kicad/common/plotters/PDF_plotter.cpp

1158 lines
36 KiB
C++

/**
* @file PDF_plotter.cpp
* @brief KiCad: specialized plotter for PDF files format
*/
/*
* This program source code file is part of KiCad, a free EDA CAD application.
*
* Copyright (C) 1992-2012 Lorenzo Marcantonio, l.marcantonio@logossrl.com
* Copyright (C) 1992-2022 KiCad Developers, see AUTHORS.txt for contributors.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, you may find one here:
* http://www.gnu.org/licenses/old-licenses/gpl-2.0.html
* or you may search the http://www.gnu.org website for the version 2 license,
* or you may write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
*/
#include <algorithm>
#include <wx/filename.h>
#include <wx/mstream.h>
#include <wx/zstream.h>
#include <advanced_config.h>
#include <eda_text.h> // for IsGotoPageHref
#include <ignore.h>
#include <macros.h>
#include <trigo.h>
#include <plotters/plotters_pslike.h>
std::string PDF_PLOTTER::encodeStringForPlotter( const wxString& aText )
{
// returns a string compatible with PDF string convention from a unicode string.
// if the initial text is only ASCII7, return the text between ( and ) for a good readability
// if the initial text is no ASCII7, return the text between < and >
// and encoded using 16 bits hexa (4 digits) by wide char (unicode 16)
std::string result;
// Is aText only ASCII7 ?
bool is_ascii7 = true;
for( size_t ii = 0; ii < aText.Len(); ii++ )
{
if( aText[ii] >= 0x7F )
{
is_ascii7 = false;
break;
}
}
if( is_ascii7 )
{
result = '(';
for( unsigned ii = 0; ii < aText.Len(); ii++ )
{
unsigned int code = aText[ii];
// These characters must be escaped
switch( code )
{
case '(':
case ')':
case '\\':
result += '\\';
KI_FALLTHROUGH;
default:
result += code;
break;
}
}
result += ')';
}
else
{
result = "<FEFF";
for( size_t ii = 0; ii < aText.Len(); ii++ )
{
unsigned int code = aText[ii];
char buffer[16];
sprintf( buffer, "%4.4X", code );
result += buffer;
}
result += '>';
}
return result;
}
bool PDF_PLOTTER::OpenFile( const wxString& aFullFilename )
{
m_filename = aFullFilename;
wxASSERT( !m_outputFile );
// Open the PDF file in binary mode
m_outputFile = wxFopen( m_filename, wxT( "wb" ) );
if( m_outputFile == nullptr )
return false ;
return true;
}
void PDF_PLOTTER::SetViewport( const VECTOR2I& aOffset, double aIusPerDecimil,
double aScale, bool aMirror )
{
m_plotMirror = aMirror;
m_plotOffset = aOffset;
m_plotScale = aScale;
m_IUsPerDecimil = aIusPerDecimil;
// The CTM is set to 1 user unit per decimal
m_iuPerDeviceUnit = 1.0 / aIusPerDecimil;
/* The paper size in this engine is handled page by page
Look in the StartPage function */
}
void PDF_PLOTTER::SetCurrentLineWidth( int aWidth, void* aData )
{
wxASSERT( m_workFile );
if( aWidth == DO_NOT_SET_LINE_WIDTH )
return;
else if( aWidth == USE_DEFAULT_LINE_WIDTH )
aWidth = m_renderSettings->GetDefaultPenWidth();
if( aWidth == 0 )
aWidth = 1;
wxASSERT_MSG( aWidth > 0, "Plotter called to set negative pen width" );
if( aWidth != m_currentPenWidth )
fprintf( m_workFile, "%g w\n", userToDeviceSize( aWidth ) );
m_currentPenWidth = aWidth;
}
void PDF_PLOTTER::emitSetRGBColor( double r, double g, double b, double a )
{
wxASSERT( m_workFile );
// PDF treats all colors as opaque, so the best we can do with alpha is generate an
// appropriate blended color assuming white paper.
if( a < 1.0 )
{
r = ( r * a ) + ( 1 - a );
g = ( g * a ) + ( 1 - a );
b = ( b * a ) + ( 1 - a );
}
fprintf( m_workFile, "%g %g %g rg %g %g %g RG\n", r, g, b, r, g, b );
}
void PDF_PLOTTER::SetDash( int aLineWidth, PLOT_DASH_TYPE aLineStyle )
{
wxASSERT( m_workFile );
switch( aLineStyle )
{
case PLOT_DASH_TYPE::DASH:
fprintf( m_workFile, "[%d %d] 0 d\n",
(int) GetDashMarkLenIU( aLineWidth ), (int) GetDashGapLenIU( aLineWidth ) );
break;
case PLOT_DASH_TYPE::DOT:
fprintf( m_workFile, "[%d %d] 0 d\n",
(int) GetDotMarkLenIU( aLineWidth ), (int) GetDashGapLenIU( aLineWidth ) );
break;
case PLOT_DASH_TYPE::DASHDOT:
fprintf( m_workFile, "[%d %d %d %d] 0 d\n",
(int) GetDashMarkLenIU( aLineWidth ), (int) GetDashGapLenIU( aLineWidth ),
(int) GetDotMarkLenIU( aLineWidth ), (int) GetDashGapLenIU( aLineWidth ) );
break;
case PLOT_DASH_TYPE::DASHDOTDOT:
fprintf( m_workFile, "[%d %d %d %d %d %d] 0 d\n",
(int) GetDashMarkLenIU( aLineWidth ), (int) GetDashGapLenIU( aLineWidth ),
(int) GetDotMarkLenIU( aLineWidth ), (int) GetDashGapLenIU( aLineWidth ),
(int) GetDotMarkLenIU( aLineWidth ), (int) GetDashGapLenIU( aLineWidth ) );
break;
default:
fputs( "[] 0 d\n", m_workFile );
}
}
void PDF_PLOTTER::Rect( const VECTOR2I& p1, const VECTOR2I& p2, FILL_T fill, int width )
{
wxASSERT( m_workFile );
VECTOR2D p1_dev = userToDeviceCoordinates( p1 );
VECTOR2D p2_dev = userToDeviceCoordinates( p2 );
SetCurrentLineWidth( width );
fprintf( m_workFile, "%g %g %g %g re %c\n", p1_dev.x, p1_dev.y,
p2_dev.x - p1_dev.x, p2_dev.y - p1_dev.y, fill == FILL_T::NO_FILL ? 'S' : 'B' );
}
void PDF_PLOTTER::Circle( const VECTOR2I& pos, int diametre, FILL_T aFill, int width )
{
wxASSERT( m_workFile );
VECTOR2D pos_dev = userToDeviceCoordinates( pos );
double radius = userToDeviceSize( diametre / 2.0 );
/* OK. Here's a trick. PDF doesn't support circles or circular angles, that's
a fact. You'll have to do with cubic beziers. These *can't* represent
circular arcs (NURBS can, beziers don't). But there is a widely known
approximation which is really good
*/
SetCurrentLineWidth( width );
// If diameter is less than width, switch to filled mode
if( aFill == FILL_T::NO_FILL && diametre < width )
{
aFill = FILL_T::FILLED_SHAPE;
SetCurrentLineWidth( 0 );
radius = userToDeviceSize( ( diametre / 2.0 ) + ( width / 2.0 ) );
}
double magic = radius * 0.551784; // You don't want to know where this come from
// This is the convex hull for the bezier approximated circle
fprintf( m_workFile, "%g %g m "
"%g %g %g %g %g %g c "
"%g %g %g %g %g %g c "
"%g %g %g %g %g %g c "
"%g %g %g %g %g %g c %c\n",
pos_dev.x - radius, pos_dev.y,
pos_dev.x - radius, pos_dev.y + magic,
pos_dev.x - magic, pos_dev.y + radius,
pos_dev.x, pos_dev.y + radius,
pos_dev.x + magic, pos_dev.y + radius,
pos_dev.x + radius, pos_dev.y + magic,
pos_dev.x + radius, pos_dev.y,
pos_dev.x + radius, pos_dev.y - magic,
pos_dev.x + magic, pos_dev.y - radius,
pos_dev.x, pos_dev.y - radius,
pos_dev.x - magic, pos_dev.y - radius,
pos_dev.x - radius, pos_dev.y - magic,
pos_dev.x - radius, pos_dev.y,
aFill == FILL_T::NO_FILL ? 's' : 'b' );
}
void PDF_PLOTTER::Arc( const VECTOR2I& aCenter, const VECTOR2I& aStart, const VECTOR2I& aEnd,
FILL_T aFill, int aWidth, int aMaxError )
{
wxASSERT( m_workFile );
/*
* Arcs are not so easily approximated by beziers (in the general case), so we approximate
* them in the old way
*/
EDA_ANGLE startAngle( aStart - aCenter );
EDA_ANGLE endAngle( aEnd - aCenter );
int radius = ( aStart - aCenter ).EuclideanNorm();
int numSegs = GetArcToSegmentCount( radius, aMaxError, FULL_CIRCLE );
EDA_ANGLE delta = ANGLE_360 / std::max( 8, numSegs );
VECTOR2I start( aStart );
VECTOR2I end( aEnd );
VECTOR2I pt;
if( startAngle > endAngle )
{
if( endAngle < ANGLE_0 )
endAngle.Normalize();
else
startAngle = startAngle.Normalize() - ANGLE_360;
}
SetCurrentLineWidth( aWidth );
VECTOR2D pos_dev = userToDeviceCoordinates( start );
fprintf( m_workFile, "%g %g m ", pos_dev.x, pos_dev.y );
for( EDA_ANGLE ii = delta; startAngle + ii < endAngle; ii += delta )
{
pt = start;
RotatePoint( pt, aCenter, -ii );
pos_dev = userToDeviceCoordinates( pt );
fprintf( m_workFile, "%g %g l ", pos_dev.x, pos_dev.y );
}
pos_dev = userToDeviceCoordinates( end );
fprintf( m_workFile, "%g %g l ", pos_dev.x, pos_dev.y );
// The arc is drawn... if not filled we stroke it, otherwise we finish
// closing the pie at the center
if( aFill == FILL_T::NO_FILL )
{
fputs( "S\n", m_workFile );
}
else
{
pos_dev = userToDeviceCoordinates( aCenter );
fprintf( m_workFile, "%g %g l b\n", pos_dev.x, pos_dev.y );
}
}
void PDF_PLOTTER::Arc( const VECTOR2I& aCenter, const EDA_ANGLE& aStartAngle,
const EDA_ANGLE& aEndAngle, int aRadius, FILL_T aFill, int aWidth )
{
wxASSERT( m_workFile );
if( aRadius <= 0 )
{
Circle( aCenter, aWidth, FILL_T::FILLED_SHAPE, 0 );
return;
}
/*
* Arcs are not so easily approximated by beziers (in the general case), so we approximate
* them in the old way
*/
EDA_ANGLE startAngle( aStartAngle );
EDA_ANGLE endAngle( aEndAngle );
VECTOR2I start;
VECTOR2I end;
const EDA_ANGLE delta( 5, DEGREES_T ); // increment to draw circles
if( startAngle > endAngle )
{
std::swap( startAngle, endAngle );
std::swap( start, end );
}
SetCurrentLineWidth( aWidth );
// Usual trig arc plotting routine...
start.x = aCenter.x + KiROUND( aRadius * (-startAngle).Cos() );
start.y = aCenter.y + KiROUND( aRadius * (-startAngle).Sin() );
VECTOR2D pos_dev = userToDeviceCoordinates( start );
fprintf( m_workFile, "%g %g m ", pos_dev.x, pos_dev.y );
for( EDA_ANGLE ii = startAngle + delta; ii < endAngle; ii += delta )
{
end.x = aCenter.x + KiROUND( aRadius * (-ii).Cos() );
end.y = aCenter.y + KiROUND( aRadius * (-ii).Sin() );
pos_dev = userToDeviceCoordinates( end );
fprintf( m_workFile, "%g %g l ", pos_dev.x, pos_dev.y );
}
end.x = aCenter.x + KiROUND( aRadius * (-endAngle).Cos() );
end.y = aCenter.y + KiROUND( aRadius * (-endAngle).Sin() );
pos_dev = userToDeviceCoordinates( end );
fprintf( m_workFile, "%g %g l ", pos_dev.x, pos_dev.y );
// The arc is drawn... if not filled we stroke it, otherwise we finish
// closing the pie at the center
if( aFill == FILL_T::NO_FILL )
{
fputs( "S\n", m_workFile );
}
else
{
pos_dev = userToDeviceCoordinates( aCenter );
fprintf( m_workFile, "%g %g l b\n", pos_dev.x, pos_dev.y );
}
}
void PDF_PLOTTER::PlotPoly( const std::vector<VECTOR2I>& aCornerList, FILL_T aFill, int aWidth,
void* aData )
{
wxASSERT( m_workFile );
if( aCornerList.size() <= 1 )
return;
SetCurrentLineWidth( aWidth );
VECTOR2D pos = userToDeviceCoordinates( aCornerList[0] );
fprintf( m_workFile, "%g %g m\n", pos.x, pos.y );
for( unsigned ii = 1; ii < aCornerList.size(); ii++ )
{
pos = userToDeviceCoordinates( aCornerList[ii] );
fprintf( m_workFile, "%g %g l\n", pos.x, pos.y );
}
// Close path and stroke(/fill)
fprintf( m_workFile, "%c\n", aFill == FILL_T::NO_FILL ? 'S' : 'b' );
}
void PDF_PLOTTER::PenTo( const VECTOR2I& pos, char plume )
{
wxASSERT( m_workFile );
if( plume == 'Z' )
{
if( m_penState != 'Z' )
{
fputs( "S\n", m_workFile );
m_penState = 'Z';
m_penLastpos.x = -1;
m_penLastpos.y = -1;
}
return;
}
if( m_penState != plume || pos != m_penLastpos )
{
VECTOR2D pos_dev = userToDeviceCoordinates( pos );
fprintf( m_workFile, "%g %g %c\n",
pos_dev.x, pos_dev.y,
( plume=='D' ) ? 'l' : 'm' );
}
m_penState = plume;
m_penLastpos = pos;
}
void PDF_PLOTTER::PlotImage( const wxImage& aImage, const VECTOR2I& aPos, double aScaleFactor )
{
wxASSERT( m_workFile );
VECTOR2I pix_size( aImage.GetWidth(), aImage.GetHeight() );
// Requested size (in IUs)
VECTOR2D drawsize( aScaleFactor * pix_size.x, aScaleFactor * pix_size.y );
// calculate the bitmap start position
VECTOR2I start( aPos.x - drawsize.x / 2, aPos.y + drawsize.y / 2 );
VECTOR2D dev_start = userToDeviceCoordinates( start );
/* PDF has an uhm... simplified coordinate system handling. There is
*one* operator to do everything (the PS concat equivalent). At least
they kept the matrix stack to save restore environments. Also images
are always emitted at the origin with a size of 1x1 user units.
What we need to do is:
1) save the CTM end establish the new one
2) plot the image
3) restore the CTM
4) profit
*/
fprintf( m_workFile, "q %g 0 0 %g %g %g cm\n", // Step 1
userToDeviceSize( drawsize.x ),
userToDeviceSize( drawsize.y ),
dev_start.x, dev_start.y );
/* An inline image is a cross between a dictionary and a stream.
A real ugly construct (compared with the elegance of the PDF
format). Also it accepts some 'abbreviations', which is stupid
since the content stream is usually compressed anyway... */
fprintf( m_workFile,
"BI\n"
" /BPC 8\n"
" /CS %s\n"
" /W %d\n"
" /H %d\n"
"ID\n", m_colorMode ? "/RGB" : "/G", pix_size.x, pix_size.y );
/* Here comes the stream (in binary!). I *could* have hex or ascii84
encoded it, but who cares? I'll go through zlib anyway */
for( int y = 0; y < pix_size.y; y++ )
{
for( int x = 0; x < pix_size.x; x++ )
{
unsigned char r = aImage.GetRed( x, y ) & 0xFF;
unsigned char g = aImage.GetGreen( x, y ) & 0xFF;
unsigned char b = aImage.GetBlue( x, y ) & 0xFF;
// PDF inline images don't support alpha, so premultiply against white background
if( aImage.HasAlpha() )
{
unsigned char alpha = aImage.GetAlpha( x, y ) & 0xFF;
if( alpha < 0xFF )
{
float a = 1.0 - ( (float) alpha / 255.0 );
r = ( int )( r + ( a * 0xFF ) ) & 0xFF;
g = ( int )( g + ( a * 0xFF ) ) & 0xFF;
b = ( int )( b + ( a * 0xFF ) ) & 0xFF;
}
}
if( aImage.HasMask() )
{
if( r == aImage.GetMaskRed() && g == aImage.GetMaskGreen()
&& b == aImage.GetMaskBlue() )
{
r = 0xFF;
g = 0xFF;
b = 0xFF;
}
}
// As usual these days, stdio buffering has to suffeeeeerrrr
if( m_colorMode )
{
putc( r, m_workFile );
putc( g, m_workFile );
putc( b, m_workFile );
}
else
{
// Greyscale conversion (CIE 1931)
unsigned char grey = KiROUND( r * 0.2126 + g * 0.7152 + b * 0.0722 );
putc( grey, m_workFile );
}
}
}
fputs( "EI Q\n", m_workFile ); // Finish step 2 and do step 3
}
int PDF_PLOTTER::allocPdfObject()
{
m_xrefTable.push_back( 0 );
return m_xrefTable.size() - 1;
}
int PDF_PLOTTER::startPdfObject(int handle)
{
wxASSERT( m_outputFile );
wxASSERT( !m_workFile );
if( handle < 0)
handle = allocPdfObject();
m_xrefTable[handle] = ftell( m_outputFile );
fprintf( m_outputFile, "%d 0 obj\n", handle );
return handle;
}
void PDF_PLOTTER::closePdfObject()
{
wxASSERT( m_outputFile );
wxASSERT( !m_workFile );
fputs( "endobj\n", m_outputFile );
}
int PDF_PLOTTER::startPdfStream( int handle )
{
wxASSERT( m_outputFile );
wxASSERT( !m_workFile );
handle = startPdfObject( handle );
// This is guaranteed to be handle+1 but needs to be allocated since
// you could allocate more object during stream preparation
m_streamLengthHandle = allocPdfObject();
if( ADVANCED_CFG::GetCfg().m_DebugPDFWriter )
{
fprintf( m_outputFile,
"<< /Length %d 0 R >>\n" // Length is deferred
"stream\n", handle + 1 );
}
else
{
fprintf( m_outputFile,
"<< /Length %d 0 R /Filter /FlateDecode >>\n" // Length is deferred
"stream\n", handle + 1 );
}
// Open a temporary file to accumulate the stream
m_workFilename = wxFileName::CreateTempFileName( "" );
m_workFile = wxFopen( m_workFilename, wxT( "w+b" ) );
wxASSERT( m_workFile );
return handle;
}
void PDF_PLOTTER::closePdfStream()
{
wxASSERT( m_workFile );
long stream_len = ftell( m_workFile );
if( stream_len < 0 )
{
wxASSERT( false );
return;
}
// Rewind the file, read in the page stream and DEFLATE it
fseek( m_workFile, 0, SEEK_SET );
unsigned char *inbuf = new unsigned char[stream_len];
int rc = fread( inbuf, 1, stream_len, m_workFile );
wxASSERT( rc == stream_len );
ignore_unused( rc );
// We are done with the temporary file, junk it
fclose( m_workFile );
m_workFile = nullptr;
::wxRemoveFile( m_workFilename );
unsigned out_count;
if( ADVANCED_CFG::GetCfg().m_DebugPDFWriter )
{
out_count = stream_len;
fwrite( inbuf, out_count, 1, m_outputFile );
}
else
{
// NULL means memos owns the memory, but provide a hint on optimum size needed.
wxMemoryOutputStream memos( nullptr, std::max( 2000l, stream_len ) ) ;
{
/* Somewhat standard parameters to compress in DEFLATE. The PDF spec is
* misleading, it says it wants a DEFLATE stream but it really want a ZLIB
* stream! (a DEFLATE stream would be generated with -15 instead of 15)
* rc = deflateInit2( &zstrm, Z_BEST_COMPRESSION, Z_DEFLATED, 15,
* 8, Z_DEFAULT_STRATEGY );
*/
wxZlibOutputStream zos( memos, wxZ_BEST_COMPRESSION, wxZLIB_ZLIB );
zos.Write( inbuf, stream_len );
} // flush the zip stream using zos destructor
wxStreamBuffer* sb = memos.GetOutputStreamBuffer();
out_count = sb->Tell();
fwrite( sb->GetBufferStart(), 1, out_count, m_outputFile );
}
delete[] inbuf;
fputs( "endstream\n", m_outputFile );
closePdfObject();
// Writing the deferred length as an indirect object
startPdfObject( m_streamLengthHandle );
fprintf( m_outputFile, "%u\n", out_count );
closePdfObject();
}
void PDF_PLOTTER::StartPage( const wxString& aPageNumber )
{
wxASSERT( m_outputFile );
wxASSERT( !m_workFile );
m_pageNumbers.push_back( aPageNumber );
// Compute the paper size in IUs
m_paperSize = m_pageInfo.GetSizeMils();
m_paperSize.x *= 10.0 / m_iuPerDeviceUnit;
m_paperSize.y *= 10.0 / m_iuPerDeviceUnit;
// Open the content stream; the page object will go later
m_pageStreamHandle = startPdfStream();
/* Now, until ClosePage *everything* must be wrote in workFile, to be
compressed later in closePdfStream */
// Default graphic settings (coordinate system, default color and line style)
fprintf( m_workFile,
"%g 0 0 %g 0 0 cm 1 J 1 j 0 0 0 rg 0 0 0 RG %g w\n",
0.0072 * plotScaleAdjX, 0.0072 * plotScaleAdjY,
userToDeviceSize( m_renderSettings->GetDefaultPenWidth() ) );
}
void PDF_PLOTTER::ClosePage()
{
wxASSERT( m_workFile );
// Close the page stream (and compress it)
closePdfStream();
// Page size is in 1/72 of inch (default user space units). Works like the bbox in postscript
// but there is no need for swapping the sizes, since PDF doesn't require a portrait page.
// We use the MediaBox but PDF has lots of other less-used boxes that could be used.
const double PTsPERMIL = 0.072;
VECTOR2D psPaperSize = VECTOR2D( m_pageInfo.GetSizeMils() ) * PTsPERMIL;
auto iuToPdfUserSpace = [&]( const VECTOR2I& aCoord ) -> VECTOR2D
{
VECTOR2D retval = VECTOR2D( aCoord ) * PTsPERMIL / SCH_IU_PER_MILS;
// PDF y=0 is at bottom of page, invert coordinate
retval.y = psPaperSize.y - retval.y;
return retval;
};
// Handle annotations (at the moment only "link" type objects)
std::vector<int> hyperlinkHandles;
// Allocate all hyperlink objects for the page and calculate their position in user space
// coordinates
for( const std::pair<BOX2I, wxString>& linkPair : m_hyperlinksInPage )
{
const BOX2I& box = linkPair.first;
const wxString& url = linkPair.second;
VECTOR2D bottomLeft = iuToPdfUserSpace( box.GetPosition() );
VECTOR2D topRight = iuToPdfUserSpace( box.GetEnd() );
BOX2D userSpaceBox;
userSpaceBox.SetOrigin( bottomLeft );
userSpaceBox.SetEnd( topRight );
hyperlinkHandles.push_back( allocPdfObject() );
m_hyperlinkHandles.insert( { hyperlinkHandles.back(), { userSpaceBox, url } } );
}
for( const std::pair<BOX2I, std::vector<wxString>>& menuPair : m_hyperlinkMenusInPage )
{
const BOX2I& box = menuPair.first;
const std::vector<wxString>& urls = menuPair.second;
VECTOR2D bottomLeft = iuToPdfUserSpace( box.GetPosition() );
VECTOR2D topRight = iuToPdfUserSpace( box.GetEnd() );
BOX2D userSpaceBox;
userSpaceBox.SetOrigin( bottomLeft );
userSpaceBox.SetEnd( topRight );
hyperlinkHandles.push_back( allocPdfObject() );
m_hyperlinkMenuHandles.insert( { hyperlinkHandles.back(), { userSpaceBox, urls } } );
}
int hyperLinkArrayHandle = -1;
// If we have added any annotation links, create an array containing all the objects
if( hyperlinkHandles.size() > 0 )
{
hyperLinkArrayHandle = startPdfObject();
bool isFirst = true;
fputs( "[", m_outputFile );
for( int handle : hyperlinkHandles )
{
if( isFirst )
isFirst = false;
else
fprintf( m_outputFile, " " );
fprintf( m_outputFile, "%d 0 R", handle );
}
fputs( "]\n", m_outputFile );
closePdfObject();
}
// Emit the page object and put it in the page list for later
m_pageHandles.push_back( startPdfObject() );
fprintf( m_outputFile,
"<<\n"
"/Type /Page\n"
"/Parent %d 0 R\n"
"/Resources <<\n"
" /ProcSet [/PDF /Text /ImageC /ImageB]\n"
" /Font %d 0 R >>\n"
"/MediaBox [0 0 %g %g]\n"
"/Contents %d 0 R\n",
m_pageTreeHandle,
m_fontResDictHandle,
psPaperSize.x,
psPaperSize.y,
m_pageStreamHandle );
if( hyperlinkHandles.size() > 0 )
fprintf( m_outputFile, "/Annots %d 0 R", hyperLinkArrayHandle );
fputs( ">>\n", m_outputFile );
closePdfObject();
// Mark the page stream as idle
m_pageStreamHandle = 0;
// Clean up
m_hyperlinksInPage.clear();
m_hyperlinkMenusInPage.clear();
}
bool PDF_PLOTTER::StartPlot( const wxString& aPageNumber )
{
wxASSERT( m_outputFile );
// First things first: the customary null object
m_xrefTable.clear();
m_xrefTable.push_back( 0 );
m_hyperlinksInPage.clear();
m_hyperlinkMenusInPage.clear();
m_hyperlinkHandles.clear();
m_hyperlinkMenuHandles.clear();
/* The header (that's easy!). The second line is binary junk required
to make the file binary from the beginning (the important thing is
that they must have the bit 7 set) */
fputs( "%PDF-1.5\n%\200\201\202\203\n", m_outputFile );
/* Allocate an entry for the page tree root, it will go in every page parent entry */
m_pageTreeHandle = allocPdfObject();
/* In the same way, the font resource dictionary is used by every page
(it *could* be inherited via the Pages tree */
m_fontResDictHandle = allocPdfObject();
/* Now, the PDF is read from the end, (more or less)... so we start
with the page stream for page 1. Other more important stuff is written
at the end */
StartPage( aPageNumber );
return true;
}
bool PDF_PLOTTER::EndPlot()
{
wxASSERT( m_outputFile );
// Close the current page (often the only one)
ClosePage();
/* We need to declare the resources we're using (fonts in particular)
The useful standard one is the Helvetica family. Adding external fonts
is *very* involved! */
struct {
const char *psname;
const char *rsname;
int font_handle;
} fontdefs[4] = {
{ "/Helvetica", "/KicadFont", 0 },
{ "/Helvetica-Oblique", "/KicadFontI", 0 },
{ "/Helvetica-Bold", "/KicadFontB", 0 },
{ "/Helvetica-BoldOblique", "/KicadFontBI", 0 }
};
/* Declare the font resources. Since they're builtin fonts, no descriptors (yay!)
We'll need metrics anyway to do any alignment (these are in the shared with
the postscript engine) */
for( int i = 0; i < 4; i++ )
{
fontdefs[i].font_handle = startPdfObject();
fprintf( m_outputFile,
"<< /BaseFont %s\n"
" /Type /Font\n"
" /Subtype /Type1\n"
/* Adobe is so Mac-based that the nearest thing to Latin1 is
the Windows ANSI encoding! */
" /Encoding /WinAnsiEncoding\n"
">>\n",
fontdefs[i].psname );
closePdfObject();
}
// Named font dictionary (was allocated, now we emit it)
startPdfObject( m_fontResDictHandle );
fputs( "<<\n", m_outputFile );
for( int i = 0; i < 4; i++ )
{
fprintf( m_outputFile, " %s %d 0 R\n",
fontdefs[i].rsname, fontdefs[i].font_handle );
}
fputs( ">>\n", m_outputFile );
closePdfObject();
for( const auto& [ linkHandle, linkPair ] : m_hyperlinkHandles )
{
const BOX2D& box = linkPair.first;
const wxString& url = linkPair.second;
startPdfObject( linkHandle );
fprintf( m_outputFile,
"<< /Type /Annot\n"
" /Subtype /Link\n"
" /Rect [%g %g %g %g]\n"
" /Border [16 16 0]\n",
box.GetLeft(), box.GetBottom(), box.GetRight(), box.GetTop() );
wxString pageNumber;
bool pageFound = false;
if( EDA_TEXT::IsGotoPageHref( url, &pageNumber ) )
{
for( size_t ii = 0; ii < m_pageNumbers.size(); ++ii )
{
if( m_pageNumbers[ii] == pageNumber )
{
fprintf( m_outputFile,
" /Dest [%d 0 R /FitB]\n"
">>\n",
m_pageHandles[ii] );
pageFound = true;
break;
}
}
if( !pageFound )
{
// destination page is not being plotted, assign the NOP action to the link
fprintf( m_outputFile,
" /A << /Type /Action /S /NOP >>\n"
">>\n" );
}
}
else
{
fprintf( m_outputFile,
" /A << /Type /Action /S /URI /URI %s >>\n"
">>\n",
encodeStringForPlotter( url ).c_str() );
}
closePdfObject();
}
for( const auto& [ menuHandle, menuPair ] : m_hyperlinkMenuHandles )
{
const BOX2D& box = menuPair.first;
const std::vector<wxString>& urls = menuPair.second;
// We currently only support menu links for internal pages. This vector holds the
// page names and numbers.
std::vector<std::pair<wxString, int>> pages;
for( const wxString& url : urls )
{
wxString pageNumber;
if( EDA_TEXT::IsGotoPageHref( url, &pageNumber ) )
{
for( size_t ii = 0; ii < m_pageNumbers.size(); ++ii )
{
if( m_pageNumbers[ii] == pageNumber )
pages.push_back( { pageNumber, ii } );
}
}
}
wxString js = wxT( "var aParams = [ " );
for( const std::pair<wxString, int>& page : pages )
{
js += wxString::Format( wxT( "{ cName: '%s', cReturn: '%d' }, " ),
wxString::Format( _( "Show Page %s" ), page.first ),
page.second );
}
js += wxT( "]; " );
js += wxT( "var cChoice = app.popUpMenuEx.apply\\( app, aParams \\); " );
js += wxT( "if\\( cChoice != null \\) this.pageNum = cChoice; " );
startPdfObject( menuHandle );
fprintf( m_outputFile,
"<< /Type /Annot\n"
" /Subtype /Link\n"
" /Rect [%g %g %g %g]\n"
" /Border [16 16 0]\n",
box.GetLeft(), box.GetBottom(), box.GetRight(), box.GetTop() );
fprintf( m_outputFile,
" /A << /Type /Action /S /JavaScript /JS (%s) >>\n"
">>\n",
js.ToStdString().c_str() );
closePdfObject();
}
/* The page tree: it's a B-tree but luckily we only have few pages!
So we use just an array... The handle was allocated at the beginning,
now we instantiate the corresponding object */
startPdfObject( m_pageTreeHandle );
fputs( "<<\n"
"/Type /Pages\n"
"/Kids [\n", m_outputFile );
for( unsigned i = 0; i < m_pageHandles.size(); i++ )
fprintf( m_outputFile, "%d 0 R\n", m_pageHandles[i] );
fprintf( m_outputFile,
"]\n"
"/Count %ld\n"
">>\n", (long) m_pageHandles.size() );
closePdfObject();
// The info dictionary
int infoDictHandle = startPdfObject();
char date_buf[250];
time_t ltime = time( nullptr );
strftime( date_buf, 250, "D:%Y%m%d%H%M%S", localtime( &ltime ) );
if( m_title.IsEmpty() )
{
// Windows uses '\' and other platforms use '/' as separator
m_title = m_filename.AfterLast( '\\');
m_title = m_title.AfterLast( '/');
}
fprintf( m_outputFile,
"<<\n"
"/Producer (KiCad PDF)\n"
"/CreationDate (%s)\n"
"/Creator %s\n"
"/Title %s\n",
date_buf,
encodeStringForPlotter( m_creator ).c_str(),
encodeStringForPlotter( m_title ).c_str() );
fputs( ">>\n", m_outputFile );
closePdfObject();
// The catalog, at last
int catalogHandle = startPdfObject();
fprintf( m_outputFile,
"<<\n"
"/Type /Catalog\n"
"/Pages %d 0 R\n"
"/Version /1.5\n"
"/PageMode /UseNone\n"
"/PageLayout /SinglePage\n"
">>\n", m_pageTreeHandle );
closePdfObject();
/* Emit the xref table (format is crucial to the byte, each entry must
be 20 bytes long, and object zero must be done in that way). Also
the offset must be kept along for the trailer */
long xref_start = ftell( m_outputFile );
fprintf( m_outputFile,
"xref\n"
"0 %ld\n"
"0000000000 65535 f \n", (long) m_xrefTable.size() );
for( unsigned i = 1; i < m_xrefTable.size(); i++ )
{
fprintf( m_outputFile, "%010ld 00000 n \n", m_xrefTable[i] );
}
// Done the xref, go for the trailer
fprintf( m_outputFile,
"trailer\n"
"<< /Size %lu /Root %d 0 R /Info %d 0 R >>\n"
"startxref\n"
"%ld\n" // The offset we saved before
"%%%%EOF\n",
(unsigned long) m_xrefTable.size(), catalogHandle, infoDictHandle, xref_start );
fclose( m_outputFile );
m_outputFile = nullptr;
return true;
}
void PDF_PLOTTER::Text( const VECTOR2I& aPos,
const COLOR4D& aColor,
const wxString& aText,
const EDA_ANGLE& aOrient,
const VECTOR2I& aSize,
enum GR_TEXT_H_ALIGN_T aH_justify,
enum GR_TEXT_V_ALIGN_T aV_justify,
int aWidth,
bool aItalic,
bool aBold,
bool aMultilineAllowed,
KIFONT::FONT* aFont,
void* aData )
{
// PDF files do not like 0 sized texts which create broken files.
if( aSize.x == 0 || aSize.y == 0 )
return;
// Render phantom text (which will be searchable) behind the stroke font. This won't
// be pixel-accurate, but it doesn't matter for searching.
int render_mode = 3; // invisible
const char *fontname = aItalic ? ( aBold ? "/KicadFontBI" : "/KicadFontI" )
: ( aBold ? "/KicadFontB" : "/KicadFont" );
// Compute the copious transformation parameters of the Current Transform Matrix
double ctm_a, ctm_b, ctm_c, ctm_d, ctm_e, ctm_f;
double wideningFactor, heightFactor;
computeTextParameters( aPos, aText, aOrient, aSize, m_plotMirror, aH_justify,
aV_justify, aWidth, aItalic, aBold, &wideningFactor, &ctm_a,
&ctm_b, &ctm_c, &ctm_d, &ctm_e, &ctm_f, &heightFactor );
SetColor( aColor );
SetCurrentLineWidth( aWidth, aData );
/* We use the full CTM instead of the text matrix because the same
coordinate system will be used for the overlining. Also the %f
for the trig part of the matrix to avoid %g going in exponential
format (which is not supported) */
fprintf( m_workFile, "q %f %f %f %f %g %g cm BT %s %g Tf %d Tr %g Tz ",
ctm_a, ctm_b, ctm_c, ctm_d, ctm_e, ctm_f,
fontname, heightFactor, render_mode, wideningFactor * 100 );
// The text must be escaped correctly
std:: string txt_pdf = encodeStringForPlotter( aText );
fprintf( m_workFile, "%s Tj ET\n", txt_pdf.c_str() );
// Restore the CTM
fputs( "Q\n", m_workFile );
// Plot the stroked text (if requested)
PLOTTER::Text( aPos, aColor, aText, aOrient, aSize, aH_justify, aV_justify, aWidth, aItalic,
aBold, aMultilineAllowed, aFont );
}
void PDF_PLOTTER::HyperlinkBox( const BOX2I& aBox, const wxString& aDestinationURL )
{
m_hyperlinksInPage.push_back( std::make_pair( aBox, aDestinationURL ) );
}
void PDF_PLOTTER::HyperlinkMenu( const BOX2I& aBox, const std::vector<wxString>& aDestURLs )
{
m_hyperlinkMenusInPage.push_back( std::make_pair( aBox, aDestURLs ) );
}