#! /usr/bin/python3

import sys, re
import html_gen as h
from process_funcs import bash
import html_report_funcs as hr

last_edit = "Last edited on 2026-06-03 20:19:13 by stolfi"

def main():

  global last_edit
  
  title = "Page ???"
  st = h.new_doc(title, "#eeffdd")
  
  
  h.section(st, 3, "Definition of <i>{k}-repeat</i>")
  
  h.parags(st, """Let {tks[0..nt-1]} be the sequence of all tokens of
  either text. For the SBJ, each token {tks[i]} is a pinyin syllable, in
  Unicode, initially with tones marked by diacritics. Recipe boundaries
  are ignored for simplicity. For the SBJ, each token is a sequence of
  lowercase EVA letters ([a-z] or '?') originally delimited by EVA
  spaces [.,-] of line breaks, ignorng line, parag, and page boundaries.
  
  For any positive {k}, we say that the text has a <i>{k}-repeat at</i> positions {i0} and
  {i1}if the following conditions hold:
  
    (0) {0 <= i0 < i1 < nt};
    
    (1) {tks[i0+r]} and {tks[i1+r]} are similar for all {r} in {0..k-1}.
    
    (2) {tks[i0-1]} and {tks[i1-1]} are not similar (or {i0 == 0});
    
    (3) {tks[i0+k]} and {tks[i1+k]} are not similar (or {i1+k == nt}); and
    
    (4) There exist no {k} repeat {i0,im} with {i0 < im < i1}.
  
  Conditions (2) and (3) exclude {k}-repeats {i0,i1} where <i>both</i> token strings are
  substrings of the strings of a {k'} repeat with {k'>k}.  It is OK is that happens to only one
  of the strings.
  
  Condition (4) means that a sequence of words that occurs 6
  times counts as 5 repeats (not 15).  
  
  For example, in the text 
    
    "x y B C D z A B C D E w r A B C D E s t B C D E u v B C D t"
     0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2
     0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9
     
  there is 
  
    a 5-repeat "A B C D E" at 06 and 13 
    a 4-repeat "B C D E" at 14 and 20
    a 3-repeat "B C D" at 02 and 07
    a 3-repeat "B C D" at 14 and 26
    
  However there is <b>not</b> a 3-repeat "A B C" at 06 and 13, nor a
  4-repeat "B C D E" at 07 and 14, nor a 3-repeat "B C D" at 02 and 20.
  
  If there is a {k}-repeat at {i0} and {i1}, the pairs of words mentioned
  in conditions (2) and (3) are its <i>stop pairs</i> -- the tokens that
  prevented the repeat from being extended further.  In the above example,
  the stop pairs of the 3-repeat at 06 and 07 are ("y", "A") and ("z", "E").
  """)

  h.section(st, 2, "Summary")

  h.parags(st, """Page ???""")

  h.section(st, 2, "All clips")

  h.parags(st, """???""")

  hr.links_section(st)

  h.output_doc(st, sys.stdout, 99, last_edit)
  return 0
  # ----------------------------------------------------------------------

main()