File psl-1983/3-1/lpt/17-utilities.lpt artifact 475c5d270b part of check-in eb17ceb7f6


PSL Manual                    7 February 1983                     Utilities
section 17.0                                                      page 17.1

                                CHAPTER 17
                                CHAPTER 17
                                CHAPTER 17
                          MISCELLANEOUS UTILITIES
                          MISCELLANEOUS UTILITIES
                          MISCELLANEOUS UTILITIES




     17.1. Introduction .  .  .  .  .  .  .  .  .  .  .  .  .  .  .    17.1
     17.2. RCREF - Cross Reference Generator for PSL Files  .  .  .    17.1
          17.2.1. Restrictions.  .  .  .  .  .  .  .  .  .  .  .  .    17.2
          17.2.2. Usage .  .  .  .  .  .  .  .  .  .  .  .  .  .  .    17.3
          17.2.3. Options  .  .  .  .  .  .  .  .  .  .  .  .  .  .    17.3
     17.3. Picture RLISP.  .  .  .  .  .  .  .  .  .  .  .  .  .  .    17.4
          17.3.1. Running PictureRLISP on HP2648A and on TEKTRONIX    17.10
                  4006-1 Terminals  .  .  .  .  .  .  .  .  .  .  .  
     17.4. Tools for Defining Macros.  .  .  .  .  .  .  .  .  .  .   17.11
          17.4.1. DefMacro .  .  .  .  .  .  .  .  .  .  .  .  .  .   17.11
          17.4.2. BackQuote.  .  .  .  .  .  .  .  .  .  .  .  .  .   17.12
          17.4.3. Sharp-Sign Macros .  .  .  .  .  .  .  .  .  .  .   17.12
          17.4.4. MacroExpand .  .  .  .  .  .  .  .  .  .  .  .  .   17.13
          17.4.5. DefLambda.  .  .  .  .  .  .  .  .  .  .  .  .  .   17.13
     17.5. Simulating a Stack .  .  .  .  .  .  .  .  .  .  .  .  .   17.14
     17.6. DefStruct .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .   17.14
          17.6.1. Options  .  .  .  .  .  .  .  .  .  .  .  .  .  .   17.17
          17.6.2. Slot Options.  .  .  .  .  .  .  .  .  .  .  .  .   17.18
          17.6.3. A Simple Example  .  .  .  .  .  .  .  .  .  .  .   17.18
     17.7. DefConst  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .   17.21
     17.8. Functions for Sorting .  .  .  .  .  .  .  .  .  .  .  .   17.22
     17.9. Hashing Cons .  .  .  .  .  .  .  .  .  .  .  .  .  .  .   17.23
     17.10. Graph-to-Tree  .  .  .  .  .  .  .  .  .  .  .  .  .  .   17.25
     17.11. Inspect Utility.  .  .  .  .  .  .  .  .  .  .  .  .  .   17.25




17.1. Introduction
17.1. Introduction
17.1. Introduction

  This chapter describes an assortment of utility packages.  Its purpose is
to  record  the  existence  and  capabilities  of  a number of tools.  More
information on existing packages can be found by looking at the current set
of HELP files (DIR PH:*.* on the DEC-20).



17.2. RCREF - Cross Reference Generator for PSL Files
17.2. RCREF - Cross Reference Generator for PSL Files
17.2. RCREF - Cross Reference Generator for PSL Files

  RCREF is a Standard LISP program for processing a set  of  Standard  LISP
function definitions to produce:


   a. A "Summary" showing:
Utilities                     7 February 1983                    PSL Manual
page 17.2                                                      section 17.2

         i. A list of files processed.
        ii. A  list  of "entry points" (functions which are not called
            or are called only by themselves).
       iii. A list of undefined functions (functions  called  but  not
            defined in this set of functions).
        iv. A  list  of  variables  that were used non-locally but not
            declared GLOBAL or FLUID before their use.
         v. A list of variables that were declared GLOBAL but used  as
            FLUIDs (i.e. bound in a function).
        vi. A  list  of  FLUID  variables  that  were  not  bound in a
            function  so  that  one  might  consider  declaring   them
            GLOBALs.
       vii. A list of all GLOBAL variables present.
      viii. A list of all FLUID variables present.
        ix. A list of all functions present.


   b. A  "global  variable  usage"  table,  showing for each non-local
      variable:


         i. Functions in which it is  used  as  a  declared  FLUID  or
            GLOBAL.
        ii. Functions in which it is used but not declared before.
       iii. Functions in which it is bound.
                                                SetQ
                                                SetQ
        iv. Functions in which it is changed by SetQ.


   c. A "function usage" table showing for each function:


         i. Where it is defined.
        ii. Functions which call this function.
       iii. Functions called by it.
        iv. Non-local variables used.


  The output is alphabetized on the first seven characters of each function
name.

  RCREF  also  checks  that functions are called with the correct number of
arguments.


17.2.1. Restrictions
17.2.1. Restrictions
17.2.1. Restrictions

  Algebraic procedures in REDUCE are treated as if they were  symbolic,  so
that  algebraic  constructs actually appear as calls to symbolic functions,
        AEval
        AEval
such as AEval.

  SYSLISP procedures are not correctly analyzed.
PSL Manual                    7 February 1983                     Utilities
section 17.2                                                      page 17.3

17.2.2. Usage
17.2.2. Usage
17.2.2. Usage

  RCREF  should  be  used in PSL:RLISP.  To make a file FILE.CRF which is a
cross reference listing for files FILE1.EX1 and FILE2.EX2 do the  following
in RLISP:

   @PSL:RLISP
   LOAD RCREF;       % RCREF is now autoloading, so this may be omitted

   OUT "file.crf";   % later, CREFOUT ...
   ON CREF;
   IN "file1.ex1","file2.ex2";
   OFF CREF;
   SHUT "file.crf";  % later CREFEND

To process more files, more IN statements may be added, or the IN statement
may be changed to include more files.


17.2.3. Options
17.2.3. Options
17.2.3. Options


               __________                                            ______
!*CREFSUMMARY [Initially: NIL]                                       switch

     If  the  switch  CREFSUMMARY  is  ON then only the summary (see 1
     above) is produced.

  Functions with the flag NOLIST are not examined or  output.    Initially,
all  Standard  LISP functions are so flagged.  (In fact, they are kept on a
list NOLIST!*, so if you wish to see references to ALL functions, then CREF
should be first loaded with the command LOAD RCREF, and this variable  then
set to NIL).  (RCREF is now autoloading.)


          __________                                                 ______
NOLIST!* [Initially: the following list]                             global

        (AND COND LIST MAX MIN OR PLUS PROG PROG2 PROGN TIMES LAMB
        ADD1 APPEND APPLY ASSOC  ATOM CAR CDR CAAR  CADR CDAR CDDR
        CAADR CADAR CADDR  CDAAR CDADR CDDAR  CDDDR CAAAAR CAAADR
        CAADDR CADAAR CADADR  CADDAR CADDDR CDAAAR  CDAADR CDADAR
        CDDAAR CDDADR CDDDAR CDDDDR  CLOSE CODEP COMPRESS CONS  CO
        DE DEFLIST  DELETE DF  DIFFERENCE DIGIT  DIVIDE DM  EJECT
        EQUAL ERROR ERRORSET EVAL EVLIS EXPAND EXPLODE EXPT FIX FI
        FLAGP FLOAT FLOATP  FLUID FLUIDP  FUNCTION GENSYM  GET GET
        GLOBAL GLOBALP  GO GREATERP  IDP INTERN  LENGTH LESSP  LIN
        LITER LPOSN MAP  MAPC MAPCAN  MAPCAR MAPCON  MAPLIST MAX2
        MEMQ MINUS MINUSP MIN2  MKVECT NCONC NOT  NULL NUMBERP ONE
        PAGELENGTH PAIR PAIRP  PLUS2 POSN PRINC  PRINT PRIN1 PRIN2
        PUT PUTD  PUTV  QUOTE QUOTIENT  RDS  READ READCH  REMAINDE
        REMFLAG REMOB  REMPROP RETURN  REVERSE RPLACA  RPLACD SASS
        SETQ STRINGP SUBLIS SUBST SUB1 TERPRI TIMES2 UNFLUID UPBV
        WRS ZEROP)
Utilities                     7 February 1983                    PSL Manual
page 17.4                                                      section 17.2

  It  should  also  be  remembered  that  in RLISP any macros with the flag
EXPAND or, if FORCE is on, without the flag NOEXPAND  are  expanded  before
the  definition  is  seen  by the cross-reference program, so this flag can
also be used to select those macros you require expanded and those  you  do
not.  The use of ON FORCE; is highly recommended for CREF.



17.3. Picture RLISP
17.3. Picture RLISP
17.3. Picture RLISP

  [??? ReWrite ???]
  [??? ReWrite ???]
  [??? ReWrite ???]

  Picture RLISP is an ALGOL-like graphics language for Teleray, HP2648a and
Tektronix,  in  which  graphics Model primitives are combined into complete
Models for display.  PRLISP is a 3D version; PRLISP2D is a faster,  smaller
2D  version  which  also  drives  more terminals.  Two demonstration files,
PR-DEMO.RED and PR-DEMO.Sl, are available  on  PU.    See  the  help  files
PH:PRLISP.HLP and PRLISP2D.HLP.

  Model primitives include:


P:={x,y,z};
          A point (y, and z may be omitted, default to 0).

PS:=P1_ P2_ ... Pn;
          A Point Set is an ordered set of Points (Polygon).

G := PS1 & PS2 & ... PSn;
          A Group of Polygons.

Point Set Modifiers
          alter the interpretation of Point Sets within their scope.

BEZIER()  causes  the  point-set  to  be  interpreted  as the specification
          points for a BEZIER curve, open pointset.

BSPLINE() does the same for a Bspline curve, closed pointset.

TRANSFORMS:
          Mostly return a transformation matrix.

Translation:
          Move   the   specified   amount   along   the   specified   axis.
          XMOVE(deltaX);            YMOVE(deltaY);           ZMOVE(deltaZ);
          MOVE(deltaX, deltaY, deltaZ);

Scale:    Scale the Model SCALE  (factor)  XSCALE(factor);  YSCALE(factor);
          ZSCALE(factor);
          SCALE1(x.scale.factor,      y.scale.factor,      z.scale.factor);
          SCALE<Scale factor>;.  Scale along all axes.
PSL Manual                    7 February 1983                     Utilities
section 17.3                                                      page 17.5

Rotation: ROT(degrees); ROT(degrees, point.specifying.axis); XROT(degrees);
          YROT(degrees); ZROT(degrees);

Window (z.eye,z.screen):
          The WINDOW primitives assume that the viewer is located along the
          z  axis looking in the positive z direction, and that the viewing
          window is to be centered on both the x and y axis.

Vwport(leftclip,rightclip,topclip,bottomclip):
          The VWPORT, which specifies the region of  the  screen  which  is
          used for display.

REPEATED (number.of.times, my.transform):
          The  Section  of the Model which is contained within the scope of
          the Repeat Specification is replicated.  Note  that  REPEATED  is
          intended  to duplicate a sub-image in several different places on
          the screen; it was not designed for animation.

Identifiers of other Models
          the Model referred to is displayed as if  it  were  part  of  the
          current Model for dynamic display.

Calls to PictureRLISP Procedures
          This Model primitive allows procedure calls to be imbedded within
          Models.    When  the  Model  interpreter  reaches  the  procedure
          identifier it calls it, passing it the portion of the Model below
          the procedure as an argument.  The current transformation  matrix
          and  the current pen position are available to such procedures as
          the  values  of  the  global  identifiers  GLOBAL!.TRANSFORM  and
          HEREPOINT.        If   normal   procedure   call   syntax,   i.e.
          proc.name (parameters), is used then the procedure is  called  at
          Model-building  time,  but  if only the procedure's identifier is
          used then the procedure is imbedded in the Model.

ERASE()   Clears the screen and leaves the cursor at the origin.

SHOW(pict)
          Takes a picture and displays it on the screen.

ESHOW (pict)
          Erases the whole screen and display "pict".

HP!.INIT(), TEK!.INIT(), TEL!.INIT()
          Initializes the operating system's view of the characteristics of
          HP2648A   terminal,   TEKTRONIX   4006-1   (also   ADM-3A    with
          Retrographics board, and Teleray-1061).


  For example, the Model
Utilities                     7 February 1983                    PSL Manual
page 17.6                                                      section 17.3

   (A _ B _ C  &  {1,2} _ B)  |  XROT (30)  |  'TRAN ;

   %
   % PictureRLISP Commands to SHOW lots of Cubes
   %
   % Outline is a Point Set defining the 20 by 20
   %   square which is part of the Cubeface
   %
   Outline := { 10, 10} _ {-10, 10} _
             {-10,-10} _ { 10,-10} _ {10, 10};

   % Cubeface also has an Arrow on it
   %
   Arrow := {0,-1} _ {0,2}  &  {-1,1} _ {0,2} _ {1,1};

   % We are ready for the Cubeface

   Cubeface   :=   (Outline & Arrow)  |  'Tranz;

   % Note the use of static clustering to keep objects
   %  meaningful as well as the quoted Cluster
   %  to the as yet undefined transformation Tranz,
   %  which results in its evaluation being
   %  deferred until SHOW time

   % and now define the Cube

   Cube   :=   Cubeface
           &  Cubeface | XROT (180)  % 180 degrees
           &  Cubeface | YROT ( 90)
           &  Cubeface | YROT (-90)
           &  Cubeface | XROT ( 90)
           &  Cubeface | XROT (-90);
   % In order to have a more pleasant look at
   % the picture shown on the screen we magnify
   % cube by 5 times.
   BigCube := Cube | SCALE 5;

   % Set up initial Z Transform for each cube face
   %
   Tranz   :=   ZMOVE (10);  % 10 units out

   %
   % GLOBAL!.TRANSFORM has been treated as a global variable.
   % GLOBAL!.TRANSFORM should be initialized as a perspective
   % transformation matrix so that a viewer can have a correct
   % look at the picture as the viewing location changed.
   % For instance, it may be set as the desired perspective
   % with a perspective window centered at the origin and
   % of screen size 60, and the observer at -300 on the z axis.
   % Currently this has been set as default perspective transformation.
PSL Manual                    7 February 1983                     Utilities
section 17.3                                                      page 17.7

   % Now draw cube
   %
   SHOW  BigCube;

   %
Utilities                     7 February 1983                    PSL Manual
page 17.8                                                      section 17.3


   % Draw it again rotated and moved left
   %
   SHOW  (BigCube | XROT 20 | YROT 30 | ZROT 10);

   % Dynamically expand the faces out
   %
   Tranz   :=   ZMOVE 12;
   %
   SHOW  (BigCube | YROT 30 | ZROT 10);

   % Now show 5 cubes, each moved further right by 80
   %
   Tranz   :=    ZMOVE 10;
   %
   SHOW (Cube | SCALE 2.5 | XMOVE (-240) | REPEATED(5, XMOVE 80));

   %
   % Now try pointset modifier.
   % Given a pointset (polygon) as control points either a BEZIER or a
   % BSPLINE curve can be drawn.
   %
   Cpts := {0,0} _ {70,-60} _ {189,-69} _ {206,33} _ {145,130} _ {48,13
          _ {0,84} $
   %
   % Now draw Bezier curve
   % Show the polygon and the Bezier curve
   %
   SHOW (Cpts & Cpts | BEZIER());

   % Now draw Bspline curve
   % Show the polygon and the Bspline curve
   %
   SHOW (Cpts & Cpts | BSPLINE());

   % Now work on the Circle
   % Given a center position and a radius a circle is drawn
   %
   SHOW ( {10,10} | CIRCLE(50));

   %
   % Define a procedure which returns a model of
   % a Cube when passed the face to be used
   %
   Symbolic Procedure Buildcube;
    List 'Buildcube;
   % put the name onto the property list
   Put('buildcube, 'pbintrp, 'Dobuildcube);
   Symbolic Procedure Dobuildcube Face$
          Face  &  Face | XROT(180)
                &  Face | YROT(90)
                &  Face | YROT(-90)
PSL Manual                    7 February 1983                     Utilities
section 17.3                                                      page 17.9

                &  Face | XROT(90)
                &  Face | XROT(-90) ;
   % just return the value of the one statement

   % Use this procedure to display 2 cubes, with and
   %  without the Arrow - first do it by calling
   %  Buildcube at time the Model is built
   %
   P := Cubeface | Buildcube() | XMOVE(-15) &
        (Outline | 'Tranz) | Buildcube() | XMOVE 15;
   %
   SHOW (P | SCALE 5);

   % Now define a procedure which returns a Model of
   %   a cube when passed the half size parameter

   Symbolic Procedure Cubemodel;
    List 'Cubemodel;
   %put the name onto the property list
   Put('Cubemodel,'Pbintrp, 'Docubemodel);
   Symbolic Procedure Docubemodel  HSize;
    << if idp HSize then HSize := eval HSize$
       { HSize,  HSize,  HSize}  _
       {-HSize,  HSize,  HSize}  _
       {-HSize, -HSize,  HSize}  _
       { HSize, -HSize,  HSize}  _
       { HSize,  HSize,  HSize}  _
       { HSize,  HSize, -HSize}  _
       {-HSize,  HSize, -HSize}  _
       {-HSize, -HSize, -HSize}  _
       { HSize, -HSize, -HSize}  _
       { HSize,  HSize, -HSize}  &
       {-HSize,  HSize, -HSize}  _
       {-HSize,  HSize,  HSize}  &
       {-HSize, -HSize, -HSize}  _
       {-HSize, -HSize,  HSize}  &
       { HSize, -HSize, -HSize}  _
       { HSize, -HSize,  HSize} >>;

   % Imbed the parameterized cube in some Models
   %
   His!.cube :=  'His!.size | Cubemodel();
   Her!.cube :=  'Her!.size | Cubemodel();
   R  :=  His!.cube | XMOVE (60)  &
         Her!.cube | XMOVE (-60) ;

   % Set up some sizes and SHOW them

   His!.size := 50;
   Her!.size := 30;
   %
   SHOW   R ;
Utilities                     7 February 1983                    PSL Manual
page 17.10                                                     section 17.3


   %
   % Set up some different sizes and SHOW them again
   %
   His!.size := 35;
   Her!.size := 60;
   %
   SHOW R;

   %
   % Now show a triangle rotated 45 degree about the z axis.
   Rotatedtriangle  :=  {0,0} _ {50,50} _
                          {100,0} _ {0,0} | Zrot (45);
   %
   SHOW Rotatedtriangle;

   %
   % Define a procedure which returns a model of a Pyramid
   % when passed 4 vertices of a pyramid.
   % Procedure Second,Third, Fourth and Fifth are primitive procedures
   % written in the source program which return the second, the third,
   % the fourth and the fifth element of a list respectively.
   % This procedure simply takes 4 points and connects the vertices to
   % show a pyramid.
   Symbolic Procedure Pyramid (Point4); %.point4 is a pointset
          Point4 &
               Third Point4 _
               Fifth Point4 _
               Second Point4 _
               Fourth Point4 ;

   % Now give a pointset indicating 4 vertices build a pyramid
   % and show it
   %
   My!.vertices := {-40,0} _ {20,-40} _ {90,20} _ {70,100};
   My!.pyramid := Pyramid Vertices;
   %
   SHOW ( My!.pyramid | XROT 30);

   %
   %  A procedure that makes a wheel with "count"
   %  spokes rotated around the z axis.
   %  in which "count" is the number specified.
   Symbolic Procedure Dowheel(spoke,count)$
       begin scalar rotatedangle$
             count := first count$
             rotatedangle := 360.0 / count$
            return (spoke | REPEATED(count, ZROT rotatedangle))
       end$
   %
   % Now draw a wheel consisting of 8 cubes
   %
PSL Manual                    7 February 1983                     Utilities
section 17.3                                                     page 17.11

   Cubeonspoke :=  (Outline | ZMOVE 10 | SCALE 2) | buildcube();
   Eight!.cubes := Cubeonspoke | XMOVE 50 | WHEEL(8);
   %
   SHOW Eight!.cubes;

   %
   %Draw a cube in which each face consists of just
   % a wheel of 8 Outlines
   %
   Flat!.Spoke := outline | XMOVE 25$
   A!.Fancy!.Cube := Flat!.Spoke | WHEEL(8) | ZMOVE 50 | Buildcube()$
   %
   SHOW A!.Fancy!.Cube;

   %
   % Redraw the fancy cube, after changing perspective by
   % moving the observer farther out along Z axis
   %
   GLOBAL!.TRANSFORM := WINDOW(-500,60);
   %
   SHOW A!.Fancy!.Cube;

   %
   % Note the flexibility resulting from the fact that
   % both Buildcube and Wheel simply take or return any
   % Model as their argument or value

  The current version of PictureRLISP runs on HP2648A graphics terminal and
TEKTRONIX  4006-1 computer display terminal.  The screen of the HP terminal
is 720 units long in  the  X  direction,  and  360  units  high  in  the  Y
direction.   The coordinate system used in HP terminal places the origin in
approximately the center of the screen, and uses a domain of  -360  to  360
and  a  range  of  -180  to  180.    Similarly, the screen of the TEKTRONIX
terminal is 1024 units long in the X direction, and 780 units high in the Y
direction.  The same origin is used but the domain is -512 to 512 in the  X
direction and the range is -390 to 390 in the Y direction.

  Procedures  HP!.INIT  and  TEK!.INIT  are  used  to  set the terminals to
graphics mode and initiate the lower level procedures on HP  and  TEKTRONIX
terminals  respectively.    Basically,  INIT  procedures  are  written  for
different terminals depending on their  specific  characteristics.    Using
INIT  procedures  keeps terminal device dependence at the user's level to a
minimum.



17.4. Tools for Defining Macros
17.4. Tools for Defining Macros
17.4. Tools for Defining Macros

  The following (and other) macro utilities are in the  file  PU:USEFUL.SL;
Utilities                     7 February 1983                    PSL Manual
page 17.12                                                     section 17.4

                                                                     1
use (LOAD USEFUL) to access.  See PH:USEFUL.HLP for more information. 


17.4.1. DefMacro
17.4.1. DefMacro
17.4.1. DefMacro


 DefMacro
 DefMacro _ __  _ ____   _ ____    __                                 _____
(DefMacro A:id  B:form  [C:form]): id                                 macro

                                              _____
                                              _____
                                              _____
     DefMacro                                 macro      DefMacro
     DefMacro                                 macro      DefMacro
     DefMacro  is  a useful tool for defining macros.  A DefMacro form
     looks like 

        (DEFMACRO <NAME> <PATTERN> <S1> <S2> ... <Sn>)

                                              ____      __
     The <PATTERN> is an S-expression made of pairs and ids.    It  is
                                             _____
                                             _____
                                             _____
                                             macro
                                             macro
     matched  against  the  arguments of the macro much like the first
                 DeSetQ
                 DeSetQ                        __
     argument to DeSetQ.  All of the  non-NIL  ids  in  <pattern>  are
     local  variables which may be used freely in the body (the <Si>).
            _____
            _____
            _____
            macro                                          ProgN
            macro                                          ProgN
     If the macro is called the <Si> are evaluated as in a ProgN  with
     the  local  variables  in  <pattern> appropriately bound, and the
                                       DefMacro
                                       DefMacro
     value  of  <Sn>  is  returned.    DefMacro  is  often  used  with
     BackQuote.


17.4.2. BackQuote
17.4.2. BackQuote
17.4.2. BackQuote

  Note  that  the special symbols described below only work in LISP syntax,
                                                       BackQuote   UnQuote
                                                       BackQuote   UnQuote
not RLISP.  In RLISP you may simply use the functions  BackQuote,  UnQuote,
    UnQuoteL                          BackQuote
    UnQuoteL                          BackQuote
and UnQuoteL.  Load USEFUL to get the BackQuote function.

                                            _____
                                            _____
                                            _____
                                      Read  macro
                                      Read  macro
  The  backquote  symbol  "`"  is  a  Read  macro which introduces a quoted
expression which may contain the unquote symbols comma "," and comma-atsign
",@".  An appropriate form consisting of the unquoted expression  calls  to
             Cons
             Cons
the function Cons and quoted expressions are produced so that the resulting
expression looks like the quoted one except that the values of the unquoted
expressions  are substituted in the appropriate place.  ",@" splices in the
value of the subsequent expression (i.e. strips  off  the  outer  layer  of
parentheses).  Thus 

   `(a (b ,x) c d ,@x e f)

is equivalent to 

   (cons 'a (cons (list 'b x) (append '(c d) (append x '(e f)))))

In particular, if x is bound to (1 2 3) this evaluates to 


_______________

  1
   Useful was written by D. Morrison.
PSL Manual                    7 February 1983                     Utilities
section 17.4                                                     page 17.13

   (a (b (1 2 3)) c d 1 2 3 e f)


 BackQuote
 BackQuote _ ____   ____                                              _____
(BackQuote A:form): form                                              macro

     Function name for back quote `.


 UnQuote
 UnQuote _ ___   _________                                            _____
(UnQuote A:any): Undefined                                            fexpr

                                                   Eval
                                                   Eval
     Function name for comma ,.  It is an error to Eval this function;
                                   BackQuote
                                   BackQuote
     it should occur only inside a BackQuote.


 UnQuoteL
 UnQuoteL _ ___   _________                                           _____
(UnQuoteL A:any): Undefined                                           fexpr

                                                             Eval
                                                             Eval
     Function  name  for comma-atsign ,@.  It is an error to Eval this
                                             BackQuote
                                             BackQuote
     function; it should only occur inside a BackQuote.


17.4.3. Sharp-Sign Macros
17.4.3. Sharp-Sign Macros
17.4.3. Sharp-Sign Macros

  USEFUL defines several MACLISP style sharp sign read macros.   Note  that
these  only work with the LISP reader, not RLISP.  Those currently included
are

  #' :  this is like the quote mark ' but is for FUNCTION instead of QUOTE.

  #/ :  this returns the numeric  form  of  the  following  character  read
without raising it.  For example #/a is 97 while #/A is 65.

  #\  :    This  is  a  read macro for the CHAR macro, described in the PSL
manual.  Not that the argument is  raised,  if  *RAISE  is  non-nil.    For
                                                              Char
                                                              Char
example,  #\a  =  #\A  =  65, while #\!a = #\(lower a) = 97.  Char has been
redefined in USEFUL to be slightly more table driven -- users can  now  add
new  "prefixes" such as META or CONTROL: just hang the appropriate function
(from integers to integers) off the char-prefix-function  property  of  the
"prefix".    A LARGE number of additional alias for various characters have
been added, including all the "standard" ASCII names like NAK and DC1.

  #. :  this causes the following expression to be evaluated at read  time.
For example, `(1 2 #.(plus 1 2) 4) reads as (1 2 3 4)

  #+ :  this reads two expressions, and passes them to the if_system macro.
That  is,  the  first  should  be a system name, and if that is the current
system the second argument is returned by the reader.   If  not,  the  next
expression is returned.

  #-:    #- is similar, but causes the second arg to be returned only if it
is NOT the current system.
Utilities                     7 February 1983                    PSL Manual
page 17.14                                                     section 17.4

17.4.4. MacroExpand
17.4.4. MacroExpand
17.4.4. MacroExpand


 MacroExpand
 MacroExpand _ ____   _ __    ____                                    _____
(MacroExpand A:form  [B:id]): form                                    macro

                                                _____
                                                _____
                                                _____
     MacroExpand                                macro
     MacroExpand                                macro
     MacroExpand is a useful tool for debugging macro definitions.  If
                            MacroExpand                 macro
                            MacroExpand                 macro
     given  one  argument,  MacroExpand expands all the macros in that
     form.  Often one wishes for more control over this process.   For
                      _____
                      _____
                      _____
                      macro                Let
                      macro                Let
     example,  if  a  macro expands into a Let, we may not wish to see
         Let
         Let
     the Let itself  expanded  to  a  lambda  expression.    Therefore
                                            MacroExpand
                                            MacroExpand
     additional  arguments  may be given to MacroExpand.  If these are
                              _____
                              _____
                              _____
                              macro
                              macro
     supplied, they should be macros, and  only  those  specified  are
     expanded.


17.4.5. DefLambda
17.4.5. DefLambda
17.4.5. DefLambda


 DefLambda
 DefLambda                                                            _____
(DefLambda ):                                                         macro

     Yet  another  little  (two  line) macro has been added to USEFUL:
     DefLambda
     DefLambda
     DefLambda.  This defines a macro much like a  substitution  macro
      ______
      ______
      ______
      smacro
      smacro
     (smacro)  except  that  it  is a lambda expression.  Thus, modulo
                                                                 ____
                                                                 ____
                                                                 ____
                                                                 expr
                                                                 expr
     redefinability, it has the same semantics as the equivalent expr.
     It is mostly intended as an easy way to open compile things.  For
     example, we would not normally  want  to  define  a  substitution
     macro  for  a constructor (NEW-FOO X) which maps into (CONS X X),
     in case X is  expensive  to  compute  or,  far  worse,  has  side
     effects.    (DEFLAMBDA  NEW-FOO  (X)  (CONS X X)) defines it as a
     macro   which   maps    (NEW-FOO    (SETQ    BAR    (BAZ)))    to
     ((LAMBDA (X) (CONS X X)) (SETQ BAR (BAZ))).



17.5. Simulating a Stack
17.5. Simulating a Stack
17.5. Simulating a Stack

  The  following macros are in the USEFUL package.  They are convenient for
                                              ____
adding and deleting things from the head of a list.


 Push
 Push ___ ___  ___ ____   ___                                         _____
(Push ITM:any  STK:list): any                                         macro

        (PUSH ITEM STACK)

     is equivalent to 

        (SETF STACK  (CONS ITEM STACK))
PSL Manual                    7 February 1983                     Utilities
section 17.5                                                     page 17.15

 Pop
 Pop ___ ____   ___                                                   _____
(Pop STK:list): any                                                   macro

        (POP STACK)

     does 

        (SETF STACK (CDR STACK))

                                        _____
     and  returns  the  item popped off STACK.  An additional argument
                        Pop
                        Pop
     may be supplied to Pop, in which case it is a variable  which  is
     SetQ
     SetQ
     SetQ'd to the popped value.



17.6. DefStruct
17.6. DefStruct
17.6. DefStruct

  (LOAD DEFSTRUCT) to use the functions described below, or FAST!-DEFSTRUCT
to  use those functions but with fast vector operations used.  DefStruct is
similar to the Spice (Common) LISP/LISP machine/MacLISP  flavor  of  struct
definitions,  and  is  expected  to be subsumed by the Mode package.  It is
                  2
implemented in PSL  as a function which builds access macros  and  fns  for
"typed"   vectors,  including  constructor  and  alterant  macros,  a  type
predicate for the structure type, and  individual  selector/assignment  fns
for   the  elements.    DefStruct  understands  a  keyword-option  oriented
structure specification.  DefStruct is now autoloading.

  First a few miscellaneous functions on types,  before  getting  into  the
depths of defining DefStructs:


 DefstructP
 DefstructP ____ __   _____ _______                                    ____
(DefstructP NAME:id): extra-boolean                                    expr

     This   is   a  predicate  that  returns  non-NIL  (the  Defstruct
                    ____
     definition) if NAME is a structured type which has  been  defined
     using Defstruct, or NIL if it is not.


 DefstructType
 DefstructType _ ______   __                                           ____
(DefstructType S:struct): id                                           expr

     This  returns  the type name field of an instance of a structured
                     _
     type, or NIL if S cannot be a Defstruct type.






_______________

  2
   Defstruct was implemented by Russ Fish.
Utilities                     7 February 1983                    PSL Manual
page 17.16                                                     section 17.6

 SubTypeP
 SubTypeP _____ __  _____ __   _______                                 ____
(SubTypeP NAME1:id  NAME2:id): boolean                                 expr

                             _____
     This  returns  true  if NAME1 is a structured type which has been
                                                      _____
     !:Included in the definition of structured type  NAME2,  possibly
     through intermediate structure definitions.  (In other words, the
                  _____                   _____
     selectors of NAME1 can be applied to NAME2.)

  Now the function which defines the beasties, in all its gory glory:


 Defstruct
 Defstruct ____ ___ _______  __ ____    ____ _____  __ ____     __    _____
(Defstruct NAME-AND-OPTIONS:{id,list}  [SLOT-DESCS:{id,list}]): id    fexpr

     Defines  a  record-structure  data  type.    A  general  call  to
     Defstruct
     Defstruct
     Defstruct looks like this: (in RLISP syntax)

        defstruct( struct-name( option-1, option-2, ... ),
                   slot-description-1,
                   slot-description-2,
                    ...
                  );

     The name of the defined structure is returned.

  Slot-descriptions are:


slot-name( default-init, slot-option-1, slot-option-2, ... )


                                __
  Struct-name and slot-name are ids.  If there are no options  following  a
name  in  a  spec,  it  can be a bare id with no option argument list.  The
default-init form is optional and may be omitted.  The default-init form is
evaluated EACH TIME a structure is to be constructed and the value is  used
as  the initial value of the slot.  Options are either a keyword id, or the
keyword followed by its argument list.  Options are described below.

                          _____
                          _____
                          _____
                          macro
                          macro
  A call to a constructor macro has the form:

   MakeThing( slot-name-1( value-expr-1 ),
              slot-name-2( value-expr-2 ),
               ... );

The slot-name:value lists override the default-init values which were  part
of  the  structure  definition.    Note that the slot-names look like unary
functions of the value, so the parens can be left off.  A call to MakeThing
with no arguments of course takes all of the default values.  The order  of
evaluation  of  the  default-init  forms and the list of assigned values is
undefined, so code should not depend upon the ordering.

  ____________ ____
  Implementors Note: Common/LispMachine Lisps define it this  way,  but  Is
this  necessary?  It wouldn't be too tough to make the order be the same as
the struct defn, or the argument order in the constructor call.  Maybe they
PSL Manual                    7 February 1983                     Utilities
section 17.6                                                     page 17.17

think  such  things  should  not  be advertised and thus constrained in the
future.  Or perhaps the theory is that  constructs  such  as  this  can  be
compiled  more  efficiently if the ordering is flexible??  Also, should the
overridden default-init forms be evaluated or not?  I think not.

               _____
               _____
               _____
               macro
               macro
  The alterant macro calls have a similar form:

   AlterThing( thing,
               slot-name-1 value-expr-1,
               slot-name-2 value-expr-2,
                ... );

The first argument evaluates to the struct to be altered.    (The  optional
parens were left off here.)  This is just a multiple-assignment form, which
eventually  goes through the slot depositors.  Remember that the slot-names
are used, not the depositor names.  (See !:Prefix,  below.)    The  altered
structure instance is returned as the value of an Alterant macro.

  Implementators  note:  Common/LispMachine Lisp defines this such that all
of the slots are  altered  in  parallel  AFTER  the  new  value  forms  are
evaluated,  but  still with the order of evaluation of the forms undefined.
This seemed to lose more than it gained, but arguments for its  worth  will
be entertained.


17.6.1. Options
17.6.1. Options
17.6.1. Options

  Structure options appear as an argument list to the struct-name.  Many of
the  options  themselves take argument lists, which are sometimes optional.
Option  ids  all  start  with  a  colon  (!:),  on  the  theory  that  this
distinguishes them from other things.

  By  default,  the names of the constructor, alterant and predicate macros
are MakeName, AlterName and  NameP.    "Name"  is  the  struct-name.    The
!:Constructor,  !:Alterant, and !:Predicate options can be used to override
the default names.  Their argument is the name to use, and a  name  of  NIL
causes the respective macro not to be defined at all.

  The  !:Creator  option  causes  a  different  form  of  constructor to be
defined, in addition to  the  regular  "Make"  constructor  (which  can  be
suppressed.)    As  in the !:Constructor option above, an argument supplies
the name of the macro, but the default name in this case is CreateName.   A
call to a Creator macro has the form:  


    CreateThing( slot-value-1, slot-value-2, ... );


___                                      ____ __ _______
All  of the slot-values of the structure must be present, in the order they
appear in the structure definition.    No  checking  is  done,  other  than
assuring that the number of values is the same as the number of slots.  For
                                                 ___  ___  ___________
obvious  reasons,  constructors  of  this  form  are  not  recommended  for
Utilities                     7 February 1983                    PSL Manual
page 17.18                                                     section 17.6

structures with many fields, or which may be expanded or modified.

  Slot selector macros may appear on either the left side or the right side
of  an  assignment.   They are by default named the same as the slot-names,
but can be given a common prefix by the !:Prefix option.  If !:Prefix  does
not  have  an  argument,  the structure name is the prefix.  If there is an
argument, it should be a string or an id whose print name is the prefix.

  The !:Include option allows building a new  structure  definition  as  an
extension of an old one.  The required argument is the name of a previously
defined  structure  type.  The access functions for the slots of the source
type also works on instances of the new type.  This can be  used  to  build
hierarchies  of  types.    The  source types contain generic information in
common to the more specific subtypes which !:Include them.

  The !:IncludeInit option takes an argument  list  of  "slot-name(default-
init)"  pairs,  like  slot-descriptors without slot-options, and files them
away to modify the default-init values for fields inherited as part of  the
!:Included structure type.


17.6.2. Slot Options
17.6.2. Slot Options
17.6.2. Slot Options

  Slot-options  include  the !:Type option, which has an argument declaring
the type of the slot as a type id or list of permissible type ids.  This is
not enforced now, but anticipates the Mode system structures.

  The !:UserGet and !:UserPut  slot-options  allow  overriding  the  simple
vector  reference and assignment semantics of the generated selector macros
with user-defined functions.  The !:UserGet FNAME is a combination  of  the
slot-name  and  a !:Prefix if applicable.  The !:UserPut FNAME is the same,
with "Put" prefixed.   One  application  of  this  capability  is  building
depositors  which  handle  the  incremental  maintenance  of  parallel data
structures as a side effect, such as automatically maintaining display file
representations of objects which are resident in a remote display processor
in parallel with modifications to the LISP structures  which  describe  the
objects.    The  Make  and Create macros bypass the depositors, while Alter
uses them.


17.6.3. A Simple Example
17.6.3. A Simple Example
17.6.3. A Simple Example

  (Input lines have a "> " prompt at the beginning.)
PSL Manual                    7 February 1983                     Utilities
section 17.6                                                     page 17.19


   > % (Do definitions twice to see what functions were defined.)
   > macro procedure TWICE u; list( 'PROGN, second u, second u );
   TWICE

   > % A definition of Complex, structure with Real and Imaginary parts
   > % Redefine to see what functions were defined.  Give 0 Init values
   > TWICE
   > Defstruct( Complex( !:Creator(Complex) ), R(0), I(0) );
   *** Function `MAKECOMPLEX' has been redefined
   *** Function `ALTERCOMPLEX' has been redefined
   *** Function `COMPLEXP' has been redefined
   *** Function `COMPLEX' has been redefined
   *** Function `R' has been redefined
   *** Function `PUTR' has been redefined
   *** Function `I' has been redefined
   *** Function `PUTI' has been redefined
   *** Defstruct `COMPLEX' has been redefined
   COMPLEX


   > C0 := MakeComplex();    % Constructor with default inits.
   [COMPLEX 0 0]

   > ComplexP C0;% Predicate.
   T

   > C1:=MakeComplex( R 1, I 2 );   % Constructor with named values.
   [COMPLEX 1 2]

   > R(C1); I(C1);% Named selectors.
   1
   2

   > C2:=Complex(3,4) % Creator with positional values.
   [COMPLEX 3 4]

   > AlterComplex( C1, R(2), I(3) );     % Alterant with named values.
   [COMPLEX 2 3]

   > C1;
   [COMPLEX 2 3]

   > R(C1):=5; I(C1):=6; % Named depositors.
   5
   6

   > C1;
   [COMPLEX 5 6]

   > % Show use of Include Option.  (Again, redef to show fns defined.)
   > TWICE
Utilities                     7 February 1983                    PSL Manual
page 17.20                                                     section 17.6

   > Defstruct( MoreComplex( !:Include(Complex) ), Z(99) );
   *** Function `MAKEMORECOMPLEX' has been redefined
   *** Function `ALTERMORECOMPLEX' has been redefined
   *** Function `MORECOMPLEXP' has been redefined
   *** Function `Z' has been redefined
   *** Function `PUTZ' has been redefined
   *** Defstruct `MORECOMPLEX' has been redefined
   MORECOMPLEX


   > M0 := MakeMoreComplex();
   [MORECOMPLEX 0 0 99]

   > M1 := MakeMoreComplex( R 1, I 2, Z 3 );
   [MORECOMPLEX 1 2 3]

   > R C1;
   5

   > R M1;
   1

   > % A more complicated example: The structures which are used in the
   > % Defstruct facility to represent defstructs.  (The EX prefix has
   > % been added to the names to protect the innocent...)
   > TWICE% Redef to show fns generated.
   > Defstruct(
   >     EXDefstructDescriptor( !:Prefix(EXDsDesc), !:Creator ),
   >DsSize(!:Type int ),   % (Upper Bound of vector.)
   >Prefix(!:Type string ),
   >SlotAlist(   !:Type alist ), % (Cdrs are SlotDescriptors.)
   >ConsName(    !:Type fnId ),
   >AltrName(    !:Type fnId ),
   >PredName(    !:Type fnId ),
   >CreateName(  !:Type fnId ),
   >Include(     !:Type typeid ),
   >InclInit(    !:Type alist )
   > );
   *** Function `MAKEEXDEFSTRUCTDESCRIPTOR' has been redefined
   *** Function `ALTEREXDEFSTRUCTDESCRIPTOR' has been redefined
   *** Function `EXDEFSTRUCTDESCRIPTORP' has been redefined
   *** Function `CREATEEXDEFSTRUCTDESCRIPTOR' has been redefined
   *** Function `EXDSDESCDSSIZE' has been redefined
   *** Function `PUTEXDSDESCDSSIZE' has been redefined
   *** Function `EXDSDESCPREFIX' has been redefined
   *** Function `PUTEXDSDESCPREFIX' has been redefined
   *** Function `EXDSDESCSLOTALIST' has been redefined
   *** Function `PUTEXDSDESCSLOTALIST' has been redefined
   *** Function `EXDSDESCCONSNAME' has been redefined
   *** Function `PUTEXDSDESCCONSNAME' has been redefined
   *** Function `EXDSDESCALTRNAME' has been redefined
   *** Function `PUTEXDSDESCALTRNAME' has been redefined
PSL Manual                    7 February 1983                     Utilities
section 17.6                                                     page 17.21

   *** Function `EXDSDESCPREDNAME' has been redefined
   *** Function `PUTEXDSDESCPREDNAME' has been redefined
   *** Function `EXDSDESCCREATENAME' has been redefined
   *** Function `PUTEXDSDESCCREATENAME' has been redefined
   *** Function `EXDSDESCINCLUDE' has been redefined
   *** Function `PUTEXDSDESCINCLUDE' has been redefined
   *** Function `EXDSDESCINCLINIT' has been redefined
   *** Function `PUTEXDSDESCINCLINIT' has been redefined
   *** Defstruct `EXDEFSTRUCTDESCRIPTOR' has been redefined
   EXDEFSTRUCTDESCRIPTOR


   > TWICE% Redef to show fns generated.
   > Defstruct(
   >     EXSlotDescriptor( !:Prefix(EXSlotDesc), !:Creator ),
   >SlotNum(     !:Type int ),
   >InitForm(    !:Type form ),
   >SlotFn(!:Type fnId ), % Selector/Depositor id.
   >SlotType(    !:Type type ), % Hm...
   >UserGet(     !:Type boolean ),
   >UserPut(     !:Type boolean )
   > );
   *** Function `MAKEEXSLOTDESCRIPTOR' has been redefined
   *** Function `ALTEREXSLOTDESCRIPTOR' has been redefined
   *** Function `EXSLOTDESCRIPTORP' has been redefined
   *** Function `CREATEEXSLOTDESCRIPTOR' has been redefined
   *** Function `EXSLOTDESCSLOTNUM' has been redefined
   *** Function `PUTEXSLOTDESCSLOTNUM' has been redefined
   *** Function `EXSLOTDESCINITFORM' has been redefined
   *** Function `PUTEXSLOTDESCINITFORM' has been redefined
   *** Function `EXSLOTDESCSLOTFN' has been redefined
   *** Function `PUTEXSLOTDESCSLOTFN' has been redefined
   *** Function `EXSLOTDESCSLOTTYPE' has been redefined
   *** Function `PUTEXSLOTDESCSLOTTYPE' has been redefined
   *** Function `EXSLOTDESCUSERGET' has been redefined
   *** Function `PUTEXSLOTDESCUSERGET' has been redefined
   *** Function `EXSLOTDESCUSERPUT' has been redefined
   *** Function `PUTEXSLOTDESCUSERPUT' has been redefined
   *** Defstruct `EXSLOTDESCRIPTOR' has been redefined
   EXSLOTDESCRIPTOR


   > END;
   NIL
Utilities                     7 February 1983                    PSL Manual
page 17.22                                                     section 17.7

17.7. DefConst
17.7. DefConst
17.7. DefConst


 DefConst
 DefConst  _ __  _ ______    _________                                _____
(DefConst [U:id  V:number]): Undefined                                macro

     DefConst
     DefConst
     DefConst  is  a  simple  means  for  defining  and using symbolic
     constants, as an alternative to the heavy-handed NEWNAM or DEFINE
     facility  in  REDUCE/RLISP.     Constants   are   defined   thus:
     DefConst(FooSize, 3); or as sequential pairs:  

        DEFCONST(FOOSIZE, 3,
                 BARSIZE, 4);


 Const
 Const _ __   ______                                                  _____
(Const U:id): number                                                  macro

                                       Const
                                       Const
     They are referred to by the macro Const, so

        CONST(FOOSIZE)

     would be replaced by 3.



17.8. Functions for Sorting
17.8. Functions for Sorting
17.8. Functions for Sorting

  The  Gsort module provides functions for sorting lists and vectors.  Some
                        __________ ________
of the functions take a comparison function as an argument.  The comparison
function takes two arguments and returns NIL if they are out of order, i.e.
if the second argument should come before the first in the  sorted  result.
Lambda expressions are acceptable as comparison functions.


 Gsort
 Gsort _____  ____ ______  ___ __  __ ________     ____ ______         ____
(Gsort TABLE:{list,vector} leq-fn:{id,function}): {list,vector}        expr

                         ____      ______     ___ __
     Returns  a  sorted  list  or  vector.    LEQ-FN is the comparison
                                                                 _____
     function used to determine the sorting order.  The original TABLE
                    Gsort
                    Gsort
     is unchanged.  Gsort uses a stable sorting algorithm.   In  other
                 _                 _                            _
     words,  if  X  appears before Y in the original table then X will
                   _                           _       _
     appear before Y in the final table unless X  and  Y  are  out  of
                                                               _     _
     order.   (An unstable sort, on the other hand, might swap X and Y
                                                       _       _
     even if they're in order.  This could happen when X  and  Y  have
     the  same  "key  field",  so  either one could come first without
     making a difference to the comparison function.)


 GmergeSort
 GmergeSort _____  ____ ______  ___ __  __ ________     ____ ______    ____
(GmergeSort TABLE:{list,vector} leq-fn:{id,function}): {list,vector}   expr

                 Gsort
                 Gsort                                 _____
     The same as Gsort, but destructively modifies the TABLE argument.
     GmergeSort                                                 Gsort
     GmergeSort                                                 Gsort
     GmergeSort has the advantage of being somewhat faster than Gsort.

     Note that you should use the value  returned  by  the  function--
PSL Manual                    7 February 1983                     Utilities
section 17.8                                                     page 17.23

     don't depend on the modified argument to give the right answer.


 IdSort
 IdSort _____  ____ ______     ____ ______                             ____
(IdSort TABLE:{list,vector}): {list,vector}                            expr

                            __
     Returns  a  table  of  ids  sorted  into alphabetical order.  The
     original  table  is  unchanged.    Case  is  not  significant  in
     determining  the  alphabetical  order.    The  table  may contain
     ______             __
     strings as well as ids.

  The following example illustrates the use of Gsort.

   1 lisp> (load gsort)
   NIL
   2 lisp> (setq X '(3 8 -7 2 1 5))
   (3 8 -7 2 1 5)
   3 lisp>   % Sort from smallest to largest.
   3 lisp> (Gsort X 'leq)
   (-7 1 2 3 5 8)
   4 lisp>   % Sort from largest to smallest.
   4 lisp> (GmergeSort X 'geq)
   (8 5 3 2 1 -7)
   5 lisp>   % Note that X was "destroyed" by GmergeSort.
   5 lisp> X
   (3 2 1 -7)
   6 lisp>
   6 lisp>   % Here's IdSort, taking a vector as its argument.
   6 lisp> (IdSort '[the quick brown fox jumped over the lazy dog])
   [BROWN DOG FOX JUMPED LAZY OVER QUICK THE THE]
   7 lisp>
   7 lisp>   % Some examples of user defined comparison functions...
   7 lisp> (setq X '(("Joe" . 20000) ("Moe" . 21000) ("Larry" . 7000)))
   (("Joe" . 20000) ("Moe" . 21000) ("Larry" . 7000))
   8 lisp>
   8 lisp>   % First, sort the list alphabetically according to name,
   8 lisp>   % using a lambda expression as the comparison function.
   8 lisp> (Gsort X
   8 lisp>     '(lambda (X Y) (string-not-greaterp (car X) (car Y))))
   (("Joe" . 20000) ("Larry" . 7000) ("Moe" . 21000))
   9 lisp>
   9 lisp>   % Now, define a comparison function that compares cdrs of
   9 lisp>   % pairs, and returns T if the first is less than or equal
   9 lisp>   % to the second.
   9 lisp> (de cdr_leq (pair1 pair2)
   9 lisp>   (leq (cdr pair1) (cdr pair2)))
   CDR_LEQ
   10 lisp>
   10 lisp>   % Use the cdr_leq function to sort X.
   10 lisp> (Gsort X 'cdr_leq)
   (("Larry" . 7000) ("Joe" . 20000) ("Moe" . 21000))
Utilities                     7 February 1983                    PSL Manual
page 17.24                                                     section 17.9

17.9. Hashing Cons
17.9. Hashing Cons
17.9. Hashing Cons

                                       HCons
                                       HCons
  HCONS  is  a  loadable  module.  The HCons function creates unique dotted
                        HCons       Eq HCons                        Eq
                        HCons _  _  Eq HCons _  _                 _ Eq    _
pairs.  In other words, HCons(A, B) Eq HCons(C, D) if and only if A Eq    C
        Eq
     _  Eq  _
and  B  Eq  D.  This allows rapid tests for equality between structures, at
the cost of expending more time in creating the structures.    The  use  of
HCons
HCons
HCons  may  also save space in cases where lists share common substructure,
since only one copy of the substructure is stored.

  Hcons
  Hcons                    ____ ____ _____
  Hcons works by keeping a pair hash table of  all  pairs  that  have  been
             HCons
             HCons
created  by  HCons.  (So the space advantage of sharing substructure may be
offset by the space consumed by table  entries.)    This  hash  table  also
allows  the  system to store property lists for pairs--in the same way that
LISP has property lists for identifiers.

                   HCons                               RplacA       RplacD
                   HCons ______ ___                    RplacA       RplacD
  Pairs created by HCons should not be modified  with  RplacA  and  RplacD.
Doing  so will make the pair hash table inconsistent, as well as being very
likely to modify structure shared with something that  you  don't  wish  to
change.  Also note that large numbers may be equal without being eq, so the
HCons                                  Eq        HCons
HCons                                  Eq        HCons
HCons  of two large numbers may not be Eq to the HCons of two other numbers
that appear to be the  same.    (Similar  warnings  hold  for  strings  and
vectors.)

  The following "user" functions are provided by HCONS:


 HCons
 HCons  _ ___    ____                                                 _____
(HCons [U:any]): pair                                                 macro

          HCons
          HCons
     The  HCons  macro  takes  one or more arguments and returns their
     "hashed cons" (right associatively).   With  two  arguments  this
                              Cons
                              Cons
     corresponds to a call of Cons.


 HList
 HList  _ ___    ____                                                 _____
(HList [U:any]): list                                                 nexpr

     HList                               List
     HList                               List
     HList is the "HCONS version" of the List function.


 HCopy
 HCopy _ ___   ___                                                    _____
(HCopy U:any): any                                                    macro

     HCopy                             Copy                      HCopy
     HCopy                             Copy                      HCopy
     HCopy is the HCONS version of the Copy function.  Note that HCopy
                                           Copy
                                           Copy
     serves  a very different purpose than Copy, which is usually used
     to copy a structure so that destructive changes can  be  made  to
                                               HCopy
                                               HCopy
     the  copy without changing the original.  HCopy only copies those
                                                                Cons
                                                                Cons
     parts  of  the  structure  which  haven't  already  been  "Consed
                  HCons
                  HCons
     together" by HCons.


 HAppend
 HAppend _ ____  _ ____   ____                                         ____
(HAppend U:list  V:list): list                                         expr

         HCons            Append
         HCons            Append
     The HCons version of Append.
PSL Manual                    7 February 1983                     Utilities
section 17.9                                                     page 17.25

 HReverse
 HReverse _ ____   ____                                                ____
(HReverse U:list): list                                                expr

         HCons            Reverse
         HCons            Reverse
     The HCons version of Reverse.

                                              Get       Put
                                              Get       Put
  The following two functions can be used to "Get" and "Put" properties for
pairs  or  identifiers.    The pairs for these functions must be created by
HCons                                    SetF
HCons                                    SetF
HCons.  These functions are known to the SetF macro.


 Extended-Put
 Extended-Put _  __ ____   ___ __  ____ ___   ___                      ____
(Extended-Put U:{id,pair}  IND:id  PROP:any): any                      expr


 Extended-Get
 Extended-Get _  __ ____   ___ ___   ___                               ____
(Extended-Get U:{id,pair}  IND:any): any                               expr



17.10. Graph-to-Tree
17.10. Graph-to-Tree
17.10. Graph-to-Tree

  GRAPH-TO-TREE is a loadable module.    For  resident  functions  printing
circular lists see Section 15.8.


 Graph-to-Tree
 Graph-to-Tree _ ____   ____                                           ____
(Graph-to-Tree A:form): form                                           expr

                    Graph-to-Tree
                    Graph-to-Tree
     The  function  Graph-to-Tree  copies  an  arbitrary s-expression,
     removing circularity.   It  does  NOT  show  non-circular  shared
                                                      Eq
                                                      Eq
     structure.    Places  where  a  substructure  is Eq to one of its
     ancestors are replaced by non-interned ids of the form <n>  where
     n  is  a  small integer.  The parent is replaced by a two element
     list of the form (<n>: u) where the  n's  match,  and  u  is  the
     (de-circularized) structure.  This is most useful in adapting any
     printer for use with circular structures.


 CPrint
 CPrint _ ___   ___                                                    ____
(CPrint A:any): NIL                                                    expr

                  CPrint
                  CPrint
     The function CPrint, also defined in the module GRAPH-TO-TREE, is
             PrettyPrint  Graph-to-Tree
             PrettyPrint  Graph-to-Tree
     simply (PrettyPrint (Graph-to-Tree X)).

  Note  that  GRAPH-TO-TREE is very embryonic.  It is MUCH more inefficient
than it needs to be, heavily consing.  A better implementation would use  a
stack  (vector)  instead  of  lists  to  hold  intermediate expressions for
comparison, and  would  not  copy  non-circular  structure.    In  addition
facilities  should  be  added  for optionally showing shared structure, for
performing the inverse  operation,  and  for  also  editing  long  or  deep
structures.    Finally,  the output representation was chosen at random and
can probably be improved, or at least brought in line with CL or some other
standard.
Utilities                     7 February 1983                    PSL Manual
page 17.26                                                    section 17.11

17.11. Inspect Utility
17.11. Inspect Utility
17.11. Inspect Utility

  INSPECT is a loadable module.  


 Inspect
 Inspect ________ ______                                               ____
(Inspect FILENAME:string):                                             expr

     This  is  a  simple  utility which scans the contents of a source
     file to tell what functions are  defined  in  it.    It  will  be
     embellished  slightly  to  permit the on-line querying of certain
                           Inspect
                           Inspect
     attributes of files.  Inspect reads one or more  files,  printing
     and collecting information on defined functions.

  Usage:

   (LOAD INSPECT)
   (INSPECT "file-name") % Scans the file, and prints proc
                         % names.  It also
                         % builds the lists ProcedureList!*
                         % FileList!* and ProcFileList!*

                         % File-Name can DSKIN other files

On  the  Fly  printing is controlled by !*PrintInspect, default is T. Other
lists built include FileList!* and  ProcFileList!*,  which  is  a  list  of
(procedure . filename) for multi-file processing.

  For more complete process, do:  

   (LOAD INSPECT)
   (OFF PRINTINSPECT)
   (INSPECTOUT)
   (DSKIN ...)
   (DSKIN ...)
   (INSPECTEND)


REDUCE Historical
REDUCE Sourceforge Project | Historical SVN Repository | GitHub Mirror | SourceHut Mirror | NotABug Mirror | Chisel Mirror | Chisel RSS ]