new-words

annotate new-words.sh @ 26:4a10c0f4510c

find changesets by author, revision, files, or words in the commit message
apostrophe support, some improvements in speed, two and three words combination support
author Igor Chubin <igor@chub.in>
date Fri May 21 01:02:21 2010 +0300 (2010-05-21)
parents d1eb7dc37feb
children 0a80b2fa3ed8
rev   line source
igor@26 1 #!/bin/bash
igor@20 2
igor@20 3 show_usage()
igor@20 4 {
igor@20 5 cat <<HELP > /dev/stderr
igor@0 6
igor@16 7 USAGE:
igor@16 8
igor@16 9 new-words [ -l lang ] [ -s ] [ ARG ]
igor@16 10
igor@16 11 SWITCHES:
igor@16 12
igor@20 13 -h print this screen
igor@20 14 -k put higher words that are similar to the known words (only for English)
igor@16 15 -l lang override language settings
igor@21 16 -n non-interactive mode (don't run vi)
igor@21 17 -m don't add marks (and don't save marks added by user)
igor@25 18 -p pages work with specified pages only (pages = start-stop/total )
igor@20 19 -s show the text statistics (percentage of known words and so on) and exit
igor@26 20 -2 -3 find 2 and 3 words' sequences
igor@16 21
igor@20 22 The language of the text can be specified also
igor@20 23 by name of the program new-words (correspondent link must be created before).
igor@20 24 For example, these calls are equivalent:
igor@0 25
igor@20 26 de-words URL
igor@0 27 new-words -l de URL
igor@0 28
igor@20 29 HELP
igor@20 30 }
igor@0 31
igor@20 32 if [ "$1" = "-h" ]
igor@20 33 then
igor@20 34 show_usage
igor@20 35 exit 0
igor@20 36 fi
igor@0 37
igor@0 38 WORK_DIR=~/.new-words/
igor@17 39 TEMP1=`mktemp /tmp/new-words-XXXXXXXXXX-temp1`
igor@17 40 TEMP2=`mktemp /tmp/new-words-XXXXXXXXXX-temp2`
igor@17 41 export ORIGINAL_TEXT=`mktemp /tmp/new-words-XXXXXXXXXX-orig`
igor@0 42 editor=${EDITOR:-vim}
igor@0 43
igor@24 44 # language detection
igor@21 45
igor@2 46 LANGUAGE=en
igor@2 47 my_name="`echo $0 | sed s@.*/@@ | sed s/-.*// `"
igor@22 48 for arg
igor@22 49 do
igor@22 50 if echo "$arg" | grep -q http://...wikipedia.org/wiki/
igor@22 51 then
igor@22 52 LANGUAGE="`echo $arg | sed s@http://@@ | sed s@.wikipedia.*@@`"
igor@22 53 fi
igor@22 54 done
igor@2 55 [ "${my_name}" = "new" ] || LANGUAGE="$my_name"
igor@24 56
igor@24 57 #----------------------------------------------------
igor@24 58 # command line options processing
igor@24 59
igor@24 60 STAT_ONLY=NO
igor@24 61 NEED_TO_USE_VOCABULARY_WHEN_SORT=NO
igor@24 62 DONT_ADD_MARKS=NO
igor@24 63 NON_INTERACTIVE_MODE=NO
igor@25 64 PART_TO_PROCESS=''
igor@26 65 GROUP_WORDS_BY_THREE=NO
igor@26 66 GROUP_WORDS_BY_TWO=NO
igor@26 67 while getopts l:skmnp:23 opt
igor@24 68 do
igor@24 69 case "$opt" in
igor@24 70 s) STAT_ONLY=YES;;
igor@24 71 k) NEED_TO_USE_VOCABULARY_WHEN_SORT=YES;;
igor@24 72 l) LANGUAGE="$OPTARG";;
igor@24 73 m) DONT_ADD_MARKS=YES;;
igor@24 74 n) NON_INTERACTIVE_MODE=YES;;
igor@25 75 p) PART_TO_PROCESS="$OPTARG";;
igor@26 76 2) GROUP_WORDS_BY_TWO=YES;;
igor@26 77 3) GROUP_WORDS_BY_THREE=YES;;
igor@24 78 \?) # unknown flag
igor@24 79 show_usage
igor@24 80 exit 1;;
igor@24 81 esac
igor@24 82 done
igor@24 83 shift `expr $OPTIND - 1`
igor@24 84
igor@0 85 if [ "$1" = "-l" ]
igor@0 86 then
igor@0 87 LANGUAGE="$2"
igor@0 88 shift 2
igor@0 89 fi
igor@24 90
igor@2 91 VOCABULARY=${LANGUAGE}.txt
igor@2 92 NOTES_FILE=notes-${LANGUAGE}.txt
igor@0 93
igor@21 94 #----------------------------------------------------
igor@21 95
igor@0 96 get_words()
igor@0 97 {
igor@1 98 tr ' ' '\n' | sed 's/--/ /g' \
igor@3 99 | sed "s/'/__APOSTROPHE__/g" \
igor@26 100 | perl -MEncode -Mutf8 -n -e '$_ = decode( "utf8", $_);y/*\r,.:#@()+=—<>$;"?!|·[]^%&/ /; binmode STDOUT, ":utf8"; print if /^[[:alpha:]'"'"'_-]*$/'\
igor@26 101 | sed "s/__APOSTROPHE__/'/g" \
igor@3 102 | tr ' ' '\n' \
igor@3 103 | tee "$1" \
igor@3 104 | grep_v_english_perl \
igor@0 105 | sort | uniq -c | awk '{if ($2!="") print;}' | sort -rn
igor@0 106 }
igor@0 107
igor@3 108 add_stat()
igor@3 109 {
igor@3 110 before="$1"
igor@3 111 after=${before}2
igor@3 112 cat > "$after"
igor@3 113 total="`wc -w $1 | awk '{print $1}'`"
igor@3 114 total_unknown="`cat $after|awk '{s=s+$1}END{print s}'`"
igor@3 115 total_known="`echo $total-$total_unknown|bc`"
igor@3 116 percentage="`echo '100*('$total-$total_unknown')'/$total | bc -l | sed 's/\\.\(.\).*/.\1/'`"
igor@18 117 #sentences="`cat $after | perl -e 'local $/; $_=<>; s@http://[a-zA-Z&_.:/0-9%?=,\#+()\[\]~-]*@@g; s@\n@@g; s@(Mr|Mrs)\.@\1POINT@g; @sentences=split /\\./;print $#sentences;'`"
igor@18 118 sentences="`cat $ORIGINAL_TEXT | perl -e 'local $/; $_=<>; s/[^.]//msg; print length($_);'`"
igor@18 119
igor@18 120
igor@16 121 if [ "$STAT_ONLY" = "YES" ]
igor@16 122 then
igor@18 123 echo "LANG KNOWN% UNKNOWN% KNOWN TOTAL WPS UWPS*10"
igor@18 124 echo "$LANGUAGE $percentage `echo \(100-$percentage\) | bc -l` $total_known $total `echo $total/$sentences|bc` `echo 10*$total_unknown/$sentences|bc` "
igor@17 125 rm $after
igor@16 126 return 0
igor@16 127 else
igor@16 128 echo "# $LANGUAGE, $percentage, <$total_known/$total>"
igor@16 129 fi
igor@16 130
igor@14 131 PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`
igor@14 132 cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@14 133 my $total=shift(@ARGV);
igor@14 134 my $total_known=shift(@ARGV);
igor@14 135 my $s=0;
igor@16 136 my $mark_line=int($total_known*100/$total/5)*5;
igor@19 137 if ($mark_line>=90) {
igor@19 138 $mark_line=int($total_known*100/$total)+1;
igor@19 139 } else { $mark_line +=5; };
igor@14 140 while(<>)
igor@14 141 {
igor@14 142 print;
igor@14 143 /^\s*([0-9]*)\s*/;
igor@14 144 $s+=$1;
igor@19 145 if (($total_known+$s)*100/$total>=$mark_line) {
igor@14 146 print "# $mark_line\n";
igor@14 147 if ($mark_line>=90) { $mark_line+=1; } else { $mark_line +=5; };
igor@14 148 }
igor@14 149 }
igor@14 150 PERL_SCRIPT
igor@14 151 perl $PERL_SCRIPT_TEMP_NAME "$total" "$total_known" "$after"
igor@14 152 rm $PERL_SCRIPT_TEMP_NAME
igor@3 153 rm $after
igor@3 154 }
igor@3 155
igor@26 156 two_and_three_words()
igor@26 157 {
igor@26 158 if [ $GROUP_WORDS_BY_THREE = NO -a $GROUP_WORDS_BY_TWO = NO ]
igor@26 159 then
igor@26 160 cat
igor@26 161 else
igor@26 162 cat
igor@26 163
igor@26 164 export GROUP_WORDS_BY_THREE
igor@26 165 export GROUP_WORDS_BY_TWO
igor@26 166 PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-two-and-three-XXXXXXXX`
igor@26 167 cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@26 168 #!/usr/bin/perl
igor@26 169 local $/;
igor@26 170 $words=<>;
igor@26 171 $words=~ s@[!?;,:#1-9".]@ @g;
igor@26 172 $words =~ s@\s+@ @g;
igor@26 173 @words = split /\s+/, $words;
igor@26 174 for ($i=0; $i<$#words-3;$i++) {
igor@26 175 my ($a,$b,$c)= ($words[$i],$words[$i+1],$words[$i+2]);
igor@26 176 if ($ENV{GROUP_WORDS_BY_THREE} eq "YES" and ($a && $b && $c)) {
igor@26 177 print "${a}_${b}_${c}\n";
igor@26 178 };
igor@26 179 if ($ENV{GROUP_WORDS_BY_TWO} eq "YES" and ($a && $b)) {
igor@26 180 print "${a}_${b}\n";
igor@26 181 };
igor@26 182 }
igor@26 183 PERL_SCRIPT
igor@26 184 perl $PERL_SCRIPT_TEMP_NAME "$ORIGINAL_TEXT"
igor@26 185 rm $PERL_SCRIPT_TEMP_NAME
igor@26 186 fi
igor@26 187 }
igor@26 188
igor@0 189 grep_v_english()
igor@0 190 {
igor@0 191 [ -e "$VOCABULARY" ] || touch "$VOCABULARY"
igor@0 192 eval $(cat $VOCABULARY | tr -d "'" | xargs -n10 echo | tr ' ' '|' | sed 's/^/egrep -xv "RRRRRRR|/' | sed 's/$/"/' | tr '\n' '|')cat
igor@0 193 }
igor@0 194
igor@0 195 grep_v_english_perl()
igor@0 196 {
igor@0 197 PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`
igor@0 198 cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@0 199 open(VOC, $ENV{VOCABULARY})
igor@0 200 or die "Can't open VOCABULARY";
igor@0 201 while (<VOC>){
igor@0 202 chomp;
igor@3 203 #s/'//g;
igor@0 204 $voc{$_}="1";
igor@0 205 }
igor@0 206 while(<>) {
igor@0 207 chomp;
igor@0 208 if (not defined($voc{$_})) { print "$_\n"; }
igor@0 209 }
igor@0 210 PERL_SCRIPT
igor@0 211 [ -e "$VOCABULARY" ] || touch "$VOCABULARY"
igor@0 212 export VOCABULARY
igor@0 213 perl $PERL_SCRIPT_TEMP_NAME
igor@0 214 rm $PERL_SCRIPT_TEMP_NAME
igor@0 215 }
igor@0 216
igor@11 217 group_words()
igor@11 218 {
igor@22 219 #if [ "$LANGUAGE" != "en" ]
igor@22 220 #then
igor@22 221 # cat
igor@22 222 # return
igor@22 223 #fi
igor@11 224 PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-group-words-XXXXXXXX`
igor@11 225 cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@11 226 #!/usr/bin/perl
igor@11 227
igor@12 228 eval {
igor@12 229 # http://stackoverflow.com/questions/251694/how-can-i-check-if-i-have-a-perl-module-before-using-it
igor@12 230 require String::Similarity;
igor@12 231 String::Similarity->import();
igor@12 232 };
igor@12 233 unless($@)
igor@12 234 {
igor@12 235 our $HAVE_String_Similarity=1;
igor@12 236 }
igor@12 237
igor@12 238 sub similar($$){
igor@12 239 my $a=shift;
igor@12 240 my $b=shift;
igor@12 241 if ($HAVE_String_Similarity) {
igor@12 242 return $Similarity{"$a $b"};
igor@12 243 }
igor@12 244 else {
igor@12 245 return 0;
igor@12 246 }
igor@12 247 }
igor@12 248
igor@11 249 sub normalize($)
igor@11 250 {
igor@22 251 if ( $ENV{LANGUAGE} eq "en" ) { return normalize_english(shift); }
igor@22 252 elsif ( $ENV{LANGUAGE} eq "de" ) { return normalize_german(shift); }
igor@22 253 else { return shift ; }
igor@22 254 }
igor@22 255
igor@22 256 sub normalize_german($)
igor@22 257 {
igor@22 258 $_=lc(shift);
igor@22 259
igor@22 260 s/heit$//; s/keit$//; s/tum$//; s/ung$//; s/nis$//;s/schaft$//; s/ist$//;
igor@22 261 s/en$//; s/er$//;
igor@22 262
igor@22 263 s/lich$//; s/ig$//;
igor@22 264 s/al$//; s/isch$//;
igor@22 265 s/ell$//; s/haft$//;
igor@22 266
igor@22 267 s/bar$//; s/sam$//; s/lich$//;
igor@22 268
igor@22 269 @prefixes=qw(
igor@22 270 ab an auf aus bei dazwischen ein fest heraus her hinaus hin los mit nach voraus vorbei vor weg weiter zurück zusammen zu
igor@22 271 be emp ent er ge miss ver zer durch über um unter wieder);
igor@24 272 @prefixes=();
igor@22 273 for $pref (@prefixes) {
igor@22 274 s/^$pref//;
igor@22 275 }
igor@22 276
igor@22 277
igor@22 278 return $_;
igor@22 279 }
igor@22 280
igor@22 281 sub normalize_english($)
igor@22 282 {
igor@11 283 $_=lc(shift);
igor@12 284
igor@11 285 s/s$//;
igor@22 286
igor@22 287 s/ation$//; s/ness$//; s/ship$//; s/ally$//; s/ance$//;s/ity$//; s/ment$//;
igor@22 288
igor@11 289 s/ed$//;
igor@12 290 s/en$//;
igor@12 291 s/er$//;
igor@12 292 s/est$//;
igor@11 293 s/ing$//;
igor@12 294
igor@12 295 s/ism$//; s/ist$//; s/ful$//; s/able$//; s/ably$//;
igor@22 296 s/ify$//; s/fy$//; s/ly$//;
igor@12 297 s/ise$//; s/ize$//;
igor@12 298
igor@12 299 s/e$//;
igor@11 300 return $_;
igor@11 301 }
igor@11 302
igor@12 303
igor@11 304 sub compare($$)
igor@11 305 {
igor@11 306 my $a=shift;
igor@11 307 my $b=shift;
igor@11 308 $a =~ s/^\s*//;
igor@11 309 $b =~ s/^\s*//;
igor@11 310 my ($a1, $a2)= split /\s+/,$a,2;
igor@11 311 my ($b1, $b2)= split /\s+/,$b,2;
igor@11 312
igor@11 313 my $cmp = $group_weight{normalize($a2)} <=> $group_weight{normalize($b2)};
igor@12 314
igor@11 315 if ($cmp) {
igor@11 316 return $cmp;
igor@11 317 }
igor@11 318 else {
igor@11 319 if (normalize($a2) ne normalize($b2)) {
igor@11 320 return normalize($a2) cmp normalize($b2);
igor@11 321 }
igor@11 322 else {
igor@11 323 return $a1 <=> $b1;
igor@11 324 }
igor@11 325 }
igor@11 326 }
igor@11 327
igor@20 328 our %Vocabulary;
igor@20 329 open(VOC, $ENV{VOCABULARY})
igor@20 330 or die "Can't open VOCABULARY";
igor@20 331 while (<VOC>){
igor@20 332 chomp;
igor@20 333 #s/'//g;
igor@20 334 $Vocabulary{normalize($_)}="1";
igor@20 335 }
igor@20 336 close(VOC);
igor@20 337
igor@20 338 @lines=<STDIN>;
igor@11 339 for $L (@lines) {
igor@11 340 chomp($L);
igor@11 341 $l=$L;
igor@11 342 $l =~ s/^\s*//;
igor@11 343 my ($a, $b)=split(/\s+/,$l,2);
igor@11 344 $group_weight{normalize($b)}+=$a;
igor@11 345 }
igor@20 346 if ($ENV{NEED_TO_USE_VOCABULARY_WHEN_SORT} eq "YES") {
igor@20 347 for $k (keys %group_weight) {
igor@20 348 if (defined($Vocabulary{$k})) {
igor@20 349 $group_weight{$k} *= 2;
igor@20 350 }
igor@20 351 }
igor@20 352 }
igor@11 353 @lines2 = sort { compare($b,$a) } @lines;
igor@11 354 for $l (@lines2) {
igor@11 355 print "$l\n";
igor@11 356 }
igor@11 357 PERL_SCRIPT
igor@20 358 export VOCABULARY
igor@20 359 export NEED_TO_USE_VOCABULARY_WHEN_SORT
igor@22 360 export LANGUAGE
igor@11 361 perl $PERL_SCRIPT_TEMP_NAME
igor@11 362 rm $PERL_SCRIPT_TEMP_NAME
igor@11 363 }
igor@11 364
igor@0 365 text_from_url()
igor@0 366 {
igor@0 367 lynx -dump "$1" | perl -p -e 's@http://[a-zA-Z&_.:/0-9%?=,#+()\[\]~-]*@@'
igor@0 368 }
igor@0 369
igor@2 370 add_marks()
igor@2 371 {
igor@2 372 PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`
igor@2 373 cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@2 374 $file = $ARGV[0];
igor@2 375 our $dict;
igor@2 376 if (open(NOTES, $ENV{NOTES_FILE})) {
igor@2 377 while(<NOTES>) {
igor@2 378 chomp;
igor@2 379 s/^\s+//;
igor@2 380 my ($a,$b)=split /\s+/,$_,2;
igor@2 381 $dict{$a}=$b;
igor@2 382 }
igor@2 383 }
igor@2 384 if (open(F, $file)) {
igor@2 385 @lines=<F>;
igor@2 386 close(F);
igor@2 387
igor@2 388 if (open(F, ">$file")) {
igor@2 389 for (@lines) {
igor@2 390 m/\s+\S+\s+(\S+)/;
igor@2 391 $name=$1;
igor@3 392 if (not /^#/ and defined($dict{$name})) {
igor@2 393 chomp;
igor@2 394 $mark=$dict{$name};
igor@2 395 $space=" "x(30-length($_));
igor@2 396 print F "$_$space$mark\n";
igor@2 397 }
igor@2 398 else {
igor@2 399 print F "$_";
igor@2 400 }
igor@2 401 }
igor@2 402 close(F);
igor@2 403 }
igor@2 404 }
igor@2 405 PERL_SCRIPT
igor@2 406 [ -e "$NOTES_FILE" ] || touch "$NOTES_FILE"
igor@2 407 export NOTES_FILE
igor@2 408 perl $PERL_SCRIPT_TEMP_NAME "$1"
igor@2 409 rm $PERL_SCRIPT_TEMP_NAME
igor@2 410 }
igor@2 411
igor@2 412 remove_marks()
igor@2 413 {
igor@2 414 PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`
igor@2 415 cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@2 416 $file = $ARGV[0];
igor@2 417 our %dict;
igor@2 418 if (open(F, $file)) {
igor@2 419 @lines=<F>;
igor@2 420 close(F);
igor@2 421
igor@2 422 if (open(F, ">$file")) {
igor@2 423 for (@lines) {
igor@2 424 chomp;
igor@3 425 if (not /^#/ and m/(\s+)(\S+)(\s+)(\S+)(\s+)(.*)/) {
igor@2 426 my $name=$4;
igor@2 427 my $comment=$6;
igor@2 428 $dict{$name}=$comment;
igor@2 429 print F "$1$2$3$4\n";
igor@2 430 }
igor@2 431 else {
igor@2 432 print F "$_\n";
igor@2 433 }
igor@2 434 }
igor@2 435 }
igor@2 436 }
igor@21 437 if (($ENV{DONT_ADD_MARKS} ne "YES") and open(NOTES, $ENV{NOTES_FILE})) {
igor@2 438 @lines=<NOTES>;
igor@2 439 close(NOTES);
igor@2 440
igor@2 441 if (open(NOTES, ">".$ENV{NOTES_FILE})) {
igor@2 442 for (@lines) {
igor@2 443 chomp;
igor@2 444 s/^\s+//;
igor@2 445 my ($a,$b)=split /\s+/,$_,2;
igor@2 446 if (not defined($dict{$a}) || ($dict{$a} eq $b)) {
igor@2 447 print NOTES "$_\n";
igor@2 448 if (defined($dict{$a})) { unset($dict{$a}); }
igor@2 449 }
igor@2 450 }
igor@2 451 for (keys %dict) {
igor@2 452 $mark=$dict{$_};
igor@2 453 $space=" "x(30-length($_));
igor@2 454 print NOTES "$_$space$mark\n";
igor@2 455 }
igor@2 456 }
igor@2 457 }
igor@2 458 PERL_SCRIPT
igor@2 459 [ -e "$NOTES_FILE" ] || touch "$NOTES_FILE"
igor@2 460 export NOTES_FILE
igor@21 461 export DONT_ADD_MARKS
igor@2 462 perl $PERL_SCRIPT_TEMP_NAME "$1"
igor@2 463 rm $PERL_SCRIPT_TEMP_NAME
igor@2 464 }
igor@2 465
igor@25 466 part()
igor@25 467 {
igor@25 468 PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-part-XXXXXXXX`
igor@25 469 cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@25 470 #!/usr/bin/perl
igor@25 471
igor@25 472 my @lines=<STDIN>;
igor@25 473 my $lines=$#lines;
igor@25 474 my $interval=$ARGV[0];
igor@25 475 if (not $interval) {
igor@25 476 print @lines;
igor@25 477 }
igor@25 478 else {
igor@25 479 my ($start,$stop,$total);
igor@25 480 if ($interval =~ m@(.*)/(.*)@) {
igor@25 481 $start = $1;
igor@25 482 $total = $2;
igor@25 483 }
igor@25 484 else {
igor@25 485 $start=$interval;
igor@25 486 $total=0;
igor@25 487 }
igor@25 488 if ($start =~ m@(.*)-(.*)@) {
igor@25 489 $start = $1;
igor@25 490 $stop = $2;
igor@25 491 }
igor@25 492 if ($start =~ m@(.*)\+(.*)@) {
igor@25 493 $start = $1;
igor@25 494 $stop = $start+$2;
igor@25 495 }
igor@25 496
igor@25 497 $start=int($lines/$total*$start);
igor@25 498 $stop=int($lines/$total*$stop);
igor@25 499
igor@25 500 for($i=$start;$i<$stop;$i++){
igor@25 501 print $lines[$i];
igor@25 502 }
igor@25 503 }
igor@25 504 PERL_SCRIPT
igor@25 505 perl $PERL_SCRIPT_TEMP_NAME "$1"
igor@25 506 rm $PERL_SCRIPT_TEMP_NAME
igor@25 507 }
igor@25 508
igor@0 509 mkdir -p $WORK_DIR
igor@8 510 oldpwd="$PWD"
igor@0 511 cd $WORK_DIR
igor@0 512 if echo "$1" | grep -q http:
igor@0 513 then
igor@11 514 text_from_url "$1"
igor@0 515 elif [ "$#" != 0 ]
igor@0 516 then
igor@13 517 if echo $1 | grep -q ^/
igor@13 518 then
igor@13 519 cat "$1"
igor@13 520 else
igor@13 521 cat "$oldpwd/$1"
igor@13 522 fi
igor@0 523 else
igor@11 524 cat
igor@11 525 fi \
igor@25 526 | part $PART_TO_PROCESS \
igor@11 527 | tee $ORIGINAL_TEXT \
igor@26 528 | two_and_three_words \
igor@11 529 | get_words ${TEMP1}-full \
igor@11 530 | group_words \
igor@11 531 | add_stat ${TEMP1}-full \
igor@11 532 | tee "$TEMP1" > "$TEMP2"
igor@0 533
igor@16 534 if [ "$STAT_ONLY" = "YES" ]
igor@0 535 then
igor@16 536 cat "$TEMP1"
igor@21 537 elif [ "$NON_INTERACTIVE_MODE" = "YES" ]
igor@21 538 then
igor@21 539 cat "$TEMP1"
igor@0 540 else
igor@21 541 [ "$DONT_ADD_MARKS" = "YES" ] || add_marks "$TEMP2"
igor@16 542 if [ "$editor" = vim ]
igor@16 543 then
igor@23 544 vim -c 'set keywordprg='"$LANGUAGE" -c 'set iskeyword=@,48-57,/,.,-,_,+,,,#,$,%,~,=,48-255' "$TEMP2" < /dev/tty > /dev/tty
igor@16 545 else
igor@16 546 echo 2
igor@16 547 $editor "$TEMP2"
igor@16 548 fi
igor@16 549 remove_marks "$TEMP2"
igor@0 550 fi
igor@2 551
igor@0 552 diff "$TEMP1" "$TEMP2" | awk '{print $3}' | sort -u >> "$VOCABULARY"
igor@5 553 rm -f "$TEMP1" "$TEMP2" "${TEMP1}-full" "$ORIGINAL_TEXT"