Return to the Index
<==

Alphabetic listing

This chapter explains how keywords (functions and commands) are specified and used, then lists them all alphabetically. Use this chapter if you know which keyword you need to use, but need to check how to use it. Each one is listed with the specification of its usage, then a description of what it does.

Note: the example programs in this chapter do not include full error handling code. This means that the programs have been kept short and easy to understand, but may fail if, for example, you enter the wrong type of value for a variable.

If you want to develop programs from these examples, it is recommended that you add some error handling code to them. An earlier chapter covers Error handling.

Typing commands, functions and arguments

Commands, functions and arguments may be typed in any combination of UPPER and lower case.

Any commands may be strung together like this, and as many of them as you like provided the total line length does not exceed 255 characters. The colon is optional before a REM statement.

How commands are specified

Commands are specified as COMMAND argument(s) where argument(s) follow the command after a space and are separated from each other by commas. The arguments may include:

For example, AT X%,Y% might be used like this: AT 15,2

How functions are specified

Functions are specified as variable=FUNCTION(argument(s)) where variable may be f% or f& for a function returning an integer or long integer result, f for a function returning a floating-point result, or f$ for a function returning a string result. The argument(s):

Eg f$=LEFT$(g$,x%) might be used like this: PRINT LEFT$(fname$,2)

If you use the wrong type of number as an argument it will, where possible, be converted. For example, you can use an integer for a floating-point argument, or a long integer for an integer argument. If the conversion is not possible for example, if you use a floating-point number for an integer argument and its value is outside the range of integers an error will be produced and the program stopped.

Some functions, such as GET, have no arguments.


ABS

Usage: a=ABS(x)

Returns the absolute value of a floating-point number that is, without any +/- sign for example ABS(-10.099) is 10.099

If x is an integer, you won't get an error, but the result will be converted to floating-point for example ABS(-6) is 6.0. Use IABS to return the absolute value as a long integer.


ACOS

Usage: a=ACOS(x)

Returns the arc cosine, or inverse cosine (COS^-1) of x.

x must be in the range -1 to +1. The number returned will be an angle in radians. To convert the angle to degrees, use the DEG function.


ADDR

Usage: a%=ADDR(variable)

Returns the address at which variable is stored in memory.

The values of different types of variables are stored in bytes starting at ADDR(variable). See PEEK for details.

See also UADD, USUB.


ADJUSTALLOC

Usage: pcelln%=ADJUSTALLOC(pcell%,off%,am%)

Opens or closes a gap at off% within the allocated cell pcell%, returning the new cell address or zero if out of memory. off% is 0 for the first byte in the cell. Opens a gap if the amount am% is positive, and closes it if negative.


ALERT

Usage any of:
r%=ALERT(m1$,m2$,b1$,b2$,b3$)
r%=ALERT(m1$,m2$,b1$,b2$)
r%=ALERT(m1$,m2$,b1$)
r%=ALERT(m1$,m2$)
r%=ALERT(m1$)

Presents an alert a simple dialog with the messages and keys specified, and waits for a response. m1$ is the message to be displayed on the first line, and m2$ on the second line. If m2$ is not supplied or if it is a null string, the second message line is left blank.

Up to three keys may be used. b1$, b2$ and b3$ are the strings (usually words) to use over the keys. b1$ appears over an Esc key, b2$ over Enter, and b3$ over Space. This means you can have Esc, or Esc and Enter, or Esc, Enter and Space keys. If no key strings are supplied, the word CONTINUE is used above an Esc key.

The key number 1 for Esc, 2 for Enter or 3 for Space is returned.


ALLOC

Usage: pcell%=ALLOC(size%)

Allocates a cell on the heap of the specified size, returning the pointer to the cell or zero if there is not enough memory.


APP

Usage:
APP name
...
ENDA

Begins definition of an OPA. name gives the name of the OPA.

See the Advanced Topics chapter for more details of OPAs.


APPEND

Usage: APPEND

Adds a new record to the end of the current data file. The record which was current is unaffected. The new record, the last in the file, becomes the current record.

The record added is made from the current values of the field variables A.field1$, A.field2$, and so on, of the current data file. If a field has not been assigned a value, zero will be assigned to it if it is a numeric field, or a null string if it is a string field.

Example:

To overwrite the current record with new field values, use UPDATE.


APPENDSPRITE

Usage:
APPENDSPRITE time%,bit$(),dx%,dy%
APPENDSPRITE time%,bit$()

Appends a single bitmap-set to the current sprite.

time% gives the duration in tenths of seconds for the bitmap-set to be displayed before going on to the next bitmap-set in the sequence.

bit$() contains the names of bitmap files in the set, or to specify no bitmap. The array must have at least 6 elements:

bit$(1) for setting black pixels
bit$(2) for clearing black pixels
bit$(3) for inverting black pixels
bit$(4) for setting grey pixels
bit$(5) for clearing grey pixels
bit$(6) for inverting grey pixels
All the bitmaps in a single bitmap-set must be the same size or `Argument' error (-2) is raised on attempting to draw the sprite. Bitmaps in different bitmap-sets may differ in size.

dx% and dy%, if supplied, are the (x,y) offsets from the sprite position to the top-left of this bitmap-set, with positive for right and down. The default value of each is zero.


ASC

Usage: a%=ASC(a$)

Returns the character code of the first character of a$.

See the character set in the back of the User Guide for the character codes. Alternatively, use A%=%char to find the code for char eg %X for `X'.

If a$ is a null string () ASC returns the value 0.

Example A%=ASC(hello) returns 104, the code for h.


ASIN

Usage: a=ASIN(x)

Returns the arc sine, or inverse sine (SIN^-1) of x.

x must be in the range -1 to +1. The number returned will be an angle in radians. To convert the angle to degrees, use the DEG function.


AT

Usage: AT x%,y%

Positions the cursor at x% characters across the text window and y% rows down. AT 1,1 always moves to the top left corner of the window. Initially, the window is the full size of the screen, but you can change its size and position with the SCREEN command.

A common use of AT is to display strings at particular positions in the text window. For example: AT 5,2 :PRINT message.

Example:


ATAN

Usage: a=ATAN(x)

Returns the arc tangent, or inverse tanget (TAN^-1) of x.

The number returned will be an angle in radians. To convert the angle to degrees, use the DEG function.


BACK

Usage: BACK

Makes the previous record in the current data file the current record.

If the current record is the first record in the file, then the current record does not change.


BEEP

Usage: BEEP time%,pitch%

Sounds the buzzer. The beep lasts for time%/32 seconds so for a beep a second long make time%=32, etc. The maximum is 3840 (2 minutes).

The pitch (frequency) of the beep is 512/(pitch%+1) KHz.

BEEP 5,300 gives a comfortably pitched beep.

If you make time% negative, BEEP first checks whether the sound system is in use (perhaps by another OPL program), and returns if it is. Otherwise, BEEP waits until the sound system is free.

Example a scale from middle C:

(Note that the IF statement should all be on the same line).

Alternatively, sound the buzzer with this statement: PRINT CHR$(7). This beeps at a fixed pitch for a fixed length of time.


BREAK

Usage: BREAK

Makes a program performing a DO...UNTIL or WHILE...ENDWH loop exit the loop and immediately execute the line following the UNTIL or ENDWH statement.

Example: (picture)

DO
 ...
 ....
 IF a=b
  BREAK
 ENDIF
 ..
 ...
 .....
UNTIL a=b
x%=3
..


BUSY

Usage - one of:
BUSY str$,c%,delay%
BUSY str$,c%
BUSY str$
BUSY OFF

BUSY str$ displays str$ in the bottom left of the screen, until BUSY OFF is called. Use this to indicate `Busy' messages, usually when an OPL program is going to be unresponsive to keypresses for a while.

If c% is given, it controls the corner in which the message appears:

c%	corner
0	top left
1	bottom left (default)
2	top right
3	bottom right
delay% specifies a delay time (in half seconds) before the message should be shown. Use this to prevent `busy' messages from continually appearing very briefly on the screen.

Only one message can be shown at a time. The string to display can be up to 19 characters long.


CACHE

Usage - one of:
CACHE init%,max%
CACHE ON
CACHE OFF

CACHE creates a procedure cache of a specified initial number of bytes init% which may grow up to the maximum size max%. You should usually TRAP this.

Once a cache has been created, CACHE OFF prevents further cacheing, although the cache is still searched when calling subsequent procedures. CACHE ON may then be used to re-enable cacheing.


CACHEHDR

Usage: CACHEHDR addr(hdr%())

Read the current cache index header into array hdr%(), which must have at least 11 integer elements.

See the `Advanced topics' chapter for more details.


CACHEREC

Usage: CACHEREC addr(rec%()),off%

Read the cache index record at offset off% into array rec%(), which must have at least 18 integer elements.

See the `Advanced topics' chapter for more details.


CACHETIDY

Usage: CACHETIDY

Remove from the cache any procedures that have returned to their callers.


CALL

Usage: e%=CALL(s%,bx%,cx%,dx%,si%,di%)

This function enables you to make operating system calls. To use it requires extensive knowledge of the Operating System and related programming techniques. The syntax of this command is included here for completeness.

The INT number itself is the least significant byte of s%. The AH value (the subfunction number) is the most significant byte of s%. The values of the other arguments are passed to the corresponding 8086 registers. The value of the AX register is returned.


CHANGESPRITE

Usage:
CHANGESPRITE ix%,time%,bit$(),dx%,dy%
CHANGESPRITE ix%,time%,bit$()

Changes the bitmap-set specified by ix% (1 for the first bitmap-set) in the current sprite, using the supplied bitmap files, offsets and duration in the same way as for APPENDSPRITE.


CHR$

Usage: a$=CHR$(x%)

Returns the character with character code x%.

You can use it to display characters not easily available from the keyboard for example, the instruction PRINT CHR$(174) displays .

The full character set is in the User Guide.


CLOSE

Usage: CLOSE

Closes the current file (that is, the one which has been OPENed and most recently USEd).

If you've used ERASE to remove some records, CLOSE recovers the memory used by the deleted records, provided it is held either in the internal memory or on a Ram SSD.


CLOSESPRITE

Usage: CLOSESPRITE id%

Closes the sprite with ID id%.


CLS

Usage: CLS

Clears the contents of the text window.

The cursor then goes to the beginning of the top line. If you have used CURSOR OFF the cursor is still positioned there, but is not displayed.


CMD$

Usage: c$=CMD$(x%)

Returns the command-line arguments passed when starting a program. Null strings may be returned. x% should be from 1 to 5. cmd$(2) to cmd$(5) are only for OPAs (OPL applications).

cmd$(1) returns the full path name used to start the running program.

cmd$(2) returns the full path name of the file to be used by an OPA application.

cmd$(3) returns C for Create file or O for Open file. If the OPA is being run with a new filename, this will return C. This happens the very first time the OPA is used, and whenever a new filename is used to run it. Otherwise, the OPA is being run with the name of an existing file, and cmd$(3) will return O.

cmd$(4) returns the alias information, if any. In practice this has no relevance for OPAs.

cmd$(5) returns the application name, as declared with the APP keyword.

See the Advanced Topics chapter for more details of OPAs.

See also GETCMD$.


COMPRESS

Usage: COMPRESS src$,dest$

Copies data file src$ to another data file dest$. If dest$ already exists, the records in src$ are appended to the end of dest$.

Deleted records are not copied. This makes COMPRESS particularly useful when copying from a Flash SSD. (The space used by deleted records on a Ram SSD or in internal memory is automatically freed when you close the file.)

If you want src$ to overwrite instead of append to dest$, use: TRAP DELETE dest$ before the COMPRESS statement.

You can use wildcards if you wish to copy more than one file at a time. But if the first name contains any wildcards, the second name must not include a filename, just the device and directory to which the files are to be copied under their original names.

Example: to copy all the data files on A: (in \OPD, the default directory) to B:\BCK\:

COMPRESS A:*.ODB,B:\BCK\

(Remember the final backslash on the directory name.)

See COPY for copying any type of file.


CONTINUE

Usage: CONTINUE

Makes a program immediately go to the UNTIL... line of a DO...UNTIL loop or the WHILE... line of a WHILE...ENDWH loop ie to the test condition.

Example: (picture)

DO
 ...
 .....
 ....
 IF a<3.5
  CONTINUE
 ENDIF
 ..
 ....
 ...
UNTIL a=b
..
See also BREAK.


COPY

Usage: COPY src$,dest$

Copies the file src$, which may be of any type, to the file dest$. Any existing file with the name dest$ is deleted. You can copy across devices.

Use the appropriate file extensions to indicate the type of file, and wildcards if you wish to copy more than one file at a time:

Example: To copy all the OPL files from internal memory (in \OPL) to B:\ME\:

COPY M:\OPL\*.OPL,B:\ME\

(Remember the final backslash on the directory name.)

See COMPRESS for more control over copying data files. If you use COPY to copy a data file, deleted records are copied and you cannot append to another data file.

There are more details of full file specifications in the Advanced Topics chapter.


COS

Usage: c=COS(x)

Returns the cosine of x, where x is an angle in radians.

To convert from degrees to radians, use the RAD function.


COUNT

Usage: c%=COUNT

Returns the number of records in the current data file.

This number will be 0 if the file is empty.


CREATE

Usage: CREATE file$,log,f1,f2,...

Creates a data file called file$.

Immediately after the CREATE statement, the file is open and can be accessed.

Example:

CREATE CLIENTS,B,NM$,PHON$

would create a data file in the internal memory with the name CLIENTS and the logical name B.


CREATESPRITE

Usage: id%=CREATESPRITE

Creates a sprite, returning the sprite ID.


CURSOR

Usage - one of:
CURSOR ON
CURSOR OFF
CURSOR id%,asc%,w%,h%
CURSOR id%,asc%,w%,h%,type%
CURSOR id%

CURSOR ON switches the text cursor on at the current cursor position. Initially, no cursor is displayed.

You can switch on a graphics cursor in a window by following CURSOR with the ID of the window. This replaces any text cursor. At the same time, you can also specify the cursor's shape, and its position relative to the baseline of text.

asc% is the ascent the number of pixels (-128 to 127) by which the top of the cursor should be above the baseline of the current font. h% and w% (both from 0 to 255) are the cursor's height and width.

If you do not specify them, the following default values are used: asc% =font ascent h% =font height w% =2

If type% is given, it can have these effects: 1 obloid 2 not flashing 4 grey

You can add these values together to combine effects eg if type% is 6 a grey non-flashing cursor is drawn.

An error is raised if id% specifies a bitmap rather than a window.

CURSOR OFF switches off any cursor.


DATETOSECS

Usage: s&=DATETOSECS(yr%,mo%,dy%,hr%,mn%,sc%)

Returns the number of seconds since 00:00 on 1 January 1970 at the date/time specified.

Raises an error for dates before 1 January 1970.

The value returned is an unsigned long integer. (Values up to +2,147,483,647, which is 03:14:07 on 19/1/2038, are returned as expected. Those from +2,147,483,648 upwards are returned as negative numbers, starting from -2,147,483,648 and increasing towards zero.)

See also SECSTODATE, HOUR, MINUTE, SECOND.


DATIM$

Usage: d$=DATIM$

Returns the current date and time from the system clock as a string for example: Fri 16 Oct 1992 16:25:30 The string returned always has this format 3 mixed-case characters for the day, then a space, then 2 digits for the day of the month, and so on.


DAY

Usage: d%=DAY

Returns the current day of the month (1 to 31) from the system clock.


DAYNAME$

Usage: d$=DAYNAME$(x%)

Converts x%,a number from 1 to 7, to the day of the week, expressed as a three letter string.

Eg d$=DAYNAME$(1) returns MON.

Example:

See also DOW.


DAYS

Usage: d&=DAYS(day%,month%,year%)

Returns the number of days since 01/01/1900.

Use this to find out the number of days between two dates.

Example:

See also dDATE, SECSTODATE.


dBUTTONS

Usage - one of:
dBUTTONS p1$,k1%,p2$,k2%,p3$,k3%
dBUTTONS p1$,k1%,p2$,k2%
dBUTTONS p1$,k1%

Defines exit keys to go at the bottom of a dialog.

From one to three exit keys may be defined. Each pair of p$ and k% specifies an exit key; p$ is the text to be displayed above it, while k% is the keycode of the key. DIALOG returns the keycode of the key pressed (in lower case for letters).

For alphabetic keys, use the % sign %A means `the code of A', and so on. An appendix lists the codes for keys (such as Tab) which are not part of the character set. If you use the code for one of these keys, its name (eg `Tab', or `Enter') will be shown in the key.

If you use a negative value for a k% argument, that key is a `Cancel' key. The corresponding positive value is used for the key to display and the value for DIALOG to return, but if you do press this key to exit, the var variables used in the commands like dEDIT, dTIME etc. will not be set.

The Esc key will always cancel a dialog box, with DIALOG returning 0. If you want to show the Esc key as one of the exit keys, use -27 as the k% argument (its keycode is 27) so that the var variables will not be set if Esc is pressed.

There can be only one dBUTTONS item per dialog, and it takes up three lines on the screen. dBUTTONS must be the last dialog command you use before DIALOG itself.

Some keypresses, such as those using the Control key, cannot be specified.

This example presents a simple query, returning `True' for Yes, or `False' for No.


dCHOICE

Usage: dCHOICE var choice%,p$,list$

Defines a choice list to go in a dialog.

p$ will be displayed on the left side of the line. list$ should contain the possible choices, separated by commas for example, Yes,No. One of these will be displayed on the right side of the line, and LEFT RIGHT can be used to move between the choices.

choice% must be a LOCAL or a GLOBAL variable. It specifies which choice should initially be shown 1 for the first choice, 2 for the second, and so on. When you finish using the dialog, choice% is given a value indicating which choice was selected again, 1 for the first choice, and so on.


dDATE

Usage: dDATE var lg&,p$,min&,max&

Defines an edit box for a date, to go in a dialog.

p$ will be displayed on the left side of the line.

lg&, which must be a LOCAL or a GLOBAL variable, specifies the date to be shown initially. Although it will appear on the screen like a normal date, for example 15/03/92, lg& must be specified as days since 1/1/1900.

min& and max& give the minimum and maximum values which are to be allowed. Again, these are in days since 1/1/1900. An error is raised if min& is higher than max&.

When you finish using the dialog, the date you entered is returned in lg&, in days since 1/1/1900.

The system setting determines whether years, months or days are displayed first.

See also DAYS, SECSTODATE.


dEDIT

Usage:
dEDIT var str$,p$,len%
dEDIT var str$,p$

Defines a string edit box, to go in a dialog.

p$ will be displayed on the left side of the line.

str$ is the string variable to edit. Its initial contents will appear in the dialog. The length used when str$ was defined is the maximum length you can type in.

len%, if supplied, gives the width of the edit box (allowing for widest possible character in the font). The string will scroll inside the edit box, if necessary. If len% is not supplied, the edit box is made wide enough for the maximum width str$ could possibly be.

See also dTEXT.


DEFAULTWIN

Usage: DEFAULTWIN mode%

Change the default window (ID=1) to enable or disable the use of grey. Initially grey cannot be used in that window.

mode%=1 enables the use of grey. mode%=0 disables the use of grey.

A side-effect of DEFAULTWIN is to clear the default window.

Using grey does use more memory than using black only.

You are advised to call DEFAULTWIN once and for all near the start of your program if you need to use grey. If it fails with `Out of memory' error, the program can then exit cleanly without losing vital information.

See also gGREY and gCREATE.


DEG

Usage: d=DEG(x)

Converts from radians to degrees.

Returns x, an angle in radians, as a number of degrees. The formula used is: 180*x/PI

All the trigonometric functions (SIN,COS etc.) work in radians, not degrees. You can use DEG to convert an angle returned by a trigonometric function back to degrees:

Example:

To convert from degrees to radians, use RAD.


DELETE

Usage: DELETE filename$

Deletes any type of file.

You can use wildcards for example, to delete all the OPL files in B:\OPL DELETE B:\OPL\*.OPL

The file type extensions are listed in the User Guide.

See also RMDIR.


dFILE

Usage: dFILE var str$,p$,f%

Defines a filename edit box, to go in a dialog. A `Disk' selector is automatically added on the line below.

p$ will be displayed on the left side of the line.

f% controls the type of file editor, and the kind of input allowed. You can add together any of the following values:

value	meaning
1	use an edit box
2	allow directory names
4	directory names only
8	disallow existing files
16	query existing files
32	allow null string input
64	don't display extension
128	obey/allow wildcards

The first of the list is the most crucial. If you add 1 into f%, you will see an file edit box, as when creating a new file. If you do not add 1, you will see the `matching file' selector, used when choosing an existing file.

If performing a `copy to' operation, you might use 1+2+16, to specify a file edit box, in which you can type the name of a directory to copy to, and which will produce a query if you type the name of an existing file.

If asking for the name of a directory to remove, you might use 4, to allow an existing directory name only.

`Query existing' is ignored if `disallow existing' is set. These two, as well as `allow null string input', only work with file edit boxes, not `matching file' selectors.

str$ is the string variable to edit. Its initial contents always control the initial drive and directory used. For a file edit box, any filename part of str$ is shown. For a `matching file' selector, you can use wildcards in the filename part (such as *.tmp) to control which filenames are matched. To do this, you must add 128 to f%. 128 also allows wildcard specifications to be entered (returned in str$), for both `matching' and `new file' selectors.

If str$ does not contain any drive or directory information, the path as set by SETPATH is used. If SETPATH has not been used, the \OPD directory on the default drive (usually M:, `Internal') is used.

With a matching file selector (as opposed to an edit box) the value 8 restricts the selection to files which match the filename/extension in str$. Matching file selectors can also use 64, in which case files with the same extension as that in str$ are shown without this extension. (Many Series 3a file selectors are like this.)

You can always press Tab to produce the full file selector with a dFILE item.

str$ must be declared to be at least 128 bytes long, or an error will be raised.


dFLOAT

Usage: dFLOAT var fp,p$,min,max

Defines an edit box for a floating-point number, to go in a dialog.

p$ will be displayed on the left side of the line.

min and max give the minimum and maximum values which are to be allowed. An error is raised if min is higher than max.

fp must be a LOCAL or a GLOBAL variable. It specifies the value to be shown initially. When you finish using the dialog, the value you entered is returned in fp.


DIALOG

Usage: n%=DIALOG

Presents the dialog prepared by dINIT and commands such as dTEXT and dCHOICE. If you complete the dialog by pressing Enter, your settings are stored in the variables specified in dTEXT, dCHOICE etc., although you can prevent this with dBUTTONS.

If you used dBUTTONS when preparing the dialog, the keycode which ended the dialog is returned. Otherwise, DIALOG returns the line number of the item which was current when Enter was pressed. The top item (or the title line, if present), has line number 1.

If you cancel the dialog by pressing Esc, the variables are not changed, and 0 is returned.


DIAMINIT

Usage: DIAMINIT pos%,str1$,str2$...

Initialises the DIAMOND list (discarding any existing list). str1$, str2$ etc. contain the text to be displayed in the status window for each item in the list.

pos% is the initial item on to which the DIAMOND indicator should be positioned, with pos%=1 specifying the first item. (Any value greater than the number of strings specifies the final item.) If pos%>=1 you must supply at least this many strings.

If pos% is not supplied or if pos%=0, or if DIAMINIT is used on its own with no arguments, no bar is defined.

If pos%=-1 the list is replaced by the icon instead in the large status window.


DIAMPOS

Usage: DIAMPOS pos%

Positions the DIAMOND indicator on the DIAMOND list.

Positioning outside the range of the items wraps around in the appropriate way. pos%=0 causes the DIAMOND symbol to disappear.


dINIT

Usage: dINIT title$ or dINIT

Prepares for definition of a dialog, cancelling any existing one. Use dTEXT, dCHOICE etc. to define each item in the dialog, then DIALOG to display the dialog.

If title$ is supplied it will be displayed at the top of the dialog, centred and with a line across the dialog below it.


DIR$

Usage:
d$=DIR$(filespec$) then
d$=DIR$()

Lists filenames, including subdirectory names, matching a file specification. You can include wildcards in the file specification. If filespec$ is just a directory name, include the final backslash on the end for example, M:\TEMP\ . Use the function like this:

Example, listing all the .DBF files in M:\DAT:


dLONG

Usage: dLONG var lg&,p$,min&,max&

Defines an edit box for a long integer, to go in a dialog.

p$ will be displayed on the left side of the line.

min& and max& give the minimum and maximum values which are to be allowed. An error is raised if min& is higher than max&.

lg& must be a LOCAL or a GLOBAL variable. It specifies the value to be shown initially. When you finish using the dialog, the value you entered is returned in lg&.


DO...UNTIL

Usage:
DO
statement
statement
.
.
.
UNTIL condition

DO forces the set of statements which follow it to execute repeatedly until the condition specified by UNTIL is met.

This is the easiest way to repeat an operation a certain number of times.


DOW

Usage: d%=DOW(day%,month%,year%)

Returns the day of the week from 1 (Monday) to 7 (Sunday) given the date.

day% must be between 1 and 31, month% from 1 to 12 and year% from 1900 to 2155.

For example, D%=DOW(4,7,1992) returns 6, meaning Saturday.


dPOSITION

Usage: dPOSITION x%,y%

Positions a dialog. Use dPOSITION at any time between dINIT and DIALOG.

dPOSITION uses two integer values. The first specifies the horizontal position, and the second, the vertical. dPOSITION -1,-1 positions to the top left of the screen; dPOSITION 1,1 to the bottom right; dPOSITION 0,0 to the centre, the usual position for dialogs.

dPOSITION 1,0, for example, positions to the right-hand edge of the screen, and centres the dialog half way up the screen.


DRAWSPRITE

Usage: DRAWSPRITE x%,y%

Draws the current sprite in the current window with top-left at pixel position x%,y%.


dTEXT

Usage: dTEXT p$,body$,t% or dTEXT p$,body$

Defines a line of text to be displayed in a dialog.

p$ will be displayed on the left side of the line, and body$ on the right side. If you only want to display a single string, use a null string () for p$, and pass the desired string in body$. It will then have the whole width of the dialog to itself. An error is raised if body$ is a null string.

body$ is normally displayed left aligned (although usually in the right column). You can override this by specifying t%:

t%	effect
0	left align body$
1	right align body$
2	centre body$
In addition, you can add any or all of the following three values to t%, for these effects:
t%	effect
$100	use bold text for body$
$200	draw a line below this
	item
$400	(allow this item to
	be selected)
Only one line can be drawn across a dialog. It will be below the last item which asks for it, whether the title from dINIT or a dTEXT item.

See also dEDIT.


dTIME

Usage: dTIME var lg&,p$,t%,min&,max&

Defines an edit box for a time, to go in a dialog.

p$ will be displayed on the left side of the line.

lg&, which must be a LOCAL or a GLOBAL variable, specifies the time to be shown initially. Although it will appear on the screen like a normal time, for example 18:27, lg& must be specified as seconds after 00:00. A value of 60 means one minute past midnight; 3600 means one o'clock, and so on.

min& and max& give the minimum and maximum values which are to be allowed. Again, these are in seconds after 00:00. An error is raised if min& is higher than max&.

When you finish using the dialog, the time you entered is returned in lg&, in seconds after 00:00.

t% specifies the type of display required, as follows:

t% time display 0 absolute time no seconds 1 absolute time with seconds 2 duration no seconds 3 duration with seconds

For example, 03:45 represents an absolute time while 3 hours 45 minutes represents a duration.

Absolute times are displayed in 24-hour or am/pm format according to the current system setting.


dXINPUT

Usage: dXINPUT var str$,p$

Defines a secret string edit box, such as for a password, to go in a dialog.

p$ will be displayed on the left side of the line.

str$ is the string variable to take the string you type. Important: str$ must be at least eight characters long.

Initially the dialog does not show any characters for the string; the initial contents of str$ are ignored. A special symbol will be displayed for each character you type, to preserve the secrecy of the string.


EDIT

Usage: EDIT a$

Displays a string variable which you can edit directly on the screen. All the usual editing keys are available the arrow keys move along the line, Esc clears the line, and so on.

When you have finished editing, press Enter to confirm the changes. If you press Enter before you have made any changes, then the string will be unaltered.

If you use EDIT in conjunction with a PRINT statement, use a comma at the end of the PRINT statement, so that the string to be edited appears on the same line as the displayed string:

... PRINT Edit address:, EDIT A.address$ UPDATE ....

TRAP EDIT If the Esc key is pressed while no text is on the input line, the `Escape key pressed' error (number -114) will be returned by ERR provided that the EDIT has been trapped. You can use this feature to enable someone to press the Esc key to escape from inputting a string.

See also INPUT, dEDIT.

ELSE(IF)/ENDA/ENDIF/ENDV/ ENDWH

See IF/APP/VECTOR/WHILE

ENTERSEND

Usage:

ret%=ENTERSEND(pobj%,m%,var p1,...)

This is the same as SEND except that, if the method leaves, the error code is returned to the caller. Otherwise the value returned is as returned by the method.


ENTERSEND0

Usage: ret%=ENTERSEND0(pobj%,m%,var p1,...)

This is the same as ENTERSEND except that, if the method does not leave, zero is returned.


EOF

Usage: e%=EOF

Finds out whether you're at the end of a file yet.

Returns -1 (true) if the end of the file has been reached, or 0 (false) if it hasn't.

When reading records from a file, you should test whether there are still records left to read, otherwise you may get an error.

Example:


ERASE

Usage: ERASE

Erases the current record in the current file.

The next record is then current. If the erased record was the last record in a file, then following this command the current record will be null and EOF will return true.


ERR

Usage: e%=ERR

Returns the number of the last error which occurred, or 0 if there has been no error.

Example:

..
PRINT Enter age in years
age::
TRAP INPUT age%
IF ERR=-1
 PRINT Number please:
 GOTO age
ENDIF
...
See also ERR$.

See the Error handling chapter for full details, including the list of error numbers and messages.


ERR$

Usage: e$=ERR$(x%)

Returns the error message for the specified error code x%.

ERR$(ERR) gives the message for the last error which occurred. Example:

TRAP OPEN B:\FILE,A,field1$
IF ERR
 PRINT ERR$(ERR)
 RETURN
ENDIF
See also ERR.

See the Error handling chapter for full details, including the list of error numbers and messages.


ESCAPE OFF

Usage: ESCAPE OFF...ESCAPE ON

ESCAPE OFF stops Psion-Esc being used to break out of the program when it is running. ESCAPE ON enables this feature again.

ESCAPE OFF takes effect only in the procedure in which it occurs, and in any sub-procedures that are called. Psion-Esc is always enabled when a program begins running.

If your program enters a loop which has no logical exit, and ESCAPE OFF has been used, you will have to return to the System screen, move to the program name under the RunOpl icon, and press the Delete key.


EVAL

Usage: d=EVAL(s$)

Evaluates the mathematical string expression s$ and returns the floating-point result. s$ may include any mathematical function or operator (eg. sin(x)/(2**3)). Note that floating-point arithmetic is always performed.

For example:

DO 
AT 10,5 :PRINT Calc:,
TRAP INPUT n$
IF n$= :CONTINUE :ENDIF
IF ERR=-114 :BREAK :ENDIF
CLS :AT 10,4
PRINT n$;=;EVAL(n$)
UNTIL 0
See also VAL.


EXIST

Usage: e%=EXIST(filename$)

Checks to see that a file exists.

Returns -1 (`True') if the file exists and 0 (`False') if it doesn't.

Use this function when creating a file to check that a file of the same name does not already exist, or when opening a file to check that it has already been created:

IF NOT EXIST(CLIENTS)
 CREATE CLIENTS,A,names$
ELSE
 OPEN CLIENTS,A,names$
ENDIF
...


EXP

Usage: e=EXP(x)

Returns ex that is, the value of the arithmetic constant e (2.71828...) raised to the power of x.


EXT

Usage: EXT name$

Gives the file extension of files used by an OPA.

This can only be used between APP and ENDA.

See the Advanced Topics chapter for more details of OPAs.


FIND

Usage: f%=FIND(a$)

Searches the current data file for fields matching a$. The search starts from the current record, so use NEXT to progress to subsequent records. FIND makes the next record containing a$ the current record and returns the number of the record found. Capitals and lower-case letters match.

You can use wildcards:
? 	matches any single character
* 	matches any group of characters.
To find a record with a field containing Dr and either BROWN or BRAUN, use: F%=FIND(*DR*BR??N*)

FIND(BROWN) will find only those records with a field consisting solely of the string BROWN.

You can only search string fields.

See also FINDFIELD.


FINDFIELD

Usage: f%=FINDFIELD(a$,start%,no%,flags%)

FINDFIELD, like FIND, finds a string, makes the record with this string the current record, and returns the number of this record.

You may experience some problems in using FINDFIELD with some versions of OPL. To ensure that problems are avoided use the line:

POKEB(peekw($1c)+7),0

immediately before each call to FINDFIELD.

a$ is the string to look for, as for FIND. start% is the number of the field to start looking in (1 for the first field), and no% is the number of fields to search in (starting from the field specified by the start%). If you want to search in all fields, use start%=1 and for no% use the number of fields you used in the OPEN/CREATE command.

flags% adds together two values:


FINDLIB

Usage: ret%=FINDLIB(var cat%, name$)

Find DYL category name$ (including .DYL extension) in the ROM. On success returns zero and writes the category handle to cat%.

FIRST

Usage:

FIRST

Positions to the first record in the current data file.


FIX$

Usage: f$=FIX$(x,y%,z%)

Returns a string representation of the number x, to y% decimal places. The string will be up to z% characters long.

Example: FIX$(123.456,2,7) returns 123.46.

See also GEN$, NUM$, SCI$.


FLT

Usage: f=FLT(x&)

Converts an integer expression (either integer or long integer) into a floating-point number. Example:

You could call this procedure like this: gamma:(FLT(a%)) if you wanted to pass it the value of an integer variable without having first to assign the integer value to a floating-point variable.

See also INT and INTF.


FONT

Usage: FONT id%,style%

Sets the text window font and style.

See `The text and graphics windows' at the end of the `Graphics' chapter for more details.


FREEALLOC

Usage: FREEALLOC pcell%

Frees a previously allocated cell at pcell%.


gAT

Usage: gAT x%,y%

Sets the current position using absolute co-ordinates. gAT 0,0 moves to the top left of the current drawable.

See also gMOVE.


gBORDER

Usage:
gBORDER flags%,width%,height%
gBORDER flags%

Draws a one-pixel wide border around the edge of the current drawable. If width% and height% are supplied, a border shape of this size is drawn with the top left corner at the current position. If they are not supplied, the border is drawn around the whole of the current drawable.

flags% controls three attributes of the border a shadow to the right and beneath, a one-pixel gap all around, and the type of corners used:

flags%		effect
1		single pixel shadow
2		gap for single pixel 
		shadow
3		double pixel shadow
4		gap for double pixel 
		shadow
$100		one-pixel gap all round
$200		more rounded corners
You can combine the values to control the three different effects. (1, 2, 3 and 4 are mutually exclusive you cannot use more than one of them.) For example, for rounded corners and a gap for a double pixel shadow, use flags%=$204.

Set flags%=0 for no shadow, no gap, and sharper corners.

For example, to de-emphasise a previously emphasised border, use gBORDER with the shadow turned off:

gBORDER 3 REM show border

GET

gBORDER 4 REM border off

...

See also gXBORDER.


gBOX

Usage: gBOX width%,height%

Draws a box from the current position, width% to the right and height% down. The current position is unaffected.


gBUTTON

Usage: gBUTTON text$,type%,w%,h%,st%

Draws a 3-D black and grey button (a key, not an application button) at the current position in a rectangle of the supplied width w% and height h%, which fully encloses the button in all its states. text$ specifies up to 64 characters to be drawn in the button in the current font and style. You must ensure that the text will fit in the button.

type%=0 it draws a Series 3 style button; type%=1 draws a Series 3a button.

The meaning of state% varies according to type%:


gCLOCK

Usage - one of:
gCLOCK ON/OFF
gCLOCK ON,mode%
gCLOCK ON,mode%,offset%
gCLOCK ON,mode%,offset%,format$
gCLOCK ON,mode%,offset%,format$,font%
gCLOCK ON,mode%,offset%,format$,font%,style%

Displays or removes a clock showing the system time. The current position in the current window is used. Only one clock may be displayed in each window.

mode% controls the type of clock. Modes 1 to 5 are provided for Series 3 compatibility, and produce Series 3 clocks: small (digital), medium (system setting), medium (analog), medium (digital), and large (analog) respectively. Other values are:

6  black and grey medium, system setting
7  black and grey medium, analog
8  second type medium, digital
9  black and grey extra large 
10 formatted digital (described below)

You can OR the value with any of these:

$10 shows the date in all except the extra large and formatted clocks

$20 shows seconds in small digital, large analog, black and grey medium analog and extra large clocks

$40 shows am/pm in small digital and black medium clocks only. $80 specifies that a clock is to be drawn in the grey plane (only for clocks that do not contain both black and grey: ie. all except the black and grey, medium, analog clock and the extra large clock).

format$, font% and style% are used only for formatted digital clocks as described below.

Note : Do not use gSCROLL to scroll the region containing a clock. When the time is updated, the old position would be used. The whole window may, however, be moved using gSETWIN.
Note : It is possible to draw clocks that include grey in windows that have no grey plane.
Digital clocks display in 24-hour or 12-hour mode according to the system-wide setting in Time. The `am/pm' flag ($40) can be used with digital clocks in 12-hour mode, and with medium analog clocks.

offset% specifies an offset in minutes from the system time to the time displayed. This allows you to display a clock showing a time other than the system time.

If these arguments are not supplied, mode% is taken as 1, and offset% as 0.

For the formatted digital clock (mode%=10), you may optionally specify font% and style% with values as for gFONT and gSTYLE. The default font for gCLOCK is the system font (value $9a). The default style is normal (0).

For the formatted digital clock, a format string (up to 255 characters long) specifies how the clock is to be displayed. The format string contains a number of format specifiers in the form of a % followed by a letter. (Upper or lower case may be used.) For example, %H means hours and %T means minutes; gCLOCK ON,10,0,h:%H, m:%T, at 11:05 pm, displays a running clock as h:23, m:05.

To make each item as abbreviated as possible, you can use a * after the %. For example, %*T at 11:05 pm abbreviates `05' to `5'. In the following list of specifiers, those which produce numbers will do so without any leading zero, if you use %* instead of %. Other abbreviations are marked, appropriately, by Abbrev:

%% =a % character

%:, %/ =time and date separators, as set in the Time application

%A =`am' or `pm' text (Abbrev: 1st letter)

%D,%W,%M =day/week/month number as two digits, 01-31, 01-53 and 01-12

%E,%N =day/month name (Abbrev: shorter form, eg 1st 3 characters in English)

%H,%I =hour in 24-hour or 12-hour format, 00-23 and 01-12

%S,%T =seconds/minutes, 00-59

%X =suffix string for day number, eg st in `1st', nd in `2nd'

%Y =year as a four digit number (Abbrev: discards the century)

%1,%2,%3 =day,month,year as ordered in the Time application. Eg Europe is Day/month/year, so %1=%D, %2=%M, %3=%Y. So to display a date in correct format use %1/%2/%3. (Abbrev: see %G/%P/%U.)

%4,%5 =day,month as ordered in the Time application.

%F,%O =toggles days/months (displayed by %1, %2 and %3) between numeric and name formats. On 9th March 1993, with European date type, %1%F%1%F%1 gives 09Tuesday09

%G,%P, %U =toggles %1, %2 and %3 between long form and abbreviation. On 9th March 1993, with European date type, %F%1%G%1%G%1 gives TuesdayTueTuesday

%L =toggles the suffix on the day number for %1/%2/%3 (in numeric form only). On 9th March 1993, with European date type, %G%1%L%1%L%1 gives 99th9

%6,%7 =hour and am/pm text according to the format selected in the Time application. With am-pm format, %6=%I and %7=%A. With 24-hour format, %6=%H and %7 gives no `am/pm' characters.

So the format string %1%/%2%/%3 automatically generates a clock with day, month and year in the order as selected in the Time application. %4%/%5 gives a clock with just day and month in selected order. Similarly, %6%:%T%:%S%7 gives a clock with hour, minute and second automatically conforming to the system configuration.

Note : Note that for those specifiers that toggle between two different options (eg. %F), the state of toggle is remembered only within one format string and not from one string to the next ie the toggle state is restored to the default setting when displaying a new clock.
As a final example, assuming that the settings in the Time application are for `Day/month/year' date format, `am-pm' time format and `:' time separator and that the time is 11:30:05 pm on 9th March 1993, %G%L%P%O%*E, %1 %2 %3 %6%:%T:%S%

generates Tue, 9th Mar 1993 11:30:05pm. With the same setup except for `Month/day/year' date format in `24-hour' mode, the same string generates Tue, Mar 9th 1993 23:30:05.


gCLOSE

Usage: gCLOSE id%

Closes the specified drawable that was previously opened by gCREATE, gCREATEBIT or gLOADBIT.

If the drawable closed was the current drawable, the default window (ID=1) becomes current.

An error is raised if you try to close the default window.


gCLS

Usage: gCLS

Clears the whole of the current drawable and sets the current position to 0,0, its top left corner.


gCOPY

Usage: gCOPY id%,x%,y%,w%,h%,mode%

Copies a rectangle of the specified size (width w%, height h%) from the point x%,y% in drawable id%, to the current position in the current drawable.

As this command can copy both set and clear pixels, the same modes are available as when displaying text. Set mode% = 0 for set, 1 for clear, 2 for invert or 3 for replace. 0, 1 and 2 act only on set pixels in the pattern; 3 copies the entire rectangle, with set and clear pixels.

The current position is not affected in either window.

gCOPY is affected by the setting of gGREY (in the current window) as follows: with gGREY 0 it copies black to black; with gGREY 1 it copies grey to grey, or black to grey if source is black only; with gGREY 2 it copies grey to grey and black to black, or black to both if source is black only.


gCREATE

Usage:
id%=gCREATE(x%,y%,w%,h%,v%)
id%=gCREATE(x%,y%,w%,h%,v%,grey%)

Creates a window with specified position and size (width w%, height h%), and makes it both current and foreground. Sets the current position to 0,0, its top left corner. If v% is 1, the window will immediately be visible; if 0, it will be invisible.

If grey% is not given or is 0, the window will not have a grey plane. If grey% is 1, it will have one.

Returns id% (2 to 8) which identifies this window for other keywords.

See also gCLOSE, gGREY, DEFAULTWIN.


gCREATEBIT

Usage: id%=gCREATEBIT(w%,h%)

Creates a bitmap with the specified width and height, and makes it the current drawable. Sets the current position to 0,0, its top left corner.

Returns id% (2 to 8) which identifies this bitmap for other keywords.

See also gCLOSE.


gDRAWOBJECT

Usage: gDRAWOBJECT type%,flags%,w%,h%

Draws the scaleable graphics object specified by type%, scaled to fit in the rectangle with top left at the current graphics cursor position and with the specified width w% and height h%.

The Series 3a has only one object type (set type%=0) a `lozenge'. This is a 3-D rounded box lit from the top left, with a shadow at bottom right and a grey body. (For an example, see the text 'City' in the top left of the World application.)

For type%=0, flags% specifies the corner roundness: 0 for normal roundness 1 for more rounded 2 for a single pixel removed from each corner.

An error is raised if the current window has no grey plane.


GEN$

Usage: g$=gen$(x,y%)

Returns a string representation of the number x. The string will be up to y% characters long.

Example GEN$(123.456,7) returns 123.456 and GEN$(243,5) returns 243

See also FIX$, NUM$, SCI$.


GET

Usage: g%=GET

Waits for a key to be pressed and returns the character code for that key.

For example, if the `A' key is pressed with Caps Lock off, the integer returned is 97 (a), or 65 (A) if `A' was pressed with the Shift key down.

The character codes of special keys, such as `Pg Dn', are given in an appendix at the back of this manual.

You can use KMOD to check whether modifier keys (Shift, Control, Psion and Caps Lock) were used.

See the User Guide for the full character set.

See also KEY.


GET$

Usage: g$=GET$

Waits until a key is pressed and then returns which key was pressed, as a string.

For example, if the `A' key is pressed in lower case mode, the string returned is a.

You can use KMOD to check whether any modifier keys (Shift, Control, Psion and Caps Lock) were used.

See also KEY$.


GETCMD$

Usage: w$=GETCMD$

Returns new command-line arguments to an OPA, after a change files or quit event has occurred. The first character of the returned string is C, O or X. If it is C or O, the rest of the string is a filename.

The first character has the following meaning:

C close down the current file, and create the specified new file.

O close down the current file, and open the specified existing file.

X close down the current file (if any) and quit the OPA.

You can only call GETCMD$ once for each system message.

See the Advanced Topics chapter for more details of OPAs.

See also CMD$.


GETEVENT

Usage: *** GETEVENT var a%()

Waits for an event to occur. Returns with a%() specifying the event. The data returned in a%() depends on the type of event that occurred. If the event is a key-press, (a%(1) AND $400) is guaranteed to be zero. For other events (a%(1) AND $400) is guaranteed to be non-zero.

If a key has been pressed:

a%(1)= keycode (as for GET)

a%(2) AND $00ff= modifier (as for KMOD)

a%(2)/256= auto-repeat count (ignored by GET)

If a program has moved to foreground: a%(1)= $401

If a program has moved to background: a%(1)= $402

If the machine has switched on:a%(1)= $403

If the Series 3a wants an OPA to change files or exit:a%(1)= $404

If the date changes:a%(1)= $405

Note: events are ignored while you are using keywords which wait for keypresses GET, GET$, EDIT, INPUT, MENU and DIALOG. If you need to use these keywords in OPAs, use LOCK ON / LOCK OFF around them.

If you do use GETEVENT you should allow for other events to be specified in the future.

For a key-press event, the modifier is returned in a%(2) and is not returned by KMOD.

Note: If a non-key event such as `foreground' occurs while a keyboard keyword such as GET, INPUT, MENU or DIALOG is being used, the event is discarded. So GETEVENT must be used if non-key events are to be monitored. (OPAs can still handle the $404 event correctly see the LOCK command for more details.)

The array (or string of integers) must be at least 6 integers long.

See also TESTEVENT, GETCMD$.


GETLIBH

Usage: cat%=GETLIBH(num%)

Convert a category number num% to a handle. If num% is zero, gets the handle for OPL.DYL.


gFILL

Usage: gFILL width%,height%,gMode%

Fills a rectangle of the specified size from the current position, according to the graphics mode specified.

The current position is unaffected.


gFONT

Usage: gFONT fontId%

Sets the font for current drawable to fontId%. The font may be one of the predefined fonts in the ROM or a user-defined font. See the `Graphics' chapter for more details of fonts.

User-defined fonts must first be loaded by gLOADFONT, which returns the fontId% needed for gFONT.

See also gLOADFONT, FONT.


gGMODE

Usage: gGMODE mode%

Sets the effect of all subsequent drawing commands gLINEBY, gBOX etc. on the current drawable.

mode%		Pixels will be:
0		set
1		cleared
2		inverted
When you first use drawing commands on a drawable, they set pixels in the drawable. Use gGMODE to change this. For example, if you have drawn a black background, you can draw a white box outline inside it with either gGMODE 1 or gGMODE 2, followed by gBOX.


gGREY

Usage: gGREY mode%

Controls whether all subsequent graphics drawing and graphics text in the current window draw to the grey plane, the black plane or to both.

mode%=0 for black plane only (default)

mode%=1 for grey plane only

mode%=2 for both planes

It is helpful to think of the black plane being in front of the grey plane, so a pixel set in both planes will appear black. See the `Graphics' chapter for details.

To enable the use of grey in the default window (ID=1) use DEFAULTWIN 1 at the start of your program. If grey is required in other windows you must create the windows with a grey plane using gCREATE.

gGREY cannot be used with bitmaps which have only one plane.

See also DEFAULTWIN and gCREATE.


gHEIGHT

Usage: height% = gHEIGHT

Returns the height of the current drawable.


gIDENTITY

Usage: id%=gIDENTITY

Returns the ID of the current drawable.

The default window has ID=1.


gINFO

Usage: gINFO var i%()

Gets general information about the current drawable and about the graphics cursor (whichever window it is in). The information is returned in the array i%() which must be at least 32 integers long.

The information is about the drawable in its current state, so eg the font information is for the current font in the current style.

The following information is returned:

i%(1)		lowest character code
i%(2)		highest character code
i%(3)		height of font
i%(4)		descent of font
i%(5)		ascent of font
i%(6)		width of '0' character
i%(7)		maximum character width
i%(8)		flags for font (see below)
i%(9-17)	name of font
i%(18)		current graphics mode(gGMODE)
i%(19)		current text mode (gTMODE)
i%(20)		current style (gSTYLE)
i%(21)		cursor state (ON=1,OFF=0)
i%(22)		ID of window containingcursor (-1 for text cursor)
i%(23)		cursor width
i%(24)		cursor height
i%(25)		cursor ascent
i%(26)		cursor x position in window
i%(27)		cursor y position in window
i%(28)		1 if drawable is a bitmap
i%(29)		cursor effects
i%(30)		gGREY setting
i%(31)		reserved (window server ID of drawable)
i%(32)		reserved

i%(8) specifies a combination of the following font characteristics:

Value: Meaning: 1 font uses standard ASCII characters (32-126) 2 font uses Code Page 850 characters (128-255) 4 font is bold 8 font is italic 16 font is serifed 32 font is monospaced $8000 font is stored expanded for quick drawing

(See HEX$ for an explanation of hexadecimal numbers.)

Use PEEK$(ADDR(i%(9))) to read the name of the font as a string.

If the cursor is on (i%(21)=1), it is visible in the window identified by i%(22).

i%(29) has bit 0 set (i%(29) AND 1) if the cursor is obloid, bit 1 set (i%(29) AND 2) if not flashing, and bit 2 set (i%(29) AND 4) if grey.

If the cursor is off (i%(21)=0), or is a text cursor (i%(22)=-1), i%(23) to i%(27) and i%(29) should be ignored.


gINVERT

Usage: gINVERT width%,height%

Inverts the rectangle width% to the right and height% down from the cursor position, except for the four corner pixels.


GIPRINT

Usage: GIPRINT str$,c% or GIPRINT str$

GIPRINT displays an information message for about two seconds, in the bottom right corner of the screen. For example, GIPRINT Not Found displays Not Found. The string you specify can be up to 64 characters. If a string is too long for the screen, it will be clipped.

If c% is given, it controls the corner in which the message appears:

c%	corner
0	top left
1	bottom left
2	top right
3	bottom right (default)

Only one message can be shown at a time. You can make the message go away for example, if a key has been pressed with GIPRINT .


gLINEBY

Usage: gLINEBY dx%,dy%

Draws a line from the current position to a point dx% to the right and dy% down. Negative dx% and dy% mean left and up respectively.

For horizontal lines, the line includes the pixel with the lower x coordinate and excludes the pixel with the higher x coordinate. Similarly for vertical lines, the line includes the pixel with the lower y coordinate and excludes the pixel with the higher y coordinate. For oblique lines (where the x and y coordinates change), the line is drawn minus one or both end points.

The current position moves to the end of the line drawn.

gLINEBY 0,0 sets the pixel at the current position.

See also gLINETO, gPOLY.


gLINETO

Usage: gLINETO x%,y%

Draws a line from the current position to the point x%,y%. The current position moves to x%,y%.

For horizontal lines, the line includes the pixel with the lower x coordinate and excludes the pixel with the higher x coordinate. Similarly for vertical lines, the line includes the pixel with the lower y coordinate and excludes the pixel with the higher y coordinate. For oblique lines (where the x and y coordinates change), the line is drawn minus one or both end points.

To plot a single point, use gLINETO to the current position (or gLINEBY 0,0).

See also gLINEBY, gPOLY.


gLOADBIT

Usage - one of:
id%=gLOADBIT(name$,write%,i%)
id%=gLOADBIT(name$,write%)
id%=gLOADBIT(name$)

Loads a bitmap from the named bitmap file and makes it the current drawable. Sets the current position to 0,0, its top left corner. If name$ has no file extension .PIC is used. The bitmap is kept as a local copy in memory.

Returns id% (2 to 8) which identifies this bitmap for other keywords.

write%=0 sets read-only access. Attempts to write to the bitmap in memory will be ignored, but the bitmap can be used by other programs without using more memory. write%=1 allows you to write to and re-save the bitmap. This is the default case.

For bitmap files which contain more than one bitmap, i% specifies which one to load. For the first bitmap, use i%=0. This is also the default value. Bitmap files saved with gSAVEBIT have only one bitmap, and this argument is not needed for them.

See also gCLOSE.


gLOADFONT

Usage: fontId%=gLOADFONT(name$)

Loads the user-defined font name$. It returns a font ID; use this with gFONT to make the current drawable use this font. If name$ does not contain a file extension, .FON is used.

gFONT itself is very efficient, so you should normally load all required fonts at the start of a program.

Note that the built-in Series 3/Series 3a fonts are automatically available, and do not need loading.

See also gUNLOADFONT.


GLOBAL

Usage: GLOBAL variables

Declares variables to be used in the current procedure (as does the LOCAL command) and (unlike LOCAL) in any procedures called by the current procedure, or procedures called by them.

The variables may be of 4 types, depending on the symbol they end with:

Array variables have a number immediately following them in brackets which specifies the number of elements in the array. Array variables may be of any type, for example: x(6),y%(5),f$(5,12),z&(3)

When declaring string arrays, you must give two numbers in the brackets. The first declares the number of elements, the second declares their maximum length. For example surname$(5,8) declares five elements, each up to 8 characters long.

Variable names may be any combination of up to 8 numbers and alphabetic letters. They must start with a letter. The length includes the % & or $ sign, but not the () in string and array variables.

More than one GLOBAL or LOCAL statement may be used, but they must be on separate lines, immediately after the procedure name.

See also LOCAL.


gMOVE

Usage: gMOVE dx%,dy%

Moves the current position dx% to the right and dy% downwards, in the current drawable.

A negative dx% causes movement to the left; a negative dy% causes upward movement.

See also gAT.


gORDER

Usage: gORDER id%,position%

Sets the window specified by id% to the selected foreground/background position, and redraws the screen. Position 1 is the foreground window, position 2 is next, and so on. Any position greater than the number of windows is interpreted as the end of the list.

On creation, a window is at position 1 in the list.

Raises an error if id% is a bitmap.

See also gRANK.


gORIGINX

Usage: x%=gORIGINX

Returns the gap between the left side of the screen and the left side of the current window.

Raises an error if the current drawable is a bitmap.


gORIGINY

Usage: y%=gORIGINY

Returns the gap between the top of the screen and the top of the current window.

Raises an error if the current drawable is a bitmap.


GOTO

Usage:
GOTO label or GOTO label::
..
..
label::

Goes to the line following the label:: and continues from there. The label


gPATT

Usage: gPATT id%,width%,height%,mode%

Fills a rectangle of the specified size from the current position with repetitions of the drawable id%.

As with gCOPY, this command can copy both set and clear pixels, so the same modes are available as when displaying text. Set mode% = 0 for set, 1 for clear, 2 for invert or 3 for replace. 0, 1 and 2 act only on set pixels in the pattern; 3 copies the entire rectangle, with set and clear pixels.

If you set id%=-1 a pre-defined grey pattern is used.

The current position is unaffected.

gPATT is affected by the setting of gGREY (in the current window) in the same way as gCOPY: with gGREY 0 it copies black to black; with gGREY 1 it copies grey to grey, or black to grey if source is black only; with gGREY 2 it copies grey to grey and black to black, or black to both if source is black only.


gPEEKLINE

Usage: gPEEKLINE id%,x%,y%,d%(),ln%

Reads a horizontal line from the black plane of the drawable id%, length ln%, starting at x%,y%. The leftmost 16 pixels are read into d%(1), with the first pixel read into the least significant bit.

The array d%() must be long enough to hold the data. You can work out the number of integers required with ((ln%+15)/16) (using whole-number division).

If you set id% to 0, this just reads from the whole screen, not from any particular window.

If you add $8000 to id%, the grey plane (not the black plane) will be peeked.


gPOLY

Usage: gPOLY a%()

Draws a sequence of lines, as if by gLINEBY and gMOVE commands.

The array is set up as follows:

a%(1)	starting x position
a%(2)	starting y position
a%(3)	number of pairs of offsets
a%(4)	dx1%
a%(5)	dy1%
a%(6)	dx2%
a%(7)	dy2% etc.
Each pair of numbers dx1%,dy1%, for example specifies a line or a movement. To draw a line, dy% is the amount to move down, while dx% is the amount to move to the right multiplied by two.

To specify a movement (ie without drawing a line) work out the dx%,dy% as for a line, then add 1 to dx%.

(For drawing/movement up or left, use negative numbers.)

gPOLY is quicker than combinations of gAT, gLINEBY and gMOVE.

Example, to draw three horizontal lines 50 pixels long at positions 20,10, 20,30 and 20,50:

a%(1)=20 :a%(2)=10 REM 20,10
a%(3)=5 REM 5 operations
REM draw right 50
a%(4)=50*2   :a%(5)=0
REM move down 20
a%(6)=0*2+1  :a%(7)=20
REM draw left 50
a%(8)=-50*2  :a%(9)=0
REM move down 20
a%(10)=0*2+1 :a%(11)=20
REM draw right 50
a%(12)=50*2  :a%(13)=0
gPOLY a%()


gPRINT

Usage: gPRINT list of expressions

Displays a list of expressions at the current position in the current drawable. All variable types are formatted as for PRINT.

Unlike PRINT, gPRINT does not end by moving to a new line. A comma between expressions is still displayed as a space, but a semi-colon has no effect. gPRINT without a list of expressions does nothing.

See also gPRINTB, gPRINTCLIP, gTWIDTH, gXPRINT, gTMODE.


gPRINTB

Usage any of:
gPRINTB t$,w%,al%,tp%,bt%,m%
gPRINTB t$,w%,al%,tp%,bt%
gPRINTB t$,w%,al%,tp%
gPRINTB t$,w%,al%
gPRINTB t$,w%

Displays text t$ in a cleared box of width w% pixels. The current position is used for the left side of the box and for the baseline of the text.

al% controls the alignment of the text in the box 1 for right aligned, 2 for left aligned, or 3 for centred.

tp% and bt% are the clearances between the text and the top/bottom of the box. Together with the current font size, they control the height of the box. An error is raised if tp% plus the font ascent is greater than 255.

m% controls the margins. For left alignment, m% is an offset from the left of the box to the start of the text. For right alignment, m% is an offset from the right of the box to the end of the text. For centering, m% is an offset from the left or right of the box to the region in which to centre, with positive m% meaning left and negative meaning right.

If values are not supplied for some arguments, these defaults are used:

al% left
tp% 0
bt% 0
m% 0

See also gPRINT, gPRINTCLIP, gTWIDTH, gXPRINT.


gPRINTCLIP

Usage: w%=gPRINTCLIP(text$,width%)

Displays text$ at the current position, displaying only as many characters as will fit inside width% pixels. Returns the number of characters displayed.

See also gPRINT, gPRINTB, gTWIDTH, gXPRINT, gTMODE.


gRANK

Usage: rank%=gRANK

Returns the foreground/background position, from 1 to 8, of the current window.

Raises an error if the current drawable is a bitmap.

See also gORDER.


gSAVEBIT

Usage:
gSAVEBIT name$,width%,height%
gSAVEBIT name$

Saves the current drawable as the named bitmap file. If width% and height% are given, then only the rectangle of that size from the current position is copied. If name$ has no file extension .PIC is used.

Saving a window to file when it includes grey will save both planes to the file black bitmap first followed by grey.


gSCROLL

Usage: gSCROLL dx%,dy%,x%,y%,wd%,ht% or gSCROLL dx%,dy%

Scrolls pixels in the current drawable by offset dx%,dy%. Positive dx% means to the right, and positive dy% means down. The drawable itself does not change its position.

If you specify a rectangle in the current drawable, at x%,y% and of size wd%,ht%, only this rectangle is scrolled.

The areas dx% wide and dy% deep which are left behind by the scroll are cleared.

The current position is not affected.


gSETWIN

Usage:
gSETWIN x%,y%,width%,height%
gSETWIN x%,y%

Changes position and, optionally, the size of the current window.

An error is raised if the current drawable is a bitmap.

The current position is unaffected.

If you use this command on the default window, you must also use the SCREEN command to ensure that the area for PRINT commands to use is wholly contained within the default window.


gSTYLE

Usage: gSTYLE style%

Sets the style of text displayed in subsequent gPRINT, gPRINTB and gPRINTCLIP commands on the current drawable.

style%		Text style:
0		normal
1		bold
2		underlined
4		inverse
8		double height
16		mono
32		italic
You can combine these styles by adding their values for example, to set bold, underlined and double height, use gSTYLE 11, as 11=1+2+8.

This command does not affect non-graphics commands, like PRINT.


gTMODE

Usage: gTMODE mode%

Sets the way characters are displayed by subsequent gPRINT and gPRINTCLIP commands on the current drawable.

mode%		Pixels will be:
0		set
1		cleared
2		inverted
3		replaced
When you first use graphics text commands on a drawable, each dot in a letter causes a pixel to be set in the drawable. This is mode%=0.

When mode% is 1 or 2, graphics text commands work in a similar way, but the pixels are cleared or inverted. When mode% is 3, entire character boxes are drawn on the screen pixels are set in the letter and cleared in the background box.

This command does not affect other text display commands.


gTWIDTH

Usage: width%=gTWIDTH(text$)

Returns the width of text$ in the current font and style.

See also gPRINT, gPRINTB, gPRINTCLIP, gXPRINT.


gUNLOADFONT

Usage: gUNLOADFONT fontId%

Unloads a user-defined font that was previously loaded using gLOADFONT. Raises an error if the font has not been loaded.

The built-in Series 3/Series 3a fonts are not held in memory and cannot be unloaded.

See also gLOADFONT.


gUPDATE

Usage one of:
gUPDATE ON
gUPDATE OFF
gUPDATE

The Series 3a screen is usually updated whenever you display anything on it. gUPDATE OFF switches off this feature. The screen will then be updated as few times as possible (though note that some keywords will always cause an update.) You can still force an update by using the gUPDATE command on its own.

This can result in a considerable speed improvement in some cases. You might, for example, use gUPDATE OFF, then a sequence of graphics commands, followed by gUPDATE. You should certainly use gUPDATE OFF if you are about to write exclusively to bitmaps.

gUPDATE ON returns to normal screen updating.

gUPDATE affects anything that displays on the screen. If you are using a lot of PRINT commands, gUPDATE OFF may make a noticeable difference in speed.

Note that with gUPDATE OFF, the location of errors which occur while the procedure is running may be incorrectly reported. For this reason, gUPDATE OFF is best used in the final stages of program development, and even then you may have to remove it to locate some errors.


gUSE

Usage: gUSE id%

Makes the drawable id% current. Graphics drawing commands will now go to this drawable. gUSE does not bring a drawable to the foreground (see gORDER).


gVISIBLE ON/OFF

Usage: gVISIBLE ON/OFF

Makes the current window visible or invisible.

Raises an error if the current drawable is a bitmap.


gWIDTH

Usage: width%=gWIDTH

Returns the width of the current drawable.


gX

Usage: x%=gX

Returns the x current position (in from the left) in the current drawable.


gXBORDER

Usage:
gXBORDER type%,flags%,w%,h%
gXBORDER type%,flags%

Draws a border in the current drawable of a specified type, fitting inside a rectangle of the specified size or with the size of the current drawable if no size is specified.

type%=0 for drawing the Series 3 type border; flags% are then as for gBORDER.

type%=1 for drawing the Series 3a 3-D grey and black border. A shadow or a gap for a shadow is always assumed. flags%=1,2,3,4 are as for gBORDER. When the shadow is enabled (1 or 3) only the grey and black parts of the border are drawn; you should pre-clear the background for the white parts. When the shadow is disabled (2 or 4) the outer and inner border lines are drawn, but the areas covered by grey/black when the shadow is enabled are now cleared. (This allows a shadow to be turned off simply by calling gXBORDER again.)

The following values of flags% apply to all border types:

0 for normal corners
OR with $100 leaves 1 pixel gap around the border.
OR with $200 for more rounded corners
OR with $400 for losing a single pixel.
If both $400 and $200 are mistakenly supplied, $200 has priority.

An error is raised if the current window has no grey plane.

See also gBORDER.


gXPRINT

Usage: gXPRINT string$,flags%

Displays string$ at the current position, with precise highlighting or underlining. The current font and style are still used, even if the style itself is inverse or underlined. Text mode 3 (replace) is used both set and cleared pixels in the text are drawn.

flags% has the following effect:

flags%		effect
0		normal, as with gPRINT
1		inverse
2		inverse, except corner pixels
3		thin inverse
4		thin inverse, except corner pixels
5		underlined
6		thin underlined
Where lines of text are separated by a single pixel, the thin options maintain the separation between lines.

gXPRINT does not support the display of a list of expressions of various types.


gY

Usage: y%=gY

Returns the y current position (down from the top) in the current drawable.


HEX$

Usage: h$=HEX$(x&)

Returns a string containing the hexadecimal (base 16) representation of integer or long integer x&.

For example HEX$(255) returns the string FF.

Notes To enter integer hexadecimal constants (16 bit) put a $ in front of them. For example $FF is 255 in decimal. (Don't confuse this use of $ with string variable names.)

To enter long integer hexadecimal constants (32 bit) put a & in front of them. For example &FFFFF is 1048575 in decimal.

Counting in hexadecimal is like this: 0 1 2 3 4 5 6 7 8 9 A B C D E F 10... A stands for decimal 10, B for decimal 11, C for decimal 12 ... up to F for decimal 15. After F comes 10, which is equivalent to decimal 16. To understand numbers greater than hexadecimal 10, again compare hexadecimals with decimals. In these examples, 10^2 means 10x10, 10^3 means 10x10x10 and so on.

253 in decimal is:
(2x10^2)+(5x10^1)+(3x10^0)
= (2x100)+(5x10)+(3x1)
= 200+50+3

By analogy, &253 in hexadecimal is: (&2x16^2)+(&5x16^1)+(&3x16^0) =(2x256)+(5x16)+(3x1) =512+80+3 = 595 in decimal.

Similarly, &A6B in hexadecimal is: (&Ax16^2)+(&6x16^1)+(&Bx16^0) =(10x256)+(6x16)+(11x1) =2560+96+11 = 2667 in decimal.

You may also find this table useful for converting between hex and decimal:

hex decimal &1 1 =16^0 &10 16 =16^1 &100 256 =16^2 &1000 4096 =16^3

For example, &20F9 is (2x&1000)+(0x&100)+(15x&10)+9
which in decimal is: (2x4096)+(0x256)+(15x16)+9 = 8441.

All hexadecimal constants are integers ($) or long integers (&). So arithmetic operations involving hexadecimal numbers behave in the usual way. For example, &3/&2 returns 1, &3/2.0 returns 1.5, 3/$2 returns 1.


HOUR

Usage: h%=HOUR

Returns the number of the current hour from the system clock as an integer between 0 and 23.


IABS

Usage: i&=IABS(x&)

Returns the absolute value, ie without any sign, of the integer or long integer expression x&.

For example IABS(-10) is 10.

See also ABS, which returns the absolute value as a floating-point value.


ICON

Usage: ICON name$

Gives the name of the bitmap file to use as the icon for an OPA.

This can only be used between APP and ENDA.

See the Advanced Topics chapter for more details of OPAs.


IF...ENDIF

Usage:
IF condition1
..
...
.
ELSEIF condition2
...
....
ELSE
..
ENDIF

Does either

or

or

After the ENDIF statement, the lines following ENDIF carry on as normal.

IF, ELSEIF, ELSE and ENDIF must be in that order.

Every IF must be matched with a closing ENDIF.

You can also have an IF...ENDIF structure within another, for example:

IF condition1
 ..
 ...
 .
ELSE
 ...
 IF condition2
 ....
 ENDIF
 ..
ENDIF
condition is an expression returning a logical value for example a


INPUT

Usage:
INPUT variable
INPUT log.field

Waits for a value to be entered at the keyboard, and then assigns the value entered to a variable or data file field.

You can edit the value as you type it in. All the usual editing keys are available the arrow keys move along the line, Esc clears the line and so on.

If inappropriate input is entered, for example a string when the input was to be assigned to an integer variable, a ? is displayed and you can try again. However, if you used TRAP INPUT, control passes on to the next line of the procedure, with the appropriate error condition being set and the value of the variable remaining unchanged.

INPUT is usually used in conjunction with a PRINT statement:

Note the commas at the end of the PRINT statements, used so that the cursor waiting for input appears on the same line as the messages.

TRAP INPUT
If a bad value is entered (for example abc for a%) in response to a TRAP INPUT, the ? is not displayed, but the ERR function can be called to return the value of the error which has occurred. If the Esc key is pressed while no text is on the input line, the `Escape key pressed' error (number -114) will be returned by ERR (provided that the INPUT has been trapped). You can use this feature to enable someone to press the Esc key to escape from inputting a value.

See also EDIT. This works like INPUT, except that it displays a string to be edited and then assigned to a variable or field. It can only be used with strings.


INT

Usage: i&=INT(x)

Returns the integer (in other words the whole number) part of the floating-point expression x. The number is returned as a long integer.

Positive numbers are rounded down, and negative numbers are rounded up for example INT(-5.9) returns -5 and INT(2.9) returns 2. If you want to round a number to the nearest integer, add 0.5 to it (or subtract 0.5 if it is negative) before you use INT.

In the Calculator, you need to use INT to pass a number to a procedure which requires a long integer parameter. This is because the Calculator passes all numbers as floating-point by default.

See also INTF.


INTF

Usage: i=INTF(x)

Used in the same way as the INT function, but the value returned is a floating-point number. For example, INTF(1234567890123.4) returns 1234567890123.0

You may also need this when an integer calculation may exceed integer range.

See also INT.

I/O functions

These functions are covered in detail in the Advanced Topics chapter.

r%=IOA(h%,f%,var status%,var a1,var a2)

The device driver opened with handle h% performs the asynchronous I/O function f% with two further arguments, a1 and a2. The argument status% is set by the device driver. An IOWAIT must be performed for each IOA.

r%=IOC(h%,f%,var status%,var a1,var a2)

Make an I/O request with guaranteed completion. This has the same form as IOA etc but it returns zero always.

r%=IOCANCEL(h%)

Cancels any outstanding asynchronous I/O request (IOC or IOA).

r%=IOCLOSE(h%)

Closes a file with the handle h%.

r%=IOOPEN(var h%,name$,mode%)

Creates or opens a file called name$. Defines h% for use by other I/O functions. mode% specifies how to open the file. For unique file creation, use IOOPEN(var h%,addr%,mode%)

r%=IOREAD(h%,addr%,maxLen%)

Reads from the file with the handle h%. address% is the address of a buffer large enough to hold a maximum of maxLen% bytes. The value returned to r% is the actual number of bytes read or, if negative, is an error value.

r%=IOSEEK(h%,mode%,var off&)

Seeks to a position in a file that has been opened for random access. mode% specifies how the offset argument off& is to be used.

IOSIGNAL

Signals an I/O function's completion.

r%=IOW(h%,func%,var a1,var a2)

The device driver opened with handle h% performs the synchronous I/O function func% with the two further arguments.


IOWAIT

Waits for an asynchronous I/O function to signal completion.

IOWAITSTAT var stat%

Waits for an asynchronous function, called with IOA, to complete.

r%=IOWRITE(h%,addr%,length%)

Writes length% bytes in a buffer at address% to the file with the handle h%.


IOYIELD

Ensures that any asynchronous handler set up with IOA


KEY

Usage: k%=KEY

Returns the character code of the last key pressed, if there has been one since the last call to the keyboard. These functions count as calling to the keyboard: INPUT, EDIT, GET, GET$, KEY and KEY$.

If no key has been pressed, zero is returned.

See Appendix D for a list of special key codes. You can use KMOD to check whether modifier keys (Shift, Control, Psion and Caps Lock) were used.

This command does not wait for a key to be pressed, unlike GET.


KEY$

Usage: k$=KEY$

Returns the last key pressed as a string, if there has been a keypress since the last use of the keyboard by INPUT, EDIT, GET, GET$, KEY or KEY$.

If no key has been pressed, a null string () is returned.

See Appendix D for a list of special key codes. You can use KMOD to check whether modifier keys (Shift, Control, Psion and Caps Lock) were used.

This command does not wait for a key to be pressed, unlike GET$.


KEYA

Usage: err%=KEYA(var stat%,var key%(1))

This is an asynchronous keyboard read function.

See the Advanced Topics chapter for details.

Cancel with KEYC.


KEYC

Usage: err%=KEYC(var stat%)

Cancels the previously called KEYA function with status stat%.

See the Advanced Topics chapter for details.


KMOD

Usage: k%=KMOD

Returns a code representing the state of the modifier keys (whether they were pressed or not) at the time of the last keyboard access, such as a KEY function. The modifiers have these codes:

			binary
2	Shift down	    10
4	Control down	   100
8	Psion down	  1000
16	Caps Lock on	 10000
If there was no modifier, the function returns 0. If a combination of modifiers was pressed, the sum of their codes is returned for example 20 is returned if Control (4) was held down and Caps Lock (16) was on.

Always use immediately after a KEY/KEY$/GET/GET$ statement.

The value returned by KMOD has one binary bit set for each modifier, as shown above. By using the logical operator AND on the value returned by KMOD you can check which of the bits are set, in order to see which modifier keys were held down. For more details on AND, see Appendix B.

Example:


LAST

Usage: LAST

Positions to the last record in a data file.


LCLOSE

Usage: LCLOSE

Closes the device opened with LOPEN. (The device is also closed automatically when a program ends.)


LEFT$

Usage: b$=LEFT$(a$,x%)

Returns the leftmost x% characters from the string a$.

For example if n$ has the value Charles, then b$=LEFT$(n$,3) assigns Cha to b$.


LEN

Usage: a%=LEN(a$)

Returns the number of characters in a$.

Eg if a$ has the value 34 Kopechnie Drive then LEN(a$) returns 18.

You might use this function to check that a data file string field is not empty before displaying:

IF LEN(A.client$)
  PRINT A.client$
ENDIF


LENALLOC

Usage: len%=LENALLOC(pcell%)

Returns the length of the previously allocated cell at pcell%.


LINKLIB

Usage: LINKLIB cat%

Link any libraries that have been loaded using LOADLIB.


LN

Usage: a=LN(x)

Returns the natural (base e) logarithm of x.

Use LOG to return the base 10 log of a number.


LOADLIB

Usage: ret%=LOADLIB(var cat%,name$,link%)

Load and optionally link a DYL that is not in the ROM. If successful, this writes the category handle to cat% and returns zero. The DYL is shared in memory if already loaded by another process.


LOADM

Usage: LOADM module$

Loads a translated OPL module so that procedures in that module can be called. Until a module is loaded with LOADM, calls to procedures in that module will give an error.

module$ is a string containing the name of the module. Specify the full file name only where necessary.

Example LOADM MODUL2

Up to 4 modules can be in memory at any one time; if you try to LOADM a fifth module, you get an error. Use UNLOADM to remove a module from memory so that you can load a different one.

By default, LOADM always uses the directory of the initial running program, or the one specified by a OPA application. It is not affected by the SETPATH command.


LOC

Usage: a%=LOC(a$,b$)

Returns an integer showing the position in a$ where b$ occurs, or zero if b$ doesn't occur in a$. The search matches upper and lower case.

Example: LOC(STANDING,AND) would return the value 3 because the substring AND starts at the third character of the string STANDING.


LOCAL

Usage: LOCAL variables

Used to declare variables which can be referenced only in the current procedure. Other procedures may use the same variable names to create new variables. Use GLOBAL to declare variables common to all called procedures.

The variables may be of 4 types, depending on the symbol they end with:

Array variables have a number immediately following them in brackets which specifies the number of elements in the array. Array variables may be of any type, for example: x(6),y%(5),f$(5,12),z&(3)

When declaring string arrays, you must give two numbers in the brackets. The first declares the number of elements, the second declares their maximum length. For example surname$(5,8) declares five elements, each up to 8 characters long.

Variable names may be any combination of up to 8 numbers and alphabetic letters. They must start with a letter. The length includes the % & or $ sign, but not the () in string and array variables.

More than one GLOBAL or LOCAL statement may be used, but they must be on separate lines, immediately after the procedure name.

See also GLOBAL and Chapter 2.


LOCK

Usage:
LOCK ON
LOCK OFF

Mark an OPA (OPL application) as locked or unlocked. When an OPA is locked with LOCK ON, the System screen will not send it events to change files or quit. If, for example, you move onto the file list in the System screen and press Delete to try to stop that running OPA, a message will appear, indicating that the OPA cannot close down at that moment.

You should use LOCK ON if your OPA uses a command, such as EDIT or DIALOG, which accesses the keyboard. You might also use it when the OPA is about to go busy for a considerable length of time, or at any other point where a clean exit is not possible. Do not forget to use LOCK OFF as soon as possible afterwards.

`Foreground', `background' and `Machine on' events may still occur while the OPA is accessing the keyboard, and will be discarded.

An OPA is initially unlocked.


LOG

Usage: a=LOG(x)

Returns the base 10 logarithm of x.

Use LN to find the base e (natural) log.


LOPEN

Usage: LOPEN device$

Opens the device to which LPRINTs are to be sent.

No LPRINTs can be sent until a device has been opened with LOPEN.

You can open any of these devices:

LOPEN REM::C:\BAK\MEMO.TXT

or on an Apple Macintosh:

LOPEN REM::HD40:ME:MEMO5

Any existing file of the name given will be overwritten when you print to it.

Only one device may be open at any one time. Use LCLOSE to close the device. (It also closes automatically when a program finishes running.)

See Appendix C.


LOWER$

Usage: b$=LOWER$(a$)

Converts any upper case characters in the string a$ to lower case and returns the completely lower case string.

Eg if a$=CLARKE, LOWER$(a$) returns the string clarke

Use UPPER$ to convert a string to upper case.


LPRINT

Usage: LPRINT list of expressions

Prints a list of items, in the same way as PRINT, except that the data is sent to the device most recently opened with LOPEN.

The expressions may be quoted strings, variables, or the evaluated results of expressions. The punctuation of the LPRINT statement (commas, semi-colons and new lines) determines the layout of the printed text, in the same way as PRINT statements.

If no device has been opened with LOPEN you will get an error.

See PRINT for displaying to the screen.

See LOPEN for opening a device for LPRINT.

See Appendix C for an overview of printing from OPL.


MAX

Usage:
m=MAX(list)
m=MAX(array(),element)

Returns the greatest of a list of numeric items.

The list can be either:

or When operating on an array, the first argument must be the array name followed by (). The second argument, separated from the first by a comma, is the number of array elements you wish to operate on for example m=MAX(arr(),3) would return the value of the largest of elements arr(1), arr(2) and arr(3).


mCARD

Usage:
mCARD title$,n1$,k1%
mCARD title$,n1$,k1%,n2$,k2% etc.

Defines a menu. When you have defined all of the menus, use MENU to display them.

title$ is the name of the menu. From one to eight items on the menu may be defined, each specified by two arguments. The first is the item name, and the second the keycode for a hot-key. This specifies a key which, when pressed together with the Psion key, will select the option. If the keycode is for an upper case key, the hot-key will use both the Shift and Psion keys.

The options can be divided into logical groups by displaying a grey line under the final option in a group. To do this, pass the negative value corresponding to the hot-key keycode for the final option in the group. For example, -%A specifies hot-key Shift-Psion-A and displays a grey line under the associated option in the menu.


MEAN

Usage:
m=MEAN(list)
m=MEAN(array(),element)

Returns the arithmetic mean (average) of a list of numeric items.

The list can be either:

or When operating on an array, the first argument must be the array name followed by (). The second argument, separated from the first by a comma, is the number of array elements you wish to operate on for example m=MEAN(arr(),3) would return the average of elements arr(1), arr(2) and arr(3).

This example displays 15.0:

a(1)=10
a(2)=15
a(3)=20
PRINT MEAN(a(),3)


MENU

Usage:
val%=MENU
val%=MENU (var init%)

Displays the menus defined by mINIT and mCARD, and waits for you to select an item. Returns the hot-key keycode of the item selected, as defined in mCARD, in lower case.

If you cancel the menu by pressing Esc, MENU returns 0.

If the name of a variable is passed is sets the initial menu and item to be highlighted. init% should be 256*(menu%)+item%; for both menu% and item%, 0 specifies the first, 1 the second and so on. If init% is 517 (=256*2+5), for example, this specifies the 6th item on the third menu.

If init% was passed, MENU writes back to init% the value for the item which was last highlighted on the menu. You can then use this value when calling the menu again. You only need to use this technique if you have more than one menu in your program.


MID$

Usage: m$=MID$(a$,x%,y%)

Returns a string comprising y% characters of a$, starting at the character at position x%.

Eg if name$=McConnell then MID$(name$,3,3) would return the string Con.


MIN

Usage:
m=MIN(list)
m=MIN(array(),element)

Returns the smallest of a list of numeric items.

The list can be either:

or

When operating on an array, the first argument must be the array name followed by (). The second argument, separated from the first by a comma, is the number of array elements you wish to operate on for example m=MIN(arr(),3) would return the minimum of elements arr(1), arr(2) and arr(3).


mINIT

Usage: mINIT

Prepares for definition of menus, cancelling any existing menus. Use mCARD to define each menu, then MENU to display them.


MINUTE

Usage: m%=MINUTE

Returns the current minute number from the system clock (0 to 59).

Eg at 8.54am MINUTE returns 54.


MKDIR

Usage: MKDIR name$

Creates a new directory. For example, MKDIR M:\MINE\TEMP creates a M:\MINE\TEMP directory, also creating M:\MINE if it is not already there.


MONTH

Usage: m%=MONTH

Returns the current month from the system clock as an integer between 1 and 12.

Eg on 12th March 1992 m%=MONTH returns 3 to m%.


MONTH$

Usage: m$=MONTH$(x%)

Converts x%, a number from 1 to 12, to the month name, expressed as a three-letter mixed case string.

Eg MONTH$(1) returns the string Jan.


NEWOBJ

Usage: pobj%=NEWOBJ(num%,clnum%)

Create a new object by category number num% belonging to the class clnum%, returning the object handle on success or zero if out of memory.


NEWOBJH

Usage: pobj%=NEWOBJH(cat%,clnum%)

Create a new object by category handle cat% belonging to the class clnum%, returning the object handle on success or zero if out of memory.


NEXT

Usage: NEXT

Positions to the next record in the current data file.

If NEXT is used after the end of a file has been reached, no error is reported but the current record is a null and the EOF function returns true.


NUM$

Usage: n$=NUM$(x,y%)

Returns a string representation of the integer part of the floating-point number x, rounded to the nearest whole number. The string will be up to y% characters wide.

See also FIX$, GEN$, SCI$.


ODBINFO

Usage: ODBINFO var info%()

Provided for advanced use only, this keyword allows you to use OS and CALL to access data file interrupt functions not accessible with OPL keywords. See the `Advanced topics' chapter for more details.


OFF

Usage:
OFF
OFF x%

Switches the Series 3a off.

When you switch back on, the statement following the OFF command is executed, for example: OFF :PRINT Hello again

If you specify an integer, x%, between 8 and 16383, the machine switches off for that number of seconds and then automatically turns back on and continues with the next line of the program (16383 is about 4 1/2 hours). However, during this time the machine may be switched on by an alarm, and of course you can turn it on with ON/Esc.

Warning: be careful how you use this command. If, due to a programming mistake, a program uses OFF in a loop, you may find it impossible to switch the Series 3a back on, and may have to reset the computer.


ONERR

Usage:
ONERR label
ONERR label::
..
..
ONERR OFF

ONERR label:: establishes an error handler in a procedure. When an error is raised, the program jumps to the label:: instead of the program stopping and an error message being displayed.

The label may be up to 8 characters long starting with a letter. It ends with a double colon (::), although you don't need to use this in the ONERR statement.

ONERR OFF disables the ONERR command, so that any errors occurring after the ONERR OFF statement no longer jump to the label.

It is advisable to use the command ONERR OFF immediately after the label:: which starts the error handling code.

See the Error handling chapter for full details.


OPEN

Usage: OPEN file$,log,f1,f2...

Opens an existing data file file$, giving it the logical file name log, and giving the fields the names f1, f2..

You need only specify those fields which you intend to update or append, though you cannot miss out a field.

The opened file is then referred to within the program by its logical name (A, B, C or D).

Up to 4 files can be open at once.

Example:

OPEN clients,A,name$,addr$

See also CREATE, USE and OPENR.


OPENR

This command works exactly like OPEN except that the opened or UPDATE the records it contains.

This means that you can run two separate programs at the same time, both sharing the same file.


OS

Usage:
a%=OS(i%, addr1%)
a%=OS(i%, addr1%, addr2%)

Calls the Operating System interrupt i%, reading the values of all returned 8086 registers and flags. The CALL function, although simpler to use, does not allow the AL register to be passed and no flags are returned, making it suitable only for certain interrupts.

The input registers are passed at the address addr1%. The output registers are returned at the address addr2% if supplied, otherwise they are returned at addr1%. Both addresses can be of an array, or of six consecutive integers.

Register values are stored sequentially as 6 integers and represent the register values in this order: AX, BX, CX, DX, SI and DI. The interrupt's function number, if required, is passed in AH.

The output array must be large enough to store the 6 integers returned in all cases, irrespective of the interrupt being called.

The value returned by OS is the flags register. The Carry flag, which is relevant in most cases, is in bit 0 of the returned value, so (a% and 1) will be `True' if Carry is set. Similarly, the Zero flag is in bit 6, the Sign flag in bit 7 and the Overflow flag in bit 10.

For example, to find cos(pi/4):

The OS function requires extensive knowledge of the Operating System and related programming techniques.

See also CALL.


PARSE$

Usage: p$=PARSE$(f$,rel$,var off%())

Returns a full file specification from the filename f$, filling in any missing information from rel$.

The offsets to the filename components in the returned string is returned in off%() which must be declared with at least 6 integers:

off%(1)	filing system offset (1 always)
off%(2)	device offset
off%(3)	path offset
off%(4)	filename offset
off%(5)	file extension offset
off%(6)	flags for wildcards in returned string

The flag values in offset%(6) are:

0 : no wildcards 1 : wildcard in filename 2 : wildcard in file extension 3 : wildcard in both

If rel$ is not itself a complete file specification, the current filing system, device and/or path are used as necessary to fill in the missing parts.

f$ and rel$ should be separate strings.

Example:

p$=PARSE(NEW,LOC::M:\ODB\*.ODB,x%())

sets p$ to LOC::M:\ODB\NEW.ODB and x%() to (1,6,8,13,16,0).


PATH

Usage: PATH name$

Gives the directory to use for an OPA's files.

This can only be used between APP and ENDA.

See the Advanced Topics chapter for more details of OPAs.


PAUSE

Usage: PAUSE x%

Pauses the program for a certain time, depending on the value of x%:

x% result

0 waits for a key to be pressed.

+ve pauses for x% twentieths of a second.

-ve pauses for x% twentieths of a second or until a key is pressed.

So PAUSE 100 would make the program pause for 100/20 = 5 seconds, and PAUSE -100 would make the program pause for 5 seconds or until a key is pressed.

If x% is 0, a GET, GET$, KEY or KEY$ will return the key press which terminated the pause. If you are not interested in this keypress, but in the one which follows it, clear the buffer after the PAUSE with a single KEY function: PAUSE 0 :KEY

You should be especially careful about this if x% is negative, since then you cannot tell whether the pause was terminated by a keypress or by the time running out.


PEEK functions

The PEEK functions find the values stored in specific bytes. Usage:

p%=PEEKB(x%) returns the integer value of the byte at address x% p%=PEEKW(x%) returns the integer at address x% p&=PEEKL(x%) returns the long integer value at address x% p=PEEKF(x%) returns the floating-point value at address x% p$=PEEK$(x%) returns the string at address x%

Usually you would find out the byte address with the ADDR function. For example, if var% has the value 7, PEEKW(ADDR(var%)) returns 7.

The different types are stored in different ways across bytes:

ADDR returns the address of the first (least significant) byte.

ADDR returns the address of the first (least significant) byte.

Each letter is stored as its character code for example, A as 65.

For example, if var$=ABC, PEEK$(ADDR(var$)) will return the string ABC. ADDR returns the address of the length byte.

You can use ADDR to find the address of the first element in an array, for example ADDR(x%()). You can also specify individual elements of the array, for example ADDR(x%(2)).

See also the POKE commands and ADDR.


PI

Usage: p=PI

Returns the value of Pi (3.14... ).


POKE commands

The POKE commands store values in specific bytes. Usage:

POKEB x%,y% stores the integer value y% (less than 256) in the single byte at address x% POKEW x%,y% stores the integer y% across two consecutive bytes, with the least significant byte in the lower address, that is x% POKEL x%,y& stores the long-integer y& in bytes starting at address x% POKEF x%,y stores the floating-point value y in bytes starting at address x% POKE$ x%,y$ stores the string y$ in bytes starting at address x%

Use ADDR to find out the address of your declared variables.

Warning: Casual use of these commands can result in the loss of data in the Series 3a.

See PEEK for more details of how the different types are stored across bytes.


POS

Usage: p%=POS

Returns the number of the current record in the current data file, from 1 (the first record) upwards.

A file can have up to 65534 records. However integers can only be in the range -32768 to +32767. Record numbers above 32767 are therefore returned like this:

record	value	returned by POS
32767		 32767
32768		-32768
32769		-32767
32770		-32766
.		.
.		.
65534		-2

To display record numbers, you can use this check:

IF POS<0 PRINT 65536+POS ELSE PRINT POS ENDIF


POSITION

Usage: POSITION x%

Makes record number x% the current record in the current data file.

If x% is greater than the number of records in the file then the EOF function will return true.


POSSPRITE

Usage: POSSPRITE x%,y%

Set the position of the current sprite to pixel x%,y%.


PRINT

Usage: PRINT list of expressions

Displays a list of expressions on the screen. The list can be punctuated in one of these ways:

On 1st January 1993, PRINT TODAY is, PRINT DAY;.;MONTH would display TODAY is 1.1

PRINT 1 1 PRINT Hello Hello PRINT Number,1 Number 1 See also LPRINT, gUPDATE, gPRINT, gPRINTB, gPRINTCLIP, gXPRINT.


RAD

Usage: r=RAD(x)

Converts x from degrees to radians.

All the trigonometric functions assume angles are specified in radians, but it may be easier for you to enter angles in degrees and then convert with RAD.

Example:

(The formula used is ^PI*x/180)

To convert from radians to degrees use DEG.


RAISE

Usage: RAISE x%

Raises an error.

The error raised is error number x%. This may be one of the errors listed in the Error handling chapter, or a new error number defined by you.

The error is handled by the error processing mechanism currently in use either OPL's own, which stops the program and displays an error message, or the ONERR handler if you have ONERR on.

For a full explanation, see the Error handling chapter.


RANDOMIZE

Usage: RANDOMIZE x&

Gives a `seed' (start-value) for RND.

Successive calls of the RND function produce a sequence of random numbers. If you use RANDOMIZE to set the seed back to what it was at the beginning of the sequence, the same sequence will be repeated.

For example, you might want to use the same `random' values to test new versions of a procedure. To do this, precede the RND statement with the statement RANDOMIZE value. Then to repeat the sequence, use RANDOMIZE value again.

Example:


REALLOC

Usage: pcelln%=REALLOC(pcell%,size%)

Change the size of a previously allocated cell at pcell% to size%, returning the new cell address or zero if there is not enough memory.


RECSIZE

Usage: r%=RECSIZE

Returns the number of bytes occupied by the current record.

Use this function to check that a record may have data added to it without overstepping the 1022-character limit.

Example:


REM

Usage: REM text

Precedes a remark you include to explain how a program works. All text after the REM up to the end of the line is ignored.

When you use REM at the end of a line you need only precede it with a space, not a space and a colon.

Examples:

INPUT a
b=a*.175 REM b=TAX
INPUT a
b=a*.175 :REM b=TAX


RENAME

Usage: RENAME file1$,file2$

Renames file1$ as file2$. You can rename any type of file.

You cannot use wildcards.

You can rename across directories RENAME \dat\x.dbf,\x.dbf is OK. If you do this, you can choose whether or not to change the name of the file.

Example:

PRINT Old name: :INPUT a$
PRINT New name: :INPUT b$
RENAME a$,b$


REPT$

Usage: r$=REPT$(a$,x%)

Returns a string comprising x% repetitions of a$.

For example, if a$=ex, r$=REPT$(a$,5) returns exexexexex.


RETURN

Usage: RETURN or RETURN variable

Terminates the execution of a procedure and returns control to the point where that procedure was called (ENDP does this automatically).

RETURN variable does this as well, but also passes the value of variable back to the calling procedure. The variable may be of any type. You can return the value of any single array element for example RETURN x%(3). You can only return one variable.

RETURN on its own, and the default return through ENDP, causes the procedure to return the value 0 or a null string.

Example: (picture)


RIGHT$

Usage: r$=RIGHT$(a$,x%)

Returns the rightmost x% characters of a$.

Example:

PRINT Enter name/ref,

INPUT c$

ref$=RIGHT$(c$,4)

name$=LEFT$(c$,LEN(c$)-4)


RMDIR

Usage: RMDIR str$

Removes the directory given by str$. You can only remove empty directories.


RND

Usage: r=RND

Returns a random floating-point number in the range 0 (inclusive) to 1 (exclusive).

To produce random numbers between 1 and n eg between 1 and 6 for a dice use the following statement: f%=1+INT(RND*n)

RND produces a different number every time it is called within a program. A fixed sequence can be generated by using RANDOMIZE. You might begin by using RANDOMIZE with an argument generated from MINUTE and SECOND (or similar), to seed the sequence differently each time.

Example:


SCI$

Usage: s$=SCI$(x,y%,z%)

Returns a string representation of x in scientific format, to y% decimal places and up to z% characters wide. Examples:

SCI$(123456,2,8)=1.23E+05 SCI$(1,2,8)=1.00E+00 SCI$(1234567,1,-8)= 1.2E+06

If the number does not fit in the width specified then the returned string contains asterisks.

If z% is negative then the string is right-justified.

See also FIX$, GEN$, NUM$.


SCREEN

Usage:
SCREEN width%,height%
SCREEN width%,height%,x%,y%

Changes the size of the window in which text is displayed. x%,y% specify the character position of the top left corner; if they are not given, the text window is centred in the screen.

An OPL program can initially display text to the whole screen.


SCREENINFO

Usage: SCREENINFO var info%()

Gets information on the text screen (as used by PRINT etc.)

This keyword allows you to mix text and graphics. It is required because while the default window is the same size as the physical screen, the text screen is slightly smaller and is centred in the default window. The few pixels gaps around the text screen, referred to as the left and top margins, depend on the font in use.

On return, the array info%(), which must have at least 10 elements, contains this information:

info%(1) left margin in pixels
info%(2) top margin in pixels
info%(3) text screen width in character units
info%(4) text screen height in character units
info%(5) reserved (window server id for default window)
info%(6) font id (FONT and gFONT)
info%(7) pixel width of text window character cell
info%(8) pixel height of text window line
info%(9), info%(10) reserved
Initially SCREENINFO returns the values for the initial text screen. Subsequently any keyword which changes the size of the text screen font, such as FONT, will change some of these values and SCREENINFO should therefore be called again.

See also FONT.


SECOND

Usage: s%=SECOND

Returns the current time in seconds from the system clock (0 to 59).

Eg at 6:00:33 SECOND returns 33.


SECSTODATE

Usage: SECSTODATE s&,var yr%,var mo%,var dy%,var hr%,var mn%,var sc%,var yrday%

Sets the variables passed by reference to the date corresponding to s&, the number of seconds since 00:00 on 1 January 1970. yrday% is set to the day in the year (1-366).

s& is an unsigned long integer. To use values greater than +2,147,483,647, subtract 4,294,967,296 from the value.

See also DATETOSECS, HOUR, MINUTE, SECOND, dDATE, DAYS.


SEND

Usage: ret%=SEND(pobj%,m%,var p1,...)

Send a message to the object pobj% to call the method number m%, passing between zero and three arguments (p1...) depending on the requirements of the method, and returning the value returned by the selected method.


SETNAME

Usage: SETNAME name$

Sets the name of the running OPA to name$ and redraws any status window, using that name below the icon.


SETPATH

Usage: SETPATH name$

Sets the current directory for file access for example, SETPATH a:\docs. LOADM continues to use the directory of the initial program, or the one specified by a OPA application, but all other file access will be to the new directory.


SIN

Usage: s=SIN(angle)

Returns the sine of angle, an angle expressed in radians.

To convert from degrees to radians, use the RAD function.


SPACE

Usage: s&=SPACE

Returns the number of free bytes on the device on which the current (open) data file is held.

Example:


SQR

Usage: s=SQR(x)

Returns the square root of x.


STATUSWIN

Usage one of:
STATUSWIN ON,type%
STATUSWIN ON
STATUSWIN OFF

Displays or removes a `permanent' status window.

If type%=1 the small status window is shown. If type%=2 the large status window is shown. STATUSWIN ON on its own displays an appropriate status window; on the Series 3a this will always be the large status window.

The permanent status window is behind all other OPL windows. In order to see it, you must use FONT (or both SCREEN and gSETWIN) to reduce the size of the text and graphics windows. You should ensure that your program does not create windows over the top of it.


STATWININFO

Usage: t%=STATWININFO(type%,var xy%())

Sets xy%(1),xy%(2), xy%(3) and xy%(4) to the top left x, top left y, width and height respectively of the specified type of status window. type%=1 is the small status window; type%=2 is the large status window; type%=3 is the Series 3 compatibility mode status window; type%=-1 is whichever status window is current.

STATWININFO returns t%, the type of the current status window (with values as for type%, or zero if there is no current status window).


STD

Usage:
s=STD(list)
s=STD(array(),element)

Returns the standard deviation of a list of numeric items.

The list can be either:

or

When operating on an array, the first argument must be the array name followed by (). The second argument, separated from the first by a comma, is the number of array elements you wish to operate on for example m=STD(arr(),3) would return the standard deviation of elements arr(1), arr(2) and arr(3).

This function gives the sample standard deviation, using the formula:

SQR((xi-x)^2/(n-1)) where x means xi/n. To convert to population standard deviation, multiply the result by SQR((n-1)/n).


STOP

Usage: STOP

Ends the running program.


STYLE

Usage: STYLE style%

Sets the text window character style. style% can be 2 for underlined, or 4 for inverse.

See `The text and graphics windows' at the end of the `Graphics' chapter for more details.


SUM

Usage:
s=SUM(list)
s=SUM(array(),element)

Returns the sum of a list of numeric items.

The list can be either:

or

When operating on an array, the first argument must be the array name followed by (). The second argument, separated from the first by a comma, is the number of array elements you wish to operate on for example m=SUM(arr(),3) would return the sum of elements arr(1), arr(2) and arr(3).


TAN

Usage: t=TAN(angle)

Returns the tangent of angle, an angle expressed in radians.

To convert from radians to degrees, use the DEG function.


TESTEVENT

Usage: t%=TESTEVENT

Returns `True' if an event has occurred, otherwise returns `False'. The event is not read by TESTEVENT it may be read with GETEVENT.


TRAP

Usage: TRAP command

TRAP is an error handling command. It may precede any of these commands:

Data file commands

APPEND, UPDATE BACK, NEXT, LAST, FIRST, POSITION USE, CREATE, OPEN, OPENR, CLOSE, DELETE

File commands

COPY, COMPRESS, ERASE, RENAME LOPEN, LCLOSE LOADM, UNLOADM

Directory commands

MKDIR, RMDIR

Data entry commands

EDIT, INPUT

Graphics commands

gSAVEBIT, gCLOSE, gUSE gUNLOADFONT, gFONT, gPATT, gCOPY

For example, TRAP FIRST.

Any error resulting from the execution of the command will be trapped. Program execution will continue at the statement after the TRAP statement, but ERR will be set to the error code.

TRAP overrides any ONERR.

See the Error handling chapter for further details.


TYPE

Usage: TYPE num%

Sets the type of an OPA, from 0 to 4, with num%. On the Series 3a you should set num% from $1000 to $1004 to set type 0 to 4 respectively; the $1000 allows a 48x48 black/grey icon to be used.

This can only be used between APP and ENDA.

See the Advanced Topics chapter for more details of OPAs.


UADD

Usage: i%=UADD(val1%, val2%)

Add val1% and val2%, as if both were unsigned integers with values from 0 to 65535. Prevents integer overflow for pointer arithmetic eg UADD(ADDR(text$),1) should be used instead of ADDR(text$)+1.

One argument would normally be a pointer and the other an offset expression.

See also USUB.


UNLOADLIB

Usage: ret%=UNLOADLIB(var cat%)

Unload a DYL from memory. If successful, returns zero.


UNLOADM

Usage: UNLOADM module$

Removes from memory the module module$ loaded with LOADM.

module$ is a string containing the name of the translated module.

The procedures in an unloaded module cannot then be called by another procedure.


UNTIL

See DO. *** UPDATE

Usage:

UPDATE

Deletes the current record in the current data file and saves the current field values as a new record at the end of the file.

This record, now the last in the file, remains the current record.

Example:

A.count=129

A.name$=Brown

UPDATE

Use APPEND to save the current field values as a new record.


UPPER$

Usage: u$=UPPER$(a$)

Converts any lower case characters in a$ to upper case, and returns the completely upper case string.

Example:

...
CLS :PRINT Y to continue
PRINT or N to stop.
g$=UPPER$(GET$)
IF g$=Y
 nextproc:
ELSEIF g$=N
 RETURN
ENDIF
...
Use LOWER$ to convert to lower case.


USE

Usage: USE logical name

Selects the data file with the logical name A, B, C or D. The file must previously have been opened with OPEN, OPENR or CREATE and not yet be closed.

All the record handling commands (such as POSITION or UPDATE) then operate on this file.


USR

Usage: u%=USR(pc%,ax%,bx%,cx%,dx%)

Executes your machine code, returning an integer. The USR code (ie the assembler code you have written) must return with a far RET, otherwise the program will crash.

The values of of ax%,bx%... are passed to the AX, BX... 8086 registers. The microprocessor then executes the machine code starting at pc%. At the end of the routine, the value in the AX register is passed back to u%.

WARNING: Casual use of this function can result in the loss of data in the Series 3a.

This example shows a simple operation, ending with a far RET:

See also USR$, ADDR, PEEK, POKE.


USR$

Usage: u$=USR$(pc%,ax%,bx%,cx%,dx%)

Executes your machine code, returning a string. The USR$ code you have written must return with a far RET, otherwise the program will crash.

The values of of ax%,bx%... are passed to the ax, bx... 8086 registers. The microprocessor then executes the machine code starting at pc%. At the end of the routine, the value in the ax register must point to a length-byte preceded string. This string is then copied to u$.

WARNING: Casual use of this function can result in the loss of data in the Series 3a.

See USR for an example. See also ADDR, PEEK, POKE.


USUB

Usage: i%=USUB(val1%, val2%)

Subtract val2% from val1%, as if both were unsigned integers with values from 0 to 65535. Prevents integer overflow for pointer arithmetic.

See also UADD.


VAL

Usage: v=VAL(numeric string)

Returns the floating-point number corresponding to a numeric string.

The string must be a valid number eg not 5.6.7 or 196f. Expressions, such as 45.6*3.1, are not allowed. Scientific notation such as 1.3E10, is OK.

Eg VAL(470.0) returns 470

See also EVAL.


VAR

Usage:
v=VAR(list)
v=VAR(array(),element)

Returns the variance of a list of numeric items.

The list can be either:

or

When operating on an array, the first argument must be the array name followed by (). The second argument, separated from the first by a comma, is the number of array elements you wish to operate on for example m=VAR(arr(),3) would return the variance of elements arr(1), arr(2) and arr(3).

This function gives the sample variance, using the formula:

(xi-x)^2/(n-1) where x means xi/n. To convert to population variance, multiply the result by (n-1)/n


VECTOR

Usage:
VECTOR I%
label1,label2,...
labelN
ENDV

VECTOR I% jumps to label number i% in the list if i% is 1 this will be the first label, and so on. The list is terminated by the ENDV statement. The list may spread over several lines, with a comma separating labels in any one line but no comma at the end of each line.

If i% is not in the range 1 to N, where N is the number of labels, the program continues with the statement after the ENDV statement.

See also GOTO.


WEEK

Usage: w%=WEEK(day%,month%,year%)

Returns the week number in which the specified day falls, as an integer between 1 and 53.

day% must be between 1 and 31, month% between 1 and 12, year% between 1900 and 2155.

Each week is taken to begin on the `Start of week' day, as specified in the Time application. When a year begins on a different day, it counts as week 1 if there are four or more days before the next week starts..


WHILE...ENDWH

Usage:
WHILE expression
..
...
..
....
ENDWH

Repeatedly performs the set of instructions between the WHILE and the ENDWH statement, so long as expression returns logical true non-zero.

If expression is not true, the program jumps to the line after the ENDWH statement.

Every WHILE must be closed with a matching ENDWH.

See also DO...UNTIL and Chapter 3.


YEAR

Usage: y%=YEAR

Returns the current year as an integer between 1900 and 2155 from the system clock.

For example, on 5th May 1992 YEAR returns 1992.


<== Return to the Index