tcl-debug.ps

Linux下的tcl语言调试器

1 10 Q
(6) 54 27 T
0 12 Q
(T) 277.22 27 T
(cl/Tk W) 284.12 27 T
(orkshop - June 10-1) 325.11 27 T
(1, 1993 - Berkeley) 426.08 27 T
(, CA) 516.68 27 T
0 14 Q
(The \322s\323 command \320 \322Step\323) 54 728.67 T
1 12 Q
0.7 (The n command executes a procedure atomical-) 54 709 P
-0.16 (ly) 54 695 P
-0.16 (.  It is possible to step into a procedure with the) 62.55 695 P
2 F
(s) 54 681 T
1 F
( command.) 58.67 681 T
1.97 (W) 54 662 P
1.97 (e\325ll rewind this scenario to just before p4 is) 64.36 662 P
(about to be executed.) 54 648 T
1 10 Q
(1) 155.92 652.8 T
3 11 Q
(4: p4 $b) 72 629.67 T
(expect5.5> s) 72 618.67 T
(7: expr 1+$x) 72 607.67 T
(expect8.6>) 72 596.67 T
1 12 Q
0.17 (\322expr 1+$x\323 is the \336rst command to be executed) 54 582 P
0.48 (inside of p4.  It is nested inside of two brackets,) 54 568 P
-0.22 (plus the procedure call of p4, so the stack level is) 54 554 P
(increased by three.) 54 540 T
0.5 (After the s command, the debugger stops before) 54 521 P
0.1 (the \336rst command in the procedure and waits for) 54 507 P
(more interactive commands.) 54 493 T
1.46 (If the command that is about to be executed is) 54 474 P
(not a procedure, then s and n behave identically) 54 460 T
(.) 282.04 460 T
1.78 (Both s and n take an optional ar) 54 441 P
1.78 (gument in the) 218.48 441 P
2.26 (form of a number describing how many com-) 54 427 P
(mands to execute.) 54 413 T
(For example:) 54 394 T
3 11 Q
(s 2) 72 375.67 T
(s 100) 72 364.67 T
(s $b) 72 353.67 T
(s [expr 2+[p4 $b]]) 72 342.67 T
1 12 Q
2.62 (The ar) 54 328 P
2.62 (guments are evaluated according to the) 87.37 328 P
1.71 (usual T) 54 314 P
1.71 (cl rules because s and n are commands) 90.51 314 P
(known to T) 54 300 T
(cl.) 108.46 300 T
1.92 (The debugger will not interrupt procedures in-) 54 281 P
1.35 (voked from the command-line.  This is usually) 54 267 P
1.99 (the desired behavior) 54 253 P
1.99 (, although it is possible to) 154.76 253 P
(change this.) 54 239 T
0 14 Q
(The \322w\323 Command \320 \322Wher) 54 213.67 T
(e\323) 226.32 213.67 T
1 12 Q
0.9 (In the current scenario, we are about to execute) 54 194 P
1.97 (\322expr 1+$x\323 in the procedure p4.  W) 54 180 P
1.97 (e can re-) 242.78 180 P
0.18 (mind ourselves of this by displaying the stack of) 54 166 P
(procedure scopes using the) 54 152 T
2 F
(w) 186.56 152 T
1 F
( command.) 194.56 152 T
3 11 Q
(7: expr 1+$x) 72 133.67 T
(expect8.6> w) 72 122.67 T
( 0: expect -D 1 debug-test.exp) 72 111.67 T
(*1: p4 1) 72 100.67 T
54 78 288 93 C
63 91 207 91 2 L
0.5 H
2 Z
0 X
0 K
N
0 0 612 792 C
1 10 Q
0 X
0 K
0.29 (1. There is no actual command to \322rewind\323 com-) 72 71.33 P
(mands, alas.) 72 59.33 T
3 11 Q
( 7: expr 1+1) 324 730.67 T
1 12 Q
0.45 (The \336rst line describes scope 0.  This is the top-) 306 716 P
1.64 (level scope of the \336le itself, and the command) 306 702 P
0.83 (used to invoke the program is shown.  The sec-) 306 688 P
6.73 (ond line describes scope 1 which is the) 306 674 P
-0.15 (invocation of procedure p4.  The last line is not a) 306 660 P
0.93 (scope but just repeats the evaluation stack level) 306 646 P
(and the command about to be executed.) 306 632 T
1.55 (Notice that when w prints commands, they are) 306 613 P
0.26 (displayed using the literal values of each param-) 306 599 P
9.91 (eter) 306 585 P
9.91 (.  In contrast, when the debugger) 323.32 585 P
0.26 (automatically prints out the next command to be) 306 571 P
-0.19 (executed, the command is printed as it was origi-) 306 557 P
2.77 (nally entered in the script.  For example, the) 306 543 P
4.36 (debugger initially stopped and printed \322expr) 306 529 P
0.68 (1+$w\323, but the same instruction shows as \322expr) 306 515 P
(1+1\323 in the output from the w command.) 306 501 T
0 14 Q
(The Curr) 306 475.67 T
(ent Scope) 362.88 475.67 T
1 12 Q
1.4 (Executing fourteen steps \050via \322s 14\323\051 brings us) 306 456 P
(to the \336rst command in procedure p3.) 306 442 T
3 11 Q
(expect8.8> s 14) 324 423.67 T
(4: set m 0) 324 412.67 T
(expect5.9> w) 324 401.67 T
( 0: expect -D 1 debug-test.exp) 324 390.67 T
( 1: p1) 324 379.67 T
( 2: p2) 324 368.67 T
(*3: p3) 324 357.67 T
( 4: set m 0) 324 346.67 T
1 12 Q
0.03 (The asterisk denotes that p3 is the) 306 332 P
2 F
0.03 (curr) 471.8 332 P
0.03 (ent scope) 492 332 P
1 F
0.03 (.) 537 332 P
0.78 (W) 306 318 P
0.78 (e can now execute T) 316.36 318 P
0.78 (cl commands appropriate) 417.2 318 P
2.56 (to the scope of p3.  This includes commands) 306 304 P
(such as global, uplevel, and upvar) 306 290 T
(.) 468.22 290 T
3 11 Q
(expect5.10> uplevel {set c}) 324 271.67 T
(4) 324 260.67 T
0 14 Q
3.37 (The \322u\323 and \322d\323 commands \320 \322Up\323) 306 237.67 P
(and \322Down\323) 306 221.67 T
1 12 Q
0.02 (The current scope can be changed by the u and d) 306 202 P
0.02 (commands.) 306 188 P
2 F
0.02 (u) 367 188 P
1 F
0.02 ( moves the current scope up, while) 373 188 P
2 F
1.26 (d) 306 174 P
1 F
1.26 ( moves it down.  Interactive variable accesses) 312 174 P
(always refer to the current scope.) 306 160 T
3 11 Q
(expect5.11> u) 324 141.67 T
(expect5.12> w) 324 130.67 T
( 0: expect -D 1 debug-test.exp) 324 119.67 T
( 1: p1) 324 108.67 T
(*2: p2) 324 97.67 T
( 3: p3) 324 86.67 T
( 4: set m 0) 324 75.67 T
(expect5.13> set c) 324 64.67 T
FMENDPAGE
%%EndPage: "6" 5
%%Page: "7" 5
612 792 0 FMBEGINPAGE
0 12 Q
0 X
0 K
(Don Libes) 72 758.33 T
(A Debugger for T) 391.87 758.33 T
(cl Applications) 481.7 758.33 T
(T) 72 27 T
(cl/Tk W) 78.9 27 T
(orkshop - June 10-1) 119.89 27 T
(1, 1993 - Berkeley) 220.85 27 T
(, CA) 311.46 27 T
1 10 Q
(7) 553 27 T
3 11 Q
(4) 90 730.67 T
1 12 Q
2.01 (Both u and d accept an ar) 72 716 P
2.01 (gument representing) 205.73 716 P
-0.15 (the number of scopes by which to move.  For ex-) 72 702 P
(ample, \322u 2\323 moves from scope 2 to scope 0.) 72 688 T
3 11 Q
(expect5.14> u 2) 90 669.67 T
(expect5.15> w) 90 658.67 T
(*0: expect -D 1 debug-test.exp) 90 647.67 T
( 1: p1) 90 636.67 T
( 2: p2) 90 625.67 T
( 3: p3) 90 614.67 T
( 4: set m 0) 90 603.67 T
1 12 Q
-0.17 (An absolute scope is also accepted in the form of) 72 589 P
(\322#\323 followed by a scope number) 72 575 T
(, such as \322#3\323.) 227.06 575 T
3 11 Q
(expect5.16> u #3) 90 556.67 T
(expect5.17> w) 90 545.67 T
( 0: expect -D 1 debug-test.exp) 90 534.67 T
( 1: p1) 90 523.67 T
( 2: p2) 90 512.67 T
(*3: p3) 90 501.67 T
( 4: set m 0) 90 490.67 T
1 12 Q
0.83 (When an absolute scope is named, either u or d) 72 476 P
0.5 (may be used, irrespective of which direction the) 72 462 P
(new scope lies.) 72 448 T
0.96 (Moving the scope does not af) 72 429 P
0.96 (fect the command) 218.15 429 P
0.25 (that is about to be executed.  If a command such) 72 415 P
0.58 (as s or n is given, the current scope is automati-) 72 401 P
5.7 (cally reset to whereever is appropriate for) 72 387 P
(execution of the new command.) 72 373 T
0 14 Q
(The \322r\323 Command \320 \322Return\323) 72 347.67 T
1 12 Q
0.45 (The) 72 328 P
2 F
0.45 (r) 94.1 328 P
1 F
0.45 ( command completes execution of the cur-) 98.77 328 P
0.28 (rent procedure.  In other words, it stops after the) 72 314 P
(current procedure returns.) 72 300 T
3 11 Q
(expect5.18> r) 90 281.67 T
(3: set d 5) 90 270.67 T
(expect4.19> w) 90 259.67 T
( 0: expect -D 1 debug-test.exp) 90 248.67 T
( 1: p1) 90 237.67 T
(*2: p2) 90 226.67 T
( 3: set d 5) 90 215.67 T
(expect4.20> r) 90 204.67 T
(2: set a 3) 90 193.67 T
(expect3.21> w) 90 182.67 T
( 0: expect -D 1 debug-test.exp) 90 171.67 T
(*1: p1) 90 160.67 T
( 2: set a 3) 90 149.67 T
(expect3.22> r) 90 138.67 T
(1: set k 7) 90 127.67 T
(expect2.23> w) 90 116.67 T
(*0: expect -D 1 debug-test.exp) 90 105.67 T
( 1: set k 7) 90 94.67 T
(expect2.24> r) 90 83.67 T
(nowhere to return to) 90 72.67 T
0 14 Q
(The \322c\323 Command \320 \322Continue\323) 324 728.67 T
1 12 Q
3.25 (The) 324 709 P
2 F
3.25 (c) 348.9 709 P
1 F
3.25 ( command lets execution of commands) 354.22 709 P
1.92 (continue without having to single-step.  In the) 324 695 P
0.43 (scenario so far) 324 681 P
0.43 (, given a command anywhere, the) 394.31 681 P
-0.22 (program would continue until the script ends and) 324 667 P
(the shell prompt appears.) 324 653 T
3 11 Q
(expect2.25> c) 342 634.67 T
(%) 342 623.67 T
1 12 Q
-0.22 (The c command is also useful in other ways.  Af-) 324 609 P
4.86 (ter setting breakpoints, the program can be) 324 595 P
-0.27 (continued until it hits a breakpoint.  The program) 324 581 P
0.4 (can also be continued until a signal occurs, such) 324 567 P
(as by the user pressing ^C.) 324 553 T
0 14 Q
(The \322b\323 Command \320 \322Br) 324 527.67 T
(eakpoint\323) 475.34 527.67 T
1 12 Q
1.6 (Prior commands have shown how to execute a) 324 508 P
1.75 (\336xed number of commands or procedure calls.) 324 494 P
2.12 (In contrast, breakpoints provide a way to stop) 324 480 P
0.97 (execution upon a condition.  The conditions in-) 324 466 P
(clude:) 324 452 T
(\245) 341.5 433 T
(line number and \336lename matching) 358.49 433 T
(\245) 341.5 417 T
(expression testing) 358.49 417 T
(\245) 341.5 401 T
(command and ar) 358.49 401 T
(gument name matching) 438.21 401 T
0 F
(Br) 324 379 T
(eakpoint by Line Number and Filename) 337.11 379 T
0 10 Q
(1) 542.02 383.8 T
1 12 Q
1.31 (Line numbers and \336lenames are the most com-) 324 360 P
1.05 (mon way to specify a breakpoint.  This form is) 324 346 P
-0.14 (correspondingly the most compact.  For example) 324 332 P
4.63 (the following command causes execution to) 324 318 P
(break before executing line 7.) 324 304 T
3 11 Q
(expect2.26> b 7) 342 285.67 T
(0) 342 274.67 T
1 12 Q
0.5 (After creation of a breakpoint, an integer identi-) 324 260 P
1.23 (fying the breakpoint is printed.  The reason for) 324 246 P
(this will be described later) 324 232 T
(.) 449.92 232 T
0.76 (By default, the line number refers to the \336le as-) 324 213 P
-0.02 (sociated with the current scope.  A \336lename may) 324 199 P
1.67 (be used to refer to a dif) 324 185 P
1.67 (ferent \336le.  A colon is) 445.05 185 P
(used to separate the \336lename and line number) 324 171 T
(.) 542.84 171 T
3 11 Q
(expect2.27> b foo.exp:7) 342 152.67 T
324 102 558 117 C
333 115 477 115 2 L
0.5 H
2 Z
0 X
0 K
N
0 0 612 792 C
1 10 Q
0 X
0 K
1.16 (1. Breakpoints by line number and \336lename are) 342 95.33 P
0.21 (not currently supported.  See \322Line Numbers\323 on) 342 83.33 P
3.62 (page) 342 71.33 P
3.62 (13 and \322Current Limitations and Future) 363.37 71.33 P
(W) 342 59.33 T
(ork\323 on page) 350.64 59.33 T
(15.) 404.76 59.33 T
FMENDPAGE
%%EndPage: "7" 6
%%Page: "8" 6
612 792 0 FMBEGINPAGE
0 12 Q
0 X
0 K
(A Debugger for T) 54 758.33 T
(cl Applications) 143.83 758.33 T
(Don Libes) 487.68 758.33 T
1 10 Q
(8) 54 27 T
0 12 Q
(T) 277.22 27 T
(cl/Tk W) 284.12 27 T
(orkshop - June 10-1) 325.11 27 T
(1, 1993 - Berkeley) 426.08 27 T
(, CA) 516.68 27 T
(Br) 54 730 T
(eakpoint by Expr) 67.11 730 T
(ession) 156.2 730 T
1 F
0.41 (It is possible to break at a line only when an ex-) 54 711 P
3.11 (pression is true.  For example, the following) 54 697 P
2.74 (command causes execution to break at line 7) 54 683 P
(only when foo is greater than three.) 54 669 T
3 11 Q
(expect2.28> b 7 if {$foo>3}) 72 650.67 T
1 12 Q
0.21 (Expressions are the usual T) 54 636 P
0.21 (cl syntax and may be) 185.9 636 P
(arbitrarily complex.) 54 622 T
0.03 (No breakpointing occurs inside of the evaluation) 54 603 P
0.24 (of breakpoint expressions \050unless another break-) 54 589 P
(point dictates this\051.) 54 575 T
0 F
(Br) 54 553 T
(eakpoint by Pattern Match) 67.11 553 T
1 F
1.35 (It is also possible to de\336ne breakpoints by pat-) 54 534 P
2.67 (tern matching on the command or ar) 54 520 P
2.67 (guments.) 244.36 520 P
-0.12 (Regular expressions are introduced by the \337ag \322-) 54 506 P
3.04 (regexp\323 \050commonly abbreviated \322-re\323\051) 54 492 P
1 10 Q
2.53 (1) 249.28 496.8 P
1 12 Q
3.04 (.  The) 254.28 492 P
2.54 (following command stops if the string p4 ap-) 54 478 P
(pears within the command:) 54 464 T
3 11 Q
(expect2.29> b -re "p4") 72 445.67 T
(0) 72 434.67 T
1 12 Q
1.48 (W) 54 420 P
1.48 (ith our sample \336le, we can see the results of) 64.84 420 P
(this:) 54 406 T
3 11 Q
(% expect -D 1 debug-test.exp) 72 387.67 T
(1: set b 1) 72 376.67 T
(expect2.1> b -re "p4") 72 365.67 T
(0) 72 354.67 T
(expect2.2> c) 72 343.67 T
(breakpoint 0: -re "p4") 72 332.67 T
(1: proc p4 {x} {) 72 321.67 T
(        return [) 72 310.67 T
(expr 5+[expr 1+$x]]) 144 299.67 T
(}) 72 288.67 T
(expect2.3> c) 72 277.67 T
(breakpoint 0: -re "p4") 72 266.67 T
(4: p4 $b) 72 255.67 T
(expect5.4> c) 72 244.67 T
(breakpoint 0: -re "p4") 72 233.67 T
(3: expr 2+[p4 $b]) 72 222.67 T
(expect4.5> c) 72 211.67 T
(breakpoint 0: -re "p4") 72 200.67 T
(2: expr 1+[expr 2+[p4 $b]]) 72 189.67 T
1 12 Q
0.14 (The \336rst breakpoint occurred upon the de\336nition) 54 175 P
-0.01 (of p4.  The second occurred when p4 was called.) 54 161 P
54 126 288 141 C
63 139 207 139 2 L
0.5 H
2 Z
0 X
0 K
N
0 0 612 792 C
1 10 Q
0 X
0 K
1.05 (1. The debugger permits all \337ags to be abbrevi-) 72 119.33 P
0.67 (ated to the smallest unique pre\336x.  For example,) 72 107.33 P
0.85 (\322-regexp\323 can actually be abbreviated \322-r\323.  The) 72 95.33 P
-0.1 (usual quoting conventions around patterns should) 72 83.33 P
0.6 (be observed.  In this example, the quotes around) 72 71.33 P
(p4 can be omitted.) 72 59.33 T
1 12 Q
0.17 (T) 306 730 P
0.17 (wo more breakpoints occurred only because p4) 312.49 730 P
(was mentioned in the command.) 306 716 T
0.05 (W) 306 697 P
0.05 (ith appropriate regular expressions, any one of) 316.84 697 P
0.8 (these can be selected by itself.  For example, to) 306 683 P
(stop only on de\336nitions:) 306 669 T
3 11 Q
(expect2.1> b -re "proc p4 ") 324 650.67 T
1 12 Q
(T) 306 636 T
(o stop only on a call to p4 itself:) 312.49 636 T
3 11 Q
(expect2.2> b -re "^p4 ") 324 617.67 T
1 12 Q
(T) 306 603 T
(o stop only on commands which call p4:) 312.49 603 T
3 11 Q
(expect2.3> b -re "\134\134\134[p4 ") 324 584.67 T
1 12 Q
0.64 (The complexity of this last example is, perhaps,) 306 570 P
1.13 (somewhat ameliorated by the unlikelihood of it) 306 556 P
1.87 (ever being used.  It is more shown simply for) 306 542 P
0.52 (completeness.  The point is, the ability to match) 306 528 P
(on regular expressions is extremely powerful.) 306 514 T
0.64 (Multi-line patterns may be matched in the usual) 306 495 P
(way \320 using characters such as \134n and \134r) 306 481 T
(.) 495.86 481 T
1 10 Q
(2) 498.85 485.8 T
1 12 Q
2.8 (Glob-style matching is available by using the) 306 462 P
-0.13 (\337ag -glob instead of -regexp.  It works exactly as) 306 448 P
1 (in T) 306 434 P
1 (cl\325) 325.82 434 P
1 (s case command.  Since glob matches an) 337.81 434 P
2.4 (entire string by default, the equivalents to the) 306 420 P