaces currently known to @value{GDBN}.
4006 Each progspace is represented by an instance of the @code{gdb.Progspace}
4009 @defvar Progspace.filename
4010 The file name of the progspace as a string.
4013 @defvar Progspace.pretty_printers
4014 The @code{pretty_printers} attribute is a list of functions. It is
4015 used to look up pretty-printers. A @code{Value} is passed to each
4016 function in order; if the function returns @code{None}, then the
4017 search continues. Otherwise, the return value should be an object
4018 which is used to format the value. @xref{Pretty Printing API}, for more
4022 @defvar Progspace.type_printers
4023 The @code{type_printers} attribute is a list of type printer objects.
4024 @xref{Type Printing API}, for more information.
4027 @defvar Progspace.frame_filters
4028 The @code{frame_filters} attribute is a dictionary of frame filter
4029 objects. @xref{Frame Filter API}, for more information.
4032 One may add arbitrary attributes to @code{gdb.Progspace} objects
4033 in the usual Python way.
4034 This is useful if, for example, one needs to do some extra record keeping
4035 associated with the program space.
4037 In this contrived example, we want to perform some processing when
4038 an objfile with a certain symbol is loaded, but we only want to do
4039 this once because it is expensive. To achieve this we record the results
4040 with the program space because we can't predict when the desired objfile
4045 def clear_objfiles_handler(event):
4046 event.progspace.expensive_computation = None
4047 def expensive(symbol):
4048 """A mock routine to perform an "expensive" computation on symbol."""
4049 print "Computing the answer to the ultimate question ..."
4051 def new_objfile_handler(event):
4052 objfile = event.new_objfile
4053 progspace = objfile.progspace
4054 if not hasattr(progspace, 'expensive_computation') or \
4055 progspace.expensive_computation is None:
4056 # We use 'main' for the symbol to keep the example simple.
4057 # Note: There's no current way to constrain the lookup
4059 symbol = gdb.lookup_global_symbol('main')
4060 if symbol is not None:
4061 progspace.expensive_computation = expensive(symbol)
4062 gdb.events.clear_objfiles.connect(clear_objfiles_handler)
4063 gdb.events.new_objfile.connect(new_objfile_handler)
4065 (gdb) file /tmp/hello
4066 Reading symbols from /tmp/hello...done.
4067 Computing the answer to the ultimate question ...
4068 (gdb) python print gdb.current_progspace().expensive_computation
4071 Starting program: /tmp/hello
4073 [Inferior 1 (process 4242) exited normally]
4076 @node Objfiles In Python
4077 @subsubsection Objfiles In Python
4079 @cindex objfiles in python
4082 @value{GDBN} loads symbols for an inferior from various
4083 symbol-containing files (@pxref{Files}). These include the primary
4084 executable file, any shared libraries used by the inferior, and any
4085 separate debug info files (@pxref{Separate Debug Files}).
4086 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
4088 The following objfile-related functions are available in the
4091 @findex gdb.current_objfile
4092 @defun gdb.current_objfile ()
4093 When auto-loading a Python script (@pxref{Python Auto-loading}), @value{GDBN}
4094 sets the ``current objfile'' to the corresponding objfile. This
4095 function returns the current objfile. If there is no current objfile,
4096 this function returns @code{None}.
4099 @findex gdb.objfiles
4100 @defun gdb.objfiles ()
4101 Return a sequence of all the objfiles current known to @value{GDBN}.
4102 @xref{Objfiles In Python}.
4105 @findex gdb.lookup_objfile
4106 @defun gdb.lookup_objfile (name @r{[}, by_build_id{]})
4107 Look up @var{name}, a file name or build ID, in the list of objfiles
4108 for the current program space (@pxref{Progspaces In Python}).
4109 If the objfile is not found throw the Python @code{ValueError} exception.
4111 If @var{name} is a relative file name, then it will match any
4112 source file name with the same trailing components. For example, if
4113 @var{name} is @samp{gcc/expr.c}, then it will match source file
4114 name of @file{/build/trunk/gcc/expr.c}, but not
4115 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
4117 If @var{by_build_id} is provided and is @code{True} then @var{name}
4118 is the build ID of the objfile. Otherwise, @var{name} is a file name.
4119 This is supported only on some operating systems, notably those which use
4120 the ELF format for binary files and the @sc{gnu} Binutils. For more details
4121 about this feature, see the description of the @option{--build-id}
4122 command-line option in @ref{Options, , Command Line Options, ld,
4126 Each objfile is represented by an instance of the @code{gdb.Objfile}
4129 @defvar Objfile.filename
4130 The file name of the objfile as a string, with symbolic links resolved.
4132 The value is @code{None} if the objfile is no longer valid.
4133 See the @code{gdb.Objfile.is_valid} method, described below.
4136 @defvar Objfile.username
4137 The file name of the objfile as specified by the user as a string.
4139 The value is @code{None} if the objfile is no longer valid.
4140 See the @code{gdb.Objfile.is_valid} method, described below.
4143 @defvar Objfile.owner
4144 For separate debug info objfiles this is the corresponding @code{gdb.Objfile}
4145 object that debug info is being provided for.
4146 Otherwise this is @code{None}.
4147 Separate debug info objfiles are added with the
4148 @code{gdb.Objfile.add_separate_debug_file} method, described below.
4151 @defvar Objfile.build_id
4152 The build ID of the objfile as a string.
4153 If the objfile does not have a build ID then the value is @code{None}.
4155 This is supported only on some operating systems, notably those which use
4156 the ELF format for binary files and the @sc{gnu} Binutils. For more details
4157 about this feature, see the description of the @option{--build-id}
4158 command-line option in @ref{Options, , Command Line Options, ld,
4162 @defvar Objfile.progspace
4163 The containing program space of the objfile as a @code{gdb.Progspace}
4164 object. @xref{Progspaces In Python}.
4167 @defvar Objfile.pretty_printers
4168 The @code{pretty_printers} attribute is a list of functions. It is
4169 used to look up pretty-printers. A @code{Value} is passed to each
4170 function in order; if the function returns @code{None}, then the
4171 search continues. Otherwise, the return value should be an object
4172 which is used to format the value. @xref{Pretty Printing API}, for more
4176 @defvar Objfile.type_printers
4177 The @code{type_printers} attribute is a list of type printer objects.
4178 @xref{Type Printing API}, for more information.
4181 @defvar Objfile.frame_filters
4182 The @code{frame_filters} attribute is a dictionary of frame filter
4183 objects. @xref{Frame Filter API}, for more information.
4186 One may add arbitrary attributes to @code{gdb.Objfile} objects
4187 in the usual Python way.
4188 This is useful if, for example, one needs to do some extra record keeping
4189 associated with the objfile.
4191 In this contrived example we record the time when @value{GDBN}
4197 def new_objfile_handler(event):
4198 # Set the time_loaded attribute of the new objfile.
4199 event.new_objfile.time_loaded = datetime.datetime.today()
4200 gdb.events.new_objfile.connect(new_objfile_handler)
4203 Reading symbols from ./hello...done.
4204 (gdb) python print gdb.objfiles()[0].time_loaded
4205 2014-10-09 11:41:36.770345
4208 A @code{gdb.Objfile} object has the following methods:
4210 @defun Objfile.is_valid ()
4211 Returns @code{True} if the @code{gdb.Objfile} object is valid,
4212 @code{False} if not. A @code{gdb.Objfile} object can become invalid
4213 if the object file it refers to is not loaded in @value{GDBN} any
4214 longer. All other @code{gdb.Objfile} methods will throw an exception
4215 if it is invalid at the time the method is called.
4218 @defun Objfile.add_separate_debug_file (file)
4219 Add @var{file} to the list of files that @value{GDBN} will search for
4220 debug information for the objfile.
4221 This is useful when the debug info has been removed from the program
4222 and stored in a separate file. @value{GDBN} has built-in support for
4223 finding separate debug info files (@pxref{Separate Debug Files}), but if
4224 the file doesn't live in one of the standard places that @value{GDBN}
4225 searches then this function can be used to add a debug info file
4226 from a different place.
4229 @node Frames In Python
4230 @subsubsection Accessing inferior stack frames from Python
4232 @cindex frames in python
4233 When the debugged program stops, @value{GDBN} is able to analyze its call
4234 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
4235 represents a frame in the stack. A @code{gdb.Frame} object is only valid
4236 while its corresponding frame exists in the inferior's stack. If you try
4237 to use an invalid frame object, @value{GDBN} will throw a @code{gdb.error}
4238 exception (@pxref{Exception Handling}).
4240 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
4244 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
4248 The following frame-related functions are available in the @code{gdb} module:
4250 @findex gdb.selected_frame
4251 @defun gdb.selected_frame ()
4252 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
4255 @findex gdb.newest_frame
4256 @defun gdb.newest_frame ()
4257 Return the newest frame object for the selected thread.
4260 @defun gdb.frame_stop_reason_string (reason)
4261 Return a string explaining the reason why @value{GDBN} stopped unwinding
4262 frames, as expressed by the given @var{reason} code (an integer, see the
4263 @code{unwind_stop_reason} method further down in this section).
4266 @findex gdb.invalidate_cached_frames
4267 @defun gdb.invalidate_cached_frames
4268 @value{GDBN} internally keeps a cache of the frames that have been
4269 unwound. This function invalidates this cache.
4271 This function should not generally be called by ordinary Python code.
4272 It is documented for the sake of completeness.
4275 A @code{gdb.Frame} object has the following methods:
4277 @defun Frame.is_valid ()
4278 Returns true if the @code{gdb.Frame} object is valid, false if not.
4279 A frame object can become invalid if the frame it refers to doesn't
4280 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
4281 an exception if it is invalid at the time the method is called.
4284 @defun Frame.name ()
4285 Returns the function name of the frame, or @code{None} if it can't be
4289 @defun Frame.architecture ()
4290 Returns the @code{gdb.Architecture} object corresponding to the frame's
4291 architecture. @xref{Architectures In Python}.
4294 @defun Frame.type ()
4295 Returns the type of the frame. The value can be one of:
4297 @item gdb.NORMAL_FRAME
4298 An ordinary stack frame.
4300 @item gdb.DUMMY_FRAME
4301 A fake stack frame that was created by @value{GDBN} when performing an
4302 inferior function call.
4304 @item gdb.INLINE_FRAME
4305 A frame representing an inlined function. The function was inlined
4306 into a @code{gdb.NORMAL_FRAME} that is older than this one.
4308 @item gdb.TAILCALL_FRAME
4309 A frame representing a tail call. @xref{Tail Call Frames}.
4311 @item gdb.SIGTRAMP_FRAME
4312 A signal trampoline frame. This is the frame created by the OS when
4313 it calls into a signal handler.
4315 @item gdb.ARCH_FRAME
4316 A fake stack frame representing a cross-architecture call.
4318 @item gdb.SENTINEL_FRAME
4319 This is like @code{gdb.NORMAL_FRAME}, but it is only used for the
4324 @defun Frame.unwind_stop_reason ()
4325 Return an integer representing the reason why it's not possible to find
4326 more frames toward the outermost frame. Use
4327 @code{gdb.frame_stop_reason_string} to convert the value returned by this
4328 function to a string. The value can be one of:
4331 @item gdb.FRAME_UNWIND_NO_REASON
4332 No particular reason (older frames should be available).
4334 @item gdb.FRAME_UNWIND_NULL_ID
4335 The previous frame's analyzer returns an invalid result. This is no
4336 longer used by @value{GDBN}, and is kept only for backward
4339 @item gdb.FRAME_UNWIND_OUTERMOST
4340 This frame is the outermost.
4342 @item gdb.FRAME_UNWIND_UNAVAILABLE
4343 Cannot unwind further, because that would require knowing the
4344 values of registers or memory that have not been collected.
4346 @item gdb.FRAME_UNWIND_INNER_ID
4347 This frame ID looks like it ought to belong to a NEXT frame,
4348 but we got it for a PREV frame. Normally, this is a sign of
4349 unwinder failure. It could also indicate stack corruption.
4351 @item gdb.FRAME_UNWIND_SAME_ID
4352 This frame has the same ID as the previous one. That means
4353 that unwinding further would almost certainly give us another
4354 frame with exactly the same ID, so break the chain. Normally,
4355 this is a sign of unwinder failure. It could also indicate
4358 @item gdb.FRAME_UNWIND_NO_SAVED_PC
4359 The frame unwinder did not find any saved PC, but we needed
4360 one to unwind further.
4362 @item gdb.FRAME_UNWIND_MEMORY_ERROR
4363 The frame unwinder caused an error while trying to access memory.
4365 @item gdb.FRAME_UNWIND_FIRST_ERROR
4366 Any stop reason greater or equal to this value indicates some kind
4367 of error. This special value facilitates writing code that tests
4368 for errors in unwinding in a way that will work correctly even if
4369 the list of the other values is modified in future @value{GDBN}
4370 versions. Using it, you could write:
4372 reason = gdb.selected_frame().unwind_stop_reason ()
4373 reason_str = gdb.frame_stop_reason_string (reason)
4374 if reason >= gdb.FRAME_UNWIND_FIRST_ERROR:
4375 print "An error occured: %s" % reason_str
4382 Returns the frame's resume address.
4385 @defun Frame.block ()
4386 Return the frame's code block. @xref{Blocks In Python}. If the frame
4387 does not have a block -- for example, if there is no debugging
4388 information for the code in question -- then this will throw an
4392 @defun Frame.function ()
4393 Return the symbol for the function corresponding to this frame.
4394 @xref{Symbols In Python}.
4397 @defun Frame.older ()
4398 Return the frame that called this frame.
4401 @defun Frame.newer ()
4402 Return the frame called by this frame.
4405 @defun Frame.find_sal ()
4406 Return the frame's symtab and line object.
4407 @xref{Symbol Tables In Python}.
4410 @defun Frame.read_register (register)
4411 Return the value of @var{register} in this frame. The @var{register}
4412 argument must be a string (e.g., @code{'sp'} or @code{'rax'}).
4413 Returns a @code{Gdb.Value} object. Throws an exception if @var{register}
4417 @defun Frame.read_var (variable @r{[}, block@r{]})
4418 Return the value of @var{variable} in this frame. If the optional
4419 argument @var{block} is provided, search for the variable from that
4420 block; otherwise start at the frame's current block (which is
4421 determined by the frame's current program counter). The @var{variable}
4422 argument must be a string or a @code{gdb.Symbol} object; @var{block} must be a
4423 @code{gdb.Block} object.
4426 @defun Frame.select ()
4427 Set this frame to be the selected frame. @xref{Stack, ,Examining the
4431 @node Blocks In Python
4432 @subsubsection Accessing blocks from Python
4434 @cindex blocks in python
4437 In @value{GDBN}, symbols are stored in blocks. A block corresponds
4438 roughly to a scope in the source code. Blocks are organized
4439 hierarchically, and are represented individually in Python as a
4440 @code{gdb.Block}. Blocks rely on debugging information being
4443 A frame has a block. Please see @ref{Frames In Python}, for a more
4444 in-depth discussion of frames.
4446 The outermost block is known as the @dfn{global block}. The global
4447 block typically holds public global variables and functions.
4449 The block nested just inside the global block is the @dfn{static
4450 block}. The static block typically holds file-scoped variables and
4453 @value{GDBN} provides a method to get a block's superblock, but there
4454 is currently no way to examine the sub-blocks of a block, or to
4455 iterate over all the blocks in a symbol table (@pxref{Symbol Tables In
4458 Here is a short example that should help explain blocks:
4461 /* This is in the global block. */
4464 /* This is in the static block. */
4465 static int file_scope;
4467 /* 'function' is in the global block, and 'argument' is
4468 in a block nested inside of 'function'. */
4469 int function (int argument)
4471 /* 'local' is in a block inside 'function'. It may or may
4472 not be in the same block as 'argument'. */
4476 /* 'inner' is in a block whose superblock is the one holding
4480 /* If this call is expanded by the compiler, you may see
4481 a nested block here whose function is 'inline_function'
4482 and whose superblock is the one holding 'inner'. */
4488 A @code{gdb.Block} is iterable. The iterator returns the symbols
4489 (@pxref{Symbols In Python}) local to the block. Python programs
4490 should not assume that a specific block object will always contain a
4491 given symbol, since changes in @value{GDBN} features and
4492 infrastructure may cause symbols move across blocks in a symbol
4495 The following block-related functions are available in the @code{gdb}
4498 @findex gdb.block_for_pc
4499 @defun gdb.block_for_pc (pc)
4500 Return the innermost @code{gdb.Block} containing the given @var{pc}
4501 value. If the block cannot be found for the @var{pc} value specified,
4502 the function will return @code{None}.
4505 A @code{gdb.Block} object has the following methods:
4507 @defun Block.is_valid ()
4508 Returns @code{True} if the @code{gdb.Block} object is valid,
4509 @code{False} if not. A block object can become invalid if the block it
4510 refers to doesn't exist anymore in the inferior. All other
4511 @code{gdb.Block} methods will throw an exception if it is invalid at
4512 the time the method is called. The block's validity is also checked
4513 during iteration over symbols of the block.
4516 A @code{gdb.Block} object has the following attributes:
4519 The start address of the block. This attribute is not writable.
4523 One past the last address that appears in the block. This attribute
4527 @defvar Block.function
4528 The name of the block represented as a @code{gdb.Symbol}. If the
4529 block is not named, then this attribute holds @code{None}. This
4530 attribute is not writable.
4532 For ordinary function blocks, the superblock is the static block.
4533 However, you should note that it is possible for a function block to
4534 have a superblock that is not the static block -- for instance this
4535 happens for an inlined function.
4538 @defvar Block.superblock
4539 The block containing this block. If this parent block does not exist,
4540 this attribute holds @code{None}. This attribute is not writable.
4543 @defvar Block.global_block
4544 The global block associated with this block. This attribute is not
4548 @defvar Block.static_block
4549 The static block associated with this block. This attribute is not
4553 @defvar Block.is_global
4554 @code{True} if the @code{gdb.Block} object is a global block,
4555 @code{False} if not. This attribute is not
4559 @defvar Block.is_static
4560 @code{True} if the @code{gdb.Block} object is a static block,
4561 @code{False} if not. This attribute is not writable.
4564 @node Symbols In Python
4565 @subsubsection Python representation of Symbols
4567 @cindex symbols in python
4570 @value{GDBN} represents every variable, function and type as an
4571 entry in a symbol table. @xref{Symbols, ,Examining the Symbol Table}.
4572 Similarly, Python represents these symbols in @value{GDBN} with the
4573 @code{gdb.Symbol} object.
4575 The following symbol-related functions are available in the @code{gdb}
4578 @findex gdb.lookup_symbol
4579 @defun gdb.lookup_symbol (name @r{[}, block @r{[}, domain@r{]]})
4580 This function searches for a symbol by name. The search scope can be
4581 restricted to the parameters defined in the optional domain and block
4584 @var{name} is the name of the symbol. It must be a string. The
4585 optional @var{block} argument restricts the search to symbols visible
4586 in that @var{block}. The @var{block} argument must be a
4587 @code{gdb.Block} object. If omitted, the block for the current frame
4588 is used. The optional @var{domain} argument restricts
4589 the search to the domain type. The @var{domain} argument must be a
4590 domain constant defined in the @code{gdb} module and described later
4593 The result is a tuple of two elements.
4594 The first element is a @code{gdb.Symbol} object or @code{None} if the symbol
4596 If the symbol is found, the second element is @code{True} if the symbol
4597 is a field of a method's object (e.g., @code{this} in C@t{++}),
4598 otherwise it is @code{False}.
4599 If the symbol is not found, the second element is @code{False}.
4602 @findex gdb.lookup_global_symbol
4603 @defun gdb.lookup_global_symbol (name @r{[}, domain@r{]})
4604 This function searches for a global symbol by name.
4605 The search scope can be restricted to by the domain argument.
4607 @var{name} is the name of the symbol. It must be a string.
4608 The optional @var{domain} argument restricts the search to the domain type.
4609 The @var{domain} argument must be a domain constant defined in the @code{gdb}
4610 module and described later in this chapter.
4612 The result is a @code{gdb.Symbol} object or @code{None} if the symbol
4616 A @code{gdb.Symbol} object has the following attributes:
4619 The type of the symbol or @code{None} if no type is recorded.
4620 This attribute is represented as a @code{gdb.Type} object.
4621 @xref{Types In Python}. This attribute is not writable.
4624 @defvar Symbol.symtab
4625 The symbol table in which the symbol appears. This attribute is
4626 represented as a @code{gdb.Symtab} object. @xref{Symbol Tables In
4627 Python}. This attribute is not writable.
4631 The line number in the source code at which the symbol was defined.
4636 The name of the symbol as a string. This attribute is not writable.
4639 @defvar Symbol.linkage_name
4640 The name of the symbol, as used by the linker (i.e., may be mangled).
4641 This attribute is not writable.
4644 @defvar Symbol.print_name
4645 The name of the symbol in a form suitable for output. This is either
4646 @code{name} or @code{linkage_name}, depending on whether the user
4647 asked @value{GDBN} to display demangled or mangled names.
4650 @defvar Symbol.addr_class
4651 The address class of the symbol. This classifies how to find the value
4652 of a symbol. Each address class is a constant defined in the
4653 @code{gdb} module and described later in this chapter.
4656 @defvar Symbol.needs_frame
4657 This is @code{True} if evaluating this symbol's value requires a frame
4658 (@pxref{Frames In Python}) and @code{False} otherwise. Typically,
4659 local variables will require a frame, but other symbols will not.
4662 @defvar Symbol.is_argument
4663 @code{True} if the symbol is an argument of a function.
4666 @defvar Symbol.is_constant
4667 @code{True} if the symbol is a constant.
4670 @defvar Symbol.is_function
4671 @code{True} if the symbol is a function or a method.
4674 @defvar Symbol.is_variable
4675 @code{True} if the symbol is a variable.
4678 A @code{gdb.Symbol} object has the following methods:
4680 @defun Symbol.is_valid ()
4681 Returns @code{True} if the @code{gdb.Symbol} object is valid,
4682 @code{False} if not. A @code{gdb.Symbol} object can become invalid if
4683 the symbol it refers to does not exist in @value{GDBN} any longer.
4684 All other @code{gdb.Symbol} methods will throw an exception if it is
4685 invalid at the time the method is called.
4688 @defun Symbol.value (@r{[}frame@r{]})
4689 Compute the value of the symbol, as a @code{gdb.Value}. For
4690 functions, this computes the address of the function, cast to the
4691 appropriate type. If the symbol requires a frame in order to compute
4692 its value, then @var{frame} must be given. If @var{frame} is not
4693 given, or if @var{frame} is invalid, then this method will throw an
4697 The available domain categories in @code{gdb.Symbol} are represented
4698 as constants in the @code{gdb} module:
4701 @vindex SYMBOL_UNDEF_DOMAIN
4702 @item gdb.SYMBOL_UNDEF_DOMAIN
4703 This is used when a domain has not been discovered or none of the
4704 following domains apply. This usually indicates an error either
4705 in the symbol information or in @value{GDBN}'s handling of symbols.
4707 @vindex SYMBOL_VAR_DOMAIN
4708 @item gdb.SYMBOL_VAR_DOMAIN
4709 This domain contains variables, function names, typedef names and enum
4712 @vindex SYMBOL_STRUCT_DOMAIN
4713 @item gdb.SYMBOL_STRUCT_DOMAIN
4714 This domain holds struct, union and enum type names.
4716 @vindex SYMBOL_LABEL_DOMAIN
4717 @item gdb.SYMBOL_LABEL_DOMAIN
4718 This domain contains names of labels (for gotos).
4720 @vindex SYMBOL_VARIABLES_DOMAIN
4721 @item gdb.SYMBOL_VARIABLES_DOMAIN
4722 This domain holds a subset of the @code{SYMBOLS_VAR_DOMAIN}; it
4723 contains everything minus functions and types.
4725 @vindex SYMBOL_FUNCTIONS_DOMAIN
4726 @item gdb.SYMBOL_FUNCTIONS_DOMAIN
4727 This domain contains all functions.
4729 @vindex SYMBOL_TYPES_DOMAIN
4730 @item gdb.SYMBOL_TYPES_DOMAIN
4731 This domain contains all types.
4734 The available address class categories in @code{gdb.Symbol} are represented
4735 as constants in the @code{gdb} module:
4738 @vindex SYMBOL_LOC_UNDEF
4739 @item gdb.SYMBOL_LOC_UNDEF
4740 If this is returned by address class, it indicates an error either in
4741 the symbol information or in @value{GDBN}'s handling of symbols.
4743 @vindex SYMBOL_LOC_CONST
4744 @item gdb.SYMBOL_LOC_CONST
4745 Value is constant int.
4747 @vindex SYMBOL_LOC_STATIC
4748 @item gdb.SYMBOL_LOC_STATIC
4749 Value is at a fixed address.
4751 @vindex SYMBOL_LOC_REGISTER
4752 @item gdb.SYMBOL_LOC_REGISTER
4753 Value is in a register.
4755 @vindex SYMBOL_LOC_ARG
4756 @item gdb.SYMBOL_LOC_ARG
4757 Value is an argument. This value is at the offset stored within the
4758 symbol inside the frame's argument list.
4760 @vindex SYMBOL_LOC_REF_ARG
4761 @item gdb.SYMBOL_LOC_REF_ARG
4762 Value address is stored in the frame's argument list. Just like
4763 @code{LOC_ARG} except that the value's address is stored at the
4764 offset, not the value itself.
4766 @vindex SYMBOL_LOC_REGPARM_ADDR
4767 @item gdb.SYMBOL_LOC_REGPARM_ADDR
4768 Value is a specified register. Just like @code{LOC_REGISTER} except
4769 the register holds the address of the argument instead of the argument
4772 @vindex SYMBOL_LOC_LOCAL
4773 @item gdb.SYMBOL_LOC_LOCAL
4774 Value is a local variable.
4776 @vindex SYMBOL_LOC_TYPEDEF
4777 @item gdb.SYMBOL_LOC_TYPEDEF
4778 Value not used. Symbols in the domain @code{SYMBOL_STRUCT_DOMAIN} all
4781 @vindex SYMBOL_LOC_BLOCK
4782 @item gdb.SYMBOL_LOC_BLOCK
4785 @vindex SYMBOL_LOC_CONST_BYTES
4786 @item gdb.SYMBOL_LOC_CONST_BYTES
4787 Value is a byte-sequence.
4789 @vindex SYMBOL_LOC_UNRESOLVED
4790 @item gdb.SYMBOL_LOC_UNRESOLVED
4791 Value is at a fixed address, but the address of the variable has to be
4792 determined from the minimal symbol table whenever the variable is
4795 @vindex SYMBOL_LOC_OPTIMIZED_OUT
4796 @item gdb.SYMBOL_LOC_OPTIMIZED_OUT
4797 The value does not actually exist in the program.
4799 @vindex SYMBOL_LOC_COMPUTED
4800 @item gdb.SYMBOL_LOC_COMPUTED
4801 The value's address is a computed location.
4804 @node Symbol Tables In Python
4805 @subsubsection Symbol table representation in Python
4807 @cindex symbol tables in python
4809 @tindex gdb.Symtab_and_line
4811 Access to symbol table data maintained by @value{GDBN} on the inferior
4812 is exposed to Python via two objects: @code{gdb.Symtab_and_line} and
4813 @code{gdb.Symtab}. Symbol table and line data for a frame is returned
4814 from the @code{find_sal} method in @code{gdb.Frame} object.
4815 @xref{Frames In Python}.
4817 For more information on @value{GDBN}'s symbol table management, see
4818 @ref{Symbols, ,Examining the Symbol Table}, for more information.
4820 A @code{gdb.Symtab_and_line} object has the following attributes:
4822 @defvar Symtab_and_line.symtab
4823 The symbol table object (@code{gdb.Symtab}) for this frame.
4824 This attribute is not writable.
4827 @defvar Symtab_and_line.pc
4828 Indicates the start of the address range occupied by code for the
4829 current source line. This attribute is not writable.
4832 @defvar Symtab_and_line.last
4833 Indicates the end of the address range occupied by code for the current
4834 source line. This attribute is not writable.
4837 @defvar Symtab_and_line.line
4838 Indicates the current line number for this object. This
4839 attribute is not writable.
4842 A @code{gdb.Symtab_and_line} object has the following methods:
4844 @defun Symtab_and_line.is_valid ()
4845 Returns @code{True} if the @code{gdb.Symtab_and_line} object is valid,
4846 @code{False} if not. A @code{gdb.Symtab_and_line} object can become
4847 invalid if the Symbol table and line object it refers to does not
4848 exist in @value{GDBN} any longer. All other
4849 @code{gdb.Symtab_and_line} methods will throw an exception if it is
4850 invalid at the time the method is called.
4853 A @code{gdb.Symtab} object has the following attributes:
4855 @defvar Symtab.filename
4856 The symbol table's source filename. This attribute is not writable.
4859 @defvar Symtab.objfile
4860 The symbol table's backing object file. @xref{Objfiles In Python}.
4861 This attribute is not writable.
4864 @defvar Symtab.producer
4865 The name and possibly version number of the program that
4866 compiled the code in the symbol table.
4867 The contents of this string is up to the compiler.
4868 If no producer information is available then @code{None} is returned.
4869 This attribute is not writable.
4872 A @code{gdb.Symtab} object has the following methods:
4874 @defun Symtab.is_valid ()
4875 Returns @code{True} if the @code{gdb.Symtab} object is valid,
4876 @code{False} if not. A @code{gdb.Symtab} object can become invalid if
4877 the symbol table it refers to does not exist in @value{GDBN} any
4878 longer. All other @code{gdb.Symtab} methods will throw an exception
4879 if it is invalid at the time the method is called.
4882 @defun Symtab.fullname ()
4883 Return the symbol table's source absolute file name.
4886 @defun Symtab.global_block ()
4887 Return the global block of the underlying symbol table.
4888 @xref{Blocks In Python}.
4891 @defun Symtab.static_block ()
4892 Return the static block of the underlying symbol table.
4893 @xref{Blocks In Python}.
4896 @defun Symtab.linetable ()
4897 Return the line table associated with the symbol table.
4898 @xref{Line Tables In Python}.
4901 @node Line Tables In Python
4902 @subsubsection Manipulating line tables using Python
4904 @cindex line tables in python
4905 @tindex gdb.LineTable
4907 Python code can request and inspect line table information from a
4908 symbol table that is loaded in @value{GDBN}. A line table is a
4909 mapping of source lines to their executable locations in memory. To
4910 acquire the line table information for a particular symbol table, use
4911 the @code{linetable} function (@pxref{Symbol Tables In Python}).
4913 A @code{gdb.LineTable} is iterable. The iterator returns
4914 @code{LineTableEntry} objects that correspond to the source line and
4915 address for each line table entry. @code{LineTableEntry} objects have
4916 the following attributes:
4918 @defvar LineTableEntry.line
4919 The source line number for this line table entry. This number
4920 corresponds to the actual line of source. This attribute is not
4924 @defvar LineTableEntry.pc
4925 The address that is associated with the line table entry where the
4926 executable code for that source line resides in memory. This
4927 attribute is not writable.
4930 As there can be multiple addresses for a single source line, you may
4931 receive multiple @code{LineTableEntry} objects with matching
4932 @code{line} attributes, but with different @code{pc} attributes. The
4933 iterator is sorted in ascending @code{pc} order. Here is a small
4934 example illustrating iterating over a line table.
4937 symtab = gdb.selected_frame().find_sal().symtab
4938 linetable = symtab.linetable()
4939 for line in linetable:
4940 print "Line: "+str(line.line)+" Address: "+hex(line.pc)
4943 This will have the following output:
4946 Line: 33 Address: 0x4005c8L
4947 Line: 37 Address: 0x4005caL
4948 Line: 39 Address: 0x4005d2L
4949 Line: 40 Address: 0x4005f8L
4950 Line: 42 Address: 0x4005ffL
4951 Line: 44 Address: 0x400608L
4952 Line: 42 Address: 0x40060cL
4953 Line: 45 Address: 0x400615L
4956 In addition to being able to iterate over a @code{LineTable}, it also
4957 has the following direct access methods:
4959 @defun LineTable.line (line)
4960 Return a Python @code{Tuple} of @code{LineTableEntry} objects for any
4961 entries in the line table for the given @var{line}, which specifies
4962 the source code line. If there are no entries for that source code
4963 @var{line}, the Python @code{None} is returned.
4966 @defun LineTable.has_line (line)
4967 Return a Python @code{Boolean} indicating whether there is an entry in
4968 the line table for this source line. Return @code{True} if an entry
4969 is found, or @code{False} if not.
4972 @defun LineTable.source_lines ()
4973 Return a Python @code{List} of the source line numbers in the symbol
4974 table. Only lines with executable code locations are returned. The
4975 contents of the @code{List} will just be the source line entries
4976 represented as Python @code{Long} values.
4979 @node Breakpoints In Python
4980 @subsubsection Manipulating breakpoints using Python
4982 @cindex breakpoints in python
4983 @tindex gdb.Breakpoint
4985 Python code can manipulate breakpoints via the @code{gdb.Breakpoint}
4988 A breakpoint can be created using one of the two forms of the
4989 @code{gdb.Breakpoint} constructor. The first one accepts a string
4990 like one would pass to the @code{break}
4991 (@pxref{Set Breaks,,Setting Breakpoints}) and @code{watch}
4992 (@pxref{Set Watchpoints, , Setting Watchpoints}) commands, and can be used to
4993 create both breakpoints and watchpoints. The second accepts separate Python
4994 arguments similar to @ref{Explicit Locations}, and can only be used to create
4997 @defun Breakpoint.__init__ (spec @r{[}, type @r{][}, wp_class @r{][}, internal @r{][}, temporary @r{][}, qualified @r{]})
4998 Create a new breakpoint according to @var{spec}, which is a string naming the
4999 location of a breakpoint, or an expression that defines a watchpoint. The
5000 string should describe a location in a format recognized by the @code{break}
5001 command (@pxref{Set Breaks,,Setting Breakpoints}) or, in the case of a
5002 watchpoint, by the @code{watch} command
5003 (@pxref{Set Watchpoints, , Setting Watchpoints}).
5005 The optional @var{type} argument specifies the type of the breakpoint to create,
5008 The optional @var{wp_class} argument defines the class of watchpoint to create,
5009 if @var{type} is @code{gdb.BP_WATCHPOINT}. If @var{wp_class} is omitted, it
5010 defaults to @code{gdb.WP_WRITE}.
5012 The optional @var{internal} argument allows the breakpoint to become invisible
5013 to the user. The breakpoint will neither be reported when created, nor will it
5014 be listed in the output from @code{info breakpoints} (but will be listed with
5015 the @code{maint info breakpoints} command).
5017 The optional @var{temporary} argument makes the breakpoint a temporary
5018 breakpoint. Temporary breakpoints are deleted after they have been hit. Any
5019 further access to the Python breakpoint after it has been hit will result in a
5020 runtime error (as that breakpoint has now been automatically deleted).
5022 The optional @var{qualified} argument is a boolean that allows interpreting
5023 the function passed in @code{spec} as a fully-qualified name. It is equivalent
5024 to @code{break}'s @code{-qualified} flag (@pxref{Linespec Locations} and
5025 @ref{Explicit Locations}).
5029 @defun Breakpoint.__init__ (@r{[} source @r{][}, function @r{][}, label @r{][}, line @r{]}, @r{][} internal @r{][}, temporary @r{][}, qualified @r{]})
5030 This second form of creating a new breakpoint specifies the explicit
5031 location (@pxref{Explicit Locations}) using keywords. The new breakpoint will
5032 be created in the specified source file @var{source}, at the specified
5033 @var{function}, @var{label} and @var{line}.
5035 @var{internal}, @var{temporary} and @var{qualified} have the same usage as
5036 explained previously.
5039 The available types are represented by constants defined in the @code{gdb}
5043 @vindex BP_BREAKPOINT
5044 @item gdb.BP_BREAKPOINT
5045 Normal code breakpoint.
5047 @vindex BP_WATCHPOINT
5048 @item gdb.BP_WATCHPOINT
5049 Watchpoint breakpoint.
5051 @vindex BP_HARDWARE_WATCHPOINT
5052 @item gdb.BP_HARDWARE_WATCHPOINT
5053 Hardware assisted watchpoint.
5055 @vindex BP_READ_WATCHPOINT
5056 @item gdb.BP_READ_WATCHPOINT
5057 Hardware assisted read watchpoint.
5059 @vindex BP_ACCESS_WATCHPOINT
5060 @item gdb.BP_ACCESS_WATCHPOINT
5061 Hardware assisted access watchpoint.
5064 The available watchpoint types represented by constants are defined in the
5070 Read only watchpoint.
5074 Write only watchpoint.
5078 Read/Write watchpoint.
5081 @defun Breakpoint.stop (self)
5082 The @code{gdb.Breakpoint} class can be sub-classed and, in
5083 particular, you may choose to implement the @code{stop} method.
5084 If this method is defined in a sub-class of @code{gdb.Breakpoint},
5085 it will be called when the inferior reaches any location of a
5086 breakpoint which instantiates that sub-class. If the method returns
5087 @code{True}, the inferior will be stopped at the location of the
5088 breakpoint, otherwise the inferior will continue.
5090 If there are multiple breakpoints at the same location with a
5091 @code{stop} method, each one will be called regardless of the
5092 return status of the previous. This ensures that all @code{stop}
5093 methods have a chance to execute at that location. In this scenario
5094 if one of the methods returns @code{True} but the others return
5095 @code{False}, the inferior will still be stopped.
5097 You should not alter the execution state of the inferior (i.e.@:, step,
5098 next, etc.), alter the current frame context (i.e.@:, change the current
5099 active frame), or alter, add or delete any breakpoint. As a general
5100 rule, you should not alter any data within @value{GDBN} or the inferior
5103 Example @code{stop} implementation:
5106 class MyBreakpoint (gdb.Breakpoint):
5108 inf_val = gdb.parse_and_eval("foo")
5115 @defun Breakpoint.is_valid ()
5116 Return @code{True} if this @code{Breakpoint} object is valid,
5117 @code{False} otherwise. A @code{Breakpoint} object can become invalid
5118 if the user deletes the breakpoint. In this case, the object still
5119 exists, but the underlying breakpoint does not. In the cases of
5120 watchpoint scope, the watchpoint remains valid even if execution of the
5121 inferior leaves the scope of that watchpoint.
5124 @defun Breakpoint.delete ()
5125 Permanently deletes the @value{GDBN} breakpoint. This also
5126 invalidates the Python @code{Breakpoint} object. Any further access
5127 to this object's attributes or methods will raise an error.
5130 @defvar Breakpoint.enabled
5131 This attribute is @code{True} if the breakpoint is enabled, and
5132 @code{False} otherwise. This attribute is writable. You can use it to enable
5133 or disable the breakpoint.
5136 @defvar Breakpoint.silent
5137 This attribute is @code{True} if the breakpoint is silent, and
5138 @code{False} otherwise. This attribute is writable.
5140 Note that a breakpoint can also be silent if it has commands and the
5141 first command is @code{silent}. This is not reported by the
5142 @code{silent} attribute.
5145 @defvar Breakpoint.pending
5146 This attribute is @code{True} if the breakpoint is pending, and
5147 @code{False} otherwise. @xref{Set Breaks}. This attribute is
5151 @anchor{python_breakpoint_thread}
5152 @defvar Breakpoint.thread
5153 If the breakpoint is thread-specific, this attribute holds the
5154 thread's global id. If the breakpoint is not thread-specific, this
5155 attribute is @code{None}. This attribute is writable.
5158 @defvar Breakpoint.task
5159 If the breakpoint is Ada task-specific, this attribute holds the Ada task
5160 id. If the breakpoint is not task-specific (or the underlying
5161 language is not Ada), this attribute is @code{None}. This attribute
5165 @defvar Breakpoint.ignore_count
5166 This attribute holds the ignore count for the breakpoint, an integer.
5167 This attribute is writable.
5170 @defvar Breakpoint.number
5171 This attribute holds the breakpoint's number --- the identifier used by
5172 the user to manipulate the breakpoint. This attribute is not writable.
5175 @defvar Breakpoint.type
5176 This attribute holds the breakpoint's type --- the identifier used to
5177 determine the actual breakpoint type or use-case. This attribute is not
5181 @defvar Breakpoint.visible
5182 This attribute tells whether the breakpoint is visible to the user
5183 when set, or when the @samp{info breakpoints} command is run. This
5184 attribute is not writable.
5187 @defvar Breakpoint.temporary
5188 This attribute indicates whether the breakpoint was created as a
5189 temporary breakpoint. Temporary breakpoints are automatically deleted
5190 after that breakpoint has been hit. Access to this attribute, and all
5191 other attributes and functions other than the @code{is_valid}
5192 function, will result in an error after the breakpoint has been hit
5193 (as it has been automatically deleted). This attribute is not
5197 @defvar Breakpoint.hit_count
5198 This attribute holds the hit count for the breakpoint, an integer.
5199 This attribute is writable, but currently it can only be set to zero.
5202 @defvar Breakpoint.location
5203 This attribute holds the location of the breakpoint, as specified by
5204 the user. It is a string. If the breakpoint does not have a location
5205 (that is, it is a watchpoint) the attribute's value is @code{None}. This
5206 attribute is not writable.
5209 @defvar Breakpoint.expression
5210 This attribute holds a breakpoint expression, as specified by
5211 the user. It is a string. If the breakpoint does not have an
5212 expression (the breakpoint is not a watchpoint) the attribute's value
5213 is @code{None}. This attribute is not writable.
5216 @defvar Breakpoint.condition
5217 This attribute holds the condition of the breakpoint, as specified by
5218 the user. It is a string. If there is no condition, this attribute's
5219 value is @code{None}. This attribute is writable.
5222 @defvar Breakpoint.commands
5223 This attribute holds the commands attached to the breakpoint. If
5224 there are commands, this attribute's value is a string holding all the
5225 commands, separated by newlines. If there are no commands, this
5226 attribute is @code{None}. This attribute is writable.
5229 @node Finish Breakpoints in Python
5230 @subsubsection Finish Breakpoints
5232 @cindex python finish breakpoints
5233 @tindex gdb.FinishBreakpoint
5235 A finish breakpoint is a temporary breakpoint set at the return address of
5236 a frame, based on the @code{finish} command. @code{gdb.FinishBreakpoint}
5237 extends @code{gdb.Breakpoint}. The underlying breakpoint will be disabled
5238 and deleted when the execution will run out of the breakpoint scope (i.e.@:
5239 @code{Breakpoint.stop} or @code{FinishBreakpoint.out_of_scope} triggered).
5240 Finish breakpoints are thread specific and must be create with the right
5243 @defun FinishBreakpoint.__init__ (@r{[}frame@r{]} @r{[}, internal@r{]})
5244 Create a finish breakpoint at the return address of the @code{gdb.Frame}
5245 object @var{frame}. If @var{frame} is not provided, this defaults to the
5246 newest frame. The optional @var{internal} argument allows the breakpoint to
5247 become invisible to the user. @xref{Breakpoints In Python}, for further
5248 details about this argument.
5251 @defun FinishBreakpoint.out_of_scope (self)
5252 In some circumstances (e.g.@: @code{longjmp}, C@t{++} exceptions, @value{GDBN}
5253 @code{return} command, @dots{}), a function may not properly terminate, and
5254 thus never hit the finish breakpoint. When @value{GDBN} notices such a
5255 situation, the @code{out_of_scope} callback will be triggered.
5257 You may want to sub-class @code{gdb.FinishBreakpoint} and override this
5261 class MyFinishBreakpoint (gdb.FinishBreakpoint)
5263 print "normal finish"
5266 def out_of_scope ():
5267 print "abnormal finish"
5271 @defvar FinishBreakpoint.return_value
5272 When @value{GDBN} is stopped at a finish breakpoint and the frame
5273 used to build the @code{gdb.FinishBreakpoint} object had debug symbols, this
5274 attribute will contain a @code{gdb.Value} object corresponding to the return
5275 value of the function. The value will be @code{None} if the function return
5276 type is @code{void} or if the return value was not computable. This attribute
5280 @node Lazy Strings In Python
5281 @subsubsection Python representation of lazy strings
5283 @cindex lazy strings in python
5284 @tindex gdb.LazyString
5286 A @dfn{lazy string} is a string whose contents is not retrieved or
5287 encoded until it is needed.
5289 A @code{gdb.LazyString} is represented in @value{GDBN} as an
5290 @code{address} that points to a region of memory, an @code{encoding}
5291 that will be used to encode that region of memory, and a @code{length}
5292 to delimit the region of memory that represents the string. The
5293 difference between a @code{gdb.LazyString} and a string wrapped within
5294 a @code{gdb.Value} is that a @code{gdb.LazyString} will be treated
5295 differently by @value{GDBN} when printing. A @code{gdb.LazyString} is
5296 retrieved and encoded during printing, while a @code{gdb.Value}
5297 wrapping a string is immediately retrieved and encoded on creation.
5299 A @code{gdb.LazyString} object has the following functions:
5301 @defun LazyString.value ()
5302 Convert the @code{gdb.LazyString} to a @code{gdb.Value}. This value
5303 will point to the string in memory, but will lose all the delayed
5304 retrieval, encoding and handling that @value{GDBN} applies to a
5305 @code{gdb.LazyString}.
5308 @defvar LazyString.address
5309 This attribute holds the address of the string. This attribute is not
5313 @defvar LazyString.length
5314 This attribute holds the length of the string in characters. If the
5315 length is -1, then the string will be fetched and encoded up to the
5316 first null of appropriate width. This attribute is not writable.
5319 @defvar LazyString.encoding
5320 This attribute holds the encoding that will be applied to the string
5321 when the string is printed by @value{GDBN}. If the encoding is not
5322 set, or contains an empty string, then @value{GDBN} will select the
5323 most appropriate encoding when the string is printed. This attribute
5327 @defvar LazyString.type
5328 This attribute holds the type that is represented by the lazy string's
5329 type. For a lazy string this is a pointer or array type. To
5330 resolve this to the lazy string's character type, use the type's
5331 @code{target} method. @xref{Types In Python}. This attribute is not
5335 @node Architectures In Python
5336 @subsubsection Python representation of architectures
5337 @cindex Python architectures
5339 @value{GDBN} uses architecture specific parameters and artifacts in a
5340 number of its various computations. An architecture is represented
5341 by an instance of the @code{gdb.Architecture} class.
5343 A @code{gdb.Architecture} class has the following methods:
5345 @defun Architecture.name ()
5346 Return the name (string value) of the architecture.
5349 @defun Architecture.disassemble (@var{start_pc} @r{[}, @var{end_pc} @r{[}, @var{count}@r{]]})
5350 Return a list of disassembled instructions starting from the memory
5351 address @var{start_pc}. The optional arguments @var{end_pc} and
5352 @var{count} determine the number of instructions in the returned list.
5353 If both the optional arguments @var{end_pc} and @var{count} are
5354 specified, then a list of at most @var{count} disassembled instructions
5355 whose start address falls in the closed memory address interval from
5356 @var{start_pc} to @var{end_pc} are returned. If @var{end_pc} is not
5357 specified, but @var{count} is specified, then @var{count} number of
5358 instructions starting from the address @var{start_pc} are returned. If
5359 @var{count} is not specified but @var{end_pc} is specified, then all
5360 instructions whose start address falls in the closed memory address
5361 interval from @var{start_pc} to @var{end_pc} are returned. If neither
5362 @var{end_pc} nor @var{count} are specified, then a single instruction at
5363 @var{start_pc} is returned. For all of these cases, each element of the
5364 returned list is a Python @code{dict} with the following string keys:
5369 The value corresponding to this key is a Python long integer capturing
5370 the memory address of the instruction.
5373 The value corresponding to this key is a string value which represents
5374 the instruction with assembly language mnemonics. The assembly
5375 language flavor used is the same as that specified by the current CLI
5376 variable @code{disassembly-flavor}. @xref{Machine Code}.
5379 The value corresponding to this key is the length (integer value) of the
5380 instruction in bytes.
5385 @node Python Auto-loading
5386 @subsection Python Auto-loading
5387 @cindex Python auto-loading
5389 When a new object file is read (for example, due to the @code{file}
5390 command, or because the inferior has loaded a shared library),
5391 @value{GDBN} will look for Python support scripts in several ways:
5392 @file{@var{objfile}-gdb.py} and @code{.debug_gdb_scripts} section.
5393 @xref{Auto-loading extensions}.
5395 The auto-loading feature is useful for supplying application-specific
5396 debugging commands and scripts.
5398 Auto-loading can be enabled or disabled,
5399 and the list of auto-loaded scripts can be printed.
5402 @anchor{set auto-load python-scripts}
5403 @kindex set auto-load python-scripts
5404 @item set auto-load python-scripts [on|off]
5405 Enable or disable the auto-loading of Python scripts.
5407 @anchor{show auto-load python-scripts}
5408 @kindex show auto-load python-scripts
5409 @item show auto-load python-scripts
5410 Show whether auto-loading of Python scripts is enabled or disabled.
5412 @anchor{info auto-load python-scripts}
5413 @kindex info auto-load python-scripts
5414 @cindex print list of auto-loaded Python scripts
5415 @item info auto-load python-scripts [@var{regexp}]
5416 Print the list of all Python scripts that @value{GDBN} auto-loaded.
5418 Also printed is the list of Python scripts that were mentioned in
5419 the @code{.debug_gdb_scripts} section and were either not found
5420 (@pxref{dotdebug_gdb_scripts section}) or were not auto-loaded due to
5421 @code{auto-load safe-path} rejection (@pxref{Auto-loading}).
5422 This is useful because their names are not printed when @value{GDBN}
5423 tries to load them and fails. There may be many of them, and printing
5424 an error message for each one is problematic.
5426 If @var{regexp} is supplied only Python scripts with matching names are printed.
5431 (gdb) info auto-load python-scripts
5433 Yes py-section-script.py
5434 full name: /tmp/py-section-script.py
5435 No my-foo-pretty-printers.py
5439 When reading an auto-loaded file or script, @value{GDBN} sets the
5440 @dfn{current objfile}. This is available via the @code{gdb.current_objfile}
5441 function (@pxref{Objfiles In Python}). This can be useful for
5442 registering objfile-specific pretty-printers and frame-filters.
5444 @node Python modules
5445 @subsection Python modules
5446 @cindex python modules
5448 @value{GDBN} comes with several modules to assist writing Python code.
5451 * gdb.printing:: Building and registering pretty-printers.
5452 * gdb.types:: Utilities for working with types.
5453 * gdb.prompt:: Utilities for prompt value substitution.
5457 @subsubsection gdb.printing
5458 @cindex gdb.printing
5460 This module provides a collection of utilities for working with
5464 @item PrettyPrinter (@var{name}, @var{subprinters}=None)
5465 This class specifies the API that makes @samp{info pretty-printer},
5466 @samp{enable pretty-printer} and @samp{disable pretty-printer} work.
5467 Pretty-printers should generally inherit from this class.
5469 @item SubPrettyPrinter (@var{name})
5470 For printers that handle multiple types, this class specifies the
5471 corresponding API for the subprinters.
5473 @item RegexpCollectionPrettyPrinter (@var{name})
5474 Utility class for handling multiple printers, all recognized via
5475 regular expressions.
5476 @xref{Writing a Pretty-Printer}, for an example.
5478 @item FlagEnumerationPrinter (@var{name})
5479 A pretty-printer which handles printing of @code{enum} values. Unlike
5480 @value{GDBN}'s built-in @code{enum} printing, this printer attempts to
5481 work properly when there is some overlap between the enumeration
5482 constants. The argument @var{name} is the name of the printer and
5483 also the name of the @code{enum} type to look up.
5485 @item register_pretty_printer (@var{obj}, @var{printer}, @var{replace}=False)
5486 Register @var{printer} with the pretty-printer list of @var{obj}.
5487 If @var{replace} is @code{True} then any existing copy of the printer
5488 is replaced. Otherwise a @code{RuntimeError} exception is raised
5489 if a printer with the same name already exists.
5493 @subsubsection gdb.types
5496 This module provides a collection of utilities for working with
5497 @code{gdb.Type} objects.
5500 @item get_basic_type (@var{type})
5501 Return @var{type} with const and volatile qualifiers stripped,
5502 and with typedefs and C@t{++} references converted to the underlying type.
5507 typedef const int const_int;
5509 const_int& foo_ref (foo);
5510 int main () @{ return 0; @}
5517 (gdb) python import gdb.types
5518 (gdb) python foo_ref = gdb.parse_and_eval("foo_ref")
5519 (gdb) python print gdb.types.get_basic_type(foo_ref.type)
5523 @item has_field (@var{type}, @var{field})
5524 Return @code{True} if @var{type}, assumed to be a type with fields
5525 (e.g., a structure or union), has field @var{field}.
5527 @item make_enum_dict (@var{enum_type})
5528 Return a Python @code{dictionary} type produced from @var{enum_type}.
5530 @item deep_items (@var{type})
5531 Returns a Python iterator similar to the standard
5532 @code{gdb.Type.iteritems} method, except that the iterator returned
5533 by @code{deep_items} will recursively traverse anonymous struct or
5534 union fields. For example:
5548 Then in @value{GDBN}:
5550 (@value{GDBP}) python import gdb.types
5551 (@value{GDBP}) python struct_a = gdb.lookup_type("struct A")
5552 (@value{GDBP}) python print struct_a.keys ()
5554 (@value{GDBP}) python print [k for k,v in gdb.types.deep_items(struct_a)]
5555 @{['a', 'b0', 'b1']@}
5558 @item get_type_recognizers ()
5559 Return a list of the enabled type recognizers for the current context.
5560 This is called by @value{GDBN} during the type-printing process
5561 (@pxref{Type Printing API}).
5563 @item apply_type_recognizers (recognizers, type_obj)
5564 Apply the type recognizers, @var{recognizers}, to the type object
5565 @var{type_obj}. If any recognizer returns a string, return that
5566 string. Otherwise, return @code{None}. This is called by
5567 @value{GDBN} during the type-printing process (@pxref{Type Printing
5570 @item register_type_printer (locus, printer)
5571 This is a convenience function to register a type printer
5572 @var{printer}. The printer must implement the type printer protocol.
5573 The @var{locus} argument is either a @code{gdb.Objfile}, in which case
5574 the printer is registered with that objfile; a @code{gdb.Progspace},
5575 in which case the printer is registered with that progspace; or
5576 @code{None}, in which case the printer is registered globally.
5579 This is a base class that implements the type printer protocol. Type
5580 printers are encouraged, but not required, to derive from this class.
5581 It defines a constructor:
5583 @defmethod TypePrinter __init__ (self, name)
5584 Initialize the type printer with the given name. The new printer
5585 starts in the enabled state.
5591 @subsubsection gdb.prompt
5594 This module provides a method for prompt value-substitution.
5597 @item substitute_prompt (@var{string})
5598 Return @var{string} with escape sequences substituted by values. Some
5599 escape sequences take arguments. You can specify arguments inside
5600 ``@{@}'' immediately following the escape sequence.
5602 The escape sequences you can pass to this function are:
5606 Substitute a backslash.
5608 Substitute an ESC character.
5610 Substitute the selected frame; an argument names a frame parameter.
5612 Substitute a newline.
5614 Substitute a parameter's value; the argument names the parameter.
5616 Substitute a carriage return.
5618 Substitute the selected thread; an argument names a thread parameter.
5620 Substitute the version of GDB.
5622 Substitute the current working directory.
5624 Begin a sequence of non-printing characters. These sequences are
5625 typically used with the ESC character, and are not counted in the string
5626 length. Example: ``\[\e[0;34m\](gdb)\[\e[0m\]'' will return a
5627 blue-colored ``(gdb)'' prompt where the length is five.
5629 End a sequence of non-printing characters.
5635 substitute_prompt (``frame: \f,
5636 print arguments: \p@{print frame-arguments@}'')
5639 @exdent will return the string:
5642 "frame: main, print arguments: scalars"