Skip to content

Commit

Permalink
. (optimization) Checking against header or footer data is now made i…
Browse files Browse the repository at this point in the history 
…n the ExtractText() method, instead of __next_instruction(), which caused too many calls to the preg_match() function.

. Added the IsPageHeaderOrFooter() method.

. Bug fix : Positive offsets between two text groups were unduly taken into account for the number of spaces to be inserted between those groups (negative offsets add spacing, while positive ones are subtracted from the current x-position).

. Bug fix : Space insertion for relative x-positioning did not take into account the last x position.
  • Loading branch information
@christian-vigh
christian-vigh committed Aug 8, 2016
1 parent 6499ecc commit c1d8d7a
Showing 1 changed file with 53 additions and 35 deletions.
88 changes: 53 additions & 35 deletions PdfToText.phpclass
View file
Original file line number Diff line number Diff line change
Expand Up @@ -617,6 +617,20 @@
. Bug fix : the __next_token() function was also returning the next character after character codes
specified within angle brackets ("<>"), which caused extra NUL values to be displayed in the output.

[Version : 1.2.37] [Date : 2016/08/08] [Author : CV]
. (optimization) Checking against header or footer data is now made in the ExtractText() method, instead
of __next_instruction(), which caused too many calls to the preg_match() function.
. Added the IsPageHeaderOrFooter() method.
. Bug fix : Positive offsets between two text groups were unduly taken into account for the number of
spaces to be inserted between those groups (negative offsets add spacing, while positive ones are
subtracted from the current x-position).
. Bug fix : Space insertion for relative x-positioning did not take into account the last x position.

http://www.fpdf.org/en/script/script37.php
https://pear.php.net/package/Crypt_RC4/docs/latest/__filesource/fsource_Crypt__Crypt_RC4-1.0.3CryptRc4.php.html
http://www.phpclasses.org/package/4957-PHP-Encrypt-and-decrypt-data-with-RC4-algorithm.html#view_files/files/25139


**************************************************************************************************************/


Expand Down Expand Up @@ -795,6 +809,31 @@ abstract class PdfObjectBase // extends Object
}


/*--------------------------------------------------------------------------------------------------------------

NAME
IsPageHeaderOrFooter - Check if the specified object contents denote a text stream.

PROTOTYPE
$status = $this -> IsText ( $object_data, $decoded_stream_data ) ;

DESCRIPTION
Checks if the specified decoded stream contents denotes header or footer data.

PARAMETERS
$stream_data (string) -
Decoded stream contents.

*-------------------------------------------------------------------------------------------------------------*/
protected function IsPageHeaderOrFooter ( $stream_data )
{
if ( preg_match ( '#/Type/Pagination/Subtype/((Header)|(Footer))#ix', $stream_data ) )
return ( true ) ;
else
return ( false ) ;
}


/*--------------------------------------------------------------------------------------------------------------

NAME
Expand Down Expand Up @@ -1058,7 +1097,7 @@ abstract class PdfObjectBase // extends Object
class PdfToText extends PdfObjectBase
{
// Current version of the class
const VERSION = "1.2.36" ;
const VERSION = "1.2.37" ;

// Pdf processing options
const PDFOPT_NONE = 0x0000 ; // No extra option
Expand Down Expand Up @@ -1474,7 +1513,9 @@ class PdfToText extends PdfObjectBase
// Plain text (well, in fact PDF drawing instructions)
else if ( $this -> IsText ( $object_data, $decoded_stream_data ) )
{
$text [ $object_number ] = $decoded_stream_data ;
// We currently ignore page headers and footers
if ( ! $this -> IsPageHeaderOrFooter ( $decoded_stream_data ) )
$text [ $object_number ] = $decoded_stream_data ;
}
else if ( self::$DEBUG > 1 )
echo "\n----------------------------------- UNRECOGNIZED #$object_number :\n$decoded_stream_data\n" ;
Expand Down Expand Up @@ -2258,6 +2299,7 @@ class PdfToText extends PdfObjectBase

// Y-coordinate of the last seen "Tm" instruction
$last_goto_y = 0 ;
$last_goto_x = 0 ;

// Y-coordinate of the last seen "Td" or "TD" relative positioning instruction
$last_relative_goto_y = 0 ;
Expand All @@ -2271,6 +2313,9 @@ class PdfToText extends PdfObjectBase
// Current font size
$current_font_size = 0 ;

// Various pre-computed variables
$separator_length = strlen ( $this -> Separator ) ;

// Current font map width, in bytes, plus a flag saying whether the current font is mapped or not
if ( isset ( $this -> FontInformationBuffer [ $current_font ] ) )
{
Expand Down Expand Up @@ -2320,13 +2365,14 @@ class PdfToText extends PdfObjectBase
if ( $instruction [ 'relative' ] )
{
// Try to put a separator if the x coordinate is non-zero
if ( $instruction [ 'x' ] >= $current_font_size )
if ( $instruction [ 'x' ] - $last_goto_x >= $current_font_size )
$result .= $this -> Separator ;

$discard_last_instruction = true ;
$extra_newlines = 0 ;
$use_same_line = ( ( $last_relative_goto_y - abs ( $instruction [ 'y' ] ) ) <= $current_font_size ) ;
$last_relative_goto_y = abs ( $instruction [ 'y' ] ) ;
$last_goto_x = $instruction [ 'x' ] ;

break ;
}
Expand Down Expand Up @@ -2439,7 +2485,7 @@ class PdfToText extends PdfObjectBase
// the current result does not end with it
if ( $same_separators )
{
if ( $this -> Separator != '' && substr ( $result, - strlen ( $this -> Separator ) ) != $this -> BlockSeparator )
if ( $this -> Separator != '' && substr ( $result, - $separator_length ) != $this -> BlockSeparator )
$result .= $this -> BlockSeparator ;
}
else
Expand Down Expand Up @@ -2539,7 +2585,7 @@ class PdfToText extends PdfObjectBase

// Handle offsets between blocks of characters
if ( isset ( $instruction [ 'offsets' ] [ $value_index ] ) &&
abs ( $instruction [ 'offsets' ] [ $value_index ] ) > $this -> MinSpaceWidth )
- ( $instruction [ 'offsets' ] [ $value_index ] ) > $this -> MinSpaceWidth )
$result .= $this -> __get_character_padding ( $instruction [ 'offsets' ] [ $value_index ] ) ;

$value_index ++ ;
Expand Down Expand Up @@ -2635,36 +2681,12 @@ class PdfToText extends PdfObjectBase
// Holds the floating-point values encountered so far
$number_stack = [] ;

// Used to ignore everything until the last "ET" token (for page headers and footers)
$ignore = false ;

// Loop through the stream of tokens
while ( ( $part = $this -> __next_token ( $data, $data_length, $index ) ) !== false )
{
$token = $part [0] ;
$next_index = $part [1] ;

// If we are in a page header or footer definition, ignore everything
if ( $ignore )
{
if ( $token == 'ET' )
$ignore = false ;

$index = $next_index ;
continue ;
}
// Otherwise, try to see if we're in a page header or footer description, something like :
// <</Attached [/Top]/Type/Pagination/Subtype/Header>>
// or :
// <</Attached [/Bottom]/Type/Pagination/Subtype/Footer>>
else if ( preg_match ( '#/Type/Pagination/Subtype/((Header)|(Footer))#ix', $token ) )
{
$ignore = true ;
$index = $next_index ;

continue ;
}

// Floating-point number : push it onto the stack
if ( ( $token [0] >= '0' && $token [0] <= '9' ) || $token [0] == '-' || $token [0] == '+' || $token [0] == '.' )
$number_stack [] = $token ;
Expand Down Expand Up @@ -2920,9 +2942,7 @@ class PdfToText extends PdfObjectBase
if ( $pos === false )
return ( false ) ;

$token = substr ( $data, $index, $pos - $index + 2 ) ;

return ( [ $token, $pos + 2 ] ) ;
return ( [ substr ( $data, $index, $pos - $index + 2 ), $pos + 2 ] ) ;
}
else
{
Expand All @@ -2931,9 +2951,7 @@ class PdfToText extends PdfObjectBase
if ( $pos === false )
return ( false ) ;

$token = substr ( $data, $index, $pos - $index + 1 ) ;

return ( [ $token, $pos + 1 ] ) ;
return ( [ substr ( $data, $index, $pos - $index + 1 ), $pos + 1 ] ) ;
}

// Tick character : consider it as a keyword, in the same way as the "TJ" or "Tj" keywords
Expand Down

0 comments on commit c1d8d7a

Please sign in to comment.