1 @c Copyright (C) 2008-2015 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}.
22 @cindex python directory
23 Python scripts used by @value{GDBN} should be installed in
24 @file{@var{data-directory}/python}, where @var{data-directory} is
25 the data directory as determined at @value{GDBN} startup (@pxref{Data Files}).
26 This directory, known as the @dfn{python directory},
27 is automatically added to the Python Search Path in order to allow
28 the Python interpreter to locate all scripts installed at this location.
30 Additionally, @value{GDBN} commands and convenience functions which
31 are written in Python and are located in the
32 @file{@var{data-directory}/python/gdb/command} or
33 @file{@var{data-directory}/python/gdb/function} directories are
34 automatically imported when @value{GDBN} starts.
37 * Python Commands:: Accessing Python from @value{GDBN}.
38 * Python API:: Accessing @value{GDBN} from Python.
39 * Python Auto-loading:: Automatically loading Python code.
40 * Python modules:: Python modules provided by @value{GDBN}.
44 @subsection Python Commands
45 @cindex python commands
46 @cindex commands to access python
48 @value{GDBN} provides two commands for accessing the Python interpreter,
49 and one related setting:
52 @kindex python-interactive
54 @item python-interactive @r{[}@var{command}@r{]}
55 @itemx pi @r{[}@var{command}@r{]}
56 Without an argument, the @code{python-interactive} command can be used
57 to start an interactive Python prompt. To return to @value{GDBN},
58 type the @code{EOF} character (e.g., @kbd{Ctrl-D} on an empty prompt).
60 Alternatively, a single-line Python command can be given as an
61 argument and evaluated. If the command is an expression, the result
62 will be printed; otherwise, nothing will be printed. For example:
65 (@value{GDBP}) python-interactive 2 + 3
71 @item python @r{[}@var{command}@r{]}
72 @itemx py @r{[}@var{command}@r{]}
73 The @code{python} command can be used to evaluate Python code.
75 If given an argument, the @code{python} command will evaluate the
76 argument as a Python command. For example:
79 (@value{GDBP}) python print 23
83 If you do not provide an argument to @code{python}, it will act as a
84 multi-line command, like @code{define}. In this case, the Python
85 script is made up of subsequent command lines, given after the
86 @code{python} command. This command list is terminated using a line
87 containing @code{end}. For example:
92 End with a line saying just "end".
98 @kindex set python print-stack
99 @item set python print-stack
100 By default, @value{GDBN} will print only the message component of a
101 Python exception when an error occurs in a Python script. This can be
102 controlled using @code{set python print-stack}: if @code{full}, then
103 full Python stack printing is enabled; if @code{none}, then Python stack
104 and message printing is disabled; if @code{message}, the default, only
105 the message component of the error is printed.
108 It is also possible to execute a Python script from the @value{GDBN}
112 @item source @file{script-name}
113 The script name must end with @samp{.py} and @value{GDBN} must be configured
114 to recognize the script language based on filename extension using
115 the @code{script-extension} setting. @xref{Extending GDB, ,Extending GDB}.
117 @item python execfile ("script-name")
118 This method is based on the @code{execfile} Python built-in function,
119 and thus is always available.
123 @subsection Python API
125 @cindex programming in python
127 You can get quick online help for @value{GDBN}'s Python API by issuing
128 the command @w{@kbd{python help (gdb)}}.
130 Functions and methods which have two or more optional arguments allow
131 them to be specified using keyword syntax. This allows passing some
132 optional arguments while skipping others. Example:
133 @w{@code{gdb.some_function ('foo', bar = 1, baz = 2)}}.
136 * Basic Python:: Basic Python Functions.
137 * Exception Handling:: How Python exceptions are translated.
138 * Values From Inferior:: Python representation of values.
139 * Types In Python:: Python representation of types.
140 * Pretty Printing API:: Pretty-printing values.
141 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
142 * Writing a Pretty-Printer:: Writing a Pretty-Printer.
143 * Type Printing API:: Pretty-printing types.
144 * Frame Filter API:: Filtering Frames.
145 * Frame Decorator API:: Decorating Frames.
146 * Writing a Frame Filter:: Writing a Frame Filter.
147 * Unwinding Frames in Python:: Writing frame unwinder.
148 * Xmethods In Python:: Adding and replacing methods of C++ classes.
149 * Xmethod API:: Xmethod types.
150 * Writing an Xmethod:: Writing an xmethod.
151 * Inferiors In Python:: Python representation of inferiors (processes)
152 * Events In Python:: Listening for events from @value{GDBN}.
153 * Threads In Python:: Accessing inferior threads from Python.
154 * Commands In Python:: Implementing new commands in Python.
155 * Parameters In Python:: Adding new @value{GDBN} parameters.
156 * Functions In Python:: Writing new convenience functions.
157 * Progspaces In Python:: Program spaces.
158 * Objfiles In Python:: Object files.
159 * Frames In Python:: Accessing inferior stack frames from Python.
160 * Blocks In Python:: Accessing blocks from Python.
161 * Symbols In Python:: Python representation of symbols.
162 * Symbol Tables In Python:: Python representation of symbol tables.
163 * Line Tables In Python:: Python representation of line tables.
164 * Breakpoints In Python:: Manipulating breakpoints using Python.
165 * Finish Breakpoints in Python:: Setting Breakpoints on function return
167 * Lazy Strings In Python:: Python representation of lazy strings.
168 * Architectures In Python:: Python representation of architectures.
172 @subsubsection Basic Python
174 @cindex python stdout
175 @cindex python pagination
176 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
177 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
178 A Python program which outputs to one of these streams may have its
179 output interrupted by the user (@pxref{Screen Size}). In this
180 situation, a Python @code{KeyboardInterrupt} exception is thrown.
182 Some care must be taken when writing Python code to run in
183 @value{GDBN}. Two things worth noting in particular:
187 @value{GDBN} install handlers for @code{SIGCHLD} and @code{SIGINT}.
188 Python code must not override these, or even change the options using
189 @code{sigaction}. If your program changes the handling of these
190 signals, @value{GDBN} will most likely stop working correctly. Note
191 that it is unfortunately common for GUI toolkits to install a
192 @code{SIGCHLD} handler.
195 @value{GDBN} takes care to mark its internal file descriptors as
196 close-on-exec. However, this cannot be done in a thread-safe way on
197 all platforms. Your Python programs should be aware of this and
198 should both create new file descriptors with the close-on-exec flag
199 set and arrange to close unneeded file descriptors before starting a
203 @cindex python functions
204 @cindex python module
206 @value{GDBN} introduces a new Python module, named @code{gdb}. All
207 methods and classes added by @value{GDBN} are placed in this module.
208 @value{GDBN} automatically @code{import}s the @code{gdb} module for
209 use in all scripts evaluated by the @code{python} command.
211 @findex gdb.PYTHONDIR
212 @defvar gdb.PYTHONDIR
213 A string containing the python directory (@pxref{Python}).
217 @defun gdb.execute (command @r{[}, from_tty @r{[}, to_string@r{]]})
218 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
219 If a GDB exception happens while @var{command} runs, it is
220 translated as described in @ref{Exception Handling,,Exception Handling}.
222 The @var{from_tty} flag specifies whether @value{GDBN} ought to consider this
223 command as having originated from the user invoking it interactively.
224 It must be a boolean value. If omitted, it defaults to @code{False}.
226 By default, any output produced by @var{command} is sent to
227 @value{GDBN}'s standard output (and to the log output if logging is
228 turned on). If the @var{to_string} parameter is
229 @code{True}, then output will be collected by @code{gdb.execute} and
230 returned as a string. The default is @code{False}, in which case the
231 return value is @code{None}. If @var{to_string} is @code{True}, the
232 @value{GDBN} virtual terminal will be temporarily set to unlimited width
233 and height, and its pagination will be disabled; @pxref{Screen Size}.
236 @findex gdb.breakpoints
237 @defun gdb.breakpoints ()
238 Return a sequence holding all of @value{GDBN}'s breakpoints.
239 @xref{Breakpoints In Python}, for more information.
242 @findex gdb.parameter
243 @defun gdb.parameter (parameter)
244 Return the value of a @value{GDBN} @var{parameter} given by its name,
245 a string; the parameter name string may contain spaces if the parameter has a
246 multi-part name. For example, @samp{print object} is a valid
249 If the named parameter does not exist, this function throws a
250 @code{gdb.error} (@pxref{Exception Handling}). Otherwise, the
251 parameter's value is converted to a Python value of the appropriate
256 @defun gdb.history (number)
257 Return a value from @value{GDBN}'s value history (@pxref{Value
258 History}). The @var{number} argument indicates which history element to return.
259 If @var{number} is negative, then @value{GDBN} will take its absolute value
260 and count backward from the last element (i.e., the most recent element) to
261 find the value to return. If @var{number} is zero, then @value{GDBN} will
262 return the most recent element. If the element specified by @var{number}
263 doesn't exist in the value history, a @code{gdb.error} exception will be
266 If no exception is raised, the return value is always an instance of
267 @code{gdb.Value} (@pxref{Values From Inferior}).
270 @findex gdb.parse_and_eval
271 @defun gdb.parse_and_eval (expression)
272 Parse @var{expression}, which must be a string, as an expression in
273 the current language, evaluate it, and return the result as a
276 This function can be useful when implementing a new command
277 (@pxref{Commands In Python}), as it provides a way to parse the
278 command's argument as an expression. It is also useful simply to
279 compute values, for example, it is the only way to get the value of a
280 convenience variable (@pxref{Convenience Vars}) as a @code{gdb.Value}.
283 @findex gdb.find_pc_line
284 @defun gdb.find_pc_line (pc)
285 Return the @code{gdb.Symtab_and_line} object corresponding to the
286 @var{pc} value. @xref{Symbol Tables In Python}. If an invalid
287 value of @var{pc} is passed as an argument, then the @code{symtab} and
288 @code{line} attributes of the returned @code{gdb.Symtab_and_line} object
289 will be @code{None} and 0 respectively.
292 @findex gdb.post_event
293 @defun gdb.post_event (event)
294 Put @var{event}, a callable object taking no arguments, into
295 @value{GDBN}'s internal event queue. This callable will be invoked at
296 some later point, during @value{GDBN}'s event processing. Events
297 posted using @code{post_event} will be run in the order in which they
298 were posted; however, there is no way to know when they will be
299 processed relative to other events inside @value{GDBN}.
301 @value{GDBN} is not thread-safe. If your Python program uses multiple
302 threads, you must be careful to only call @value{GDBN}-specific
303 functions in the @value{GDBN} thread. @code{post_event} ensures
307 (@value{GDBP}) python
311 > def __init__(self, message):
312 > self.message = message;
313 > def __call__(self):
314 > gdb.write(self.message)
316 >class MyThread1 (threading.Thread):
318 > gdb.post_event(Writer("Hello "))
320 >class MyThread2 (threading.Thread):
322 > gdb.post_event(Writer("World\n"))
327 (@value{GDBP}) Hello World
332 @defun gdb.write (string @r{[}, stream{]})
333 Print a string to @value{GDBN}'s paginated output stream. The
334 optional @var{stream} determines the stream to print to. The default
335 stream is @value{GDBN}'s standard output stream. Possible stream
342 @value{GDBN}'s standard output stream.
347 @value{GDBN}'s standard error stream.
352 @value{GDBN}'s log stream (@pxref{Logging Output}).
355 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
356 call this function and will automatically direct the output to the
362 Flush the buffer of a @value{GDBN} paginated stream so that the
363 contents are displayed immediately. @value{GDBN} will flush the
364 contents of a stream automatically when it encounters a newline in the
365 buffer. The optional @var{stream} determines the stream to flush. The
366 default stream is @value{GDBN}'s standard output stream. Possible
373 @value{GDBN}'s standard output stream.
378 @value{GDBN}'s standard error stream.
383 @value{GDBN}'s log stream (@pxref{Logging Output}).
387 Flushing @code{sys.stdout} or @code{sys.stderr} will automatically
388 call this function for the relevant stream.
391 @findex gdb.target_charset
392 @defun gdb.target_charset ()
393 Return the name of the current target character set (@pxref{Character
394 Sets}). This differs from @code{gdb.parameter('target-charset')} in
395 that @samp{auto} is never returned.
398 @findex gdb.target_wide_charset
399 @defun gdb.target_wide_charset ()
400 Return the name of the current target wide character set
401 (@pxref{Character Sets}). This differs from
402 @code{gdb.parameter('target-wide-charset')} in that @samp{auto} is
406 @findex gdb.solib_name
407 @defun gdb.solib_name (address)
408 Return the name of the shared library holding the given @var{address}
409 as a string, or @code{None}.
412 @findex gdb.decode_line
413 @defun gdb.decode_line @r{[}expression@r{]}
414 Return locations of the line specified by @var{expression}, or of the
415 current line if no argument was given. This function returns a Python
416 tuple containing two elements. The first element contains a string
417 holding any unparsed section of @var{expression} (or @code{None} if
418 the expression has been fully parsed). The second element contains
419 either @code{None} or another tuple that contains all the locations
420 that match the expression represented as @code{gdb.Symtab_and_line}
421 objects (@pxref{Symbol Tables In Python}). If @var{expression} is
422 provided, it is decoded the way that @value{GDBN}'s inbuilt
423 @code{break} or @code{edit} commands do (@pxref{Specify Location}).
426 @defun gdb.prompt_hook (current_prompt)
429 If @var{prompt_hook} is callable, @value{GDBN} will call the method
430 assigned to this operation before a prompt is displayed by
433 The parameter @code{current_prompt} contains the current @value{GDBN}
434 prompt. This method must return a Python string, or @code{None}. If
435 a string is returned, the @value{GDBN} prompt will be set to that
436 string. If @code{None} is returned, @value{GDBN} will continue to use
439 Some prompts cannot be substituted in @value{GDBN}. Secondary prompts
440 such as those used by readline for command input, and annotation
441 related prompts are prohibited from being changed.
444 @node Exception Handling
445 @subsubsection Exception Handling
446 @cindex python exceptions
447 @cindex exceptions, python
449 When executing the @code{python} command, Python exceptions
450 uncaught within the Python code are translated to calls to
451 @value{GDBN} error-reporting mechanism. If the command that called
452 @code{python} does not handle the error, @value{GDBN} will
453 terminate it and print an error message containing the Python
454 exception name, the associated value, and the Python call stack
455 backtrace at the point where the exception was raised. Example:
458 (@value{GDBP}) python print foo
459 Traceback (most recent call last):
460 File "<string>", line 1, in <module>
461 NameError: name 'foo' is not defined
464 @value{GDBN} errors that happen in @value{GDBN} commands invoked by
465 Python code are converted to Python exceptions. The type of the
466 Python exception depends on the error.
470 This is the base class for most exceptions generated by @value{GDBN}.
471 It is derived from @code{RuntimeError}, for compatibility with earlier
472 versions of @value{GDBN}.
474 If an error occurring in @value{GDBN} does not fit into some more
475 specific category, then the generated exception will have this type.
477 @item gdb.MemoryError
478 This is a subclass of @code{gdb.error} which is thrown when an
479 operation tried to access invalid memory in the inferior.
481 @item KeyboardInterrupt
482 User interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
483 prompt) is translated to a Python @code{KeyboardInterrupt} exception.
486 In all cases, your exception handler will see the @value{GDBN} error
487 message as its value and the Python call stack backtrace at the Python
488 statement closest to where the @value{GDBN} error occured as the
492 When implementing @value{GDBN} commands in Python via @code{gdb.Command},
493 it is useful to be able to throw an exception that doesn't cause a
494 traceback to be printed. For example, the user may have invoked the
495 command incorrectly. Use the @code{gdb.GdbError} exception
496 to handle this case. Example:
500 >class HelloWorld (gdb.Command):
501 > """Greet the whole world."""
502 > def __init__ (self):
503 > super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_USER)
504 > def invoke (self, args, from_tty):
505 > argv = gdb.string_to_argv (args)
506 > if len (argv) != 0:
507 > raise gdb.GdbError ("hello-world takes no arguments")
508 > print "Hello, World!"
512 hello-world takes no arguments
515 @node Values From Inferior
516 @subsubsection Values From Inferior
517 @cindex values from inferior, with Python
518 @cindex python, working with values from inferior
520 @cindex @code{gdb.Value}
521 @value{GDBN} provides values it obtains from the inferior program in
522 an object of type @code{gdb.Value}. @value{GDBN} uses this object
523 for its internal bookkeeping of the inferior's values, and for
524 fetching values when necessary.
526 Inferior values that are simple scalars can be used directly in
527 Python expressions that are valid for the value's data type. Here's
528 an example for an integer or floating-point value @code{some_val}:
535 As result of this, @code{bar} will also be a @code{gdb.Value} object
536 whose values are of the same type as those of @code{some_val}. Valid
537 Python operations can also be performed on @code{gdb.Value} objects
538 representing a @code{struct} or @code{class} object. For such cases,
539 the overloaded operator (if present), is used to perform the operation.
540 For example, if @code{val1} and @code{val2} are @code{gdb.Value} objects
541 representing instances of a @code{class} which overloads the @code{+}
542 operator, then one can use the @code{+} operator in their Python script
550 The result of the operation @code{val3} is also a @code{gdb.Value}
551 object corresponding to the value returned by the overloaded @code{+}
552 operator. In general, overloaded operators are invoked for the
553 following operations: @code{+} (binary addition), @code{-} (binary
554 subtraction), @code{*} (multiplication), @code{/}, @code{%}, @code{<<},
555 @code{>>}, @code{|}, @code{&}, @code{^}.
557 Inferior values that are structures or instances of some class can
558 be accessed using the Python @dfn{dictionary syntax}. For example, if
559 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
560 can access its @code{foo} element with:
563 bar = some_val['foo']
566 @cindex getting structure elements using gdb.Field objects as subscripts
567 Again, @code{bar} will also be a @code{gdb.Value} object. Structure
568 elements can also be accessed by using @code{gdb.Field} objects as
569 subscripts (@pxref{Types In Python}, for more information on
570 @code{gdb.Field} objects). For example, if @code{foo_field} is a
571 @code{gdb.Field} object corresponding to element @code{foo} of the above
572 structure, then @code{bar} can also be accessed as follows:
575 bar = some_val[foo_field]
578 A @code{gdb.Value} that represents a function can be executed via
579 inferior function call. Any arguments provided to the call must match
580 the function's prototype, and must be provided in the order specified
583 For example, @code{some_val} is a @code{gdb.Value} instance
584 representing a function that takes two integers as arguments. To
585 execute this function, call it like so:
588 result = some_val (10,20)
591 Any values returned from a function call will be stored as a
594 The following attributes are provided:
596 @defvar Value.address
597 If this object is addressable, this read-only attribute holds a
598 @code{gdb.Value} object representing the address. Otherwise,
599 this attribute holds @code{None}.
602 @cindex optimized out value in Python
603 @defvar Value.is_optimized_out
604 This read-only boolean attribute is true if the compiler optimized out
605 this value, thus it is not available for fetching from the inferior.
609 The type of this @code{gdb.Value}. The value of this attribute is a
610 @code{gdb.Type} object (@pxref{Types In Python}).
613 @defvar Value.dynamic_type
614 The dynamic type of this @code{gdb.Value}. This uses C@t{++} run-time
615 type information (@acronym{RTTI}) to determine the dynamic type of the
616 value. If this value is of class type, it will return the class in
617 which the value is embedded, if any. If this value is of pointer or
618 reference to a class type, it will compute the dynamic type of the
619 referenced object, and return a pointer or reference to that type,
620 respectively. In all other cases, it will return the value's static
623 Note that this feature will only work when debugging a C@t{++} program
624 that includes @acronym{RTTI} for the object in question. Otherwise,
625 it will just return the static type of the value as in @kbd{ptype foo}
626 (@pxref{Symbols, ptype}).
629 @defvar Value.is_lazy
630 The value of this read-only boolean attribute is @code{True} if this
631 @code{gdb.Value} has not yet been fetched from the inferior.
632 @value{GDBN} does not fetch values until necessary, for efficiency.
636 myval = gdb.parse_and_eval ('somevar')
639 The value of @code{somevar} is not fetched at this time. It will be
640 fetched when the value is needed, or when the @code{fetch_lazy}
644 The following methods are provided:
646 @defun Value.__init__ (@var{val})
647 Many Python values can be converted directly to a @code{gdb.Value} via
648 this object initializer. Specifically:
652 A Python boolean is converted to the boolean type from the current
656 A Python integer is converted to the C @code{long} type for the
657 current architecture.
660 A Python long is converted to the C @code{long long} type for the
661 current architecture.
664 A Python float is converted to the C @code{double} type for the
665 current architecture.
668 A Python string is converted to a target string in the current target
669 language using the current target encoding.
670 If a character cannot be represented in the current target encoding,
671 then an exception is thrown.
673 @item @code{gdb.Value}
674 If @code{val} is a @code{gdb.Value}, then a copy of the value is made.
676 @item @code{gdb.LazyString}
677 If @code{val} is a @code{gdb.LazyString} (@pxref{Lazy Strings In
678 Python}), then the lazy string's @code{value} method is called, and
683 @defun Value.cast (type)
684 Return a new instance of @code{gdb.Value} that is the result of
685 casting this instance to the type described by @var{type}, which must
686 be a @code{gdb.Type} object. If the cast cannot be performed for some
687 reason, this method throws an exception.
690 @defun Value.dereference ()
691 For pointer data types, this method returns a new @code{gdb.Value} object
692 whose contents is the object pointed to by the pointer. For example, if
693 @code{foo} is a C pointer to an @code{int}, declared in your C program as
700 then you can use the corresponding @code{gdb.Value} to access what
701 @code{foo} points to like this:
704 bar = foo.dereference ()
707 The result @code{bar} will be a @code{gdb.Value} object holding the
708 value pointed to by @code{foo}.
710 A similar function @code{Value.referenced_value} exists which also
711 returns @code{gdb.Value} objects corresonding to the values pointed to
712 by pointer values (and additionally, values referenced by reference
713 values). However, the behavior of @code{Value.dereference}
714 differs from @code{Value.referenced_value} by the fact that the
715 behavior of @code{Value.dereference} is identical to applying the C
716 unary operator @code{*} on a given value. For example, consider a
717 reference to a pointer @code{ptrref}, declared in your C@t{++} program
725 intptr &ptrref = ptr;
728 Though @code{ptrref} is a reference value, one can apply the method
729 @code{Value.dereference} to the @code{gdb.Value} object corresponding
730 to it and obtain a @code{gdb.Value} which is identical to that
731 corresponding to @code{val}. However, if you apply the method
732 @code{Value.referenced_value}, the result would be a @code{gdb.Value}
733 object identical to that corresponding to @code{ptr}.
736 py_ptrref = gdb.parse_and_eval ("ptrref")
737 py_val = py_ptrref.dereference ()
738 py_ptr = py_ptrref.referenced_value ()
741 The @code{gdb.Value} object @code{py_val} is identical to that
742 corresponding to @code{val}, and @code{py_ptr} is identical to that
743 corresponding to @code{ptr}. In general, @code{Value.dereference} can
744 be applied whenever the C unary operator @code{*} can be applied
745 to the corresponding C value. For those cases where applying both
746 @code{Value.dereference} and @code{Value.referenced_value} is allowed,
747 the results obtained need not be identical (as we have seen in the above
748 example). The results are however identical when applied on
749 @code{gdb.Value} objects corresponding to pointers (@code{gdb.Value}
750 objects with type code @code{TYPE_CODE_PTR}) in a C/C@t{++} program.
753 @defun Value.referenced_value ()
754 For pointer or reference data types, this method returns a new
755 @code{gdb.Value} object corresponding to the value referenced by the
756 pointer/reference value. For pointer data types,
757 @code{Value.dereference} and @code{Value.referenced_value} produce
758 identical results. The difference between these methods is that
759 @code{Value.dereference} cannot get the values referenced by reference
760 values. For example, consider a reference to an @code{int}, declared
761 in your C@t{++} program as
769 then applying @code{Value.dereference} to the @code{gdb.Value} object
770 corresponding to @code{ref} will result in an error, while applying
771 @code{Value.referenced_value} will result in a @code{gdb.Value} object
772 identical to that corresponding to @code{val}.
775 py_ref = gdb.parse_and_eval ("ref")
776 er_ref = py_ref.dereference () # Results in error
777 py_val = py_ref.referenced_value () # Returns the referenced value
780 The @code{gdb.Value} object @code{py_val} is identical to that
781 corresponding to @code{val}.
784 @defun Value.reference_value ()
785 Return a @code{gdb.Value} object which is a reference to the value
786 encapsulated by this instance.
789 @defun Value.const_value ()
790 Return a @code{gdb.Value} object which is a @code{const} version of the
791 value encapsulated by this instance.
794 @defun Value.dynamic_cast (type)
795 Like @code{Value.cast}, but works as if the C@t{++} @code{dynamic_cast}
796 operator were used. Consult a C@t{++} reference for details.
799 @defun Value.reinterpret_cast (type)
800 Like @code{Value.cast}, but works as if the C@t{++} @code{reinterpret_cast}
801 operator were used. Consult a C@t{++} reference for details.
804 @defun Value.string (@r{[}encoding@r{[}, errors@r{[}, length@r{]]]})
805 If this @code{gdb.Value} represents a string, then this method
806 converts the contents to a Python string. Otherwise, this method will
809 Values are interpreted as strings according to the rules of the
810 current language. If the optional length argument is given, the
811 string will be converted to that length, and will include any embedded
812 zeroes that the string may contain. Otherwise, for languages
813 where the string is zero-terminated, the entire string will be
816 For example, in C-like languages, a value is a string if it is a pointer
817 to or an array of characters or ints of type @code{wchar_t}, @code{char16_t},
820 If the optional @var{encoding} argument is given, it must be a string
821 naming the encoding of the string in the @code{gdb.Value}, such as
822 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
823 the same encodings as the corresponding argument to Python's
824 @code{string.decode} method, and the Python codec machinery will be used
825 to convert the string. If @var{encoding} is not given, or if
826 @var{encoding} is the empty string, then either the @code{target-charset}
827 (@pxref{Character Sets}) will be used, or a language-specific encoding
828 will be used, if the current language is able to supply one.
830 The optional @var{errors} argument is the same as the corresponding
831 argument to Python's @code{string.decode} method.
833 If the optional @var{length} argument is given, the string will be
834 fetched and converted to the given length.
837 @defun Value.lazy_string (@r{[}encoding @r{[}, length@r{]]})
838 If this @code{gdb.Value} represents a string, then this method
839 converts the contents to a @code{gdb.LazyString} (@pxref{Lazy Strings
840 In Python}). Otherwise, this method will throw an exception.
842 If the optional @var{encoding} argument is given, it must be a string
843 naming the encoding of the @code{gdb.LazyString}. Some examples are:
844 @samp{ascii}, @samp{iso-8859-6} or @samp{utf-8}. If the
845 @var{encoding} argument is an encoding that @value{GDBN} does
846 recognize, @value{GDBN} will raise an error.
848 When a lazy string is printed, the @value{GDBN} encoding machinery is
849 used to convert the string during printing. If the optional
850 @var{encoding} argument is not provided, or is an empty string,
851 @value{GDBN} will automatically select the encoding most suitable for
852 the string type. For further information on encoding in @value{GDBN}
853 please see @ref{Character Sets}.
855 If the optional @var{length} argument is given, the string will be
856 fetched and encoded to the length of characters specified. If
857 the @var{length} argument is not provided, the string will be fetched
858 and encoded until a null of appropriate width is found.
861 @defun Value.fetch_lazy ()
862 If the @code{gdb.Value} object is currently a lazy value
863 (@code{gdb.Value.is_lazy} is @code{True}), then the value is
864 fetched from the inferior. Any errors that occur in the process
865 will produce a Python exception.
867 If the @code{gdb.Value} object is not a lazy value, this method
870 This method does not return a value.
874 @node Types In Python
875 @subsubsection Types In Python
876 @cindex types in Python
877 @cindex Python, working with types
880 @value{GDBN} represents types from the inferior using the class
883 The following type-related functions are available in the @code{gdb}
886 @findex gdb.lookup_type
887 @defun gdb.lookup_type (name @r{[}, block@r{]})
888 This function looks up a type by its @var{name}, which must be a string.
890 If @var{block} is given, then @var{name} is looked up in that scope.
891 Otherwise, it is searched for globally.
893 Ordinarily, this function will return an instance of @code{gdb.Type}.
894 If the named type cannot be found, it will throw an exception.
897 If the type is a structure or class type, or an enum type, the fields
898 of that type can be accessed using the Python @dfn{dictionary syntax}.
899 For example, if @code{some_type} is a @code{gdb.Type} instance holding
900 a structure type, you can access its @code{foo} field with:
903 bar = some_type['foo']
906 @code{bar} will be a @code{gdb.Field} object; see below under the
907 description of the @code{Type.fields} method for a description of the
908 @code{gdb.Field} class.
910 An instance of @code{Type} has the following attributes:
913 The type code for this type. The type code will be one of the
914 @code{TYPE_CODE_} constants defined below.
918 The name of this type. If this type has no name, then @code{None}
923 The size of this type, in target @code{char} units. Usually, a
924 target's @code{char} type will be an 8-bit byte. However, on some
925 unusual platforms, this type may have a different size.
929 The tag name for this type. The tag name is the name after
930 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
931 languages have this concept. If this type has no tag name, then
932 @code{None} is returned.
935 The following methods are provided:
937 @defun Type.fields ()
938 For structure and union types, this method returns the fields. Range
939 types have two fields, the minimum and maximum values. Enum types
940 have one field per enum constant. Function and method types have one
941 field per parameter. The base types of C@t{++} classes are also
942 represented as fields. If the type has no fields, or does not fit
943 into one of these categories, an empty sequence will be returned.
945 Each field is a @code{gdb.Field} object, with some pre-defined attributes:
948 This attribute is not available for @code{enum} or @code{static}
949 (as in C@t{++} or Java) fields. The value is the position, counting
950 in bits, from the start of the containing type.
953 This attribute is only available for @code{enum} fields, and its value
954 is the enumeration member's integer representation.
957 The name of the field, or @code{None} for anonymous fields.
960 This is @code{True} if the field is artificial, usually meaning that
961 it was provided by the compiler and not the user. This attribute is
962 always provided, and is @code{False} if the field is not artificial.
965 This is @code{True} if the field represents a base class of a C@t{++}
966 structure. This attribute is always provided, and is @code{False}
967 if the field is not a base class of the type that is the argument of
968 @code{fields}, or if that type was not a C@t{++} class.
971 If the field is packed, or is a bitfield, then this will have a
972 non-zero value, which is the size of the field in bits. Otherwise,
973 this will be zero; in this case the field's size is given by its type.
976 The type of the field. This is usually an instance of @code{Type},
977 but it can be @code{None} in some situations.
980 The type which contains this field. This is an instance of
985 @defun Type.array (@var{n1} @r{[}, @var{n2}@r{]})
986 Return a new @code{gdb.Type} object which represents an array of this
987 type. If one argument is given, it is the inclusive upper bound of
988 the array; in this case the lower bound is zero. If two arguments are
989 given, the first argument is the lower bound of the array, and the
990 second argument is the upper bound of the array. An array's length
991 must not be negative, but the bounds can be.
994 @defun Type.vector (@var{n1} @r{[}, @var{n2}@r{]})
995 Return a new @code{gdb.Type} object which represents a vector of this
996 type. If one argument is given, it is the inclusive upper bound of
997 the vector; in this case the lower bound is zero. If two arguments are
998 given, the first argument is the lower bound of the vector, and the
999 second argument is the upper bound of the vector. A vector's length
1000 must not be negative, but the bounds can be.
1002 The difference between an @code{array} and a @code{vector} is that
1003 arrays behave like in C: when used in expressions they decay to a pointer
1004 to the first element whereas vectors are treated as first class values.
1007 @defun Type.const ()
1008 Return a new @code{gdb.Type} object which represents a
1009 @code{const}-qualified variant of this type.
1012 @defun Type.volatile ()
1013 Return a new @code{gdb.Type} object which represents a
1014 @code{volatile}-qualified variant of this type.
1017 @defun Type.unqualified ()
1018 Return a new @code{gdb.Type} object which represents an unqualified
1019 variant of this type. That is, the result is neither @code{const} nor
1023 @defun Type.range ()
1024 Return a Python @code{Tuple} object that contains two elements: the
1025 low bound of the argument type and the high bound of that type. If
1026 the type does not have a range, @value{GDBN} will raise a
1027 @code{gdb.error} exception (@pxref{Exception Handling}).
1030 @defun Type.reference ()
1031 Return a new @code{gdb.Type} object which represents a reference to this
1035 @defun Type.pointer ()
1036 Return a new @code{gdb.Type} object which represents a pointer to this
1040 @defun Type.strip_typedefs ()
1041 Return a new @code{gdb.Type} that represents the real type,
1042 after removing all layers of typedefs.
1045 @defun Type.target ()
1046 Return a new @code{gdb.Type} object which represents the target type
1049 For a pointer type, the target type is the type of the pointed-to
1050 object. For an array type (meaning C-like arrays), the target type is
1051 the type of the elements of the array. For a function or method type,
1052 the target type is the type of the return value. For a complex type,
1053 the target type is the type of the elements. For a typedef, the
1054 target type is the aliased type.
1056 If the type does not have a target, this method will throw an
1060 @defun Type.template_argument (n @r{[}, block@r{]})
1061 If this @code{gdb.Type} is an instantiation of a template, this will
1062 return a new @code{gdb.Value} or @code{gdb.Type} which represents the
1063 value of the @var{n}th template argument (indexed starting at 0).
1065 If this @code{gdb.Type} is not a template type, or if the type has fewer
1066 than @var{n} template arguments, this will throw an exception.
1067 Ordinarily, only C@t{++} code will have template types.
1069 If @var{block} is given, then @var{name} is looked up in that scope.
1070 Otherwise, it is searched for globally.
1073 @defun Type.optimized_out ()
1074 Return @code{gdb.Value} instance of this type whose value is optimized
1075 out. This allows a frame decorator to indicate that the value of an
1076 argument or a local variable is not known.
1079 Each type has a code, which indicates what category this type falls
1080 into. The available type categories are represented by constants
1081 defined in the @code{gdb} module:
1084 @vindex TYPE_CODE_PTR
1085 @item gdb.TYPE_CODE_PTR
1086 The type is a pointer.
1088 @vindex TYPE_CODE_ARRAY
1089 @item gdb.TYPE_CODE_ARRAY
1090 The type is an array.
1092 @vindex TYPE_CODE_STRUCT
1093 @item gdb.TYPE_CODE_STRUCT
1094 The type is a structure.
1096 @vindex TYPE_CODE_UNION
1097 @item gdb.TYPE_CODE_UNION
1098 The type is a union.
1100 @vindex TYPE_CODE_ENUM
1101 @item gdb.TYPE_CODE_ENUM
1102 The type is an enum.
1104 @vindex TYPE_CODE_FLAGS
1105 @item gdb.TYPE_CODE_FLAGS
1106 A bit flags type, used for things such as status registers.
1108 @vindex TYPE_CODE_FUNC
1109 @item gdb.TYPE_CODE_FUNC
1110 The type is a function.
1112 @vindex TYPE_CODE_INT
1113 @item gdb.TYPE_CODE_INT
1114 The type is an integer type.
1116 @vindex TYPE_CODE_FLT
1117 @item gdb.TYPE_CODE_FLT
1118 A floating point type.
1120 @vindex TYPE_CODE_VOID
1121 @item gdb.TYPE_CODE_VOID
1122 The special type @code{void}.
1124 @vindex TYPE_CODE_SET
1125 @item gdb.TYPE_CODE_SET
1128 @vindex TYPE_CODE_RANGE
1129 @item gdb.TYPE_CODE_RANGE
1130 A range type, that is, an integer type with bounds.
1132 @vindex TYPE_CODE_STRING
1133 @item gdb.TYPE_CODE_STRING
1134 A string type. Note that this is only used for certain languages with
1135 language-defined string types; C strings are not represented this way.
1137 @vindex TYPE_CODE_BITSTRING
1138 @item gdb.TYPE_CODE_BITSTRING
1139 A string of bits. It is deprecated.
1141 @vindex TYPE_CODE_ERROR
1142 @item gdb.TYPE_CODE_ERROR
1143 An unknown or erroneous type.
1145 @vindex TYPE_CODE_METHOD
1146 @item gdb.TYPE_CODE_METHOD
1147 A method type, as found in C@t{++} or Java.
1149 @vindex TYPE_CODE_METHODPTR
1150 @item gdb.TYPE_CODE_METHODPTR
1151 A pointer-to-member-function.
1153 @vindex TYPE_CODE_MEMBERPTR
1154 @item gdb.TYPE_CODE_MEMBERPTR
1155 A pointer-to-member.
1157 @vindex TYPE_CODE_REF
1158 @item gdb.TYPE_CODE_REF
1161 @vindex TYPE_CODE_CHAR
1162 @item gdb.TYPE_CODE_CHAR
1165 @vindex TYPE_CODE_BOOL
1166 @item gdb.TYPE_CODE_BOOL
1169 @vindex TYPE_CODE_COMPLEX
1170 @item gdb.TYPE_CODE_COMPLEX
1171 A complex float type.
1173 @vindex TYPE_CODE_TYPEDEF
1174 @item gdb.TYPE_CODE_TYPEDEF
1175 A typedef to some other type.
1177 @vindex TYPE_CODE_NAMESPACE
1178 @item gdb.TYPE_CODE_NAMESPACE
1179 A C@t{++} namespace.
1181 @vindex TYPE_CODE_DECFLOAT
1182 @item gdb.TYPE_CODE_DECFLOAT
1183 A decimal floating point type.
1185 @vindex TYPE_CODE_INTERNAL_FUNCTION
1186 @item gdb.TYPE_CODE_INTERNAL_FUNCTION
1187 A function internal to @value{GDBN}. This is the type used to represent
1188 convenience functions.
1191 Further support for types is provided in the @code{gdb.types}
1192 Python module (@pxref{gdb.types}).
1194 @node Pretty Printing API
1195 @subsubsection Pretty Printing API
1196 @cindex python pretty printing api
1198 An example output is provided (@pxref{Pretty Printing}).
1200 A pretty-printer is just an object that holds a value and implements a
1201 specific interface, defined here.
1203 @defun pretty_printer.children (self)
1204 @value{GDBN} will call this method on a pretty-printer to compute the
1205 children of the pretty-printer's value.
1207 This method must return an object conforming to the Python iterator
1208 protocol. Each item returned by the iterator must be a tuple holding
1209 two elements. The first element is the ``name'' of the child; the
1210 second element is the child's value. The value can be any Python
1211 object which is convertible to a @value{GDBN} value.
1213 This method is optional. If it does not exist, @value{GDBN} will act
1214 as though the value has no children.
1217 @defun pretty_printer.display_hint (self)
1218 The CLI may call this method and use its result to change the
1219 formatting of a value. The result will also be supplied to an MI
1220 consumer as a @samp{displayhint} attribute of the variable being
1223 This method is optional. If it does exist, this method must return a
1226 Some display hints are predefined by @value{GDBN}:
1230 Indicate that the object being printed is ``array-like''. The CLI
1231 uses this to respect parameters such as @code{set print elements} and
1232 @code{set print array}.
1235 Indicate that the object being printed is ``map-like'', and that the
1236 children of this value can be assumed to alternate between keys and
1240 Indicate that the object being printed is ``string-like''. If the
1241 printer's @code{to_string} method returns a Python string of some
1242 kind, then @value{GDBN} will call its internal language-specific
1243 string-printing function to format the string. For the CLI this means
1244 adding quotation marks, possibly escaping some characters, respecting
1245 @code{set print elements}, and the like.
1249 @defun pretty_printer.to_string (self)
1250 @value{GDBN} will call this method to display the string
1251 representation of the value passed to the object's constructor.
1253 When printing from the CLI, if the @code{to_string} method exists,
1254 then @value{GDBN} will prepend its result to the values returned by
1255 @code{children}. Exactly how this formatting is done is dependent on
1256 the display hint, and may change as more hints are added. Also,
1257 depending on the print settings (@pxref{Print Settings}), the CLI may
1258 print just the result of @code{to_string} in a stack trace, omitting
1259 the result of @code{children}.
1261 If this method returns a string, it is printed verbatim.
1263 Otherwise, if this method returns an instance of @code{gdb.Value},
1264 then @value{GDBN} prints this value. This may result in a call to
1265 another pretty-printer.
1267 If instead the method returns a Python value which is convertible to a
1268 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
1269 the resulting value. Again, this may result in a call to another
1270 pretty-printer. Python scalars (integers, floats, and booleans) and
1271 strings are convertible to @code{gdb.Value}; other types are not.
1273 Finally, if this method returns @code{None} then no further operations
1274 are peformed in this method and nothing is printed.
1276 If the result is not one of these types, an exception is raised.
1279 @value{GDBN} provides a function which can be used to look up the
1280 default pretty-printer for a @code{gdb.Value}:
1282 @findex gdb.default_visualizer
1283 @defun gdb.default_visualizer (value)
1284 This function takes a @code{gdb.Value} object as an argument. If a
1285 pretty-printer for this value exists, then it is returned. If no such
1286 printer exists, then this returns @code{None}.
1289 @node Selecting Pretty-Printers
1290 @subsubsection Selecting Pretty-Printers
1291 @cindex selecting python pretty-printers
1293 The Python list @code{gdb.pretty_printers} contains an array of
1294 functions or callable objects that have been registered via addition
1295 as a pretty-printer. Printers in this list are called @code{global}
1296 printers, they're available when debugging all inferiors.
1297 Each @code{gdb.Progspace} contains a @code{pretty_printers} attribute.
1298 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
1301 Each function on these lists is passed a single @code{gdb.Value}
1302 argument and should return a pretty-printer object conforming to the
1303 interface definition above (@pxref{Pretty Printing API}). If a function
1304 cannot create a pretty-printer for the value, it should return
1307 @value{GDBN} first checks the @code{pretty_printers} attribute of each
1308 @code{gdb.Objfile} in the current program space and iteratively calls
1309 each enabled lookup routine in the list for that @code{gdb.Objfile}
1310 until it receives a pretty-printer object.
1311 If no pretty-printer is found in the objfile lists, @value{GDBN} then
1312 searches the pretty-printer list of the current program space,
1313 calling each enabled function until an object is returned.
1314 After these lists have been exhausted, it tries the global
1315 @code{gdb.pretty_printers} list, again calling each enabled function until an
1318 The order in which the objfiles are searched is not specified. For a
1319 given list, functions are always invoked from the head of the list,
1320 and iterated over sequentially until the end of the list, or a printer
1323 For various reasons a pretty-printer may not work.
1324 For example, the underlying data structure may have changed and
1325 the pretty-printer is out of date.
1327 The consequences of a broken pretty-printer are severe enough that
1328 @value{GDBN} provides support for enabling and disabling individual
1329 printers. For example, if @code{print frame-arguments} is on,
1330 a backtrace can become highly illegible if any argument is printed
1331 with a broken printer.
1333 Pretty-printers are enabled and disabled by attaching an @code{enabled}
1334 attribute to the registered function or callable object. If this attribute
1335 is present and its value is @code{False}, the printer is disabled, otherwise
1336 the printer is enabled.
1338 @node Writing a Pretty-Printer
1339 @subsubsection Writing a Pretty-Printer
1340 @cindex writing a pretty-printer
1342 A pretty-printer consists of two parts: a lookup function to detect
1343 if the type is supported, and the printer itself.
1345 Here is an example showing how a @code{std::string} printer might be
1346 written. @xref{Pretty Printing API}, for details on the API this class
1350 class StdStringPrinter(object):
1351 "Print a std::string"
1353 def __init__(self, val):
1356 def to_string(self):
1357 return self.val['_M_dataplus']['_M_p']
1359 def display_hint(self):
1363 And here is an example showing how a lookup function for the printer
1364 example above might be written.
1367 def str_lookup_function(val):
1368 lookup_tag = val.type.tag
1369 if lookup_tag == None:
1371 regex = re.compile("^std::basic_string<char,.*>$")
1372 if regex.match(lookup_tag):
1373 return StdStringPrinter(val)
1377 The example lookup function extracts the value's type, and attempts to
1378 match it to a type that it can pretty-print. If it is a type the
1379 printer can pretty-print, it will return a printer object. If not, it
1380 returns @code{None}.
1382 We recommend that you put your core pretty-printers into a Python
1383 package. If your pretty-printers are for use with a library, we
1384 further recommend embedding a version number into the package name.
1385 This practice will enable @value{GDBN} to load multiple versions of
1386 your pretty-printers at the same time, because they will have
1389 You should write auto-loaded code (@pxref{Python Auto-loading}) such that it
1390 can be evaluated multiple times without changing its meaning. An
1391 ideal auto-load file will consist solely of @code{import}s of your
1392 printer modules, followed by a call to a register pretty-printers with
1393 the current objfile.
1395 Taken as a whole, this approach will scale nicely to multiple
1396 inferiors, each potentially using a different library version.
1397 Embedding a version number in the Python package name will ensure that
1398 @value{GDBN} is able to load both sets of printers simultaneously.
1399 Then, because the search for pretty-printers is done by objfile, and
1400 because your auto-loaded code took care to register your library's
1401 printers with a specific objfile, @value{GDBN} will find the correct
1402 printers for the specific version of the library used by each
1405 To continue the @code{std::string} example (@pxref{Pretty Printing API}),
1406 this code might appear in @code{gdb.libstdcxx.v6}:
1409 def register_printers(objfile):
1410 objfile.pretty_printers.append(str_lookup_function)
1414 And then the corresponding contents of the auto-load file would be:
1417 import gdb.libstdcxx.v6
1418 gdb.libstdcxx.v6.register_printers(gdb.current_objfile())
1421 The previous example illustrates a basic pretty-printer.
1422 There are a few things that can be improved on.
1423 The printer doesn't have a name, making it hard to identify in a
1424 list of installed printers. The lookup function has a name, but
1425 lookup functions can have arbitrary, even identical, names.
1427 Second, the printer only handles one type, whereas a library typically has
1428 several types. One could install a lookup function for each desired type
1429 in the library, but one could also have a single lookup function recognize
1430 several types. The latter is the conventional way this is handled.
1431 If a pretty-printer can handle multiple data types, then its
1432 @dfn{subprinters} are the printers for the individual data types.
1434 The @code{gdb.printing} module provides a formal way of solving these
1435 problems (@pxref{gdb.printing}).
1436 Here is another example that handles multiple types.
1438 These are the types we are going to pretty-print:
1441 struct foo @{ int a, b; @};
1442 struct bar @{ struct foo x, y; @};
1445 Here are the printers:
1449 """Print a foo object."""
1451 def __init__(self, val):
1454 def to_string(self):
1455 return ("a=<" + str(self.val["a"]) +
1456 "> b=<" + str(self.val["b"]) + ">")
1459 """Print a bar object."""
1461 def __init__(self, val):
1464 def to_string(self):
1465 return ("x=<" + str(self.val["x"]) +
1466 "> y=<" + str(self.val["y"]) + ">")
1469 This example doesn't need a lookup function, that is handled by the
1470 @code{gdb.printing} module. Instead a function is provided to build up
1471 the object that handles the lookup.
1476 def build_pretty_printer():
1477 pp = gdb.printing.RegexpCollectionPrettyPrinter(
1479 pp.add_printer('foo', '^foo$', fooPrinter)
1480 pp.add_printer('bar', '^bar$', barPrinter)
1484 And here is the autoload support:
1489 gdb.printing.register_pretty_printer(
1490 gdb.current_objfile(),
1491 my_library.build_pretty_printer())
1494 Finally, when this printer is loaded into @value{GDBN}, here is the
1495 corresponding output of @samp{info pretty-printer}:
1498 (gdb) info pretty-printer
1505 @node Type Printing API
1506 @subsubsection Type Printing API
1507 @cindex type printing API for Python
1509 @value{GDBN} provides a way for Python code to customize type display.
1510 This is mainly useful for substituting canonical typedef names for
1513 @cindex type printer
1514 A @dfn{type printer} is just a Python object conforming to a certain
1515 protocol. A simple base class implementing the protocol is provided;
1516 see @ref{gdb.types}. A type printer must supply at least:
1518 @defivar type_printer enabled
1519 A boolean which is True if the printer is enabled, and False
1520 otherwise. This is manipulated by the @code{enable type-printer}
1521 and @code{disable type-printer} commands.
1524 @defivar type_printer name
1525 The name of the type printer. This must be a string. This is used by
1526 the @code{enable type-printer} and @code{disable type-printer}
1530 @defmethod type_printer instantiate (self)
1531 This is called by @value{GDBN} at the start of type-printing. It is
1532 only called if the type printer is enabled. This method must return a
1533 new object that supplies a @code{recognize} method, as described below.
1537 When displaying a type, say via the @code{ptype} command, @value{GDBN}
1538 will compute a list of type recognizers. This is done by iterating
1539 first over the per-objfile type printers (@pxref{Objfiles In Python}),
1540 followed by the per-progspace type printers (@pxref{Progspaces In
1541 Python}), and finally the global type printers.
1543 @value{GDBN} will call the @code{instantiate} method of each enabled
1544 type printer. If this method returns @code{None}, then the result is
1545 ignored; otherwise, it is appended to the list of recognizers.
1547 Then, when @value{GDBN} is going to display a type name, it iterates
1548 over the list of recognizers. For each one, it calls the recognition
1549 function, stopping if the function returns a non-@code{None} value.
1550 The recognition function is defined as:
1552 @defmethod type_recognizer recognize (self, type)
1553 If @var{type} is not recognized, return @code{None}. Otherwise,
1554 return a string which is to be printed as the name of @var{type}.
1555 The @var{type} argument will be an instance of @code{gdb.Type}
1556 (@pxref{Types In Python}).
1559 @value{GDBN} uses this two-pass approach so that type printers can
1560 efficiently cache information without holding on to it too long. For
1561 example, it can be convenient to look up type information in a type
1562 printer and hold it for a recognizer's lifetime; if a single pass were
1563 done then type printers would have to make use of the event system in
1564 order to avoid holding information that could become stale as the
1567 @node Frame Filter API
1568 @subsubsection Filtering Frames.
1569 @cindex frame filters api
1571 Frame filters are Python objects that manipulate the visibility of a
1572 frame or frames when a backtrace (@pxref{Backtrace}) is printed by
1575 Only commands that print a backtrace, or, in the case of @sc{gdb/mi}
1576 commands (@pxref{GDB/MI}), those that return a collection of frames
1577 are affected. The commands that work with frame filters are:
1579 @code{backtrace} (@pxref{backtrace-command,, The backtrace command}),
1580 @code{-stack-list-frames}
1581 (@pxref{-stack-list-frames,, The -stack-list-frames command}),
1582 @code{-stack-list-variables} (@pxref{-stack-list-variables,, The
1583 -stack-list-variables command}), @code{-stack-list-arguments}
1584 @pxref{-stack-list-arguments,, The -stack-list-arguments command}) and
1585 @code{-stack-list-locals} (@pxref{-stack-list-locals,, The
1586 -stack-list-locals command}).
1588 A frame filter works by taking an iterator as an argument, applying
1589 actions to the contents of that iterator, and returning another
1590 iterator (or, possibly, the same iterator it was provided in the case
1591 where the filter does not perform any operations). Typically, frame
1592 filters utilize tools such as the Python's @code{itertools} module to
1593 work with and create new iterators from the source iterator.
1594 Regardless of how a filter chooses to apply actions, it must not alter
1595 the underlying @value{GDBN} frame or frames, or attempt to alter the
1596 call-stack within @value{GDBN}. This preserves data integrity within
1597 @value{GDBN}. Frame filters are executed on a priority basis and care
1598 should be taken that some frame filters may have been executed before,
1599 and that some frame filters will be executed after.
1601 An important consideration when designing frame filters, and well
1602 worth reflecting upon, is that frame filters should avoid unwinding
1603 the call stack if possible. Some stacks can run very deep, into the
1604 tens of thousands in some cases. To search every frame when a frame
1605 filter executes may be too expensive at that step. The frame filter
1606 cannot know how many frames it has to iterate over, and it may have to
1607 iterate through them all. This ends up duplicating effort as
1608 @value{GDBN} performs this iteration when it prints the frames. If
1609 the filter can defer unwinding frames until frame decorators are
1610 executed, after the last filter has executed, it should. @xref{Frame
1611 Decorator API}, for more information on decorators. Also, there are
1612 examples for both frame decorators and filters in later chapters.
1613 @xref{Writing a Frame Filter}, for more information.
1615 The Python dictionary @code{gdb.frame_filters} contains key/object
1616 pairings that comprise a frame filter. Frame filters in this
1617 dictionary are called @code{global} frame filters, and they are
1618 available when debugging all inferiors. These frame filters must
1619 register with the dictionary directly. In addition to the
1620 @code{global} dictionary, there are other dictionaries that are loaded
1621 with different inferiors via auto-loading (@pxref{Python
1622 Auto-loading}). The two other areas where frame filter dictionaries
1623 can be found are: @code{gdb.Progspace} which contains a
1624 @code{frame_filters} dictionary attribute, and each @code{gdb.Objfile}
1625 object which also contains a @code{frame_filters} dictionary
1628 When a command is executed from @value{GDBN} that is compatible with
1629 frame filters, @value{GDBN} combines the @code{global},
1630 @code{gdb.Progspace} and all @code{gdb.Objfile} dictionaries currently
1631 loaded. All of the @code{gdb.Objfile} dictionaries are combined, as
1632 several frames, and thus several object files, might be in use.
1633 @value{GDBN} then prunes any frame filter whose @code{enabled}
1634 attribute is @code{False}. This pruned list is then sorted according
1635 to the @code{priority} attribute in each filter.
1637 Once the dictionaries are combined, pruned and sorted, @value{GDBN}
1638 creates an iterator which wraps each frame in the call stack in a
1639 @code{FrameDecorator} object, and calls each filter in order. The
1640 output from the previous filter will always be the input to the next
1643 Frame filters have a mandatory interface which each frame filter must
1644 implement, defined here:
1646 @defun FrameFilter.filter (iterator)
1647 @value{GDBN} will call this method on a frame filter when it has
1648 reached the order in the priority list for that filter.
1650 For example, if there are four frame filters:
1661 The order that the frame filters will be called is:
1664 Filter3 -> Filter2 -> Filter1 -> Filter4
1667 Note that the output from @code{Filter3} is passed to the input of
1668 @code{Filter2}, and so on.
1670 This @code{filter} method is passed a Python iterator. This iterator
1671 contains a sequence of frame decorators that wrap each
1672 @code{gdb.Frame}, or a frame decorator that wraps another frame
1673 decorator. The first filter that is executed in the sequence of frame
1674 filters will receive an iterator entirely comprised of default
1675 @code{FrameDecorator} objects. However, after each frame filter is
1676 executed, the previous frame filter may have wrapped some or all of
1677 the frame decorators with their own frame decorator. As frame
1678 decorators must also conform to a mandatory interface, these
1679 decorators can be assumed to act in a uniform manner (@pxref{Frame
1682 This method must return an object conforming to the Python iterator
1683 protocol. Each item in the iterator must be an object conforming to
1684 the frame decorator interface. If a frame filter does not wish to
1685 perform any operations on this iterator, it should return that
1688 This method is not optional. If it does not exist, @value{GDBN} will
1689 raise and print an error.
1692 @defvar FrameFilter.name
1693 The @code{name} attribute must be Python string which contains the
1694 name of the filter displayed by @value{GDBN} (@pxref{Frame Filter
1695 Management}). This attribute may contain any combination of letters
1696 or numbers. Care should be taken to ensure that it is unique. This
1697 attribute is mandatory.
1700 @defvar FrameFilter.enabled
1701 The @code{enabled} attribute must be Python boolean. This attribute
1702 indicates to @value{GDBN} whether the frame filter is enabled, and
1703 should be considered when frame filters are executed. If
1704 @code{enabled} is @code{True}, then the frame filter will be executed
1705 when any of the backtrace commands detailed earlier in this chapter
1706 are executed. If @code{enabled} is @code{False}, then the frame
1707 filter will not be executed. This attribute is mandatory.
1710 @defvar FrameFilter.priority
1711 The @code{priority} attribute must be Python integer. This attribute
1712 controls the order of execution in relation to other frame filters.
1713 There are no imposed limits on the range of @code{priority} other than
1714 it must be a valid integer. The higher the @code{priority} attribute,
1715 the sooner the frame filter will be executed in relation to other
1716 frame filters. Although @code{priority} can be negative, it is
1717 recommended practice to assume zero is the lowest priority that a
1718 frame filter can be assigned. Frame filters that have the same
1719 priority are executed in unsorted order in that priority slot. This
1720 attribute is mandatory.
1723 @node Frame Decorator API
1724 @subsubsection Decorating Frames.
1725 @cindex frame decorator api
1727 Frame decorators are sister objects to frame filters (@pxref{Frame
1728 Filter API}). Frame decorators are applied by a frame filter and can
1729 only be used in conjunction with frame filters.
1731 The purpose of a frame decorator is to customize the printed content
1732 of each @code{gdb.Frame} in commands where frame filters are executed.
1733 This concept is called decorating a frame. Frame decorators decorate
1734 a @code{gdb.Frame} with Python code contained within each API call.
1735 This separates the actual data contained in a @code{gdb.Frame} from
1736 the decorated data produced by a frame decorator. This abstraction is
1737 necessary to maintain integrity of the data contained in each
1740 Frame decorators have a mandatory interface, defined below.
1742 @value{GDBN} already contains a frame decorator called
1743 @code{FrameDecorator}. This contains substantial amounts of
1744 boilerplate code to decorate the content of a @code{gdb.Frame}. It is
1745 recommended that other frame decorators inherit and extend this
1746 object, and only to override the methods needed.
1748 @defun FrameDecorator.elided (self)
1750 The @code{elided} method groups frames together in a hierarchical
1751 system. An example would be an interpreter, where multiple low-level
1752 frames make up a single call in the interpreted language. In this
1753 example, the frame filter would elide the low-level frames and present
1754 a single high-level frame, representing the call in the interpreted
1755 language, to the user.
1757 The @code{elided} function must return an iterable and this iterable
1758 must contain the frames that are being elided wrapped in a suitable
1759 frame decorator. If no frames are being elided this function may
1760 return an empty iterable, or @code{None}. Elided frames are indented
1761 from normal frames in a @code{CLI} backtrace, or in the case of
1762 @code{GDB/MI}, are placed in the @code{children} field of the eliding
1765 It is the frame filter's task to also filter out the elided frames from
1766 the source iterator. This will avoid printing the frame twice.
1769 @defun FrameDecorator.function (self)
1771 This method returns the name of the function in the frame that is to
1774 This method must return a Python string describing the function, or
1777 If this function returns @code{None}, @value{GDBN} will not print any
1778 data for this field.
1781 @defun FrameDecorator.address (self)
1783 This method returns the address of the frame that is to be printed.
1785 This method must return a Python numeric integer type of sufficient
1786 size to describe the address of the frame, or @code{None}.
1788 If this function returns a @code{None}, @value{GDBN} will not print
1789 any data for this field.
1792 @defun FrameDecorator.filename (self)
1794 This method returns the filename and path associated with this frame.
1796 This method must return a Python string containing the filename and
1797 the path to the object file backing the frame, or @code{None}.
1799 If this function returns a @code{None}, @value{GDBN} will not print
1800 any data for this field.
1803 @defun FrameDecorator.line (self):
1805 This method returns the line number associated with the current
1806 position within the function addressed by this frame.
1808 This method must return a Python integer type, or @code{None}.
1810 If this function returns a @code{None}, @value{GDBN} will not print
1811 any data for this field.
1814 @defun FrameDecorator.frame_args (self)
1817 This method must return an iterable, or @code{None}. Returning an
1818 empty iterable, or @code{None} means frame arguments will not be
1819 printed for this frame. This iterable must contain objects that
1820 implement two methods, described here.
1822 This object must implement a @code{argument} method which takes a
1823 single @code{self} parameter and must return a @code{gdb.Symbol}
1824 (@pxref{Symbols In Python}), or a Python string. The object must also
1825 implement a @code{value} method which takes a single @code{self}
1826 parameter and must return a @code{gdb.Value} (@pxref{Values From
1827 Inferior}), a Python value, or @code{None}. If the @code{value}
1828 method returns @code{None}, and the @code{argument} method returns a
1829 @code{gdb.Symbol}, @value{GDBN} will look-up and print the value of
1830 the @code{gdb.Symbol} automatically.
1835 class SymValueWrapper():
1837 def __init__(self, symbol, value):
1847 class SomeFrameDecorator()
1850 def frame_args(self):
1853 block = self.inferior_frame.block()
1857 # Iterate over all symbols in a block. Only add
1858 # symbols that are arguments.
1860 if not sym.is_argument:
1862 args.append(SymValueWrapper(sym,None))
1864 # Add example synthetic argument.
1865 args.append(SymValueWrapper(``foo'', 42))
1871 @defun FrameDecorator.frame_locals (self)
1873 This method must return an iterable or @code{None}. Returning an
1874 empty iterable, or @code{None} means frame local arguments will not be
1875 printed for this frame.
1877 The object interface, the description of the various strategies for
1878 reading frame locals, and the example are largely similar to those
1879 described in the @code{frame_args} function, (@pxref{frame_args,,The
1880 frame filter frame_args function}). Below is a modified example:
1883 class SomeFrameDecorator()
1886 def frame_locals(self):
1889 block = self.inferior_frame.block()
1893 # Iterate over all symbols in a block. Add all
1894 # symbols, except arguments.
1898 vars.append(SymValueWrapper(sym,None))
1900 # Add an example of a synthetic local variable.
1901 vars.append(SymValueWrapper(``bar'', 99))
1907 @defun FrameDecorator.inferior_frame (self):
1909 This method must return the underlying @code{gdb.Frame} that this
1910 frame decorator is decorating. @value{GDBN} requires the underlying
1911 frame for internal frame information to determine how to print certain
1912 values when printing a frame.
1915 @node Writing a Frame Filter
1916 @subsubsection Writing a Frame Filter
1917 @cindex writing a frame filter
1919 There are three basic elements that a frame filter must implement: it
1920 must correctly implement the documented interface (@pxref{Frame Filter
1921 API}), it must register itself with @value{GDBN}, and finally, it must
1922 decide if it is to work on the data provided by @value{GDBN}. In all
1923 cases, whether it works on the iterator or not, each frame filter must
1924 return an iterator. A bare-bones frame filter follows the pattern in
1925 the following example.
1930 class FrameFilter():
1933 # Frame filter attribute creation.
1935 # 'name' is the name of the filter that GDB will display.
1937 # 'priority' is the priority of the filter relative to other
1940 # 'enabled' is a boolean that indicates whether this filter is
1941 # enabled and should be executed.
1947 # Register this frame filter with the global frame_filters
1949 gdb.frame_filters[self.name] = self
1951 def filter(self, frame_iter):
1952 # Just return the iterator.
1956 The frame filter in the example above implements the three
1957 requirements for all frame filters. It implements the API, self
1958 registers, and makes a decision on the iterator (in this case, it just
1959 returns the iterator untouched).
1961 The first step is attribute creation and assignment, and as shown in
1962 the comments the filter assigns the following attributes: @code{name},
1963 @code{priority} and whether the filter should be enabled with the
1964 @code{enabled} attribute.
1966 The second step is registering the frame filter with the dictionary or
1967 dictionaries that the frame filter has interest in. As shown in the
1968 comments, this filter just registers itself with the global dictionary
1969 @code{gdb.frame_filters}. As noted earlier, @code{gdb.frame_filters}
1970 is a dictionary that is initialized in the @code{gdb} module when
1971 @value{GDBN} starts. What dictionary a filter registers with is an
1972 important consideration. Generally, if a filter is specific to a set
1973 of code, it should be registered either in the @code{objfile} or
1974 @code{progspace} dictionaries as they are specific to the program
1975 currently loaded in @value{GDBN}. The global dictionary is always
1976 present in @value{GDBN} and is never unloaded. Any filters registered
1977 with the global dictionary will exist until @value{GDBN} exits. To
1978 avoid filters that may conflict, it is generally better to register
1979 frame filters against the dictionaries that more closely align with
1980 the usage of the filter currently in question. @xref{Python
1981 Auto-loading}, for further information on auto-loading Python scripts.
1983 @value{GDBN} takes a hands-off approach to frame filter registration,
1984 therefore it is the frame filter's responsibility to ensure
1985 registration has occurred, and that any exceptions are handled
1986 appropriately. In particular, you may wish to handle exceptions
1987 relating to Python dictionary key uniqueness. It is mandatory that
1988 the dictionary key is the same as frame filter's @code{name}
1989 attribute. When a user manages frame filters (@pxref{Frame Filter
1990 Management}), the names @value{GDBN} will display are those contained
1991 in the @code{name} attribute.
1993 The final step of this example is the implementation of the
1994 @code{filter} method. As shown in the example comments, we define the
1995 @code{filter} method and note that the method must take an iterator,
1996 and also must return an iterator. In this bare-bones example, the
1997 frame filter is not very useful as it just returns the iterator
1998 untouched. However this is a valid operation for frame filters that
1999 have the @code{enabled} attribute set, but decide not to operate on
2002 In the next example, the frame filter operates on all frames and
2003 utilizes a frame decorator to perform some work on the frames.
2004 @xref{Frame Decorator API}, for further information on the frame
2005 decorator interface.
2007 This example works on inlined frames. It highlights frames which are
2008 inlined by tagging them with an ``[inlined]'' tag. By applying a
2009 frame decorator to all frames with the Python @code{itertools imap}
2010 method, the example defers actions to the frame decorator. Frame
2011 decorators are only processed when @value{GDBN} prints the backtrace.
2013 This introduces a new decision making topic: whether to perform
2014 decision making operations at the filtering step, or at the printing
2015 step. In this example's approach, it does not perform any filtering
2016 decisions at the filtering step beyond mapping a frame decorator to
2017 each frame. This allows the actual decision making to be performed
2018 when each frame is printed. This is an important consideration, and
2019 well worth reflecting upon when designing a frame filter. An issue
2020 that frame filters should avoid is unwinding the stack if possible.
2021 Some stacks can run very deep, into the tens of thousands in some
2022 cases. To search every frame to determine if it is inlined ahead of
2023 time may be too expensive at the filtering step. The frame filter
2024 cannot know how many frames it has to iterate over, and it would have
2025 to iterate through them all. This ends up duplicating effort as
2026 @value{GDBN} performs this iteration when it prints the frames.
2028 In this example decision making can be deferred to the printing step.
2029 As each frame is printed, the frame decorator can examine each frame
2030 in turn when @value{GDBN} iterates. From a performance viewpoint,
2031 this is the most appropriate decision to make as it avoids duplicating
2032 the effort that the printing step would undertake anyway. Also, if
2033 there are many frame filters unwinding the stack during filtering, it
2034 can substantially delay the printing of the backtrace which will
2035 result in large memory usage, and a poor user experience.
2038 class InlineFilter():
2041 self.name = "InlinedFrameFilter"
2044 gdb.frame_filters[self.name] = self
2046 def filter(self, frame_iter):
2047 frame_iter = itertools.imap(InlinedFrameDecorator,
2052 This frame filter is somewhat similar to the earlier example, except
2053 that the @code{filter} method applies a frame decorator object called
2054 @code{InlinedFrameDecorator} to each element in the iterator. The
2055 @code{imap} Python method is light-weight. It does not proactively
2056 iterate over the iterator, but rather creates a new iterator which
2057 wraps the existing one.
2059 Below is the frame decorator for this example.
2062 class InlinedFrameDecorator(FrameDecorator):
2064 def __init__(self, fobj):
2065 super(InlinedFrameDecorator, self).__init__(fobj)
2068 frame = fobj.inferior_frame()
2069 name = str(frame.name())
2071 if frame.type() == gdb.INLINE_FRAME:
2072 name = name + " [inlined]"
2077 This frame decorator only defines and overrides the @code{function}
2078 method. It lets the supplied @code{FrameDecorator}, which is shipped
2079 with @value{GDBN}, perform the other work associated with printing
2082 The combination of these two objects create this output from a
2086 #0 0x004004e0 in bar () at inline.c:11
2087 #1 0x00400566 in max [inlined] (b=6, a=12) at inline.c:21
2088 #2 0x00400566 in main () at inline.c:31
2091 So in the case of this example, a frame decorator is applied to all
2092 frames, regardless of whether they may be inlined or not. As
2093 @value{GDBN} iterates over the iterator produced by the frame filters,
2094 @value{GDBN} executes each frame decorator which then makes a decision
2095 on what to print in the @code{function} callback. Using a strategy
2096 like this is a way to defer decisions on the frame content to printing
2099 @subheading Eliding Frames
2101 It might be that the above example is not desirable for representing
2102 inlined frames, and a hierarchical approach may be preferred. If we
2103 want to hierarchically represent frames, the @code{elided} frame
2104 decorator interface might be preferable.
2106 This example approaches the issue with the @code{elided} method. This
2107 example is quite long, but very simplistic. It is out-of-scope for
2108 this section to write a complete example that comprehensively covers
2109 all approaches of finding and printing inlined frames. However, this
2110 example illustrates the approach an author might use.
2112 This example comprises of three sections.
2115 class InlineFrameFilter():
2118 self.name = "InlinedFrameFilter"
2121 gdb.frame_filters[self.name] = self
2123 def filter(self, frame_iter):
2124 return ElidingInlineIterator(frame_iter)
2127 This frame filter is very similar to the other examples. The only
2128 difference is this frame filter is wrapping the iterator provided to
2129 it (@code{frame_iter}) with a custom iterator called
2130 @code{ElidingInlineIterator}. This again defers actions to when
2131 @value{GDBN} prints the backtrace, as the iterator is not traversed
2134 The iterator for this example is as follows. It is in this section of
2135 the example where decisions are made on the content of the backtrace.
2138 class ElidingInlineIterator:
2139 def __init__(self, ii):
2140 self.input_iterator = ii
2146 frame = next(self.input_iterator)
2148 if frame.inferior_frame().type() != gdb.INLINE_FRAME:
2152 eliding_frame = next(self.input_iterator)
2153 except StopIteration:
2155 return ElidingFrameDecorator(eliding_frame, [frame])
2158 This iterator implements the Python iterator protocol. When the
2159 @code{next} function is called (when @value{GDBN} prints each frame),
2160 the iterator checks if this frame decorator, @code{frame}, is wrapping
2161 an inlined frame. If it is not, it returns the existing frame decorator
2162 untouched. If it is wrapping an inlined frame, it assumes that the
2163 inlined frame was contained within the next oldest frame,
2164 @code{eliding_frame}, which it fetches. It then creates and returns a
2165 frame decorator, @code{ElidingFrameDecorator}, which contains both the
2166 elided frame, and the eliding frame.
2169 class ElidingInlineDecorator(FrameDecorator):
2171 def __init__(self, frame, elided_frames):
2172 super(ElidingInlineDecorator, self).__init__(frame)
2174 self.elided_frames = elided_frames
2177 return iter(self.elided_frames)
2180 This frame decorator overrides one function and returns the inlined
2181 frame in the @code{elided} method. As before it lets
2182 @code{FrameDecorator} do the rest of the work involved in printing
2183 this frame. This produces the following output.
2186 #0 0x004004e0 in bar () at inline.c:11
2187 #2 0x00400529 in main () at inline.c:25
2188 #1 0x00400529 in max (b=6, a=12) at inline.c:15
2191 In that output, @code{max} which has been inlined into @code{main} is
2192 printed hierarchically. Another approach would be to combine the
2193 @code{function} method, and the @code{elided} method to both print a
2194 marker in the inlined frame, and also show the hierarchical
2197 @node Unwinding Frames in Python
2198 @subsubsection Unwinding Frames in Python
2199 @cindex unwinding frames in Python
2201 In @value{GDBN} terminology ``unwinding'' is the process of finding
2202 the previous frame (that is, caller's) from the current one. An
2203 unwinder has three methods. The first one checks if it can handle
2204 given frame (``sniff'' it). For the frames it can sniff an unwinder
2205 provides two additional methods: it can return frame's ID, and it can
2206 fetch registers from the previous frame. A running @value{GDBN}
2207 mantains a list of the unwinders and calls each unwinder's sniffer in
2208 turn until it finds the one that recognizes the current frame. There
2209 is an API to register an unwinder.
2211 The unwinders that come with @value{GDBN} handle standard frames.
2212 However, mixed language applications (for example, an application
2213 running Java Virtual Machine) sometimes use frame layouts that cannot
2214 be handled by the @value{GDBN} unwinders. You can write Python code
2215 that can handle such custom frames.
2217 You implement a frame unwinder in Python as a class with which has two
2218 attributes, @code{name} and @code{enabled}, with obvious meanings, and
2219 a single method @code{__call__}, which examines a given frame and
2220 returns an object (an instance of @code{gdb.UnwindInfo class)}
2221 describing it. If an unwinder does not recognize a frame, it should
2222 return @code{None}. The code in @value{GDBN} that enables writing
2223 unwinders in Python uses this object to return frame's ID and previous
2224 frame registers when @value{GDBN} core asks for them.
2226 @subheading Unwinder Input
2228 An object passed to an unwinder (a @code{gdb.PendingFrame} instance)
2229 provides a method to read frame's registers:
2231 @defun PendingFrame.read_register (reg)
2232 This method returns the contents of the register @var{regn} in the
2233 frame as a @code{gdb.Value} object. @var{reg} can be either a
2234 register number or a register name; the values are platform-specific.
2235 They are usually found in the corresponding
2236 @file{@var{platform}-tdep.h} file in the @value{GDBN} source tree.
2239 It also provides a factory method to create a @code{gdb.UnwindInfo}
2240 instance to be returned to @value{GDBN}:
2242 @defun PendingFrame.create_unwind_info (frame_id)
2243 Returns a new @code{gdb.UnwindInfo} instance identified by given
2244 @var{frame_id}. The argument is used to build @value{GDBN}'s frame ID
2245 using one of functions provided by @value{GDBN}. @var{frame_id}'s attributes
2246 determine which function will be used, as follows:
2249 @item sp, pc, special
2250 @code{frame_id_build_special (@var{frame_id}.sp, @var{frame_id}.pc, @var{frame_id}.special)}
2253 @code{frame_id_build (@var{frame_id}.sp, @var{frame_id}.pc)}
2255 This is the most common case.
2258 @code{frame_id_build_wild (@var{frame_id}.sp)}
2260 The attribute values should be @code{gdb.Value}
2264 @subheading Unwinder Output: UnwindInfo
2266 Use @code{PendingFrame.create_unwind_info} method described above to
2267 create a @code{gdb.UnwindInfo} instance. Use the following method to
2268 specify caller registers that have been saved in this frame:
2270 @defun gdb.UnwindInfo.add_saved_register (reg, value)
2271 @var{reg} identifies the register. It can be a number or a name, just
2272 as for the @code{PendingFrame.read_register} method above.
2273 @var{value} is a register value (a @code{gdb.Value} object).
2276 @subheading Unwinder Skeleton Code
2278 @value{GDBN} comes with the module containing the base @code{Unwinder}
2279 class. Derive your unwinder class from it and structure the code as
2283 from gdb.unwinders import Unwinder
2285 class FrameId(object):
2286 def __init__(self, sp, pc):
2291 class MyUnwinder(Unwinder):
2293 supe(MyUnwinder, self).__init___(<expects unwinder name argument>)
2295 def __call__(pending_frame):
2296 if not <we recognize frame>:
2298 # Create UnwindInfo. Usually the frame is identified by the stack
2299 # pointer and the program counter.
2300 sp = pending_frame.read_register(<SP number>)
2301 pc = pending_frame.read_register(<PC number>)
2302 unwind_info = pending_frame.create_unwind_info(FrameId(sp, pc))
2304 # Find the values of the registers in the caller's frame and
2305 # save them in the result:
2306 unwind_info.add_saved_register(<register>, <value>)
2309 # Return the result:
2314 @subheading Registering a Unwinder
2316 An object file, a program space, and the @value{GDBN} proper can have
2317 unwinders registered with it.
2319 The @code{gdb.unwinders} module provides the function to register a
2322 @defun gdb.unwinder.register_unwinder (locus, unwinder, replace=False)
2323 @var{locus} is specifies an object file or a program space to which
2324 @var{unwinder} is added. Passing @code{None} or @code{gdb} adds
2325 @var{unwinder} to the @value{GDBN}'s global unwinder list. The newly
2326 added @var{unwinder} will be called before any other unwinder from the
2327 same locus. Two unwinders in the same locus cannot have the same
2328 name. An attempt to add a unwinder with already existing name raises
2329 an exception unless @var{replace} is @code{True}, in which case the
2330 old unwinder is deleted.
2333 @subheading Unwinder Precedence
2335 @value{GDBN} first calls the unwinders from all the object files in no
2336 particular order, then the unwinders from the current program space,
2337 and finally the unwinders from @value{GDBN}.
2339 @node Xmethods In Python
2340 @subsubsection Xmethods In Python
2341 @cindex xmethods in Python
2343 @dfn{Xmethods} are additional methods or replacements for existing
2344 methods of a C@t{++} class. This feature is useful for those cases
2345 where a method defined in C@t{++} source code could be inlined or
2346 optimized out by the compiler, making it unavailable to @value{GDBN}.
2347 For such cases, one can define an xmethod to serve as a replacement
2348 for the method defined in the C@t{++} source code. @value{GDBN} will
2349 then invoke the xmethod, instead of the C@t{++} method, to
2350 evaluate expressions. One can also use xmethods when debugging
2351 with core files. Moreover, when debugging live programs, invoking an
2352 xmethod need not involve running the inferior (which can potentially
2353 perturb its state). Hence, even if the C@t{++} method is available, it
2354 is better to use its replacement xmethod if one is defined.
2356 The xmethods feature in Python is available via the concepts of an
2357 @dfn{xmethod matcher} and an @dfn{xmethod worker}. To
2358 implement an xmethod, one has to implement a matcher and a
2359 corresponding worker for it (more than one worker can be
2360 implemented, each catering to a different overloaded instance of the
2361 method). Internally, @value{GDBN} invokes the @code{match} method of a
2362 matcher to match the class type and method name. On a match, the
2363 @code{match} method returns a list of matching @emph{worker} objects.
2364 Each worker object typically corresponds to an overloaded instance of
2365 the xmethod. They implement a @code{get_arg_types} method which
2366 returns a sequence of types corresponding to the arguments the xmethod
2367 requires. @value{GDBN} uses this sequence of types to perform
2368 overload resolution and picks a winning xmethod worker. A winner
2369 is also selected from among the methods @value{GDBN} finds in the
2370 C@t{++} source code. Next, the winning xmethod worker and the
2371 winning C@t{++} method are compared to select an overall winner. In
2372 case of a tie between a xmethod worker and a C@t{++} method, the
2373 xmethod worker is selected as the winner. That is, if a winning
2374 xmethod worker is found to be equivalent to the winning C@t{++}
2375 method, then the xmethod worker is treated as a replacement for
2376 the C@t{++} method. @value{GDBN} uses the overall winner to invoke the
2377 method. If the winning xmethod worker is the overall winner, then
2378 the corresponding xmethod is invoked via the @code{__call__} method
2379 of the worker object.
2381 If one wants to implement an xmethod as a replacement for an
2382 existing C@t{++} method, then they have to implement an equivalent
2383 xmethod which has exactly the same name and takes arguments of
2384 exactly the same type as the C@t{++} method. If the user wants to
2385 invoke the C@t{++} method even though a replacement xmethod is
2386 available for that method, then they can disable the xmethod.
2388 @xref{Xmethod API}, for API to implement xmethods in Python.
2389 @xref{Writing an Xmethod}, for implementing xmethods in Python.
2392 @subsubsection Xmethod API
2395 The @value{GDBN} Python API provides classes, interfaces and functions
2396 to implement, register and manipulate xmethods.
2397 @xref{Xmethods In Python}.
2399 An xmethod matcher should be an instance of a class derived from
2400 @code{XMethodMatcher} defined in the module @code{gdb.xmethod}, or an
2401 object with similar interface and attributes. An instance of
2402 @code{XMethodMatcher} has the following attributes:
2405 The name of the matcher.
2409 A boolean value indicating whether the matcher is enabled or disabled.
2413 A list of named methods managed by the matcher. Each object in the list
2414 is an instance of the class @code{XMethod} defined in the module
2415 @code{gdb.xmethod}, or any object with the following attributes:
2420 Name of the xmethod which should be unique for each xmethod
2421 managed by the matcher.
2424 A boolean value indicating whether the xmethod is enabled or
2429 The class @code{XMethod} is a convenience class with same
2430 attributes as above along with the following constructor:
2432 @defun XMethod.__init__ (self, name)
2433 Constructs an enabled xmethod with name @var{name}.
2438 The @code{XMethodMatcher} class has the following methods:
2440 @defun XMethodMatcher.__init__ (self, name)
2441 Constructs an enabled xmethod matcher with name @var{name}. The
2442 @code{methods} attribute is initialized to @code{None}.
2445 @defun XMethodMatcher.match (self, class_type, method_name)
2446 Derived classes should override this method. It should return a
2447 xmethod worker object (or a sequence of xmethod worker
2448 objects) matching the @var{class_type} and @var{method_name}.
2449 @var{class_type} is a @code{gdb.Type} object, and @var{method_name}
2450 is a string value. If the matcher manages named methods as listed in
2451 its @code{methods} attribute, then only those worker objects whose
2452 corresponding entries in the @code{methods} list are enabled should be
2456 An xmethod worker should be an instance of a class derived from
2457 @code{XMethodWorker} defined in the module @code{gdb.xmethod},
2458 or support the following interface:
2460 @defun XMethodWorker.get_arg_types (self)
2461 This method returns a sequence of @code{gdb.Type} objects corresponding
2462 to the arguments that the xmethod takes. It can return an empty
2463 sequence or @code{None} if the xmethod does not take any arguments.
2464 If the xmethod takes a single argument, then a single
2465 @code{gdb.Type} object corresponding to it can be returned.
2468 @defun XMethodWorker.get_result_type (self, *args)
2469 This method returns a @code{gdb.Type} object representing the type
2470 of the result of invoking this xmethod.
2471 The @var{args} argument is the same tuple of arguments that would be
2472 passed to the @code{__call__} method of this worker.
2475 @defun XMethodWorker.__call__ (self, *args)
2476 This is the method which does the @emph{work} of the xmethod. The
2477 @var{args} arguments is the tuple of arguments to the xmethod. Each
2478 element in this tuple is a gdb.Value object. The first element is
2479 always the @code{this} pointer value.
2482 For @value{GDBN} to lookup xmethods, the xmethod matchers
2483 should be registered using the following function defined in the module
2486 @defun register_xmethod_matcher (locus, matcher, replace=False)
2487 The @code{matcher} is registered with @code{locus}, replacing an
2488 existing matcher with the same name as @code{matcher} if
2489 @code{replace} is @code{True}. @code{locus} can be a
2490 @code{gdb.Objfile} object (@pxref{Objfiles In Python}), or a
2491 @code{gdb.Progspace} object (@pxref{Progspaces In Python}), or
2492 @code{None}. If it is @code{None}, then @code{matcher} is registered
2496 @node Writing an Xmethod
2497 @subsubsection Writing an Xmethod
2498 @cindex writing xmethods in Python
2500 Implementing xmethods in Python will require implementing xmethod
2501 matchers and xmethod workers (@pxref{Xmethods In Python}). Consider
2502 the following C@t{++} class:
2508 MyClass (int a) : a_(a) @{ @}
2510 int geta (void) @{ return a_; @}
2511 int operator+ (int b);
2518 MyClass::operator+ (int b)
2525 Let us define two xmethods for the class @code{MyClass}, one
2526 replacing the method @code{geta}, and another adding an overloaded
2527 flavor of @code{operator+} which takes a @code{MyClass} argument (the
2528 C@t{++} code above already has an overloaded @code{operator+}
2529 which takes an @code{int} argument). The xmethod matcher can be
2533 class MyClass_geta(gdb.xmethod.XMethod):
2535 gdb.xmethod.XMethod.__init__(self, 'geta')
2537 def get_worker(self, method_name):
2538 if method_name == 'geta':
2539 return MyClassWorker_geta()
2542 class MyClass_sum(gdb.xmethod.XMethod):
2544 gdb.xmethod.XMethod.__init__(self, 'sum')
2546 def get_worker(self, method_name):
2547 if method_name == 'operator+':
2548 return MyClassWorker_plus()
2551 class MyClassMatcher(gdb.xmethod.XMethodMatcher):
2553 gdb.xmethod.XMethodMatcher.__init__(self, 'MyClassMatcher')
2554 # List of methods 'managed' by this matcher
2555 self.methods = [MyClass_geta(), MyClass_sum()]
2557 def match(self, class_type, method_name):
2558 if class_type.tag != 'MyClass':
2561 for method in self.methods:
2563 worker = method.get_worker(method_name)
2565 workers.append(worker)
2571 Notice that the @code{match} method of @code{MyClassMatcher} returns
2572 a worker object of type @code{MyClassWorker_geta} for the @code{geta}
2573 method, and a worker object of type @code{MyClassWorker_plus} for the
2574 @code{operator+} method. This is done indirectly via helper classes
2575 derived from @code{gdb.xmethod.XMethod}. One does not need to use the
2576 @code{methods} attribute in a matcher as it is optional. However, if a
2577 matcher manages more than one xmethod, it is a good practice to list the
2578 xmethods in the @code{methods} attribute of the matcher. This will then
2579 facilitate enabling and disabling individual xmethods via the
2580 @code{enable/disable} commands. Notice also that a worker object is
2581 returned only if the corresponding entry in the @code{methods} attribute
2582 of the matcher is enabled.
2584 The implementation of the worker classes returned by the matcher setup
2585 above is as follows:
2588 class MyClassWorker_geta(gdb.xmethod.XMethodWorker):
2589 def get_arg_types(self):
2592 def get_result_type(self, obj):
2593 return gdb.lookup_type('int')
2595 def __call__(self, obj):
2599 class MyClassWorker_plus(gdb.xmethod.XMethodWorker):
2600 def get_arg_types(self):
2601 return gdb.lookup_type('MyClass')
2603 def get_result_type(self, obj):
2604 return gdb.lookup_type('int')
2606 def __call__(self, obj, other):
2607 return obj['a_'] + other['a_']
2610 For @value{GDBN} to actually lookup a xmethod, it has to be
2611 registered with it. The matcher defined above is registered with
2612 @value{GDBN} globally as follows:
2615 gdb.xmethod.register_xmethod_matcher(None, MyClassMatcher())
2618 If an object @code{obj} of type @code{MyClass} is initialized in C@t{++}
2626 then, after loading the Python script defining the xmethod matchers
2627 and workers into @code{GDBN}, invoking the method @code{geta} or using
2628 the operator @code{+} on @code{obj} will invoke the xmethods
2639 Consider another example with a C++ template class:
2646 MyTemplate () : dsize_(10), data_ (new T [10]) @{ @}
2647 ~MyTemplate () @{ delete [] data_; @}
2649 int footprint (void)
2651 return sizeof (T) * dsize_ + sizeof (MyTemplate<T>);
2660 Let us implement an xmethod for the above class which serves as a
2661 replacement for the @code{footprint} method. The full code listing
2662 of the xmethod workers and xmethod matchers is as follows:
2665 class MyTemplateWorker_footprint(gdb.xmethod.XMethodWorker):
2666 def __init__(self, class_type):
2667 self.class_type = class_type
2669 def get_arg_types(self):
2672 def get_result_type(self):
2673 return gdb.lookup_type('int')
2675 def __call__(self, obj):
2676 return (self.class_type.sizeof +
2678 self.class_type.template_argument(0).sizeof)
2681 class MyTemplateMatcher_footprint(gdb.xmethod.XMethodMatcher):
2683 gdb.xmethod.XMethodMatcher.__init__(self, 'MyTemplateMatcher')
2685 def match(self, class_type, method_name):
2686 if (re.match('MyTemplate<[ \t\n]*[_a-zA-Z][ _a-zA-Z0-9]*>',
2688 method_name == 'footprint'):
2689 return MyTemplateWorker_footprint(class_type)
2692 Notice that, in this example, we have not used the @code{methods}
2693 attribute of the matcher as the matcher manages only one xmethod. The
2694 user can enable/disable this xmethod by enabling/disabling the matcher
2697 @node Inferiors In Python
2698 @subsubsection Inferiors In Python
2699 @cindex inferiors in Python
2701 @findex gdb.Inferior
2702 Programs which are being run under @value{GDBN} are called inferiors
2703 (@pxref{Inferiors and Programs}). Python scripts can access
2704 information about and manipulate inferiors controlled by @value{GDBN}
2705 via objects of the @code{gdb.Inferior} class.
2707 The following inferior-related functions are available in the @code{gdb}
2710 @defun gdb.inferiors ()
2711 Return a tuple containing all inferior objects.
2714 @defun gdb.selected_inferior ()
2715 Return an object representing the current inferior.
2718 A @code{gdb.Inferior} object has the following attributes:
2720 @defvar Inferior.num
2721 ID of inferior, as assigned by GDB.
2724 @defvar Inferior.pid
2725 Process ID of the inferior, as assigned by the underlying operating
2729 @defvar Inferior.was_attached
2730 Boolean signaling whether the inferior was created using `attach', or
2731 started by @value{GDBN} itself.
2734 A @code{gdb.Inferior} object has the following methods:
2736 @defun Inferior.is_valid ()
2737 Returns @code{True} if the @code{gdb.Inferior} object is valid,
2738 @code{False} if not. A @code{gdb.Inferior} object will become invalid
2739 if the inferior no longer exists within @value{GDBN}. All other
2740 @code{gdb.Inferior} methods will throw an exception if it is invalid
2741 at the time the method is called.
2744 @defun Inferior.threads ()
2745 This method returns a tuple holding all the threads which are valid
2746 when it is called. If there are no valid threads, the method will
2747 return an empty tuple.
2750 @findex Inferior.read_memory
2751 @defun Inferior.read_memory (address, length)
2752 Read @var{length} bytes of memory from the inferior, starting at
2753 @var{address}. Returns a buffer object, which behaves much like an array
2754 or a string. It can be modified and given to the
2755 @code{Inferior.write_memory} function. In @code{Python} 3, the return
2756 value is a @code{memoryview} object.
2759 @findex Inferior.write_memory
2760 @defun Inferior.write_memory (address, buffer @r{[}, length@r{]})
2761 Write the contents of @var{buffer} to the inferior, starting at
2762 @var{address}. The @var{buffer} parameter must be a Python object
2763 which supports the buffer protocol, i.e., a string, an array or the
2764 object returned from @code{Inferior.read_memory}. If given, @var{length}
2765 determines the number of bytes from @var{buffer} to be written.
2768 @findex gdb.search_memory
2769 @defun Inferior.search_memory (address, length, pattern)
2770 Search a region of the inferior memory starting at @var{address} with
2771 the given @var{length} using the search pattern supplied in
2772 @var{pattern}. The @var{pattern} parameter must be a Python object
2773 which supports the buffer protocol, i.e., a string, an array or the
2774 object returned from @code{gdb.read_memory}. Returns a Python @code{Long}
2775 containing the address where the pattern was found, or @code{None} if
2776 the pattern could not be found.
2779 @node Events In Python
2780 @subsubsection Events In Python
2781 @cindex inferior events in Python
2783 @value{GDBN} provides a general event facility so that Python code can be
2784 notified of various state changes, particularly changes that occur in
2787 An @dfn{event} is just an object that describes some state change. The
2788 type of the object and its attributes will vary depending on the details
2789 of the change. All the existing events are described below.
2791 In order to be notified of an event, you must register an event handler
2792 with an @dfn{event registry}. An event registry is an object in the
2793 @code{gdb.events} module which dispatches particular events. A registry
2794 provides methods to register and unregister event handlers:
2796 @defun EventRegistry.connect (object)
2797 Add the given callable @var{object} to the registry. This object will be
2798 called when an event corresponding to this registry occurs.
2801 @defun EventRegistry.disconnect (object)
2802 Remove the given @var{object} from the registry. Once removed, the object
2803 will no longer receive notifications of events.
2809 def exit_handler (event):
2810 print "event type: exit"
2811 print "exit code: %d" % (event.exit_code)
2813 gdb.events.exited.connect (exit_handler)
2816 In the above example we connect our handler @code{exit_handler} to the
2817 registry @code{events.exited}. Once connected, @code{exit_handler} gets
2818 called when the inferior exits. The argument @dfn{event} in this example is
2819 of type @code{gdb.ExitedEvent}. As you can see in the example the
2820 @code{ExitedEvent} object has an attribute which indicates the exit code of
2823 The following is a listing of the event registries that are available and
2824 details of the events they emit:
2829 Emits @code{gdb.ThreadEvent}.
2831 Some events can be thread specific when @value{GDBN} is running in non-stop
2832 mode. When represented in Python, these events all extend
2833 @code{gdb.ThreadEvent}. Note, this event is not emitted directly; instead,
2834 events which are emitted by this or other modules might extend this event.
2835 Examples of these events are @code{gdb.BreakpointEvent} and
2836 @code{gdb.ContinueEvent}.
2838 @defvar ThreadEvent.inferior_thread
2839 In non-stop mode this attribute will be set to the specific thread which was
2840 involved in the emitted event. Otherwise, it will be set to @code{None}.
2843 Emits @code{gdb.ContinueEvent} which extends @code{gdb.ThreadEvent}.
2845 This event indicates that the inferior has been continued after a stop. For
2846 inherited attribute refer to @code{gdb.ThreadEvent} above.
2849 Emits @code{events.ExitedEvent} which indicates that the inferior has exited.
2850 @code{events.ExitedEvent} has two attributes:
2851 @defvar ExitedEvent.exit_code
2852 An integer representing the exit code, if available, which the inferior
2853 has returned. (The exit code could be unavailable if, for example,
2854 @value{GDBN} detaches from the inferior.) If the exit code is unavailable,
2855 the attribute does not exist.
2857 @defvar ExitedEvent inferior
2858 A reference to the inferior which triggered the @code{exited} event.
2862 Emits @code{gdb.StopEvent} which extends @code{gdb.ThreadEvent}.
2864 Indicates that the inferior has stopped. All events emitted by this registry
2865 extend StopEvent. As a child of @code{gdb.ThreadEvent}, @code{gdb.StopEvent}
2866 will indicate the stopped thread when @value{GDBN} is running in non-stop
2867 mode. Refer to @code{gdb.ThreadEvent} above for more details.
2869 Emits @code{gdb.SignalEvent} which extends @code{gdb.StopEvent}.
2871 This event indicates that the inferior or one of its threads has received as
2872 signal. @code{gdb.SignalEvent} has the following attributes:
2874 @defvar SignalEvent.stop_signal
2875 A string representing the signal received by the inferior. A list of possible
2876 signal values can be obtained by running the command @code{info signals} in
2877 the @value{GDBN} command prompt.
2880 Also emits @code{gdb.BreakpointEvent} which extends @code{gdb.StopEvent}.
2882 @code{gdb.BreakpointEvent} event indicates that one or more breakpoints have
2883 been hit, and has the following attributes:
2885 @defvar BreakpointEvent.breakpoints
2886 A sequence containing references to all the breakpoints (type
2887 @code{gdb.Breakpoint}) that were hit.
2888 @xref{Breakpoints In Python}, for details of the @code{gdb.Breakpoint} object.
2890 @defvar BreakpointEvent.breakpoint
2891 A reference to the first breakpoint that was hit.
2892 This function is maintained for backward compatibility and is now deprecated
2893 in favor of the @code{gdb.BreakpointEvent.breakpoints} attribute.
2896 @item events.new_objfile
2897 Emits @code{gdb.NewObjFileEvent} which indicates that a new object file has
2898 been loaded by @value{GDBN}. @code{gdb.NewObjFileEvent} has one attribute:
2900 @defvar NewObjFileEvent.new_objfile
2901 A reference to the object file (@code{gdb.Objfile}) which has been loaded.
2902 @xref{Objfiles In Python}, for details of the @code{gdb.Objfile} object.
2905 @item events.clear_objfiles
2906 Emits @code{gdb.ClearObjFilesEvent} which indicates that the list of object
2907 files for a program space has been reset.
2908 @code{gdb.ClearObjFilesEvent} has one attribute:
2910 @defvar ClearObjFilesEvent.progspace
2911 A reference to the program space (@code{gdb.Progspace}) whose objfile list has
2912 been cleared. @xref{Progspaces In Python}.
2915 @item events.inferior_call_pre
2916 Emits @code{gdb.InferiorCallPreEvent} which indicates that a function in
2917 the inferior is about to be called.
2919 @defvar InferiorCallPreEvent.ptid
2920 The thread in which the call will be run.
2923 @defvar InferiorCallPreEvent.address
2924 The location of the function to be called.
2927 @item events.inferior_call_post
2928 Emits @code{gdb.InferiorCallPostEvent} which indicates that a function in
2929 the inferior has returned.
2931 @defvar InferiorCallPostEvent.ptid
2932 The thread in which the call was run.
2935 @defvar InferiorCallPostEvent.address
2936 The location of the function that was called.
2939 @item events.memory_changed
2940 Emits @code{gdb.MemoryChangedEvent} which indicates that the memory of the
2941 inferior has been modified by the @value{GDBN} user, for instance via a
2942 command like @w{@code{set *addr = value}}. The event has the following
2945 @defvar MemoryChangedEvent.address
2946 The start address of the changed region.
2949 @defvar MemoryChangedEvent.length
2950 Length in bytes of the changed region.
2953 @item events.register_changed
2954 Emits @code{gdb.RegisterChangedEvent} which indicates that a register in the
2955 inferior has been modified by the @value{GDBN} user.
2957 @defvar RegisterChangedEvent.frame
2958 A gdb.Frame object representing the frame in which the register was modified.
2960 @defvar RegisterChangedEvent.regnum
2961 Denotes which register was modified.
2966 @node Threads In Python
2967 @subsubsection Threads In Python
2968 @cindex threads in python
2970 @findex gdb.InferiorThread
2971 Python scripts can access information about, and manipulate inferior threads
2972 controlled by @value{GDBN}, via objects of the @code{gdb.InferiorThread} class.
2974 The following thread-related functions are available in the @code{gdb}
2977 @findex gdb.selected_thread
2978 @defun gdb.selected_thread ()
2979 This function returns the thread object for the selected thread. If there
2980 is no selected thread, this will return @code{None}.
2983 A @code{gdb.InferiorThread} object has the following attributes:
2985 @defvar InferiorThread.name
2986 The name of the thread. If the user specified a name using
2987 @code{thread name}, then this returns that name. Otherwise, if an
2988 OS-supplied name is available, then it is returned. Otherwise, this
2989 returns @code{None}.
2991 This attribute can be assigned to. The new value must be a string
2992 object, which sets the new name, or @code{None}, which removes any
2993 user-specified thread name.
2996 @defvar InferiorThread.num
2997 ID of the thread, as assigned by GDB.
3000 @defvar InferiorThread.ptid
3001 ID of the thread, as assigned by the operating system. This attribute is a
3002 tuple containing three integers. The first is the Process ID (PID); the second
3003 is the Lightweight Process ID (LWPID), and the third is the Thread ID (TID).
3004 Either the LWPID or TID may be 0, which indicates that the operating system
3005 does not use that identifier.
3008 A @code{gdb.InferiorThread} object has the following methods:
3010 @defun InferiorThread.is_valid ()
3011 Returns @code{True} if the @code{gdb.InferiorThread} object is valid,
3012 @code{False} if not. A @code{gdb.InferiorThread} object will become
3013 invalid if the thread exits, or the inferior that the thread belongs
3014 is deleted. All other @code{gdb.InferiorThread} methods will throw an
3015 exception if it is invalid at the time the method is called.
3018 @defun InferiorThread.switch ()
3019 This changes @value{GDBN}'s currently selected thread to the one represented
3023 @defun InferiorThread.is_stopped ()
3024 Return a Boolean indicating whether the thread is stopped.
3027 @defun InferiorThread.is_running ()
3028 Return a Boolean indicating whether the thread is running.
3031 @defun InferiorThread.is_exited ()
3032 Return a Boolean indicating whether the thread is exited.
3035 @node Commands In Python
3036 @subsubsection Commands In Python
3038 @cindex commands in python
3039 @cindex python commands
3040 You can implement new @value{GDBN} CLI commands in Python. A CLI
3041 command is implemented using an instance of the @code{gdb.Command}
3042 class, most commonly using a subclass.
3044 @defun Command.__init__ (name, @var{command_class} @r{[}, @var{completer_class} @r{[}, @var{prefix}@r{]]})
3045 The object initializer for @code{Command} registers the new command
3046 with @value{GDBN}. This initializer is normally invoked from the
3047 subclass' own @code{__init__} method.
3049 @var{name} is the name of the command. If @var{name} consists of
3050 multiple words, then the initial words are looked for as prefix
3051 commands. In this case, if one of the prefix commands does not exist,
3052 an exception is raised.
3054 There is no support for multi-line commands.
3056 @var{command_class} should be one of the @samp{COMMAND_} constants
3057 defined below. This argument tells @value{GDBN} how to categorize the
3058 new command in the help system.
3060 @var{completer_class} is an optional argument. If given, it should be
3061 one of the @samp{COMPLETE_} constants defined below. This argument
3062 tells @value{GDBN} how to perform completion for this command. If not
3063 given, @value{GDBN} will attempt to complete using the object's
3064 @code{complete} method (see below); if no such method is found, an
3065 error will occur when completion is attempted.
3067 @var{prefix} is an optional argument. If @code{True}, then the new
3068 command is a prefix command; sub-commands of this command may be
3071 The help text for the new command is taken from the Python
3072 documentation string for the command's class, if there is one. If no
3073 documentation string is provided, the default value ``This command is
3074 not documented.'' is used.
3077 @cindex don't repeat Python command
3078 @defun Command.dont_repeat ()
3079 By default, a @value{GDBN} command is repeated when the user enters a
3080 blank line at the command prompt. A command can suppress this
3081 behavior by invoking the @code{dont_repeat} method. This is similar
3082 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
3085 @defun Command.invoke (argument, from_tty)
3086 This method is called by @value{GDBN} when this command is invoked.
3088 @var{argument} is a string. It is the argument to the command, after
3089 leading and trailing whitespace has been stripped.
3091 @var{from_tty} is a boolean argument. When true, this means that the
3092 command was entered by the user at the terminal; when false it means
3093 that the command came from elsewhere.
3095 If this method throws an exception, it is turned into a @value{GDBN}
3096 @code{error} call. Otherwise, the return value is ignored.
3098 @findex gdb.string_to_argv
3099 To break @var{argument} up into an argv-like string use
3100 @code{gdb.string_to_argv}. This function behaves identically to
3101 @value{GDBN}'s internal argument lexer @code{buildargv}.
3102 It is recommended to use this for consistency.
3103 Arguments are separated by spaces and may be quoted.
3107 print gdb.string_to_argv ("1 2\ \\\"3 '4 \"5' \"6 '7\"")
3108 ['1', '2 "3', '4 "5', "6 '7"]
3113 @cindex completion of Python commands
3114 @defun Command.complete (text, word)
3115 This method is called by @value{GDBN} when the user attempts
3116 completion on this command. All forms of completion are handled by
3117 this method, that is, the @key{TAB} and @key{M-?} key bindings
3118 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
3121 The arguments @var{text} and @var{word} are both strings; @var{text}
3122 holds the complete command line up to the cursor's location, while
3123 @var{word} holds the last word of the command line; this is computed
3124 using a word-breaking heuristic.
3126 The @code{complete} method can return several values:
3129 If the return value is a sequence, the contents of the sequence are
3130 used as the completions. It is up to @code{complete} to ensure that the
3131 contents actually do complete the word. A zero-length sequence is
3132 allowed, it means that there were no completions available. Only
3133 string elements of the sequence are used; other elements in the
3134 sequence are ignored.
3137 If the return value is one of the @samp{COMPLETE_} constants defined
3138 below, then the corresponding @value{GDBN}-internal completion
3139 function is invoked, and its result is used.
3142 All other results are treated as though there were no available
3147 When a new command is registered, it must be declared as a member of
3148 some general class of commands. This is used to classify top-level
3149 commands in the on-line help system; note that prefix commands are not
3150 listed under their own category but rather that of their top-level
3151 command. The available classifications are represented by constants
3152 defined in the @code{gdb} module:
3155 @findex COMMAND_NONE
3156 @findex gdb.COMMAND_NONE
3157 @item gdb.COMMAND_NONE
3158 The command does not belong to any particular class. A command in
3159 this category will not be displayed in any of the help categories.
3161 @findex COMMAND_RUNNING
3162 @findex gdb.COMMAND_RUNNING
3163 @item gdb.COMMAND_RUNNING
3164 The command is related to running the inferior. For example,
3165 @code{start}, @code{step}, and @code{continue} are in this category.
3166 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
3167 commands in this category.
3169 @findex COMMAND_DATA
3170 @findex gdb.COMMAND_DATA
3171 @item gdb.COMMAND_DATA
3172 The command is related to data or variables. For example,
3173 @code{call}, @code{find}, and @code{print} are in this category. Type
3174 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
3177 @findex COMMAND_STACK
3178 @findex gdb.COMMAND_STACK
3179 @item gdb.COMMAND_STACK
3180 The command has to do with manipulation of the stack. For example,
3181 @code{backtrace}, @code{frame}, and @code{return} are in this
3182 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
3183 list of commands in this category.
3185 @findex COMMAND_FILES
3186 @findex gdb.COMMAND_FILES
3187 @item gdb.COMMAND_FILES
3188 This class is used for file-related commands. For example,
3189 @code{file}, @code{list} and @code{section} are in this category.
3190 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
3191 commands in this category.
3193 @findex COMMAND_SUPPORT
3194 @findex gdb.COMMAND_SUPPORT
3195 @item gdb.COMMAND_SUPPORT
3196 This should be used for ``support facilities'', generally meaning
3197 things that are useful to the user when interacting with @value{GDBN},
3198 but not related to the state of the inferior. For example,
3199 @code{help}, @code{make}, and @code{shell} are in this category. Type
3200 @kbd{help support} at the @value{GDBN} prompt to see a list of
3201 commands in this category.
3203 @findex COMMAND_STATUS
3204 @findex gdb.COMMAND_STATUS
3205 @item gdb.COMMAND_STATUS
3206 The command is an @samp{info}-related command, that is, related to the
3207 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
3208 and @code{show} are in this category. Type @kbd{help status} at the
3209 @value{GDBN} prompt to see a list of commands in this category.
3211 @findex COMMAND_BREAKPOINTS
3212 @findex gdb.COMMAND_BREAKPOINTS
3213 @item gdb.COMMAND_BREAKPOINTS
3214 The command has to do with breakpoints. For example, @code{break},
3215 @code{clear}, and @code{delete} are in this category. Type @kbd{help
3216 breakpoints} at the @value{GDBN} prompt to see a list of commands in
3219 @findex COMMAND_TRACEPOINTS
3220 @findex gdb.COMMAND_TRACEPOINTS
3221 @item gdb.COMMAND_TRACEPOINTS
3222 The command has to do with tracepoints. For example, @code{trace},
3223 @code{actions}, and @code{tfind} are in this category. Type
3224 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
3225 commands in this category.
3227 @findex COMMAND_USER
3228 @findex gdb.COMMAND_USER
3229 @item gdb.COMMAND_USER
3230 The command is a general purpose command for the user, and typically
3231 does not fit in one of the other categories.
3232 Type @kbd{help user-defined} at the @value{GDBN} prompt to see
3233 a list of commands in this category, as well as the list of gdb macros
3234 (@pxref{Sequences}).
3236 @findex COMMAND_OBSCURE
3237 @findex gdb.COMMAND_OBSCURE
3238 @item gdb.COMMAND_OBSCURE
3239 The command is only used in unusual circumstances, or is not of
3240 general interest to users. For example, @code{checkpoint},
3241 @code{fork}, and @code{stop} are in this category. Type @kbd{help
3242 obscure} at the @value{GDBN} prompt to see a list of commands in this
3245 @findex COMMAND_MAINTENANCE
3246 @findex gdb.COMMAND_MAINTENANCE
3247 @item gdb.COMMAND_MAINTENANCE
3248 The command is only useful to @value{GDBN} maintainers. The
3249 @code{maintenance} and @code{flushregs} commands are in this category.
3250 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
3251 commands in this category.
3254 A new command can use a predefined completion function, either by
3255 specifying it via an argument at initialization, or by returning it
3256 from the @code{complete} method. These predefined completion
3257 constants are all defined in the @code{gdb} module:
3260 @vindex COMPLETE_NONE
3261 @item gdb.COMPLETE_NONE
3262 This constant means that no completion should be done.
3264 @vindex COMPLETE_FILENAME
3265 @item gdb.COMPLETE_FILENAME
3266 This constant means that filename completion should be performed.
3268 @vindex COMPLETE_LOCATION
3269 @item gdb.COMPLETE_LOCATION
3270 This constant means that location completion should be done.
3271 @xref{Specify Location}.
3273 @vindex COMPLETE_COMMAND
3274 @item gdb.COMPLETE_COMMAND
3275 This constant means that completion should examine @value{GDBN}
3278 @vindex COMPLETE_SYMBOL
3279 @item gdb.COMPLETE_SYMBOL
3280 This constant means that completion should be done using symbol names
3283 @vindex COMPLETE_EXPRESSION
3284 @item gdb.COMPLETE_EXPRESSION
3285 This constant means that completion should be done on expressions.
3286 Often this means completing on symbol names, but some language
3287 parsers also have support for completing on field names.
3290 The following code snippet shows how a trivial CLI command can be
3291 implemented in Python:
3294 class HelloWorld (gdb.Command):
3295 """Greet the whole world."""
3297 def __init__ (self):
3298 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_USER)
3300 def invoke (self, arg, from_tty):
3301 print "Hello, World!"
3306 The last line instantiates the class, and is necessary to trigger the
3307 registration of the command with @value{GDBN}. Depending on how the
3308 Python code is read into @value{GDBN}, you may need to import the
3309 @code{gdb} module explicitly.
3311 @node Parameters In Python
3312 @subsubsection Parameters In Python
3314 @cindex parameters in python
3315 @cindex python parameters
3316 @tindex gdb.Parameter
3318 You can implement new @value{GDBN} parameters using Python. A new
3319 parameter is implemented as an instance of the @code{gdb.Parameter}
3322 Parameters are exposed to the user via the @code{set} and
3323 @code{show} commands. @xref{Help}.
3325 There are many parameters that already exist and can be set in
3326 @value{GDBN}. Two examples are: @code{set follow fork} and
3327 @code{set charset}. Setting these parameters influences certain
3328 behavior in @value{GDBN}. Similarly, you can define parameters that
3329 can be used to influence behavior in custom Python scripts and commands.
3331 @defun Parameter.__init__ (name, @var{command-class}, @var{parameter-class} @r{[}, @var{enum-sequence}@r{]})
3332 The object initializer for @code{Parameter} registers the new
3333 parameter with @value{GDBN}. This initializer is normally invoked
3334 from the subclass' own @code{__init__} method.
3336 @var{name} is the name of the new parameter. If @var{name} consists
3337 of multiple words, then the initial words are looked for as prefix
3338 parameters. An example of this can be illustrated with the
3339 @code{set print} set of parameters. If @var{name} is
3340 @code{print foo}, then @code{print} will be searched as the prefix
3341 parameter. In this case the parameter can subsequently be accessed in
3342 @value{GDBN} as @code{set print foo}.
3344 If @var{name} consists of multiple words, and no prefix parameter group
3345 can be found, an exception is raised.
3347 @var{command-class} should be one of the @samp{COMMAND_} constants
3348 (@pxref{Commands In Python}). This argument tells @value{GDBN} how to
3349 categorize the new parameter in the help system.
3351 @var{parameter-class} should be one of the @samp{PARAM_} constants
3352 defined below. This argument tells @value{GDBN} the type of the new
3353 parameter; this information is used for input validation and
3356 If @var{parameter-class} is @code{PARAM_ENUM}, then
3357 @var{enum-sequence} must be a sequence of strings. These strings
3358 represent the possible values for the parameter.
3360 If @var{parameter-class} is not @code{PARAM_ENUM}, then the presence
3361 of a fourth argument will cause an exception to be thrown.
3363 The help text for the new parameter is taken from the Python
3364 documentation string for the parameter's class, if there is one. If
3365 there is no documentation string, a default value is used.
3368 @defvar Parameter.set_doc
3369 If this attribute exists, and is a string, then its value is used as
3370 the help text for this parameter's @code{set} command. The value is
3371 examined when @code{Parameter.__init__} is invoked; subsequent changes
3375 @defvar Parameter.show_doc
3376 If this attribute exists, and is a string, then its value is used as
3377 the help text for this parameter's @code{show} command. The value is
3378 examined when @code{Parameter.__init__} is invoked; subsequent changes
3382 @defvar Parameter.value
3383 The @code{value} attribute holds the underlying value of the
3384 parameter. It can be read and assigned to just as any other
3385 attribute. @value{GDBN} does validation when assignments are made.
3388 There are two methods that should be implemented in any
3389 @code{Parameter} class. These are:
3391 @defun Parameter.get_set_string (self)
3392 @value{GDBN} will call this method when a @var{parameter}'s value has
3393 been changed via the @code{set} API (for example, @kbd{set foo off}).
3394 The @code{value} attribute has already been populated with the new
3395 value and may be used in output. This method must return a string.
3398 @defun Parameter.get_show_string (self, svalue)
3399 @value{GDBN} will call this method when a @var{parameter}'s
3400 @code{show} API has been invoked (for example, @kbd{show foo}). The
3401 argument @code{svalue} receives the string representation of the
3402 current value. This method must return a string.
3405 When a new parameter is defined, its type must be specified. The
3406 available types are represented by constants defined in the @code{gdb}
3410 @findex PARAM_BOOLEAN
3411 @findex gdb.PARAM_BOOLEAN
3412 @item gdb.PARAM_BOOLEAN
3413 The value is a plain boolean. The Python boolean values, @code{True}
3414 and @code{False} are the only valid values.
3416 @findex PARAM_AUTO_BOOLEAN
3417 @findex gdb.PARAM_AUTO_BOOLEAN
3418 @item gdb.PARAM_AUTO_BOOLEAN
3419 The value has three possible states: true, false, and @samp{auto}. In
3420 Python, true and false are represented using boolean constants, and
3421 @samp{auto} is represented using @code{None}.
3423 @findex PARAM_UINTEGER
3424 @findex gdb.PARAM_UINTEGER
3425 @item gdb.PARAM_UINTEGER
3426 The value is an unsigned integer. The value of 0 should be
3427 interpreted to mean ``unlimited''.
3429 @findex PARAM_INTEGER
3430 @findex gdb.PARAM_INTEGER
3431 @item gdb.PARAM_INTEGER
3432 The value is a signed integer. The value of 0 should be interpreted
3433 to mean ``unlimited''.
3435 @findex PARAM_STRING
3436 @findex gdb.PARAM_STRING
3437 @item gdb.PARAM_STRING
3438 The value is a string. When the user modifies the string, any escape
3439 sequences, such as @samp{\t}, @samp{\f}, and octal escapes, are
3440 translated into corresponding characters and encoded into the current
3443 @findex PARAM_STRING_NOESCAPE
3444 @findex gdb.PARAM_STRING_NOESCAPE
3445 @item gdb.PARAM_STRING_NOESCAPE
3446 The value is a string. When the user modifies the string, escapes are
3447 passed through untranslated.
3449 @findex PARAM_OPTIONAL_FILENAME
3450 @findex gdb.PARAM_OPTIONAL_FILENAME
3451 @item gdb.PARAM_OPTIONAL_FILENAME
3452 The value is a either a filename (a string), or @code{None}.
3454 @findex PARAM_FILENAME
3455 @findex gdb.PARAM_FILENAME
3456 @item gdb.PARAM_FILENAME
3457 The value is a filename. This is just like
3458 @code{PARAM_STRING_NOESCAPE}, but uses file names for completion.
3460 @findex PARAM_ZINTEGER
3461 @findex gdb.PARAM_ZINTEGER
3462 @item gdb.PARAM_ZINTEGER
3463 The value is an integer. This is like @code{PARAM_INTEGER}, except 0
3464 is interpreted as itself.
3467 @findex gdb.PARAM_ENUM
3468 @item gdb.PARAM_ENUM
3469 The value is a string, which must be one of a collection string
3470 constants provided when the parameter is created.
3473 @node Functions In Python
3474 @subsubsection Writing new convenience functions
3476 @cindex writing convenience functions
3477 @cindex convenience functions in python
3478 @cindex python convenience functions
3479 @tindex gdb.Function
3481 You can implement new convenience functions (@pxref{Convenience Vars})
3482 in Python. A convenience function is an instance of a subclass of the
3483 class @code{gdb.Function}.
3485 @defun Function.__init__ (name)
3486 The initializer for @code{Function} registers the new function with
3487 @value{GDBN}. The argument @var{name} is the name of the function,
3488 a string. The function will be visible to the user as a convenience
3489 variable of type @code{internal function}, whose name is the same as
3490 the given @var{name}.
3492 The documentation for the new function is taken from the documentation
3493 string for the new class.
3496 @defun Function.invoke (@var{*args})
3497 When a convenience function is evaluated, its arguments are converted
3498 to instances of @code{gdb.Value}, and then the function's
3499 @code{invoke} method is called. Note that @value{GDBN} does not
3500 predetermine the arity of convenience functions. Instead, all
3501 available arguments are passed to @code{invoke}, following the
3502 standard Python calling convention. In particular, a convenience
3503 function can have default values for parameters without ill effect.
3505 The return value of this method is used as its value in the enclosing
3506 expression. If an ordinary Python value is returned, it is converted
3507 to a @code{gdb.Value} following the usual rules.
3510 The following code snippet shows how a trivial convenience function can
3511 be implemented in Python:
3514 class Greet (gdb.Function):
3515 """Return string to greet someone.
3516 Takes a name as argument."""
3518 def __init__ (self):
3519 super (Greet, self).__init__ ("greet")
3521 def invoke (self, name):
3522 return "Hello, %s!" % name.string ()
3527 The last line instantiates the class, and is necessary to trigger the
3528 registration of the function with @value{GDBN}. Depending on how the
3529 Python code is read into @value{GDBN}, you may need to import the
3530 @code{gdb} module explicitly.
3532 Now you can use the function in an expression:
3535 (gdb) print $greet("Bob")
3539 @node Progspaces In Python
3540 @subsubsection Program Spaces In Python
3542 @cindex progspaces in python
3543 @tindex gdb.Progspace
3545 A program space, or @dfn{progspace}, represents a symbolic view
3546 of an address space.
3547 It consists of all of the objfiles of the program.
3548 @xref{Objfiles In Python}.
3549 @xref{Inferiors and Programs, program spaces}, for more details
3550 about program spaces.
3552 The following progspace-related functions are available in the
3555 @findex gdb.current_progspace
3556 @defun gdb.current_progspace ()
3557 This function returns the program space of the currently selected inferior.
3558 @xref{Inferiors and Programs}.
3561 @findex gdb.progspaces
3562 @defun gdb.progspaces ()
3563 Return a sequence of all the progspaces currently known to @value{GDBN}.
3566 Each progspace is represented by an instance of the @code{gdb.Progspace}
3569 @defvar Progspace.filename
3570 The file name of the progspace as a string.
3573 @defvar Progspace.pretty_printers
3574 The @code{pretty_printers} attribute is a list of functions. It is
3575 used to look up pretty-printers. A @code{Value} is passed to each
3576 function in order; if the function returns @code{None}, then the
3577 search continues. Otherwise, the return value should be an object
3578 which is used to format the value. @xref{Pretty Printing API}, for more
3582 @defvar Progspace.type_printers
3583 The @code{type_printers} attribute is a list of type printer objects.
3584 @xref{Type Printing API}, for more information.
3587 @defvar Progspace.frame_filters
3588 The @code{frame_filters} attribute is a dictionary of frame filter
3589 objects. @xref{Frame Filter API}, for more information.
3592 One may add arbitrary attributes to @code{gdb.Progspace} objects
3593 in the usual Python way.
3594 This is useful if, for example, one needs to do some extra record keeping
3595 associated with the program space.
3597 In this contrived example, we want to perform some processing when
3598 an objfile with a certain symbol is loaded, but we only want to do
3599 this once because it is expensive. To achieve this we record the results
3600 with the program space because we can't predict when the desired objfile
3605 def clear_objfiles_handler(event):
3606 event.progspace.expensive_computation = None
3607 def expensive(symbol):
3608 """A mock routine to perform an "expensive" computation on symbol."""
3609 print "Computing the answer to the ultimate question ..."
3611 def new_objfile_handler(event):
3612 objfile = event.new_objfile
3613 progspace = objfile.progspace
3614 if not hasattr(progspace, 'expensive_computation') or \
3615 progspace.expensive_computation is None:
3616 # We use 'main' for the symbol to keep the example simple.
3617 # Note: There's no current way to constrain the lookup
3619 symbol = gdb.lookup_global_symbol('main')
3620 if symbol is not None:
3621 progspace.expensive_computation = expensive(symbol)
3622 gdb.events.clear_objfiles.connect(clear_objfiles_handler)
3623 gdb.events.new_objfile.connect(new_objfile_handler)
3625 (gdb) file /tmp/hello
3626 Reading symbols from /tmp/hello...done.
3627 Computing the answer to the ultimate question ...
3628 (gdb) python print gdb.current_progspace().expensive_computation
3631 Starting program: /tmp/hello
3633 [Inferior 1 (process 4242) exited normally]
3636 @node Objfiles In Python
3637 @subsubsection Objfiles In Python
3639 @cindex objfiles in python
3642 @value{GDBN} loads symbols for an inferior from various
3643 symbol-containing files (@pxref{Files}). These include the primary
3644 executable file, any shared libraries used by the inferior, and any
3645 separate debug info files (@pxref{Separate Debug Files}).
3646 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
3648 The following objfile-related functions are available in the
3651 @findex gdb.current_objfile
3652 @defun gdb.current_objfile ()
3653 When auto-loading a Python script (@pxref{Python Auto-loading}), @value{GDBN}
3654 sets the ``current objfile'' to the corresponding objfile. This
3655 function returns the current objfile. If there is no current objfile,
3656 this function returns @code{None}.
3659 @findex gdb.objfiles
3660 @defun gdb.objfiles ()
3661 Return a sequence of all the objfiles current known to @value{GDBN}.
3662 @xref{Objfiles In Python}.
3665 @findex gdb.lookup_objfile
3666 @defun gdb.lookup_objfile (name @r{[}, by_build_id{]})
3667 Look up @var{name}, a file name or build ID, in the list of objfiles
3668 for the current program space (@pxref{Progspaces In Python}).
3669 If the objfile is not found throw the Python @code{ValueError} exception.
3671 If @var{name} is a relative file name, then it will match any
3672 source file name with the same trailing components. For example, if
3673 @var{name} is @samp{gcc/expr.c}, then it will match source file
3674 name of @file{/build/trunk/gcc/expr.c}, but not
3675 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
3677 If @var{by_build_id} is provided and is @code{True} then @var{name}
3678 is the build ID of the objfile. Otherwise, @var{name} is a file name.
3679 This is supported only on some operating systems, notably those which use
3680 the ELF format for binary files and the @sc{gnu} Binutils. For more details
3681 about this feature, see the description of the @option{--build-id}
3682 command-line option in @ref{Options, , Command Line Options, ld.info,
3686 Each objfile is represented by an instance of the @code{gdb.Objfile}
3689 @defvar Objfile.filename
3690 The file name of the objfile as a string, with symbolic links resolved.
3692 The value is @code{None} if the objfile is no longer valid.
3693 See the @code{gdb.Objfile.is_valid} method, described below.
3696 @defvar Objfile.username
3697 The file name of the objfile as specified by the user as a string.
3699 The value is @code{None} if the objfile is no longer valid.
3700 See the @code{gdb.Objfile.is_valid} method, described below.
3703 @defvar Objfile.owner
3704 For separate debug info objfiles this is the corresponding @code{gdb.Objfile}
3705 object that debug info is being provided for.
3706 Otherwise this is @code{None}.
3707 Separate debug info objfiles are added with the
3708 @code{gdb.Objfile.add_separate_debug_file} method, described below.
3711 @defvar Objfile.build_id
3712 The build ID of the objfile as a string.
3713 If the objfile does not have a build ID then the value is @code{None}.
3715 This is supported only on some operating systems, notably those which use
3716 the ELF format for binary files and the @sc{gnu} Binutils. For more details
3717 about this feature, see the description of the @option{--build-id}
3718 command-line option in @ref{Options, , Command Line Options, ld.info,
3722 @defvar Objfile.progspace
3723 The containing program space of the objfile as a @code{gdb.Progspace}
3724 object. @xref{Progspaces In Python}.
3727 @defvar Objfile.pretty_printers
3728 The @code{pretty_printers} attribute is a list of functions. It is
3729 used to look up pretty-printers. A @code{Value} is passed to each
3730 function in order; if the function returns @code{None}, then the
3731 search continues. Otherwise, the return value should be an object
3732 which is used to format the value. @xref{Pretty Printing API}, for more
3736 @defvar Objfile.type_printers
3737 The @code{type_printers} attribute is a list of type printer objects.
3738 @xref{Type Printing API}, for more information.
3741 @defvar Objfile.frame_filters
3742 The @code{frame_filters} attribute is a dictionary of frame filter
3743 objects. @xref{Frame Filter API}, for more information.
3746 One may add arbitrary attributes to @code{gdb.Objfile} objects
3747 in the usual Python way.
3748 This is useful if, for example, one needs to do some extra record keeping
3749 associated with the objfile.
3751 In this contrived example we record the time when @value{GDBN}
3757 def new_objfile_handler(event):
3758 # Set the time_loaded attribute of the new objfile.
3759 event.new_objfile.time_loaded = datetime.datetime.today()
3760 gdb.events.new_objfile.connect(new_objfile_handler)
3763 Reading symbols from ./hello...done.
3764 (gdb) python print gdb.objfiles()[0].time_loaded
3765 2014-10-09 11:41:36.770345
3768 A @code{gdb.Objfile} object has the following methods:
3770 @defun Objfile.is_valid ()
3771 Returns @code{True} if the @code{gdb.Objfile} object is valid,
3772 @code{False} if not. A @code{gdb.Objfile} object can become invalid
3773 if the object file it refers to is not loaded in @value{GDBN} any
3774 longer. All other @code{gdb.Objfile} methods will throw an exception
3775 if it is invalid at the time the method is called.
3778 @defun Objfile.add_separate_debug_file (file)
3779 Add @var{file} to the list of files that @value{GDBN} will search for
3780 debug information for the objfile.
3781 This is useful when the debug info has been removed from the program
3782 and stored in a separate file. @value{GDBN} has built-in support for
3783 finding separate debug info files (@pxref{Separate Debug Files}), but if
3784 the file doesn't live in one of the standard places that @value{GDBN}
3785 searches then this function can be used to add a debug info file
3786 from a different place.
3789 @node Frames In Python
3790 @subsubsection Accessing inferior stack frames from Python.
3792 @cindex frames in python
3793 When the debugged program stops, @value{GDBN} is able to analyze its call
3794 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
3795 represents a frame in the stack. A @code{gdb.Frame} object is only valid
3796 while its corresponding frame exists in the inferior's stack. If you try
3797 to use an invalid frame object, @value{GDBN} will throw a @code{gdb.error}
3798 exception (@pxref{Exception Handling}).
3800 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
3804 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
3808 The following frame-related functions are available in the @code{gdb} module:
3810 @findex gdb.selected_frame
3811 @defun gdb.selected_frame ()
3812 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
3815 @findex gdb.newest_frame
3816 @defun gdb.newest_frame ()
3817 Return the newest frame object for the selected thread.
3820 @defun gdb.frame_stop_reason_string (reason)
3821 Return a string explaining the reason why @value{GDBN} stopped unwinding
3822 frames, as expressed by the given @var{reason} code (an integer, see the
3823 @code{unwind_stop_reason} method further down in this section).
3826 A @code{gdb.Frame} object has the following methods:
3828 @defun Frame.is_valid ()
3829 Returns true if the @code{gdb.Frame} object is valid, false if not.
3830 A frame object can become invalid if the frame it refers to doesn't
3831 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
3832 an exception if it is invalid at the time the method is called.
3835 @defun Frame.name ()
3836 Returns the function name of the frame, or @code{None} if it can't be
3840 @defun Frame.architecture ()
3841 Returns the @code{gdb.Architecture} object corresponding to the frame's
3842 architecture. @xref{Architectures In Python}.
3845 @defun Frame.type ()
3846 Returns the type of the frame. The value can be one of:
3848 @item gdb.NORMAL_FRAME
3849 An ordinary stack frame.
3851 @item gdb.DUMMY_FRAME
3852 A fake stack frame that was created by @value{GDBN} when performing an
3853 inferior function call.
3855 @item gdb.INLINE_FRAME
3856 A frame representing an inlined function. The function was inlined
3857 into a @code{gdb.NORMAL_FRAME} that is older than this one.
3859 @item gdb.TAILCALL_FRAME
3860 A frame representing a tail call. @xref{Tail Call Frames}.
3862 @item gdb.SIGTRAMP_FRAME
3863 A signal trampoline frame. This is the frame created by the OS when
3864 it calls into a signal handler.
3866 @item gdb.ARCH_FRAME
3867 A fake stack frame representing a cross-architecture call.
3869 @item gdb.SENTINEL_FRAME
3870 This is like @code{gdb.NORMAL_FRAME}, but it is only used for the
3875 @defun Frame.unwind_stop_reason ()
3876 Return an integer representing the reason why it's not possible to find
3877 more frames toward the outermost frame. Use
3878 @code{gdb.frame_stop_reason_string} to convert the value returned by this
3879 function to a string. The value can be one of:
3882 @item gdb.FRAME_UNWIND_NO_REASON
3883 No particular reason (older frames should be available).
3885 @item gdb.FRAME_UNWIND_NULL_ID
3886 The previous frame's analyzer returns an invalid result. This is no
3887 longer used by @value{GDBN}, and is kept only for backward
3890 @item gdb.FRAME_UNWIND_OUTERMOST
3891 This frame is the outermost.
3893 @item gdb.FRAME_UNWIND_UNAVAILABLE
3894 Cannot unwind further, because that would require knowing the
3895 values of registers or memory that have not been collected.
3897 @item gdb.FRAME_UNWIND_INNER_ID
3898 This frame ID looks like it ought to belong to a NEXT frame,
3899 but we got it for a PREV frame. Normally, this is a sign of
3900 unwinder failure. It could also indicate stack corruption.
3902 @item gdb.FRAME_UNWIND_SAME_ID
3903 This frame has the same ID as the previous one. That means
3904 that unwinding further would almost certainly give us another
3905 frame with exactly the same ID, so break the chain. Normally,
3906 this is a sign of unwinder failure. It could also indicate
3909 @item gdb.FRAME_UNWIND_NO_SAVED_PC
3910 The frame unwinder did not find any saved PC, but we needed
3911 one to unwind further.
3913 @item gdb.FRAME_UNWIND_MEMORY_ERROR
3914 The frame unwinder caused an error while trying to access memory.
3916 @item gdb.FRAME_UNWIND_FIRST_ERROR
3917 Any stop reason greater or equal to this value indicates some kind
3918 of error. This special value facilitates writing code that tests
3919 for errors in unwinding in a way that will work correctly even if
3920 the list of the other values is modified in future @value{GDBN}
3921 versions. Using it, you could write:
3923 reason = gdb.selected_frame().unwind_stop_reason ()
3924 reason_str = gdb.frame_stop_reason_string (reason)
3925 if reason >= gdb.FRAME_UNWIND_FIRST_ERROR:
3926 print "An error occured: %s" % reason_str
3933 Returns the frame's resume address.
3936 @defun Frame.block ()
3937 Return the frame's code block. @xref{Blocks In Python}.
3940 @defun Frame.function ()
3941 Return the symbol for the function corresponding to this frame.
3942 @xref{Symbols In Python}.
3945 @defun Frame.older ()
3946 Return the frame that called this frame.
3949 @defun Frame.newer ()
3950 Return the frame called by this frame.
3953 @defun Frame.find_sal ()
3954 Return the frame's symtab and line object.
3955 @xref{Symbol Tables In Python}.
3958 @defun Frame.read_register (register)
3959 Return the value of @var{register} in this frame. The @var{register}
3960 argument must be a string (e.g., @code{'sp'} or @code{'rax'}).
3961 Returns a @code{Gdb.Value} object. Throws an exception if @var{register}
3965 @defun Frame.read_var (variable @r{[}, block@r{]})
3966 Return the value of @var{variable} in this frame. If the optional
3967 argument @var{block} is provided, search for the variable from that
3968 block; otherwise start at the frame's current block (which is
3969 determined by the frame's current program counter). The @var{variable}
3970 argument must be a string or a @code{gdb.Symbol} object; @var{block} must be a
3971 @code{gdb.Block} object.
3974 @defun Frame.select ()
3975 Set this frame to be the selected frame. @xref{Stack, ,Examining the
3979 @node Blocks In Python
3980 @subsubsection Accessing blocks from Python.
3982 @cindex blocks in python
3985 In @value{GDBN}, symbols are stored in blocks. A block corresponds
3986 roughly to a scope in the source code. Blocks are organized
3987 hierarchically, and are represented individually in Python as a
3988 @code{gdb.Block}. Blocks rely on debugging information being
3991 A frame has a block. Please see @ref{Frames In Python}, for a more
3992 in-depth discussion of frames.
3994 The outermost block is known as the @dfn{global block}. The global
3995 block typically holds public global variables and functions.
3997 The block nested just inside the global block is the @dfn{static
3998 block}. The static block typically holds file-scoped variables and
4001 @value{GDBN} provides a method to get a block's superblock, but there
4002 is currently no way to examine the sub-blocks of a block, or to
4003 iterate over all the blocks in a symbol table (@pxref{Symbol Tables In
4006 Here is a short example that should help explain blocks:
4009 /* This is in the global block. */
4012 /* This is in the static block. */
4013 static int file_scope;
4015 /* 'function' is in the global block, and 'argument' is
4016 in a block nested inside of 'function'. */
4017 int function (int argument)
4019 /* 'local' is in a block inside 'function'. It may or may
4020 not be in the same block as 'argument'. */
4024 /* 'inner' is in a block whose superblock is the one holding
4028 /* If this call is expanded by the compiler, you may see
4029 a nested block here whose function is 'inline_function'
4030 and whose superblock is the one holding 'inner'. */
4036 A @code{gdb.Block} is iterable. The iterator returns the symbols
4037 (@pxref{Symbols In Python}) local to the block. Python programs
4038 should not assume that a specific block object will always contain a
4039 given symbol, since changes in @value{GDBN} features and
4040 infrastructure may cause symbols move across blocks in a symbol
4043 The following block-related functions are available in the @code{gdb}
4046 @findex gdb.block_for_pc
4047 @defun gdb.block_for_pc (pc)
4048 Return the innermost @code{gdb.Block} containing the given @var{pc}
4049 value. If the block cannot be found for the @var{pc} value specified,
4050 the function will return @code{None}.
4053 A @code{gdb.Block} object has the following methods:
4055 @defun Block.is_valid ()
4056 Returns @code{True} if the @code{gdb.Block} object is valid,
4057 @code{False} if not. A block object can become invalid if the block it
4058 refers to doesn't exist anymore in the inferior. All other
4059 @code{gdb.Block} methods will throw an exception if it is invalid at
4060 the time the method is called. The block's validity is also checked
4061 during iteration over symbols of the block.
4064 A @code{gdb.Block} object has the following attributes:
4067 The start address of the block. This attribute is not writable.
4071 The end address of the block. This attribute is not writable.
4074 @defvar Block.function
4075 The name of the block represented as a @code{gdb.Symbol}. If the
4076 block is not named, then this attribute holds @code{None}. This
4077 attribute is not writable.
4079 For ordinary function blocks, the superblock is the static block.
4080 However, you should note that it is possible for a function block to
4081 have a superblock that is not the static block -- for instance this
4082 happens for an inlined function.
4085 @defvar Block.superblock
4086 The block containing this block. If this parent block does not exist,
4087 this attribute holds @code{None}. This attribute is not writable.
4090 @defvar Block.global_block
4091 The global block associated with this block. This attribute is not
4095 @defvar Block.static_block
4096 The static block associated with this block. This attribute is not
4100 @defvar Block.is_global
4101 @code{True} if the @code{gdb.Block} object is a global block,
4102 @code{False} if not. This attribute is not
4106 @defvar Block.is_static
4107 @code{True} if the @code{gdb.Block} object is a static block,
4108 @code{False} if not. This attribute is not writable.
4111 @node Symbols In Python
4112 @subsubsection Python representation of Symbols.
4114 @cindex symbols in python
4117 @value{GDBN} represents every variable, function and type as an
4118 entry in a symbol table. @xref{Symbols, ,Examining the Symbol Table}.
4119 Similarly, Python represents these symbols in @value{GDBN} with the
4120 @code{gdb.Symbol} object.
4122 The following symbol-related functions are available in the @code{gdb}
4125 @findex gdb.lookup_symbol
4126 @defun gdb.lookup_symbol (name @r{[}, block @r{[}, domain@r{]]})
4127 This function searches for a symbol by name. The search scope can be
4128 restricted to the parameters defined in the optional domain and block
4131 @var{name} is the name of the symbol. It must be a string. The
4132 optional @var{block} argument restricts the search to symbols visible
4133 in that @var{block}. The @var{block} argument must be a
4134 @code{gdb.Block} object. If omitted, the block for the current frame
4135 is used. The optional @var{domain} argument restricts
4136 the search to the domain type. The @var{domain} argument must be a
4137 domain constant defined in the @code{gdb} module and described later
4140 The result is a tuple of two elements.
4141 The first element is a @code{gdb.Symbol} object or @code{None} if the symbol
4143 If the symbol is found, the second element is @code{True} if the symbol
4144 is a field of a method's object (e.g., @code{this} in C@t{++}),
4145 otherwise it is @code{False}.
4146 If the symbol is not found, the second element is @code{False}.
4149 @findex gdb.lookup_global_symbol
4150 @defun gdb.lookup_global_symbol (name @r{[}, domain@r{]})
4151 This function searches for a global symbol by name.
4152 The search scope can be restricted to by the domain argument.
4154 @var{name} is the name of the symbol. It must be a string.
4155 The optional @var{domain} argument restricts the search to the domain type.
4156 The @var{domain} argument must be a domain constant defined in the @code{gdb}
4157 module and described later in this chapter.
4159 The result is a @code{gdb.Symbol} object or @code{None} if the symbol
4163 A @code{gdb.Symbol} object has the following attributes:
4166 The type of the symbol or @code{None} if no type is recorded.
4167 This attribute is represented as a @code{gdb.Type} object.
4168 @xref{Types In Python}. This attribute is not writable.
4171 @defvar Symbol.symtab
4172 The symbol table in which the symbol appears. This attribute is
4173 represented as a @code{gdb.Symtab} object. @xref{Symbol Tables In
4174 Python}. This attribute is not writable.
4178 The line number in the source code at which the symbol was defined.
4183 The name of the symbol as a string. This attribute is not writable.
4186 @defvar Symbol.linkage_name
4187 The name of the symbol, as used by the linker (i.e., may be mangled).
4188 This attribute is not writable.
4191 @defvar Symbol.print_name
4192 The name of the symbol in a form suitable for output. This is either
4193 @code{name} or @code{linkage_name}, depending on whether the user
4194 asked @value{GDBN} to display demangled or mangled names.
4197 @defvar Symbol.addr_class
4198 The address class of the symbol. This classifies how to find the value
4199 of a symbol. Each address class is a constant defined in the
4200 @code{gdb} module and described later in this chapter.
4203 @defvar Symbol.needs_frame
4204 This is @code{True} if evaluating this symbol's value requires a frame
4205 (@pxref{Frames In Python}) and @code{False} otherwise. Typically,
4206 local variables will require a frame, but other symbols will not.
4209 @defvar Symbol.is_argument
4210 @code{True} if the symbol is an argument of a function.
4213 @defvar Symbol.is_constant
4214 @code{True} if the symbol is a constant.
4217 @defvar Symbol.is_function
4218 @code{True} if the symbol is a function or a method.
4221 @defvar Symbol.is_variable
4222 @code{True} if the symbol is a variable.
4225 A @code{gdb.Symbol} object has the following methods:
4227 @defun Symbol.is_valid ()
4228 Returns @code{True} if the @code{gdb.Symbol} object is valid,
4229 @code{False} if not. A @code{gdb.Symbol} object can become invalid if
4230 the symbol it refers to does not exist in @value{GDBN} any longer.
4231 All other @code{gdb.Symbol} methods will throw an exception if it is
4232 invalid at the time the method is called.
4235 @defun Symbol.value (@r{[}frame@r{]})
4236 Compute the value of the symbol, as a @code{gdb.Value}. For
4237 functions, this computes the address of the function, cast to the
4238 appropriate type. If the symbol requires a frame in order to compute
4239 its value, then @var{frame} must be given. If @var{frame} is not
4240 given, or if @var{frame} is invalid, then this method will throw an
4244 The available domain categories in @code{gdb.Symbol} are represented
4245 as constants in the @code{gdb} module:
4248 @vindex SYMBOL_UNDEF_DOMAIN
4249 @item gdb.SYMBOL_UNDEF_DOMAIN
4250 This is used when a domain has not been discovered or none of the
4251 following domains apply. This usually indicates an error either
4252 in the symbol information or in @value{GDBN}'s handling of symbols.
4254 @vindex SYMBOL_VAR_DOMAIN
4255 @item gdb.SYMBOL_VAR_DOMAIN
4256 This domain contains variables, function names, typedef names and enum
4259 @vindex SYMBOL_STRUCT_DOMAIN
4260 @item gdb.SYMBOL_STRUCT_DOMAIN
4261 This domain holds struct, union and enum type names.
4263 @vindex SYMBOL_LABEL_DOMAIN
4264 @item gdb.SYMBOL_LABEL_DOMAIN
4265 This domain contains names of labels (for gotos).
4267 @vindex SYMBOL_VARIABLES_DOMAIN
4268 @item gdb.SYMBOL_VARIABLES_DOMAIN
4269 This domain holds a subset of the @code{SYMBOLS_VAR_DOMAIN}; it
4270 contains everything minus functions and types.
4272 @vindex SYMBOL_FUNCTIONS_DOMAIN
4273 @item gdb.SYMBOL_FUNCTION_DOMAIN
4274 This domain contains all functions.
4276 @vindex SYMBOL_TYPES_DOMAIN
4277 @item gdb.SYMBOL_TYPES_DOMAIN
4278 This domain contains all types.
4281 The available address class categories in @code{gdb.Symbol} are represented
4282 as constants in the @code{gdb} module:
4285 @vindex SYMBOL_LOC_UNDEF
4286 @item gdb.SYMBOL_LOC_UNDEF
4287 If this is returned by address class, it indicates an error either in
4288 the symbol information or in @value{GDBN}'s handling of symbols.
4290 @vindex SYMBOL_LOC_CONST
4291 @item gdb.SYMBOL_LOC_CONST
4292 Value is constant int.
4294 @vindex SYMBOL_LOC_STATIC
4295 @item gdb.SYMBOL_LOC_STATIC
4296 Value is at a fixed address.
4298 @vindex SYMBOL_LOC_REGISTER
4299 @item gdb.SYMBOL_LOC_REGISTER
4300 Value is in a register.
4302 @vindex SYMBOL_LOC_ARG
4303 @item gdb.SYMBOL_LOC_ARG
4304 Value is an argument. This value is at the offset stored within the
4305 symbol inside the frame's argument list.
4307 @vindex SYMBOL_LOC_REF_ARG
4308 @item gdb.SYMBOL_LOC_REF_ARG
4309 Value address is stored in the frame's argument list. Just like
4310 @code{LOC_ARG} except that the value's address is stored at the
4311 offset, not the value itself.
4313 @vindex SYMBOL_LOC_REGPARM_ADDR
4314 @item gdb.SYMBOL_LOC_REGPARM_ADDR
4315 Value is a specified register. Just like @code{LOC_REGISTER} except
4316 the register holds the address of the argument instead of the argument
4319 @vindex SYMBOL_LOC_LOCAL
4320 @item gdb.SYMBOL_LOC_LOCAL
4321 Value is a local variable.
4323 @vindex SYMBOL_LOC_TYPEDEF
4324 @item gdb.SYMBOL_LOC_TYPEDEF
4325 Value not used. Symbols in the domain @code{SYMBOL_STRUCT_DOMAIN} all
4328 @vindex SYMBOL_LOC_BLOCK
4329 @item gdb.SYMBOL_LOC_BLOCK
4332 @vindex SYMBOL_LOC_CONST_BYTES
4333 @item gdb.SYMBOL_LOC_CONST_BYTES
4334 Value is a byte-sequence.
4336 @vindex SYMBOL_LOC_UNRESOLVED
4337 @item gdb.SYMBOL_LOC_UNRESOLVED
4338 Value is at a fixed address, but the address of the variable has to be
4339 determined from the minimal symbol table whenever the variable is
4342 @vindex SYMBOL_LOC_OPTIMIZED_OUT
4343 @item gdb.SYMBOL_LOC_OPTIMIZED_OUT
4344 The value does not actually exist in the program.
4346 @vindex SYMBOL_LOC_COMPUTED
4347 @item gdb.SYMBOL_LOC_COMPUTED
4348 The value's address is a computed location.
4351 @node Symbol Tables In Python
4352 @subsubsection Symbol table representation in Python.
4354 @cindex symbol tables in python
4356 @tindex gdb.Symtab_and_line
4358 Access to symbol table data maintained by @value{GDBN} on the inferior
4359 is exposed to Python via two objects: @code{gdb.Symtab_and_line} and
4360 @code{gdb.Symtab}. Symbol table and line data for a frame is returned
4361 from the @code{find_sal} method in @code{gdb.Frame} object.
4362 @xref{Frames In Python}.
4364 For more information on @value{GDBN}'s symbol table management, see
4365 @ref{Symbols, ,Examining the Symbol Table}, for more information.
4367 A @code{gdb.Symtab_and_line} object has the following attributes:
4369 @defvar Symtab_and_line.symtab
4370 The symbol table object (@code{gdb.Symtab}) for this frame.
4371 This attribute is not writable.
4374 @defvar Symtab_and_line.pc
4375 Indicates the start of the address range occupied by code for the
4376 current source line. This attribute is not writable.
4379 @defvar Symtab_and_line.last
4380 Indicates the end of the address range occupied by code for the current
4381 source line. This attribute is not writable.
4384 @defvar Symtab_and_line.line
4385 Indicates the current line number for this object. This
4386 attribute is not writable.
4389 A @code{gdb.Symtab_and_line} object has the following methods:
4391 @defun Symtab_and_line.is_valid ()
4392 Returns @code{True} if the @code{gdb.Symtab_and_line} object is valid,
4393 @code{False} if not. A @code{gdb.Symtab_and_line} object can become
4394 invalid if the Symbol table and line object it refers to does not
4395 exist in @value{GDBN} any longer. All other
4396 @code{gdb.Symtab_and_line} methods will throw an exception if it is
4397 invalid at the time the method is called.
4400 A @code{gdb.Symtab} object has the following attributes:
4402 @defvar Symtab.filename
4403 The symbol table's source filename. This attribute is not writable.
4406 @defvar Symtab.objfile
4407 The symbol table's backing object file. @xref{Objfiles In Python}.
4408 This attribute is not writable.
4411 @defvar Symtab.producer
4412 The name and possibly version number of the program that
4413 compiled the code in the symbol table.
4414 The contents of this string is up to the compiler.
4415 If no producer information is available then @code{None} is returned.
4416 This attribute is not writable.
4419 A @code{gdb.Symtab} object has the following methods:
4421 @defun Symtab.is_valid ()
4422 Returns @code{True} if the @code{gdb.Symtab} object is valid,
4423 @code{False} if not. A @code{gdb.Symtab} object can become invalid if
4424 the symbol table it refers to does not exist in @value{GDBN} any
4425 longer. All other @code{gdb.Symtab} methods will throw an exception
4426 if it is invalid at the time the method is called.
4429 @defun Symtab.fullname ()
4430 Return the symbol table's source absolute file name.
4433 @defun Symtab.global_block ()
4434 Return the global block of the underlying symbol table.
4435 @xref{Blocks In Python}.
4438 @defun Symtab.static_block ()
4439 Return the static block of the underlying symbol table.
4440 @xref{Blocks In Python}.
4443 @defun Symtab.linetable ()
4444 Return the line table associated with the symbol table.
4445 @xref{Line Tables In Python}.
4448 @node Line Tables In Python
4449 @subsubsection Manipulating line tables using Python
4451 @cindex line tables in python
4452 @tindex gdb.LineTable
4454 Python code can request and inspect line table information from a
4455 symbol table that is loaded in @value{GDBN}. A line table is a
4456 mapping of source lines to their executable locations in memory. To
4457 acquire the line table information for a particular symbol table, use
4458 the @code{linetable} function (@pxref{Symbol Tables In Python}).
4460 A @code{gdb.LineTable} is iterable. The iterator returns
4461 @code{LineTableEntry} objects that correspond to the source line and
4462 address for each line table entry. @code{LineTableEntry} objects have
4463 the following attributes:
4465 @defvar LineTableEntry.line
4466 The source line number for this line table entry. This number
4467 corresponds to the actual line of source. This attribute is not
4471 @defvar LineTableEntry.pc
4472 The address that is associated with the line table entry where the
4473 executable code for that source line resides in memory. This
4474 attribute is not writable.
4477 As there can be multiple addresses for a single source line, you may
4478 receive multiple @code{LineTableEntry} objects with matching
4479 @code{line} attributes, but with different @code{pc} attributes. The
4480 iterator is sorted in ascending @code{pc} order. Here is a small
4481 example illustrating iterating over a line table.
4484 symtab = gdb.selected_frame().find_sal().symtab
4485 linetable = symtab.linetable()
4486 for line in linetable:
4487 print "Line: "+str(line.line)+" Address: "+hex(line.pc)
4490 This will have the following output:
4493 Line: 33 Address: 0x4005c8L
4494 Line: 37 Address: 0x4005caL
4495 Line: 39 Address: 0x4005d2L
4496 Line: 40 Address: 0x4005f8L
4497 Line: 42 Address: 0x4005ffL
4498 Line: 44 Address: 0x400608L
4499 Line: 42 Address: 0x40060cL
4500 Line: 45 Address: 0x400615L
4503 In addition to being able to iterate over a @code{LineTable}, it also
4504 has the following direct access methods:
4506 @defun LineTable.line (line)
4507 Return a Python @code{Tuple} of @code{LineTableEntry} objects for any
4508 entries in the line table for the given @var{line}, which specifies
4509 the source code line. If there are no entries for that source code
4510 @var{line}, the Python @code{None} is returned.
4513 @defun LineTable.has_line (line)
4514 Return a Python @code{Boolean} indicating whether there is an entry in
4515 the line table for this source line. Return @code{True} if an entry
4516 is found, or @code{False} if not.
4519 @defun LineTable.source_lines ()
4520 Return a Python @code{List} of the source line numbers in the symbol
4521 table. Only lines with executable code locations are returned. The
4522 contents of the @code{List} will just be the source line entries
4523 represented as Python @code{Long} values.
4526 @node Breakpoints In Python
4527 @subsubsection Manipulating breakpoints using Python
4529 @cindex breakpoints in python
4530 @tindex gdb.Breakpoint
4532 Python code can manipulate breakpoints via the @code{gdb.Breakpoint}
4535 @defun Breakpoint.__init__ (spec @r{[}, type @r{[}, wp_class @r{[},internal @r{[},temporary@r{]]]]})
4536 Create a new breakpoint according to @var{spec}, which is a string
4537 naming the location of the breakpoint, or an expression that defines a
4538 watchpoint. The contents can be any location recognized by the
4539 @code{break} command, or in the case of a watchpoint, by the
4540 @code{watch} command. The optional @var{type} denotes the breakpoint
4541 to create from the types defined later in this chapter. This argument
4542 can be either @code{gdb.BP_BREAKPOINT} or @code{gdb.BP_WATCHPOINT}; it
4543 defaults to @code{gdb.BP_BREAKPOINT}. The optional @var{internal}
4544 argument allows the breakpoint to become invisible to the user. The
4545 breakpoint will neither be reported when created, nor will it be
4546 listed in the output from @code{info breakpoints} (but will be listed
4547 with the @code{maint info breakpoints} command). The optional
4548 @var{temporary} argument makes the breakpoint a temporary breakpoint.
4549 Temporary breakpoints are deleted after they have been hit. Any
4550 further access to the Python breakpoint after it has been hit will
4551 result in a runtime error (as that breakpoint has now been
4552 automatically deleted). The optional @var{wp_class} argument defines
4553 the class of watchpoint to create, if @var{type} is
4554 @code{gdb.BP_WATCHPOINT}. If a watchpoint class is not provided, it
4555 is assumed to be a @code{gdb.WP_WRITE} class.
4558 @defun Breakpoint.stop (self)
4559 The @code{gdb.Breakpoint} class can be sub-classed and, in
4560 particular, you may choose to implement the @code{stop} method.
4561 If this method is defined in a sub-class of @code{gdb.Breakpoint},
4562 it will be called when the inferior reaches any location of a
4563 breakpoint which instantiates that sub-class. If the method returns
4564 @code{True}, the inferior will be stopped at the location of the
4565 breakpoint, otherwise the inferior will continue.
4567 If there are multiple breakpoints at the same location with a
4568 @code{stop} method, each one will be called regardless of the
4569 return status of the previous. This ensures that all @code{stop}
4570 methods have a chance to execute at that location. In this scenario
4571 if one of the methods returns @code{True} but the others return
4572 @code{False}, the inferior will still be stopped.
4574 You should not alter the execution state of the inferior (i.e.@:, step,
4575 next, etc.), alter the current frame context (i.e.@:, change the current
4576 active frame), or alter, add or delete any breakpoint. As a general
4577 rule, you should not alter any data within @value{GDBN} or the inferior
4580 Example @code{stop} implementation:
4583 class MyBreakpoint (gdb.Breakpoint):
4585 inf_val = gdb.parse_and_eval("foo")
4592 The available watchpoint types represented by constants are defined in the
4598 Read only watchpoint.
4602 Write only watchpoint.
4606 Read/Write watchpoint.
4609 @defun Breakpoint.is_valid ()
4610 Return @code{True} if this @code{Breakpoint} object is valid,
4611 @code{False} otherwise. A @code{Breakpoint} object can become invalid
4612 if the user deletes the breakpoint. In this case, the object still
4613 exists, but the underlying breakpoint does not. In the cases of
4614 watchpoint scope, the watchpoint remains valid even if execution of the
4615 inferior leaves the scope of that watchpoint.
4618 @defun Breakpoint.delete ()
4619 Permanently deletes the @value{GDBN} breakpoint. This also
4620 invalidates the Python @code{Breakpoint} object. Any further access
4621 to this object's attributes or methods will raise an error.
4624 @defvar Breakpoint.enabled
4625 This attribute is @code{True} if the breakpoint is enabled, and
4626 @code{False} otherwise. This attribute is writable. You can use it to enable
4627 or disable the breakpoint.
4630 @defvar Breakpoint.silent
4631 This attribute is @code{True} if the breakpoint is silent, and
4632 @code{False} otherwise. This attribute is writable.
4634 Note that a breakpoint can also be silent if it has commands and the
4635 first command is @code{silent}. This is not reported by the
4636 @code{silent} attribute.
4639 @defvar Breakpoint.thread
4640 If the breakpoint is thread-specific, this attribute holds the thread
4641 id. If the breakpoint is not thread-specific, this attribute is
4642 @code{None}. This attribute is writable.
4645 @defvar Breakpoint.task
4646 If the breakpoint is Ada task-specific, this attribute holds the Ada task
4647 id. If the breakpoint is not task-specific (or the underlying
4648 language is not Ada), this attribute is @code{None}. This attribute
4652 @defvar Breakpoint.ignore_count
4653 This attribute holds the ignore count for the breakpoint, an integer.
4654 This attribute is writable.
4657 @defvar Breakpoint.number
4658 This attribute holds the breakpoint's number --- the identifier used by
4659 the user to manipulate the breakpoint. This attribute is not writable.
4662 @defvar Breakpoint.type
4663 This attribute holds the breakpoint's type --- the identifier used to
4664 determine the actual breakpoint type or use-case. This attribute is not
4668 @defvar Breakpoint.visible
4669 This attribute tells whether the breakpoint is visible to the user
4670 when set, or when the @samp{info breakpoints} command is run. This
4671 attribute is not writable.
4674 @defvar Breakpoint.temporary
4675 This attribute indicates whether the breakpoint was created as a
4676 temporary breakpoint. Temporary breakpoints are automatically deleted
4677 after that breakpoint has been hit. Access to this attribute, and all
4678 other attributes and functions other than the @code{is_valid}
4679 function, will result in an error after the breakpoint has been hit
4680 (as it has been automatically deleted). This attribute is not
4684 The available types are represented by constants defined in the @code{gdb}
4688 @vindex BP_BREAKPOINT
4689 @item gdb.BP_BREAKPOINT
4690 Normal code breakpoint.
4692 @vindex BP_WATCHPOINT
4693 @item gdb.BP_WATCHPOINT
4694 Watchpoint breakpoint.
4696 @vindex BP_HARDWARE_WATCHPOINT
4697 @item gdb.BP_HARDWARE_WATCHPOINT
4698 Hardware assisted watchpoint.
4700 @vindex BP_READ_WATCHPOINT
4701 @item gdb.BP_READ_WATCHPOINT
4702 Hardware assisted read watchpoint.
4704 @vindex BP_ACCESS_WATCHPOINT
4705 @item gdb.BP_ACCESS_WATCHPOINT
4706 Hardware assisted access watchpoint.
4709 @defvar Breakpoint.hit_count
4710 This attribute holds the hit count for the breakpoint, an integer.
4711 This attribute is writable, but currently it can only be set to zero.
4714 @defvar Breakpoint.location
4715 This attribute holds the location of the breakpoint, as specified by
4716 the user. It is a string. If the breakpoint does not have a location
4717 (that is, it is a watchpoint) the attribute's value is @code{None}. This
4718 attribute is not writable.
4721 @defvar Breakpoint.expression
4722 This attribute holds a breakpoint expression, as specified by
4723 the user. It is a string. If the breakpoint does not have an
4724 expression (the breakpoint is not a watchpoint) the attribute's value
4725 is @code{None}. This attribute is not writable.
4728 @defvar Breakpoint.condition
4729 This attribute holds the condition of the breakpoint, as specified by
4730 the user. It is a string. If there is no condition, this attribute's
4731 value is @code{None}. This attribute is writable.
4734 @defvar Breakpoint.commands
4735 This attribute holds the commands attached to the breakpoint. If
4736 there are commands, this attribute's value is a string holding all the
4737 commands, separated by newlines. If there are no commands, this
4738 attribute is @code{None}. This attribute is not writable.
4741 @node Finish Breakpoints in Python
4742 @subsubsection Finish Breakpoints
4744 @cindex python finish breakpoints
4745 @tindex gdb.FinishBreakpoint
4747 A finish breakpoint is a temporary breakpoint set at the return address of
4748 a frame, based on the @code{finish} command. @code{gdb.FinishBreakpoint}
4749 extends @code{gdb.Breakpoint}. The underlying breakpoint will be disabled
4750 and deleted when the execution will run out of the breakpoint scope (i.e.@:
4751 @code{Breakpoint.stop} or @code{FinishBreakpoint.out_of_scope} triggered).
4752 Finish breakpoints are thread specific and must be create with the right
4755 @defun FinishBreakpoint.__init__ (@r{[}frame@r{]} @r{[}, internal@r{]})
4756 Create a finish breakpoint at the return address of the @code{gdb.Frame}
4757 object @var{frame}. If @var{frame} is not provided, this defaults to the
4758 newest frame. The optional @var{internal} argument allows the breakpoint to
4759 become invisible to the user. @xref{Breakpoints In Python}, for further
4760 details about this argument.
4763 @defun FinishBreakpoint.out_of_scope (self)
4764 In some circumstances (e.g.@: @code{longjmp}, C@t{++} exceptions, @value{GDBN}
4765 @code{return} command, @dots{}), a function may not properly terminate, and
4766 thus never hit the finish breakpoint. When @value{GDBN} notices such a
4767 situation, the @code{out_of_scope} callback will be triggered.
4769 You may want to sub-class @code{gdb.FinishBreakpoint} and override this
4773 class MyFinishBreakpoint (gdb.FinishBreakpoint)
4775 print "normal finish"
4778 def out_of_scope ():
4779 print "abnormal finish"
4783 @defvar FinishBreakpoint.return_value
4784 When @value{GDBN} is stopped at a finish breakpoint and the frame
4785 used to build the @code{gdb.FinishBreakpoint} object had debug symbols, this
4786 attribute will contain a @code{gdb.Value} object corresponding to the return
4787 value of the function. The value will be @code{None} if the function return
4788 type is @code{void} or if the return value was not computable. This attribute
4792 @node Lazy Strings In Python
4793 @subsubsection Python representation of lazy strings.
4795 @cindex lazy strings in python
4796 @tindex gdb.LazyString
4798 A @dfn{lazy string} is a string whose contents is not retrieved or
4799 encoded until it is needed.
4801 A @code{gdb.LazyString} is represented in @value{GDBN} as an
4802 @code{address} that points to a region of memory, an @code{encoding}
4803 that will be used to encode that region of memory, and a @code{length}
4804 to delimit the region of memory that represents the string. The
4805 difference between a @code{gdb.LazyString} and a string wrapped within
4806 a @code{gdb.Value} is that a @code{gdb.LazyString} will be treated
4807 differently by @value{GDBN} when printing. A @code{gdb.LazyString} is
4808 retrieved and encoded during printing, while a @code{gdb.Value}
4809 wrapping a string is immediately retrieved and encoded on creation.
4811 A @code{gdb.LazyString} object has the following functions:
4813 @defun LazyString.value ()
4814 Convert the @code{gdb.LazyString} to a @code{gdb.Value}. This value
4815 will point to the string in memory, but will lose all the delayed
4816 retrieval, encoding and handling that @value{GDBN} applies to a
4817 @code{gdb.LazyString}.
4820 @defvar LazyString.address
4821 This attribute holds the address of the string. This attribute is not
4825 @defvar LazyString.length
4826 This attribute holds the length of the string in characters. If the
4827 length is -1, then the string will be fetched and encoded up to the
4828 first null of appropriate width. This attribute is not writable.
4831 @defvar LazyString.encoding
4832 This attribute holds the encoding that will be applied to the string
4833 when the string is printed by @value{GDBN}. If the encoding is not
4834 set, or contains an empty string, then @value{GDBN} will select the
4835 most appropriate encoding when the string is printed. This attribute
4839 @defvar LazyString.type
4840 This attribute holds the type that is represented by the lazy string's
4841 type. For a lazy string this will always be a pointer type. To
4842 resolve this to the lazy string's character type, use the type's
4843 @code{target} method. @xref{Types In Python}. This attribute is not
4847 @node Architectures In Python
4848 @subsubsection Python representation of architectures
4849 @cindex Python architectures
4851 @value{GDBN} uses architecture specific parameters and artifacts in a
4852 number of its various computations. An architecture is represented
4853 by an instance of the @code{gdb.Architecture} class.
4855 A @code{gdb.Architecture} class has the following methods:
4857 @defun Architecture.name ()
4858 Return the name (string value) of the architecture.
4861 @defun Architecture.disassemble (@var{start_pc} @r{[}, @var{end_pc} @r{[}, @var{count}@r{]]})
4862 Return a list of disassembled instructions starting from the memory
4863 address @var{start_pc}. The optional arguments @var{end_pc} and
4864 @var{count} determine the number of instructions in the returned list.
4865 If both the optional arguments @var{end_pc} and @var{count} are
4866 specified, then a list of at most @var{count} disassembled instructions
4867 whose start address falls in the closed memory address interval from
4868 @var{start_pc} to @var{end_pc} are returned. If @var{end_pc} is not
4869 specified, but @var{count} is specified, then @var{count} number of
4870 instructions starting from the address @var{start_pc} are returned. If
4871 @var{count} is not specified but @var{end_pc} is specified, then all
4872 instructions whose start address falls in the closed memory address
4873 interval from @var{start_pc} to @var{end_pc} are returned. If neither
4874 @var{end_pc} nor @var{count} are specified, then a single instruction at
4875 @var{start_pc} is returned. For all of these cases, each element of the
4876 returned list is a Python @code{dict} with the following string keys:
4881 The value corresponding to this key is a Python long integer capturing
4882 the memory address of the instruction.
4885 The value corresponding to this key is a string value which represents
4886 the instruction with assembly language mnemonics. The assembly
4887 language flavor used is the same as that specified by the current CLI
4888 variable @code{disassembly-flavor}. @xref{Machine Code}.
4891 The value corresponding to this key is the length (integer value) of the
4892 instruction in bytes.
4897 @node Python Auto-loading
4898 @subsection Python Auto-loading
4899 @cindex Python auto-loading
4901 When a new object file is read (for example, due to the @code{file}
4902 command, or because the inferior has loaded a shared library),
4903 @value{GDBN} will look for Python support scripts in several ways:
4904 @file{@var{objfile}-gdb.py} and @code{.debug_gdb_scripts} section.
4905 @xref{Auto-loading extensions}.
4907 The auto-loading feature is useful for supplying application-specific
4908 debugging commands and scripts.
4910 Auto-loading can be enabled or disabled,
4911 and the list of auto-loaded scripts can be printed.
4914 @anchor{set auto-load python-scripts}
4915 @kindex set auto-load python-scripts
4916 @item set auto-load python-scripts [on|off]
4917 Enable or disable the auto-loading of Python scripts.
4919 @anchor{show auto-load python-scripts}
4920 @kindex show auto-load python-scripts
4921 @item show auto-load python-scripts
4922 Show whether auto-loading of Python scripts is enabled or disabled.
4924 @anchor{info auto-load python-scripts}
4925 @kindex info auto-load python-scripts
4926 @cindex print list of auto-loaded Python scripts
4927 @item info auto-load python-scripts [@var{regexp}]
4928 Print the list of all Python scripts that @value{GDBN} auto-loaded.
4930 Also printed is the list of Python scripts that were mentioned in
4931 the @code{.debug_gdb_scripts} section and were either not found
4932 (@pxref{dotdebug_gdb_scripts section}) or were not auto-loaded due to
4933 @code{auto-load safe-path} rejection (@pxref{Auto-loading}).
4934 This is useful because their names are not printed when @value{GDBN}
4935 tries to load them and fails. There may be many of them, and printing
4936 an error message for each one is problematic.
4938 If @var{regexp} is supplied only Python scripts with matching names are printed.
4943 (gdb) info auto-load python-scripts
4945 Yes py-section-script.py
4946 full name: /tmp/py-section-script.py
4947 No my-foo-pretty-printers.py
4951 When reading an auto-loaded file or script, @value{GDBN} sets the
4952 @dfn{current objfile}. This is available via the @code{gdb.current_objfile}
4953 function (@pxref{Objfiles In Python}). This can be useful for
4954 registering objfile-specific pretty-printers and frame-filters.
4956 @node Python modules
4957 @subsection Python modules
4958 @cindex python modules
4960 @value{GDBN} comes with several modules to assist writing Python code.
4963 * gdb.printing:: Building and registering pretty-printers.
4964 * gdb.types:: Utilities for working with types.
4965 * gdb.prompt:: Utilities for prompt value substitution.
4969 @subsubsection gdb.printing
4970 @cindex gdb.printing
4972 This module provides a collection of utilities for working with
4976 @item PrettyPrinter (@var{name}, @var{subprinters}=None)
4977 This class specifies the API that makes @samp{info pretty-printer},
4978 @samp{enable pretty-printer} and @samp{disable pretty-printer} work.
4979 Pretty-printers should generally inherit from this class.
4981 @item SubPrettyPrinter (@var{name})
4982 For printers that handle multiple types, this class specifies the
4983 corresponding API for the subprinters.
4985 @item RegexpCollectionPrettyPrinter (@var{name})
4986 Utility class for handling multiple printers, all recognized via
4987 regular expressions.
4988 @xref{Writing a Pretty-Printer}, for an example.
4990 @item FlagEnumerationPrinter (@var{name})
4991 A pretty-printer which handles printing of @code{enum} values. Unlike
4992 @value{GDBN}'s built-in @code{enum} printing, this printer attempts to
4993 work properly when there is some overlap between the enumeration
4994 constants. The argument @var{name} is the name of the printer and
4995 also the name of the @code{enum} type to look up.
4997 @item register_pretty_printer (@var{obj}, @var{printer}, @var{replace}=False)
4998 Register @var{printer} with the pretty-printer list of @var{obj}.
4999 If @var{replace} is @code{True} then any existing copy of the printer
5000 is replaced. Otherwise a @code{RuntimeError} exception is raised
5001 if a printer with the same name already exists.
5005 @subsubsection gdb.types
5008 This module provides a collection of utilities for working with
5009 @code{gdb.Type} objects.
5012 @item get_basic_type (@var{type})
5013 Return @var{type} with const and volatile qualifiers stripped,
5014 and with typedefs and C@t{++} references converted to the underlying type.
5019 typedef const int const_int;
5021 const_int& foo_ref (foo);
5022 int main () @{ return 0; @}
5029 (gdb) python import gdb.types
5030 (gdb) python foo_ref = gdb.parse_and_eval("foo_ref")
5031 (gdb) python print gdb.types.get_basic_type(foo_ref.type)
5035 @item has_field (@var{type}, @var{field})
5036 Return @code{True} if @var{type}, assumed to be a type with fields
5037 (e.g., a structure or union), has field @var{field}.
5039 @item make_enum_dict (@var{enum_type})
5040 Return a Python @code{dictionary} type produced from @var{enum_type}.
5042 @item deep_items (@var{type})
5043 Returns a Python iterator similar to the standard
5044 @code{gdb.Type.iteritems} method, except that the iterator returned
5045 by @code{deep_items} will recursively traverse anonymous struct or
5046 union fields. For example:
5060 Then in @value{GDBN}:
5062 (@value{GDBP}) python import gdb.types
5063 (@value{GDBP}) python struct_a = gdb.lookup_type("struct A")
5064 (@value{GDBP}) python print struct_a.keys ()
5066 (@value{GDBP}) python print [k for k,v in gdb.types.deep_items(struct_a)]
5067 @{['a', 'b0', 'b1']@}
5070 @item get_type_recognizers ()
5071 Return a list of the enabled type recognizers for the current context.
5072 This is called by @value{GDBN} during the type-printing process
5073 (@pxref{Type Printing API}).
5075 @item apply_type_recognizers (recognizers, type_obj)
5076 Apply the type recognizers, @var{recognizers}, to the type object
5077 @var{type_obj}. If any recognizer returns a string, return that
5078 string. Otherwise, return @code{None}. This is called by
5079 @value{GDBN} during the type-printing process (@pxref{Type Printing
5082 @item register_type_printer (locus, printer)
5083 This is a convenience function to register a type printer
5084 @var{printer}. The printer must implement the type printer protocol.
5085 The @var{locus} argument is either a @code{gdb.Objfile}, in which case
5086 the printer is registered with that objfile; a @code{gdb.Progspace},
5087 in which case the printer is registered with that progspace; or
5088 @code{None}, in which case the printer is registered globally.
5091 This is a base class that implements the type printer protocol. Type
5092 printers are encouraged, but not required, to derive from this class.
5093 It defines a constructor:
5095 @defmethod TypePrinter __init__ (self, name)
5096 Initialize the type printer with the given name. The new printer
5097 starts in the enabled state.
5103 @subsubsection gdb.prompt
5106 This module provides a method for prompt value-substitution.
5109 @item substitute_prompt (@var{string})
5110 Return @var{string} with escape sequences substituted by values. Some
5111 escape sequences take arguments. You can specify arguments inside
5112 ``@{@}'' immediately following the escape sequence.
5114 The escape sequences you can pass to this function are:
5118 Substitute a backslash.
5120 Substitute an ESC character.
5122 Substitute the selected frame; an argument names a frame parameter.
5124 Substitute a newline.
5126 Substitute a parameter's value; the argument names the parameter.
5128 Substitute a carriage return.
5130 Substitute the selected thread; an argument names a thread parameter.
5132 Substitute the version of GDB.
5134 Substitute the current working directory.
5136 Begin a sequence of non-printing characters. These sequences are
5137 typically used with the ESC character, and are not counted in the string
5138 length. Example: ``\[\e[0;34m\](gdb)\[\e[0m\]'' will return a
5139 blue-colored ``(gdb)'' prompt where the length is five.
5141 End a sequence of non-printing characters.
5147 substitute_prompt (``frame: \f,
5148 print arguments: \p@{print frame-arguments@}'')
5151 @exdent will return the string:
5154 "frame: main, print arguments: scalars"