Merge pull request #4 from FOSSEE/develop

merging dev into master

16 files changed, 5277 insertions, 0 deletions

diff --git a/common/gal/cairo/cairo_compositor.cpp b/common/gal/cairo/cairo_compositor.cpp
new file mode 100644
index 0000000..f81cb5a
--- /dev/null
+++ b/common/gal/cairo/cairo_compositor.cpp
@@ -0,0 +1,159 @@
+/*
+ * This program source code file is part of KiCad, a free EDA CAD application.
+ *
+ * Copyright (C) 2013 CERN
+ * @author Maciej Suminski <maciej.suminski@cern.ch>
+ *
+ * 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:O//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
+ */
+
+/**
+ * @file cairo_compositor.cpp
+ * @brief Class that handles multitarget rendering (ie. to different textures/surfaces) and
+ * later compositing into a single image (Cairo flavour).
+ */
+
+#include <gal/cairo/cairo_compositor.h>
+#include <wx/log.h>
+
+using namespace KIGFX;
+
+CAIRO_COMPOSITOR::CAIRO_COMPOSITOR( cairo_t** aMainContext ) :
+    m_current( 0 ), m_currentContext( aMainContext ), m_mainContext( *aMainContext )
+{
+    // Do not have uninitialized members:
+    cairo_matrix_init_identity( &m_matrix );
+    m_stride = 0;
+    m_bufferSize = 0;
+}
+
+
+CAIRO_COMPOSITOR::~CAIRO_COMPOSITOR()
+{
+    clean();
+}
+
+
+void CAIRO_COMPOSITOR::Initialize()
+{
+    // Nothing has to be done
+}
+
+
+void CAIRO_COMPOSITOR::Resize( unsigned int aWidth, unsigned int aHeight )
+{
+    clean();
+
+    assert( aWidth > 0 );
+    assert( aHeight > 0 );
+
+    m_width  = aWidth;
+    m_height = aHeight;
+
+    m_stride     = cairo_format_stride_for_width( CAIRO_FORMAT_ARGB32, m_width );
+    m_bufferSize = m_stride * m_height;
+}
+
+
+unsigned int CAIRO_COMPOSITOR::CreateBuffer()
+{
+    // Pixel storage
+    BitmapPtr bitmap( new unsigned int[m_bufferSize] );
+
+    memset( bitmap.get(), 0x00, m_bufferSize * sizeof(int) );
+
+    // Create the Cairo surface
+    cairo_surface_t* surface = cairo_image_surface_create_for_data(
+                                                        (unsigned char*) bitmap.get(),
+                                                        CAIRO_FORMAT_ARGB32, m_width,
+                                                        m_height, m_stride );
+    cairo_t* context = cairo_create( surface );
+#ifdef __WXDEBUG__
+    cairo_status_t status = cairo_status( context );
+    wxASSERT_MSG( status == CAIRO_STATUS_SUCCESS, wxT( "Cairo context creation error" ) );
+#endif /* __WXDEBUG__ */
+
+    // Set default settings for the buffer
+    cairo_set_antialias( context, CAIRO_ANTIALIAS_SUBPIXEL );
+    cairo_set_line_join( context, CAIRO_LINE_JOIN_ROUND );
+    cairo_set_line_cap( context, CAIRO_LINE_CAP_ROUND );
+
+    // Use the same transformation matrix as the main context
+    cairo_get_matrix( m_mainContext, &m_matrix );
+    cairo_set_matrix( context, &m_matrix );
+
+    // Store the new buffer
+    CAIRO_BUFFER buffer = { context, surface, bitmap };
+    m_buffers.push_back( buffer );
+
+    return usedBuffers();
+}
+
+
+void CAIRO_COMPOSITOR::SetBuffer( unsigned int aBufferHandle )
+{
+    wxASSERT_MSG( aBufferHandle <= usedBuffers(), wxT( "Tried to use a not existing buffer" ) );
+
+    // Get currently used transformation matrix, so it can be applied to the new buffer
+    cairo_get_matrix( *m_currentContext, &m_matrix );
+
+    m_current = aBufferHandle - 1;
+    *m_currentContext = m_buffers[m_current].context;
+
+    // Apply the current transformation matrix
+    cairo_set_matrix( *m_currentContext, &m_matrix );
+}
+
+
+void CAIRO_COMPOSITOR::ClearBuffer()
+{
+    // Clear the pixel storage
+    memset( m_buffers[m_current].bitmap.get(), 0x00, m_bufferSize * sizeof(int) );
+}
+
+
+void CAIRO_COMPOSITOR::DrawBuffer( unsigned int aBufferHandle )
+{
+    wxASSERT_MSG( aBufferHandle <= usedBuffers(), wxT( "Tried to use a not existing buffer" ) );
+
+    // Reset the transformation matrix, so it is possible to composite images using
+    // screen coordinates instead of world coordinates
+    cairo_get_matrix( m_mainContext, &m_matrix );
+    cairo_identity_matrix( m_mainContext );
+
+    // Draw the selected buffer contents
+    cairo_set_source_surface( m_mainContext, m_buffers[aBufferHandle - 1].surface, 0.0, 0.0 );
+    cairo_paint( m_mainContext );
+
+    // Restore the transformation matrix
+    cairo_set_matrix( m_mainContext, &m_matrix );
+}
+
+
+void CAIRO_COMPOSITOR::clean()
+{
+    CAIRO_BUFFERS::const_iterator it;
+
+    for( it = m_buffers.begin(); it != m_buffers.end(); ++it )
+    {
+        cairo_destroy( it->context );
+        cairo_surface_destroy( it->surface );
+    }
+
+    m_buffers.clear();
+}
diff --git a/common/gal/cairo/cairo_gal.cpp b/common/gal/cairo/cairo_gal.cpp
new file mode 100644
index 0000000..b7ead4c
--- /dev/null
+++ b/common/gal/cairo/cairo_gal.cpp
@@ -0,0 +1,1074 @@
+/*
+ * This program source code file is part of KICAD, a free EDA CAD application.
+ *
+ * Copyright (C) 2012 Torsten Hueter, torstenhtr <at> gmx.de
+ * Copyright (C) 2012 Kicad Developers, see change_log.txt for contributors.
+ *
+ * CAIRO_GAL - Graphics Abstraction Layer for Cairo
+ *
+ * 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 <wx/image.h>
+#include <wx/log.h>
+
+#include <gal/cairo/cairo_gal.h>
+#include <gal/cairo/cairo_compositor.h>
+#include <gal/definitions.h>
+
+#include <limits>
+
+using namespace KIGFX;
+
+
+const float CAIRO_GAL::LAYER_ALPHA = 0.8;
+
+
+CAIRO_GAL::CAIRO_GAL( wxWindow* aParent, wxEvtHandler* aMouseListener,
+        wxEvtHandler* aPaintListener, const wxString& aName ) :
+    wxWindow( aParent, wxID_ANY, wxDefaultPosition, wxDefaultSize, wxEXPAND, aName )
+{
+    parentWindow  = aParent;
+    mouseListener = aMouseListener;
+    paintListener = aPaintListener;
+
+    // Initialize the flags
+    isGrouping          = false;
+    isInitialized       = false;
+    isDeleteSavedPixels = false;
+    validCompositor     = false;
+    groupCounter        = 0;
+
+    // Connecting the event handlers
+    Connect( wxEVT_PAINT,       wxPaintEventHandler( CAIRO_GAL::onPaint ) );
+
+    // Mouse events are skipped to the parent
+    Connect( wxEVT_MOTION,          wxMouseEventHandler( CAIRO_GAL::skipMouseEvent ) );
+    Connect( wxEVT_LEFT_DOWN,       wxMouseEventHandler( CAIRO_GAL::skipMouseEvent ) );
+    Connect( wxEVT_LEFT_UP,         wxMouseEventHandler( CAIRO_GAL::skipMouseEvent ) );
+    Connect( wxEVT_LEFT_DCLICK,     wxMouseEventHandler( CAIRO_GAL::skipMouseEvent ) );
+    Connect( wxEVT_MIDDLE_DOWN,     wxMouseEventHandler( CAIRO_GAL::skipMouseEvent ) );
+    Connect( wxEVT_MIDDLE_UP,       wxMouseEventHandler( CAIRO_GAL::skipMouseEvent ) );
+    Connect( wxEVT_MIDDLE_DCLICK,   wxMouseEventHandler( CAIRO_GAL::skipMouseEvent ) );
+    Connect( wxEVT_RIGHT_DOWN,      wxMouseEventHandler( CAIRO_GAL::skipMouseEvent ) );
+    Connect( wxEVT_RIGHT_UP,        wxMouseEventHandler( CAIRO_GAL::skipMouseEvent ) );
+    Connect( wxEVT_RIGHT_DCLICK,    wxMouseEventHandler( CAIRO_GAL::skipMouseEvent ) );
+    Connect( wxEVT_MOUSEWHEEL,      wxMouseEventHandler( CAIRO_GAL::skipMouseEvent ) );
+#if defined _WIN32 || defined _WIN64
+    Connect( wxEVT_ENTER_WINDOW,    wxMouseEventHandler( CAIRO_GAL::skipMouseEvent ) );
+#endif
+
+    SetSize( aParent->GetSize() );
+    screenSize = VECTOR2I( aParent->GetSize() );
+
+    cursorPixels = NULL;
+    cursorPixelsSaved = NULL;
+    initCursor();
+
+    // Grid color settings are different in Cairo and OpenGL
+    SetGridColor( COLOR4D( 0.1, 0.1, 0.1, 0.8 ) );
+
+    // Allocate memory for pixel storage
+    allocateBitmaps();
+
+    initSurface();
+}
+
+
+CAIRO_GAL::~CAIRO_GAL()
+{
+    deinitSurface();
+    deleteBitmaps();
+
+    delete cursorPixels;
+    delete cursorPixelsSaved;
+
+    ClearCache();
+}
+
+
+void CAIRO_GAL::BeginDrawing()
+{
+    initSurface();
+
+    if( !validCompositor )
+        setCompositor();
+
+    compositor->SetMainContext( context );
+    compositor->SetBuffer( mainBuffer );
+
+    // Cairo grouping prevents display of overlapping items on the same layer in the lighter color
+    cairo_push_group( currentContext );
+}
+
+
+void CAIRO_GAL::EndDrawing()
+{
+    // Force remaining objects to be drawn
+    Flush();
+
+    // Cairo grouping prevents display of overlapping items on the same layer in the lighter color
+    cairo_pop_group_to_source( currentContext );
+    cairo_paint_with_alpha( currentContext, LAYER_ALPHA );
+
+    // Merge buffers on the screen
+    compositor->DrawBuffer( mainBuffer );
+    compositor->DrawBuffer( overlayBuffer );
+
+    // This code was taken from the wxCairo example - it's not the most efficient one
+    // Here is a good place for optimizations
+
+    // Now translate the raw context data from the format stored
+    // by cairo into a format understood by wxImage.
+    unsigned char* wxOutputPtr = wxOutput;
+
+    for( size_t count = 0; count < bufferSize; count++ )
+    {
+        unsigned int value = bitmapBuffer[count];
+        *wxOutputPtr++ = ( value >> 16 ) & 0xff;  // Red pixel
+        *wxOutputPtr++ = ( value >> 8 ) & 0xff;   // Green pixel
+        *wxOutputPtr++ = value & 0xff;            // Blue pixel
+    }
+
+    wxImage      img( screenSize.x, screenSize.y, (unsigned char*) wxOutput, true );
+    wxBitmap     bmp( img );
+    wxClientDC   client_dc( this );
+    wxBufferedDC dc;
+    dc.Init( &client_dc, bmp );
+
+    // Now it is the time to blit the mouse cursor
+    blitCursor( dc );
+
+    deinitSurface();
+}
+
+
+void CAIRO_GAL::DrawLine( const VECTOR2D& aStartPoint, const VECTOR2D& aEndPoint )
+{
+    cairo_move_to( currentContext, aStartPoint.x, aStartPoint.y );
+    cairo_line_to( currentContext, aEndPoint.x, aEndPoint.y );
+    isElementAdded = true;
+}
+
+
+void CAIRO_GAL::DrawSegment( const VECTOR2D& aStartPoint, const VECTOR2D& aEndPoint,
+                             double aWidth )
+{
+    if( isFillEnabled )
+    {
+        // Filled tracks mode
+        SetLineWidth( aWidth );
+
+        cairo_move_to( currentContext, (double) aStartPoint.x, (double) aStartPoint.y );
+        cairo_line_to( currentContext, (double) aEndPoint.x, (double) aEndPoint.y );
+    }
+    else
+    {
+        // Outline mode for tracks
+        VECTOR2D startEndVector = aEndPoint - aStartPoint;
+        double   lineAngle      = atan2( startEndVector.y, startEndVector.x );
+        double   lineLength     = startEndVector.EuclideanNorm();
+
+        cairo_save( currentContext );
+
+        cairo_translate( currentContext, aStartPoint.x, aStartPoint.y );
+        cairo_rotate( currentContext, lineAngle );
+
+        cairo_arc( currentContext, 0.0,        0.0, aWidth / 2.0,  M_PI / 2.0, 3.0 * M_PI / 2.0 );
+        cairo_arc( currentContext, lineLength, 0.0, aWidth / 2.0, -M_PI / 2.0, M_PI / 2.0 );
+
+        cairo_move_to( currentContext, 0.0,        aWidth / 2.0 );
+        cairo_line_to( currentContext, lineLength, aWidth / 2.0 );
+
+        cairo_move_to( currentContext, 0.0,        -aWidth / 2.0 );
+        cairo_line_to( currentContext, lineLength, -aWidth / 2.0 );
+
+        cairo_restore( currentContext );
+    }
+
+    isElementAdded = true;
+}
+
+
+void CAIRO_GAL::DrawCircle( const VECTOR2D& aCenterPoint, double aRadius )
+{
+    // A circle is drawn using an arc
+    cairo_new_sub_path( currentContext );
+    cairo_arc( currentContext, aCenterPoint.x, aCenterPoint.y, aRadius, 0.0, 2 * M_PI );
+
+    isElementAdded = true;
+}
+
+
+void CAIRO_GAL::DrawArc( const VECTOR2D& aCenterPoint, double aRadius, double aStartAngle,
+                         double aEndAngle )
+{
+    SWAP( aStartAngle, >, aEndAngle );
+
+    cairo_new_sub_path( currentContext );
+    cairo_arc( currentContext, aCenterPoint.x, aCenterPoint.y, aRadius, aStartAngle, aEndAngle );
+
+    if( isFillEnabled )
+    {
+        VECTOR2D startPoint( cos( aStartAngle ) * aRadius + aCenterPoint.x,
+                             sin( aStartAngle ) * aRadius + aCenterPoint.y );
+        VECTOR2D endPoint( cos( aEndAngle ) * aRadius + aCenterPoint.x,
+                           sin( aEndAngle ) * aRadius + aCenterPoint.y );
+
+        cairo_move_to( currentContext, aCenterPoint.x, aCenterPoint.y );
+        cairo_line_to( currentContext, startPoint.x, startPoint.y );
+        cairo_line_to( currentContext, endPoint.x, endPoint.y );
+        cairo_close_path( currentContext );
+    }
+
+    isElementAdded = true;
+}
+
+
+void CAIRO_GAL::DrawRectangle( const VECTOR2D& aStartPoint, const VECTOR2D& aEndPoint )
+{
+    // Calculate the diagonal points
+    VECTOR2D diagonalPointA( aEndPoint.x,  aStartPoint.y );
+    VECTOR2D diagonalPointB( aStartPoint.x, aEndPoint.y );
+
+    // The path is composed from 4 segments
+    cairo_move_to( currentContext, aStartPoint.x, aStartPoint.y );
+    cairo_line_to( currentContext, diagonalPointA.x, diagonalPointA.y );
+    cairo_line_to( currentContext, aEndPoint.x, aEndPoint.y );
+    cairo_line_to( currentContext, diagonalPointB.x, diagonalPointB.y );
+    cairo_close_path( currentContext );
+
+    isElementAdded = true;
+}
+
+
+void CAIRO_GAL::DrawCurve( const VECTOR2D& aStartPoint, const VECTOR2D& aControlPointA,
+                           const VECTOR2D& aControlPointB, const VECTOR2D& aEndPoint )
+{
+    cairo_move_to( currentContext, aStartPoint.x, aStartPoint.y );
+    cairo_curve_to( currentContext, aControlPointA.x, aControlPointA.y, aControlPointB.x,
+                    aControlPointB.y, aEndPoint.x, aEndPoint.y );
+    cairo_line_to( currentContext, aEndPoint.x, aEndPoint.y );
+
+    isElementAdded = true;
+}
+
+
+void CAIRO_GAL::ResizeScreen( int aWidth, int aHeight )
+{
+    screenSize = VECTOR2I( aWidth, aHeight );
+
+    // Recreate the bitmaps
+    deleteBitmaps();
+    allocateBitmaps();
+
+    if( validCompositor )
+        compositor->Resize( aWidth, aHeight );
+
+    validCompositor = false;
+
+    SetSize( wxSize( aWidth, aHeight ) );
+}
+
+
+bool CAIRO_GAL::Show( bool aShow )
+{
+    bool s = wxWindow::Show( aShow );
+
+    if( aShow )
+        wxWindow::Raise();
+
+    return s;
+}
+
+
+void CAIRO_GAL::Flush()
+{
+    storePath();
+}
+
+
+void CAIRO_GAL::ClearScreen( const COLOR4D& aColor )
+{
+    backgroundColor = aColor;
+    cairo_set_source_rgb( currentContext, aColor.r, aColor.g, aColor.b );
+    cairo_rectangle( currentContext, 0.0, 0.0, screenSize.x, screenSize.y );
+    cairo_fill( currentContext );
+}
+
+
+void CAIRO_GAL::SetIsFill( bool aIsFillEnabled )
+{
+    storePath();
+    isFillEnabled = aIsFillEnabled;
+
+    if( isGrouping )
+    {
+        GROUP_ELEMENT groupElement;
+        groupElement.command = CMD_SET_FILL;
+        groupElement.boolArgument = aIsFillEnabled;
+        currentGroup->push_back( groupElement );
+    }
+}
+
+
+void CAIRO_GAL::SetIsStroke( bool aIsStrokeEnabled )
+{
+    storePath();
+    isStrokeEnabled = aIsStrokeEnabled;
+
+    if( isGrouping )
+    {
+        GROUP_ELEMENT groupElement;
+        groupElement.command = CMD_SET_STROKE;
+        groupElement.boolArgument = aIsStrokeEnabled;
+        currentGroup->push_back( groupElement );
+    }
+}
+
+
+void CAIRO_GAL::SetStrokeColor( const COLOR4D& aColor )
+{
+    storePath();
+    strokeColor = aColor;
+
+    if( isGrouping )
+    {
+        GROUP_ELEMENT groupElement;
+        groupElement.command = CMD_SET_STROKECOLOR;
+        groupElement.arguments[0] = strokeColor.r;
+        groupElement.arguments[1] = strokeColor.g;
+        groupElement.arguments[2] = strokeColor.b;
+        groupElement.arguments[3] = strokeColor.a;
+        currentGroup->push_back( groupElement );
+    }
+}
+
+
+void CAIRO_GAL::SetFillColor( const COLOR4D& aColor )
+{
+    storePath();
+    fillColor = aColor;
+
+    if( isGrouping )
+    {
+        GROUP_ELEMENT groupElement;
+        groupElement.command = CMD_SET_FILLCOLOR;
+        groupElement.arguments[0] = fillColor.r;
+        groupElement.arguments[1] = fillColor.g;
+        groupElement.arguments[2] = fillColor.b;
+        groupElement.arguments[3] = fillColor.a;
+        currentGroup->push_back( groupElement );
+    }
+}
+
+
+void CAIRO_GAL::SetLineWidth( double aLineWidth )
+{
+    storePath();
+
+    lineWidth = aLineWidth;
+
+    if( isGrouping )
+    {
+        GROUP_ELEMENT groupElement;
+        groupElement.command = CMD_SET_LINE_WIDTH;
+        groupElement.arguments[0] = aLineWidth;
+        currentGroup->push_back( groupElement );
+    }
+    else
+    {
+        // Make lines appear at least 1 pixel wide, no matter of zoom
+        double x = 1.0, y = 1.0;
+        cairo_device_to_user_distance( currentContext, &x, &y );
+        double minWidth = std::min( fabs( x ), fabs( y ) );
+        cairo_set_line_width( currentContext, std::max( aLineWidth, minWidth ) );
+    }
+}
+
+
+void CAIRO_GAL::SetLayerDepth( double aLayerDepth )
+{
+    super::SetLayerDepth( aLayerDepth );
+
+    if( isInitialized )
+    {
+        storePath();
+
+        cairo_pop_group_to_source( currentContext );
+        cairo_paint_with_alpha( currentContext, LAYER_ALPHA );
+
+        cairo_push_group( currentContext );
+    }
+}
+
+
+void CAIRO_GAL::Transform( const MATRIX3x3D& aTransformation )
+{
+    cairo_matrix_t cairoTransformation;
+
+    cairo_matrix_init( &cairoTransformation,
+                       aTransformation.m_data[0][0],
+                       aTransformation.m_data[1][0],
+                       aTransformation.m_data[0][1],
+                       aTransformation.m_data[1][1],
+                       aTransformation.m_data[0][2],
+                       aTransformation.m_data[1][2] );
+
+    cairo_transform( currentContext, &cairoTransformation );
+}
+
+
+void CAIRO_GAL::Rotate( double aAngle )
+{
+    storePath();
+
+    if( isGrouping )
+    {
+        GROUP_ELEMENT groupElement;
+        groupElement.command = CMD_ROTATE;
+        groupElement.arguments[0] = aAngle;
+        currentGroup->push_back( groupElement );
+    }
+    else
+    {
+        cairo_rotate( currentContext, aAngle );
+    }
+}
+
+
+void CAIRO_GAL::Translate( const VECTOR2D& aTranslation )
+{
+    storePath();
+
+    if( isGrouping )
+    {
+        GROUP_ELEMENT groupElement;
+        groupElement.command = CMD_TRANSLATE;
+        groupElement.arguments[0] = aTranslation.x;
+        groupElement.arguments[1] = aTranslation.y;
+        currentGroup->push_back( groupElement );
+    }
+    else
+    {
+        cairo_translate( currentContext, aTranslation.x, aTranslation.y );
+    }
+}
+
+
+void CAIRO_GAL::Scale( const VECTOR2D& aScale )
+{
+    storePath();
+
+    if( isGrouping )
+    {
+        GROUP_ELEMENT groupElement;
+        groupElement.command = CMD_SCALE;
+        groupElement.arguments[0] = aScale.x;
+        groupElement.arguments[1] = aScale.y;
+        currentGroup->push_back( groupElement );
+    }
+    else
+    {
+        cairo_scale( currentContext, aScale.x, aScale.y );
+    }
+}
+
+
+void CAIRO_GAL::Save()
+{
+    storePath();
+
+    if( isGrouping )
+    {
+        GROUP_ELEMENT groupElement;
+        groupElement.command = CMD_SAVE;
+        currentGroup->push_back( groupElement );
+    }
+    else
+    {
+        cairo_save( currentContext );
+    }
+}
+
+
+void CAIRO_GAL::Restore()
+{
+    storePath();
+
+    if( isGrouping )
+    {
+        GROUP_ELEMENT groupElement;
+        groupElement.command = CMD_RESTORE;
+        currentGroup->push_back( groupElement );
+    }
+    else
+    {
+        cairo_restore( currentContext );
+    }
+}
+
+
+int CAIRO_GAL::BeginGroup()
+{
+    initSurface();
+
+    // If the grouping is started: the actual path is stored in the group, when
+    // a attribute was changed or when grouping stops with the end group method.
+    storePath();
+
+    GROUP group;
+    int groupNumber = getNewGroupNumber();
+    groups.insert( std::make_pair( groupNumber, group ) );
+    currentGroup = &groups[groupNumber];
+    isGrouping   = true;
+
+    return groupNumber;
+}
+
+
+void CAIRO_GAL::EndGroup()
+{
+    storePath();
+    isGrouping = false;
+
+    deinitSurface();
+}
+
+
+void CAIRO_GAL::DrawGroup( int aGroupNumber )
+{
+    // This method implements a small Virtual Machine - all stored commands
+    // are executed; nested calling is also possible
+
+    storePath();
+
+    for( GROUP::iterator it = groups[aGroupNumber].begin();
+         it != groups[aGroupNumber].end(); ++it )
+    {
+        switch( it->command )
+        {
+        case CMD_SET_FILL:
+            isFillEnabled = it->boolArgument;
+            break;
+
+        case CMD_SET_STROKE:
+            isStrokeEnabled = it->boolArgument;
+            break;
+
+        case CMD_SET_FILLCOLOR:
+            fillColor = COLOR4D( it->arguments[0], it->arguments[1], it->arguments[2],
+                                 it->arguments[3] );
+            break;
+
+        case CMD_SET_STROKECOLOR:
+            strokeColor = COLOR4D( it->arguments[0], it->arguments[1], it->arguments[2],
+                                   it->arguments[3] );
+            break;
+
+        case CMD_SET_LINE_WIDTH:
+            {
+                // Make lines appear at least 1 pixel wide, no matter of zoom
+                double x = 1.0, y = 1.0;
+                cairo_device_to_user_distance( currentContext, &x, &y );
+                double minWidth = std::min( fabs( x ), fabs( y ) );
+                cairo_set_line_width( currentContext, std::max( it->arguments[0], minWidth ) );
+            }
+            break;
+
+
+        case CMD_STROKE_PATH:
+            cairo_set_source_rgb( currentContext, strokeColor.r, strokeColor.g, strokeColor.b );
+            cairo_append_path( currentContext, it->cairoPath );
+            cairo_stroke( currentContext );
+            break;
+
+        case CMD_FILL_PATH:
+            cairo_set_source_rgb( currentContext, fillColor.r, fillColor.g, fillColor.b );
+            cairo_append_path( currentContext, it->cairoPath );
+            cairo_fill( currentContext );
+            break;
+
+        case CMD_TRANSFORM:
+            cairo_matrix_t matrix;
+            cairo_matrix_init( &matrix, it->arguments[0], it->arguments[1], it->arguments[2],
+                               it->arguments[3], it->arguments[4], it->arguments[5] );
+            cairo_transform( currentContext, &matrix );
+            break;
+
+        case CMD_ROTATE:
+            cairo_rotate( currentContext, it->arguments[0] );
+            break;
+
+        case CMD_TRANSLATE:
+            cairo_translate( currentContext, it->arguments[0], it->arguments[1] );
+            break;
+
+        case CMD_SCALE:
+            cairo_scale( currentContext, it->arguments[0], it->arguments[1] );
+            break;
+
+        case CMD_SAVE:
+            cairo_save( currentContext );
+            break;
+
+        case CMD_RESTORE:
+            cairo_restore( currentContext );
+            break;
+
+        case CMD_CALL_GROUP:
+            DrawGroup( it->intArgument );
+            break;
+        }
+    }
+}
+
+
+void CAIRO_GAL::ChangeGroupColor( int aGroupNumber, const COLOR4D& aNewColor )
+{
+    storePath();
+
+    for( GROUP::iterator it = groups[aGroupNumber].begin();
+         it != groups[aGroupNumber].end(); ++it )
+    {
+        if( it->command == CMD_SET_FILLCOLOR || it->command == CMD_SET_STROKECOLOR )
+        {
+            it->arguments[0] = aNewColor.r;
+            it->arguments[1] = aNewColor.g;
+            it->arguments[2] = aNewColor.b;
+            it->arguments[3] = aNewColor.a;
+        }
+    }
+}
+
+
+void CAIRO_GAL::ChangeGroupDepth( int aGroupNumber, int aDepth )
+{
+    // Cairo does not have any possibilities to change the depth coordinate of stored items,
+    // it depends only on the order of drawing
+}
+
+
+void CAIRO_GAL::DeleteGroup( int aGroupNumber )
+{
+    storePath();
+
+    // Delete the Cairo paths
+    std::deque<GROUP_ELEMENT>::iterator it, end;
+
+    for( it = groups[aGroupNumber].begin(), end = groups[aGroupNumber].end(); it != end; ++it )
+    {
+        if( it->command == CMD_FILL_PATH || it->command == CMD_STROKE_PATH )
+        {
+            cairo_path_destroy( it->cairoPath );
+        }
+    }
+
+    // Delete the group
+    groups.erase( aGroupNumber );
+}
+
+
+void CAIRO_GAL::ClearCache()
+{
+    for( int i = groups.size() - 1; i >= 0; --i )
+    {
+        DeleteGroup( i );
+    }
+}
+
+
+void CAIRO_GAL::SaveScreen()
+{
+    // Copy the current bitmap to the backup buffer
+    int offset = 0;
+
+    for( int j = 0; j < screenSize.y; j++ )
+    {
+        for( int i = 0; i < stride; i++ )
+        {
+            bitmapBufferBackup[offset + i] = bitmapBuffer[offset + i];
+            offset += stride;
+        }
+    }
+}
+
+
+void CAIRO_GAL::RestoreScreen()
+{
+    int offset = 0;
+
+    for( int j = 0; j < screenSize.y; j++ )
+    {
+        for( int i = 0; i < stride; i++ )
+        {
+            bitmapBuffer[offset + i] = bitmapBufferBackup[offset + i];
+            offset += stride;
+        }
+    }
+}
+
+
+void CAIRO_GAL::SetTarget( RENDER_TARGET aTarget )
+{
+    // If the compositor is not set, that means that there is a recaching process going on
+    // and we do not need the compositor now
+    if( !validCompositor )
+        return;
+
+    // Cairo grouping prevents display of overlapping items on the same layer in the lighter color
+    if( isInitialized )
+    {
+        storePath();
+
+        cairo_pop_group_to_source( currentContext );
+        cairo_paint_with_alpha( currentContext, LAYER_ALPHA );
+    }
+
+    switch( aTarget )
+    {
+    default:
+    case TARGET_CACHED:
+    case TARGET_NONCACHED:
+        compositor->SetBuffer( mainBuffer );
+        break;
+
+    case TARGET_OVERLAY:
+        compositor->SetBuffer( overlayBuffer );
+        break;
+    }
+
+    if( isInitialized )
+        cairo_push_group( currentContext );
+
+    currentTarget = aTarget;
+}
+
+
+RENDER_TARGET CAIRO_GAL::GetTarget() const
+{
+    return currentTarget;
+}
+
+
+void CAIRO_GAL::ClearTarget( RENDER_TARGET aTarget )
+{
+    // Save the current state
+    unsigned int currentBuffer = compositor->GetBuffer();
+
+    switch( aTarget )
+    {
+    // Cached and noncached items are rendered to the same buffer
+    default:
+    case TARGET_CACHED:
+    case TARGET_NONCACHED:
+        compositor->SetBuffer( mainBuffer );
+        break;
+
+    case TARGET_OVERLAY:
+        compositor->SetBuffer( overlayBuffer );
+        break;
+    }
+
+    compositor->ClearBuffer();
+
+    // Restore the previous state
+    compositor->SetBuffer( currentBuffer );
+}
+
+
+void CAIRO_GAL::SetCursorSize( unsigned int aCursorSize )
+{
+    GAL::SetCursorSize( aCursorSize );
+    initCursor();
+}
+
+
+void CAIRO_GAL::DrawCursor( const VECTOR2D& aCursorPosition )
+{
+    // Now we should only store the position of the mouse cursor
+    // The real drawing routines are in blitCursor()
+    cursorPosition = VECTOR2D( aCursorPosition.x - cursorSize / 2,
+                               aCursorPosition.y - cursorSize / 2 );
+}
+
+
+void CAIRO_GAL::drawGridLine( const VECTOR2D& aStartPoint, const VECTOR2D& aEndPoint )
+{
+    cairo_move_to( currentContext, aStartPoint.x, aStartPoint.y );
+    cairo_line_to( currentContext, aEndPoint.x, aEndPoint.y );
+    cairo_set_source_rgb( currentContext, gridColor.r, gridColor.g, gridColor.b );
+    cairo_stroke( currentContext );
+}
+
+
+void CAIRO_GAL::storePath()
+{
+    if( isElementAdded )
+    {
+        isElementAdded = false;
+
+        if( !isGrouping )
+        {
+            if( isFillEnabled )
+            {
+                cairo_set_source_rgb( currentContext, fillColor.r, fillColor.g, fillColor.b );
+                cairo_fill_preserve( currentContext );
+            }
+
+            if( isStrokeEnabled )
+            {
+                cairo_set_source_rgb( currentContext, strokeColor.r, strokeColor.g,
+                                      strokeColor.b );
+                cairo_stroke_preserve( currentContext );
+            }
+        }
+        else
+        {
+            // Copy the actual path, append it to the global path list
+            // then check, if the path needs to be stroked/filled and
+            // add this command to the group list;
+            if( isStrokeEnabled )
+            {
+                GROUP_ELEMENT groupElement;
+                groupElement.cairoPath = cairo_copy_path( currentContext );
+                groupElement.command   = CMD_STROKE_PATH;
+                currentGroup->push_back( groupElement );
+            }
+
+            if( isFillEnabled )
+            {
+                GROUP_ELEMENT groupElement;
+                groupElement.cairoPath = cairo_copy_path( currentContext );
+                groupElement.command   = CMD_FILL_PATH;
+                currentGroup->push_back( groupElement );
+            }
+        }
+
+        cairo_new_path( currentContext );
+    }
+}
+
+
+void CAIRO_GAL::onPaint( wxPaintEvent& WXUNUSED( aEvent ) )
+{
+    PostPaint();
+}
+
+
+void CAIRO_GAL::skipMouseEvent( wxMouseEvent& aEvent )
+{
+    // Post the mouse event to the event listener registered in constructor, if any
+    if( mouseListener )
+        wxPostEvent( mouseListener, aEvent );
+}
+
+
+void CAIRO_GAL::initCursor()
+{
+    if( cursorPixels )
+        delete cursorPixels;
+
+    if( cursorPixelsSaved )
+        delete cursorPixelsSaved;
+
+    cursorPixels      = new wxBitmap( cursorSize, cursorSize );
+    cursorPixelsSaved = new wxBitmap( cursorSize, cursorSize );
+
+    wxMemoryDC cursorShape( *cursorPixels );
+
+    cursorShape.SetBackground( *wxTRANSPARENT_BRUSH );
+    wxColour color( cursorColor.r * cursorColor.a * 255, cursorColor.g * cursorColor.a * 255,
+                    cursorColor.b * cursorColor.a * 255, 255 );
+    wxPen pen = wxPen( color );
+    cursorShape.SetPen( pen );
+    cursorShape.Clear();
+
+    cursorShape.DrawLine( 0, cursorSize / 2, cursorSize, cursorSize / 2 );
+    cursorShape.DrawLine( cursorSize / 2, 0, cursorSize / 2, cursorSize );
+}
+
+
+void CAIRO_GAL::blitCursor( wxBufferedDC& clientDC )
+{
+    if( !isCursorEnabled )
+        return;
+
+    wxMemoryDC cursorSave( *cursorPixelsSaved );
+    wxMemoryDC cursorShape( *cursorPixels );
+
+    if( !isDeleteSavedPixels )
+    {
+        // Restore pixels that were overpainted by the previous cursor
+        clientDC.Blit( savedCursorPosition.x, savedCursorPosition.y,
+                       cursorSize, cursorSize, &cursorSave, 0, 0 );
+    }
+    else
+    {
+        isDeleteSavedPixels = false;
+    }
+
+    // Store pixels that are going to be overpainted
+    VECTOR2D cursorScreen = ToScreen( cursorPosition ) - cursorSize / 2.0f;
+    cursorSave.Blit( 0, 0, cursorSize, cursorSize, &clientDC, cursorScreen.x, cursorScreen.y );
+
+    // Draw the cursor
+    clientDC.Blit( cursorScreen.x, cursorScreen.y, cursorSize, cursorSize,
+                   &cursorShape, 0, 0, wxOR );
+
+    savedCursorPosition.x = (wxCoord) cursorScreen.x;
+    savedCursorPosition.y = (wxCoord) cursorScreen.y;
+}
+
+
+void CAIRO_GAL::allocateBitmaps()
+{
+    // Create buffer, use the system independent Cairo context backend
+    stride     = cairo_format_stride_for_width( GAL_FORMAT, screenSize.x );
+    bufferSize = stride * screenSize.y;
+
+    bitmapBuffer        = new unsigned int[bufferSize];
+    bitmapBufferBackup  = new unsigned int[bufferSize];
+    wxOutput            = new unsigned char[bufferSize * 3];
+}
+
+
+void CAIRO_GAL::deleteBitmaps()
+{
+    delete[] bitmapBuffer;
+    delete[] bitmapBufferBackup;
+    delete[] wxOutput;
+}
+
+
+void CAIRO_GAL::initSurface()
+{
+    if( isInitialized )
+        return;
+
+    // Create the Cairo surface
+    surface = cairo_image_surface_create_for_data( (unsigned char*) bitmapBuffer, GAL_FORMAT,
+                                                   screenSize.x, screenSize.y, stride );
+    context = cairo_create( surface );
+#ifdef __WXDEBUG__
+    cairo_status_t status = cairo_status( context );
+    wxASSERT_MSG( status == CAIRO_STATUS_SUCCESS, wxT( "Cairo context creation error" ) );
+#endif /* __WXDEBUG__ */
+    currentContext = context;
+
+    cairo_set_antialias( context, CAIRO_ANTIALIAS_SUBPIXEL );
+
+    // Clear the screen
+    ClearScreen( backgroundColor );
+
+    // Compute the world <-> screen transformations
+    ComputeWorldScreenMatrix();
+
+    cairo_matrix_init( &cairoWorldScreenMatrix, worldScreenMatrix.m_data[0][0],
+                       worldScreenMatrix.m_data[1][0], worldScreenMatrix.m_data[0][1],
+                       worldScreenMatrix.m_data[1][1], worldScreenMatrix.m_data[0][2],
+                       worldScreenMatrix.m_data[1][2] );
+
+    cairo_set_matrix( context, &cairoWorldScreenMatrix );
+
+    // Start drawing with a new path
+    cairo_new_path( context );
+    isElementAdded = true;
+
+    cairo_set_line_join( context, CAIRO_LINE_JOIN_ROUND );
+    cairo_set_line_cap( context, CAIRO_LINE_CAP_ROUND );
+
+    lineWidth = 0;
+
+    isDeleteSavedPixels = true;
+    isInitialized = true;
+}
+
+
+void CAIRO_GAL::deinitSurface()
+{
+    if( !isInitialized )
+        return;
+
+    // Destroy Cairo objects
+    cairo_destroy( context );
+    cairo_surface_destroy( surface );
+
+    isInitialized = false;
+}
+
+
+void CAIRO_GAL::setCompositor()
+{
+    // Recreate the compositor with the new Cairo context
+    compositor.reset( new CAIRO_COMPOSITOR( &currentContext ) );
+    compositor->Resize( screenSize.x, screenSize.y );
+
+    // Prepare buffers
+    mainBuffer = compositor->CreateBuffer();
+    overlayBuffer = compositor->CreateBuffer();
+
+    validCompositor = true;
+}
+
+
+void CAIRO_GAL::drawPoly( const std::deque<VECTOR2D>& aPointList )
+{
+    // Iterate over the point list and draw the segments
+    std::deque<VECTOR2D>::const_iterator it = aPointList.begin();
+
+    cairo_move_to( currentContext, it->x, it->y );
+
+    for( ++it; it != aPointList.end(); ++it )
+    {
+        cairo_line_to( currentContext, it->x, it->y );
+    }
+
+    isElementAdded = true;
+}
+
+
+void CAIRO_GAL::drawPoly( const VECTOR2D aPointList[], int aListSize )
+{
+    // Iterate over the point list and draw the segments
+    const VECTOR2D* ptr = aPointList;
+
+    cairo_move_to( currentContext, ptr->x, ptr->y );
+
+    for( int i = 0; i < aListSize; ++i )
+    {
+        ++ptr;
+        cairo_line_to( currentContext, ptr->x, ptr->y );
+    }
+
+    isElementAdded = true;
+}
+
+
+unsigned int CAIRO_GAL::getNewGroupNumber()
+{
+    wxASSERT_MSG( groups.size() < std::numeric_limits<unsigned int>::max(),
+                  wxT( "There are no free slots to store a group" ) );
+
+    while( groups.find( groupCounter ) != groups.end() )
+    {
+        groupCounter++;
+    }
+
+    return groupCounter++;
+}
diff --git a/common/gal/color4d.cpp b/common/gal/color4d.cpp
new file mode 100644
index 0000000..3b509d3
--- /dev/null
+++ b/common/gal/color4d.cpp
@@ -0,0 +1,179 @@
+/*
+ * This program source code file is part of KICAD, a free EDA CAD application.
+ *
+ * Copyright (C) 2012 Torsten Hueter, torstenhtr <at> gmx.de
+ * Copyright (C) 2012 Kicad Developers, see change_log.txt for contributors.
+ *
+ * Color class
+ *
+ * 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 <gal/color4d.h>
+
+using namespace KIGFX;
+
+COLOR4D::COLOR4D( EDA_COLOR_T aColor )
+{
+    r = g_ColorRefs[aColor].m_Red / 255.0;
+    g = g_ColorRefs[aColor].m_Green / 255.0;
+    b = g_ColorRefs[aColor].m_Blue / 255.0;
+    a = 1.0;
+}
+
+
+#ifdef WX_COMPATIBILITY
+    COLOR4D::COLOR4D( const wxColour& aColor )
+    {
+        r = aColor.Red();
+        g = aColor.Green();
+        b = aColor.Blue();
+        a = aColor.Alpha();
+    }
+#endif
+
+
+const bool COLOR4D::operator==( const COLOR4D& aColor )
+{
+    return a == aColor.a && r == aColor.r && g == aColor.g && b == aColor.b;
+}
+
+
+const bool COLOR4D::operator!=( const COLOR4D& aColor )
+{
+    return a != aColor.a || r != aColor.r || g != aColor.g || b != aColor.b;
+}
+
+
+void COLOR4D::ToHSV( double& aOutH, double& aOutS, double& aOutV ) const
+{
+    double min, max, delta;
+
+    min = r < g ? r : g;
+    min = min < b ? min : b;
+
+    max = r > g ? r : g;
+    max = max > b ? max : b;
+
+    aOutV = max;                                // v
+    delta = max - min;
+
+    if( max > 0.0 )
+    {
+        aOutS = ( delta / max );                  // s
+    }
+    else
+    {
+        // r = g = b = 0                        // s = 0, v is undefined
+        aOutS = 0.0;
+        aOutH = NAN;                            // its now undefined
+        return;
+    }
+
+    if( r >= max )                          // > is bogus, just keeps compiler happy
+        aOutH = ( g - b ) / delta;          // between yellow & magenta
+    else if( g >= max )
+        aOutH = 2.0 + ( b - r ) / delta;    // between cyan & yellow
+    else
+        aOutH = 4.0 + ( r - g ) / delta;    // between magenta & cyan
+
+    aOutH *= 60.0;                          // degrees
+
+    if( aOutH < 0.0 )
+        aOutH += 360.0;
+}
+
+
+void COLOR4D::FromHSV( double aInH, double aInS, double aInV )
+{
+    double hh, p, q, t, ff;
+    long i;
+
+    if( aInS <= 0.0 )    // < is bogus, just shuts up warnings
+    {
+        r = aInV;
+        g = aInV;
+        b = aInV;
+        return;
+    }
+
+    hh = aInH;
+
+    if( hh >= 360.0 )
+        hh = 0.0;
+
+    hh /= 60.0;
+
+    i = (long) hh;
+    ff = hh - i;
+
+    p = aInV * ( 1.0 - aInS );
+    q = aInV * ( 1.0 - ( aInS * ff ) );
+    t = aInV * ( 1.0 - ( aInS * ( 1.0 - ff ) ) );
+
+    switch( i )
+    {
+    case 0:
+        r = aInV;
+        g = t;
+        b = p;
+        break;
+
+    case 1:
+        r = q;
+        g = aInV;
+        b = p;
+        break;
+
+    case 2:
+        r = p;
+        g = aInV;
+        b = t;
+        break;
+
+    case 3:
+        r = p;
+        g = q;
+        b = aInV;
+        break;
+
+    case 4:
+        r = t;
+        g = p;
+        b = aInV;
+        break;
+
+    case 5:
+    default:
+        r = aInV;
+        g = p;
+        b = q;
+        break;
+    }
+}
+
+
+COLOR4D& COLOR4D::Saturate( double aFactor )
+{
+    double h, s, v;
+
+    ToHSV( h, s, v );
+    FromHSV( h, aFactor, 1.0 );
+
+    return *this;
+}
diff --git a/common/gal/graphics_abstraction_layer.cpp b/common/gal/graphics_abstraction_layer.cpp
new file mode 100644
index 0000000..d9e9eac
--- /dev/null
+++ b/common/gal/graphics_abstraction_layer.cpp
@@ -0,0 +1,240 @@
+/*
+ * This program source code file is part of KICAD, a free EDA CAD application.
+ *
+ * Copyright (C) 2012 Torsten Hueter, torstenhtr <at> gmx.de
+ * Copyright (C) 2012 Kicad Developers, see change_log.txt for contributors.
+ *
+ * Graphics Abstraction Layer (GAL) - base class
+ *
+ * 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 <wx/log.h>
+
+#include <gal/graphics_abstraction_layer.h>
+#include <gal/definitions.h>
+
+using namespace KIGFX;
+
+
+const double GAL::METRIC_UNIT_LENGTH = 1e9;
+
+
+GAL::GAL() :
+    strokeFont( this )
+{
+    // Set the default values for the internal variables
+    SetIsFill( false );
+    SetIsStroke( true );
+    SetFillColor( COLOR4D( 0.0, 0.0, 0.0, 0.0 ) );
+    SetStrokeColor( COLOR4D( 1.0, 1.0, 1.0, 1.0 ) );
+    SetLookAtPoint( VECTOR2D( 0, 0 ) );
+    SetZoomFactor( 1.0 );
+    SetWorldUnitLength( 1.0 / METRIC_UNIT_LENGTH * 2.54 );   // 1 inch in nanometers
+    SetScreenDPI( 106 );                                     // Display resolution setting
+    SetDepthRange( VECTOR2D( GAL::MIN_DEPTH, GAL::MAX_DEPTH ) );
+    SetLayerDepth( 0.0 );
+    SetFlip( false, false );
+    SetLineWidth( 1.0 );
+    ComputeWorldScale();
+
+    // Set grid defaults
+    SetGridVisibility( true );
+    SetGridStyle( GRID_STYLE_LINES );
+    SetGridDrawThreshold( 10 );
+    SetCoarseGrid( 10 );
+    SetGridLineWidth( 0.5 );
+
+    // Initialize the cursor shape
+    SetCursorColor( COLOR4D( 1.0, 1.0, 1.0, 1.0 ) );
+    SetCursorSize( 80 );
+    SetCursorEnabled( false );
+
+    strokeFont.LoadNewStrokeFont( newstroke_font, newstroke_font_bufsize );
+}
+
+
+GAL::~GAL()
+{
+}
+
+
+void GAL::SetTextAttributes( const EDA_TEXT* aText )
+{
+    strokeFont.SetGlyphSize( VECTOR2D( aText->GetSize() ) );
+    strokeFont.SetHorizontalJustify( aText->GetHorizJustify() );
+    strokeFont.SetVerticalJustify( aText->GetVertJustify() );
+    strokeFont.SetBold( aText->IsBold() );
+    strokeFont.SetItalic( aText->IsItalic() );
+    strokeFont.SetMirrored( aText->IsMirrored() );
+}
+
+
+void GAL::ComputeWorldScreenMatrix()
+{
+    ComputeWorldScale();
+
+    worldScreenMatrix.SetIdentity();
+
+    MATRIX3x3D translation;
+    translation.SetIdentity();
+    translation.SetTranslation( 0.5 * VECTOR2D( screenSize ) );
+
+    MATRIX3x3D scale;
+    scale.SetIdentity();
+    scale.SetScale( VECTOR2D( worldScale, worldScale ) );
+
+    MATRIX3x3D flip;
+    flip.SetIdentity();
+    flip.SetScale( VECTOR2D( flipX, flipY ) );
+
+    MATRIX3x3D lookat;
+    lookat.SetIdentity();
+    lookat.SetTranslation( -lookAtPoint );
+
+    worldScreenMatrix = translation * flip * scale * lookat * worldScreenMatrix;
+    screenWorldMatrix = worldScreenMatrix.Inverse();
+}
+
+
+void GAL::DrawGrid()
+{
+    if( !gridVisibility )
+        return;
+
+    SetTarget( TARGET_NONCACHED );
+
+    // Draw the grid
+    // For the drawing the start points, end points and increments have
+    // to be calculated in world coordinates
+    VECTOR2D worldStartPoint = screenWorldMatrix * VECTOR2D( 0.0, 0.0 );
+    VECTOR2D worldEndPoint   = screenWorldMatrix * VECTOR2D( screenSize );
+
+    int gridScreenSizeDense  = KiROUND( gridSize.x * worldScale );
+    int gridScreenSizeCoarse = KiROUND( gridSize.x * static_cast<double>( gridTick ) * worldScale );
+
+    // Compute the line marker or point radius of the grid
+    double marker = 2.0 * gridLineWidth / worldScale;
+    double doubleMarker = 2.0 * marker;
+
+    // Check if the grid would not be too dense
+    if( std::max( gridScreenSizeDense, gridScreenSizeCoarse ) > gridDrawThreshold )
+    {
+        // Compute grid variables
+        int gridStartX  = KiROUND( worldStartPoint.x / gridSize.x );
+        int gridEndX    = KiROUND( worldEndPoint.x / gridSize.x );
+        int gridStartY  = KiROUND( worldStartPoint.y / gridSize.y );
+        int gridEndY    = KiROUND( worldEndPoint.y / gridSize.y );
+
+        assert( gridEndX >= gridStartX );
+        assert( gridEndY >= gridStartY );
+
+        // Correct the index, else some lines are not correctly painted
+        gridStartX -= std::abs( gridOrigin.x / gridSize.x ) + 1;
+        gridStartY -= std::abs( gridOrigin.y / gridSize.y ) + 1;
+        gridEndX += std::abs( gridOrigin.x / gridSize.x ) + 1;
+        gridEndY += std::abs( gridOrigin.y / gridSize.y ) + 1;
+
+        // Draw the grid behind all other layers
+        SetLayerDepth( depthRange.y * 0.75 );
+
+        if( gridStyle == GRID_STYLE_LINES )
+        {
+            SetIsFill( false );
+            SetIsStroke( true );
+            SetStrokeColor( gridColor );
+
+            // Now draw the grid, every coarse grid line gets the double width
+
+            // Vertical lines
+            for( int j = gridStartY; j < gridEndY; j += 1 )
+            {
+                if( j % gridTick == 0 && gridScreenSizeDense > gridDrawThreshold )
+                    SetLineWidth( doubleMarker );
+                else
+                    SetLineWidth( marker );
+
+                if( ( j % gridTick == 0 && gridScreenSizeCoarse > gridDrawThreshold )
+                    || gridScreenSizeDense > gridDrawThreshold )
+                {
+                    drawGridLine( VECTOR2D( gridStartX * gridSize.x, j * gridSize.y + gridOrigin.y ),
+                                  VECTOR2D( gridEndX * gridSize.x,   j * gridSize.y + gridOrigin.y ) );
+                }
+            }
+
+            // Horizontal lines
+            for( int i = gridStartX; i < gridEndX; i += 1 )
+            {
+                if( i % gridTick == 0 && gridScreenSizeDense > gridDrawThreshold )
+                    SetLineWidth( doubleMarker );
+                else
+                    SetLineWidth( marker );
+
+                if( ( i % gridTick == 0 && gridScreenSizeCoarse > gridDrawThreshold )
+                    || gridScreenSizeDense > gridDrawThreshold )
+                {
+                    drawGridLine( VECTOR2D( i * gridSize.x + gridOrigin.x, gridStartY * gridSize.y ),
+                                  VECTOR2D( i * gridSize.x + gridOrigin.x, gridEndY * gridSize.y ) );
+                }
+            }
+        }
+        else    // Dotted grid
+        {
+            bool tickX, tickY;
+            SetIsFill( true );
+            SetIsStroke( false );
+            SetFillColor( gridColor );
+
+            for( int j = gridStartY; j < gridEndY; j += 1 )
+            {
+                if( j % gridTick == 0 && gridScreenSizeDense > gridDrawThreshold )
+                    tickY = true;
+                else
+                    tickY = false;
+
+                for( int i = gridStartX; i < gridEndX; i += 1 )
+                {
+                    if( i % gridTick == 0 && gridScreenSizeDense > gridDrawThreshold )
+                        tickX = true;
+                    else
+                        tickX = false;
+
+                    if( tickX || tickY || gridScreenSizeDense > gridDrawThreshold )
+                    {
+                        double radius = ( tickX && tickY ) ? doubleMarker : marker;
+                        DrawRectangle( VECTOR2D( i * gridSize.x - radius + gridOrigin.x,
+                                                 j * gridSize.y - radius + gridOrigin.y ),
+                                       VECTOR2D( i * gridSize.x + radius + gridOrigin.x,
+                                                 j * gridSize.y + radius + gridOrigin.y ) );
+                    }
+                }
+            }
+        }
+    }
+}
+
+
+VECTOR2D GAL::GetGridPoint( const VECTOR2D& aPoint ) const
+{
+    return VECTOR2D( KiROUND( ( aPoint.x - gridOffset.x ) / gridSize.x ) * gridSize.x + gridOffset.x,
+                     KiROUND( ( aPoint.y - gridOffset.y ) / gridSize.y ) * gridSize.y + gridOffset.y );
+}
+
+const int GAL::MIN_DEPTH = -1024;
+const int GAL::MAX_DEPTH = 1023;
+const int GAL::GRID_DEPTH = MAX_DEPTH - 1;
diff --git a/common/gal/opengl/cached_container.cpp b/common/gal/opengl/cached_container.cpp
new file mode 100644
index 0000000..8cf3513
--- /dev/null
+++ b/common/gal/opengl/cached_container.cpp
@@ -0,0 +1,547 @@
+/*
+ * This program source code file is part of KiCad, a free EDA CAD application.
+ *
+ * Copyright (C) 2013 CERN
+ * @author Maciej Suminski <maciej.suminski@cern.ch>
+ *
+ * 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
+ */
+
+/**
+ * @file cached_container.cpp
+ * @brief Class to store instances of VERTEX with caching. It allows storing VERTEX objects and
+ * associates them with VERTEX_ITEMs. This leads to a possibility of caching vertices data in the
+ * GPU memory and a fast reuse of that data.
+ */
+
+#include <gal/opengl/cached_container.h>
+#include <gal/opengl/vertex_manager.h>
+#include <gal/opengl/vertex_item.h>
+#include <gal/opengl/shader.h>
+#include <confirm.h>
+#include <wx/log.h>
+#include <list>
+#ifdef __WXDEBUG__
+#include <profile.h>
+#endif /* __WXDEBUG__ */
+
+using namespace KIGFX;
+
+CACHED_CONTAINER::CACHED_CONTAINER( unsigned int aSize ) :
+    VERTEX_CONTAINER( aSize ), m_item( NULL )
+{
+    // In the beginning there is only free space
+    m_freeChunks.insert( CHUNK( aSize, 0 ) );
+
+    // Do not have uninitialized members:
+    m_chunkSize = 0;
+    m_chunkOffset = 0;
+    m_itemSize = 0;
+}
+
+
+void CACHED_CONTAINER::SetItem( VERTEX_ITEM* aItem )
+{
+    wxASSERT( aItem != NULL );
+
+    m_item      = aItem;
+    m_itemSize  = m_item->GetSize();
+    m_chunkSize = m_itemSize;
+
+    if( m_itemSize == 0 )
+        m_items.insert( m_item ); // The item was not stored before
+    else
+        m_chunkOffset = m_item->GetOffset();
+
+#if CACHED_CONTAINER_TEST > 1
+    wxLogDebug( wxT( "Adding/editing item 0x%08lx (size %d)" ), (long) m_item, m_itemSize );
+#endif
+}
+
+
+void CACHED_CONTAINER::FinishItem()
+{
+    wxASSERT( m_item != NULL );
+    wxASSERT( m_item->GetSize() == m_itemSize );
+
+    // Finishing the previously edited item
+    if( m_itemSize < m_chunkSize )
+    {
+        // There is some not used but reserved memory left, so we should return it to the pool
+        int itemOffset = m_item->GetOffset();
+
+        // Add the not used memory back to the pool
+        m_freeChunks.insert( CHUNK( m_chunkSize - m_itemSize, itemOffset + m_itemSize ) );
+        m_freeSpace += ( m_chunkSize - m_itemSize );
+        // mergeFreeChunks();   // veery slow and buggy
+    }
+
+#if CACHED_CONTAINER_TEST > 1
+    wxLogDebug( wxT( "Finishing item 0x%08lx (size %d)" ), (long) m_item, m_itemSize );
+    test();
+    m_item = NULL;    // electric fence
+#endif
+}
+
+
+VERTEX* CACHED_CONTAINER::Allocate( unsigned int aSize )
+{
+    wxASSERT( m_item != NULL );
+
+    if( m_failed )
+        return NULL;
+
+    if( m_itemSize + aSize > m_chunkSize )
+    {
+        // There is not enough space in the currently reserved chunk, so we have to resize it
+
+        // Reserve a bigger memory chunk for the current item and
+        // make it multiple of 3 to store triangles
+        m_chunkSize = ( 2 * m_itemSize ) + aSize + ( 3 - aSize % 3 );
+        // Save the current size before reallocating
+        m_chunkOffset = reallocate( m_chunkSize );
+
+        if( m_chunkOffset > m_currentSize )
+        {
+            m_failed = true;
+            return NULL;
+        }
+    }
+
+    VERTEX* reserved = &m_vertices[m_chunkOffset + m_itemSize];
+    m_itemSize += aSize;
+    // Now the item officially possesses the memory chunk
+    m_item->setSize( m_itemSize );
+
+    // The content has to be updated
+    m_dirty = true;
+
+#if CACHED_CONTAINER_TEST > 1
+    test();
+#endif
+#if CACHED_CONTAINER_TEST > 2
+    showFreeChunks();
+    showReservedChunks();
+#endif
+
+    return reserved;
+}
+
+
+void CACHED_CONTAINER::Delete( VERTEX_ITEM* aItem )
+{
+    wxASSERT( aItem != NULL );
+    wxASSERT( m_items.find( aItem ) != m_items.end() );
+
+    int size   = aItem->GetSize();
+    int offset = aItem->GetOffset();
+
+#if CACHED_CONTAINER_TEST > 1
+    wxLogDebug( wxT( "Removing 0x%08lx (size %d offset %d)" ), (long) aItem, size, offset );
+#endif
+
+    // Insert a free memory chunk entry in the place where item was stored
+    if( size > 0 )
+    {
+        m_freeChunks.insert( CHUNK( size, offset ) );
+        m_freeSpace += size;
+        // Indicate that the item is not stored in the container anymore
+        aItem->setSize( 0 );
+    }
+
+    m_items.erase( aItem );
+
+#if CACHED_CONTAINER_TEST > 1
+    test();
+#endif
+
+    // Dynamic memory freeing, there is no point in holding
+    // a large amount of memory when there is no use for it
+    if( m_freeSpace > ( 0.75 * m_currentSize ) && m_currentSize > m_initialSize )
+    {
+        resizeContainer( 0.5 * m_currentSize );
+    }
+}
+
+
+void CACHED_CONTAINER::Clear()
+{
+    // Change size to the default one
+    m_vertices = static_cast<VERTEX*>( realloc( m_vertices,
+                                                m_initialSize * sizeof( VERTEX ) ) );
+
+    // Reset state variables
+    m_freeSpace     = m_initialSize;
+    m_currentSize   = m_initialSize;
+    m_failed = false;
+
+    // Set the size of all the stored VERTEX_ITEMs to 0, so it is clear that they are not held
+    // in the container anymore
+    ITEMS::iterator it;
+
+    for( it = m_items.begin(); it != m_items.end(); ++it )
+    {
+        ( *it )->setSize( 0 );
+    }
+
+    m_items.clear();
+
+
+    // Now there is only free space left
+    m_freeChunks.clear();
+    m_freeChunks.insert( CHUNK( m_freeSpace, 0 ) );
+}
+
+
+unsigned int CACHED_CONTAINER::reallocate( unsigned int aSize )
+{
+    wxASSERT( aSize > 0 );
+
+#if CACHED_CONTAINER_TEST > 2
+    wxLogDebug( wxT( "Resize 0x%08lx from %d to %d" ), (long) m_item, m_itemSize, aSize );
+#endif
+
+    // Is there enough space to store vertices?
+    if( m_freeSpace < aSize )
+    {
+        bool result;
+
+        // Would it be enough to double the current space?
+        if( aSize < m_freeSpace + m_currentSize )
+        {
+            // Yes: exponential growing
+            result = resizeContainer( m_currentSize * 2 );
+        }
+        else
+        {
+            // No: grow to the nearest greater power of 2
+            result = resizeContainer( pow( 2, ceil( log2( m_currentSize * 2 + aSize ) ) ) );
+        }
+
+        if( !result )
+            return UINT_MAX;
+    }
+
+    // Look for the free space chunk of at least given size
+    FREE_CHUNK_MAP::iterator newChunk = m_freeChunks.lower_bound( aSize );
+
+    if( newChunk == m_freeChunks.end() )
+    {
+        // In the case when there is enough space to store the vertices,
+        // but the free space is not continous we should defragment the container
+        if( !defragment() )
+            return UINT_MAX;
+
+        // Update the current offset
+        m_chunkOffset = m_item->GetOffset();
+
+        // We can take the first free chunk, as there is only one after defragmentation
+        // and we can be sure that it provides enough space to store the object
+        newChunk = m_freeChunks.begin();
+    }
+
+    // Parameters of the allocated cuhnk
+    unsigned int chunkSize   = newChunk->first;
+    unsigned int chunkOffset = newChunk->second;
+
+    wxASSERT( chunkSize >= aSize );
+    wxASSERT( chunkOffset < m_currentSize );
+
+    // Check if the item was previously stored in the container
+    if( m_itemSize > 0 )
+    {
+#if CACHED_CONTAINER_TEST > 3
+        wxLogDebug( wxT( "Moving 0x%08x from 0x%08x to 0x%08x" ),
+                    (int) m_item, oldChunkOffset, chunkOffset );
+#endif
+        // The item was reallocated, so we have to copy all the old data to the new place
+        memcpy( &m_vertices[chunkOffset], &m_vertices[m_chunkOffset], m_itemSize * VertexSize );
+
+        // Free the space previously used by the chunk
+        wxASSERT( m_itemSize > 0 );
+        m_freeChunks.insert( CHUNK( m_itemSize, m_chunkOffset ) );
+        m_freeSpace += m_itemSize;
+    }
+
+    // Remove the allocated chunk from the free space pool
+    m_freeChunks.erase( newChunk );
+
+    // If there is some space left, return it to the pool - add an entry for it
+    if( chunkSize > aSize )
+    {
+        m_freeChunks.insert( CHUNK( chunkSize - aSize, chunkOffset + aSize ) );
+    }
+
+    m_freeSpace -= aSize;
+    // mergeFreeChunks();   // veery slow and buggy
+
+    m_item->setOffset( chunkOffset );
+
+    return chunkOffset;
+}
+
+
+bool CACHED_CONTAINER::defragment( VERTEX* aTarget )
+{
+#if CACHED_CONTAINER_TEST > 0
+    wxLogDebug( wxT( "Defragmenting" ) );
+
+    prof_counter totalTime;
+    prof_start( &totalTime );
+#endif
+
+    if( aTarget == NULL )
+    {
+        // No target was specified, so we have to reallocate our own space
+        int size = m_currentSize * sizeof( VERTEX );
+        aTarget = static_cast<VERTEX*>( malloc( size ) );
+
+        if( aTarget == NULL )
+        {
+            DisplayError( NULL, wxString::Format(
+                          wxT( "CACHED_CONTAINER::defragment: Run out of memory (malloc %d bytes)" ),
+                          size ) );
+            return false;
+        }
+    }
+
+    int newOffset = 0;
+    ITEMS::iterator it, it_end;
+
+    for( it = m_items.begin(), it_end = m_items.end(); it != it_end; ++it )
+    {
+        VERTEX_ITEM* item = *it;
+        int itemOffset    = item->GetOffset();
+        int itemSize      = item->GetSize();
+
+        // Move an item to the new container
+        memcpy( &aTarget[newOffset], &m_vertices[itemOffset], itemSize * VertexSize );
+
+        // Update new offset
+        item->setOffset( newOffset );
+
+        // Move to the next free space
+        newOffset += itemSize;
+    }
+
+    free( m_vertices );
+    m_vertices = aTarget;
+
+    // Now there is only one big chunk of free memory
+    m_freeChunks.clear();
+    wxASSERT( m_freeSpace > 0 );
+    m_freeChunks.insert( CHUNK( m_freeSpace, m_currentSize - m_freeSpace ) );
+
+#if CACHED_CONTAINER_TEST > 0
+    prof_end( &totalTime );
+
+    wxLogDebug( wxT( "Defragmented the container storing %d vertices / %.1f ms" ),
+                m_currentSize - m_freeSpace, totalTime.msecs() );
+#endif
+
+    return true;
+}
+
+
+void CACHED_CONTAINER::mergeFreeChunks()
+{
+    if( m_freeChunks.size() <= 1 ) // There are no chunks that can be merged
+        return;
+
+#if CACHED_CONTAINER_TEST > 0
+    prof_counter totalTime;
+    prof_start( &totalTime );
+#endif
+
+    // Reversed free chunks map - this one stores chunk size with its offset as the key
+    std::list<CHUNK> freeChunks;
+
+    FREE_CHUNK_MAP::const_iterator it, it_end;
+
+    for( it = m_freeChunks.begin(), it_end = m_freeChunks.end(); it != it_end; ++it )
+    {
+        freeChunks.push_back( std::make_pair( it->second, it->first ) );
+    }
+
+    m_freeChunks.clear();
+    freeChunks.sort();
+
+    std::list<CHUNK>::const_iterator itf, itf_end;
+    unsigned int offset = freeChunks.front().first;
+    unsigned int size   = freeChunks.front().second;
+    freeChunks.pop_front();
+
+    for( itf = freeChunks.begin(), itf_end = freeChunks.end(); itf != itf_end; ++itf )
+    {
+        if( itf->first == offset + size )
+        {
+            // These chunks can be merged, so just increase the current chunk size and go on
+            size += itf->second;
+        }
+        else
+        {
+            // These chunks cannot be merged
+            // So store the previous one
+            m_freeChunks.insert( std::make_pair( size, offset ) );
+            // and let's check the next chunk
+            offset = itf->first;
+            size   = itf->second;
+
+        }
+    }
+
+    // Add the last one
+    m_freeChunks.insert( std::make_pair( size, offset ) );
+
+#if CACHED_CONTAINER_TEST > 0
+    prof_end( &totalTime );
+
+    wxLogDebug( wxT( "Merged free chunks / %.1f ms" ), totalTime.msecs() );
+#endif
+
+    test();
+}
+
+
+bool CACHED_CONTAINER::resizeContainer( unsigned int aNewSize )
+{
+    wxASSERT( aNewSize != m_currentSize );
+
+#if CACHED_CONTAINER_TEST > 0
+    wxLogDebug( wxT( "Resizing container from %d to %d" ), m_currentSize, aNewSize );
+#endif
+
+    VERTEX* newContainer;
+
+    if( aNewSize < m_currentSize )
+    {
+        // Shrinking container
+        // Sanity check, no shrinking if we cannot fit all the data
+        if( reservedSpace() > aNewSize )
+            return false;
+
+        int size = aNewSize * sizeof( VERTEX );
+        newContainer = static_cast<VERTEX*>( malloc( size ) );
+
+        if( newContainer == NULL )
+        {
+            DisplayError( NULL, wxString::Format(
+                          wxT( "CACHED_CONTAINER::resizeContainer:\n"
+                               "Run out of memory (malloc %d bytes)" ),
+                          size ) );
+            return false;
+        }
+
+        // Defragment directly to the new, smaller container
+        defragment( newContainer );
+
+        // We have to correct freeChunks after defragmentation
+        m_freeChunks.clear();
+        wxASSERT( aNewSize - reservedSpace() > 0 );
+        m_freeChunks.insert( CHUNK( aNewSize - reservedSpace(), reservedSpace() ) );
+    }
+    else
+    {
+        // Enlarging container
+        int size = aNewSize * sizeof( VERTEX );
+        newContainer = static_cast<VERTEX*>( realloc( m_vertices, size ) );
+
+        if( newContainer == NULL )
+        {
+            DisplayError( NULL, wxString::Format(
+                          wxT( "CACHED_CONTAINER::resizeContainer:\n"
+                               "Run out of memory (realloc from %d to %d bytes)" ),
+                          m_currentSize * sizeof( VERTEX ), size ) );
+            return false;
+        }
+
+        // Add an entry for the new memory chunk at the end of the container
+        m_freeChunks.insert( CHUNK( aNewSize - m_currentSize, m_currentSize ) );
+    }
+
+    m_vertices = newContainer;
+
+    m_freeSpace   += ( aNewSize - m_currentSize );
+    m_currentSize = aNewSize;
+
+    return true;
+}
+
+
+#ifdef CACHED_CONTAINER_TEST
+void CACHED_CONTAINER::showFreeChunks()
+{
+    FREE_CHUNK_MAP::iterator it;
+
+    wxLogDebug( wxT( "Free chunks:" ) );
+
+    for( it = m_freeChunks.begin(); it != m_freeChunks.end(); ++it )
+    {
+        unsigned int offset = getChunkOffset( *it );
+        unsigned int size   = getChunkSize( *it );
+        wxASSERT( size > 0 );
+
+        wxLogDebug( wxT( "[0x%08x-0x%08x] (size %d)" ),
+                    offset, offset + size - 1, size );
+    }
+}
+
+
+void CACHED_CONTAINER::showReservedChunks()
+{
+    ITEMS::iterator it;
+
+    wxLogDebug( wxT( "Reserved chunks:" ) );
+
+    for( it = m_items.begin(); it != m_items.end(); ++it )
+    {
+        VERTEX_ITEM* item   = *it;
+        unsigned int offset = item->GetOffset();
+        unsigned int size   = item->GetSize();
+        wxASSERT( size > 0 );
+
+        wxLogDebug( wxT( "[0x%08x-0x%08x] @ 0x%08lx (size %d)" ),
+                    offset, offset + size - 1, (long) item, size );
+    }
+}
+
+
+void CACHED_CONTAINER::test()
+{
+    // Free space check
+    unsigned int freeSpace = 0;
+    FREE_CHUNK_MAP::iterator itf;
+
+    for( itf = m_freeChunks.begin(); itf != m_freeChunks.end(); ++itf )
+        freeSpace += getChunkSize( *itf );
+
+    wxASSERT( freeSpace == m_freeSpace );
+
+    // Reserved space check
+    /*unsigned int reservedSpace = 0;
+    ITEMS::iterator itr;
+    for( itr = m_items.begin(); itr != m_items.end(); ++itr )
+        reservedSpace += ( *itr )->GetSize();
+    reservedSpace += m_itemSize;    // Add the current chunk size
+
+    wxASSERT( ( freeSpace + reservedSpace ) == m_currentSize );*/
+
+    // Overlapping check TBD
+}
+
+#endif /* CACHED_CONTAINER_TEST */
diff --git a/common/gal/opengl/gpu_manager.cpp b/common/gal/opengl/gpu_manager.cpp
new file mode 100644
index 0000000..be34a56
--- /dev/null
+++ b/common/gal/opengl/gpu_manager.cpp
@@ -0,0 +1,302 @@
+/*
+ * This program source code file is part of KiCad, a free EDA CAD application.
+ *
+ * Copyright (C) 2013 CERN
+ * @author Maciej Suminski <maciej.suminski@cern.ch>
+ *
+ * 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
+ */
+
+/**
+ * @file gpu_manager.cpp
+ * @brief Class to handle uploading vertices and indices to GPU in drawing purposes.
+ */
+
+#include <gal/opengl/gpu_manager.h>
+#include <gal/opengl/cached_container.h>
+#include <gal/opengl/noncached_container.h>
+#include <gal/opengl/shader.h>
+#include <typeinfo>
+#include <wx/msgdlg.h>
+#include <confirm.h>
+#ifdef PROFILE
+#include <profile.h>
+#include <wx/debug.h>
+#include <wx/log.h>
+#endif /* PROFILE */
+
+using namespace KIGFX;
+
+GPU_MANAGER* GPU_MANAGER::MakeManager( VERTEX_CONTAINER* aContainer )
+{
+    if( typeid( *aContainer ) == typeid( CACHED_CONTAINER ) )
+        return new GPU_CACHED_MANAGER( aContainer );
+    else if( typeid( *aContainer ) == typeid( NONCACHED_CONTAINER ) )
+        return new GPU_NONCACHED_MANAGER( aContainer );
+
+    wxASSERT_MSG( false, wxT( "Not handled container type" ) );
+    return NULL;
+}
+
+
+GPU_MANAGER::GPU_MANAGER( VERTEX_CONTAINER* aContainer ) :
+    m_isDrawing( false ), m_container( aContainer ), m_shader( NULL ), m_shaderAttrib( 0 )
+{
+}
+
+
+GPU_MANAGER::~GPU_MANAGER()
+{
+}
+
+
+void GPU_MANAGER::SetShader( SHADER& aShader )
+{
+    m_shader = &aShader;
+
+    m_shaderAttrib = m_shader->GetAttribute( "attrShaderParams" );
+
+    if( m_shaderAttrib == -1 )
+    {
+        DisplayError( NULL, wxT( "Could not get the shader attribute location" ) );
+    }
+}
+
+
+// Cached manager
+GPU_CACHED_MANAGER::GPU_CACHED_MANAGER( VERTEX_CONTAINER* aContainer ) :
+    GPU_MANAGER( aContainer ), m_buffersInitialized( false ), m_indicesPtr( NULL ),
+    m_verticesBuffer( 0 ), m_indicesBuffer( 0 ), m_indicesSize( 0 ), m_indicesCapacity( 0 )
+{
+    // Allocate the biggest possible buffer for indices
+    resizeIndices( aContainer->GetSize() );
+}
+
+
+GPU_CACHED_MANAGER::~GPU_CACHED_MANAGER()
+{
+    if( m_buffersInitialized )
+    {
+        glBindBuffer( GL_ARRAY_BUFFER, 0 );
+        glDeleteBuffers( 1, &m_verticesBuffer );
+        glDeleteBuffers( 1, &m_indicesBuffer );
+    }
+}
+
+
+void GPU_CACHED_MANAGER::Initialize()
+{
+    wxASSERT( !m_buffersInitialized );
+
+    if( !m_buffersInitialized )
+    {
+        glGenBuffers( 1, &m_verticesBuffer );
+        glGenBuffers( 1, &m_indicesBuffer );
+        m_buffersInitialized = true;
+    }
+}
+
+
+void GPU_CACHED_MANAGER::BeginDrawing()
+{
+    wxASSERT( !m_isDrawing );
+
+    if( m_container->IsDirty() )
+        uploadToGpu();
+
+    // Number of vertices to be drawn in the EndDrawing()
+    m_indicesSize = 0;
+    // Set the indices pointer to the beginning of the indices-to-draw buffer
+    m_indicesPtr = m_indices.get();
+
+    m_isDrawing = true;
+}
+
+
+void GPU_CACHED_MANAGER::DrawIndices( unsigned int aOffset, unsigned int aSize )
+{
+    wxASSERT( m_isDrawing );
+
+    // Copy indices of items that should be drawn to GPU memory
+    for( unsigned int i = aOffset; i < aOffset + aSize; *m_indicesPtr++ = i++ );
+
+    m_indicesSize += aSize;
+}
+
+
+void GPU_CACHED_MANAGER::DrawAll()
+{
+    wxASSERT( m_isDrawing );
+
+    m_indicesSize = m_container->GetSize();
+    for( unsigned int i = 0; i < m_indicesSize; *m_indicesPtr++ = i++ );
+}
+
+
+void GPU_CACHED_MANAGER::EndDrawing()
+{
+    wxASSERT( m_isDrawing );
+
+    // Prepare buffers
+    glEnableClientState( GL_VERTEX_ARRAY );
+    glEnableClientState( GL_COLOR_ARRAY );
+
+    // Bind vertices data buffers
+    glBindBuffer( GL_ARRAY_BUFFER, m_verticesBuffer );
+    glVertexPointer( CoordStride, GL_FLOAT, VertexSize, 0 );
+    glColorPointer( ColorStride, GL_UNSIGNED_BYTE, VertexSize, (GLvoid*) ColorOffset );
+
+    if( m_shader != NULL )    // Use shader if applicable
+    {
+        m_shader->Use();
+        glEnableVertexAttribArray( m_shaderAttrib );
+        glVertexAttribPointer( m_shaderAttrib, ShaderStride, GL_FLOAT, GL_FALSE,
+                               VertexSize, (GLvoid*) ShaderOffset );
+    }
+
+    glBindBuffer( GL_ELEMENT_ARRAY_BUFFER, m_indicesBuffer );
+    glBufferData( GL_ELEMENT_ARRAY_BUFFER, m_indicesSize * sizeof(int), (GLvoid*) m_indices.get(), GL_DYNAMIC_DRAW );
+
+    glDrawElements( GL_TRIANGLES, m_indicesSize, GL_UNSIGNED_INT, 0 );
+
+    glBindBuffer( GL_ARRAY_BUFFER, 0 );
+    glBindBuffer( GL_ELEMENT_ARRAY_BUFFER, 0 );
+
+    // Deactivate vertex array
+    glDisableClientState( GL_COLOR_ARRAY );
+    glDisableClientState( GL_VERTEX_ARRAY );
+
+    if( m_shader != NULL )
+    {
+        glDisableVertexAttribArray( m_shaderAttrib );
+        m_shader->Deactivate();
+    }
+
+    m_isDrawing = false;
+}
+
+
+void GPU_CACHED_MANAGER::uploadToGpu()
+{
+#ifdef PROFILE
+    prof_counter totalTime;
+    prof_start( &totalTime );
+#endif /* PROFILE  */
+
+    if( !m_buffersInitialized )
+        Initialize();
+
+    int bufferSize    = m_container->GetSize();
+    GLfloat* vertices = (GLfloat*) m_container->GetAllVertices();
+
+    // Upload vertices coordinates and shader types to GPU memory
+    glBindBuffer( GL_ARRAY_BUFFER, m_verticesBuffer );
+    glBufferData( GL_ARRAY_BUFFER, bufferSize * VertexSize, vertices, GL_STATIC_DRAW );
+    glBindBuffer( GL_ARRAY_BUFFER, 0 );
+
+    // Allocate the biggest possible buffer for indices
+    resizeIndices( bufferSize );
+
+    if( glGetError() != GL_NO_ERROR )
+        DisplayError( NULL, wxT( "Error during data upload to the GPU memory" ) );
+
+#ifdef PROFILE
+    prof_end( &totalTime );
+
+    wxLogDebug( wxT( "Uploading %d vertices to GPU / %.1f ms" ), bufferSize, totalTime.msecs() );
+#endif /* PROFILE */
+}
+
+
+void GPU_CACHED_MANAGER::resizeIndices( unsigned int aNewSize )
+{
+    if( aNewSize > m_indicesCapacity )
+    {
+        m_indicesCapacity = aNewSize;
+        m_indices.reset( new GLuint[m_indicesCapacity] );
+    }
+}
+
+
+// Noncached manager
+GPU_NONCACHED_MANAGER::GPU_NONCACHED_MANAGER( VERTEX_CONTAINER* aContainer ) :
+    GPU_MANAGER( aContainer )
+{
+}
+
+
+void GPU_NONCACHED_MANAGER::Initialize()
+{
+    // Nothing has to be intialized
+}
+
+
+void GPU_NONCACHED_MANAGER::BeginDrawing()
+{
+    // Nothing has to be prepared
+}
+
+
+void GPU_NONCACHED_MANAGER::DrawIndices( unsigned int aOffset, unsigned int aSize )
+{
+    wxASSERT_MSG( false, wxT( "Not implemented yet" ) );
+}
+
+
+void GPU_NONCACHED_MANAGER::DrawAll()
+{
+    // This is the default use case, nothing has to be done
+    // The real rendering takes place in the EndDrawing() function
+}
+
+
+void GPU_NONCACHED_MANAGER::EndDrawing()
+{
+    VERTEX*  vertices       = m_container->GetAllVertices();
+    GLfloat* coordinates    = (GLfloat*) ( vertices );
+    GLubyte* colors         = (GLubyte*) ( vertices ) + ColorOffset;
+
+    // Prepare buffers
+    glEnableClientState( GL_VERTEX_ARRAY );
+    glEnableClientState( GL_COLOR_ARRAY );
+
+    glVertexPointer( CoordStride, GL_FLOAT, VertexSize, coordinates );
+    glColorPointer( ColorStride, GL_UNSIGNED_BYTE, VertexSize, colors );
+
+    if( m_shader != NULL )    // Use shader if applicable
+    {
+        GLfloat* shaders = (GLfloat*) ( vertices ) + ShaderOffset / sizeof(GLfloat);
+
+        m_shader->Use();
+        glEnableVertexAttribArray( m_shaderAttrib );
+        glVertexAttribPointer( m_shaderAttrib, ShaderStride, GL_FLOAT, GL_FALSE,
+                               VertexSize, shaders );
+    }
+
+    glDrawArrays( GL_TRIANGLES, 0, m_container->GetSize() );
+
+    // Deactivate vertex array
+    glDisableClientState( GL_COLOR_ARRAY );
+    glDisableClientState( GL_VERTEX_ARRAY );
+
+    if( m_shader != NULL )
+    {
+        glDisableVertexAttribArray( m_shaderAttrib );
+        m_shader->Deactivate();
+    }
+}
diff --git a/common/gal/opengl/noncached_container.cpp b/common/gal/opengl/noncached_container.cpp
new file mode 100644
index 0000000..c902c47
--- /dev/null
+++ b/common/gal/opengl/noncached_container.cpp
@@ -0,0 +1,89 @@
+/*
+ * This program source code file is part of KiCad, a free EDA CAD application.
+ *
+ * Copyright (C) 2013 CERN
+ * @author Maciej Suminski <maciej.suminski@cern.ch>
+ *
+ * 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
+ */
+
+/**
+ * @file noncached_container.cpp
+ * @brief Class to store instances of VERTEX without caching. It allows a fast one-frame drawing
+ * and then clearing the buffer and starting from scratch.
+ */
+
+#include <gal/opengl/noncached_container.h>
+#include <cstdlib>
+
+using namespace KIGFX;
+
+NONCACHED_CONTAINER::NONCACHED_CONTAINER( unsigned int aSize ) :
+    VERTEX_CONTAINER( aSize ), m_freePtr( 0 )
+{
+}
+
+
+NONCACHED_CONTAINER::~NONCACHED_CONTAINER()
+{
+}
+
+
+void NONCACHED_CONTAINER::SetItem( VERTEX_ITEM* aItem )
+{
+    // Nothing has to be done, as the noncached container
+    // does not care about VERTEX_ITEMs ownership
+}
+
+
+VERTEX* NONCACHED_CONTAINER::Allocate( unsigned int aSize )
+{
+    if( m_freeSpace < aSize )
+    {
+        // Double the space
+        VERTEX* newVertices = static_cast<VERTEX*>( realloc( m_vertices,
+                                                             m_currentSize * 2 *
+                                                             sizeof(VERTEX) ) );
+
+        if( newVertices != NULL )
+        {
+            m_vertices    = newVertices;
+            m_freeSpace   += m_currentSize;
+            m_currentSize *= 2;
+        }
+        else
+        {
+            return NULL;
+        }
+    }
+
+    VERTEX* freeVertex = &m_vertices[m_freePtr];
+
+    // Move to the next free chunk
+    m_freePtr   += aSize;
+    m_freeSpace -= aSize;
+
+    return freeVertex;
+}
+
+
+void NONCACHED_CONTAINER::Clear()
+{
+    m_freePtr   = 0;
+    m_freeSpace = m_currentSize;
+}
diff --git a/common/gal/opengl/opengl_compositor.cpp b/common/gal/opengl/opengl_compositor.cpp
new file mode 100644
index 0000000..9f35d9e
--- /dev/null
+++ b/common/gal/opengl/opengl_compositor.cpp
@@ -0,0 +1,291 @@
+/*
+ * This program source code file is part of KiCad, a free EDA CAD application.
+ *
+ * Copyright (C) 2013-2015 CERN
+ * @author Maciej Suminski <maciej.suminski@cern.ch>
+ *
+ * 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
+ */
+
+/**
+ * @file opengl_compositor.cpp
+ * @brief Class that handles multitarget rendering (i.e. to different textures/surfaces) and
+ * later compositing into a single image (OpenGL flavour).
+ */
+
+#include <gal/opengl/opengl_compositor.h>
+
+#include <stdexcept>
+#include <cassert>
+
+using namespace KIGFX;
+
+OPENGL_COMPOSITOR::OPENGL_COMPOSITOR() :
+    m_initialized( false ), m_current( 0 ), m_currentFbo( DIRECT_RENDERING )
+{
+    // Avoid not initialized members:
+    m_framebuffer = 0;
+    m_depthBuffer = 0;
+}
+
+
+OPENGL_COMPOSITOR::~OPENGL_COMPOSITOR()
+{
+    if( m_initialized )
+        clean();
+}
+
+
+void OPENGL_COMPOSITOR::Initialize()
+{
+    if( m_initialized )
+        return;
+
+    // We need framebuffer objects for drawing the screen contents
+    // Generate framebuffer and a depth buffer
+    glGenFramebuffersEXT( 1, &m_framebuffer );
+    glBindFramebufferEXT( GL_FRAMEBUFFER, m_framebuffer );
+    m_currentFbo = m_framebuffer;
+
+    // Allocate memory for the depth buffer
+    // Attach the depth buffer to the framebuffer
+    glGenRenderbuffersEXT( 1, &m_depthBuffer );
+    glBindRenderbufferEXT( GL_RENDERBUFFER_EXT, m_depthBuffer );
+
+    // Use here a size of 24 bits for the depth buffer, 8 bits for the stencil buffer
+    // this is required later for anti-aliasing
+    glRenderbufferStorageEXT( GL_RENDERBUFFER_EXT, GL_DEPTH_COMPONENT16, m_width, m_height );
+    glFramebufferRenderbufferEXT( GL_FRAMEBUFFER_EXT, GL_DEPTH_ATTACHMENT,
+                                  GL_RENDERBUFFER_EXT, m_depthBuffer );
+
+    // Unbind the framebuffer, so by default all the rendering goes directly to the display
+    glBindFramebufferEXT( GL_FRAMEBUFFER_EXT, DIRECT_RENDERING );
+    m_currentFbo = DIRECT_RENDERING;
+
+    m_initialized = true;
+}
+
+
+void OPENGL_COMPOSITOR::Resize( unsigned int aWidth, unsigned int aHeight )
+{
+    if( m_initialized )
+        clean();
+
+    m_width  = aWidth;
+    m_height = aHeight;
+}
+
+
+unsigned int OPENGL_COMPOSITOR::CreateBuffer()
+{
+    assert( m_initialized );
+
+    unsigned int maxBuffers;
+
+    // Get the maximum number of buffers
+    glGetIntegerv( GL_MAX_COLOR_ATTACHMENTS, (GLint*) &maxBuffers );
+
+    if( usedBuffers() >= maxBuffers )
+    {
+        throw std::runtime_error("Cannot create more framebuffers. OpenGL rendering "
+                        "backend requires at least 3 framebuffers. You may try to update/change "
+                        "your graphic drivers.");
+    }
+
+    // GL_COLOR_ATTACHMENTn are consecutive integers
+    GLuint attachmentPoint = GL_COLOR_ATTACHMENT0 + usedBuffers();
+    GLuint textureTarget;
+
+    // Generate the texture for the pixel storage
+    glGenTextures( 1, &textureTarget );
+    glBindTexture( GL_TEXTURE_2D, textureTarget );
+
+    // Set texture parameters
+    glTexEnvf( GL_TEXTURE_ENV, GL_TEXTURE_ENV_MODE, GL_MODULATE );
+    glTexImage2D( GL_TEXTURE_2D, 0, GL_RGBA8, m_width, m_height, 0, GL_RGBA,
+                  GL_UNSIGNED_BYTE, NULL );
+    glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST );
+    glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST );
+
+    // Bind the texture to the specific attachment point, clear and rebind the screen
+    glBindFramebufferEXT( GL_FRAMEBUFFER_EXT, m_framebuffer );
+    m_currentFbo = m_framebuffer;
+    glFramebufferTexture2DEXT( GL_FRAMEBUFFER_EXT, attachmentPoint,
+                               GL_TEXTURE_2D, textureTarget, 0 );
+
+    // Check the status, exit if the framebuffer can't be created
+    GLenum status = glCheckFramebufferStatusEXT( GL_FRAMEBUFFER_EXT );
+
+    if( status != GL_FRAMEBUFFER_COMPLETE_EXT )
+    {
+        switch( status )
+        {
+        case GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT_EXT:
+            throw std::runtime_error( "Cannot create the framebuffer." );
+            break;
+
+        case GL_FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT_EXT:
+            throw std::runtime_error(  "The framebuffer attachment points are incomplete." );
+            break;
+
+        case GL_FRAMEBUFFER_INCOMPLETE_DRAW_BUFFER_EXT:
+            throw std::runtime_error(  "The framebuffer does not have at least one "
+                    "image attached to it." );
+            break;
+
+        case GL_FRAMEBUFFER_INCOMPLETE_READ_BUFFER_EXT:
+            throw std::runtime_error(  "The framebuffer read buffer is incomplete." );
+            break;
+
+        case GL_FRAMEBUFFER_UNSUPPORTED_EXT:
+            throw std::runtime_error(  "The combination of internal formats of the attached "
+                    "images violates an implementation-dependent set of restrictions." );
+            break;
+
+        case GL_FRAMEBUFFER_INCOMPLETE_MULTISAMPLE_EXT:
+            throw std::runtime_error(  "GL_RENDERBUFFER_SAMPLES is not the same for "
+                    "all attached renderbuffers" );
+            break;
+
+        case GL_FRAMEBUFFER_INCOMPLETE_LAYER_TARGETS_EXT:
+            throw std::runtime_error(  "Framebuffer incomplete layer targets errors." );
+            break;
+
+        default:
+            throw std::runtime_error(  "Cannot create the framebuffer." );
+            break;
+        }
+
+        return 0;
+    }
+
+    ClearBuffer();
+
+    // Return to direct rendering (we were asked only to create a buffer, not switch to one)
+    glBindFramebufferEXT( GL_FRAMEBUFFER_EXT, DIRECT_RENDERING );
+    m_currentFbo = DIRECT_RENDERING;
+
+    // Store the new buffer
+    OPENGL_BUFFER buffer = { textureTarget, attachmentPoint };
+    m_buffers.push_back( buffer );
+
+    return usedBuffers();
+}
+
+
+void OPENGL_COMPOSITOR::SetBuffer( unsigned int aBufferHandle )
+{
+    if( aBufferHandle > usedBuffers() )
+        return;
+
+    // Change the rendering destination to the selected attachment point
+    if( aBufferHandle == DIRECT_RENDERING )
+    {
+        m_currentFbo = DIRECT_RENDERING;
+    }
+    else if( m_currentFbo != m_framebuffer )
+    {
+        glBindFramebufferEXT( GL_FRAMEBUFFER_EXT, m_framebuffer );
+        m_currentFbo = m_framebuffer;
+    }
+
+    if( m_currentFbo != DIRECT_RENDERING )
+    {
+        m_current = aBufferHandle - 1;
+        glDrawBuffer( m_buffers[m_current].attachmentPoint );
+    }
+}
+
+
+void OPENGL_COMPOSITOR::ClearBuffer()
+{
+    assert( m_initialized );
+
+    glClearColor( 0.0f, 0.0f, 0.0f, 0.0f );
+    glClear( GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT );
+}
+
+
+void OPENGL_COMPOSITOR::DrawBuffer( unsigned int aBufferHandle )
+{
+    assert( m_initialized );
+    assert( aBufferHandle != 0 && aBufferHandle <= usedBuffers() );
+
+    // Switch to the main framebuffer and blit the scene
+    glBindFramebufferEXT( GL_FRAMEBUFFER, DIRECT_RENDERING );
+    m_currentFbo = DIRECT_RENDERING;
+
+    // Depth test has to be disabled to make transparency working
+    glDisable( GL_DEPTH_TEST );
+    glBlendFunc( GL_ONE, GL_ONE_MINUS_SRC_ALPHA );
+
+    // Enable texturing and bind the main texture
+    glEnable( GL_TEXTURE_2D );
+    glBindTexture( GL_TEXTURE_2D, m_buffers[aBufferHandle - 1].textureTarget );
+
+    // Draw a full screen quad with the texture
+    glMatrixMode( GL_MODELVIEW );
+    glPushMatrix();
+    glLoadIdentity();
+    glMatrixMode( GL_PROJECTION );
+    glPushMatrix();
+    glLoadIdentity();
+
+    glBegin( GL_TRIANGLES );
+    glTexCoord2f( 0.0f,  1.0f );
+    glVertex2f(  -1.0f, -1.0f );
+    glTexCoord2f( 1.0f,  1.0f );
+    glVertex2f(   1.0f, -1.0f );
+    glTexCoord2f( 1.0f,  0.0f );
+    glVertex2f(   1.0f,  1.0f );
+
+    glTexCoord2f( 0.0f,  1.0f );
+    glVertex2f(  -1.0f, -1.0f );
+    glTexCoord2f( 1.0f,  0.0f );
+    glVertex2f(   1.0f,  1.0f );
+    glTexCoord2f( 0.0f,  0.0f );
+    glVertex2f(  -1.0f,  1.0f );
+    glEnd();
+
+    glPopMatrix();
+    glMatrixMode( GL_MODELVIEW );
+    glPopMatrix();
+}
+
+
+void OPENGL_COMPOSITOR::clean()
+{
+    assert( m_initialized );
+
+    glBindFramebufferEXT( GL_FRAMEBUFFER, DIRECT_RENDERING );
+    m_currentFbo = DIRECT_RENDERING;
+
+    OPENGL_BUFFERS::const_iterator it;
+
+    for( it = m_buffers.begin(); it != m_buffers.end(); ++it )
+    {
+        glDeleteTextures( 1, &it->textureTarget );
+    }
+
+    m_buffers.clear();
+
+    glDeleteFramebuffersEXT( 1, &m_framebuffer );
+    glDeleteRenderbuffersEXT( 1, &m_depthBuffer );
+
+    m_initialized = false;
+}
diff --git a/common/gal/opengl/opengl_gal.cpp b/common/gal/opengl/opengl_gal.cpp
new file mode 100644
index 0000000..945161f
--- /dev/null
+++ b/common/gal/opengl/opengl_gal.cpp
@@ -0,0 +1,1204 @@
+/*
+ * This program source code file is part of KICAD, a free EDA CAD application.
+ *
+ * Copyright (C) 2012 Torsten Hueter, torstenhtr <at> gmx.de
+ * Copyright (C) 2012 Kicad Developers, see change_log.txt for contributors.
+ * Copyright (C) 2013-2016 CERN
+ * @author Maciej Suminski <maciej.suminski@cern.ch>
+ *
+ * Graphics Abstraction Layer (GAL) for OpenGL
+ *
+ * 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 <gal/opengl/opengl_gal.h>
+#include <gal/definitions.h>
+
+#include <wx/log.h>
+#include <macros.h>
+#ifdef __WXDEBUG__
+#include <profile.h>
+#endif /* __WXDEBUG__ */
+
+#include <limits>
+#include <boost/bind.hpp>
+
+using namespace KIGFX;
+
+static void InitTesselatorCallbacks( GLUtesselator* aTesselator );
+const int glAttributes[] = { WX_GL_RGBA, WX_GL_DOUBLEBUFFER, WX_GL_DEPTH_SIZE, 8, 0 };
+wxGLContext* OPENGL_GAL::glContext = NULL;
+
+OPENGL_GAL::OPENGL_GAL( wxWindow* aParent, wxEvtHandler* aMouseListener,
+                        wxEvtHandler* aPaintListener, const wxString& aName ) :
+    wxGLCanvas( aParent, wxID_ANY, (int*) glAttributes, wxDefaultPosition, wxDefaultSize,
+                wxEXPAND, aName ),
+    mouseListener( aMouseListener ),
+    paintListener( aPaintListener ),
+    cachedManager( true ),
+    nonCachedManager( false ),
+    overlayManager( false )
+{
+    if( glContext == NULL )
+        glContext = new wxGLContext( this );
+
+    // Check if OpenGL requirements are met
+    runTest();
+
+    // Make VBOs use shaders
+    cachedManager.SetShader( shader );
+    nonCachedManager.SetShader( shader );
+    overlayManager.SetShader( shader );
+
+    // Initialize the flags
+    isFramebufferInitialized = false;
+    isGrouping               = false;
+    groupCounter             = 0;
+
+#ifdef RETINA_OPENGL_PATCH
+    SetViewWantsBestResolution( true );
+#endif
+
+    // Connecting the event handlers
+    Connect( wxEVT_PAINT,           wxPaintEventHandler( OPENGL_GAL::onPaint ) );
+
+    // Mouse events are skipped to the parent
+    Connect( wxEVT_MOTION,          wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+    Connect( wxEVT_LEFT_DOWN,       wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+    Connect( wxEVT_LEFT_UP,         wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+    Connect( wxEVT_LEFT_DCLICK,     wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+    Connect( wxEVT_MIDDLE_DOWN,     wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+    Connect( wxEVT_MIDDLE_UP,       wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+    Connect( wxEVT_MIDDLE_DCLICK,   wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+    Connect( wxEVT_RIGHT_DOWN,      wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+    Connect( wxEVT_RIGHT_UP,        wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+    Connect( wxEVT_RIGHT_DCLICK,    wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+    Connect( wxEVT_MOUSEWHEEL,      wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+#ifdef USE_OSX_MAGNIFY_EVENT
+    Connect( wxEVT_MAGNIFY,         wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+#endif
+#if defined _WIN32 || defined _WIN64
+    Connect( wxEVT_ENTER_WINDOW,    wxMouseEventHandler( OPENGL_GAL::skipMouseEvent ) );
+#endif
+
+    SetSize( aParent->GetSize() );
+    screenSize = VECTOR2I( aParent->GetSize() );
+
+    // Grid color settings are different in Cairo and OpenGL
+    SetGridColor( COLOR4D( 0.8, 0.8, 0.8, 0.1 ) );
+
+    // Tesselator initialization
+    tesselator = gluNewTess();
+    InitTesselatorCallbacks( tesselator );
+
+    if( tesselator == NULL )
+        throw std::runtime_error( "Could not create the tesselator" );
+
+    gluTessProperty( tesselator, GLU_TESS_WINDING_RULE, GLU_TESS_WINDING_POSITIVE );
+
+    currentManager = &nonCachedManager;
+}
+
+
+OPENGL_GAL::~OPENGL_GAL()
+{
+    glFlush();
+
+    gluDeleteTess( tesselator );
+    ClearCache();
+}
+
+
+void OPENGL_GAL::BeginDrawing()
+{
+    if( !IsShownOnScreen() )
+        return;
+
+    SetCurrent( *glContext );
+    clientDC = new wxClientDC( this );
+
+#ifdef RETINA_OPENGL_PATCH
+    const float scaleFactor = GetBackingScaleFactor();
+#else
+    const float scaleFactor = 1.0f;
+#endif
+
+    // Set up the view port
+    glMatrixMode( GL_PROJECTION );
+    glLoadIdentity();
+    glViewport( 0, 0, (GLsizei) screenSize.x * scaleFactor, (GLsizei) screenSize.y * scaleFactor );
+
+    // Create the screen transformation
+    glOrtho( 0, (GLint) screenSize.x, 0, (GLsizei) screenSize.y,
+             -depthRange.x, -depthRange.y );
+
+    if( !isFramebufferInitialized )
+    {
+        // Prepare rendering target buffers
+        compositor.Initialize();
+        mainBuffer = compositor.CreateBuffer();
+        overlayBuffer = compositor.CreateBuffer();
+
+        isFramebufferInitialized = true;
+    }
+
+    // Disable 2D Textures
+    glDisable( GL_TEXTURE_2D );
+
+    // Enable the depth buffer
+    glEnable( GL_DEPTH_TEST );
+    glDepthFunc( GL_LESS );
+
+    // Setup blending, required for transparent objects
+    glEnable( GL_BLEND );
+    glBlendFunc( GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA );
+
+    glMatrixMode( GL_MODELVIEW );
+
+    // Set up the world <-> screen transformation
+    ComputeWorldScreenMatrix();
+    GLdouble matrixData[16] = { 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1 };
+    matrixData[0]   = worldScreenMatrix.m_data[0][0];
+    matrixData[1]   = worldScreenMatrix.m_data[1][0];
+    matrixData[2]   = worldScreenMatrix.m_data[2][0];
+    matrixData[4]   = worldScreenMatrix.m_data[0][1];
+    matrixData[5]   = worldScreenMatrix.m_data[1][1];
+    matrixData[6]   = worldScreenMatrix.m_data[2][1];
+    matrixData[12]  = worldScreenMatrix.m_data[0][2];
+    matrixData[13]  = worldScreenMatrix.m_data[1][2];
+    matrixData[14]  = worldScreenMatrix.m_data[2][2];
+    glLoadMatrixd( matrixData );
+
+    // Set defaults
+    SetFillColor( fillColor );
+    SetStrokeColor( strokeColor );
+
+    // Unbind buffers - set compositor for direct drawing
+    compositor.SetBuffer( OPENGL_COMPOSITOR::DIRECT_RENDERING );
+
+    // Remove all previously stored items
+    nonCachedManager.Clear();
+    overlayManager.Clear();
+
+    cachedManager.BeginDrawing();
+    nonCachedManager.BeginDrawing();
+    overlayManager.BeginDrawing();
+}
+
+
+void OPENGL_GAL::EndDrawing()
+{
+    // Cached & non-cached containers are rendered to the same buffer
+    compositor.SetBuffer( mainBuffer );
+    nonCachedManager.EndDrawing();
+    cachedManager.EndDrawing();
+
+    // Overlay container is rendered to a different buffer
+    compositor.SetBuffer( overlayBuffer );
+    overlayManager.EndDrawing();
+
+    // Be sure that the framebuffer is not colorized (happens on specific GPU&drivers combinations)
+    glColor4d( 1.0, 1.0, 1.0, 1.0 );
+
+    // Draw the remaining contents, blit the rendering targets to the screen, swap the buffers
+    compositor.DrawBuffer( mainBuffer );
+    compositor.DrawBuffer( overlayBuffer );
+    blitCursor();
+
+    glFlush();
+    SwapBuffers();
+
+    delete clientDC;
+}
+
+
+void OPENGL_GAL::DrawLine( const VECTOR2D& aStartPoint, const VECTOR2D& aEndPoint )
+{
+    const VECTOR2D  startEndVector = aEndPoint - aStartPoint;
+    double          lineAngle = startEndVector.Angle();
+
+    currentManager->Color( strokeColor.r, strokeColor.g, strokeColor.b, strokeColor.a );
+
+    drawLineQuad( aStartPoint, aEndPoint );
+
+    // Line caps
+    if( lineWidth > 1.0 )
+    {
+        drawFilledSemiCircle( aStartPoint, lineWidth / 2, lineAngle + M_PI / 2 );
+        drawFilledSemiCircle( aEndPoint,   lineWidth / 2, lineAngle - M_PI / 2 );
+    }
+}
+
+
+void OPENGL_GAL::DrawSegment( const VECTOR2D& aStartPoint, const VECTOR2D& aEndPoint,
+                              double aWidth )
+{
+    VECTOR2D startEndVector = aEndPoint - aStartPoint;
+    double   lineAngle      = startEndVector.Angle();
+
+    if( isFillEnabled )
+    {
+        // Filled tracks
+        currentManager->Color( fillColor.r, fillColor.g, fillColor.b, fillColor.a );
+
+        SetLineWidth( aWidth );
+        drawLineQuad( aStartPoint, aEndPoint );
+
+        // Draw line caps
+        drawFilledSemiCircle( aStartPoint, aWidth / 2, lineAngle + M_PI / 2 );
+        drawFilledSemiCircle( aEndPoint,   aWidth / 2, lineAngle - M_PI / 2 );
+    }
+    else
+    {
+        // Outlined tracks
+        double lineLength = startEndVector.EuclideanNorm();
+
+        currentManager->Color( strokeColor.r, strokeColor.g, strokeColor.b, strokeColor.a );
+
+        Save();
+
+        currentManager->Translate( aStartPoint.x, aStartPoint.y, 0.0 );
+        currentManager->Rotate( lineAngle, 0.0f, 0.0f, 1.0f );
+
+        drawLineQuad( VECTOR2D( 0.0,         aWidth / 2.0 ),
+                      VECTOR2D( lineLength,  aWidth / 2.0 ) );
+
+        drawLineQuad( VECTOR2D( 0.0,        -aWidth / 2.0 ),
+                      VECTOR2D( lineLength, -aWidth / 2.0 ) );
+
+        // Draw line caps
+        drawStrokedSemiCircle( VECTOR2D( 0.0, 0.0 ), aWidth / 2, M_PI / 2 );
+        drawStrokedSemiCircle( VECTOR2D( lineLength, 0.0 ), aWidth / 2, -M_PI / 2 );
+
+        Restore();
+    }
+}
+
+
+void OPENGL_GAL::DrawCircle( const VECTOR2D& aCenterPoint, double aRadius )
+{
+    if( isFillEnabled )
+    {
+        currentManager->Color( fillColor.r, fillColor.g, fillColor.b, fillColor.a );
+
+        /* Draw a triangle that contains the circle, then shade it leaving only the circle.
+         *  Parameters given to setShader are indices of the triangle's vertices
+         *  (if you want to understand more, check the vertex shader source [shader.vert]).
+         *  Shader uses this coordinates to determine if fragments are inside the circle or not.
+         *       v2
+         *       /\
+         *      //\\
+         *  v0 /_\/_\ v1
+         */
+        currentManager->Shader( SHADER_FILLED_CIRCLE, 1.0 );
+        currentManager->Vertex( aCenterPoint.x - aRadius * sqrt( 3.0f ),            // v0
+                                aCenterPoint.y - aRadius, layerDepth );
+
+        currentManager->Shader( SHADER_FILLED_CIRCLE, 2.0 );
+        currentManager->Vertex( aCenterPoint.x + aRadius * sqrt( 3.0f ),             // v1
+                                aCenterPoint.y - aRadius, layerDepth );
+
+        currentManager->Shader( SHADER_FILLED_CIRCLE, 3.0 );
+        currentManager->Vertex( aCenterPoint.x, aCenterPoint.y + aRadius * 2.0f,    // v2
+                                layerDepth );
+    }
+
+    if( isStrokeEnabled )
+    {
+        currentManager->Color( strokeColor.r, strokeColor.g, strokeColor.b, strokeColor.a );
+
+        /* Draw a triangle that contains the circle, then shade it leaving only the circle.
+         *  Parameters given to setShader are indices of the triangle's vertices
+         *  (if you want to understand more, check the vertex shader source [shader.vert]).
+         *  and the line width. Shader uses this coordinates to determine if fragments are
+         *  inside the circle or not.
+         *       v2
+         *       /\
+         *      //\\
+         *  v0 /_\/_\ v1
+         */
+        double outerRadius = aRadius + ( lineWidth / 2 );
+        currentManager->Shader( SHADER_STROKED_CIRCLE, 1.0, aRadius, lineWidth );
+        currentManager->Vertex( aCenterPoint.x - outerRadius * sqrt( 3.0f ),            // v0
+                                aCenterPoint.y - outerRadius, layerDepth );
+
+        currentManager->Shader( SHADER_STROKED_CIRCLE, 2.0, aRadius, lineWidth );
+        currentManager->Vertex( aCenterPoint.x + outerRadius * sqrt( 3.0f ),            // v1
+                                aCenterPoint.y - outerRadius, layerDepth );
+
+        currentManager->Shader( SHADER_STROKED_CIRCLE, 3.0, aRadius, lineWidth );
+        currentManager->Vertex( aCenterPoint.x, aCenterPoint.y + outerRadius * 2.0f,    // v2
+                                layerDepth );
+    }
+}
+
+
+void OPENGL_GAL::DrawArc( const VECTOR2D& aCenterPoint, double aRadius, double aStartAngle,
+                          double aEndAngle )
+{
+    if( aRadius <= 0 )
+        return;
+
+    // Swap the angles, if start angle is greater than end angle
+    SWAP( aStartAngle, >, aEndAngle );
+
+    Save();
+    currentManager->Translate( aCenterPoint.x, aCenterPoint.y, 0.0 );
+
+    if( isStrokeEnabled )
+    {
+        const double alphaIncrement = 2.0 * M_PI / CIRCLE_POINTS;
+        currentManager->Color( strokeColor.r, strokeColor.g, strokeColor.b, strokeColor.a );
+
+        VECTOR2D p( cos( aStartAngle ) * aRadius, sin( aStartAngle ) * aRadius );
+        double alpha;
+
+        for( alpha = aStartAngle + alphaIncrement; alpha <= aEndAngle; alpha += alphaIncrement )
+        {
+            VECTOR2D p_next( cos( alpha ) * aRadius, sin( alpha ) * aRadius );
+            DrawLine( p, p_next );
+
+            p = p_next;
+        }
+
+        // Draw the last missing part
+        if( alpha != aEndAngle )
+        {
+            VECTOR2D p_last( cos( aEndAngle ) * aRadius, sin( aEndAngle ) * aRadius );
+            DrawLine( p, p_last );
+        }
+    }
+
+    if( isFillEnabled )
+    {
+        const double alphaIncrement = 2 * M_PI / CIRCLE_POINTS;
+        double alpha;
+        currentManager->Color( fillColor.r, fillColor.g, fillColor.b, fillColor.a );
+        currentManager->Shader( SHADER_NONE );
+
+        // Triangle fan
+        for( alpha = aStartAngle; ( alpha + alphaIncrement ) < aEndAngle; )
+        {
+            currentManager->Vertex( 0.0, 0.0, 0.0 );
+            currentManager->Vertex( cos( alpha ) * aRadius, sin( alpha ) * aRadius, 0.0 );
+            alpha += alphaIncrement;
+            currentManager->Vertex( cos( alpha ) * aRadius, sin( alpha ) * aRadius, 0.0 );
+        }
+
+        // The last missing triangle
+        const VECTOR2D endPoint( cos( aEndAngle ) * aRadius, sin( aEndAngle ) * aRadius );
+        currentManager->Vertex( 0.0, 0.0, 0.0 );
+        currentManager->Vertex( cos( alpha ) * aRadius, sin( alpha ) * aRadius, 0.0 );
+        currentManager->Vertex( endPoint.x,    endPoint.y,     0.0 );
+    }
+
+    Restore();
+}
+
+
+void OPENGL_GAL::DrawRectangle( const VECTOR2D& aStartPoint, const VECTOR2D& aEndPoint )
+{
+    // Compute the diagonal points of the rectangle
+    VECTOR2D diagonalPointA( aEndPoint.x, aStartPoint.y );
+    VECTOR2D diagonalPointB( aStartPoint.x, aEndPoint.y );
+
+    // Stroke the outline
+    if( isStrokeEnabled )
+    {
+        currentManager->Color( strokeColor.r, strokeColor.g, strokeColor.b, strokeColor.a );
+
+        std::deque<VECTOR2D> pointList;
+        pointList.push_back( aStartPoint );
+        pointList.push_back( diagonalPointA );
+        pointList.push_back( aEndPoint );
+        pointList.push_back( diagonalPointB );
+        pointList.push_back( aStartPoint );
+        DrawPolyline( pointList );
+    }
+
+    // Fill the rectangle
+    if( isFillEnabled )
+    {
+        currentManager->Shader( SHADER_NONE );
+        currentManager->Color( fillColor.r, fillColor.g, fillColor.b, fillColor.a );
+
+        currentManager->Vertex( aStartPoint.x, aStartPoint.y, layerDepth );
+        currentManager->Vertex( diagonalPointA.x, diagonalPointA.y, layerDepth );
+        currentManager->Vertex( aEndPoint.x, aEndPoint.y, layerDepth );
+
+        currentManager->Vertex( aStartPoint.x, aStartPoint.y, layerDepth );
+        currentManager->Vertex( aEndPoint.x, aEndPoint.y, layerDepth );
+        currentManager->Vertex( diagonalPointB.x, diagonalPointB.y, layerDepth );
+    }
+}
+
+
+void OPENGL_GAL::DrawPolyline( const std::deque<VECTOR2D>& aPointList )
+{
+    if( aPointList.size() < 2 )
+        return;
+
+    currentManager->Color( strokeColor.r, strokeColor.g, strokeColor.b, strokeColor.a );
+
+    std::deque<VECTOR2D>::const_iterator it = aPointList.begin();
+
+    // Start from the second point
+    for( ++it; it != aPointList.end(); ++it )
+    {
+        const VECTOR2D startEndVector = ( *it - *( it - 1 ) );
+        double lineAngle = startEndVector.Angle();
+
+        drawLineQuad( *( it - 1 ), *it );
+
+        // There is no need to draw line caps on both ends of polyline's segments
+        drawFilledSemiCircle( *( it - 1 ), lineWidth / 2, lineAngle + M_PI / 2 );
+    }
+
+    // ..and now - draw the ending cap
+    const VECTOR2D startEndVector = ( *( it - 1 ) - *( it - 2 ) );
+    double lineAngle = startEndVector.Angle();
+    drawFilledSemiCircle( *( it - 1 ), lineWidth / 2, lineAngle - M_PI / 2 );
+}
+
+
+void OPENGL_GAL::DrawPolyline( const VECTOR2D aPointList[], int aListSize )
+{
+    if( aListSize < 2 )
+        return;
+
+    currentManager->Color( strokeColor.r, strokeColor.g, strokeColor.b, strokeColor.a );
+
+    // Start from the second point
+    for( int i = 1; i < aListSize; ++i )
+    {
+        const VECTOR2D startEndVector = ( aPointList[i] - aPointList[i - 1] );
+        double lineAngle = startEndVector.Angle();
+
+        drawLineQuad( aPointList[i - 1], aPointList[i] );
+
+        // There is no need to draw line caps on both ends of polyline's segments
+        drawFilledSemiCircle( aPointList[i - 1], lineWidth / 2, lineAngle + M_PI / 2 );
+    }
+
+    // ..and now - draw the ending cap
+    const VECTOR2D startEndVector = ( aPointList[aListSize - 1] - aPointList[aListSize - 2] );
+    double lineAngle = startEndVector.Angle();
+    drawFilledSemiCircle( aPointList[aListSize - 1], lineWidth / 2, lineAngle - M_PI / 2 );
+}
+
+
+void OPENGL_GAL::DrawPolygon( const std::deque<VECTOR2D>& aPointList )
+{
+    currentManager->Shader( SHADER_NONE );
+    currentManager->Color( fillColor.r, fillColor.g, fillColor.b, fillColor.a );
+
+    // Any non convex polygon needs to be tesselated
+    // for this purpose the GLU standard functions are used
+    TessParams params = { currentManager, tessIntersects };
+    gluTessBeginPolygon( tesselator, &params );
+    gluTessBeginContour( tesselator );
+
+    boost::shared_array<GLdouble> points( new GLdouble[3 * aPointList.size()] );
+    int v = 0;
+
+    for( std::deque<VECTOR2D>::const_iterator it = aPointList.begin(); it != aPointList.end(); ++it )
+    {
+        points[v]     = it->x;
+        points[v + 1] = it->y;
+        points[v + 2] = layerDepth;
+        gluTessVertex( tesselator, &points[v], &points[v] );
+        v += 3;
+    }
+
+    gluTessEndContour( tesselator );
+    gluTessEndPolygon( tesselator );
+
+    // Free allocated intersecting points
+    tessIntersects.clear();
+
+    // vertexList destroyed here
+}
+
+
+void OPENGL_GAL::DrawPolygon( const VECTOR2D aPointList[], int aListSize )
+{
+    currentManager->Shader( SHADER_NONE );
+    currentManager->Color( fillColor.r, fillColor.g, fillColor.b, fillColor.a );
+
+    // Any non convex polygon needs to be tesselated
+    // for this purpose the GLU standard functions are used
+    TessParams params = { currentManager, tessIntersects };
+    gluTessBeginPolygon( tesselator, &params );
+    gluTessBeginContour( tesselator );
+
+    boost::shared_array<GLdouble> points( new GLdouble[3 * aListSize] );
+    int v = 0;
+    const VECTOR2D* ptr = aPointList;
+
+    for( int i = 0; i < aListSize; ++i )
+    {
+        points[v]     = ptr->x;
+        points[v + 1] = ptr->y;
+        points[v + 2] = layerDepth;
+        gluTessVertex( tesselator, &points[v], &points[v] );
+        ++ptr;
+        v += 3;
+    }
+
+    gluTessEndContour( tesselator );
+    gluTessEndPolygon( tesselator );
+
+    // Free allocated intersecting points
+    tessIntersects.clear();
+
+    // vertexList destroyed here
+}
+
+
+void OPENGL_GAL::DrawCurve( const VECTOR2D& aStartPoint, const VECTOR2D& aControlPointA,
+                            const VECTOR2D& aControlPointB, const VECTOR2D& aEndPoint )
+{
+    // FIXME The drawing quality needs to be improved
+    // FIXME Perhaps choose a quad/triangle strip instead?
+    // FIXME Brute force method, use a better (recursive?) algorithm
+
+    std::deque<VECTOR2D> pointList;
+
+    double t  = 0.0;
+    double dt = 1.0 / (double) CURVE_POINTS;
+
+    for( int i = 0; i <= CURVE_POINTS; i++ )
+    {
+        double omt  = 1.0 - t;
+        double omt2 = omt * omt;
+        double omt3 = omt * omt2;
+        double t2   = t * t;
+        double t3   = t * t2;
+
+        VECTOR2D vertex = omt3 * aStartPoint + 3.0 * t * omt2 * aControlPointA
+                          + 3.0 * t2 * omt * aControlPointB + t3 * aEndPoint;
+
+        pointList.push_back( vertex );
+
+        t += dt;
+    }
+
+    DrawPolyline( pointList );
+}
+
+
+void OPENGL_GAL::ResizeScreen( int aWidth, int aHeight )
+{
+    screenSize = VECTOR2I( aWidth, aHeight );
+
+#ifdef RETINA_OPENGL_PATCH
+    const float scaleFactor = GetBackingScaleFactor();
+#else
+    const float scaleFactor = 1.0f;
+#endif
+
+    // Resize framebuffers
+    compositor.Resize( aWidth * scaleFactor, aHeight * scaleFactor );
+    isFramebufferInitialized = false;
+
+    wxGLCanvas::SetSize( aWidth, aHeight );
+}
+
+
+bool OPENGL_GAL::Show( bool aShow )
+{
+    bool s = wxGLCanvas::Show( aShow );
+
+    if( aShow )
+        wxGLCanvas::Raise();
+
+    return s;
+}
+
+
+void OPENGL_GAL::Flush()
+{
+    glFlush();
+}
+
+
+void OPENGL_GAL::ClearScreen( const COLOR4D& aColor )
+{
+    // Clear screen
+    glClearColor( aColor.r, aColor.g, aColor.b, aColor.a );
+    glClear( GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT );
+}
+
+
+void OPENGL_GAL::Transform( const MATRIX3x3D& aTransformation )
+{
+    GLdouble matrixData[16] = { 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1 };
+
+    matrixData[0]   = aTransformation.m_data[0][0];
+    matrixData[1]   = aTransformation.m_data[1][0];
+    matrixData[2]   = aTransformation.m_data[2][0];
+    matrixData[4]   = aTransformation.m_data[0][1];
+    matrixData[5]   = aTransformation.m_data[1][1];
+    matrixData[6]   = aTransformation.m_data[2][1];
+    matrixData[12]  = aTransformation.m_data[0][2];
+    matrixData[13]  = aTransformation.m_data[1][2];
+    matrixData[14]  = aTransformation.m_data[2][2];
+
+    glMultMatrixd( matrixData );
+}
+
+
+void OPENGL_GAL::Rotate( double aAngle )
+{
+    currentManager->Rotate( aAngle, 0.0f, 0.0f, 1.0f );
+}
+
+
+void OPENGL_GAL::Translate( const VECTOR2D& aVector )
+{
+    currentManager->Translate( aVector.x, aVector.y, 0.0f );
+}
+
+
+void OPENGL_GAL::Scale( const VECTOR2D& aScale )
+{
+    currentManager->Scale( aScale.x, aScale.y, 0.0f );
+}
+
+
+void OPENGL_GAL::Save()
+{
+    currentManager->PushMatrix();
+}
+
+
+void OPENGL_GAL::Restore()
+{
+    currentManager->PopMatrix();
+}
+
+
+int OPENGL_GAL::BeginGroup()
+{
+    isGrouping = true;
+
+    boost::shared_ptr<VERTEX_ITEM> newItem( new VERTEX_ITEM( cachedManager ) );
+    int groupNumber = getNewGroupNumber();
+    groups.insert( std::make_pair( groupNumber, newItem ) );
+
+    return groupNumber;
+}
+
+
+void OPENGL_GAL::EndGroup()
+{
+    cachedManager.FinishItem();
+    isGrouping = false;
+}
+
+
+void OPENGL_GAL::DrawGroup( int aGroupNumber )
+{
+    cachedManager.DrawItem( *groups[aGroupNumber] );
+}
+
+
+void OPENGL_GAL::ChangeGroupColor( int aGroupNumber, const COLOR4D& aNewColor )
+{
+    cachedManager.ChangeItemColor( *groups[aGroupNumber], aNewColor );
+}
+
+
+void OPENGL_GAL::ChangeGroupDepth( int aGroupNumber, int aDepth )
+{
+    cachedManager.ChangeItemDepth( *groups[aGroupNumber], aDepth );
+}
+
+
+void OPENGL_GAL::DeleteGroup( int aGroupNumber )
+{
+    // Frees memory in the container as well
+    groups.erase( aGroupNumber );
+}
+
+
+void OPENGL_GAL::ClearCache()
+{
+    groups.clear();
+    cachedManager.Clear();
+}
+
+
+void OPENGL_GAL::SaveScreen()
+{
+    wxASSERT_MSG( false, wxT( "Not implemented yet" ) );
+}
+
+
+void OPENGL_GAL::RestoreScreen()
+{
+    wxASSERT_MSG( false, wxT( "Not implemented yet" ) );
+}
+
+
+void OPENGL_GAL::SetTarget( RENDER_TARGET aTarget )
+{
+    switch( aTarget )
+    {
+    default:
+    case TARGET_CACHED:
+        currentManager = &cachedManager;
+        break;
+
+    case TARGET_NONCACHED:
+        currentManager = &nonCachedManager;
+        break;
+
+    case TARGET_OVERLAY:
+        currentManager = &overlayManager;
+        break;
+    }
+
+    currentTarget = aTarget;
+}
+
+
+RENDER_TARGET OPENGL_GAL::GetTarget() const
+{
+    return currentTarget;
+}
+
+
+void OPENGL_GAL::ClearTarget( RENDER_TARGET aTarget )
+{
+    // Save the current state
+    unsigned int oldTarget = compositor.GetBuffer();
+
+    switch( aTarget )
+    {
+    // Cached and noncached items are rendered to the same buffer
+    default:
+    case TARGET_CACHED:
+    case TARGET_NONCACHED:
+        compositor.SetBuffer( mainBuffer );
+        break;
+
+    case TARGET_OVERLAY:
+        compositor.SetBuffer( overlayBuffer );
+        break;
+    }
+
+    compositor.ClearBuffer();
+
+    // Restore the previous state
+    compositor.SetBuffer( oldTarget );
+}
+
+
+void OPENGL_GAL::DrawCursor( const VECTOR2D& aCursorPosition )
+{
+    // Now we should only store the position of the mouse cursor
+    // The real drawing routines are in blitCursor()
+    VECTOR2D screenCursor = worldScreenMatrix * aCursorPosition;
+
+    cursorPosition = screenWorldMatrix * VECTOR2D( screenCursor.x, screenSize.y - screenCursor.y );
+}
+
+
+void OPENGL_GAL::drawGridLine( const VECTOR2D& aStartPoint, const VECTOR2D& aEndPoint )
+{
+    compositor.SetBuffer( mainBuffer );
+
+    // We do not need a very precise comparison here (the lineWidth is set by GAL::DrawGrid())
+    if( fabs( lineWidth - 2.0 * gridLineWidth / worldScale ) < 0.1 )
+        glLineWidth( 1.0 );
+    else
+        glLineWidth( 2.0 );
+
+    glColor4d( gridColor.r, gridColor.g, gridColor.b, gridColor.a );
+
+    glBegin( GL_LINES );
+    glVertex3d( aStartPoint.x, aStartPoint.y, layerDepth );
+    glVertex3d( aEndPoint.x, aEndPoint.y, layerDepth );
+    glEnd();
+
+    // Restore the default color, so textures will be drawn properly
+    glColor4d( 1.0, 1.0, 1.0, 1.0 );
+}
+
+
+void OPENGL_GAL::drawLineQuad( const VECTOR2D& aStartPoint, const VECTOR2D& aEndPoint )
+{
+    /* Helper drawing:                   ____--- v3       ^
+     *                           ____---- ...   \          \
+     *                   ____----      ...       \   end    \
+     *     v1    ____----           ...    ____----          \ width
+     *       ----                ...___----        \          \
+     *       \             ___...--                 \          v
+     *        \    ____----...                ____---- v2
+     *         ----     ...           ____----
+     *  start   \    ...      ____----
+     *           \... ____----
+     *            ----
+     *            v0
+     * dots mark triangles' hypotenuses
+     */
+
+    VECTOR2D startEndVector = aEndPoint - aStartPoint;
+    double   lineLength     = startEndVector.EuclideanNorm();
+
+    if( lineLength <= 0.0 )
+        return;
+
+    double   scale          = 0.5 * lineWidth / lineLength;
+
+    // The perpendicular vector also needs transformations
+    glm::vec4 vector = currentManager->GetTransformation() *
+                       glm::vec4( -startEndVector.y * scale, startEndVector.x * scale, 0.0, 0.0 );
+
+    // Line width is maintained by the vertex shader
+    currentManager->Shader( SHADER_LINE, vector.x, vector.y, lineWidth );
+    currentManager->Vertex( aStartPoint.x, aStartPoint.y, layerDepth );    // v0
+
+    currentManager->Shader( SHADER_LINE, -vector.x, -vector.y, lineWidth );
+    currentManager->Vertex( aStartPoint.x, aStartPoint.y, layerDepth );    // v1
+
+    currentManager->Shader( SHADER_LINE, -vector.x, -vector.y, lineWidth );
+    currentManager->Vertex( aEndPoint.x, aEndPoint.y, layerDepth );        // v3
+
+    currentManager->Shader( SHADER_LINE, vector.x, vector.y, lineWidth );
+    currentManager->Vertex( aStartPoint.x, aStartPoint.y, layerDepth );    // v0
+
+    currentManager->Shader( SHADER_LINE, -vector.x, -vector.y, lineWidth );
+    currentManager->Vertex( aEndPoint.x, aEndPoint.y, layerDepth );        // v3
+
+    currentManager->Shader( SHADER_LINE, vector.x, vector.y, lineWidth );
+    currentManager->Vertex( aEndPoint.x, aEndPoint.y, layerDepth );        // v2
+}
+
+
+void OPENGL_GAL::drawSemiCircle( const VECTOR2D& aCenterPoint, double aRadius, double aAngle )
+{
+    if( isFillEnabled )
+    {
+        currentManager->Color( fillColor.r, fillColor.g, fillColor.b, fillColor.a );
+        drawFilledSemiCircle( aCenterPoint, aRadius, aAngle );
+    }
+
+    if( isStrokeEnabled )
+    {
+        currentManager->Color( strokeColor.r, strokeColor.g, strokeColor.b, strokeColor.a );
+        drawStrokedSemiCircle( aCenterPoint, aRadius, aAngle );
+    }
+}
+
+
+void OPENGL_GAL::drawFilledSemiCircle( const VECTOR2D& aCenterPoint, double aRadius,
+                                       double aAngle )
+{
+    Save();
+    currentManager->Translate( aCenterPoint.x, aCenterPoint.y, 0.0f );
+    currentManager->Rotate( aAngle, 0.0f, 0.0f, 1.0f );
+
+    /* Draw a triangle that contains the semicircle, then shade it to leave only
+     * the semicircle. Parameters given to setShader are indices of the triangle's vertices
+     * (if you want to understand more, check the vertex shader source [shader.vert]).
+     * Shader uses these coordinates to determine if fragments are inside the semicircle or not.
+     *       v2
+     *       /\
+     *      /__\
+     *  v0 //__\\ v1
+     */
+    currentManager->Shader( SHADER_FILLED_CIRCLE, 4.0f );
+    currentManager->Vertex( -aRadius * 3.0f / sqrt( 3.0f ), 0.0f, layerDepth );     // v0
+
+    currentManager->Shader( SHADER_FILLED_CIRCLE, 5.0f );
+    currentManager->Vertex( aRadius * 3.0f / sqrt( 3.0f ), 0.0f, layerDepth );      // v1
+
+    currentManager->Shader( SHADER_FILLED_CIRCLE, 6.0f );
+    currentManager->Vertex( 0.0f, aRadius * 2.0f, layerDepth );                     // v2
+
+    Restore();
+}
+
+
+void OPENGL_GAL::drawStrokedSemiCircle( const VECTOR2D& aCenterPoint, double aRadius,
+                                        double aAngle )
+{
+    double outerRadius = aRadius + ( lineWidth / 2 );
+
+    Save();
+    currentManager->Translate( aCenterPoint.x, aCenterPoint.y, 0.0f );
+    currentManager->Rotate( aAngle, 0.0f, 0.0f, 1.0f );
+
+    /* Draw a triangle that contains the semicircle, then shade it to leave only
+     * the semicircle. Parameters given to setShader are indices of the triangle's vertices
+     * (if you want to understand more, check the vertex shader source [shader.vert]), the
+     * radius and the line width. Shader uses these coordinates to determine if fragments are
+     * inside the semicircle or not.
+     *       v2
+     *       /\
+     *      /__\
+     *  v0 //__\\ v1
+     */
+    currentManager->Shader( SHADER_STROKED_CIRCLE, 4.0f, aRadius, lineWidth );
+    currentManager->Vertex( -outerRadius * 3.0f / sqrt( 3.0f ), 0.0f, layerDepth );     // v0
+
+    currentManager->Shader( SHADER_STROKED_CIRCLE, 5.0f, aRadius, lineWidth );
+    currentManager->Vertex( outerRadius * 3.0f / sqrt( 3.0f ), 0.0f, layerDepth );      // v1
+
+    currentManager->Shader( SHADER_STROKED_CIRCLE, 6.0f, aRadius, lineWidth );
+    currentManager->Vertex( 0.0f, outerRadius * 2.0f, layerDepth );                     // v2
+
+    Restore();
+}
+
+
+void OPENGL_GAL::onPaint( wxPaintEvent& WXUNUSED( aEvent ) )
+{
+    PostPaint();
+}
+
+
+void OPENGL_GAL::skipMouseEvent( wxMouseEvent& aEvent )
+{
+    // Post the mouse event to the event listener registered in constructor, if any
+    if( mouseListener )
+        wxPostEvent( mouseListener, aEvent );
+}
+
+
+void OPENGL_GAL::blitCursor()
+{
+    if( !isCursorEnabled )
+        return;
+
+    compositor.SetBuffer( OPENGL_COMPOSITOR::DIRECT_RENDERING );
+
+    VECTOR2D cursorBegin  = cursorPosition - cursorSize / ( 2 * worldScale );
+    VECTOR2D cursorEnd    = cursorPosition + cursorSize / ( 2 * worldScale );
+    VECTOR2D cursorCenter = ( cursorBegin + cursorEnd ) / 2;
+
+    glDisable( GL_TEXTURE_2D );
+    glLineWidth( 1.0 );
+    glColor4d( cursorColor.r, cursorColor.g, cursorColor.b, cursorColor.a );
+
+    glBegin( GL_LINES );
+    glVertex2d( cursorCenter.x, cursorBegin.y );
+    glVertex2d( cursorCenter.x, cursorEnd.y );
+
+    glVertex2d( cursorBegin.x, cursorCenter.y );
+    glVertex2d( cursorEnd.x, cursorCenter.y );
+    glEnd();
+}
+
+
+unsigned int OPENGL_GAL::getNewGroupNumber()
+{
+    wxASSERT_MSG( groups.size() < std::numeric_limits<unsigned int>::max(),
+                  wxT( "There are no free slots to store a group" ) );
+
+    while( groups.find( groupCounter ) != groups.end() )
+    {
+        groupCounter++;
+    }
+
+    return groupCounter++;
+}
+
+
+bool OPENGL_GAL::runTest()
+{
+    wxDialog dlgtest( GetParent(), -1, wxT( "opengl test" ), wxPoint( 50, 50 ),
+                      wxDLG_UNIT( GetParent(), wxSize( 50, 50 ) ) );
+    OPENGL_TEST* test = new OPENGL_TEST( &dlgtest, this );
+
+    dlgtest.Raise();         // on Linux, on some windows managers (Unity for instance) this is needed to actually show the dialog
+    dlgtest.ShowModal();
+    bool result = test->IsOk();
+
+    if( !result )
+        throw std::runtime_error( test->GetError() );
+
+    return result;
+}
+
+
+OPENGL_GAL::OPENGL_TEST::OPENGL_TEST( wxDialog* aParent, OPENGL_GAL* aGal ) :
+    wxGLCanvas( aParent, wxID_ANY, glAttributes, wxDefaultPosition,
+                wxDefaultSize, 0, wxT( "GLCanvas" ) ),
+    m_parent( aParent ), m_gal( aGal ), m_tested( false ), m_result( false )
+{
+    m_timeoutTimer.SetOwner( this );
+    Connect( wxEVT_PAINT, wxPaintEventHandler( OPENGL_GAL::OPENGL_TEST::Render ) );
+    Connect( wxEVT_TIMER, wxTimerEventHandler( OPENGL_GAL::OPENGL_TEST::OnTimeout ) );
+    m_parent->Connect( wxEVT_PAINT, wxPaintEventHandler( OPENGL_GAL::OPENGL_TEST::OnDialogPaint ), NULL, this );
+}
+
+
+void OPENGL_GAL::OPENGL_TEST::Render( wxPaintEvent& WXUNUSED( aEvent ) )
+{
+    if( !m_tested )
+    {
+        if( !IsShownOnScreen() )
+            return;
+
+        m_timeoutTimer.Stop();
+        m_result = true;    // Assume everything is fine, until proven otherwise
+
+        // One test is enough - close the testing dialog when the test is finished
+        Disconnect( wxEVT_PAINT, wxPaintEventHandler( OPENGL_GAL::OPENGL_TEST::Render ) );
+        CallAfter( boost::bind( &wxDialog::EndModal, m_parent, wxID_NONE ) );
+
+        SetCurrent( *OPENGL_GAL::glContext );
+        GLenum err = glewInit();
+
+        if( GLEW_OK != err )
+        {
+            error( (const char*) glewGetErrorString( err ) );
+            return;
+        }
+        else
+        {
+            wxLogDebug( wxString( wxT( "Status: Using GLEW " ) ) +
+                        FROM_UTF8( (char*) glewGetString( GLEW_VERSION ) ) );
+        }
+
+        // Check the OpenGL version (minimum 2.1 is required)
+        if( GLEW_VERSION_2_1 )
+        {
+            wxLogInfo( wxT( "OpenGL 2.1 supported." ) );
+        }
+        else
+        {
+            error( "OpenGL 2.1 or higher is required!" );
+            return;
+        }
+
+        // Framebuffers have to be supported
+        if( !GLEW_EXT_framebuffer_object )
+        {
+            error( "Framebuffer objects are not supported!" );
+            return;
+        }
+
+        // Vertex buffer has to be supported
+        if( !GLEW_ARB_vertex_buffer_object )
+        {
+            error( "Vertex buffer objects are not supported!" );
+            return;
+        }
+
+        // Prepare shaders
+        if( !m_gal->shader.LoadBuiltinShader( 0, SHADER_TYPE_VERTEX ) )
+        {
+            error( "Cannot compile vertex shader!" );
+            return;
+        }
+
+        if( !m_gal->shader.LoadBuiltinShader( 1, SHADER_TYPE_FRAGMENT ) )
+        {
+            error( "Cannot compile fragment shader!" );
+            return;
+        }
+
+        if( !m_gal->shader.Link() )
+        {
+            error( "Cannot link the shaders!" );
+            return;
+        }
+
+        m_tested = true;
+    }
+}
+
+
+void OPENGL_GAL::OPENGL_TEST::OnTimeout( wxTimerEvent& aEvent )
+{
+    error( "Could not create OpenGL canvas" );
+    m_parent->EndModal( wxID_NONE );
+}
+
+
+void OPENGL_GAL::OPENGL_TEST::OnDialogPaint( wxPaintEvent& aEvent )
+{
+    // GL canvas may never appear on the screen (e.g. due to missing GL extensions), and the test
+    // will not be run. Therefore give at most a second to perform the test, otherwise we conclude
+    // it has failed.
+    // Also, wxWidgets OnShow event is triggered before a window is shown, therefore here we use
+    // OnPaint event, which is executed when a window is actually visible.
+    m_timeoutTimer.StartOnce( 1000 );
+}
+
+
+void OPENGL_GAL::OPENGL_TEST::error( const std::string& aError )
+{
+    m_timeoutTimer.Stop();
+    m_result = false;
+    m_tested = true;
+    m_error = aError;
+}
+
+// ------------------------------------- // Callback functions for the tesselator // ------------------------------------- // Compare Redbook Chapter 11
+void CALLBACK VertexCallback( GLvoid* aVertexPtr, void* aData )
+{
+    GLdouble* vertex = static_cast<GLdouble*>( aVertexPtr );
+    OPENGL_GAL::TessParams* param = static_cast<OPENGL_GAL::TessParams*>( aData );
+    VERTEX_MANAGER* vboManager = param->vboManager;
+
+    if( vboManager )
+        vboManager->Vertex( vertex[0], vertex[1], vertex[2] );
+}
+
+
+void CALLBACK CombineCallback( GLdouble coords[3],
+                               GLdouble* vertex_data[4],
+                               GLfloat weight[4], GLdouble** dataOut, void* aData )
+{
+    GLdouble* vertex = new GLdouble[3];
+    OPENGL_GAL::TessParams* param = static_cast<OPENGL_GAL::TessParams*>( aData );
+
+    // Save the pointer so we can delete it later
+    param->intersectPoints.push_back( boost::shared_array<GLdouble>( vertex ) );
+
+    memcpy( vertex, coords, 3 * sizeof(GLdouble) );
+
+    *dataOut = vertex;
+}
+
+
+void CALLBACK EdgeCallback( GLboolean aEdgeFlag )
+{
+    // This callback is needed to force GLU tesselator to use triangles only
+}
+
+
+void CALLBACK ErrorCallback( GLenum aErrorCode )
+{
+    //throw std::runtime_error( std::string( "Tessellation error: " ) +
+                              //std::string( (const char*) gluErrorString( aErrorCode ) );
+}
+
+
+static void InitTesselatorCallbacks( GLUtesselator* aTesselator )
+{
+    gluTessCallback( aTesselator, GLU_TESS_VERTEX_DATA,  ( void (CALLBACK*)() )VertexCallback );
+    gluTessCallback( aTesselator, GLU_TESS_COMBINE_DATA, ( void (CALLBACK*)() )CombineCallback );
+    gluTessCallback( aTesselator, GLU_TESS_EDGE_FLAG,    ( void (CALLBACK*)() )EdgeCallback );
+    gluTessCallback( aTesselator, GLU_TESS_ERROR,        ( void (CALLBACK*)() )ErrorCallback );
+}
diff --git a/common/gal/opengl/shader.cpp b/common/gal/opengl/shader.cpp
new file mode 100644
index 0000000..228d53d
--- /dev/null
+++ b/common/gal/opengl/shader.cpp
@@ -0,0 +1,272 @@
+/*
+ * This program source code file is part of KICAD, a free EDA CAD application.
+ *
+ * Copyright (C) 2012 Torsten Hueter, torstenhtr <at> gmx.de
+ * Copyright (C) 2012 Kicad Developers, see change_log.txt for contributors.
+ *
+ * Graphics Abstraction Layer (GAL) for OpenGL
+ *
+ * Shader class
+ *
+ * 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 <iostream>
+#include <fstream>
+#include <stdexcept>
+
+#include <cstring>
+#include <cassert>
+
+#include <gal/opengl/shader.h>
+#include "shader_src.h"
+
+using namespace KIGFX;
+
+SHADER::SHADER() :
+    isProgramCreated( false ),
+    isShaderLinked( false ),
+    active( false ),
+    maximumVertices( 4 ),
+    geomInputType( GL_LINES ),
+    geomOutputType( GL_LINES )
+
+{
+    // Do not have uninitialized members:
+    programNumber = 0;
+}
+
+
+SHADER::~SHADER()
+{
+    if( isProgramCreated )
+    {
+        // Delete the shaders and the program
+        for( std::deque<GLuint>::iterator it = shaderNumbers.begin(); it != shaderNumbers.end();
+             ++it )
+        {
+            glDeleteShader( *it );
+        }
+
+        glDeleteProgram( programNumber );
+    }
+}
+
+
+bool SHADER::LoadBuiltinShader( unsigned int aShaderNumber, SHADER_TYPE aShaderType )
+{
+    if( aShaderNumber >= shaders_number )
+        return false;
+
+    return addSource( std::string( shaders_src[aShaderNumber] ), aShaderType );
+}
+
+
+bool SHADER::LoadShaderFromFile( const std::string& aShaderSourceName, SHADER_TYPE aShaderType )
+{
+    // Load shader sources
+    const std::string shaderSource = readSource( aShaderSourceName );
+
+    return addSource( shaderSource, aShaderType );
+}
+
+
+void SHADER::ConfigureGeometryShader( GLuint maxVertices, GLuint geometryInputType,
+                                      GLuint geometryOutputType )
+{
+    maximumVertices = maxVertices;
+    geomInputType   = geometryInputType;
+    geomOutputType  = geometryOutputType;
+}
+
+
+bool SHADER::Link()
+{
+    // Shader linking
+    glLinkProgram( programNumber );
+    programInfo( programNumber );
+
+    // Check the Link state
+    glGetObjectParameterivARB( programNumber, GL_OBJECT_LINK_STATUS_ARB,
+                               (GLint*) &isShaderLinked );
+
+#ifdef DEBUG
+    if( !isShaderLinked )
+    {
+        int maxLength;
+        glGetProgramiv( programNumber, GL_INFO_LOG_LENGTH, &maxLength );
+        maxLength = maxLength + 1;
+        char* linkInfoLog = new char[maxLength];
+        glGetProgramInfoLog( programNumber, maxLength, &maxLength, linkInfoLog );
+        std::cerr << "Shader linking error:" << std::endl;
+        std::cerr << linkInfoLog;
+        delete[] linkInfoLog;
+    }
+#endif /* DEBUG */
+
+    return isShaderLinked;
+}
+
+
+int SHADER::AddParameter( const std::string& aParameterName )
+{
+    GLint location = glGetUniformLocation( programNumber, aParameterName.c_str() );
+
+    if( location != -1 )
+        parameterLocation.push_back( location );
+
+    return location;
+}
+
+
+void SHADER::SetParameter( int parameterNumber, float value ) const
+{
+    glUniform1f( parameterLocation[parameterNumber], value );
+}
+
+
+void SHADER::SetParameter( int parameterNumber, int value ) const
+{
+    glUniform1i( parameterLocation[parameterNumber], value );
+}
+
+
+int SHADER::GetAttribute( std::string aAttributeName ) const
+{
+    return glGetAttribLocation( programNumber, aAttributeName.c_str() );
+}
+
+
+void SHADER::programInfo( GLuint aProgram )
+{
+    GLint glInfoLogLength = 0;
+    GLint writtenChars    = 0;
+
+    // Get the length of the info string
+    glGetProgramiv( aProgram, GL_INFO_LOG_LENGTH, &glInfoLogLength );
+
+    // Print the information
+    if( glInfoLogLength > 2 )
+    {
+        GLchar* glInfoLog = new GLchar[glInfoLogLength];
+        glGetProgramInfoLog( aProgram, glInfoLogLength, &writtenChars, glInfoLog );
+
+        std::cerr << glInfoLog << std::endl;
+
+        delete[] glInfoLog;
+    }
+}
+
+
+void SHADER::shaderInfo( GLuint aShader )
+{
+    GLint glInfoLogLength = 0;
+    GLint writtenChars    = 0;
+
+    // Get the length of the info string
+    glGetShaderiv( aShader, GL_INFO_LOG_LENGTH, &glInfoLogLength );
+
+    // Print the information
+    if( glInfoLogLength > 2 )
+    {
+        GLchar* glInfoLog = new GLchar[glInfoLogLength];
+        glGetShaderInfoLog( aShader, glInfoLogLength, &writtenChars, glInfoLog );
+
+        std::cerr << glInfoLog << std::endl;
+
+        delete[] glInfoLog;
+    }
+}
+
+
+std::string SHADER::readSource( std::string aShaderSourceName )
+{
+    // Open the shader source for reading
+    std::ifstream inputFile( aShaderSourceName.c_str(), std::ifstream::in );
+    std::string   shaderSource;
+
+    if( !inputFile )
+        throw std::runtime_error( "Can't read the shader source: " + aShaderSourceName );
+
+    std::string shaderSourceLine;
+
+    // Read all lines from the text file
+    while( getline( inputFile, shaderSourceLine ) )
+    {
+        shaderSource += shaderSourceLine;
+        shaderSource += "\n";
+    }
+
+    return shaderSource;
+}
+
+
+bool SHADER::addSource( const std::string& aShaderSource, SHADER_TYPE aShaderType )
+{
+    assert( !isShaderLinked );
+
+    // Create the program
+    if( !isProgramCreated )
+    {
+        programNumber    = glCreateProgram();
+        isProgramCreated = true;
+    }
+
+    // Create a shader
+    GLuint shaderNumber = glCreateShader( aShaderType );
+    shaderNumbers.push_back( shaderNumber );
+
+    // Get the program info
+    programInfo( programNumber );
+
+    // Copy to char array
+    char* source = new char[aShaderSource.size() + 1];
+    strncpy( source, aShaderSource.c_str(), aShaderSource.size() + 1 );
+    const char** source_ = (const char**) ( &source );
+
+    // Attach the source
+    glShaderSource( shaderNumber, 1, source_, NULL );
+    programInfo( programNumber );
+
+    // Delete the allocated char array
+    delete[] source;
+
+    // Compile and attach shader to the program
+    glCompileShader( shaderNumber );
+    GLint status;
+    glGetShaderiv( shaderNumber, GL_COMPILE_STATUS, &status );
+
+    if( status != GL_TRUE )
+    {
+        shaderInfo( shaderNumber );
+        throw std::runtime_error( "Shader compilation error" );
+    }
+
+    glAttachShader( programNumber, shaderNumber );
+    programInfo( programNumber );
+
+    // Special handling for the geometry shader
+    if( aShaderType == SHADER_TYPE_GEOMETRY )
+    {
+        glProgramParameteriEXT( programNumber, GL_GEOMETRY_VERTICES_OUT_EXT, maximumVertices );
+        glProgramParameteriEXT( programNumber, GL_GEOMETRY_INPUT_TYPE_EXT, geomInputType );
+        glProgramParameteriEXT( programNumber, GL_GEOMETRY_OUTPUT_TYPE_EXT, geomOutputType );
+    }
+
+    return true;
+}
diff --git a/common/gal/opengl/shader.frag b/common/gal/opengl/shader.frag
new file mode 100644
index 0000000..7d7d96c
--- /dev/null
+++ b/common/gal/opengl/shader.frag
@@ -0,0 +1,75 @@
+/*
+ * This program source code file is part of KICAD, a free EDA CAD application.
+ *
+ * Copyright (C) 2013 CERN
+ * @author Maciej Suminski <maciej.suminski@cern.ch>
+ *
+ * Fragment shader
+ *
+ * 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
+ */
+
+#version 120
+
+// Shader types
+const float SHADER_LINE                 = 1.0;
+const float SHADER_FILLED_CIRCLE        = 2.0;
+const float SHADER_STROKED_CIRCLE       = 3.0;
+
+varying vec4 shaderParams;
+varying vec2 circleCoords;
+
+void filledCircle( vec2 aCoord )
+{
+    if( dot( aCoord, aCoord ) < 1.0 )
+        gl_FragColor = gl_Color;
+    else
+        discard;
+}
+
+
+void strokedCircle( vec2 aCoord, float aRadius, float aWidth )
+{
+    float outerRadius = aRadius + ( aWidth / 2 );
+    float innerRadius = aRadius - ( aWidth / 2 );
+    float relWidth = innerRadius / outerRadius;
+
+    if( ( dot( aCoord, aCoord ) < 1.0 ) &&
+        ( dot( aCoord, aCoord ) > relWidth * relWidth ) )
+        gl_FragColor = gl_Color;
+    else
+        discard;
+}
+
+
+void main()
+{
+    if( shaderParams[0] == SHADER_FILLED_CIRCLE )
+    {
+        filledCircle( circleCoords );
+    }
+    else if( shaderParams[0] == SHADER_STROKED_CIRCLE )
+    {
+        strokedCircle( circleCoords, shaderParams[2], shaderParams[3] );
+    }
+    else
+    {
+        // Simple pass-through
+        gl_FragColor = gl_Color;
+    }
+}
diff --git a/common/gal/opengl/shader.vert b/common/gal/opengl/shader.vert
new file mode 100644
index 0000000..24521a4
--- /dev/null
+++ b/common/gal/opengl/shader.vert
@@ -0,0 +1,95 @@
+/*
+ * This program source code file is part of KICAD, a free EDA CAD application.
+ *
+ * Copyright (C) 2013 CERN
+ * @author Maciej Suminski <maciej.suminski@cern.ch>
+ *
+ * Vertex shader
+ *
+ * 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
+ */
+
+#version 120
+
+// Shader types
+const float SHADER_LINE                 = 1.0;
+const float SHADER_FILLED_CIRCLE        = 2.0;
+const float SHADER_STROKED_CIRCLE       = 3.0;
+
+// Minimum line width
+const float MIN_WIDTH = 1.0;
+
+attribute vec4 attrShaderParams;
+varying vec4 shaderParams;
+varying vec2 circleCoords;
+
+void main()
+{
+    // Pass attributes to the fragment shader
+    shaderParams = attrShaderParams;
+
+    if( shaderParams[0] == SHADER_LINE )
+    {
+        float lineWidth = shaderParams[3];
+        float worldScale = gl_ModelViewMatrix[0][0];
+
+        // Make lines appear to be at least 1 pixel wide
+        if( worldScale * lineWidth < MIN_WIDTH )
+            gl_Position = gl_ModelViewProjectionMatrix *
+                ( gl_Vertex + vec4( shaderParams.yz * MIN_WIDTH / ( worldScale * lineWidth ), 0.0, 0.0 ) );
+        else
+            gl_Position = gl_ModelViewProjectionMatrix *
+                ( gl_Vertex + vec4( shaderParams.yz, 0.0, 0.0 ) );
+    }
+    else if( ( shaderParams[0] == SHADER_STROKED_CIRCLE ) ||
+             ( shaderParams[0] == SHADER_FILLED_CIRCLE  ) )
+    {
+        // Compute relative circle coordinates basing on indices
+        // Circle
+        if( shaderParams[1] == 1.0 )
+            circleCoords = vec2( -sqrt( 3.0 ), -1.0 );
+        else if( shaderParams[1] == 2.0 )
+            circleCoords = vec2( sqrt( 3.0 ), -1.0 );
+        else if( shaderParams[1] == 3.0 )
+            circleCoords = vec2( 0.0, 2.0 );
+
+        // Semicircle
+        else if( shaderParams[1] == 4.0 )
+            circleCoords = vec2( -3.0 / sqrt( 3.0 ), 0.0 );
+        else if( shaderParams[1] == 5.0 )
+            circleCoords = vec2( 3.0 / sqrt( 3.0 ), 0.0 );
+        else if( shaderParams[1] == 6.0 )
+            circleCoords = vec2( 0.0, 2.0 );
+
+        // Make the line appear to be at least 1 pixel wide
+        float lineWidth = shaderParams[3];
+        float worldScale = gl_ModelViewMatrix[0][0];
+
+        if( worldScale * lineWidth < MIN_WIDTH )
+            shaderParams[3] = shaderParams[3] / ( worldScale * lineWidth );
+
+        gl_Position = ftransform();
+    }
+    else
+    {
+        // Pass through the coordinates like in the fixed pipeline
+        gl_Position = ftransform();
+    }
+
+    gl_FrontColor = gl_Color;
+}
diff --git a/common/gal/opengl/vertex_container.cpp b/common/gal/opengl/vertex_container.cpp
new file mode 100644
index 0000000..fa41ecb
--- /dev/null
+++ b/common/gal/opengl/vertex_container.cpp
@@ -0,0 +1,60 @@
+/*
+ * This program source code file is part of KiCad, a free EDA CAD application.
+ *
+ * Copyright (C) 2013 CERN
+ * @author Maciej Suminski <maciej.suminski@cern.ch>
+ *
+ * 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
+ */
+
+/**
+ * @file vertex_container.cpp
+ * @brief Class to store vertices and handle transfers between system memory and GPU memory.
+ */
+
+#include <gal/opengl/vertex_container.h>
+#include <gal/opengl/cached_container.h>
+#include <gal/opengl/noncached_container.h>
+#include <gal/opengl/shader.h>
+#include <cstdlib>
+#include <cstring>
+
+using namespace KIGFX;
+
+VERTEX_CONTAINER* VERTEX_CONTAINER::MakeContainer( bool aCached )
+{
+    if( aCached )
+        return new CACHED_CONTAINER;
+    else
+        return new NONCACHED_CONTAINER;
+}
+
+
+VERTEX_CONTAINER::VERTEX_CONTAINER( unsigned int aSize ) :
+    m_freeSpace( aSize ), m_currentSize( aSize ), m_initialSize( aSize ),
+    m_failed( false ), m_dirty( true )
+{
+    m_vertices = static_cast<VERTEX*>( malloc( aSize * sizeof( VERTEX ) ) );
+    memset( m_vertices, 0x00, aSize * sizeof( VERTEX ) );
+}
+
+
+VERTEX_CONTAINER::~VERTEX_CONTAINER()
+{
+    free( m_vertices );
+}
diff --git a/common/gal/opengl/vertex_item.cpp b/common/gal/opengl/vertex_item.cpp
new file mode 100644
index 0000000..a6fd352
--- /dev/null
+++ b/common/gal/opengl/vertex_item.cpp
@@ -0,0 +1,53 @@
+/*
+ * This program source code file is part of KiCad, a free EDA CAD application.
+ *
+ * Copyright (C) 2013 CERN
+ * @author Maciej Suminski <maciej.suminski@cern.ch>
+ *
+ * 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
+ */
+
+/**
+ * @file vertex_item.cpp
+ * @brief Class to handle an item held in a container.
+ */
+
+#include <gal/opengl/vertex_item.h>
+#include <gal/opengl/vertex_manager.h>
+#include <cstring>
+
+using namespace KIGFX;
+
+VERTEX_ITEM::VERTEX_ITEM( const VERTEX_MANAGER& aManager ) :
+    m_manager( aManager ), m_offset( 0 ), m_size( 0 )
+{
+    // As the item is created, we are going to modify it, so call to SetItem() is needed
+    m_manager.SetItem( *this );
+}
+
+
+VERTEX_ITEM::~VERTEX_ITEM()
+{
+    m_manager.FreeItem( *this );
+}
+
+
+VERTEX* VERTEX_ITEM::GetVertices() const
+{
+    return m_manager.GetVertices( *this );
+}
diff --git a/common/gal/opengl/vertex_manager.cpp b/common/gal/opengl/vertex_manager.cpp
new file mode 100644
index 0000000..d055b52
--- /dev/null
+++ b/common/gal/opengl/vertex_manager.cpp
@@ -0,0 +1,234 @@
+/*
+ * This program source code file is part of KiCad, a free EDA CAD application.
+ *
+ * Copyright (C) 2013 CERN
+ * @author Maciej Suminski <maciej.suminski@cern.ch>
+ *
+ * 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
+ */
+
+/**
+ * @file vertex_manager.cpp
+ * @brief Class to control vertex container and GPU with possibility of emulating old-style OpenGL
+ * 1.0 state machine using modern OpenGL methods.
+ */
+
+#include <gal/opengl/vertex_manager.h>
+#include <gal/opengl/cached_container.h>
+#include <gal/opengl/noncached_container.h>
+#include <gal/opengl/gpu_manager.h>
+#include <gal/opengl/vertex_item.h>
+#include <confirm.h>
+
+using namespace KIGFX;
+
+VERTEX_MANAGER::VERTEX_MANAGER( bool aCached ) :
+    m_noTransform( true ), m_transform( 1.0f )
+{
+    m_container.reset( VERTEX_CONTAINER::MakeContainer( aCached ) );
+    m_gpu.reset( GPU_MANAGER::MakeManager( m_container.get() ) );
+
+    // There is no shader used by default
+    for( unsigned int i = 0; i < ShaderStride; ++i )
+        m_shader[i] = 0.0f;
+}
+
+
+void VERTEX_MANAGER::Vertex( GLfloat aX, GLfloat aY, GLfloat aZ ) const
+{
+    // flag to avoid hanging by calling DisplayError too many times:
+    static bool show_err = true;
+
+    // Obtain the pointer to the vertex in the currently used container
+    VERTEX* newVertex = m_container->Allocate( 1 );
+
+    if( newVertex == NULL )
+    {
+        if( show_err )
+        {
+            DisplayError( NULL, wxT( "VERTEX_MANAGER::Vertex: Vertex allocation error" ) );
+            show_err = false;
+        }
+
+        return;
+    }
+
+    putVertex( *newVertex, aX, aY, aZ );
+}
+
+
+void VERTEX_MANAGER::Vertices( const VERTEX aVertices[], unsigned int aSize ) const
+{
+    // flag to avoid hanging by calling DisplayError too many times:
+    static bool show_err = true;
+
+    // Obtain pointer to the vertex in currently used container
+    VERTEX* newVertex = m_container->Allocate( aSize );
+
+    if( newVertex == NULL )
+    {
+        if( show_err )
+        {
+            DisplayError( NULL, wxT( "VERTEX_MANAGER::Vertices: Vertex allocation error" ) );
+            show_err = false;
+        }
+
+        return;
+    }
+
+    // Put vertices in already allocated memory chunk
+    for( unsigned int i = 0; i < aSize; ++i )
+    {
+        putVertex( newVertex[i], aVertices[i].x, aVertices[i].y, aVertices[i].z );
+    }
+}
+
+
+void VERTEX_MANAGER::SetItem( VERTEX_ITEM& aItem ) const
+{
+    m_container->SetItem( &aItem );
+}
+
+
+void VERTEX_MANAGER::FinishItem() const
+{
+    m_container->FinishItem();
+}
+
+
+void VERTEX_MANAGER::FreeItem( VERTEX_ITEM& aItem ) const
+{
+    m_container->Delete( &aItem );
+}
+
+
+void VERTEX_MANAGER::ChangeItemColor( const VERTEX_ITEM& aItem, const COLOR4D& aColor ) const
+{
+    unsigned int size   = aItem.GetSize();
+    unsigned int offset = aItem.GetOffset();
+
+    VERTEX* vertex = m_container->GetVertices( offset );
+
+    for( unsigned int i = 0; i < size; ++i )
+    {
+        vertex->r = aColor.r * 255.0;
+        vertex->g = aColor.g * 255.0;
+        vertex->b = aColor.b * 255.0;
+        vertex->a = aColor.a * 255.0;
+        vertex++;
+    }
+
+    m_container->SetDirty();
+}
+
+
+void VERTEX_MANAGER::ChangeItemDepth( const VERTEX_ITEM& aItem, GLfloat aDepth ) const
+{
+    unsigned int size   = aItem.GetSize();
+    unsigned int offset = aItem.GetOffset();
+
+    VERTEX* vertex = m_container->GetVertices( offset );
+
+    for( unsigned int i = 0; i < size; ++i )
+    {
+        vertex->z = aDepth;
+        vertex++;
+    }
+
+    m_container->SetDirty();
+}
+
+
+VERTEX* VERTEX_MANAGER::GetVertices( const VERTEX_ITEM& aItem ) const
+{
+    if( aItem.GetSize() == 0 )
+        return NULL; // The item is not stored in the container
+
+    return m_container->GetVertices( aItem.GetOffset() );
+}
+
+
+void VERTEX_MANAGER::SetShader( SHADER& aShader ) const
+{
+    m_gpu->SetShader( aShader );
+}
+
+
+void VERTEX_MANAGER::Clear() const
+{
+    m_container->Clear();
+}
+
+
+void VERTEX_MANAGER::BeginDrawing() const
+{
+    m_gpu->BeginDrawing();
+}
+
+
+void VERTEX_MANAGER::DrawItem( const VERTEX_ITEM& aItem ) const
+{
+    int size = aItem.GetSize();
+
+    if( size > 0 )
+    {
+        int offset = aItem.GetOffset();
+        m_gpu->DrawIndices( offset, size );
+    }
+}
+
+
+void VERTEX_MANAGER::EndDrawing() const
+{
+    m_gpu->EndDrawing();
+}
+
+
+void VERTEX_MANAGER::putVertex( VERTEX& aTarget, GLfloat aX, GLfloat aY, GLfloat aZ ) const
+{
+    // Modify the vertex according to the currently used transformations
+    if( m_noTransform )
+    {
+        // Simply copy coordinates, when the transform matrix is the identity matrix
+        aTarget.x = aX;
+        aTarget.y = aY;
+        aTarget.z = aZ;
+    }
+    else
+    {
+        // Apply transformations
+        glm::vec4 transVertex( aX, aY, aZ, 1.0f );
+        transVertex = m_transform * transVertex;
+
+        aTarget.x = transVertex.x;
+        aTarget.y = transVertex.y;
+        aTarget.z = transVertex.z;
+    }
+
+    // Apply currently used color
+    aTarget.r = m_color[0];
+    aTarget.g = m_color[1];
+    aTarget.b = m_color[2];
+    aTarget.a = m_color[3];
+
+    // Apply currently used shader
+    for( unsigned int j = 0; j < ShaderStride; ++j )
+    {
+        aTarget.shader[j] = m_shader[j];
+    }
+}
diff --git a/common/gal/stroke_font.cpp b/common/gal/stroke_font.cpp
new file mode 100644
index 0000000..636f75e
--- /dev/null
+++ b/common/gal/stroke_font.cpp
@@ -0,0 +1,403 @@
+/*
+ * This program source code file is part of KICAD, a free EDA CAD application.
+ *
+ * Copyright (C) 2012 Torsten Hueter, torstenhtr <at> gmx.de
+ * Copyright (C) 2012 Kicad Developers, see change_log.txt for contributors.
+ * Copyright (C) 2013 CERN
+ * @author Maciej Suminski <maciej.suminski@cern.ch>
+ *
+ * Stroke font class
+ *
+ * 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 <gal/stroke_font.h>
+#include <gal/graphics_abstraction_layer.h>
+#include <wx/string.h>
+
+using namespace KIGFX;
+
+const double STROKE_FONT::INTERLINE_PITCH_RATIO = 1.5;
+const double STROKE_FONT::OVERBAR_HEIGHT = 1.22;
+const double STROKE_FONT::BOLD_FACTOR = 1.3;
+const double STROKE_FONT::STROKE_FONT_SCALE = 1.0 / 21.0;
+const double STROKE_FONT::ITALIC_TILT = 1.0 / 8;
+
+STROKE_FONT::STROKE_FONT( GAL* aGal ) :
+    m_gal( aGal ),
+    m_bold( false ),
+    m_italic( false ),
+    m_mirrored( false ),
+    m_overbar( false )
+{
+    // Default values
+    m_glyphSize = VECTOR2D( 10.0, 10.0 );
+    m_verticalJustify   = GR_TEXT_VJUSTIFY_BOTTOM;
+    m_horizontalJustify = GR_TEXT_HJUSTIFY_LEFT;
+}
+
+
+bool STROKE_FONT::LoadNewStrokeFont( const char* const aNewStrokeFont[], int aNewStrokeFontSize )
+{
+    m_glyphs.clear();
+    m_glyphBoundingBoxes.clear();
+    m_glyphs.resize( aNewStrokeFontSize );
+    m_glyphBoundingBoxes.resize( aNewStrokeFontSize );
+
+    for( int j = 0; j < aNewStrokeFontSize; j++ )
+    {
+        GLYPH    glyph;
+        double   glyphStartX = 0.0;
+        double   glyphEndX = 0.0;
+        VECTOR2D glyphBoundingX;
+
+        std::deque<VECTOR2D> pointList;
+
+        int i = 0;
+
+        while( aNewStrokeFont[j][i] )
+        {
+            VECTOR2D    point( 0.0, 0.0 );
+            char        coordinate[2] = { 0, };
+
+            for( int k = 0; k < 2; k++ )
+            {
+                coordinate[k] = aNewStrokeFont[j][i + k];
+            }
+
+            if( i < 2 )
+            {
+                // The first two values contain the width of the char
+                glyphStartX     = ( coordinate[0] - 'R' ) * STROKE_FONT_SCALE;
+                glyphEndX       = ( coordinate[1] - 'R' ) * STROKE_FONT_SCALE;
+                glyphBoundingX  = VECTOR2D( 0, glyphEndX - glyphStartX );
+            }
+            else if( ( coordinate[0] == ' ' ) && ( coordinate[1] == 'R' ) )
+            {
+                // Raise pen
+                if( pointList.size() > 0 )
+                    glyph.push_back( pointList );
+
+                pointList.clear();
+            }
+            else
+            {
+                // Every coordinate description of the Hershey format has an offset,
+                // it has to be subtracted
+                point.x = (double) ( coordinate[0] - 'R' ) * STROKE_FONT_SCALE - glyphStartX;
+				// -10 is here to keep GAL rendering consistent with the legacy gfx stuff
+                point.y = (double) ( coordinate[1] - 'R' - 10) * STROKE_FONT_SCALE;
+                pointList.push_back( point );
+            }
+
+            i += 2;
+        }
+
+        if( pointList.size() > 0 )
+            glyph.push_back( pointList );
+
+        m_glyphs[j] = glyph;
+
+        // Compute the bounding box of the glyph
+        m_glyphBoundingBoxes[j] = computeBoundingBox( glyph, glyphBoundingX );
+    }
+
+    return true;
+}
+
+
+int STROKE_FONT::getInterline() const
+{
+    return KiROUND( m_glyphSize.y * INTERLINE_PITCH_RATIO ) + m_gal->GetLineWidth();
+}
+
+
+BOX2D STROKE_FONT::computeBoundingBox( const GLYPH& aGLYPH, const VECTOR2D& aGLYPHBoundingX ) const
+{
+    BOX2D boundingBox;
+
+    std::deque<VECTOR2D> boundingPoints;
+
+    boundingPoints.push_back( VECTOR2D( aGLYPHBoundingX.x, 0 ) );
+    boundingPoints.push_back( VECTOR2D( aGLYPHBoundingX.y, 0 ) );
+
+    for( GLYPH::const_iterator pointListIt = aGLYPH.begin(); pointListIt != aGLYPH.end(); ++pointListIt )
+    {
+        for( std::deque<VECTOR2D>::const_iterator pointIt = pointListIt->begin();
+                pointIt != pointListIt->end(); ++pointIt )
+        {
+            boundingPoints.push_back( VECTOR2D( aGLYPHBoundingX.x, pointIt->y ) );
+        }
+    }
+
+    boundingBox.Compute( boundingPoints );
+
+    return boundingBox;
+}
+
+
+void STROKE_FONT::Draw( const UTF8& aText, const VECTOR2D& aPosition, double aRotationAngle )
+{
+    if( aText.empty() )
+        return;
+
+    // Context needs to be saved before any transformations
+    m_gal->Save();
+
+    m_gal->Translate( aPosition );
+    m_gal->Rotate( -aRotationAngle );
+
+    // Single line height
+    int lineHeight = getInterline( );
+    int lineCount = linesCount( aText );
+
+    // align the 1st line of text
+    switch( m_verticalJustify )
+    {
+    case GR_TEXT_VJUSTIFY_TOP:
+        m_gal->Translate( VECTOR2D( 0, m_glyphSize.y ) );
+        break;
+
+    case GR_TEXT_VJUSTIFY_CENTER:
+        m_gal->Translate( VECTOR2D( 0, m_glyphSize.y / 2.0 ) );
+        break;
+
+    case GR_TEXT_VJUSTIFY_BOTTOM:
+        break;
+
+    default:
+        break;
+    }
+
+    if( lineCount > 1 )
+    {
+        switch( m_verticalJustify )
+        {
+        case GR_TEXT_VJUSTIFY_TOP:
+            break;
+
+        case GR_TEXT_VJUSTIFY_CENTER:
+            m_gal->Translate( VECTOR2D(0, -( lineCount - 1 ) * lineHeight / 2) );
+            break;
+
+        case GR_TEXT_VJUSTIFY_BOTTOM:
+            m_gal->Translate( VECTOR2D(0, -( lineCount - 1 ) * lineHeight ) );
+            break;
+        }
+    }
+
+    m_gal->SetIsStroke( true );
+    m_gal->SetIsFill( false );
+
+    if( m_bold )
+        m_gal->SetLineWidth( m_gal->GetLineWidth() * BOLD_FACTOR );
+
+    // Split multiline strings into separate ones and draw them line by line
+    size_t  begin = 0;
+    size_t  newlinePos = aText.find( '\n' );
+
+    while( newlinePos != aText.npos )
+    {
+        size_t length = newlinePos - begin;
+
+        drawSingleLineText( aText.substr( begin, length ) );
+        m_gal->Translate( VECTOR2D( 0.0, lineHeight ) );
+
+        begin = newlinePos + 1;
+        newlinePos = aText.find( '\n', begin );
+    }
+
+    // Draw the last (or the only one) line
+    if( !aText.empty() )
+        drawSingleLineText( aText.substr( begin ) );
+
+    m_gal->Restore();
+}
+
+
+void STROKE_FONT::drawSingleLineText( const UTF8& aText )
+{
+    // By default the overbar is turned off
+    m_overbar = false;
+
+    double      xOffset;
+    VECTOR2D    glyphSize( m_glyphSize );
+    double      overbar_italic_comp = 0.0;
+
+    // Compute the text size
+    VECTOR2D textSize = computeTextSize( aText );
+
+    m_gal->Save();
+
+    // Adjust the text position to the given alignment
+    switch( m_horizontalJustify )
+    {
+    case GR_TEXT_HJUSTIFY_CENTER:
+        m_gal->Translate( VECTOR2D( -textSize.x / 2.0, 0 ) );
+        break;
+
+    case GR_TEXT_HJUSTIFY_RIGHT:
+        if( !m_mirrored )
+            m_gal->Translate( VECTOR2D( -textSize.x, 0 ) );
+        break;
+
+    case GR_TEXT_HJUSTIFY_LEFT:
+        if( m_mirrored )
+            m_gal->Translate( VECTOR2D( -textSize.x, 0 ) );
+        break;
+
+    default:
+        break;
+    }
+
+    if( m_mirrored )
+    {
+        // In case of mirrored text invert the X scale of points and their X direction
+        // (m_glyphSize.x) and start drawing from the position where text normally should end
+        // (textSize.x)
+        xOffset = textSize.x;
+        glyphSize.x = -m_glyphSize.x;
+    }
+    else
+    {
+        xOffset = 0.0;
+    }
+
+    // The overbar is indented inward at the beginning of an italicized section, but
+    // must not be indented on subsequent letters to ensure that the bar segments
+    // overlap.
+    bool last_had_overbar = false;
+
+    for( UTF8::uni_iter chIt = aText.ubegin(), end = aText.uend(); chIt < end; ++chIt )
+    {
+        // Toggle overbar
+        if( *chIt == '~' )
+        {
+            if( ++chIt >= end )
+                break;
+
+            if( *chIt != '~' )      // It was a single tilda, it toggles overbar
+                m_overbar = !m_overbar;
+
+            // If it is a double tilda, just process the second one
+        }
+
+        int dd = *chIt - ' ';
+
+        if( dd >= (int) m_glyphBoundingBoxes.size() || dd < 0 )
+            dd = '?' - ' ';
+
+        GLYPH& glyph = m_glyphs[dd];
+        BOX2D& bbox  = m_glyphBoundingBoxes[dd];
+
+        if( m_overbar && m_italic )
+        {
+            if( m_mirrored )
+            {
+                overbar_italic_comp = (-m_glyphSize.y * OVERBAR_HEIGHT) / ITALIC_TILT;
+            }
+            else
+            {
+                overbar_italic_comp = (m_glyphSize.y * OVERBAR_HEIGHT) / ITALIC_TILT;
+            }
+        }
+
+        if( m_overbar )
+        {
+            double overbar_start_x = xOffset;
+            double overbar_start_y = -m_glyphSize.y * OVERBAR_HEIGHT;
+            double overbar_end_x = xOffset + glyphSize.x * bbox.GetEnd().x;
+            double overbar_end_y = -m_glyphSize.y * OVERBAR_HEIGHT;
+
+            if( !last_had_overbar )
+            {
+                overbar_start_x += overbar_italic_comp;
+                last_had_overbar = true;
+            }
+
+            VECTOR2D startOverbar( overbar_start_x, overbar_start_y );
+            VECTOR2D endOverbar( overbar_end_x, overbar_end_y );
+
+            m_gal->DrawLine( startOverbar, endOverbar );
+        }
+        else
+        {
+            last_had_overbar = false;
+        }
+
+        for( GLYPH::iterator pointListIt = glyph.begin(); pointListIt != glyph.end();
+             ++pointListIt )
+        {
+            std::deque<VECTOR2D> pointListScaled;
+
+            for( std::deque<VECTOR2D>::iterator pointIt = pointListIt->begin();
+                 pointIt != pointListIt->end(); ++pointIt )
+            {
+                VECTOR2D pointPos( pointIt->x * glyphSize.x + xOffset, pointIt->y * glyphSize.y );
+
+                if( m_italic )
+                {
+                    // FIXME should be done other way - referring to the lowest Y value of point
+                    // because now italic fonts are translated a bit
+                    if( m_mirrored )
+                        pointPos.x += pointPos.y * 0.1;
+                    else
+                        pointPos.x -= pointPos.y * 0.1;
+                }
+
+                pointListScaled.push_back( pointPos );
+            }
+
+            m_gal->DrawPolyline( pointListScaled );
+        }
+
+        xOffset += glyphSize.x * bbox.GetEnd().x;
+    }
+
+    m_gal->Restore();
+}
+
+
+VECTOR2D STROKE_FONT::computeTextSize( const UTF8& aText ) const
+{
+    VECTOR2D result = VECTOR2D( 0.0, m_glyphSize.y );
+
+    for( UTF8::uni_iter it = aText.ubegin(), end = aText.uend(); it < end; ++it )
+    {
+        wxASSERT_MSG( *it != '\n',
+                      wxT( "This function is intended to work with single line strings" ) );
+
+        // If it is double tilda, then it is displayed as a single tilda
+        // If it is single tilda, then it is toggling overbar, so we need to skip it
+        if( *it == '~' )
+        {
+            if( ++it >= end )
+                break;
+        }
+
+        // Index in the bounding boxes table
+        int dd = *it - ' ';
+
+        if( dd >= (int) m_glyphBoundingBoxes.size() || dd < 0 )
+            dd = '?' - ' ';
+
+        result.x += m_glyphSize.x * m_glyphBoundingBoxes[dd].GetEnd().x;
+    }
+
+    return result;
+}