/ REXX / / ========================================================================== / / BXC - Benford Analysis Tool (Classic REXX for ArcaOS 5.1.1) / / Original BASIC version by Jason S. Page / / Enhanced Classic REXX version for ArcaOS / / ========================================================================== /

/ -------------------------------------------------------------------------- / / Robust RexxUtil Loading / / -------------------------------------------------------------------------- / if RxFuncQuery('SysFileExists') then do call RxFuncAdd 'SysLoadFuncs', 'RexxUtil', 'SysLoadFuncs' call SysLoadFuncs end

if RxFuncQuery('SysFileExists') then do say "Error: Could not load REXXUTIL.DLL functions (SysFileExists)." say "Please ensure OS/2 RexxUtil is installed and in your LIBPATH." exit 99 end

/ -------------------------------------------------------------------------- / / Initialize Benford's Law Constants / / -------------------------------------------------------------------------- / numeric digits 12

benford.0 = 9 benford.1 = 30.1 benford.2 = 17.6 benford.3 = 12.5 benford.4 = 9.7 benford.5 = 7.9 benford.6 = 6.7 benford.7 = 5.8 benford.8 = 5.1 benford.9 = 4.6

/ -------------------------------------------------------------------------- / / Initialize Global Variables / / -------------------------------------------------------------------------- / toke = 0 animate = 0 animate_interval = 100 total_processed = 0 block_char = 'DB'x

/ -------------------------------------------------------------------------- / / Parse Command Line Arguments / / -------------------------------------------------------------------------- / parse arg params

load1 = "" a12 = "" prog = "" col = "" anim = ""

call ParseFlag "-f", params load1 = result

call ParseFlag "-d", params a12 = result

call ParseFlag "-l", params prog = result

call ParseFlag "-c", params col = result

call ParseFlag "-a", params anim = result

/ -------------------------------------------------------------------------- / / Show Help if Requested / / -------------------------------------------------------------------------- / if pos("-h", params) > 0 | pos("--help", params) > 0 then do call ShowHelp exit end

/ -------------------------------------------------------------------------- / / Handle Animation Flag / / -------------------------------------------------------------------------- / if length(anim) > 0 then do animate = 1 if datatype(anim, 'W') then animate_interval = anim end

/ -------------------------------------------------------------------------- / / Interactive Mode if Missing Required Arguments / / -------------------------------------------------------------------------- / if length(load1) = 0 | length(a12) = 0 | length(prog) = 0 | length(col) = 0 then do say "Error: Missing required flags." say "" say "Interactive mode:"

call SysFileTree "*.dat", "fileList", "FO"
do i = 1 to fileList.0
    say "Found: " || fileList.i
end

call charout , "Load file: "
parse pull load1

call charout , "[A]ll Digits or [1]st Digit: "
parse pull a12

call charout , "Capture Average Every (default: 10000) Points: "
parse pull prog

call charout , "Which Column Number (0 = single column): "
parse pull col

call charout , "Enable animation? [y/n]: "
parse pull anim_choice

if translate(left(anim_choice, 1)) = "Y" then do
    animate = 1
    call charout , "Update interval (default: 100): "
    parse pull anim_input
    if datatype(anim_input, 'W') then animate_interval = anim_input
end

end

/ -------------------------------------------------------------------------- / / Set Defaults / / -------------------------------------------------------------------------- / if length(prog) = 0 then prog = "10000" if length(col) = 0 then col = "0" if col = "0" then col = "1"

/ -------------------------------------------------------------------------- / / Display Configuration / / -------------------------------------------------------------------------- / say "" say "Configuration:" say " File: " || load1 say " Digits: " || a12 say " Sample size: " || prog say " Column: " || col if animate = 1 then say " Animation: Enabled (update every " || animate_interval || " records)" else say " Animation: Disabled" say ""

/ -------------------------------------------------------------------------- / / Download File if URL / / -------------------------------------------------------------------------- / if left(translate(load1), 3) = "FTP" | left(translate(load1), 4) = "HTTP" then do say "Downloading file..." address CMD "wget -c " || load1 || " 2>&1"

last_slash = lastpos('/', load1)
if last_slash > 0 then load1 = substr(load1, last_slash + 1)

say "Assuming downloaded file is: " || load1

end

/ -------------------------------------------------------------------------- / / Verify File Exists / / -------------------------------------------------------------------------- / if \SysFileExists(load1) then do say "Error: File '" || load1 || "' not found!" exit 1 end

/ -------------------------------------------------------------------------- / / Create Log Files / / -------------------------------------------------------------------------- / logvalues = load1 || "" || a12 || "-" || prog || "-.log" logpercent = load1 || "" || a12 || "-" || prog || "_.log"

call lineout log_values, , 1 call lineout log_percent, , 1

/ -------------------------------------------------------------------------- / / Initialize Counters / / -------------------------------------------------------------------------- / c. = 0 count_total = 0

/ -------------------------------------------------------------------------- / / Display Header / / -------------------------------------------------------------------------- / say "" say "========================================================================" say "Benford X-C Forensics Digital Analysis Tool (ArcaOS 5.1.1)" say "========================================================================" say ""

if animate = 0 then do say " Record, 1, 2, 3, 4, 5, 6, 7, 8, 9" end

/ -------------------------------------------------------------------------- / / Main Processing Loop / / -------------------------------------------------------------------------- / infile = load1

do while lines(infile) > 0 line = linein(infile) toke = toke + 1 total_processed = total_processed + 1

if length(line) = 0 then iterate

ot = line

/* Extract column if needed */
if col > 1 then do
    call ExtractColumn line, col
    ot = result
end

/* Clean to digits only */
call ExtractDigits ot
ot = result

if length(ot) = 0 then iterate

/* Determine position based on digit mode */
if translate(left(a12, 1)) = "A" then
    position = length(ot)
else
    position = 1

/* Count digits */
do i = 1 to position
    digit = substr(ot, i, 1)
    if digit >= 1 & digit <= 9 then do
        count_total = count_total + 1
        c.digit = c.digit + 1
    end
end

/* Animate if enabled */
if animate = 1 & (total_processed // animate_interval = 0) & count_total > 0 then do
    call DrawAnimatedChart count_total, total_processed, prog
end

/* Write sample when reaching sample size */
if count_total >= prog then do
    p. = 0
    do i = 1 to 9
        if count_total > 0 then
            p.i = format((c.i * 100) / count_total, 2, 2)
        else
            p.i = 0
    end

    cc1 = ot || "," || toke || "," || c.1 || "," || c.2 || "," || c.3 || "," || c.4 || "," || c.5 || "," || c.6 || "," || c.7 || "," || c.8 || "," || c.9
    cc2 = ot || "," || toke || "," || p.1 || "," || p.2 || "," || p.3 || "," || p.4 || "," || p.5 || "," || p.6 || "," || p.7 || "," || p.8 || "," || p.9

    if animate = 0 then do
        say "#: " || cc1
        say "%: " || cc2
    end

    call lineout log_values, cc1
    call lineout log_percent, cc2

    /* Reset counters */
    count_total = 0
    do i = 1 to 9
        c.i = 0
    end
end

end

/ -------------------------------------------------------------------------- / / Close Files / / -------------------------------------------------------------------------- / call stream infile, 'C', 'CLOSE' call lineout log_values call lineout log_percent

if animate = 1 then say ""

/ -------------------------------------------------------------------------- / / Display Completion Message / / -------------------------------------------------------------------------- / say "" say "========================================================================" say "Analysis Complete" say "========================================================================" say "Total records processed: " || toke say ""

say "Calculating final statistics and chi-squared test..." call GenerateFinalReport log_percent, load1, a12, prog

say "" say "Output files:" say " Values log: " || log_values say " Percent log: " || log_percent say ""

exit 0

/ ========================================================================== / / Subroutines / / ========================================================================== /

/ -------------------------------------------------------------------------- / / ShowHelp - Display usage information / / -------------------------------------------------------------------------- / ShowHelp: say "==================================" say "bxc - Benford Analysis Tool (ArcaOS)" say "==================================" say "Usage: bxc -f [file] -d [1|all] -l [length] -c [column] -a [interval]" say "" say "Required flags:" say " -f [file] Data file to analyze (or URL)" say " -d [1|all] Analyze first digit (1) or all digits (all)" say " -l [number] Sample pool length (default: 10000)" say " -c [number] Column number (0 for single column)" say "" say "Optional flags:" say " -a [interval] Enable animated graph (default: 100)" say " -h, --help Show this help message" say "" say "Example:" say " bxc -f data.dat -d 1 -l 10000 -c 1" return

/ -------------------------------------------------------------------------- / / ParseFlag - Extract flag value from command line / / -------------------------------------------------------------------------- / ParseFlag: procedure parse arg flag, cmdline result = "" ppos = pos(flag, cmdline) if ppos > 0 then do start = ppos + length(flag) + 1 rest = substr(cmdline, start) parse var rest val " -" rest_ignored result = strip(val) end return result

/ -------------------------------------------------------------------------- / / ExtractColumn - Extract specified column from CSV line / / -------------------------------------------------------------------------- / ExtractColumn: procedure parse arg line, col_num

if col_num <= 1 then return line

loop_col = 1
temp_line = line
found_val = ""

do while loop_col <= col_num
    parse var temp_line this_val "," temp_line
    if loop_col = col_num then found_val = this_val
    loop_col = loop_col + 1
end

return found_val

/ -------------------------------------------------------------------------- / / ExtractDigits - Extract only numeric digits from string / / -------------------------------------------------------------------------- / ExtractDigits: procedure parse arg text

clean = ""
do i = 1 to length(text)
    char = substr(text, i, 1)
    if datatype(char, 'N') then clean = clean || char
end

return clean

/ -------------------------------------------------------------------------- / / DrawAnimatedChart - Display live animated chart / / -------------------------------------------------------------------------- / DrawAnimatedChart: procedure expose c. benford. block_char parse arg total_seg, total_proc, sample_size

call SysCls

say "========================================================================"
say "Benford X-C Live Analysis - Animated View"
say "========================================================================"
say "Processing: " || total_proc || " records | Sample pool: " || sample_size || " | Current: " || total_seg
say "------------------------------------------------------------------------"
say ""
say "Digit  Actual   Expected  Deviation  Chart"
say "-----  -------  --------  ---------  ---------------------------------"

do i = 1 to 9
    if total_seg > 0 then
        pct = (c.i * 100) / total_seg
    else
        pct = 0
    call PrintDigitRow i, pct, benford.i
end

say "========================================================================"
return

/ -------------------------------------------------------------------------- / / PrintDigitRow - Display single digit row in chart / / -------------------------------------------------------------------------- / PrintDigitRow: procedure expose block_char parse arg digit, actual, expected

deviation = actual - expected
bar_len = actual / 2
if bar_len > 50 then bar_len = 50

bar = copies(block_char, trunc(bar_len))

s_actual = format(actual, 2, 1) || "%"
s_expect = format(expected, 2, 1) || "%"

sign = " "
if deviation >= 0 then sign = "+"
s_dev = sign || format(deviation, 2, 1) || "%"

say "  " || digit || "    " || left(s_actual, 7) || "  " || left(s_expect, 8) || "  " || left(s_dev, 9) || "  " || bar
return

/ -------------------------------------------------------------------------- / / GenerateFinalReport - Create final statistical report / / -------------------------------------------------------------------------- / GenerateFinalReport: procedure expose benford. block_char parse arg percent_file, data_file, mode, samp

sum. = 0
count = 0

/* Check if file exists and has content */
if \SysFileExists(percent_file) then return

/* Read and calculate averages */
call stream percent_file, 'C', 'CLOSE'

do while lines(percent_file) > 0
    line = linein(percent_file)
    if length(line) = 0 then iterate

    parse var line toss "," recd "," p.1 "," p.2 "," p.3 "," p.4 "," p.5 "," p.6 "," p.7 "," p.8 "," p.9

    do i = 1 to 9
        sum.i = sum.i + p.i
    end
    count = count + 1
end
call stream percent_file, 'C', 'CLOSE'

if count > 0 then do
    chi_squared = 0

    say "Final Statistics (Averaged across all samples):"
    say "------------------------------------------------"
    say "Digit  Actual   Expected  Deviation"
    say "-----  -------  --------  ---------"

    do i = 1 to 9
        avg = sum.i / count
        dev = avg - benford.i
        chi_part = (dev * dev) / benford.i
        chi_squared = chi_squared + chi_part

        s_avg = format(avg, 2, 1) || "%"
        s_exp = format(benford.i, 2, 1) || "%"
        s_dev = format(dev, 2, 1) || "%"

        say "  " || i || "    " || left(s_avg, 7) || "  " || left(s_exp, 8) || "  " || s_dev
    end

    say ""
    say "Chi-Squared Statistic: " || format(chi_squared, 4, 4)
    say ""

    /* Interpret results */
    if chi_squared < 15.51 then do
        say "Result: Data FITS Benford's Law (95% confidence)"
        say "        No significant deviation detected."
    end
    else do
        say "Result: Data DOES NOT fit Benford's Law (95% confidence)"
        say "        Significant deviation detected - possible fraud indicator!"
    end

    say ""
    say "Generating ASCII chart..."
    call GenerateChart percent_file
end
return

/ -------------------------------------------------------------------------- / / GenerateChart - Create ASCII chart output file / / -------------------------------------------------------------------------- / GenerateChart: procedure expose block_char parse arg percent_file

chart_file = "chart_" || percent_file || ".txt"

call lineout chart_file, , 1

call stream percent_file, 'C', 'CLOSE'
sample_count = 0

do while lines(percent_file) > 0
    line = linein(percent_file)
    if length(line) = 0 then iterate

    parse var line toss "," recd "," p.1 "," p.2 "," p.3 "," p.4 "," p.5 "," p.6 "," p.7 "," p.8 "," p.9

    sample_count = sample_count + 1
    call lineout chart_file, "========================================================================"
    call lineout chart_file, "Sample #" || sample_count || " | Record: " || recd || " | Value: " || toss
    call lineout chart_file, "------------------------------------------------------------------------"

    do i = 1 to 9
        call WriteChartLine chart_file, i, p.i
    end
    call lineout chart_file, ""
end

call stream percent_file, 'C', 'CLOSE'
call stream chart_file, 'C', 'CLOSE'

say "ASCII chart saved to: " || chart_file
say ""
say "Preview (first sample):"
call DisplayHead chart_file, 15
say ""
say "Preview (last sample):"
call DisplayTail chart_file, 12
return

/ -------------------------------------------------------------------------- / / WriteChartLine - Write single chart line to file / / -------------------------------------------------------------------------- / WriteChartLine: procedure expose block_char parse arg file_name, digit, percentage

bar_len = percentage
if bar_len > 100 then bar_len = 100

bar = copies(digit, trunc(bar_len))

line = "  " || digit || " | " || format(percentage, 2, 1) || "% | " || bar
call lineout file_name, line
return

/ -------------------------------------------------------------------------- / / DisplayHead - Display first N lines of file / / -------------------------------------------------------------------------- / DisplayHead: procedure parse arg file, num_lines

cnt = 0
call stream file, 'C', 'CLOSE'

do while lines(file) > 0 & cnt < num_lines
    say linein(file)
    cnt = cnt + 1
end

call stream file, 'C', 'CLOSE'
return

/ -------------------------------------------------------------------------- / / DisplayTail - Display last N lines of file / / -------------------------------------------------------------------------- / DisplayTail: procedure parse arg file, num_lines

/* Read entire file into stem */
call stream file, 'C', 'CLOSE'

fileContent.0 = 0
do while lines(file) > 0
    idx = fileContent.0 + 1
    fileContent.idx = linein(file)
    fileContent.0 = idx
end

call stream file, 'C', 'CLOSE'

/* Display last N lines */
start_line = fileContent.0 - num_lines + 1
if start_line < 1 then start_line = 1

do i = start_line to fileContent.0
    say fileContent.i
end
return