In order to reduce the chance of transcription errors, my approach was to prove the peals as I went along, using MicroSiril, which was also used to generate output that could be inserted straight into the html file. This is a bit tedious, since stedman does not lend itself to a general-purpose proving program, and the results, although acceptable, are far from ideal.
Having processed about forty compositions by this means, I noticed that I was short of peals by Paul Mounsey. Following a request for some, he furnished me with an envelope stuffed full, and it was at this point that I decided I needed something to make my job a bit easier.
The result is SPAFT, which was designed with the following requirements in mind:
If the touch does not come round, the normal reason will be an error in the input (which should be apparent from the course-heads), so details of falseness are irrelevant. If it does come round, and there is any falseness, the six-ends of the courses containing at least the first repeated row should be output, with the false sixes highlighted.
I think I have more or less succeeded with these aims. The software is available here as a 32-bit widows version for download. This requires libraries msvcp50.dll and msvcrt.dll. If you need these, then download and unzip redist.exe.
Alternatively, send an for the source code, which uses the Standard Template Library (STL).
The program runs from the command line, and takes a single input parameter - the filename. Output is written to the standard output stream, and can be redirected to a file if required, e.g:
spaft peal.in > peal.outIf no input fileneame is specified, input is taken from standard input. This means that the microSIRIL interface can be used.
An Example input file is as follows:
{ 5009 by PABS }
"by Philip A B Saddleton" {title}
#11 { Cinques (default, so not required) }
!2q { Start from 2nd row of a quick six (default !4q) }
<6.19> { Displayed calling positions (optional) }
{ white space, stops and new lines interchangeable separators }
a[1.5.7.8.10.11.13.14.15.16.17 (20)] { course length defaults to 2n }
A[6.19 19 6.19 19 19 6.19 19 6.19 19]
b[2.5.6.13.14.15.16.19]
B[6.19 19 19 6.19 19 19 6.19 19]
2.13s.15.16.19
B
_ {extra underline}
b A
This should be fairly self-explanatory when compared with the
output, which is then:
5009 Stedman Cinques by Philip A B Saddleton 3124567890E 6 19 ----------------- 12546378E90 (a) ----------------- 5241 - 2 | 423165 - 3 |A 3214 - 2 | 1243 - 2 | ----------------- 2145638709E (b) ----------------- 413562 - 3 | 312564 - 3 |B 2153 - 2 | ----------------- 12456378E90 2.13s.15.16.19 125364 B ----------------- 2154638709E (b) 214365 A ----------------- (a) = 1.5.7.8.10.11.13.14.15.16.17 (20 sixes) (b) = 2.5.6.13.14.15.16.19 Start from rounds as the second row of a quick sixHere are a few other examples:
4284 Stedman Caters by Thomas Hooley (incorrect version taken from Comp 501) 123456789 1 5 6 8 15 ------------------------ 241356978 - s - | 4321 s - | 3142 s - | 1324 s s - |A 3412 s - | 4231 s - | 2143 s - | ------------------------ 2314 - 214365897 (a) 132564789 - - - - ------------------------ 3152 s s - | 123465 - - - | ------------------------ | 241365978 - s - | | 4321 s - | | 3142 s - | | 1324 s s - | | 132465978 | | 341527689 | | 345718296 | | 473859162 | | 478936521 | | 794682315 | | s 796243851 * | | 927365418 | |B 923571684 |A | 259138746 | | 251894367 | | 582416973 | | 584627139 | | 865743291 | | 867359412 | | - 638974512 | | 639481725 | | 346192857 | | 341265978 | | 3412 s - | | 4231 s - | | 2143 s - | | ------------------------ 214365978 - 132549678 135927486 391758264 397816542 - 983675142 986534721 859462317 854291673 528147936 521783469 275316894 273659148 762934581 769428315 * - 647893215 648371952 436185729 431562897 431562897 - - - 142365789 B 431562987 (b) 142365798 B 123456789 (c) ------------------------ (a) = 1.4.7s.10s.12s (16 sixes) (b) = 1.5.6.7s.15.18 (c) = 1s.2.4.9s.10.11.13 (14 sixes) Start from rounds as the last row of a quick six 150 repeated rows Rows 1750 and 2119 are both 679243851 Rounds occurs after 4284 changes
I am not the first to go down this path. Nearly 30 years ago
Frank King produced a program called SORE (Stedman or Erin), for
which the emphasis was on a fairly natural input style. However
the aims were somewhat different, with a priority on memory and
speed requirements that are hardly necessary today.
Following a request from Richard Allton, I have produced a similar program to prove Grandsire. Most of the code is the same, but because of the way the input is parsed it was simpler to create a new program than make it another input option. Download GRAFT.
Input is as you might expect by analogy to SPAFT, but a few points may need clarification:
COMPBASE allows both SPAFT and GRAFT definitions to be contained in a single database, along with the ability to call external programs. Any further development of SPAFT and GRAFT will be to this software, which I hope will eventually expand to include other types of composition (e.g. twin-bob stedman, or spliced). There is a rudimentary help included. See the sample input for examples. In order to run microsiril, download CBPROVE.BAT and copy PROVE.COM to the working folder.
Download compbase.zip
This page created by Philip Saddleton
Last updated 27 April 2004