| 1 | ## -*- Mode: CPerl -*- | ||
| 2 | ## | ||
| 3 | ## File: DTA::CAB::Format::Text.pm | ||
| 4 | ## Author: Bryan Jurish <moocow@cpan.org> | ||
| 5 | ## Description: Datum parser: verbose human-readable text | ||
| 6 | |||
| 7 | package DTA::CAB::Format::Text; | ||
| 8 | use DTA::CAB::Format; | ||
| 9 | use DTA::CAB::Format::TT; | ||
| 10 | use DTA::CAB::Datum ':all'; | ||
| 11 | use IO::File; | ||
| 12 | use Carp; | ||
| 13 | use strict; | ||
| 14 | |||
| 15 | ##============================================================================== | ||
| 16 | ## Globals | ||
| 17 | ##============================================================================== | ||
| 18 | |||
| 19 | our @ISA = qw(DTA::CAB::Format::TT); | ||
| 20 | |||
| 21 | BEGIN { | ||
| 22 | DTA::CAB::Format->registerFormat(name=>__PACKAGE__, filenameRegex=>qr/\.(?i:txt|text|cab\-txt|cab\-text)$/); | ||
| 23 | DTA::CAB::Format->registerFormat(name=>__PACKAGE__, short=>'txt'); | ||
| 24 | } | ||
| 25 | |||
| 26 | ##============================================================================== | ||
| 27 | ## Constructors etc. | ||
| 28 | ##============================================================================== | ||
| 29 | |||
| 30 | ## $fmt = CLASS_OR_OBJ->new(%args) | ||
| 31 | ## + object structure: assumed HASH | ||
| 32 | ## ( | ||
| 33 | ## ##---- Input | ||
| 34 | ## doc => $doc, ##-- buffered input document | ||
| 35 | ## | ||
| 36 | ## ##---- Output | ||
| 37 | ## #level => $formatLevel, ##-- output formatting level: n/a | ||
| 38 | ## #outbuf => $stringBuffer, ##-- buffered output | ||
| 39 | ## | ||
| 40 | ## ##---- Common | ||
| 41 | ## utf8 => $bool, ##-- default: 1 | ||
| 42 | ## defaultFieldName => $name, ##-- default name for unnamed fields; parsed into @{$tok->{other}{$name}}; default='' | ||
| 43 | ## ) | ||
| 44 | ## + inherited from DTA::CAB::Format::TT | ||
| 45 | |||
| 46 | ##============================================================================== | ||
| 47 | ## Methods: Persistence | ||
| 48 | ##============================================================================== | ||
| 49 | |||
| 50 | ##============================================================================== | ||
| 51 | ## Methods: I/O: Block-wise | ||
| 52 | ##============================================================================== | ||
| 53 | |||
| 54 | ## \@blocks = $fmt->blockScanBody(\$buf,\%opts) | ||
| 55 | ## + scans $filename for block boundaries according to \%opts | ||
| 56 | sub blockScanBody { | ||
| 57 | my ($fmt,$bufr,$opts) = @_; | ||
| 58 | |||
| 59 | ##-- scan blocks into head, body, foot | ||
| 60 | my $bsize = $opts->{bsize} // $opts->{size} // 1048576; | ||
| 61 | my $fsize = $opts->{ifsize}; | ||
| 62 | my $eob = $opts->{eob} =~ /^s/i ? 's' : 'w'; | ||
| 63 | my $blocks = []; | ||
| 64 | |||
| 65 | my ($off0,$off1,$blk); | ||
| 66 | for ($off0=$opts->{ihead}[0]+$opts->{ihead}[1]; $off0 < $fsize; $off0=$off1) { | ||
| 67 | push(@$blocks, $blk={ioff=>$off0}); | ||
| 68 | pos($$bufr) = ($off0+$bsize < $fsize ? $off0+$bsize : $fsize); | ||
| 69 | if ($eob eq 's' ? $$bufr=~m/\n{2,}/sg : $$bufr=~m/\n{1,}(?!\t)/sg) { | ||
| 70 | $off1 = $+[0]; | ||
| 71 | $blk->{eos} = $+[0]-$-[0] > 1 ? 1 : 0; | ||
| 72 | } else { | ||
| 73 | $off1 = $fsize; | ||
| 74 | $blk->{eos} = 1; | ||
| 75 | } | ||
| 76 | $blk->{ilen} = $off1-$off0; | ||
| 77 | } | ||
| 78 | |||
| 79 | return $blocks; | ||
| 80 | } | ||
| 81 | |||
| 82 | ##============================================================================== | ||
| 83 | ## Methods: Input | ||
| 84 | ##============================================================================== | ||
| 85 | |||
| 86 | ##-------------------------------------------------------------- | ||
| 87 | ## Methods: Input: Local | ||
| 88 | |||
| 89 | ## $fmt = $fmt->parseTextString(\$string) | ||
| 90 | BEGIN { *parseTTString = \&parseTextString; } | ||
| 91 | sub parseTextString { | ||
| 92 | my ($fmt,$src) = @_; | ||
| 93 | $$src =~ s/\r?\n\t\+?/\t/sg; | ||
| 94 | return DTA::CAB::Format::TT::parseTTString($fmt,$src); | ||
| 95 | } | ||
| 96 | |||
| 97 | ##-------------------------------------------------------------- | ||
| 98 | ## Methods: Input: Generic API | ||
| 99 | |||
| 100 | |||
| 101 | ##============================================================================== | ||
| 102 | ## Methods: Output | ||
| 103 | ##============================================================================== | ||
| 104 | |||
| 105 | ##-------------------------------------------------------------- | ||
| 106 | ## Methods: Output: Generic | ||
| 107 | |||
| 108 | ## $ext = $fmt->defaultExtension() | ||
| 109 | ## + returns default filename extension for this format | ||
| 110 | sub defaultExtension { return '.txt'; } | ||
| 111 | |||
| 112 | ##-------------------------------------------------------------- | ||
| 113 | ## Methods: Output: output selection | ||
| 114 | |||
| 115 | ## \$buf = $fmt->token2buf($tok,\$buf) | ||
| 116 | ## + buffer output for a single token | ||
| 117 | ## + override converts from TT->Text | ||
| 118 | sub token2buf { | ||
| 119 | my $bufr = DTA::CAB::Format::TT::token2buf(@_); | ||
| 120 | $$bufr =~ s/(?<!\n)\t/\n\t/sg; | ||
| 121 | $$bufr =~ s/^\t\[/\t+\[/mg; | ||
| 122 | return $bufr; | ||
| 123 | } | ||
| 124 | |||
| 125 | ##-------------------------------------------------------------- | ||
| 126 | ## Methods: Output: Generic API | ||
| 127 | |||
| 128 | ## $fmt = $fmt->putToken($tok) | ||
| 129 | ## + appends $tok to output buffer | ||
| 130 | ## + INHERITED from Format::TT | ||
| 131 | |||
| 132 | ## $fmt = $fmt->putSentence($sent) | ||
| 133 | ## + concatenates formatted tokens, adding sentence-id comment if available | ||
| 134 | ## + inherited from DTA::CAB::Format::TT | ||
| 135 | |||
| 136 | ## $out = $fmt->formatDocument($doc) | ||
| 137 | ## + concatenates formatted sentences, adding document 'xmlbase' comment if available | ||
| 138 | ## + inherited from DTA::CAB::Format::TT | ||
| 139 | |||
| 140 | 1; ##-- be happy | ||
| 141 | |||
| 142 | __END__ |