Inno VB system 

This isn't actually one of my downloads but so associated with VB6 project distribution that I thought it fits in best here.

Inno Setup provides a scripting language to compile a setup program for Windows applications.

For Visual Basic users, this is a useful and very versatile alternative to the "Package & Deployment Wizard", (PDW), particularly for packages with a large number of files. One of the main problems with PDW is that files have to be added one-at-time to the basic selection and there is no easy facility to create sub-folders in the installed package.

Inno allows both of these problems to be solved and has a host of other facilities not available with PDW. In fact, my only criticism is that there are so many possibilities that there is a very steep learning curve.

• Inno Setup is a freeware package which can be downloaded from here
• The on-line help file is available here
• There is also an independent project translation package, InnoScript, from Randem Systems available here

For anyone who's interested, I've outlined my use of Inno Setup for VB6 packages on this linked page.

 Downloads 

 Project structure 

It's worth spending a little time considering the structure of your project before starting to create an installation utility.

Because of Inno's ability to add folders and their contents, it's advantageous to put as much as you can into folders so that the individual files don't have to be defined in the Inno script file. In fact, you could put everything into one folder which would then be the only thing to be defined in the script and which would become the operational folder of the installed package.

In particular, I place all the graphics required by the package into a 'graphics' folder rather than have them in the same folder as the VB code.

When I have a project working to my satisfaction, I create a folder called 'Publish' or 'Package' and copy a clean version of all the material into it.

 Creating the Inno script 

 Run the VB6 Package & Deployment Wizard 

Whilst you can just write the script or use the Inno Script Wizard, it's useful to actually start from the 'setup.lst' file produced by the VB Wizard. This will give you information on the dependencies of your package if you're not sure of what these may be. This can be a problem: if you're using the File System Object in your package, you may be aware that you'll need 'scrrun.dll' to utilise facets of VBscript but I wasn't aware until I ran PDW that you also need 'msvcrt.dll'!

Just add the minium of data when you use the VB wizard and only add the [project].exe file. The wizard also creates a 'Support' folder which will have copies of the .ocx and other files that are required. You should access these copies in the Inno script rather than the ones in your system folder/s. If you try to do the latter, you'll get a warning when you try to compile the script.

 Using a script template 

I recommend that you set up a simple script to be used as a template and just alter the files, project name, etc. for each package you create.

The details of the one I use are explained on this linked page.

 Other help 

Other sources of help with Inno Setup scripting are,

• Jordan Russell's Knowledge base
• The INNO setup runtime installer for VB6
• Fred's Shack's Quick Guide to InnoSetup5

 Downloads 

 HTML Help 

This package provides an alternative to the MS HTML Help Workshop and the WinHelp Workshop to allow you to create HTML Help files for VB packages where you have complete control of the styling of all aspects of the display.

 Downloads 

The MAP EDITOR package allows you to create "hot-spot" areas on images for .HTM files.

As well as defining the type and coordinates of the area, the program produces HTML code for the action required:

• A link,
• A Pop-Up,
• Opening a New Window or
• An inactive area.

Maps can be stored and there is also an extensive "Help" file included.

 Downloads 

3D polar graph

2D Parametric function

The GRAPHING package allows you to draw graphs for nine types of function.

These are functions of the form:-

• z = f(x, y),
• x = f₁(u,v), y = f₂(u,v), z = f₃(u,v)
• r = f(a, b),
• x = f₁(t), y = f₂(t), z = f₃(t)
• r = f₁(t), a = f₂(t), z = f₃(t)
• z = f(r, a),
• y = f(x),
• r = f(a) and
• x = f₁(t), y = f₂(t)

The functions can be plotted without or with axes which can be scaled and the resolution of the graph can be changed to facilitate speed of plotting.

Up to 10 different functions of the same type can be overlaid with each being drawn in a different colour.

The 3D functions can be displayed and printed in one of two stereo forms: stereo-pairs or red/blue anaglyphs.

Functions can be stored and any number of different storage files created. There is also an extensive "Help" file included.

The package now allows 3D surface plotting with some forms of graph, courtesy of the Surface Viewer control by Bright Ideas Software

z = f(x,y) with hidden lines removed

3D Parametric function

Stereo-pair

Have a look at the Graph gallery

or download the GRAPHING package, (3.38M).

Other graphing sites:-

Open GL 3D graphing

Pacific Tech

3D theory

Study Works, 3D graphing

Curvus Pro - the reference for 2D curves and 3D graphs

Bright Ideas software, 3D graphs and calculus software

Cool graphs

 Downloads 

	z = Cos(x+2*pi*cos(y/2))+Cos(y+2*pi*Cos(x/2)) x = -9 to 9, y = -9 to 9		z = (x*x-y*y)^2/(x*x+y*y) x = -9 to 9, y = -9 to 9 Psuedo-Bessell function
	Surface plot of z = 4*(x*x-y*y)^2/(x*x+y*y) x = -5 to 5, y = -5 to 5 Psuedo-Bessell function		z = Cos(x+2*pi*Cos(y/2)) x = -2pi to 2pi, y = -2pi to 2pi
	z = Exp(Sin(x)*Sin(y)*3)/4 x = -2pi to 2pi, y = -2pi to 2pi		z = Cos(x*x-y*y) x = -pi to pi, y = -pi to pi
	x = 2*Cos(t) + Sin(2*t) y = Sin(t) + Cos(2*t) z = t*Cos(t)/10 t = -30 to 30		x = 2*Cos(t) + Cos(2*t) y = 2*Sin(t) + Sin(2*t) z = t*Sin(t) t = -50 to 50
	r = Sqr(4*Cos(a + b)) a = 0 to 2pi, b = 0 to 2pi		r = Sin(2*a) + Sin(2*b) a = 0 to 2pi, b = 0 to 2pi
	r = Sin(a) + Sin(b) a = -pi to pi, b = -pi to pi		r = 1/Sqr(a*a + b*b) a = 0 to 2pi, b = 0 to 2pi
	Red/blue anaglyph r = sin(a + b) - cos(a + b) a = 0 to 2pi b = 0 to 2pi		x=cos(t) + t^2*sin(t) y=sin(t) + t^2*cos(t) t = -8pi to 8pi
	z = e^(-r*r)*(sin(2*r)-r*cos(4*a)) r = 0 to 2 a = -2pi to 2pi		r =2 a = t z =cos(7*t) t = -pi to pi

 Downloads 

	r = (a + b)/(a*a + b*b) a = 0 to 2pi, b = 0 to 2pi		r = 1/(a + b) a = 0 to 2pi, b = 0 to 2pi
	r = (a + b)/a*b a = 0 to 2pi, b = 0 to 2pi		r = a + b a = 0 to 2pi, b = 0 to 2pi
	y = Sin(8/(x^2 + 1)) x = -5pi to 5 pi		y = Sin(x)+Sin(3*x)/3+Sin(5*x)/5+Sin(7*x)/7 +Sin(9*x)/9+Sin(11*x)/11+Sin(13*x)/13 Square wave synthesis x = -3pi to 3pi
	r = a^2 Archimedes spiral a = 0 to 10pi		r = 1/a Hyperbolic spiral a = 0 to 2pi
	r = 4*Cos(a)*(Sin(a)^2) - Cos(a) Trifolium a = -2pi to 2pi		r = 2*Cos(a) + n, (n = 1, 2 & 3) Limaçon of Pascal a = 0 to 2pi
	x = Cos(t) + t*Sin(t) y = Sin(t) + t*Cos(t) t = -2pi to 2pi		x = 2*Cos(t) + Cos(2*t) y = 2*Sin(t) + Sin(2*t) Deltoid t = 0 to 2pi
	x = (t + 1)/(t^3 + 1) y = (t + 1)/(t^2 + 1) t = -100 to 100		x = (t^2 - 1)/(t^4 + 1) y = (t^2 - 1)/(t^3 + 1) t = -100 to 100

 Downloads 

This version lets you display a particular biomorph and all its relations produced by a change of 1 in each gene value or a set of 25 random mutations from a particular biomorph, as shown on the left.

Each type of display can be printed, either as a grid display or as a full-size biomorph, and they can be saved in storage files of which any number can be created.

A sample storage file is included in the package and there is also an extensive "Help" file appended.

Other Biomorph sites:-

Biomorph

Biomorph Breeder

Biomorph viewer

3D-Biomorphs

Biomorphs

 Downloads 

Mandelbröt set

Julia enlargement

This version allows you to draw the Mandelbröt sets for ten different iterative functions and their associated Julia sets.

Function	Real part	Imaginary part
z = z² + c	x² - y²	2xy
z = z³ + c	x³ - 3xy²	3x²y - y³
z = z² + z + c	x² - y² + x	y(2x + 1)
z = z² - z + c	x² - y² - x	y(2x - 1)
z = z(z² - 1) + c	x³ - 3xy² - x	3x²y - y - y³
z = z²(z - 1) + c	x³ - x² - 3xy² + y²	3x²y - 2xy - y³
z = z² - z - 1 + c	x² - y² - x - 1	y(2x - 1)
z = z⁴ + c	x⁴ - 6x²y² + y⁴	4(x³y - xy³)
z = z(z + i) + c	x² - y² - y	x(2y + 1)
z = z²(z + i) + c	x³ - 3xy² - 2xy	3x²y + x² - y² - y³

Areas for enlargement are selected by dragging the mouse over a display and displays can be printed and stored, either as the basic data used to generate the display or as the actual display data.

The illustrations are areas of the sets produced by the function z = z²(z - 1) + c

A sample storage file is included in the package and there is also an extensive "Help" file appended.

The current version has had a number of modifications added, the most important of which is that you can now enter your own equations.

Julia set

Mandelbröt enlargement

Have a look at the Mandelbröt and Julia gallery
or download the MANDELBRÖT package, (1.58M).

Other Mandelbröt sites:-

Julia and Mandelbröt Sets

Images of the Filled Julia Set for the Riemann Zeta Function

Fractal Explorer Mandelbröt and Julia sets (by Fabio Cesari)

Julia Set, Mu-Ency at MROB

Mandelbröt Set and Julia Sets

Mandelbröt And Julia Set Explorer

Julia and Mandelbröt Set Bibliography

The Fractory Make-Your-Own Julia Sets

Mandel and Julia

Mandelbröt and Julia sets

 Downloads 

	Mandelbröt z = z² + c (-1.40E-1, 8.90E-1), 4.95E-5		Mandelbröt z = z² + c (-7.22E-1, 3.03E-1), 6.51E-3		Julia z = z(z² - 1) + c (-1.52, -1.45), 3.23 (-0.6171, 0.4553i)		Julia z = z(z² - 1) + c (-2.94E-1, 2.52E-1), 1.53E-3 (-0.6171, 0.4553i)
	Mandelbröt z = z² - z - 1 + c (9.19E-1, 1.69E-2), 1.69E-1		Mandelbröt z = z² - z - 1 + c (9.49E-1, 1.33E-1), 3.42E-2		Julia z = z²(z - 1) + c (-1.20, -1.50), 3.00 (0.9541, 0.5385i)		Julia z = z²(z - 1) + c (-5.29E-1, -6.70E-2), 2.17E-1 (0.9541, 0.5385i)
	Mandelbröt z = z⁴ + c (--1.40, -1.30), 2.60		Mandelbröt z = z⁴ + c (-6.85E-1, 3.53E-1), 1.23E-3		Julia z = z⁴ + c (-1.50, -1.50), 3.00 (-0.6841, 0.3534i)		Julia z = z⁴ + c (-8.76E-1, 1.32E-1), 1.20E-1 (-0.6841, 0.3534i)

 Downloads 

	Mandelbröt z = z(z + i) + c (-2.5, -2.5), 4.00		Mandelbröt z = z(z + i) + c (-3.47E-2, -1.90), 1.42E-1		Mandelbröt z = z(z + i) + c (7.32-2, -1.20), 1.00E-2		Mandelbröt z = z(z + i) + c (8.13E-2, -1.02), 4.49E-4
	Julia z = z(z + i) + c (-2.00, --2.50), 4.00 (0.0816, -1.0176i)		Julia z = z(z + i) + c (8.66E-1, 2.12E-1), 1.05E-1 (0.0816, -1.0176i)		Julia z = z(z + i) + c (9.62E-1, 2.45E-1), 2.91E-3 (0.0816, -1.0176i)		Julia z = z(z + i) + c 9.64E-1, 2.48E-1), 8.51E-5 (0.0816, -1.0176i)
	Mandelbröt z = z²(z + i) + c (-1.80, -2.20), 3.60		Mandelbröt z = z²(z + i) + c (-1.85E0-1, -1.041), 7.77E-2		Mandelbröt z = z²(z + i) + c (-1.43E-1, -9.98E-1), 2.51E-3		Mandelbröt z = z²(z + i) + c -1.41E-1, -9.99E-1), 1.16E-4
	Julia z = z²(z + i) + c (-1.50, -1.80), 3.00 (-0.1413, -0.9976i)		Julia z = z²(z + i) + c (7.70E-1, -1.51E-1), 1.02E-1 (-0.1413, -0.9976i)		Julia z = z²(z + i) + c (8.48E-1, 1.88E-1), 4.54E-3 (-0.1413, -0.9976i)		Julia z = z²(z + i) + c 8.49E-1, 1.90E-1), 3.43E-4 (-0.1413, -0.9976i)

 Downloads 

The package allows investigation of a range of statistical operations by the use of an interactive, command driven system.

Data can be entered from the keyboard, generated by commands or imported from Excel workbooks or Access databases.

A simple interpreted language is included to allow the programming of commands.

Data can be stored and re-loaded as can sets of commands or programs.

The package includes an extensive Help file which illustrates the various operations and gives sample programs including that shown in the Command window on the right.

 Downloads 

The command syntax requires that NO SPACES are used within a command. A list of abbreviated commands can be added to the display by right-clicking the text above the command window.

Where there are commands whose only difference is that one operates on a block of data whilst the other processes a column, the command is prefaced by 'b' or 'c' to indicate which is which.

e.g.'b_range' finds the ranges across the rows of a block of data whilst 'c_range' finds the range of the data within the specified column.

The commands are shown with the command in bold and the required parameters in italics. This is followed by a description of the command and an illustrative example.

Some commands require a variable name as a parameter to return the result of an operation.

Descriptions and examples of the commands can be found by clicking on the 'Commands' links in the following list.

The interpretation of the commands and the evaluation of expressions make use of the Microsoft document, Q86688, which has been expanded to deal with more user variables.

 Downloads 

 CLEARING DATA 

clear_all
Clears all data to zero.

b_clear(first column number, last column number)
Clears the block of columns to zero.
e.g. b_clear(3,6) will zero all values in columns 3 through to 6.

 DATA STORAGE 

save_data( first column number,second column number)
This saves the block of data defined by the two column numbers.
e.g. save_data(1,4) will save all the data in columns 1 to 4.

load_data(column number)
This loads a block of data, starting the storage at the specified column number. The data is added below any existing data in the columns concerned,
e.g. load_data(2) will load a block of data, starting the storage in column 2 and continuing for the number of columns in the block.

If STATS data is being re-loaded, a parameter of zero will re-load it into the same columns from which it was saved. Otherwise, it's loaded into the specified column onwards. A parameter of zero will give an error with Access or Excel data.

Executing the command first displays the data file selection screen.

Three types of data file can be loaded.

• MS Access databases, (.MDB files),
• MS Excel workbooks, (.XLS files), and
• STATS custom datafiles, (.SDF files).

 DATA PRODUCTION 

enter_data(number of items, first column number, last column number)
Allows the specified number of items to be entered from the keyboard into a block of columns, a row at a time.
e.g. enter_data(50,1,3) will allow the entry of 50 rows of data into columns 1 to 3.

If data already exists in the block the entered data is added to the existing so, if this is not wanted, clear_all or b_clear should first be used to remove the existing data.
Clicking the CANCEL button will stop the process; saving the last complete row of data.

seq(initial value,geometric factor,arithmetic factor,number of items,column number)
This will produce a sequence of numbers in the specified column using the first three parameters in a recurence relationship of the form u_n+1 = g u_n + a where u₁ is the initial value and g and a the geometric and arithmetic factors.
e.g. seq(1,2,1,7,4) will produce the sequence 1, 3, 7, 15, 31, . . . in column 4.

fib(initial value,second value,first term factor,second term factor,number of items,column number)
This will produce a sequence of numbers in the specified column using a recurence relationship of the form u_n+2 = a u_n + b u_n+1 where u₁ is the initial value, u₂ is the second value, a is the first term factor and b is the second term factor.
e.g. fib(1,2,1,1,20,1) will produce the Fibonacci sequence, 1, 2, 3, 5, 8, 13 . . ., for 20 terms in column 1.

function(function,start value,end value,step value ,x-data column number,function column number)
This will produce a sequence of numbers where the values in the column correspond to the X-data generated by the start value, the end value and the step value and the Function column contains the corresponding values of the function.
The function must be entered with 'x' as the variable using standard BASIC syntax.
e.g. function(x^2,1,8,2,1,3) will produce the values 1,3,5,7 in column 1 and the corresponding squares in column 3.

concat(column1,column2)
Extends the data in the first column by adding on the data from the second.
e.g. concat(3,5) will add the data in column 5 to that already in column 3.

The largest number of data creation commands are to produce sets of random numbers.

random_i(lower limit,upper limit,number,first column number, last column number)
This produces a set of random integers with a rectangular distribution between the specified limits in the designated block of columns.
e.g. random_i(1,6,50,3,5) will simulate 50 rolls of a die in columns 3 to 5.

random_r(lower limit,upper limit,number,first column number, last column number)
This produces a set of random real numbers with a rectangular distribution between the specified limits in the designated block of columns.
e.g. random_r(2.5,3.5,10,1,2) will generate 10 numbers in each of columns 1 and 2 with values between 2.5 and 3.5

random_n(mean,standard deviation,number,first column number,last column number)
This generates a set of normally distributed random real numbers with the specified standard deviation and mean in the designated block of columns.
e.g. random_n(20,2,50,3,5) will generate 50 numbers in columns 3 to 5 where the numbers in each column have a mean of 20 and a standard deviation of 2.

random_be(probability,number,first column number,last column number)
This generates a set of random ones and zeroes in the designated block of columns according to a Bernoulli distribution.
e.g. random_be(0.35,10,7,9) will produce a set of 10 0's and 1's in each of columns 7 to 9 where the probability of a 1 is 0.35.

random_bi(trials,probability,number,first column number,last column number)
This generates random integers with a Binomial distribution specified by the 'trials' and probability values in the designated columns.
e.g. random_bi(10,0.8,20,5,8) will produce 20 numbers in each of columns 5 to 8 where each value is the number of successes in 10 trials where the probability of a success was 0.8.

random_p(mean,number,first column number,last column number)
This generates random integers with a Poisson's distribution which represents the number of occurences with the specified mean within a unit interval.
e.g. random_p(10,100,1,2) will produce 100 random integers in each of columns 1 and 2 which have a mean value of 10 and a Poisson distribution.

random_x(mean,number,first column number,last column number)
This generates random integers with an exponential distribution which represents the number of intervals with the specified mean value between two events.
e.g. random_x(4,20,3,5) will produce 20 random integers with a mean of 4 in each of columns 3 to 5 with an exponential distribution.

 Downloads 

display(output device,first column number,last column number)
This will output the contents of the specified columns to either the screen or printer depending on the output device parameter, (screen or printer).
e.g. display(screen,1,6) will display the contents of columns 1 to 6 on the VDU.

A screen display has buttons to select the NEXT and PREVious pages, when applicable, and a PRINT button.
If the data can be identified as integer it is displayed with no decimal points. Otherwise, it is rounded to 4 d.p. no matter how many places there are.

The maximum size of a display page is 53 rows of 7 columns.

value(object name)
This will display on the screen the current value of the object concerned which can be a program variable or an element of data using the st[row:column] notation.
e.g. value(pp[2]) will display the current value of the second element of the pp[ ] array.

histogram(dimensions,type,steps,first column number,last column number)
This will display on the screen histograms of the data in the selected columns.
The acceptable values for the first two parameters are:-

e.g. histogram(2D,hist,6,1,3) will display a 2-dimensional histogram with three sets of bars representing the data in columns 1 to 3, split into 6 steps.

The display has a PRINT and CLOSE button.

N.B.3D displays can be rotated around their axes by holding down the CTRL key, left-clicking the mouse and dragging.

point_plot(axes,first column number,last column number)
This will display on the screen scattergrams of the data in the columns concerned.
e.g. point_plot(axes,1,3) will display scattergrams of the data in columns 1 to 3.

The 'axes' parameter, (axes or no_axes), defines whether axes are to be drawn for this plot.
Essentially, this allows you to overlay several sets of data by selecting 'axes' for an initial plot and then 'no_axes' for a subsequent plot or plots.

If you then use point_plot(no_axes,6,7), the plots for the data in columns 6 and 7 will be added to the display.

The display has a PRINT and CLOSE button.

Each set of data is plotted using a different symbol up to the maximum of 8 sets of data.

The maximum number of points which will be displayed is 30. If there are more than this in a selected column, a sample is taken rather than having an error message displayed.

If there are different number of points in a block of columns, it's the first column which is used to set the number for display.

graph(axes,markers,x-data column number,first y_data column number,last y_axis column number)
The data in the x-axis column is plotted against that of each of the columns in the block of y-axis data.
e.g. graph(axes,markers,1,2,4) will draw graphs with point markers, where the X-data is that in column 1 and there are three graphs corresponding to Y-data in columns 2 to 4.

The acceptable values for the first two parameters are:-

The axes ranges are calculated from the X-data column and the first of the Y-data columns. The actual range is rationalised to give ranges in simple numbers and the axes do not pass through the origin if this is not within the range of the data.

The X-data column can be included within the block but will be skipped as the graphs are drawn. e.g. graph(axes,markers,3,1,6) is acceptable: column 3 provides the X-data and graphs will be drawn using columns 1, 2, 4, 5 and 6 as the Y-data.

best_line(axes,type,x-data column number,first y_data column number,last y_data column number)
This will find the best line that fits the sets of data: the data in the x-axis column being plotted against that of the block of y-axis data.
e.g. best_line(axes,quad,1,2,4) will draw the best quadratics for the data sets where the X-data is that in column 1 and there are three graphs corresponding to Y-data in columns 2 to 4.

After each line is drawn, there is a display of the equation of the line concerned, the mean relative residual, (MRR), and the symbol used for the data points.
Each residual = data - function value and the MRR is defined as Σ(absolute residuals)/Σ(absolute data).

The acceptable values for the first two parameters are:-

The X-data column can be included within the block but will be skipped as the graphs are drawn. e.g. best_line(axes,linear,2,1,4) is acceptable: column 2 provides the X-data and graphs will be drawn using columns 1, 3 and 4 as the Y-data.

find_line(x-data column number,y_data column number)
This plots all of the available lines of best-fit for the data in the specified columns.

The mean relative residual for each line is stored and, if the graphs are printed, this information is added to the display as described above for the 'best_line' command and the minimum M.R.R. is listed as an aid in deciding which is the best line for this data.

A check is made at the start to see if the X or Y ranges include zeroes and, if they do, the two power graphs, y = a(b^x) and y = a(x^r), are excluded.
The first marker, , is used for the data and each line is drawn in a different colour.

e.g. find_line(1,2) will draw the four lines for the data in columns 1 and 2.

When all the graphs have been drawn, the equation of the one giving the smallest MRR is appended to the data display which, in this case, is the Fibonacci series!

 Downloads 

b_sum(first column number,last column number,results column number))
This adds the values across a row and stores the result in the 'results' column.
e.g. b_sum(1,3,4) will add the values in each row of columns 1 to 3 and store the result in the corresponding row of column 4.

c_sum(column number,label)
This adds the values in the designated column and stores the answer with the specified variable name.
e.g. c_sum(5,c) will add the values in column 5 and store the result with the program variable name, c.

diff(first column number,second column number,results column number)
This subtracts the values in the second column form those in the first and stores the results in the results column.
e.g. diff(6,3,7) will subtract the values in column 3 from those in column 6 and store the results in column 7.

inc(amount,column number,results column number)
This will add the amount to each item in the specified column and store the result in the results column.
e.g. inc(-1.25,1,2) will subtract 1.25 from each item in column 1 and store the results in column 2.

scale(amount,column number,results column number)
This will multiply each item in the specified column by the amount and store the results in the results column.
e.g. scale(0.87,3,5)) will multiply each item in column 3 by 0.87 and store the results in column 5.

b_min_max(first column number,second column number,minimum column number,maximum column number)
This finds the maximum and minimum values in the specified block of data and stored the results in the minimum and maximum columns.
e.g. b_min_max(1,4,5,6) will find the minimum and maximum values in each row of columns 1 to 4; storing the minima in column 5 and the maxima in column 6.

c_min_max(the column number,variable for minimum,variable for maximum)
This finds the maximum and minimum values in the designated column and stored the results in the specified variables.
e.g. c_min_max(3,pp[0],pp[1]) will find the minimum and maximum values in column 3 and store the results in the array values pp[0] and pp[1].

b_range(the first column number,the last column number,the results column)
This finds the range of values across each row of the designated block and stores the value in the results column.
e.g.
b_range(3,5,6) will find the range of values across a row from column 3 to column 5 and store the result in column 6.

c_range(the column number,the result variable)
This finds the range of values within the designated column and stores the result with the specified variable name.
e.g. c_range(4,b) will find the range of values in column 4 and store the result as b.

b_mean(the first column number,the last column number,the results column)
This finds the mean of the values across each row of the designated block and stores the value in the results column.
e.g. b_mean(1,4,5) will find the mean of the values across a row from column 1 to column 4 and store the result in column 5.

c_mean(the column number,the variable for the result)
This finds the mean of the values in the designated column and stores the result with the specified variable name.
e.g. c_mean(7,d) will find the mean of the values in column 7 and store the answer as variable d.

parsum(the first column number,the second column number)
This finds the partial sums, (cumulative totals), of the data in the first column and stores the results in the second column.
e.g. parsum(4,6) will find the partial sums of the data in column 4 and store them in column 6.

sigma(the first column number,the second column number,label)
This finds the sum of the products of the values in the specified columns and stores the result with the designated variable name.
e.g. sigma(1,5,d) will multiply each element in column 1 by the corresponding item in column 5 and store the total of the products with the label d.

replace(lower limit,upper limit,replacement,the column number,the results column number)
This checks the data in the first column and, if the value, v, is such that lower limit ≤ v ≤ upper limit, v is replaced by the replacement value in the results column. Otherwise, the original value is transferred to the results column.
e.g. replace(1.1,2.3,0,1,2) will transfer the data from column 1 to column2, replacing the value by zero if 1.1 ≤ value ≤ 2.3

s_d(the column number,label for the result)
This finds the Standard Deviation of the data in the designated column and stored the value with the specified variable name.
e.g. s_d(5,x) will find the Standard Deviation of the data in column 5 and store it as variable x.
N.B. This will produce a result for any set of data but will only have validity if the data is normally distributed.

dist(type,interval,the column number,the results column number,label for the initial value, (the minimum))
e.g. dist(c_d_f,10,4,5,d) will find the cumulative distribution of the data in column 4 over 10 steps and store the results in column 5 with the minimum value being stored as variable d.

This was developed as part of the 'histogram' command but may have some use in its own right.

The 'type' parameter takes the same values as in 'histogram'. The command will find the distribution of the data in the designated column and stores the results in the results column with the initial, (minimum), value stored with the specified label.

sort(the number of fields,the order,the first column number,the last column number,the first results column number,the key column numbers)
The command allows a block of data to be sorted on any number of key columns which can be used in any order.
e.g. sort(3,ascend,1,4,5,1,2,4) will sort the data in the block of columns 1 to 4 in ascending order using as the 3 key fields column 1, then column 2 and finally column 4. The sorted data is stored in a block starting at column 5.

The 'order' parameter takes the values ascend or descend.

If the sorted data is to replace the original set, then the results column number is made the same as the intial.
e.g. sort(3,ascend,1,4,1,1,2,4) for the example shown above.

regress(type,the X-data column number,the Y-data column number)
A form is displayed which allows the entry of X-values and displays the corresponding Y-value on the basis of regression onto the designated line.

e.g. regress(quad,1,2) will find the quadratic of best fit to the data in columns 1 and 2 and display a form for the entry of X-values.

The 'type' parameter takes the same values as in 'best_line'.

round(type,number of figures,the source column number,the results column number)
The items in the source column are rounded according to the specification and stored in the results column.

The 'type' parameter takes the values 'dp' or 'sf'.

e.g. round(sf,3,1,2) will round the data in column 1 to 3 significant figures; storing the results in column 2.

 Dowloads 

Several significance tests can be carried out using the basic commands, e.g. those involving means, so those included as specific commands are the ones which would be difficult to do via a series of arithmetic commands.

Probability and distribution tables will have to be used as it isn't thought to be efficient to tie up your PC's memory with pages of tables.

The tests will give results for any set of data but they will only have significance if the data is of a nature to which the test can be applied.

correlate(first column number,last column number,label))
This finds the product-moment correlation coefficient between the data in the designated columns and stores the result with the specified variable name.
e.g. correlate(3,4,x) will find the product-moment correlation coefficient between the data in columns 3 and 4 and store it as variable x.

chi_test(observed frequency column number,expected frequency column number,label))
This carries out a χ² test on the data in the first column by comparison with that in the second. The statistic, χ², is stored with the specified variable name.
e.g. chi_test(1,2,x) will compare the data in column 1 with the expected frequency of this data in column 2 and store the coefficient as variable x.

t_test(column number,quoted mean,label))
This carries out the 'Student's' T-test on the data in the column, which is compared with the quoted mean. The result is stored with the specified variable name.
e.g. t_test(1,2.3,c) will compare the data in column 1 which may be part of a set which has a mean value of 2.3; storing the result as variable c.

m_w_test(tails,end,first column number,second column number,label))
This carries out a Mann-Whitney test on the data where the acceptable values for 'tails' and 'end' are shown below.

e.g. m_w_test(one,high,1,2,a) will carry out a single-tailed test on the high end of the distribution of the data in columns 1 and 2; storing the result as the variable a.

Even though the 'end' parameter has no meaning in a two-tailed test, a value must be entered to avoid a syntax error. In fact, the algorithm assumes a symmetrical distribution, carries out the single-tailed test selected and multiplies the result by 2!

 Downloads 

There is a limited set of programing commands which can be used to automate the use of STATS commands.

 PROGRAM STATEMENTS 

The application has a number of in-built numeric variables which can be used to store the results of operations or in programs.

These are:-

In the last two cases it is essential to use square brackets, [ ], rather than round ones.

The program also allows you to use pi, (p), and e

 Downloads 

It may be that my references to 'BASIC syntax' might not mean much to some people.

It used to be that every small computer came equipped with BASIC but, as this has not been the case for some years now, I felt it might be useful to give a brief outline of the BASIC syntax you need to use to enter equations for the 'function' command or to create expressions for a program or command.

BASIC is an acronym for Beginners All-purpose Symbolic Instruction Code and it is the language that most of us over cut our programming teeth on. Over and you were likely to have started programming in binary via a set of switches!

The operators and functions which you can use are listed below. They are all case sensitive.

 OPERATORS 

* The results of logical operations return -1 for True and 0 for False.

† These operators can be applied to values as well as conditions.
Hover over the reference symbols for examples.

‡ These operators can generate 1's from zeroes. As all of the bits in the number/s are processed, (32 bits in the case of integers), any leading zeroes in the values used will generate 1's and so the result is likely to be a negative number.

Hover over the symbols for examples.

 FUNCTIONS 

With BASIC, the main rule is that everything has to be stated explicitly. e.g. we all know that 2x means "Two multiplied by x" but this has to be written in full as '2*x' in BASIC.

A second point is that, as all parts of an equation have to be entered in line, far more parentheses are needed than is the case with written algebra. An equation which would usually be written as "1 over 1 + x" has to be entered as 1/(1+x).

Whilst the following is a rule of basic arithmetic not BASIC, a common mistake is to misinterpret expressions such as 2 + 5 x 3 and expect the result '21' when, in fact, it's 17. [21 = (2 + 5) x 3].

 Downloads 

Whilst being OK for the occasional macro, VBA has its limits for full projects. The syntax is more limited than that of VB and I've found stability problems with Excel VBA. However, Excel is a very useful data storage medium and the Excel classes were designed to allow this use from within a VB based application. However, I'm in the process of replacing Excel with LibreOffice Calc, (see the LibreOffice VB6 section), and may re-write these classes to work with LO.

There are two classes, ExcelApplication and ExcelWorkbook.

This opens Excel if necessary and creates a collection for ExcelWorkbook objects. Ony one ExcelApplication object can exist at a time.

This allows workbooks to be opened and manipulated. Books are defined as 'With Events' so that use can be made of Excel events.

The class has the following methods and properties.

The class contains a procedure for only one event, 'xlBk_SheetActivate', which calls a main program procedure, 'select_sheet_action', whenever a new sheet is selected. However, it is easy to add code for other events using these as templates.

 Downloads 

These classes are designed to allow the layout, preview and printing of text from within a VB application.

There are two classes,

PPpages, which is used to create a set of pages and deals with operations which apply to the WHOLE SET, and

PPpage, which deals with the methods and properties applicable to a SINGLE page.

 PPpages 

This class has the following Methods and Properties,

† These methods exist to be accessed by the Preview Form and should not otherwise be used.

The units used for page sizes can be set to either inches or mm. by means of a flag variable.

 PPpage 

This class has the following Methods and Properties,

† This method exists only to transfer page data from a PPpages object and should not otherwise be used.

add_text is a basic text display routine. It will add a line of text formatted Left, Right or Centre but the text is truncated if longer than a line width.

add_f_text is a more versitile text display routine which will wrap excess text onto a new line, allow Full justification as well as Left, Right and Centre and will recognise the following, in-line font formating codes,

* More detail on the operation of these methods is given here.

 Downloads 

 Basic addition of text 

add_text(tx, hp, tm, ByRef p_d, Optional vp, Optional alg)

If the text is longer than the available page width as defined by the vertical margins and any horizontal offset set by 'hp', an error message is displayed but as much of the line will be printed as will fit on the page.

The line terminator flag is a Boolean which will terminate the text line if True. Setting it False allows further text to be added to the same line.

The alignment setting is really of little use! If you're going to add extra text on the same line, i.e. the termination flag is clear, only Left alignment can be allowed as is the case if there is already some text on the line, i.e. 'a non-zero 'hp'.

The vertical parameter, 'vp', which is set to default to zero, is an addition to the current printing depth, 'p_d', which can be used to add line spacing.

All text formating has to be carried out using the 'font' or 'set_font' methods.

add_f_text(tx, Optional p_d = 0, Optional tm As Boolean = False, Optional alg = -1, Optional p_l As Single = 0,
Optional vp As Single = 0, Optional h_p As Single = 0)

If the printing depth is defaulted, the current internal value will be used. If a non-zero value is entered for the parameter, the internal value is reset to this figure.

The line terminator flag is a Boolean which will terminate the line at the end of the text block if True. The default of False allows further text to be added to the same line, (See PART LINE, below).

The alignment, 'alg', defaults to -1 which means that the current page alignment is used. This is either the default of Left or that set by the 'set_align' method.

The PART LINE value is used to return the position of the text on the line at the end of a block which can then be used to complete this line with further text, provided that the termination flag was set to False. N.B. If full justification is used, the first part of such a line, (i.e. the last part of the previous section), will NOT be justified.

The vertical parameter, 'vp', which is set to default to zero, is an addition to the current printing depth, 'p_d', which can be used to add line spacing. i.e. setting it to the line height will give double line spacing.

The horizontal offset, 'h_p', is added as in INDENT from the LEFT margin. If LEFT or FULL JUSTIFICATION is used with a non-zero PART LINE value, this offset is ignored. If a NEGATIVE value is used, this is interpreted as an indent from both the LEFT and RIGHT margins.

If your text requires a 'back-slash' character to be displayed, just enter two, e.g. \\ There is no problem in using forward slashes.

Codes are sequential and are only reset by \n/. i.e. if you set text to italic, \i/, and later to bold, \b/, your display will be both italic and bold. To have bold without italic, you'd have to use \nb/.

The use of 'y' to select the Symbol font depends on that font being available on the machine. This is an unknown for distributed software but most machines will have 'Symbol'. The facility exists for Mathematical/Scientific text which may require these symbols and further symbols can be had by the use of code 'w' for 'WingDings' which should be available on all machines.

The text printing routines will recognise the VB constants, vbTab, vbCr, vbLf and vbCrLf although the last three will start the text for the following line at the far left of the page, (NOT at the Left margin), and the 'terminate' command is a better way to start a new line. Special characters are best entered from the keyboard using ALT plus the three or four digit code, e.g. © is given by ALT plus 0169.

The inclusion of a backspace code, 'k', is to allow the production of vertically aligned subscripts and superscripts for Maths and Science text.
e.g. which was produced by "\y/m\ds/3\kp/n\n/" but which would now be better done as "μ\s/3\kp/n\n/"
The QB colour codes are as shown below. The colour descriptions are weird! i.e. 'Red' is a greyish red whilst 'Light Red' is red. 'White' is pale grey and 'Bright White' is white.

VisualBasic's default text display is positioned at the TOP and LEFT of the character bounding boxes which is not what we want so the 'SetTextAlign' API call is used to vertically position characters at the BOTTOM of the character as a Word Processor would display them.

These text addtion routines DON'T constitute a word processor. e.g. if text is displayed and then the margins or other page properties changed, the existing text display isn't altered; only that which follows the changes in your 'create_page' routine.

I'm considering the possibility of wrapping text around a graphic.

add_graphic(g_path As String, Optional gw As Single = 0, Optional gh As Single = 0, Optional hh As Integer = -1, Optional vv As Integer = 0)

If the width, 'gw', is defaulted to zero, the width is set to that of the image. This may be too big to fit on a page and so an error message will be displayed.

If 'gh' is set to its default, the graphic height is set to keep the proportions of the original, which will be the usual situation. Therefore, defaulting both 'gw' and 'gh' will display the graphic at its original size.

The negative values for fractions of the page width and height are required as fractional unit values might be used when the units are set to inches.

One routine is used for both operations.

head_foot(flg, Optional ltx As String = "", Optional ctx As String = "", Optional rtx As String = "", Optional f_nm As String = "", Optional f_sz As Single = 0, Optional f_mb As Boolean = False))

Any of the three items of text can be omitted as can the font data. In the latter case, the default settings are used.

Specialist data can be displayed by using the following codes,

No check is made to see if the three pieces of text overlap!

There is also no check made that your specified footer margin won't push the depth of the text off the bottom of a page. It isn't possible to do this because it isn't known what size of font you're going to use for the footer text. In 'preview' mode, too great a depth for the footer will just result in it vanishing. However, when printing, this can lead to a situation where tens of printed pages will be generated, so beware!

 Downloads 

The standard MsgBox and InputBox displays are a little limited. The font is fixed and the colours are the Windows colours and, whilst an Input box can be positioned, you can't place a MsgBox.

Whilst these are not major problems, these classes give a little more scope.

There are two classes, MsgBx and mbx.

 MsgBx  This stores general parameters which apply to all boxes such as font and colour.

 METHODS .

 mbx  This produces the boxes.

 METHODS 

The class doesn't provide the 'Context' or 'Help' attributes of the standard functions.

 Downloads 

Evaluation functions are ubiquitous in scripts but rare in compiled languages because it's far easier to carry out the evaluation of a string with an interpreter. This creates problems if you're writing an interactive application, particularly a mathematical one, as you're liable to want the user to enter equations, etc.

This class overcomes that problem by providing a method for a VB application to evaluate a string.

It's based on the published Microsoft code with additions to allow the recognition of more variables and functions.

The class is called eval which has a single method of the same name, eval.

The class recognises the variables a, b, c, d, l, m, x and y, (although only x is available in the demonstration), and the constants pi and e as well as the usual operators, trig. and inverse trig. functions, the hyperbolic functions, log₂, (log2), and log₁₀, (log10).

 Downloads 

There are reports of problems using the CommonDialog control with different platforms and that Microsoft has declared it to be 'unsafe'!

There are certainly problems when trying to use it with Office 2003, VBA applications running under XP.

Several codings are available to replace the CD control with API calls and this class is based on code from a number of sources.

The CDC_class class has the following self-explanitory methods,

ShowOpen,

ShowSave,

ShowPrinter,

ShowColor,

ShowFont,

ShowPageSetupDlg

It has one property, FontData, which can be used to return the selected font size, colour, etc.

The demonstration allows the use of each of the methods.

A complete set of constants is included for all aspects of the API calls even 'though some aren't used by the existing code. I found some values so difficult to find that I decided to include them all.

The ShowPrinter method applies the specified data to the selected printer: the other methods require the main program to do this.

A modified version of the class, CDC_Excel_class, is included as a text file for use with VBA code but this is not recommended as there is little control over a printer from VBA as a Printer object can't be defined. This is one of the reasons for me switching to LO rather than MS Office

 Downloads 

The sort class has code for six sorting algorithms and has the following methods,

The main demonstrations allow the production of either a set of integer data or string data which can be sorted and timed by each of the methods.

The multi-field, Shell sort could be adapted for any of the other algorithms.

The multi-field demonstration generates an integer or string data set with up to ten fields which can be sorted on any selection of these ten, taken in any order by the MF_srt method.

There is also an option to sort names according to the UK English conventions and a data file, name.dta, to demonstrate this.

To allow for different types of data, the code is written using variant variables but could be re-dimensioned to fit specific data types so as to reduce memory usage and speed up the sorts.

 Downloads 

This is even more esoteric than the 'Message Box' class but, if you write Help files, it can be useful to have a standardised 'Help About' form which doesn't need any programming other than defining its colour and graphics, as it picks up its information from the package properties.

It would be easy to extend the customisation of the form to include font name and size.

It contains two versions of the About class code; 'about.cls' which provides memory information for 32bit systems and 'about_64.cls' for 64 bit systems.

 Downloads 

The Graphic size class allows the Width, Height and, (in most cases), the colour depth of a graphic file to be read.

The original class was created by David Crowell, (davidc@qtm.net, http://www.qtm.net/~davidc), although these links are no longer active.

The file formats it can detect are:-

.BMP, .JPG, .GIF, .TIF, .PNG, .ICO_[1], .PCX, .PSD, .PSB, .TGA_[2] and the Adobe group _[3], .EPS, .PS and .AI

The class has the following Method and Properties,

ReadImageInfo, which reads the image information.

ImageType, which gives one of the following as the Image type code.

• itUNKNOWN = 0
• itGIF = 1
• itJPEG = 2
• itPNG = 3
• itBMP = 4
• itICO = 5
• itTIFF = 6
• itPCX = 7
• itTGA = 8
• itPSD = 9
• itPSB = 10
• itEPS = 11
• itPS = 12
• itAI = 13

Width,The image width in pixels.

Height,The image height in pixels.

Depth,The colour depth in bits/pixel.

 Downloads 

Visual Basic is designed to be used with other MS applications such as Excel by virtue of the addition of various references.

However, no such facilities are provided for free Office packages such as OpenOffice or LibreOffice.

This section provides information on how to use VB6 with LibreOffice and a set of procedures and functions based on this information. Whilst these are written for VB 6.0, as this is all I have, I imagine that they can easily be translated to the versions released after 1998 and the various Net versions.

The package consists of the set of files listed below. The VB6 project hasn't been compiled so that you can make your own edits,

The files can be downloaded as a .ZIP document, (LO_VB6.zip, 242K)