| 1 | @comment -*-texinfo-*- |
| 2 | @c This is part of the GNU Emacs Lisp Reference Manual. |
| 3 | @c Copyright (C) 1992, 1993, 1994, 1998, 1999 Free Software Foundation, Inc. |
| 4 | @c See the file elisp.texi for copying conditions. |
| 5 | |
| 6 | @c This file can also be used by an independent Edebug User |
| 7 | @c Manual in which case the Edebug node below should be used |
| 8 | @c with the following links to the Bugs section and to the top level: |
| 9 | |
| 10 | @c , Bugs and Todo List, Top, Top |
| 11 | |
| 12 | @node Edebug, Syntax Errors, Debugger, Debugging |
| 13 | @section Edebug |
| 14 | @cindex Edebug mode |
| 15 | |
| 16 | @cindex Edebug |
| 17 | Edebug is a source-level debugger for Emacs Lisp programs with which |
| 18 | you can: |
| 19 | |
| 20 | @itemize @bullet |
| 21 | @item |
| 22 | Step through evaluation, stopping before and after each expression. |
| 23 | |
| 24 | @item |
| 25 | Set conditional or unconditional breakpoints. |
| 26 | |
| 27 | @item |
| 28 | Stop when a specified condition is true (the global break event). |
| 29 | |
| 30 | @item |
| 31 | Trace slow or fast, stopping briefly at each stop point, or |
| 32 | at each breakpoint. |
| 33 | |
| 34 | @item |
| 35 | Display expression results and evaluate expressions as if outside of |
| 36 | Edebug. |
| 37 | |
| 38 | @item |
| 39 | Automatically re-evaluate a list of expressions and |
| 40 | display their results each time Edebug updates the display. |
| 41 | |
| 42 | @item |
| 43 | Output trace info on function enter and exit. |
| 44 | |
| 45 | @item |
| 46 | Stop when an error occurs. |
| 47 | |
| 48 | @item |
| 49 | Display a backtrace, omitting Edebug's own frames. |
| 50 | |
| 51 | @item |
| 52 | Specify argument evaluation for macros and defining forms. |
| 53 | |
| 54 | @item |
| 55 | Obtain rudimentary coverage testing and frequency counts. |
| 56 | @end itemize |
| 57 | |
| 58 | The first three sections below should tell you enough about Edebug to |
| 59 | enable you to use it. |
| 60 | |
| 61 | @menu |
| 62 | * Using Edebug:: Introduction to use of Edebug. |
| 63 | * Instrumenting:: You must instrument your code |
| 64 | in order to debug it with Edebug. |
| 65 | * Modes: Edebug Execution Modes. Execution modes, stopping more or less often. |
| 66 | * Jumping:: Commands to jump to a specified place. |
| 67 | * Misc: Edebug Misc. Miscellaneous commands. |
| 68 | * Breakpoints:: Setting breakpoints to make the program stop. |
| 69 | * Trapping Errors:: Trapping errors with Edebug. |
| 70 | * Views: Edebug Views. Views inside and outside of Edebug. |
| 71 | * Eval: Edebug Eval. Evaluating expressions within Edebug. |
| 72 | * Eval List:: Expressions whose values are displayed |
| 73 | each time you enter Edebug. |
| 74 | * Printing in Edebug:: Customization of printing. |
| 75 | * Trace Buffer:: How to produce trace output in a buffer. |
| 76 | * Coverage Testing:: How to test evaluation coverage. |
| 77 | * The Outside Context:: Data that Edebug saves and restores. |
| 78 | * Instrumenting Macro Calls:: Specifying how to handle macro calls. |
| 79 | * Options: Edebug Options. Option variables for customizing Edebug. |
| 80 | @end menu |
| 81 | |
| 82 | @node Using Edebug |
| 83 | @subsection Using Edebug |
| 84 | |
| 85 | To debug a Lisp program with Edebug, you must first @dfn{instrument} |
| 86 | the Lisp code that you want to debug. A simple way to do this is to |
| 87 | first move point into the definition of a function or macro and then do |
| 88 | @kbd{C-u C-M-x} (@code{eval-defun} with a prefix argument). See |
| 89 | @ref{Instrumenting}, for alternative ways to instrument code. |
| 90 | |
| 91 | Once a function is instrumented, any call to the function activates |
| 92 | Edebug. Depending on which Edebug execution mode you have selected, |
| 93 | activating Edebug may stop execution and let you step through the |
| 94 | function, or it may update the display and continue execution while |
| 95 | checking for debugging commands. The default execution mode is step, |
| 96 | which stops execution. @xref{Edebug Execution Modes}. |
| 97 | |
| 98 | Within Edebug, you normally view an Emacs buffer showing the source of |
| 99 | the Lisp code you are debugging. This is referred to as the @dfn{source |
| 100 | code buffer}, and it is temporarily read-only. |
| 101 | |
| 102 | An arrow at the left margin indicates the line where the function is |
| 103 | executing. Point initially shows where within the line the function is |
| 104 | executing, but this ceases to be true if you move point yourself. |
| 105 | |
| 106 | If you instrument the definition of @code{fac} (shown below) and then |
| 107 | execute @code{(fac 3)}, here is what you would normally see. Point is |
| 108 | at the open-parenthesis before @code{if}. |
| 109 | |
| 110 | @example |
| 111 | (defun fac (n) |
| 112 | =>@point{}(if (< 0 n) |
| 113 | (* n (fac (1- n))) |
| 114 | 1)) |
| 115 | @end example |
| 116 | |
| 117 | @cindex stop points |
| 118 | The places within a function where Edebug can stop execution are called |
| 119 | @dfn{stop points}. These occur both before and after each subexpression |
| 120 | that is a list, and also after each variable reference. |
| 121 | Here we use periods to show the stop points in the function |
| 122 | @code{fac}: |
| 123 | |
| 124 | @example |
| 125 | (defun fac (n) |
| 126 | .(if .(< 0 n.). |
| 127 | .(* n. .(fac (1- n.).).). |
| 128 | 1).) |
| 129 | @end example |
| 130 | |
| 131 | The special commands of Edebug are available in the source code buffer |
| 132 | in addition to the commands of Emacs Lisp mode. For example, you can |
| 133 | type the Edebug command @key{SPC} to execute until the next stop point. |
| 134 | If you type @key{SPC} once after entry to @code{fac}, here is the |
| 135 | display you will see: |
| 136 | |
| 137 | @example |
| 138 | (defun fac (n) |
| 139 | =>(if @point{}(< 0 n) |
| 140 | (* n (fac (1- n))) |
| 141 | 1)) |
| 142 | @end example |
| 143 | |
| 144 | When Edebug stops execution after an expression, it displays the |
| 145 | expression's value in the echo area. |
| 146 | |
| 147 | Other frequently used commands are @kbd{b} to set a breakpoint at a stop |
| 148 | point, @kbd{g} to execute until a breakpoint is reached, and @kbd{q} to |
| 149 | exit Edebug and return to the top-level command loop. Type @kbd{?} to |
| 150 | display a list of all Edebug commands. |
| 151 | |
| 152 | @node Instrumenting |
| 153 | @subsection Instrumenting for Edebug |
| 154 | |
| 155 | In order to use Edebug to debug Lisp code, you must first |
| 156 | @dfn{instrument} the code. Instrumenting code inserts additional code |
| 157 | into it, to invoke Edebug at the proper places. |
| 158 | |
| 159 | @kindex C-M-x |
| 160 | @findex eval-defun (Edebug) |
| 161 | Once you have loaded Edebug, the command @kbd{C-M-x} |
| 162 | (@code{eval-defun}) is redefined so that when invoked with a prefix |
| 163 | argument on a definition, it instruments the definition before |
| 164 | evaluating it. (The source code itself is not modified.) If the |
| 165 | variable @code{edebug-all-defs} is non-@code{nil}, that inverts the |
| 166 | meaning of the prefix argument: in this case, @kbd{C-M-x} instruments the |
| 167 | definition @emph{unless} it has a prefix argument. The default value of |
| 168 | @code{edebug-all-defs} is @code{nil}. The command @kbd{M-x |
| 169 | edebug-all-defs} toggles the value of the variable |
| 170 | @code{edebug-all-defs}. |
| 171 | |
| 172 | @findex eval-region @r{(Edebug)} |
| 173 | @findex eval-current-buffer @r{(Edebug)} |
| 174 | If @code{edebug-all-defs} is non-@code{nil}, then the commands |
| 175 | @code{eval-region}, @code{eval-current-buffer}, and @code{eval-buffer} |
| 176 | also instrument any definitions they evaluate. Similarly, |
| 177 | @code{edebug-all-forms} controls whether @code{eval-region} should |
| 178 | instrument @emph{any} form, even non-defining forms. This doesn't apply |
| 179 | to loading or evaluations in the minibuffer. The command @kbd{M-x |
| 180 | edebug-all-forms} toggles this option. |
| 181 | |
| 182 | @findex edebug-eval-top-level-form |
| 183 | Another command, @kbd{M-x edebug-eval-top-level-form}, is available to |
| 184 | instrument any top-level form regardless of the values of |
| 185 | @code{edebug-all-defs} and @code{edebug-all-forms}. |
| 186 | |
| 187 | While Edebug is active, the command @kbd{I} |
| 188 | (@code{edebug-instrument-callee}) instruments the definition of the |
| 189 | function or macro called by the list form after point, if is not already |
| 190 | instrumented. This is possible only if Edebug knows where to find the |
| 191 | source for that function; for this reading, after loading Edebug, |
| 192 | @code{eval-region} records the position of every definition it |
| 193 | evaluates, even if not instrumenting it. See also the @kbd{i} command |
| 194 | (@pxref{Jumping}), which steps into the call after instrumenting the |
| 195 | function. |
| 196 | |
| 197 | @cindex special forms (Edebug) |
| 198 | @cindex interactive commands (Edebug) |
| 199 | @cindex anonymous lambda expressions (Edebug) |
| 200 | @cindex Common Lisp (Edebug) |
| 201 | @pindex cl.el @r{(Edebug)} |
| 202 | @pindex cl-specs.el |
| 203 | Edebug knows how to instrument all the standard special forms, |
| 204 | @code{interactive} forms with an expression argument, anonymous lambda |
| 205 | expressions, and other defining forms. However, Edebug cannot determine |
| 206 | on its own what a user-defined macro will do with the arguments of a |
| 207 | macro call, so you must provide that information; see @ref{Instrumenting |
| 208 | Macro Calls}, for details. |
| 209 | |
| 210 | When Edebug is about to instrument code for the first time in a |
| 211 | session, it runs the hook @code{edebug-setup-hook}, then sets it to |
| 212 | @code{nil}. You can use this to load Edebug specifications |
| 213 | (@pxref{Instrumenting Macro Calls}) associated with a package you are |
| 214 | using, but only when you use Edebug. |
| 215 | |
| 216 | @findex eval-expression @r{(Edebug)} |
| 217 | To remove instrumentation from a definition, simply re-evaluate its |
| 218 | definition in a way that does not instrument. There are two ways of |
| 219 | evaluating forms that never instrument them: from a file with |
| 220 | @code{load}, and from the minibuffer with @code{eval-expression} |
| 221 | (@kbd{M-:}). |
| 222 | |
| 223 | If Edebug detects a syntax error while instrumenting, it leaves point |
| 224 | at the erroneous code and signals an @code{invalid-read-syntax} error. |
| 225 | |
| 226 | @xref{Edebug Eval}, for other evaluation functions available |
| 227 | inside of Edebug. |
| 228 | |
| 229 | @node Edebug Execution Modes |
| 230 | @subsection Edebug Execution Modes |
| 231 | |
| 232 | @cindex Edebug execution modes |
| 233 | Edebug supports several execution modes for running the program you are |
| 234 | debugging. We call these alternatives @dfn{Edebug execution modes}; do |
| 235 | not confuse them with major or minor modes. The current Edebug execution mode |
| 236 | determines how far Edebug continues execution before stopping---whether |
| 237 | it stops at each stop point, or continues to the next breakpoint, for |
| 238 | example---and how much Edebug displays the progress of the evaluation |
| 239 | before it stops. |
| 240 | |
| 241 | Normally, you specify the Edebug execution mode by typing a command to |
| 242 | continue the program in a certain mode. Here is a table of these |
| 243 | commands; all except for @kbd{S} resume execution of the program, at |
| 244 | least for a certain distance. |
| 245 | |
| 246 | @table @kbd |
| 247 | @item S |
| 248 | Stop: don't execute any more of the program, but wait for more |
| 249 | Edebug commands (@code{edebug-stop}). |
| 250 | |
| 251 | @item @key{SPC} |
| 252 | Step: stop at the next stop point encountered (@code{edebug-step-mode}). |
| 253 | |
| 254 | @item n |
| 255 | Next: stop at the next stop point encountered after an expression |
| 256 | (@code{edebug-next-mode}). Also see @code{edebug-forward-sexp} in |
| 257 | @ref{Edebug Misc}. |
| 258 | |
| 259 | @item t |
| 260 | Trace: pause one second at each Edebug stop point (@code{edebug-trace-mode}). |
| 261 | |
| 262 | @item T |
| 263 | Rapid trace: update the display at each stop point, but don't actually |
| 264 | pause (@code{edebug-Trace-fast-mode}). |
| 265 | |
| 266 | @item g |
| 267 | Go: run until the next breakpoint (@code{edebug-go-mode}). @xref{Breakpoints}. |
| 268 | |
| 269 | @item c |
| 270 | Continue: pause one second at each breakpoint, and then continue |
| 271 | (@code{edebug-continue-mode}). |
| 272 | |
| 273 | @item C |
| 274 | Rapid continue: move point to each breakpoint, but don't pause |
| 275 | (@code{edebug-Continue-fast-mode}). |
| 276 | |
| 277 | @item G |
| 278 | Go non-stop: ignore breakpoints (@code{edebug-Go-nonstop-mode}). You |
| 279 | can still stop the program by typing @kbd{S}, or any editing command. |
| 280 | @end table |
| 281 | |
| 282 | In general, the execution modes earlier in the above list run the |
| 283 | program more slowly or stop sooner than the modes later in the list. |
| 284 | |
| 285 | While executing or tracing, you can interrupt the execution by typing |
| 286 | any Edebug command. Edebug stops the program at the next stop point and |
| 287 | then executes the command you typed. For example, typing @kbd{t} during |
| 288 | execution switches to trace mode at the next stop point. You can use |
| 289 | @kbd{S} to stop execution without doing anything else. |
| 290 | |
| 291 | If your function happens to read input, a character you type intending |
| 292 | to interrupt execution may be read by the function instead. You can |
| 293 | avoid such unintended results by paying attention to when your program |
| 294 | wants input. |
| 295 | |
| 296 | @cindex keyboard macros (Edebug) |
| 297 | Keyboard macros containing the commands in this section do not |
| 298 | completely work: exiting from Edebug, to resume the program, loses track |
| 299 | of the keyboard macro. This is not easy to fix. Also, defining or |
| 300 | executing a keyboard macro outside of Edebug does not affect commands |
| 301 | inside Edebug. This is usually an advantage. See also the |
| 302 | @code{edebug-continue-kbd-macro} option (@pxref{Edebug Options}). |
| 303 | |
| 304 | When you enter a new Edebug level, the initial execution mode comes from |
| 305 | the value of the variable @code{edebug-initial-mode}. By default, this |
| 306 | specifies step mode. Note that you may reenter the same Edebug level |
| 307 | several times if, for example, an instrumented function is called |
| 308 | several times from one command. |
| 309 | |
| 310 | |
| 311 | @node Jumping |
| 312 | @subsection Jumping |
| 313 | |
| 314 | The commands described in this section execute until they reach a |
| 315 | specified location. All except @kbd{i} make a temporary breakpoint to |
| 316 | establish the place to stop, then switch to go mode. Any other |
| 317 | breakpoint reached before the intended stop point will also stop |
| 318 | execution. @xref{Breakpoints}, for the details on breakpoints. |
| 319 | |
| 320 | These commands may fail to work as expected in case of nonlocal exit, |
| 321 | as that can bypass the temporary breakpoint where you expected the |
| 322 | program to stop. |
| 323 | |
| 324 | @table @kbd |
| 325 | @item h |
| 326 | Proceed to the stop point near where point is (@code{edebug-goto-here}). |
| 327 | |
| 328 | @item f |
| 329 | Run the program forward over one expression |
| 330 | (@code{edebug-forward-sexp}). |
| 331 | |
| 332 | @item o |
| 333 | Run the program until the end of the containing sexp. |
| 334 | |
| 335 | @item i |
| 336 | Step into the function or macro called by the form after point. |
| 337 | @end table |
| 338 | |
| 339 | The @kbd{h} command proceeds to the stop point near the current location |
| 340 | of point, using a temporary breakpoint. See @ref{Breakpoints}, for more |
| 341 | information about breakpoints. |
| 342 | |
| 343 | The @kbd{f} command runs the program forward over one expression. More |
| 344 | precisely, it sets a temporary breakpoint at the position that |
| 345 | @kbd{C-M-f} would reach, then executes in go mode so that the program |
| 346 | will stop at breakpoints. |
| 347 | |
| 348 | With a prefix argument @var{n}, the temporary breakpoint is placed |
| 349 | @var{n} sexps beyond point. If the containing list ends before @var{n} |
| 350 | more elements, then the place to stop is after the containing |
| 351 | expression. |
| 352 | |
| 353 | You must check that the position @kbd{C-M-f} finds is a place that the |
| 354 | program will really get to. In @code{cond}, for example, this may not |
| 355 | be true. |
| 356 | |
| 357 | For flexibility, the @kbd{f} command does @code{forward-sexp} starting |
| 358 | at point, rather than at the stop point. If you want to execute one |
| 359 | expression @emph{from the current stop point}, first type @kbd{w}, to |
| 360 | move point there, and then type @kbd{f}. |
| 361 | |
| 362 | The @kbd{o} command continues ``out of'' an expression. It places a |
| 363 | temporary breakpoint at the end of the sexp containing point. If the |
| 364 | containing sexp is a function definition itself, @kbd{o} continues until |
| 365 | just before the last sexp in the definition. If that is where you are |
| 366 | now, it returns from the function and then stops. In other words, this |
| 367 | command does not exit the currently executing function unless you are |
| 368 | positioned after the last sexp. |
| 369 | |
| 370 | The @kbd{i} command steps into the function or macro called by the list |
| 371 | form after point, and stops at its first stop point. Note that the form |
| 372 | need not be the one about to be evaluated. But if the form is a |
| 373 | function call about to be evaluated, remember to use this command before |
| 374 | any of the arguments are evaluated, since otherwise it will be too late. |
| 375 | |
| 376 | The @kbd{i} command instruments the function or macro it's supposed to |
| 377 | step into, if it isn't instrumented already. This is convenient, but keep |
| 378 | in mind that the function or macro remains instrumented unless you explicitly |
| 379 | arrange to deinstrument it. |
| 380 | |
| 381 | @node Edebug Misc |
| 382 | @subsection Miscellaneous Edebug Commands |
| 383 | |
| 384 | Some miscellaneous Edebug commands are described here. |
| 385 | |
| 386 | @table @kbd |
| 387 | @item ? |
| 388 | Display the help message for Edebug (@code{edebug-help}). |
| 389 | |
| 390 | @item C-] |
| 391 | Abort one level back to the previous command level |
| 392 | (@code{abort-recursive-edit}). |
| 393 | |
| 394 | @item q |
| 395 | Return to the top level editor command loop (@code{top-level}). This |
| 396 | exits all recursive editing levels, including all levels of Edebug |
| 397 | activity. However, instrumented code protected with |
| 398 | @code{unwind-protect} or @code{condition-case} forms may resume |
| 399 | debugging. |
| 400 | |
| 401 | @item Q |
| 402 | Like @kbd{q}, but don't stop even for protected code |
| 403 | (@code{top-level-nonstop}). |
| 404 | |
| 405 | @item r |
| 406 | Redisplay the most recently known expression result in the echo area |
| 407 | (@code{edebug-previous-result}). |
| 408 | |
| 409 | @item d |
| 410 | Display a backtrace, excluding Edebug's own functions for clarity |
| 411 | (@code{edebug-backtrace}). |
| 412 | |
| 413 | You cannot use debugger commands in the backtrace buffer in Edebug as |
| 414 | you would in the standard debugger. |
| 415 | |
| 416 | The backtrace buffer is killed automatically when you continue |
| 417 | execution. |
| 418 | @end table |
| 419 | |
| 420 | You can invoke commands from Edebug that activate Edebug again |
| 421 | recursively. Whenever Edebug is active, you can quit to the top level |
| 422 | with @kbd{q} or abort one recursive edit level with @kbd{C-]}. You can |
| 423 | display a backtrace of all the pending evaluations with @kbd{d}. |
| 424 | |
| 425 | @node Breakpoints |
| 426 | @subsection Breakpoints |
| 427 | |
| 428 | @cindex breakpoints |
| 429 | Edebug's step mode stops execution when the next stop point is reached. |
| 430 | There are three other ways to stop Edebug execution once it has started: |
| 431 | breakpoints, the global break condition, and source breakpoints. |
| 432 | |
| 433 | While using Edebug, you can specify @dfn{breakpoints} in the program you |
| 434 | are testing: these are places where execution should stop. You can set a |
| 435 | breakpoint at any stop point, as defined in @ref{Using Edebug}. For |
| 436 | setting and unsetting breakpoints, the stop point that is affected is |
| 437 | the first one at or after point in the source code buffer. Here are the |
| 438 | Edebug commands for breakpoints: |
| 439 | |
| 440 | @table @kbd |
| 441 | @item b |
| 442 | Set a breakpoint at the stop point at or after point |
| 443 | (@code{edebug-set-breakpoint}). If you use a prefix argument, the |
| 444 | breakpoint is temporary---it turns off the first time it stops the |
| 445 | program. |
| 446 | |
| 447 | @item u |
| 448 | Unset the breakpoint (if any) at the stop point at or after |
| 449 | point (@code{edebug-unset-breakpoint}). |
| 450 | |
| 451 | @item x @var{condition} @key{RET} |
| 452 | Set a conditional breakpoint which stops the program only if |
| 453 | @var{condition} evaluates to a non-@code{nil} value |
| 454 | (@code{edebug-set-conditional-breakpoint}). With a prefix argument, the |
| 455 | breakpoint is temporary. |
| 456 | |
| 457 | @item B |
| 458 | Move point to the next breakpoint in the current definition |
| 459 | (@code{edebug-next-breakpoint}). |
| 460 | @end table |
| 461 | |
| 462 | While in Edebug, you can set a breakpoint with @kbd{b} and unset one |
| 463 | with @kbd{u}. First move point to the Edebug stop point of your choice, |
| 464 | then type @kbd{b} or @kbd{u} to set or unset a breakpoint there. |
| 465 | Unsetting a breakpoint where none has been set has no effect. |
| 466 | |
| 467 | Re-evaluating or reinstrumenting a definition removes all of its |
| 468 | previous breakpoints. |
| 469 | |
| 470 | A @dfn{conditional breakpoint} tests a condition each time the program |
| 471 | gets there. Any errors that occur as a result of evaluating the |
| 472 | condition are ignored, as if the result were @code{nil}. To set a |
| 473 | conditional breakpoint, use @kbd{x}, and specify the condition |
| 474 | expression in the minibuffer. Setting a conditional breakpoint at a |
| 475 | stop point that has a previously established conditional breakpoint puts |
| 476 | the previous condition expression in the minibuffer so you can edit it. |
| 477 | |
| 478 | You can make a conditional or unconditional breakpoint |
| 479 | @dfn{temporary} by using a prefix argument with the command to set the |
| 480 | breakpoint. When a temporary breakpoint stops the program, it is |
| 481 | automatically unset. |
| 482 | |
| 483 | Edebug always stops or pauses at a breakpoint, except when the Edebug |
| 484 | mode is Go-nonstop. In that mode, it ignores breakpoints entirely. |
| 485 | |
| 486 | To find out where your breakpoints are, use the @kbd{B} command, which |
| 487 | moves point to the next breakpoint following point, within the same |
| 488 | function, or to the first breakpoint if there are no following |
| 489 | breakpoints. This command does not continue execution---it just moves |
| 490 | point in the buffer. |
| 491 | |
| 492 | @menu |
| 493 | * Global Break Condition:: Breaking on an event. |
| 494 | * Source Breakpoints:: Embedding breakpoints in source code. |
| 495 | @end menu |
| 496 | |
| 497 | |
| 498 | @node Global Break Condition |
| 499 | @subsubsection Global Break Condition |
| 500 | |
| 501 | @cindex stopping on events |
| 502 | @cindex global break condition |
| 503 | A @dfn{global break condition} stops execution when a specified |
| 504 | condition is satisfied, no matter where that may occur. Edebug |
| 505 | evaluates the global break condition at every stop point; if it |
| 506 | evaluates to a non-@code{nil} value, then execution stops or pauses |
| 507 | depending on the execution mode, as if a breakpoint had been hit. If |
| 508 | evaluating the condition gets an error, execution does not stop. |
| 509 | |
| 510 | @findex edebug-set-global-break-condition |
| 511 | The condition expression is stored in |
| 512 | @code{edebug-global-break-condition}. You can specify a new expression |
| 513 | using the @kbd{X} command (@code{edebug-set-global-break-condition}). |
| 514 | |
| 515 | The global break condition is the simplest way to find where in your |
| 516 | code some event occurs, but it makes code run much more slowly. So you |
| 517 | should reset the condition to @code{nil} when not using it. |
| 518 | |
| 519 | @node Source Breakpoints |
| 520 | @subsubsection Source Breakpoints |
| 521 | |
| 522 | @findex edebug |
| 523 | @cindex source breakpoints |
| 524 | All breakpoints in a definition are forgotten each time you |
| 525 | reinstrument it. If you wish to make a breakpoint that won't be |
| 526 | forgotten, you can write a @dfn{source breakpoint}, which is simply a |
| 527 | call to the function @code{edebug} in your source code. You can, of |
| 528 | course, make such a call conditional. For example, in the @code{fac} |
| 529 | function, you can insert the first line as shown below, to stop when the |
| 530 | argument reaches zero: |
| 531 | |
| 532 | @example |
| 533 | (defun fac (n) |
| 534 | (if (= n 0) (edebug)) |
| 535 | (if (< 0 n) |
| 536 | (* n (fac (1- n))) |
| 537 | 1)) |
| 538 | @end example |
| 539 | |
| 540 | When the @code{fac} definition is instrumented and the function is |
| 541 | called, the call to @code{edebug} acts as a breakpoint. Depending on |
| 542 | the execution mode, Edebug stops or pauses there. |
| 543 | |
| 544 | If no instrumented code is being executed when @code{edebug} is called, |
| 545 | that function calls @code{debug}. |
| 546 | @c This may not be a good idea anymore. |
| 547 | |
| 548 | @node Trapping Errors |
| 549 | @subsection Trapping Errors |
| 550 | |
| 551 | Emacs normally displays an error message when an error is signaled and |
| 552 | not handled with @code{condition-case}. While Edebug is active and |
| 553 | executing instrumented code, it normally responds to all unhandled |
| 554 | errors. You can customize this with the options @code{edebug-on-error} |
| 555 | and @code{edebug-on-quit}; see @ref{Edebug Options}. |
| 556 | |
| 557 | When Edebug responds to an error, it shows the last stop point |
| 558 | encountered before the error. This may be the location of a call to a |
| 559 | function which was not instrumented, and within which the error actually |
| 560 | occurred. For an unbound variable error, the last known stop point |
| 561 | might be quite distant from the offending variable reference. In that |
| 562 | case, you might want to display a full backtrace (@pxref{Edebug Misc}). |
| 563 | |
| 564 | @c Edebug should be changed for the following: -- dan |
| 565 | If you change @code{debug-on-error} or @code{debug-on-quit} while |
| 566 | Edebug is active, these changes will be forgotten when Edebug becomes |
| 567 | inactive. Furthermore, during Edebug's recursive edit, these variables |
| 568 | are bound to the values they had outside of Edebug. |
| 569 | |
| 570 | @node Edebug Views |
| 571 | @subsection Edebug Views |
| 572 | |
| 573 | These Edebug commands let you view aspects of the buffer and window |
| 574 | status as they were before entry to Edebug. The outside window |
| 575 | configuration is the collection of windows and contents that were in |
| 576 | effect outside of Edebug. |
| 577 | |
| 578 | @table @kbd |
| 579 | @item v |
| 580 | Temporarily view the outside window configuration |
| 581 | (@code{edebug-view-outside}). |
| 582 | |
| 583 | @item p |
| 584 | Temporarily display the outside current buffer with point at its outside |
| 585 | position (@code{edebug-bounce-point}). With a prefix argument @var{n}, |
| 586 | pause for @var{n} seconds instead. |
| 587 | |
| 588 | @item w |
| 589 | Move point back to the current stop point in the source code buffer |
| 590 | (@code{edebug-where}). |
| 591 | |
| 592 | If you use this command in a different window displaying the same |
| 593 | buffer, that window will be used instead to display the current |
| 594 | definition in the future. |
| 595 | |
| 596 | @item W |
| 597 | @c Its function is not simply to forget the saved configuration -- dan |
| 598 | Toggle whether Edebug saves and restores the outside window |
| 599 | configuration (@code{edebug-toggle-save-windows}). |
| 600 | |
| 601 | With a prefix argument, @code{W} only toggles saving and restoring of |
| 602 | the selected window. To specify a window that is not displaying the |
| 603 | source code buffer, you must use @kbd{C-x X W} from the global keymap. |
| 604 | @end table |
| 605 | |
| 606 | You can view the outside window configuration with @kbd{v} or just |
| 607 | bounce to the point in the current buffer with @kbd{p}, even if |
| 608 | it is not normally displayed. After moving point, you may wish to jump |
| 609 | back to the stop point with @kbd{w} from a source code buffer. |
| 610 | |
| 611 | Each time you use @kbd{W} to turn saving @emph{off}, Edebug forgets the |
| 612 | saved outside window configuration---so that even if you turn saving |
| 613 | back @emph{on}, the current window configuration remains unchanged when |
| 614 | you next exit Edebug (by continuing the program). However, the |
| 615 | automatic redisplay of @samp{*edebug*} and @samp{*edebug-trace*} may |
| 616 | conflict with the buffers you wish to see unless you have enough windows |
| 617 | open. |
| 618 | |
| 619 | @node Edebug Eval |
| 620 | @subsection Evaluation |
| 621 | |
| 622 | While within Edebug, you can evaluate expressions ``as if'' Edebug |
| 623 | were not running. Edebug tries to be invisible to the expression's |
| 624 | evaluation and printing. Evaluation of expressions that cause side |
| 625 | effects will work as expected, except for changes to data that Edebug |
| 626 | explicitly saves and restores. @xref{The Outside Context}, for details |
| 627 | on this process. |
| 628 | |
| 629 | @table @kbd |
| 630 | @item e @var{exp} @key{RET} |
| 631 | Evaluate expression @var{exp} in the context outside of Edebug |
| 632 | (@code{edebug-eval-expression}). That is, Edebug tries to minimize its |
| 633 | interference with the evaluation. |
| 634 | |
| 635 | @item M-: @var{exp} @key{RET} |
| 636 | Evaluate expression @var{exp} in the context of Edebug itself. |
| 637 | |
| 638 | @item C-x C-e |
| 639 | Evaluate the expression before point, in the context outside of Edebug |
| 640 | (@code{edebug-eval-last-sexp}). |
| 641 | @end table |
| 642 | |
| 643 | @cindex lexical binding (Edebug) |
| 644 | Edebug supports evaluation of expressions containing references to |
| 645 | lexically bound symbols created by the following constructs in |
| 646 | @file{cl.el} (version 2.03 or later): @code{lexical-let}, |
| 647 | @code{macrolet}, and @code{symbol-macrolet}. |
| 648 | |
| 649 | @node Eval List |
| 650 | @subsection Evaluation List Buffer |
| 651 | |
| 652 | You can use the @dfn{evaluation list buffer}, called @samp{*edebug*}, to |
| 653 | evaluate expressions interactively. You can also set up the |
| 654 | @dfn{evaluation list} of expressions to be evaluated automatically each |
| 655 | time Edebug updates the display. |
| 656 | |
| 657 | @table @kbd |
| 658 | @item E |
| 659 | Switch to the evaluation list buffer @samp{*edebug*} |
| 660 | (@code{edebug-visit-eval-list}). |
| 661 | @end table |
| 662 | |
| 663 | In the @samp{*edebug*} buffer you can use the commands of Lisp |
| 664 | Interaction mode (@pxref{Lisp Interaction,,, emacs, The GNU Emacs |
| 665 | Manual}) as well as these special commands: |
| 666 | |
| 667 | @table @kbd |
| 668 | @item C-j |
| 669 | Evaluate the expression before point, in the outside context, and insert |
| 670 | the value in the buffer (@code{edebug-eval-print-last-sexp}). |
| 671 | |
| 672 | @item C-x C-e |
| 673 | Evaluate the expression before point, in the context outside of Edebug |
| 674 | (@code{edebug-eval-last-sexp}). |
| 675 | |
| 676 | @item C-c C-u |
| 677 | Build a new evaluation list from the contents of the buffer |
| 678 | (@code{edebug-update-eval-list}). |
| 679 | |
| 680 | @item C-c C-d |
| 681 | Delete the evaluation list group that point is in |
| 682 | (@code{edebug-delete-eval-item}). |
| 683 | |
| 684 | @item C-c C-w |
| 685 | Switch back to the source code buffer at the current stop point |
| 686 | (@code{edebug-where}). |
| 687 | @end table |
| 688 | |
| 689 | You can evaluate expressions in the evaluation list window with |
| 690 | @kbd{C-j} or @kbd{C-x C-e}, just as you would in @samp{*scratch*}; |
| 691 | but they are evaluated in the context outside of Edebug. |
| 692 | |
| 693 | The expressions you enter interactively (and their results) are lost |
| 694 | when you continue execution; but you can set up an @dfn{evaluation list} |
| 695 | consisting of expressions to be evaluated each time execution stops. |
| 696 | |
| 697 | @cindex evaluation list group |
| 698 | To do this, write one or more @dfn{evaluation list groups} in the |
| 699 | evaluation list buffer. An evaluation list group consists of one or |
| 700 | more Lisp expressions. Groups are separated by comment lines. |
| 701 | |
| 702 | The command @kbd{C-c C-u} (@code{edebug-update-eval-list}) rebuilds the |
| 703 | evaluation list, scanning the buffer and using the first expression of |
| 704 | each group. (The idea is that the second expression of the group is the |
| 705 | value previously computed and displayed.) |
| 706 | |
| 707 | Each entry to Edebug redisplays the evaluation list by inserting each |
| 708 | expression in the buffer, followed by its current value. It also |
| 709 | inserts comment lines so that each expression becomes its own group. |
| 710 | Thus, if you type @kbd{C-c C-u} again without changing the buffer text, |
| 711 | the evaluation list is effectively unchanged. |
| 712 | |
| 713 | If an error occurs during an evaluation from the evaluation list, the |
| 714 | error message is displayed in a string as if it were the result. |
| 715 | Therefore, expressions that use variables not currently valid do not |
| 716 | interrupt your debugging. |
| 717 | |
| 718 | Here is an example of what the evaluation list window looks like after |
| 719 | several expressions have been added to it: |
| 720 | |
| 721 | @smallexample |
| 722 | (current-buffer) |
| 723 | #<buffer *scratch*> |
| 724 | ;--------------------------------------------------------------- |
| 725 | (selected-window) |
| 726 | #<window 16 on *scratch*> |
| 727 | ;--------------------------------------------------------------- |
| 728 | (point) |
| 729 | 196 |
| 730 | ;--------------------------------------------------------------- |
| 731 | bad-var |
| 732 | "Symbol's value as variable is void: bad-var" |
| 733 | ;--------------------------------------------------------------- |
| 734 | (recursion-depth) |
| 735 | 0 |
| 736 | ;--------------------------------------------------------------- |
| 737 | this-command |
| 738 | eval-last-sexp |
| 739 | ;--------------------------------------------------------------- |
| 740 | @end smallexample |
| 741 | |
| 742 | To delete a group, move point into it and type @kbd{C-c C-d}, or simply |
| 743 | delete the text for the group and update the evaluation list with |
| 744 | @kbd{C-c C-u}. To add a new expression to the evaluation list, insert |
| 745 | the expression at a suitable place, insert a new comment line, then type |
| 746 | @kbd{C-c C-u}. You need not insert dashes in the comment line---its |
| 747 | contents don't matter. |
| 748 | |
| 749 | After selecting @samp{*edebug*}, you can return to the source code |
| 750 | buffer with @kbd{C-c C-w}. The @samp{*edebug*} buffer is killed when |
| 751 | you continue execution, and recreated next time it is needed. |
| 752 | |
| 753 | @node Printing in Edebug |
| 754 | @subsection Printing in Edebug |
| 755 | |
| 756 | @cindex printing (Edebug) |
| 757 | @cindex printing circular structures |
| 758 | @pindex cust-print |
| 759 | If an expression in your program produces a value containing circular |
| 760 | list structure, you may get an error when Edebug attempts to print it. |
| 761 | |
| 762 | One way to cope with circular structure is to set @code{print-length} |
| 763 | or @code{print-level} to truncate the printing. Edebug does this for |
| 764 | you; it binds @code{print-length} and @code{print-level} to 50 if they |
| 765 | were @code{nil}. (Actually, the variables @code{edebug-print-length} |
| 766 | and @code{edebug-print-level} specify the values to use within Edebug.) |
| 767 | @xref{Output Variables}. |
| 768 | |
| 769 | @defopt edebug-print-length |
| 770 | If non-@code{nil}, Edebug binds @code{print-length} to this value while |
| 771 | printing results. The default value is @code{50}. |
| 772 | @end defopt |
| 773 | |
| 774 | @defopt edebug-print-level |
| 775 | If non-@code{nil}, Edebug binds @code{print-level} to this value while |
| 776 | printing results. The default value is @code{50}. |
| 777 | @end defopt |
| 778 | |
| 779 | You can also print circular structures and structures that share |
| 780 | elements more informatively by binding @code{print-circle} |
| 781 | to a non-@code{nil} value. |
| 782 | |
| 783 | Here is an example of code that creates a circular structure: |
| 784 | |
| 785 | @example |
| 786 | (setq a '(x y)) |
| 787 | (setcar a a) |
| 788 | @end example |
| 789 | |
| 790 | @noindent |
| 791 | Custom printing prints this as @samp{Result: #1=(#1# y)}. The |
| 792 | @samp{#1=} notation labels the structure that follows it with the label |
| 793 | @samp{1}, and the @samp{#1#} notation references the previously labeled |
| 794 | structure. This notation is used for any shared elements of lists or |
| 795 | vectors. |
| 796 | |
| 797 | @defopt edebug-print-circle |
| 798 | If non-@code{nil}, Edebug binds @code{print-circle} to this value while |
| 799 | printing results. The default value is @code{nil}. |
| 800 | @end defopt |
| 801 | |
| 802 | Other programs can also use custom printing; see @file{cust-print.el} |
| 803 | for details. |
| 804 | |
| 805 | @node Trace Buffer |
| 806 | @subsection Trace Buffer |
| 807 | @cindex trace buffer |
| 808 | |
| 809 | Edebug can record an execution trace, storing it in a buffer named |
| 810 | @samp{*edebug-trace*}. This is a log of function calls and returns, |
| 811 | showing the function names and their arguments and values. To enable |
| 812 | trace recording, set @code{edebug-trace} to a non-@code{nil} value. |
| 813 | |
| 814 | Making a trace buffer is not the same thing as using trace execution |
| 815 | mode (@pxref{Edebug Execution Modes}). |
| 816 | |
| 817 | When trace recording is enabled, each function entry and exit adds |
| 818 | lines to the trace buffer. A function entry record consists of |
| 819 | @samp{::::@{}, followed by the function name and argument values. A |
| 820 | function exit record consists of @samp{::::@}}, followed by the function |
| 821 | name and result of the function. |
| 822 | |
| 823 | The number of @samp{:}s in an entry shows its recursion depth. You |
| 824 | can use the braces in the trace buffer to find the matching beginning or |
| 825 | end of function calls. |
| 826 | |
| 827 | @findex edebug-print-trace-before |
| 828 | @findex edebug-print-trace-after |
| 829 | You can customize trace recording for function entry and exit by |
| 830 | redefining the functions @code{edebug-print-trace-before} and |
| 831 | @code{edebug-print-trace-after}. |
| 832 | |
| 833 | @defmac edebug-tracing string body@dots{} |
| 834 | This macro requests additional trace information around the execution |
| 835 | of the @var{body} forms. The argument @var{string} specifies text |
| 836 | to put in the trace buffer. All the arguments are evaluated, and |
| 837 | @code{edebug-tracing} returns the value of the last form in @var{body}. |
| 838 | @end defmac |
| 839 | |
| 840 | @defun edebug-trace format-string &rest format-args |
| 841 | This function inserts text in the trace buffer. It computes the text |
| 842 | with @code{(apply 'format @var{format-string} @var{format-args})}. |
| 843 | It also appends a newline to separate entries. |
| 844 | @end defun |
| 845 | |
| 846 | @code{edebug-tracing} and @code{edebug-trace} insert lines in the |
| 847 | trace buffer whenever they are called, even if Edebug is not active. |
| 848 | Adding text to the trace buffer also scrolls its window to show the last |
| 849 | lines inserted. |
| 850 | |
| 851 | @node Coverage Testing |
| 852 | @subsection Coverage Testing |
| 853 | |
| 854 | @cindex coverage testing |
| 855 | @cindex frequency counts |
| 856 | @cindex performance analysis |
| 857 | Edebug provides rudimentary coverage testing and display of execution |
| 858 | frequency. |
| 859 | |
| 860 | Coverage testing works by comparing the result of each expression with |
| 861 | the previous result; each form in the program is considered ``covered'' |
| 862 | if it has returned two different values since you began testing coverage |
| 863 | in the current Emacs session. Thus, to do coverage testing on your |
| 864 | program, execute it under various conditions and note whether it behaves |
| 865 | correctly; Edebug will tell you when you have tried enough different |
| 866 | conditions that each form has returned two different values. |
| 867 | |
| 868 | Coverage testing makes execution slower, so it is only done if |
| 869 | @code{edebug-test-coverage} is non-@code{nil}. Frequency counting is |
| 870 | performed for all execution of an instrumented function, even if the |
| 871 | execution mode is Go-nonstop, and regardless of whether coverage testing |
| 872 | is enabled. |
| 873 | |
| 874 | Use @kbd{M-x edebug-display-freq-count} to display both the |
| 875 | coverage information and the frequency counts for a definition. |
| 876 | |
| 877 | @deffn Command edebug-display-freq-count |
| 878 | This command displays the frequency count data for each line of the |
| 879 | current definition. |
| 880 | |
| 881 | The frequency counts appear as comment lines after each line of code, |
| 882 | and you can undo all insertions with one @code{undo} command. The |
| 883 | counts appear under the @samp{(} before an expression or the @samp{)} |
| 884 | after an expression, or on the last character of a variable. To |
| 885 | simplify the display, a count is not shown if it is equal to the |
| 886 | count of an earlier expression on the same line. |
| 887 | |
| 888 | The character @samp{=} following the count for an expression says that |
| 889 | the expression has returned the same value each time it was evaluated. |
| 890 | In other words, it is not yet ``covered'' for coverage testing purposes. |
| 891 | |
| 892 | To clear the frequency count and coverage data for a definition, |
| 893 | simply reinstrument it with @code{eval-defun}. |
| 894 | @end deffn |
| 895 | |
| 896 | For example, after evaluating @code{(fac 5)} with a source |
| 897 | breakpoint, and setting @code{edebug-test-coverage} to @code{t}, when |
| 898 | the breakpoint is reached, the frequency data looks like this: |
| 899 | |
| 900 | @example |
| 901 | (defun fac (n) |
| 902 | (if (= n 0) (edebug)) |
| 903 | ;#6 1 0 =5 |
| 904 | (if (< 0 n) |
| 905 | ;#5 = |
| 906 | (* n (fac (1- n))) |
| 907 | ;# 5 0 |
| 908 | 1)) |
| 909 | ;# 0 |
| 910 | @end example |
| 911 | |
| 912 | The comment lines show that @code{fac} was called 6 times. The |
| 913 | first @code{if} statement returned 5 times with the same result each |
| 914 | time; the same is true of the condition on the second @code{if}. |
| 915 | The recursive call of @code{fac} did not return at all. |
| 916 | |
| 917 | |
| 918 | @node The Outside Context |
| 919 | @subsection The Outside Context |
| 920 | |
| 921 | Edebug tries to be transparent to the program you are debugging, but it |
| 922 | does not succeed completely. Edebug also tries to be transparent when |
| 923 | you evaluate expressions with @kbd{e} or with the evaluation list |
| 924 | buffer, by temporarily restoring the outside context. This section |
| 925 | explains precisely what context Edebug restores, and how Edebug fails to |
| 926 | be completely transparent. |
| 927 | |
| 928 | @menu |
| 929 | * Checking Whether to Stop:: When Edebug decides what to do. |
| 930 | * Edebug Display Update:: When Edebug updates the display. |
| 931 | * Edebug Recursive Edit:: When Edebug stops execution. |
| 932 | @end menu |
| 933 | |
| 934 | @node Checking Whether to Stop |
| 935 | @subsubsection Checking Whether to Stop |
| 936 | |
| 937 | Whenever Edebug is entered, it needs to save and restore certain data |
| 938 | before even deciding whether to make trace information or stop the |
| 939 | program. |
| 940 | |
| 941 | @itemize @bullet |
| 942 | @item |
| 943 | @code{max-lisp-eval-depth} and @code{max-specpdl-size} are both |
| 944 | incremented once to reduce Edebug's impact on the stack. You could, |
| 945 | however, still run out of stack space when using Edebug. |
| 946 | |
| 947 | @item |
| 948 | The state of keyboard macro execution is saved and restored. While |
| 949 | Edebug is active, @code{executing-macro} is bound to |
| 950 | @code{edebug-continue-kbd-macro}. |
| 951 | |
| 952 | @end itemize |
| 953 | |
| 954 | |
| 955 | @node Edebug Display Update |
| 956 | @subsubsection Edebug Display Update |
| 957 | |
| 958 | @c This paragraph is not filled, because LaLiberte's conversion script |
| 959 | @c needs an xref to be on just one line. |
| 960 | When Edebug needs to display something (e.g., in trace mode), it saves |
| 961 | the current window configuration from ``outside'' Edebug |
| 962 | (@pxref{Window Configurations}). When you exit Edebug (by continuing |
| 963 | the program), it restores the previous window configuration. |
| 964 | |
| 965 | Emacs redisplays only when it pauses. Usually, when you continue |
| 966 | execution, the program re-enters Edebug at a breakpoint or after |
| 967 | stepping, without pausing or reading input in between. In such cases, |
| 968 | Emacs never gets a chance to redisplay the ``outside'' configuration. |
| 969 | Consequently, what you see is the same window configuration as the last |
| 970 | time Edebug was active, with no interruption. |
| 971 | |
| 972 | Entry to Edebug for displaying something also saves and restores the |
| 973 | following data (though some of them are deliberately not restored if an |
| 974 | error or quit signal occurs). |
| 975 | |
| 976 | @itemize @bullet |
| 977 | @item |
| 978 | @cindex current buffer point and mark (Edebug) |
| 979 | Which buffer is current, and the positions of point and the mark in the |
| 980 | current buffer, are saved and restored. |
| 981 | |
| 982 | @item |
| 983 | @cindex window configuration (Edebug) |
| 984 | The outside window configuration is saved and restored if |
| 985 | @code{edebug-save-windows} is non-@code{nil} (@pxref{Edebug Display Update}). |
| 986 | |
| 987 | The window configuration is not restored on error or quit, but the |
| 988 | outside selected window @emph{is} reselected even on error or quit in |
| 989 | case a @code{save-excursion} is active. If the value of |
| 990 | @code{edebug-save-windows} is a list, only the listed windows are saved |
| 991 | and restored. |
| 992 | |
| 993 | The window start and horizontal scrolling of the source code buffer are |
| 994 | not restored, however, so that the display remains coherent within Edebug. |
| 995 | |
| 996 | @item |
| 997 | The value of point in each displayed buffer is saved and restored if |
| 998 | @code{edebug-save-displayed-buffer-points} is non-@code{nil}. |
| 999 | |
| 1000 | @item |
| 1001 | The variables @code{overlay-arrow-position} and |
| 1002 | @code{overlay-arrow-string} are saved and restored. So you can safely |
| 1003 | invoke Edebug from the recursive edit elsewhere in the same buffer. |
| 1004 | |
| 1005 | @item |
| 1006 | @code{cursor-in-echo-area} is locally bound to @code{nil} so that |
| 1007 | the cursor shows up in the window. |
| 1008 | @end itemize |
| 1009 | |
| 1010 | @node Edebug Recursive Edit |
| 1011 | @subsubsection Edebug Recursive Edit |
| 1012 | |
| 1013 | When Edebug is entered and actually reads commands from the user, it |
| 1014 | saves (and later restores) these additional data: |
| 1015 | |
| 1016 | @itemize @bullet |
| 1017 | @item |
| 1018 | The current match data. @xref{Match Data}. |
| 1019 | |
| 1020 | @item |
| 1021 | @code{last-command}, @code{this-command}, @code{last-command-char}, |
| 1022 | @code{last-input-char}, @code{last-input-event}, |
| 1023 | @code{last-command-event}, @code{last-event-frame}, |
| 1024 | @code{last-nonmenu-event}, and @code{track-mouse}. Commands used within |
| 1025 | Edebug do not affect these variables outside of Edebug. |
| 1026 | |
| 1027 | The key sequence returned by @code{this-command-keys} is changed by |
| 1028 | executing commands within Edebug and there is no way to reset |
| 1029 | the key sequence from Lisp. |
| 1030 | |
| 1031 | Edebug cannot save and restore the value of |
| 1032 | @code{unread-command-events}. Entering Edebug while this variable has a |
| 1033 | nontrivial value can interfere with execution of the program you are |
| 1034 | debugging. |
| 1035 | |
| 1036 | @item |
| 1037 | Complex commands executed while in Edebug are added to the variable |
| 1038 | @code{command-history}. In rare cases this can alter execution. |
| 1039 | |
| 1040 | @item |
| 1041 | Within Edebug, the recursion depth appears one deeper than the recursion |
| 1042 | depth outside Edebug. This is not true of the automatically updated |
| 1043 | evaluation list window. |
| 1044 | |
| 1045 | @item |
| 1046 | @code{standard-output} and @code{standard-input} are bound to @code{nil} |
| 1047 | by the @code{recursive-edit}, but Edebug temporarily restores them during |
| 1048 | evaluations. |
| 1049 | |
| 1050 | @item |
| 1051 | The state of keyboard macro definition is saved and restored. While |
| 1052 | Edebug is active, @code{defining-kbd-macro} is bound to |
| 1053 | @code{edebug-continue-kbd-macro}. |
| 1054 | @end itemize |
| 1055 | |
| 1056 | @node Instrumenting Macro Calls |
| 1057 | @subsection Instrumenting Macro Calls |
| 1058 | |
| 1059 | When Edebug instruments an expression that calls a Lisp macro, it needs |
| 1060 | additional information about the macro to do the job properly. This is |
| 1061 | because there is no a-priori way to tell which subexpressions of the |
| 1062 | macro call are forms to be evaluated. (Evaluation may occur explicitly |
| 1063 | in the macro body, or when the resulting expansion is evaluated, or any |
| 1064 | time later.) |
| 1065 | |
| 1066 | Therefore, you must define an Edebug specification for each macro |
| 1067 | that Edebug will encounter, to explain the format of calls to that |
| 1068 | macro. To do this, add an @code{edebug} declaration to the macro |
| 1069 | definition. Here is a simple example that shows the specification for |
| 1070 | the @code{for} example macro (@pxref{Argument Evaluation}). |
| 1071 | |
| 1072 | @example |
| 1073 | (defmacro for (var from init to final do &rest body) |
| 1074 | "Execute a simple \"for\" loop. |
| 1075 | For example, (for i from 1 to 10 do (print i))." |
| 1076 | (declare (edebug symbolp "from" form "to" form "do" &rest form)) |
| 1077 | ...) |
| 1078 | @end example |
| 1079 | |
| 1080 | @defspec declare (edebug @var{specification}) |
| 1081 | Specify which expressions of a call to the macro in which the |
| 1082 | declaration appears are forms to be evaluated. For simple macros, the |
| 1083 | @var{specification} often looks very similar to the formal argument list |
| 1084 | of the macro definition, but specifications are much more general than |
| 1085 | macro arguments. |
| 1086 | @end defspec |
| 1087 | |
| 1088 | You can also define an edebug specification for a macro separately |
| 1089 | from the macro definition with @code{def-edebug-spec}. Adding |
| 1090 | @code{edebug} declarations is preferred, and more convenient, for |
| 1091 | macro definitions in Lisp, but @code{def-edebug-spec} makes it |
| 1092 | possible to define Edebug specifications for special forms implemented |
| 1093 | in C. |
| 1094 | |
| 1095 | @deffn Macro def-edebug-spec macro specification |
| 1096 | Specify which expressions of a call to macro @var{macro} are forms to be |
| 1097 | evaluated. @var{specification} should be the edebug specification. |
| 1098 | It is not evaluated. |
| 1099 | |
| 1100 | The @var{macro} argument can actually be any symbol, not just a macro |
| 1101 | name. |
| 1102 | @end deffn |
| 1103 | |
| 1104 | Here is a table of the possibilities for @var{specification} and how each |
| 1105 | directs processing of arguments. |
| 1106 | |
| 1107 | @table @asis |
| 1108 | @item @code{t} |
| 1109 | All arguments are instrumented for evaluation. |
| 1110 | |
| 1111 | @item @code{0} |
| 1112 | None of the arguments is instrumented. |
| 1113 | |
| 1114 | @item a symbol |
| 1115 | The symbol must have an Edebug specification which is used instead. |
| 1116 | This indirection is repeated until another kind of specification is |
| 1117 | found. This allows you to inherit the specification from another macro. |
| 1118 | |
| 1119 | @item a list |
| 1120 | The elements of the list describe the types of the arguments of a |
| 1121 | calling form. The possible elements of a specification list are |
| 1122 | described in the following sections. |
| 1123 | @end table |
| 1124 | |
| 1125 | @menu |
| 1126 | * Specification List:: How to specify complex patterns of evaluation. |
| 1127 | * Backtracking:: What Edebug does when matching fails. |
| 1128 | * Specification Examples:: To help understand specifications. |
| 1129 | @end menu |
| 1130 | |
| 1131 | |
| 1132 | @node Specification List |
| 1133 | @subsubsection Specification List |
| 1134 | |
| 1135 | @cindex Edebug specification list |
| 1136 | A @dfn{specification list} is required for an Edebug specification if |
| 1137 | some arguments of a macro call are evaluated while others are not. Some |
| 1138 | elements in a specification list match one or more arguments, but others |
| 1139 | modify the processing of all following elements. The latter, called |
| 1140 | @dfn{specification keywords}, are symbols beginning with @samp{&} (such |
| 1141 | as @code{&optional}). |
| 1142 | |
| 1143 | A specification list may contain sublists which match arguments that are |
| 1144 | themselves lists, or it may contain vectors used for grouping. Sublists |
| 1145 | and groups thus subdivide the specification list into a hierarchy of |
| 1146 | levels. Specification keywords apply only to the remainder of the |
| 1147 | sublist or group they are contained in. |
| 1148 | |
| 1149 | When a specification list involves alternatives or repetition, matching |
| 1150 | it against an actual macro call may require backtracking. |
| 1151 | @xref{Backtracking}, for more details. |
| 1152 | |
| 1153 | Edebug specifications provide the power of regular expression matching, |
| 1154 | plus some context-free grammar constructs: the matching of sublists with |
| 1155 | balanced parentheses, recursive processing of forms, and recursion via |
| 1156 | indirect specifications. |
| 1157 | |
| 1158 | Here's a table of the possible elements of a specification list, with |
| 1159 | their meanings: |
| 1160 | |
| 1161 | @table @code |
| 1162 | @item sexp |
| 1163 | A single unevaluated Lisp object, which is not instrumented. |
| 1164 | @c an "expression" is not necessarily intended for evaluation. |
| 1165 | |
| 1166 | @item form |
| 1167 | A single evaluated expression, which is instrumented. |
| 1168 | |
| 1169 | @item place |
| 1170 | @findex edebug-unwrap |
| 1171 | A place to store a value, as in the Common Lisp @code{setf} construct. |
| 1172 | |
| 1173 | @item body |
| 1174 | Short for @code{&rest form}. See @code{&rest} below. |
| 1175 | |
| 1176 | @item function-form |
| 1177 | A function form: either a quoted function symbol, a quoted lambda |
| 1178 | expression, or a form (that should evaluate to a function symbol or |
| 1179 | lambda expression). This is useful when an argument that's a lambda |
| 1180 | expression might be quoted with @code{quote} rather than |
| 1181 | @code{function}, since it instruments the body of the lambda expression |
| 1182 | either way. |
| 1183 | |
| 1184 | @item lambda-expr |
| 1185 | A lambda expression with no quoting. |
| 1186 | |
| 1187 | @item &optional |
| 1188 | @kindex &optional @r{(Edebug)} |
| 1189 | All following elements in the specification list are optional; as soon |
| 1190 | as one does not match, Edebug stops matching at this level. |
| 1191 | |
| 1192 | To make just a few elements optional followed by non-optional elements, |
| 1193 | use @code{[&optional @var{specs}@dots{}]}. To specify that several |
| 1194 | elements must all match or none, use @code{&optional |
| 1195 | [@var{specs}@dots{}]}. See the @code{defun} example below. |
| 1196 | |
| 1197 | @item &rest |
| 1198 | @kindex &rest @r{(Edebug)} |
| 1199 | All following elements in the specification list are repeated zero or |
| 1200 | more times. In the last repetition, however, it is not a problem if the |
| 1201 | expression runs out before matching all of the elements of the |
| 1202 | specification list. |
| 1203 | |
| 1204 | To repeat only a few elements, use @code{[&rest @var{specs}@dots{}]}. |
| 1205 | To specify several elements that must all match on every repetition, use |
| 1206 | @code{&rest [@var{specs}@dots{}]}. |
| 1207 | |
| 1208 | @item &or |
| 1209 | @kindex &or @r{(Edebug)} |
| 1210 | Each of the following elements in the specification list is an |
| 1211 | alternative. One of the alternatives must match, or the @code{&or} |
| 1212 | specification fails. |
| 1213 | |
| 1214 | Each list element following @code{&or} is a single alternative. To |
| 1215 | group two or more list elements as a single alternative, enclose them in |
| 1216 | @code{[@dots{}]}. |
| 1217 | |
| 1218 | @item ¬ |
| 1219 | @kindex ¬ @r{(Edebug)} |
| 1220 | Each of the following elements is matched as alternatives as if by using |
| 1221 | @code{&or}, but if any of them match, the specification fails. If none |
| 1222 | of them match, nothing is matched, but the @code{¬} specification |
| 1223 | succeeds. |
| 1224 | |
| 1225 | @item &define |
| 1226 | @kindex &define @r{(Edebug)} |
| 1227 | Indicates that the specification is for a defining form. The defining |
| 1228 | form itself is not instrumented (that is, Edebug does not stop before and |
| 1229 | after the defining form), but forms inside it typically will be |
| 1230 | instrumented. The @code{&define} keyword should be the first element in |
| 1231 | a list specification. |
| 1232 | |
| 1233 | @item nil |
| 1234 | This is successful when there are no more arguments to match at the |
| 1235 | current argument list level; otherwise it fails. See sublist |
| 1236 | specifications and the backquote example below. |
| 1237 | |
| 1238 | @item gate |
| 1239 | @cindex preventing backtracking |
| 1240 | No argument is matched but backtracking through the gate is disabled |
| 1241 | while matching the remainder of the specifications at this level. This |
| 1242 | is primarily used to generate more specific syntax error messages. See |
| 1243 | @ref{Backtracking}, for more details. Also see the @code{let} example |
| 1244 | below. |
| 1245 | |
| 1246 | @item @var{other-symbol} |
| 1247 | @cindex indirect specifications |
| 1248 | Any other symbol in a specification list may be a predicate or an |
| 1249 | indirect specification. |
| 1250 | |
| 1251 | If the symbol has an Edebug specification, this @dfn{indirect |
| 1252 | specification} should be either a list specification that is used in |
| 1253 | place of the symbol, or a function that is called to process the |
| 1254 | arguments. The specification may be defined with @code{def-edebug-spec} |
| 1255 | just as for macros. See the @code{defun} example below. |
| 1256 | |
| 1257 | Otherwise, the symbol should be a predicate. The predicate is called |
| 1258 | with the argument and the specification fails if the predicate returns |
| 1259 | @code{nil}. In either case, that argument is not instrumented. |
| 1260 | |
| 1261 | Some suitable predicates include @code{symbolp}, @code{integerp}, |
| 1262 | @code{stringp}, @code{vectorp}, and @code{atom}. |
| 1263 | |
| 1264 | @item [@var{elements}@dots{}] |
| 1265 | @cindex [@dots{}] (Edebug) |
| 1266 | A vector of elements groups the elements into a single @dfn{group |
| 1267 | specification}. Its meaning has nothing to do with vectors. |
| 1268 | |
| 1269 | @item "@var{string}" |
| 1270 | The argument should be a symbol named @var{string}. This specification |
| 1271 | is equivalent to the quoted symbol, @code{'@var{symbol}}, where the name |
| 1272 | of @var{symbol} is the @var{string}, but the string form is preferred. |
| 1273 | |
| 1274 | @item (vector @var{elements}@dots{}) |
| 1275 | The argument should be a vector whose elements must match the |
| 1276 | @var{elements} in the specification. See the backquote example below. |
| 1277 | |
| 1278 | @item (@var{elements}@dots{}) |
| 1279 | Any other list is a @dfn{sublist specification} and the argument must be |
| 1280 | a list whose elements match the specification @var{elements}. |
| 1281 | |
| 1282 | @cindex dotted lists (Edebug) |
| 1283 | A sublist specification may be a dotted list and the corresponding list |
| 1284 | argument may then be a dotted list. Alternatively, the last @sc{cdr} of a |
| 1285 | dotted list specification may be another sublist specification (via a |
| 1286 | grouping or an indirect specification, e.g., @code{(spec . [(more |
| 1287 | specs@dots{})])}) whose elements match the non-dotted list arguments. |
| 1288 | This is useful in recursive specifications such as in the backquote |
| 1289 | example below. Also see the description of a @code{nil} specification |
| 1290 | above for terminating such recursion. |
| 1291 | |
| 1292 | Note that a sublist specification written as @code{(specs . nil)} |
| 1293 | is equivalent to @code{(specs)}, and @code{(specs . |
| 1294 | (sublist-elements@dots{}))} is equivalent to @code{(specs |
| 1295 | sublist-elements@dots{})}. |
| 1296 | @end table |
| 1297 | |
| 1298 | @c Need to document extensions with &symbol and :symbol |
| 1299 | |
| 1300 | Here is a list of additional specifications that may appear only after |
| 1301 | @code{&define}. See the @code{defun} example below. |
| 1302 | |
| 1303 | @table @code |
| 1304 | @item name |
| 1305 | The argument, a symbol, is the name of the defining form. |
| 1306 | |
| 1307 | A defining form is not required to have a name field; and it may have |
| 1308 | multiple name fields. |
| 1309 | |
| 1310 | @item :name |
| 1311 | This construct does not actually match an argument. The element |
| 1312 | following @code{:name} should be a symbol; it is used as an additional |
| 1313 | name component for the definition. You can use this to add a unique, |
| 1314 | static component to the name of the definition. It may be used more |
| 1315 | than once. |
| 1316 | |
| 1317 | @item arg |
| 1318 | The argument, a symbol, is the name of an argument of the defining form. |
| 1319 | However, lambda-list keywords (symbols starting with @samp{&}) |
| 1320 | are not allowed. |
| 1321 | |
| 1322 | @item lambda-list |
| 1323 | @cindex lambda-list (Edebug) |
| 1324 | This matches a lambda list---the argument list of a lambda expression. |
| 1325 | |
| 1326 | @item def-body |
| 1327 | The argument is the body of code in a definition. This is like |
| 1328 | @code{body}, described above, but a definition body must be instrumented |
| 1329 | with a different Edebug call that looks up information associated with |
| 1330 | the definition. Use @code{def-body} for the highest level list of forms |
| 1331 | within the definition. |
| 1332 | |
| 1333 | @item def-form |
| 1334 | The argument is a single, highest-level form in a definition. This is |
| 1335 | like @code{def-body}, except use this to match a single form rather than |
| 1336 | a list of forms. As a special case, @code{def-form} also means that |
| 1337 | tracing information is not output when the form is executed. See the |
| 1338 | @code{interactive} example below. |
| 1339 | @end table |
| 1340 | |
| 1341 | @node Backtracking |
| 1342 | @subsubsection Backtracking in Specifications |
| 1343 | |
| 1344 | @cindex backtracking |
| 1345 | @cindex syntax error (Edebug) |
| 1346 | If a specification fails to match at some point, this does not |
| 1347 | necessarily mean a syntax error will be signaled; instead, |
| 1348 | @dfn{backtracking} will take place until all alternatives have been |
| 1349 | exhausted. Eventually every element of the argument list must be |
| 1350 | matched by some element in the specification, and every required element |
| 1351 | in the specification must match some argument. |
| 1352 | |
| 1353 | When a syntax error is detected, it might not be reported until much |
| 1354 | later after higher-level alternatives have been exhausted, and with the |
| 1355 | point positioned further from the real error. But if backtracking is |
| 1356 | disabled when an error occurs, it can be reported immediately. Note |
| 1357 | that backtracking is also reenabled automatically in several situations; |
| 1358 | it is reenabled when a new alternative is established by |
| 1359 | @code{&optional}, @code{&rest}, or @code{&or}, or at the start of |
| 1360 | processing a sublist, group, or indirect specification. The effect of |
| 1361 | enabling or disabling backtracking is limited to the remainder of the |
| 1362 | level currently being processed and lower levels. |
| 1363 | |
| 1364 | Backtracking is disabled while matching any of the |
| 1365 | form specifications (that is, @code{form}, @code{body}, @code{def-form}, and |
| 1366 | @code{def-body}). These specifications will match any form so any error |
| 1367 | must be in the form itself rather than at a higher level. |
| 1368 | |
| 1369 | Backtracking is also disabled after successfully matching a quoted |
| 1370 | symbol or string specification, since this usually indicates a |
| 1371 | recognized construct. But if you have a set of alternative constructs that |
| 1372 | all begin with the same symbol, you can usually work around this |
| 1373 | constraint by factoring the symbol out of the alternatives, e.g., |
| 1374 | @code{["foo" &or [first case] [second case] ...]}. |
| 1375 | |
| 1376 | Most needs are satisfied by these two ways that bactracking is |
| 1377 | automatically disabled, but occasionally it is useful to explicitly |
| 1378 | disable backtracking by using the @code{gate} specification. This is |
| 1379 | useful when you know that no higher alternatives could apply. See the |
| 1380 | example of the @code{let} specification. |
| 1381 | |
| 1382 | @node Specification Examples |
| 1383 | @subsubsection Specification Examples |
| 1384 | |
| 1385 | It may be easier to understand Edebug specifications by studying |
| 1386 | the examples provided here. |
| 1387 | |
| 1388 | A @code{let} special form has a sequence of bindings and a body. Each |
| 1389 | of the bindings is either a symbol or a sublist with a symbol and |
| 1390 | optional expression. In the specification below, notice the @code{gate} |
| 1391 | inside of the sublist to prevent backtracking once a sublist is found. |
| 1392 | |
| 1393 | @example |
| 1394 | (def-edebug-spec let |
| 1395 | ((&rest |
| 1396 | &or symbolp (gate symbolp &optional form)) |
| 1397 | body)) |
| 1398 | @end example |
| 1399 | |
| 1400 | Edebug uses the following specifications for @code{defun} and |
| 1401 | @code{defmacro} and the associated argument list and @code{interactive} |
| 1402 | specifications. It is necessary to handle interactive forms specially |
| 1403 | since an expression argument it is actually evaluated outside of the |
| 1404 | function body. |
| 1405 | |
| 1406 | @smallexample |
| 1407 | (def-edebug-spec defmacro defun) ; @r{Indirect ref to @code{defun} spec.} |
| 1408 | (def-edebug-spec defun |
| 1409 | (&define name lambda-list |
| 1410 | [&optional stringp] ; @r{Match the doc string, if present.} |
| 1411 | [&optional ("interactive" interactive)] |
| 1412 | def-body)) |
| 1413 | |
| 1414 | (def-edebug-spec lambda-list |
| 1415 | (([&rest arg] |
| 1416 | [&optional ["&optional" arg &rest arg]] |
| 1417 | &optional ["&rest" arg] |
| 1418 | ))) |
| 1419 | |
| 1420 | (def-edebug-spec interactive |
| 1421 | (&optional &or stringp def-form)) ; @r{Notice: @code{def-form}} |
| 1422 | @end smallexample |
| 1423 | |
| 1424 | The specification for backquote below illustrates how to match |
| 1425 | dotted lists and use @code{nil} to terminate recursion. It also |
| 1426 | illustrates how components of a vector may be matched. (The actual |
| 1427 | specification defined by Edebug does not support dotted lists because |
| 1428 | doing so causes very deep recursion that could fail.) |
| 1429 | |
| 1430 | @smallexample |
| 1431 | (def-edebug-spec ` (backquote-form)) ; @r{Alias just for clarity.} |
| 1432 | |
| 1433 | (def-edebug-spec backquote-form |
| 1434 | (&or ([&or "," ",@@"] &or ("quote" backquote-form) form) |
| 1435 | (backquote-form . [&or nil backquote-form]) |
| 1436 | (vector &rest backquote-form) |
| 1437 | sexp)) |
| 1438 | @end smallexample |
| 1439 | |
| 1440 | |
| 1441 | @node Edebug Options |
| 1442 | @subsection Edebug Options |
| 1443 | |
| 1444 | These options affect the behavior of Edebug: |
| 1445 | |
| 1446 | @defopt edebug-setup-hook |
| 1447 | Functions to call before Edebug is used. Each time it is set to a new |
| 1448 | value, Edebug will call those functions once and then |
| 1449 | @code{edebug-setup-hook} is reset to @code{nil}. You could use this to |
| 1450 | load up Edebug specifications associated with a package you are using |
| 1451 | but only when you also use Edebug. |
| 1452 | @xref{Instrumenting}. |
| 1453 | @end defopt |
| 1454 | |
| 1455 | @defopt edebug-all-defs |
| 1456 | If this is non-@code{nil}, normal evaluation of defining forms such as |
| 1457 | @code{defun} and @code{defmacro} instruments them for Edebug. This |
| 1458 | applies to @code{eval-defun}, @code{eval-region}, @code{eval-buffer}, |
| 1459 | and @code{eval-current-buffer}. |
| 1460 | |
| 1461 | Use the command @kbd{M-x edebug-all-defs} to toggle the value of this |
| 1462 | option. @xref{Instrumenting}. |
| 1463 | @end defopt |
| 1464 | |
| 1465 | @defopt edebug-all-forms |
| 1466 | If this is non-@code{nil}, the commands @code{eval-defun}, |
| 1467 | @code{eval-region}, @code{eval-buffer}, and @code{eval-current-buffer} |
| 1468 | instrument all forms, even those that don't define anything. |
| 1469 | This doesn't apply to loading or evaluations in the minibuffer. |
| 1470 | |
| 1471 | Use the command @kbd{M-x edebug-all-forms} to toggle the value of this |
| 1472 | option. @xref{Instrumenting}. |
| 1473 | @end defopt |
| 1474 | |
| 1475 | @defopt edebug-save-windows |
| 1476 | If this is non-@code{nil}, Edebug saves and restores the window |
| 1477 | configuration. That takes some time, so if your program does not care |
| 1478 | what happens to the window configurations, it is better to set this |
| 1479 | variable to @code{nil}. |
| 1480 | |
| 1481 | If the value is a list, only the listed windows are saved and |
| 1482 | restored. |
| 1483 | |
| 1484 | You can use the @kbd{W} command in Edebug to change this variable |
| 1485 | interactively. @xref{Edebug Display Update}. |
| 1486 | @end defopt |
| 1487 | |
| 1488 | @defopt edebug-save-displayed-buffer-points |
| 1489 | If this is non-@code{nil}, Edebug saves and restores point in all |
| 1490 | displayed buffers. |
| 1491 | |
| 1492 | Saving and restoring point in other buffers is necessary if you are |
| 1493 | debugging code that changes the point of a buffer which is displayed in |
| 1494 | a non-selected window. If Edebug or the user then selects the window, |
| 1495 | point in that buffer will move to the window's value of point. |
| 1496 | |
| 1497 | Saving and restoring point in all buffers is expensive, since it |
| 1498 | requires selecting each window twice, so enable this only if you need |
| 1499 | it. @xref{Edebug Display Update}. |
| 1500 | @end defopt |
| 1501 | |
| 1502 | @defopt edebug-initial-mode |
| 1503 | If this variable is non-@code{nil}, it specifies the initial execution |
| 1504 | mode for Edebug when it is first activated. Possible values are |
| 1505 | @code{step}, @code{next}, @code{go}, @code{Go-nonstop}, @code{trace}, |
| 1506 | @code{Trace-fast}, @code{continue}, and @code{Continue-fast}. |
| 1507 | |
| 1508 | The default value is @code{step}. |
| 1509 | @xref{Edebug Execution Modes}. |
| 1510 | @end defopt |
| 1511 | |
| 1512 | @defopt edebug-trace |
| 1513 | Non-@code{nil} means display a trace of function entry and exit. |
| 1514 | Tracing output is displayed in a buffer named @samp{*edebug-trace*}, one |
| 1515 | function entry or exit per line, indented by the recursion level. |
| 1516 | |
| 1517 | The default value is @code{nil}. |
| 1518 | |
| 1519 | Also see @code{edebug-tracing}, in @ref{Trace Buffer}. |
| 1520 | @end defopt |
| 1521 | |
| 1522 | @defopt edebug-test-coverage |
| 1523 | If non-@code{nil}, Edebug tests coverage of all expressions debugged. |
| 1524 | @xref{Coverage Testing}. |
| 1525 | @end defopt |
| 1526 | |
| 1527 | @defopt edebug-continue-kbd-macro |
| 1528 | If non-@code{nil}, continue defining or executing any keyboard macro |
| 1529 | that is executing outside of Edebug. Use this with caution since it is not |
| 1530 | debugged. |
| 1531 | @xref{Edebug Execution Modes}. |
| 1532 | @end defopt |
| 1533 | |
| 1534 | @defopt edebug-on-error |
| 1535 | Edebug binds @code{debug-on-error} to this value, if |
| 1536 | @code{debug-on-error} was previously @code{nil}. @xref{Trapping |
| 1537 | Errors}. |
| 1538 | @end defopt |
| 1539 | |
| 1540 | @defopt edebug-on-quit |
| 1541 | Edebug binds @code{debug-on-quit} to this value, if |
| 1542 | @code{debug-on-quit} was previously @code{nil}. @xref{Trapping |
| 1543 | Errors}. |
| 1544 | @end defopt |
| 1545 | |
| 1546 | If you change the values of @code{edebug-on-error} or |
| 1547 | @code{edebug-on-quit} while Edebug is active, their values won't be used |
| 1548 | until the @emph{next} time Edebug is invoked via a new command. |
| 1549 | @c Not necessarily a deeper command level. |
| 1550 | @c A new command is not precisely true, but that is close enough -- dan |
| 1551 | |
| 1552 | @defopt edebug-global-break-condition |
| 1553 | If non-@code{nil}, an expression to test for at every stop point. |
| 1554 | If the result is non-nil, then break. Errors are ignored. |
| 1555 | @xref{Global Break Condition}. |
| 1556 | @end defopt |