| 1 |
#!/usr/bin/perl |
| 2 |
|
| 3 |
# condense -- condense cross-reference only foldoc entries |
| 4 |
# |
| 5 |
# Takes as sole argument a raw foldoc Dictionary data file. |
| 6 |
# |
| 7 |
# Prints to stdout the file with cross-reference only entries |
| 8 |
# removed and their headword appended with $separator onto their |
| 9 |
# cross-referenced entry headword. |
| 10 |
# |
| 11 |
# Assumes that headwords are unindented. |
| 12 |
|
| 13 |
# Changelog: |
| 14 |
$version = 0.1; # Thu Aug 28 09:39:18 EDT 2003 |
| 15 |
$version = 0.2; # Thu Aug 28 09:57:07 EDT 2003 |
| 16 |
# Added a $separator variable. |
| 17 |
|
| 18 |
$separator = "%%%"; |
| 19 |
|
| 20 |
# verbosity level controls report to strerr: |
| 21 |
# 0 no report |
| 22 |
# 1 counts reported |
| 23 |
# 2 quirks also listed |
| 24 |
# 3 condensed headwords also listed |
| 25 |
|
| 26 |
$verbose = 1; |
| 27 |
die "Usage: $0 data_file\n" unless @ARGV == 1; |
| 28 |
|
| 29 |
open DATA, $ARGV[0] or die "Can't open file $ARGV[0]\n"; |
| 30 |
|
| 31 |
# prime $new_headword |
| 32 |
while($_=<DATA> and ! /^\S/) {} |
| 33 |
chomp($new_headword = $_); |
| 34 |
|
| 35 |
# scan the dictionary |
| 36 |
while (defined $new_headword) { |
| 37 |
$hw = $new_headword; |
| 38 |
undef $new_headword; |
| 39 |
$entry = ""; |
| 40 |
++$headword{$hw}; |
| 41 |
while($_=<DATA> and ! /^\S/) { $entry .= $_; } |
| 42 |
chomp($new_headword = $_); |
| 43 |
if($entry =~ /^\s*\{\s*(.*\S)\s*\}\s*$/) { |
| 44 |
$xref{$hw} = $1; |
| 45 |
push @ordered_keys, $hw; |
| 46 |
} |
| 47 |
} |
| 48 |
|
| 49 |
# shorten chains and analyze for errors |
| 50 |
for $hw (@ordered_keys) { |
| 51 |
if ( $headword{$hw} > 1 ) { |
| 52 |
push @duplicate, $hw unless $duplicate{$hw}; |
| 53 |
++$duplicate{$hw}; |
| 54 |
next; |
| 55 |
} |
| 56 |
$target = $xref{$hw}; |
| 57 |
if (! $headword{$target}) { |
| 58 |
push @missing, $hw; |
| 59 |
next; |
| 60 |
} |
| 61 |
%chain = undef; |
| 62 |
++$chain{$target}; |
| 63 |
while (defined $xref{$target} and $headword{$xref{$target}} |
| 64 |
and $chain{$target} == 1) { |
| 65 |
$target = $xref{$target}; |
| 66 |
++$chain{$target}; |
| 67 |
} |
| 68 |
if ($chain{$target} > 1) { |
| 69 |
push @loop, $hw; |
| 70 |
next; |
| 71 |
} |
| 72 |
++$remove{$hw}; |
| 73 |
push @ordered_remove, $hw; |
| 74 |
$headword_expansion{$target} .= $separator . $hw; |
| 75 |
} |
| 76 |
|
| 77 |
# report |
| 78 |
if ($verbose > 0) { |
| 79 |
warn scalar @ordered_remove, " condensable cross-references\n"; |
| 80 |
} |
| 81 |
if ($verbose > 2 ) { |
| 82 |
for(sort @ordered_remove) { warn" $_\n"; } |
| 83 |
} |
| 84 |
if ($verbose > 0) { |
| 85 |
warn scalar @missing, " entries with dangling cross-reference\n"; |
| 86 |
} |
| 87 |
if ($verbose > 1 ) { |
| 88 |
for(sort @missing) { warn" $_\n"; } |
| 89 |
} |
| 90 |
if ($verbose > 0) { |
| 91 |
warn scalar @duplicate, " duplicate entries with cross-reference\n"; |
| 92 |
} |
| 93 |
if ($verbose > 1 ) { |
| 94 |
for(sort @duplicate) { warn" $_\n"; } |
| 95 |
} |
| 96 |
if ($verbose > 0) { |
| 97 |
warn scalar @loop, " looped chains of cross-references\n"; |
| 98 |
} |
| 99 |
if ($verbose > 1 ) { |
| 100 |
for(sort @loop) { warn" $_\n"; } |
| 101 |
} |
| 102 |
|
| 103 |
# starting over, prime $new_headword |
| 104 |
seek DATA, 0, 0; |
| 105 |
while($_=<DATA> and ! /^\S/) {print} |
| 106 |
chomp($new_headword = $_); |
| 107 |
|
| 108 |
# and write the condensed dictionary |
| 109 |
while (defined $new_headword) { |
| 110 |
$headword = $new_headword; |
| 111 |
undef $new_headword; |
| 112 |
if ($remove{$headword}) { |
| 113 |
while($_=<DATA> and ! /^\S/) { } |
| 114 |
chomp($new_headword = $_); |
| 115 |
} else { |
| 116 |
print $headword, $headword_expansion{$headword}, "\n"; |
| 117 |
while($_=<DATA> and ! /^\S/) { print; } |
| 118 |
chomp($new_headword = $_); |
| 119 |
} |
| 120 |
} |