1 
2 The indexing programs here are modified from the versions printed in
3 Bentley & Kernighan, EP-ODD V1 #1.  The programs are also described in
4 AT&T Bell Laboratories Computing Science Technical Report No. 128,
5 ``Tools for Printing Indexes''.
6 
7 Changes from the published version derive from further experience
8 after the paper was frozen, plus some cleanup and corrections by Joe
9 Kruskal (to whom many thanks), plus some very local features for
10 printing the AMPL book.
11 
12 
13 USING THE PROGRAMS
14 
15 install
16 	makes the appropriate files executable.  since this file
17 	is not executable, use by typing "sh <install"
18 ix.test
19 see.terms
20 	these two files provide test input.  make.index produces
21 	standard output and files
22 		foo[1-9] foo.regular foo.see foo.hier foo.all
23 	from ix.test and see.terms.  to make sure things
24 	work when you first unbundle this file, type
25 		sh <install
26 		make.index ix.test >foo.ix
27 		troff -ms foo.ix >foo.out
28 	and then examine the troff output foo.out
29 cleanup
30 	removes the garbage files left around for debugging
31 
32 
33 CHANGES FROM THE PAPER
34 
35 make.index
36 	handles "see" file see.terms.  A line like
37 		algorithms<tab>searching, sorting
38 	generates in the final index
39 		algorithms see searching, sorting
40 	a 3rd field of %also makes it
41 		algorithms see also searching, sorting
42 doclean
43 deroman
44 range.prep
45 	minor change to defend against bug in some versions of "sort"
46 rotate
47 	moved here (and changed as necessary) to remove subtle bug.
48 	see check.data below
49 range.sort
50 	-u option on sort removes duplicate entries on same page
51 range.collapse
52 reroman
53 	page number concatenation removed from here ...
54 num.collapse
55 	and moved to here.   also commas between numbers now
56 	inserted here (to make see terms easier)
57 gen.key
58 	literals protected differently in gsub commands.
59 	rules for non-alpha index terms slightly richer:
60 		purely nonalphabetic lines first
61 		lines with leading digits next
62 		ordinary lines last
63 see.prep
64 	changed to match changes above, and to rely on font-changing {}
65 final.sort
66 	uses -d option for "telephone directory" order.
67 hierarchy
68 	a rather special purpose version to replace runs of items
69 	with a common one or two word prefix and replace them by
70 	a head word and indented lines.
71 	this also does some rearrangement to bring see terms to the top,
72 	and terms with formatting info to the bottom;  this is not
73 	always the right thing to do.
74 format
75 	letter changes (.YY) determined by first letter.
76 	minor rearrangement of how output line is created.
77 	commas no longer added here.
78 	[Some systems have a disk-formatting program called format.]
79 check.data
80 	new program that catches subtle errors in the data