Shareware Super Platinum 8

home *** CD-ROM | disk | FTP | other *** search

/ Shareware Super Platinum 8 / Shareware Super Platinum 8.iso / mac / PROGTOOL / GUILIB20.ZIP;1 / DOCS.ZIP / GUI.DOC < prev next >

Wrap

Text File | 1993-01-16 | 44.6 KB | 1,309 lines

**************************************************************************** Documentation for GUI_LIB Library for Borland C++ and Turbo C++ Version 2.0 (c) Copyright 1993 - David S. Reinhart Associates (717) 284-3736 **************************************************************************** All files included on this distribution diskette are (c) Copyright 1993, David S. Reinhart Associates Congratulations! You have just received one of the finest graphics libraries available for use with Borland C++ and Borland's Turbo C++ compilers. Using the GUI library will enable you to create graphic based applications to run in DOS just as easily as creating text based applications. The objects included in the library are all familiar objects, found in many commercially available software packages. They include pushbuttons, icons, text entry fields, and bitmaps to name just a few. Some of the features in this library are found nowhere else! Now you can harness the power and intuitiveness of the Graphic User Interface for use in your applications. From paint programs to databases, all of your programs can benefit from the use of graphics. If you have not already done so, I would suggest that you stop reading this documentation for a moment to check out the demo provided on the distribution diskette. The name of the demo is simply DEMO.EXE. In order to run the demo you must first have loaded your mouse driver. Any Microsoft, Logitech, or compatible mouse will be fine. From the DOS prompt type DEMO <ENTER>. The demo highlights most of the objects in the library. After viewing the demo, return to the documentation. SHAREWARE **************************************************************************** The GUI library and it's included files are being distributed as Shareware. They are complete and not crippled in any way. You may use these files for a 30 day period free of charge. You may not, however, distribute any works created with these files in any way, shape or form until you have registered your copy of the package. If after 30 days you decide to continue using the files in this package you are obligated to register with the author. The concept of Shareware helps to keep the cost of quality software low. Please do your part by registering this software. Use of the software contained in this package after 30 days without registration will bring about sever penalties. This software package is comparable with many retail libraries costing hundreds of dollars. It is through the concept of Shareware that I am able to bring this software to you at a much reduced cost. What's more, registered users will receive one upgrade free of cost rather than at an upgrade cost. The price of this package is as follows: Single user license - $35 includes free copy of the next major upgrade of the package. Corporate user license - $60 includes free copy of the next major upgrade of the package. Please use the registration form included with this package to register your copy. The form can be found in the file REGISTER.DOC. Thank you If you like this product you may also wish to order the ObjectEase library also from this author. This library makes it easy to migrate into the realm of C++ programming by offering the widest range of classes available in any library at any cost. Include a note on your registration form to receive a demo of the ObjectEase library. Use of the GUI library In order to utilize the objects and functions in the GUI library you need only add the GUI.LIB library to your project, and include the header file in your program code. You can then declare object instances and call the functions as you would with any other library. You must, of course, first initialize the graphics mode with a call to initgraph(). The best way to do this is by using one of the member functions of the Screen class, new in version 2.0. The GUI.LIB library file was compiled under the large memory model using Borland C++ 3.1 and should be included only in projects that are also to be compiled for the large memory model. This is because the use of graphics can be very memory intensive and the large memory model gives the program access to all of the available RAM, not just a single 64K data segment. This documentation is divided into sections describing each of the various graphic objects followed by a description and examples of all of the methods that may be used with the object. Please take special note of the example fragments as they may be of great help to you in getting to know the proper use of these objects and methods. I'll start by describing the objects new in version 2.0. If you are familiar with version 1.0, you should also check out the new editors included in this version, as well as the file EDITORS.DOC which describes their use. **************************************************************************** SCREEN **************************************************************************** The screen class was devised to put the hardware into any of several graphics modes and to initialize several other variables that come into play when programming for the graphics mode. I would highly recommend using this class instead of calling initgraph() directly. You should still call closegraph() at the end of your program to restore the hardware to it's original state. Here is the declaration of the Screen class from the header file: class Screen { protected: static int huge alwayszero(); public: Screen(); ~Screen(); int VGA_480_16(); int VGA_350_16(); int VGA_200_16(); int VGA_200_256(); void fill(int); }; As you can see, this class is really only useful for VGA modes, but I think that this is where 95% of all graphics programming is done anyway. Class Screen is very simple to use. You should declare an instance of this class as a global variable. Consider this short example: Screen screen; void main() { screen.VGA_200_256(); ... That's really all there is to it. The screen object will (re)init- ialize the mouse driver if one is present. This is because I have found that Logitech mice are not always capable of reaching the entire screen if they are not initialized AFTER the screen changes modes. Use the fill(int) member function with an integer paremeter re- presenting the color you would like to clear the screen with. For example: Screen screen; void main() { screen.VGA_200_256(); screen.fill(15); //clears the screen to white ... *************************************************************************** MCURSOR *************************************************************************** The Mcursor object controls and gets information about the mouse cursor. This is the only new class that will cause you to do some re- coding if you use version 2.0 of the GUI library to re-compile programs which you have written with version 1.0. Several predefined shapes for the graphics mouse cursor have been included in the GUI library, however, at this time there is no provision for including your own designs with the Mcursor class. **************************************************************************** Note: The instance "Mcursor the_mouse;" has been pre-defined in the header file. This is because many of the objects in this library make use of the Mcursor object on their own. You need to declare this same instance of the Mcursor class within your code or the mouse will not function as you expect. When you declare an Mcursor instance in your code simply declare it as external like this: extern Mcursor the_mouse; //you MUST name your instance //the_mouse **************************************************************************** int Mcursor::init() Initializes the mouse. This function must be called before trying to use any of the other mouse functions. If a mouse driver is detected the function will return 1, if no mouse driver is present it will return 0. example: extern Mcursor the_mouse; if(!the_mouse.init()) { puts("No mouse driver detected"); exit(0); } The Mcursor instance is immediately put into a state of "unarmed." You must "arm" the instance before it can be displayed. void Mcursor::show() Makes the mouse cursor visible on the screen. example: extern Mcursor the_mouse; the_mouse.init(); the_mouse.show(); void Mcursor::hide() Makes the mouse cursor invisible. example: extern Mcursor the_mouse; the_mouse.init(); getch(); the_mouse.hide(); void Mcursor::unarm() Makes the mouse cursor invisible, and keeps other object instances from turning it back on. Any object that needs to draw itself on the screen will turn off the mouse cursor while it is manipulating the screen and then turns it back on when it is finished. If you don't want an object to have the capability of restoring the mouse cursor to the screen you must call Mcursor::unarm(). example: extern Mcursor the_mouse; the_mouse.init(); the_mouse.unarm(); //hides the mouse cursor //and keeps it from //being restored by //other objects //instances. void Mcursor::arm() Restores the mouse cursor to the screen and allows other object instances to restore it to the screen when they have finished updating the screen. YOU MUST CALL THIS MEMBER FUNCTION AFTER INITIALIZING THE MOUSE BECAUSE IT IS UNARMED UPON INITIALIZATION! example: extern Mcursor the_mouse; the_mouse.init(); the_mouse.unarm(); ...do some screen manipulation the_mouse.arm(); void Mcursor::set_hor_bounds(int left,int right) Sets the horizontal bounds of the mouse cursor movement. example: extern Mcursor the_mouse; the_mouse.init(); the_mouse.set_hor_bounds(100,getmaxx()-100); // the mouse cursor now cannot move closer than //100 pixels to the sides of the screen. void Mcursor::set_ver_bounds(int top,int bottom) Sets the vertical bounds of the mouse cursor movement. example: extern Mcursor the_mouse; the_mouse.init(); the_mouse.set_ver_bounds(100,getmaxy()-100); //the mouse cursor now cannot move closer than //100 pixels to the top or bottom of the screen. void Mcursor::position(int x,int y) Moves the mouse to the specified coordinates on the screen. example: extern Mcursor the_mouse; the_mouse.init(); the_mouse.position(0,0); //the mouse cursor is now in the upper left //corner of the screen. void Mcursor::conditional_off(int x,int y,int x1,int y1) Declares a portion of the screen where the mouse cursor will not be visible. example: extern Mcursor the_mouse; the_mouse.init(); the_mouse.conditional_off(0,0,100,100); //the mouse cursor will not be visible in the //upper left corner of the screen. int Mcursor::LBP() Returns 1 if the left mouse button is pressed, 0 otherwise. example: extern Mcursor the_mouse; int done=0; the_mouse.init(); while(!done) { if(the_mouse.LBP()) done=1; } printf("\a"); int Mcursor::RBP() Returns 1 if the right mouse button is pressed, 0 otherwise. example: extern Mcursor the_mouse; int done=0; the_mouse.init(); while(!done) { if(the_mouse.RBP()) done=1; } printf("\a"); int Mcursor::mx() Returns an integer representing the x coordinate of the mouse cursor. example: extern Mcursor the_mouse; int x; the_mouse.init(); x=the_mouse.mx(); int Mcursor::my() Returns an integer representing the y coordinate of the mouse cursor. example: extern Mcursor the_mouse; int y; the_mouse.init(); y=the_mouse.my(); void Mcursor::changeto(int) Changes the shape of the graphics mouse cursor. Has no effect on the text mode mouse cursor. This function first determines the current shape of the mouse cursor. If the current shape is the same as the one you are requesting it to change to the function will exit with no action taken. example: extern Mcursor the_mouse; the_mouse.init(); the_mouse.changeto(CROSSHAIR); Note: The following cursor shapes are predefined and may be used as the parameter to the Mcursor::changeto(int) function: #define UPARROW 1 #define DOT 2 #define PENCIL 3 #define CROSSHAIR 4 #define ARROW 5 #define FINGER 6 #define POINT 7 #define CLOCK 8 #define DISK 9 #define IBAR 10 #define PAINTCAN 11 #define HAND 12 #define ERASOR 13 #define GUNSIGHT 14 #define SCISSORS 15 #define JAWS 16 These are the member functions of the Mcursor class. I'm sure you'll find them as easy to implement as they are powerful to use. In these days of the Graphical User Interface, mouse routines are becoming a standard even in text based applications. These tools out the power of the mouse in your hands. *************************************************************************** BITMAPS *************************************************************************** The Bitmap object is used to display and capture graphic images. There are two editors included in this package to create bitmap images. The files are in the Borland getimage() format, which is certainly not the most compressed format for image storage. Future versions of this library are sure to include Bitmap objects with various graphic file formats. The bitmap editors are 16IMAGE.EXE and 256IMAGE.EXE for creating 16 and 256 color bitmaps respectively. Bitmap::Bitmap() The default constructor. Bitmap::~Bitmap() The default destructor. void Bitmap::init(int x,int y) Initializes a bitmap object. The x and y parameter represent the x,y coordinates for displaying the bitmap. If this coordinate pair is not known when you initialize the bitmap instance, simply initialize it to 0,0. void Bitmap::changexy(int x,int y) Changes the x,y coordinates at which to display the bitmap. void Bitmap::load(char* name) Loads the bitmap image specified in the name parameter from disk. This file must have been created by one of the bitmap editors included with the GUI library, or by a previous call to Bitmap::save(char*). void Bitmap::save(char* name) Saves a bitmap image to disk. Under most circumstances, this will be an image that has been captured using the Bitmap::capture(int,int,int,int) member function. void Bitmap::capture(int x,int y,int x1,int y1) Captures the area bounded by the rectangle x,y,x1,y1 to a Bitmap object. This area MUST NOT exceed 64K as this is the maximum size for a Borland getimage() buffer. void Bitmap::moveto(int x,int y) Hides a displyed Bitmap object and re-displays it at the new x,y coordinates. void Bitmap::hide() Removes a displayed Bitmap object from the screen. void Bitmap::show_COPY() Displays a Bitmap image using the COPY_PUT mode. void Bitmap::show_XOR() Displays a Bitmap object using the XOR_PUT mode. void Bitmap::show_OR() Displays a Bitmap object using the OR_PUT mode. void Bitmap::show_NOT() Displays a Bitmap using the NOT_PUT mode. void Bitmap::show_AND() Displays a Bitmap object using the AND_PUT mode. int Bitmap::isshown() Returns 1 if the Bitmap object is displayed on the screen, 0 otherwise. int Bitmap::hit() Returns 1 if the mouse cursor is on the Bitmap object, 0 otherwise. int Bitmap::xsize() Returns an integer representing the horizontal size of the Bitmap. int Bitmap::ysize() Returns an integer representing the vertical size of the Bitmap. int Bitmap::bitmapx() Returns the upper left x coordinate of the Bitmap. int Bitmap::bitmapy() Returns the upper left y coordinate of the Bitmap. int Bitmap::LBSCLK() Returns 1 if the left mouse button has been pressed while the mouse cursor was on the Bitmap object, 0 otherwise int Bitmap::LBDCLK() Returns 1 if the left mouse button was pressed twice while the mouse cursor was on the Bitmap object. The button must have been pressed two times within 1/2 of a second or else the timer resets. *************************************************************************** GCHECKBOXES *************************************************************************** A Gcheckbox can be used to allow the user to make multiple selections. Call "init()" with the following three parameters to initialize a Gcheckbox. The x int coordinate, the y int coordinate, and a char * representing the descriptive text to be displayed with the Gcheckbox. Call "show()" to display the Gcheckbox on the screen. Calling "check()" and "uncheck()" will toggle the state of the Gcheckbox, and calling "is_checked()" will tell you its current state. The "hit()" member function returns true if the mouse cursor is positioned on the Gcheckbox. Follow this example: Gcheckbox mybox; int done=0; mybox.init(100,100,"This is a Gcheckbox"); mybox.show(); while(!done) { if(the_mouse.LBP()) { if(mybox.hit()) { if(mybox.is_checked()) mybox.uncheck(); else mybox.check(); } } } You don't have to use the mouse as the trigger to toggle states. Any keypress or key combination that you choose can be used to toggle the state. **************************************************************************** GRADIO **************************************************************************** Gradio is the graphics mode version of the radio button. Basically the way a radio button works is this: First, you must have more than one of them. If you only need one use a Gcheckbox instead. One of the series of Gradios is initially selected. When the user selects a different radio button, the first one becomes de-selected, and the new one becomes selected, and so on. Remember, unlike a series of Gcheckboxes, only ONE Gradio can be selected at a time. Initialize a Gradio by calling "init()" with three parameters. The first two are ints specifying the x and y coordinates respectively to display the Gradio, and the third is a char* representing the descriptive text to be displayed to the right of the Gradio. After initializing the Gradio you can show it with a call to "show()". Like a Gcheckbox you can toggle the state of a Gradio with calls to "check()" and "uncheck()", and test if it has been hit with the mouse cursor by calling "hit()". Calling "is_checked()" will tell you the current state of the Gradio. Let's look at an example: Gradio rb1,rb2; int done=0; rb1.init(100,100,"This is a radio button"); rb2.init(100,120,"So is this"); rb1.show(); rb2.show(); while(!done) { if(the_mouse.LBP()) { if(rb1.hit()) { if(rb1.is_checked()) { rb1.uncheck(); rb2.check(); } else { rb1.check(); rb2.uncheck(); } if(rb2.hit()) { if(rb2.is_checked()) { rb2.uncheck(); rb1.check(); } else { rb2.check(); rb1.uncheck(); } } } } *************************************************************************** SOUNDQ *************************************************************************** The SoundQ class is used for playing music and sounds in the background while foreground processes continue to execute. This is a feature that you will commonly find in use in game programs, but can be useful in any application. If you're like me, you have probably tried to find some way to do this before, but the necessary functions are simply not available as part of the standard C or C++ libraries. Now you have the tools... you have the POWER...the power of the OBJECT! NOTE: When you include an instance of the SoundQ class in your program the timer chip will be reset to tick 500 times per second. The normal rate for this chip is 18.2 times per second. Clock functions will continue to operate normally as the SoundQ object issues a clock tick at the normal rate of 18.2 times per second, however, you may find that you need to make adjustments in any delays issued by the "void delay(long)" function that comes with the standard library. Let's take a look at the details of the SoundQ class... SoundQ::SoundQ The default constructor. SoundQ::~SoundQ The default destructor. void SoundQ::adjust_speed(float) Adjusts the speed of the sounds being played in the background. The default value is 1.0. Increasing this value will slow down the playback, decreasing it will of course speed up the playback. Changing this value to 2.0 will effectively play back the sounds at half speed. example: SoundQ sq; sq.adjust_speed(2.0); void SoundQ::play(int freq,int duration) Places notes in the queue to be played by the SoundQ object. The notes placed in the queue will begin to play immediately. Remember that the timer tick has been reset to tick 500 times per second by the SoundQ object, therefore, to make a note sound for 1 second the duration parameter must be 500. example: SoundQ sq; sq.play(440,100); Note: The following have been defined in the header file, and make the programming of music much simpler: #define C 523 Frequency of Middle C #define CS 554 C Sharp #define D 587 D #define DS 622 D Sharp #define E 659 E #define F 698 F #define FS 740 F Sharp #define G 784 G #define GS 831 G Sharp #define A 880 A #define AS 932 A Sharp (or B Flat) #define B 988 B #define C1 1046 C one octave above middle C #define SN 32 Duration of sixteenth note #define EN 63 Eighth note #define QN 125 Quarter note #define HN 250 Half note #define WN 500 Whole note #define ENT 20 Duration of eighth note triplet #define QNT 41 Duration of Quarter note triplet #define HNT 83 Duration of Half note triplet #define BN 30000,5 Place between two notes of the same frequency to clearly define the sounding of two notes. #define SR 30000,32 Plays a sixteenth note rest #define ER 30000,63 Plays an eighth note rest #define QR 30000,125 Plays a quarter note rest #define HR 30000,250 Plays a half note rest #define WR 30000,500 Plays a whole note rest #define NM 30000,1 Stops the sounding of a tone. Use at the end of a song. Let's use these defines to play the C Major Scale with quarter notes... SoundQ sq; sq.play(C,QN); sq.play(D,QN); sq.play(E,QN); sq.play(F,QN); sq.play(G,QN); sq.play(A,QN); sq.play(B,QN); sq.play(C1,QN); sq.play(NM); //used to stop the sound These defined values can be multiplied or divided to play sounds from different octaves. Remember that once you have created a song you only need to use SoundQ::adjust_speed to vary the playback speed of the song. **************************************************************************** MISCELLANEOUS **************************************************************************** The following functions, while not part of any class structure, are included in this version of the library, and will be of great use to you. void dlay(int ticks); Causes a delay of the specified number of clock ticks. Clock ticks occur 18.2 times per second. long getticks(); Returns the time as a long number representing the number of clock ticks since the preceding midnight. int altkey(); Returns 1 if an ALT key is pressed, 0 otherwise. int ctrlkey(); Returns 1 if a CTRL key is pressed, 0 otherwise. int lshiftkey(); Returns 1 if the LEFT SHIFT key is pressed, 0 otherwise. int rshiftkey(); Returns 1 if the RIGHT SHIFT key is pressed, 0 otherwise. void flushkeys(); Clears any and all keypresses from the keyboard buffer. unsigned char getvidmode(); Returns an unsigned char value representing the current video mode. **************************************************************************** PANELS **************************************************************************** Panels can be associated with pop-up windows in a text environment. They serve to contain and draw attention to other objects such as text or pushbuttons. The class Panel is declared in the header file GROBJECT.H as: class Panel:public Point { protected: int w; WIDTH int h; HEIGHT int in_or_out; CONCAVE or CONVEX int thick_or_thin; BORDER STYLE public: Panel(); ~Panel(); virtual void show(); void init(int xpos,int ypos,int width,int height, int inorout,int thickorthin); }; The class POINT from which Panel is derived is the base class for all of the graphic objects in this library. In order to facilitate the initialization of the inorout and thickorthin variables the definitions IN, OUT, THICK and THIN may be used. These are defined in the header file. To declare an instance of class PANEL and display it in your application follow the example in the following code fragment. Panel mypanel; mypanel.init(10,10,100,100,IN,THICK); mypanel.show(); It's just that simple. Remember that the graphics mode must first be initialized by a call to initgraph. If you are unfamiliar with this function consult your compiler documentation. All of the objects in this library contain the method "init". You must call "init" with the proper parameters before trying to display the object. There's no telling what might happen if you try to display an uninitialized object. **************************************************************************** BEVELS **************************************************************************** Bevels are just like panels with a more decorative border. All calls to their methods are identical except that a call to Bevel.init() takes one less parameter. You do not need to specify IN or OUT because Bevel borders always go out. Here is the definition. I'll skip an example as you can easily follow the Panel example and make appropriate substitutions. class Bevel:public Point { private: int w; WIDTH int h; HEIGHT int thick_or_thin; STYLE OF BORDER Panel outerbevel; Panel innerbevel; public: void init(int,int,int,int,int); virtual void show(); }; **************************************************************************** PUSHBUTTONS **************************************************************************** Pushbuttons simulate keyboard keys on the screen. They serve to get user input for a variety of purposes. They are very intuitive and can display either a text title or a graphic image. The header file definition of the class Button is: class Button:public Point { protected: int state, UP or DOWN int sizex, WIDTH int sizey; HEIGHT char btntxt[40]; TEXT ID int file_text; TEXT ID or GRAPHIC IMAGE void *picture; BUFFER CONTAINING GRAPHIC IMAGE void getpic(char*); public: Button(); ~Button(); virtual void show(); virtual void press(); void init(int xpos,int ypos,char* text,int); int hit(); }; To declare an instance of class Button you need to supply the parameters for the upper left x and y positions, a string which either represents the text to be displayed on the button or the name of the file containing the image to be used on the button, and an integer flag representing whether or not the last parameter refers to text or a filename. The definitions TEXT and IMAGE can be used in this parameter and are defined in the header file. Consider this example: Button textbutton; textbutton.init(10,10," EXIT ",TEXT); textbutton.show(); This would create and display a button that said " EXIT " on it. Consider this: Button graphbutton; graphbutton.init(10,10,"EXIT",IMAGE); graphbutton.show(); This would create and display a button containing the image stored in the file "EXIT.BTN". The file extension should not be included in the parameter as ".BTN" is the assumed extension. The graphic file must be present on the disk before trying to call this function. Using the ICONEDIT program supplied with this package you can very easily create these graphics to be included on buttons. Note that you do not need to specify the width and height of buttons. If you are using text labels the button width is determined automatically to accomodate up to 40 characters, and if your are using graphics the button defaults to 20x20. This is the size of the button graphics created by the ICONEDIT program. Pressing the button is usually done with the mouse cursor, however, you may find it useful to use some other trigger to press the button. The button method "hit()" determines if the mouse cursor is within the boundaries of the button. For a detailed discussion on the use of the mouse functions see the appropriate section of this documentation. If you don't understand all of the functions in this next code fragment you can find explanations elsewhere in this document. Button mybutton; DECLARES INSTANCE OF BUTTON mybutton.init(10,10,"EXIT",TEXT); INITIALIZES MYBUTTON mybutton.show(); DISPLAYS MYBUTTON ON SCREEN while(!kbhit()) { DO THIS WHILE NO KEYS ARE HIT if(leftmousekeypressed()) { IF THE LEFT MOUSE KEY IS PRESSED if(mybutton.hit()) { IF MOUSE CURSOR IS ON MYBUTTON mybutton.press(); DISPLAY MYBUTTON AS PRESSED while(leftmousekeypressed()); WAIT UNTIL LEFT MOUSE KEY HAS BEEN RELEASED mybutton.show(); DISPLAY MYBUTTON IN NORMAL STATE } } } Immediately after restoring the button to the normal (undepressed) state the program can branch to take whatever action the button calls for. By using the Button method "press()" all of the redrawing and repositioning of the text and/or graphic images is handled automatically by the GUI_LIB library. **************************************************************************** COLORBUTTONS **************************************************************************** Colorbuttons don't do much except tell you when they are hit and return their current color. This is very useful if you are coding a paint program and need some way to set the current foreground and background colors. I'm sure you can find some other good uses for this class as well. The header definition of class Colorbutton is: class Colorbutton:public Point { protected: int color; int width; int height; public: void init(int,int,int,int,int); void show(int); int hit(); int getcolor(); }; Call "init()" with integer parameters for the x and y position, the width, the height, and the color of the button. Call "show()" with and integer value representing the color of the rectangle you wish to have surrounding the Colorbutton. By manipulating this value you can effectively show an "active" Colorbutton in a row of Colorbuttons. "Hit()" simply returns whether or not the mouse cursor is in the Colorbutton area, and "getcolor()" return the color of the Colorbutton. A quick way to make a color selection bar is as follows: Colorbutton colorarray[16]; for(int i=0;i<16;i++) { colorarray[i].init(100+(i*20),100,20,20,i); colorarray[i].show(15); } **************************************************************************** ICONS **************************************************************************** Icons can be used in much the same way as buttons. They will always contain a graphic image, although text can be a part of the graphic. They are not 3 dimensional like pushbuttons. The graphic images for these Icon objects are created using the ICONEDIT program. This program produces 32x32 pixel icons that can be single or multi-framed. More on this in a moment. The definition of the Icon class is: class Icon:public Point { protected: int state; SELECTED OR NOT public: void far *picture; BUFFER CONTAINING GRAPHIC Icon(); ~Icon(); void init(int xpos,int ypos,char* fname); void show(); void choose(); int hit(); int ispressed(); }; Like the other graphic objects, the Icon class must first be initialized with a call to "init()." Init() takes three parameter, the x and y coordinates of the upper left corner, and the file name of the file containing the graphic image for the icon. The filename parameter should be given without an extension as the default extension ".ICN" is assumed. The file named in this parameter must exist in the current directory. Let's look at a short example: Icon myicon; myicon.init(10,10,"paint"); myicon.show(); This example declares an instance of class Icon using the file "paint.icn" as the graphic, and then displays the icon. As with the pushbutton, the method "hit()" determines if the mouse cursor is within the boundaries of the icon. If it is, and if a mouse key is pressed, we can change the appearance of the icon to mark it as selected using the method "choose()." Here's a quick example, similar to that presented for the pushbutton: Icon myicon; myicon.init(10,10,"paint"); myicon.show(); while(!kbhit()) { if(leftmousekeypressed()) { if(myicon.hit()) { if(!myicon.ispressed()) { myicon.choose(); while(leftmousekeypressed()); myicon.show(); } } } } Usually when an icon has been selected you will want to make sure that the last icon to be selected gets reset to its normal state with a call to "show()." Whatever action needs to be taken as a result of the icon's selection can be done after the call to "choose()." The method "ispressed()" simply returns whether or not the icon is already displayed in its selected state. **************************************************************************** ACTICONS **************************************************************************** Acticons are exactly like Icons except that instead of simply reversing their image to mark them as the selected icon, they become animated. Acticons are created using the ICONEDIT program. They are created in the same manner as an Icon, but instead of just a single frame graphic, they contain several (up to 32) frames. Acticons are defined in the header file as follows: class Acticon:public Icon { protected: void *picture[32]; GRAPHICS BUFFERS int state; SELECTED OR NOT int numpix; NUMBER OF FRAMES public: Acticon(); ~Acticon(); void init(int,int,char*); void show(int); void choose(); int ispressed(); void animate(int); void backforth(int); }; Like Icons, they are initialized with a call to "init()" with parameters for the upper left x and y coordinates, and the file name containing the graphic images. When you call "show()", however, you must supply an integer parameter specifying which frame to display. Usually you will want to make this frame 0 because if you call "choose()" the negative image of frame 0 will be displayed by default. For example: Acticon myicon; myicon.init(10,10,"paint"); myicon.show(0); Just like an Icon except for the parameter in the call to "show()." If you want to use choose to show the reverse image of the Acticon it is done the same as it is for class Icon. If however you want to animate the Acticon you must call "animate()" or "backforth()." Both of these methods will put the Acticon in motion. "Animate()" will display the frames in ascending order and then restart at frame 0. "Backforth()" will display the frames in ascending order, but when it reaches the last frame it will then display the frames in reverse order back to frame 0. Both of these methods take a single parameter which specifies the delay between displaying each frame. This delay is expressed in eighteenths of a second. The lower the value, the faster the animation. You do not need to specify the number of frames in the Acticon as this information is contained in the graphic file created by ICONEDIT. Let's take a look: Acticon myicon; myicon.init(10,10,"paint"); myicon.show(0); while(!kbhit()) { if(leftmousekeypressed()) { if(myicon.hit()) { while(leftmousekeypressed()) myicon.animate(3); } } } In this code fragment, if the left mouse key is pressed while the cursor is on the Acticon, then the Acticon will animate for as long as the left mouse key is pressed. Note that "animate()" and "backforth()" will only change the animation by one frame, so for continuous motion you must include calls to these methods within some sort of loop. **************************************************************************** GSTRINGS **************************************************************************** No, it's not something a stripper wears, the G in Gstring simply stands for graphic. These object are used for getting text input while in graphics mode. They display their own input fields and can handle all of the text editing keys like arrows, backspace, delete, etc... Class Gstring is defined as: class Gstring:public Tstring { protected: int curpos; CURSOR POSITION IN STRING int curson; CURSOR ON OR OFF void showcurs(); void hidecurs(); public: void init(int xpos,int ypos,int length,int caps); void show(); void input(); void get_input(); void get_form_input(); int isshown(); void check_for_blink(); }; Once again a call to "init()" is necessary to begin. The necessary parameters are the x and y coordinates of the upper left corner of the input field, the length of the field, and an integer flag to indicate whether of not to force the input into all capital letters. Any nonzero value will force all caps. Notice that Gstring is not derived from class Point as are all the other object, but is derived from class Tstring. Tstring is not a part of this graphics library, but performs the same functions as Gstring only in text mode. Two methods that are part of Tstring that you might find useful are "preset()" and "reset()." Call "preset()" with a string parameter that you wish to have displayed in the input field by default, and "reset()" with no parameters to clear the input field. For example, if you wanted the input field to be displayed with the preset value of "Married" you would use code similar to the following: Gstring mystring; mystring.init(100,10,10,0); mystring.preset("Married"); mystring.show(); If you don't use the "preset()" method the input field will just be blank, probably what you want in most cases anyway. To allow user input just add the statement: mystring.get_input(); This will allow text input until the user presses either <ESCAPE> or <ENTER>. You can test for the <ESCAPE> key with the Tstring method "escapehit()." Let's put it all together... Gstring mystring; char *the_string; mystring.init(100,10,10,0); mystring.get_input(); if(mystring.escapehit()) exit(0); else strcpy(the_string,mystring.getstring()); The Tstring method "getstring()" returns a char * to the actual text string that was entered by the user. **************************************************************************** GMENUBUTTON and GMENU **************************************************************************** The classes Gmenubutton and Gmenu provide the capabilities for you to include pulldown menus in you graphics applications. The Gmenubutton is simply the text that will appear in the menu bar whereas the Gmenu is the moving bar menu that will pop up when a Gmenubutton is selected. Since these two objects are so closely related I will discuss them together. Here are the definitions: class Gmenu { protected: int on; int x,y,w,h; int num; gitemarray gitems; int menuchoice; int oldbarx,oldbary; void *ptr; void *menubar; public: Gmenu(); ~Gmenu(); void init(int xloc,int yloc,int numentries, gitemarray gitem); int show(); void hide(); int isshown(); }; class Gmenubutton { protected: int on; int x,y; int offfgd,offbgd; int onfgd,onbgd; char id[20]; public: Gmenubutton(); ~Gmenubutton(); void init(int xloc,int yloc,int ffgd,int fbgd, int nfgd,int nbgd,char txt[20]); void show(); void press(); int hit(); }; The data element gitemarray is a predefined two dimensional array containing the strings that will appear in the Gmenu. The maximum number of items in a Gmenu is 10, and the width of the pop-up menu is fixed at 100 pixels. When initializing the strings in the gitemarray always start at index 1 rather than index 0. Like this: gitemarray menu1array; strcpy(menu1array[1],"ITEM 1"); strcpy(menu1array[2],"ITEM 2"); etc... A Gmenubutton is initialized with parameters for x and y screen locations, normal state fgd and bgd colors, selected state fgd and bgd colors, and the text to be displayed. A Gmenu is initialized with parameters for the x and y screen locations, the number of entries to be in the menu, and the gitemarray containing the item strings. The Gmenu is popped-up with a call to "show()" which returns an integer representing the index of the gitemarray the bar was on when the selection was made. If no selection was made "show()" returns 11. Let's look at an example to hopefully clear this all up. gitemarray itemarray; Gmenu mymenu; Gmenubutton mymenubutton; int menuchoice; strcpy(itemarray[1],"ITEM 1"); strcpy(itemarray[2],"ITEM 2"); strcpy(itemarray[3],"ITEM 3"); strcpy(itemarray[4],"ITEM 4"); setfillstyle(SOLID_FILL,15); MAKE A MENU BAR AT TOP OF SCREEN bar(0,0,getmaxx(),10); mymenubutton.init(0,0,0,15,15,0,"MENU 1"); mymenubutton.show(); mymenu.init(0,11,4,itemarray); if(leftmousekeypressed()) { if(mymenubutton.hit()) { mymenubutton.press(); choice=mymenu.show(); mymenu.hide(); mymenubutton.show(); } } In this example the integer variable choice will contain the number of the item that was selected. Execution can then branch to the appropriate functions based on this value using case statements. ************************************************************************** PRINTING TEXT *************************************************************************** Although not part of a class library, there is a text output function included in the library that you will find most useful. The "outtextxy()" function which is part of the Borland library will only take a char* as the parameter to write to the screen. You cannot give it a variable to format into a string for you. There is a function in the GUI_LIB library called gprintxy() that will format variables into printable text for you. The definition is: gprintxy(int,int,char *fmt,...) You can use this function just as you would the "printf()" function in text mode. So for example you could write: int age=27; gprintxy(100,100,"%d",age); "gprintxy()" will format the variable and print it to the screen correctly. The two integer parameters represent the x and y position to print the string. Remember that this function is dependent on the current text justification settings. Use the Borland library function "settextjustify()" to alter these settings. Don't forget to use the registration form in the file REGISTER.DOC to register your copy of this library package. I feel that it is good quality and very useful software that is well worth the price that I am asking. Thank you for using software from David S. Reinhart Associates