Changeset 78

Timestamp:

04/27/08 22:17:53 (2 months ago)

Author:

bkuhn

Message:

r93@hughes: bkuhn | 2008-04-27 22:17:14 -0400

• Various improvements to handle autoformat of paragraphs and skipping
  various parts.
  

Files:

• trunk/apps/pdfdiff/Perl/lib/PDF/Extract/Text.pm (modified) (4 diffs)

trunk/apps/pdfdiff/Perl/lib/PDF/Extract/Text.pm

@@ -36 +36 @@
-
-use PDF::OCR::Thorough;
-
+  qw(autoformat break_wrap);
-use Text::Aspell;
-
@@ -139 +139 @@
-  my $cutOffLen = $avgLen - 5;
-  for (my $ii = 0; $ii < @lines; $ii++) {
-
+
+
+
-      # Assume that any line that starts with ten spaces or more is a
-      # title, heading or other stand alone unit of some sort.
+
+      $newText .= "\n\n" if ($newText !~ /\n\n$/s or $inPara);
+
-      $newText .= $lines->[$ii] . "\n";
-      # Add another newline if one doesn't follow
@@ -152 +157 @@
-        if ($lines->[$ii] =~ /\-$/);
-
-my $curLen = length($lines->[$ii]);
+$curLen = length($lines->[$ii]);  # May have changed
-    if ($curLen <= $cutOffLen) {
-      $newText .= $lines->[$ii] . "\n";
-      # Add another newline if one doesn't follow so the para is separated
-      $newText .= "\n" unless $lines->[$ii+1] =~ /^\s*$/;
+      $inPara = 0;
-    } else {
-      $newText .= $lines->[$ii] . " ";
+      $inPara = 1;
-    }
-  }
-
+
+
-}
-
@@ -167 +175 @@
-  my($self, $origFirstLine, $origSecondLine) = @_;
-  my ($firstLine, $secondLine) = ($origFirstLine, $origSecondLine);
-)\s+
+\s+[\[\(,]*)
-    my $word = $2;
-S+)([\s\.\,]+.*)$/$2
+w+)([\s\.\,\)\]]+)(.*)$/$3
-      $word .= $1;
-
+
+
+
+
-        if ($self->{speller}->check($word));
-    }