Doc/lib/libpdb.tex

   1 \chapter{The Python Debugger}
   2 \label{module-pdb}
   3 \stmodindex{pdb}
   4 \index{debugging}
   5
   6
   7 The module \code{pdb} defines an interactive source code debugger for
   8 Python programs.  It supports setting
   9 (conditional) breakpoints and single stepping
  10 at the source line level, inspection of stack frames, source code
  11 listing, and evaluation of arbitrary Python code in the context of any
  12 stack frame.  It also supports post-mortem debugging and can be called
  13 under program control.
  14
  15 The debugger is extensible --- it is actually defined as a class
  16 \class{Pdb}.
  17 \withsubitem{(class in pdb)}{\ttindex{Pdb}}
  18 This is currently undocumented but easily understood by reading the
  19 source.  The extension interface uses the (also undocumented) modules
  20 \module{bdb}\refstmodindex{bdb} and \module{cmd}\refstmodindex{cmd}.
  21
  22 A primitive windowing version of the debugger also exists --- this is
  23 module \module{wdb}, which requires \module{stdwin} (see the chapter
  24 on STDWIN specific modules).
  25 \refbimodindex{stdwin}
  26 \refstmodindex{wdb}
  27
  28 The debugger's prompt is \samp{(Pdb) }.
  29 Typical usage to run a program under control of the debugger is:
  30
  31 \begin{verbatim}
  32 >>> import pdb
  33 >>> import mymodule
  34 >>> pdb.run('mymodule.test()')
  35 > <string>(0)?()
  36 (Pdb) continue
  37 > <string>(1)?()
  38 (Pdb) continue
  39 NameError: 'spam'
  40 > <string>(1)?()
  41 (Pdb)
  42 \end{verbatim}
  43
  44 \file{pdb.py} can also be invoked as
  45 a script to debug other scripts.  For example:
  46
  47 \begin{verbatim}
  48 python /usr/local/lib/python1.5/pdb.py myscript.py
  49 \end{verbatim}
  50
  51 Typical usage to inspect a crashed program is:
  52
  53 \begin{verbatim}
  54 >>> import pdb
  55 >>> import mymodule
  56 >>> mymodule.test()
  57 Traceback (innermost last):
  58   File "<stdin>", line 1, in ?
  59   File "./mymodule.py", line 4, in test
  60     test2()
  61   File "./mymodule.py", line 3, in test2
  62     print spam
  63 NameError: spam
  64 >>> pdb.pm()
  65 > ./mymodule.py(3)test2()
  66 -> print spam
  67 (Pdb)
  68 \end{verbatim}
  69
  70 The module defines the following functions; each enters the debugger
  71 in a slightly different way:
  72
  73 \begin{funcdesc}{run}{statement\optional{, globals\optional{, locals}}}
  74 Execute the \var{statement} (given as a string) under debugger
  75 control.  The debugger prompt appears before any code is executed; you
  76 can set breakpoints and type \code{continue}, or you can step through
  77 the statement using \code{step} or \code{next} (all these commands are
  78 explained below).  The optional \var{globals} and \var{locals}
  79 arguments specify the environment in which the code is executed; by
  80 default the dictionary of the module \code{__main__} is used.  (See
  81 the explanation of the \code{exec} statement or the \code{eval()}
  82 built-in function.)
  83 \end{funcdesc}
  84
  85 \begin{funcdesc}{runeval}{expression\optional{, globals\optional{, locals}}}
  86 Evaluate the \var{expression} (given as a a string) under debugger
  87 control.  When \code{runeval()} returns, it returns the value of the
  88 expression.  Otherwise this function is similar to
  89 \code{run()}.
  90 \end{funcdesc}
  91
  92 \begin{funcdesc}{runcall}{function\optional{, argument, ...}}
  93 Call the \var{function} (a function or method object, not a string)
  94 with the given arguments.  When \code{runcall()} returns, it returns
  95 whatever the function call returned.  The debugger prompt appears as
  96 soon as the function is entered.
  97 \end{funcdesc}
  98
  99 \begin{funcdesc}{set_trace}{}
 100 Enter the debugger at the calling stack frame.  This is useful to
 101 hard-code a breakpoint at a given point in a program, even if the code
 102 is not otherwise being debugged (e.g. when an assertion fails).
 103 \end{funcdesc}
 104
 105 \begin{funcdesc}{post_mortem}{traceback}
 106 Enter post-mortem debugging of the given \var{traceback} object.
 107 \end{funcdesc}
 108
 109 \begin{funcdesc}{pm}{}
 110 Enter post-mortem debugging of the traceback found in
 111 \code{sys.last_traceback}.
 112 \end{funcdesc}
 113
 114 \section{Debugger Commands}
 115
 116 The debugger recognizes the following commands.  Most commands can be
 117 abbreviated to one or two letters; e.g. ``\code{h(elp)}'' means that
 118 either ``\code{h}'' or ``\code{help}'' can be used to enter the help
 119 command (but not ``\code{he}'' or ``\code{hel}'', nor ``\code{H}'' or
 120 ``\code{Help} or ``\code{HELP}'').  Arguments to commands must be
 121 separated by whitespace (spaces or tabs).  Optional arguments are
 122 enclosed in square brackets (``\code{[]}'') in the command syntax; the
 123 square brackets must not be typed.  Alternatives in the command syntax
 124 are separated by a vertical bar (``\code{|}'').
 125
 126 Entering a blank line repeats the last command entered.  Exception: if
 127 the last command was a ``\code{list}'' command, the next 11 lines are
 128 listed.
 129
 130 Commands that the debugger doesn't recognize are assumed to be Python
 131 statements and are executed in the context of the program being
 132 debugged.  Python statements can also be prefixed with an exclamation
 133 point (``\code{!}'').  This is a powerful way to inspect the program
 134 being debugged; it is even possible to change a variable or call a
 135 function.  When an
 136 exception occurs in such a statement, the exception name is printed
 137 but the debugger's state is not changed.
 138
 139 \begin{description}
 140
 141 \item[h(elp) \optional{\var{command}}]
 142
 143 Without argument, print the list of available commands.  With a
 144 \var{command} as argument, print help about that command.  \samp{help
 145 pdb} displays the full documentation file; if the environment variable
 146 \code{PAGER} is defined, the file is piped through that command
 147 instead.  Since the \var{command} argument must be an identifier,
 148 \samp{help exec} must be entered to get help on the \samp{!} command.
 149
 150 \item[w(here)]
 151
 152 Print a stack trace, with the most recent frame at the bottom.  An
 153 arrow indicates the current frame, which determines the context of
 154 most commands.
 155
 156 \item[d(own)]
 157
 158 Move the current frame one level down in the stack trace
 159 (to an older frame).
 160
 161 \item[u(p)]
 162
 163 Move the current frame one level up in the stack trace
 164 (to a newer frame).
 165
 166 \item[b(reak) \optional{\var{lineno}\code{\Large|}\var{function}%
 167               \optional{, \code{'}\var{condition}\code{'}}}]
 168
 169 With a \var{lineno} argument, set a break there in the current
 170 file.  With a \var{function} argument, set a break at the entry of
 171 that function.  Without argument, list all breaks.
 172 If a second argument is present, it is a string (included in string
 173 quotes!) specifying an expression which must evaluate to true before
 174 the breakpoint is honored.
 175
 176 \item[cl(ear) \optional{\var{lineno}}]
 177
 178 With a \var{lineno} argument, clear that break in the current file.
 179 Without argument, clear all breaks (but first ask confirmation).
 180
 181 \item[s(tep)]
 182
 183 Execute the current line, stop at the first possible occasion
 184 (either in a function that is called or on the next line in the
 185 current function).
 186
 187 \item[n(ext)]
 188
 189 Continue execution until the next line in the current function
 190 is reached or it returns.  (The difference between \code{next} and
 191 \code{step} is that \code{step} stops inside a called function, while
 192 \code{next} executes called functions at (nearly) full speed, only
 193 stopping at the next line in the current function.)
 194
 195 \item[r(eturn)]
 196
 197 Continue execution until the current function returns.
 198
 199 \item[c(ont(inue))]
 200
 201 Continue execution, only stop when a breakpoint is encountered.
 202
 203 \item[l(ist) \optional{\var{first\optional{, last}}}]
 204
 205 List source code for the current file.  Without arguments, list 11
 206 lines around the current line or continue the previous listing.  With
 207 one argument, list 11 lines around at that line.  With two arguments,
 208 list the given range; if the second argument is less than the first,
 209 it is interpreted as a count.
 210
 211 \item[a(rgs)]
 212
 213 Print the argument list of the current function.
 214
 215 \item[p \var{expression}]
 216
 217 Evaluate the \var{expression} in the current context and print its
 218 value.  (Note: \code{print} can also be used, but is not a debugger
 219 command --- this executes the Python \code{print} statement.)
 220
 221 \item[\optional{!}\var{statement}]
 222
 223 Execute the (one-line) \var{statement} in the context of
 224 the current stack frame.
 225 The exclamation point can be omitted unless the first word
 226 of the statement resembles a debugger command.
 227 To set a global variable, you can prefix the assignment
 228 command with a ``\code{global}'' command on the same line, e.g.:
 229
 230 \begin{verbatim}
 231 (Pdb) global list_options; list_options = ['-l']
 232 (Pdb)
 233 \end{verbatim}
 234
 235 \item[q(uit)]
 236
 237 Quit from the debugger.
 238 The program being executed is aborted.
 239
 240 \end{description}
 241
 242 \section{How It Works}
 243
 244 Some changes were made to the interpreter:
 245
 246 \begin{itemize}
 247 \item \code{sys.settrace(\var{func})} sets the global trace function
 248 \item there can also a local trace function (see later)
 249 \end{itemize}
 250
 251 Trace functions have three arguments: \var{frame}, \var{event}, and
 252 \var{arg}. \var{frame} is the current stack frame.  \var{event} is a
 253 string: \code{'call'}, \code{'line'}, \code{'return'} or
 254 \code{'exception'}.  \var{arg} depends on the event type.
 255
 256 The global trace function is invoked (with \var{event} set to
 257 \code{'call'}) whenever a new local scope is entered; it should return
 258 a reference to the local trace function to be used that scope, or
 259 \code{None} if the scope shouldn't be traced.
 260
 261 The local trace function should return a reference to itself (or to
 262 another function for further tracing in that scope), or \code{None} to
 263 turn off tracing in that scope.
 264
 265 Instance methods are accepted (and very useful!) as trace functions.
 266
 267 The events have the following meaning:
 268
 269 \begin{description}
 270
 271 \item[\code{'call'}]
 272 A function is called (or some other code block entered).  The global
 273 trace function is called; arg is the argument list to the function;
 274 the return value specifies the local trace function.
 275
 276 \item[\code{'line'}]
 277 The interpreter is about to execute a new line of code (sometimes
 278 multiple line events on one line exist).  The local trace function is
 279 called; arg in None; the return value specifies the new local trace
 280 function.
 281
 282 \item[\code{'return'}]
 283 A function (or other code block) is about to return.  The local trace
 284 function is called; arg is the value that will be returned.  The trace
 285 function's return value is ignored.
 286
 287 \item[\code{'exception'}]
 288 An exception has occurred.  The local trace function is called; arg is
 289 a triple (exception, value, traceback); the return value specifies the
 290 new local trace function
 291
 292 \end{description}
 293
 294 Note that as an exception is propagated down the chain of callers, an
 295 \code{'exception'} event is generated at each level.
 296
 297 For more information on code and frame objects, refer to the
 298 \emph{Python Reference Manual}.