drc_test_provider_connection_width.cpp

Go to the documentation of this file.

/*
 * This program source code file is part of KiCad, a free EDA CAD application.
 *
 * Copyright (C) 2022 KiCad Developers.
 *
 * 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 3
 * 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/gpl-3.0.html
 * or you may search the http://www.gnu.org website for the version 3 license,
 * or you may write to the Free Software Foundation, Inc.,
 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
 */
 
#include <algorithm>
#include <atomic>
#include <deque>
#include <optional>
#include <utility>
 
#include <wx/debug.h>
 
#include <board.h>
#include <board_connected_item.h>
#include <board_design_settings.h>
#include <drc/drc_rule.h>
#include <drc/drc_item.h>
#include <drc/drc_test_provider.h>
#include <drc/drc_rtree.h>
#include <drc/drc_rule_condition.h>
#include <footprint.h>
#include <geometry/seg.h>
#include <geometry/shape_poly_set.h>
#include <math/box2.h>
#include <math/vector2d.h>
#include <pcb_shape.h>
#include <progress_reporter.h>
#include <core/thread_pool.h>
#include <pcb_track.h>
#include <pad.h>
#include <zone.h>
 
/*
 Checks for copper connections that are less than the specified minimum width
 
 Errors generated:
 - DRCE_CONNECTION_WIDTH
*/
 
struct NETCODE_LAYER_CACHE_KEY
{
 int Netcode;
 PCB_LAYER_ID Layer;
 
 bool operator==(const NETCODE_LAYER_CACHE_KEY& other) const
 {
 return Netcode == other.Netcode && Layer == other.Layer;
 }
};
 
 
namespace std
{
 template <>
 struct hash<NETCODE_LAYER_CACHE_KEY>
 {
 std::size_t operator()( const NETCODE_LAYER_CACHE_KEY& k ) const
 {
 constexpr std::size_t prime = 19937;
 
 return hash<int>()( k.Netcode ) ^ ( hash<int>()( k.Layer ) * prime );
 }
 };
}
 
 
class DRC_TEST_PROVIDER_CONNECTION_WIDTH : public DRC_TEST_PROVIDER
{
public:
 DRC_TEST_PROVIDER_CONNECTION_WIDTH()
 {
 }
 
 virtual ~DRC_TEST_PROVIDER_CONNECTION_WIDTH()
 {
 }
 
 virtual bool Run() override;
 
 virtual const wxString GetName() const override
 {
 return wxT( "copper width" );
 };
 
 virtual const wxString GetDescription() const override
 {
 return wxT( "Checks copper nets for connections less than a specified minimum" );
 }
 
private:
 wxString layerDesc( PCB_LAYER_ID aLayer );
};
 
 
class POLYGON_TEST
{
public:
 POLYGON_TEST( int aLimit ) :
 m_limit( aLimit )
 {
 };
 
 bool FindPairs( const SHAPE_LINE_CHAIN& aPoly )
 {
 m_hits.clear();
 m_vertices.clear();
 m_bbox = aPoly.BBox();
 
 createList( aPoly );
 
 m_vertices.front().updateList();
 
 Vertex* p = m_vertices.front().next;
 std::set<Vertex*> all_hits;
 
 while( p != &m_vertices.front() )
 {
 Vertex* match = nullptr;
 
 // Only run the expensive search if we don't already have a match for the point
 if( ( all_hits.empty() || all_hits.count( p ) == 0 ) && ( match = getKink( p ) ) != nullptr )
 {
 if( !all_hits.count( match ) && m_hits.emplace( p->i, match->i ).second )
 {
 all_hits.emplace( p );
 all_hits.emplace( match );
 all_hits.emplace( p->next );
 all_hits.emplace( p->prev );
 all_hits.emplace( match->next );
 all_hits.emplace( match->prev );
 }
 }
 
 p = p->next;
 }
 
 return !m_hits.empty();
 }
 
 std::set<std::pair<int, int>>& GetVertices()
 {
 return m_hits;
 }
 
 
private:
 struct Vertex
 {
 Vertex( int aIndex, double aX, double aY, POLYGON_TEST* aParent ) :
 i( aIndex ),
 x( aX ),
 y( aY ),
 parent( aParent )
 {
 }
 
 Vertex& operator=( const Vertex& ) = delete;
 Vertex& operator=( Vertex&& ) = delete;
 
 bool operator==( const Vertex& rhs ) const
 {
 return this->x == rhs.x && this->y == rhs.y;
 }
 bool operator!=( const Vertex& rhs ) const { return !( *this == rhs ); }
 
 void remove()
 {
 next->prev = prev;
 prev->next = next;
 
 if( prevZ )
 prevZ->nextZ = nextZ;
 
 if( nextZ )
 nextZ->prevZ = prevZ;
 
 next = nullptr;
 prev = nullptr;
 nextZ = nullptr;
 prevZ = nullptr;
 }
 
 void updateOrder()
 {
 if( !z )
 z = parent->zOrder( x, y );
 }
 
 void updateList()
 {
 Vertex* p = next;
 
 while( p != this )
 {
 if( *p == *p->next )
 {
 p = p->prev;
 p->next->remove();
 
 if( p == p->next )
 break;
 }
 
 p->updateOrder();
 p = p->next;
 };
 
 updateOrder();
 zSort();
 }
 
 void zSort()
 {
 std::deque<Vertex*> queue;
 
 queue.push_back( this );
 
 for( Vertex* p = next; p && p != this; p = p->next )
 queue.push_back( p );
 
 std::sort( queue.begin(), queue.end(), []( const Vertex* a, const Vertex* b )
 {
 if( a->z != b->z )
 return a->z < b->z;
 
 if( a->x != b->x )
 return a->x < b->x;
 
 if( a->y != b->y )
 return a->y < b->y;
 
 return a->i < b->i;
 } );
 
 Vertex* prev_elem = nullptr;
 
 for( Vertex* elem : queue )
 {
 if( prev_elem )
 prev_elem->nextZ = elem;
 
 elem->prevZ = prev_elem;
 prev_elem = elem;
 }
 
 prev_elem->nextZ = nullptr;
 }
 
 const int i;
 const double x;
 const double y;
 POLYGON_TEST* parent;
 
 // previous and next vertices nodes in a polygon ring
 Vertex* prev = nullptr;
 Vertex* next = nullptr;
 
 // z-order curve value
 int32_t z = 0;
 
 // previous and next nodes in z-order
 Vertex* prevZ = nullptr;
 Vertex* nextZ = nullptr;
 };
 
 int32_t zOrder( const double aX, const double aY ) const
 {
 int32_t x = static_cast<int32_t>( 32767.0 * ( aX - m_bbox.GetX() ) / m_bbox.GetWidth() );
 int32_t y = static_cast<int32_t>( 32767.0 * ( aY - m_bbox.GetY() ) / m_bbox.GetHeight() );
 
 x = ( x | ( x << 8 ) ) & 0x00FF00FF;
 x = ( x | ( x << 4 ) ) & 0x0F0F0F0F;
 x = ( x | ( x << 2 ) ) & 0x33333333;
 x = ( x | ( x << 1 ) ) & 0x55555555;
 
 y = ( y | ( y << 8 ) ) & 0x00FF00FF;
 y = ( y | ( y << 4 ) ) & 0x0F0F0F0F;
 y = ( y | ( y << 2 ) ) & 0x33333333;
 y = ( y | ( y << 1 ) ) & 0x55555555;
 
 return x | ( y << 1 );
 }
 
 constexpr bool same_point( const Vertex* aA, const Vertex* aB ) const
 {
 return aA && aB && aA->x == aB->x && aA->y == aB->y;
 }
 
 Vertex* getNextOutlineVertex( const Vertex* aPt ) const
 {
 Vertex* nz = aPt->nextZ;
 Vertex* pz = aPt->prevZ;
 
 // If we hit a fracture point, we want to continue around the
 // edge we are working on and not switch to the pair edge
 // However, this will depend on which direction the initial
 // fracture hit is. If we find that we skip directly to
 // a new fracture point, then we know that we are proceeding
 // in the wrong direction from the fracture and should
 // fall through to the next point
 if( same_point( aPt, nz )
 && aPt->y == aPt->next->y )
 {
 return nz->next;
 }
 
 if( same_point( aPt, pz )
 && aPt->y == aPt->next->y )
 {
 return pz->next;
 }
 
 return aPt->next;
 }
 
 Vertex* getPrevOutlineVertex( const Vertex* aPt ) const
 {
 Vertex* nz = aPt->nextZ;
 Vertex* pz = aPt->prevZ;
 
 // If we hit a fracture point, we want to continue around the
 // edge we are working on and not switch to the pair edge
 // However, this will depend on which direction the initial
 // fracture hit is. If we find that we skip directly to
 // a new fracture point, then we know that we are proceeding
 // in the wrong direction from the fracture and should
 // fall through to the next point
 if( same_point( aPt, nz )
 && aPt->y == aPt->prev->y)
 {
 return nz->prev;
 }
 
 if( same_point( aPt, pz )
 && aPt->y == aPt->prev->y )
 {
 return pz->prev;
 }
 
 return aPt->prev;
 
 }
 
 bool isSubstantial( const Vertex* aA, const Vertex* aB ) const
 {
 bool x_change = false;
 bool y_change = false;
 
 // This is a failsafe in case of invalid lists. Never check
 // more than the total number of points in m_vertices
 size_t checked = 0;
 size_t total_pts = m_vertices.size();
 
 const Vertex* p0 = aA;
 const Vertex* p = getNextOutlineVertex( p0 );
 
 while( !same_point( p, aB ) && checked < total_pts && !( x_change && y_change ) )
 {
 double diff_x = std::abs( p->x - p0->x );
 double diff_y = std::abs( p->y - p0->y );
 
 // Floating point zeros can have a negative sign, so we need to
 // ensure that only substantive diversions count for a direction
 // change
 if( diff_x > m_limit )
 x_change = true;
 
 if( diff_y > m_limit )
 y_change = true;
 
 p = getNextOutlineVertex( p );
 
 ++checked;
 }
 
 wxCHECK_MSG( checked < total_pts, false, wxT( "Invalid polygon detected. Missing points to check" ) );
 
 if( !x_change || !y_change )
 return false;
 
 p = getPrevOutlineVertex( p0 );
 
 x_change = false;
 y_change = false;
 checked = 0;
 
 while( !same_point( p, aB ) && checked < total_pts && !( x_change && y_change ) )
 {
 double diff_x = std::abs( p->x - p0->x );
 double diff_y = std::abs( p->y - p0->y );
 
 // Floating point zeros can have a negative sign, so we need to
 // ensure that only substantive diversions count for a direction
 // change
 if( diff_x > m_limit )
 x_change = true;
 
 if( diff_y > m_limit )
 y_change = true;
 
 p = getPrevOutlineVertex( p );
 
 ++checked;
 }
 
 wxCHECK_MSG( checked < total_pts, false, wxT( "Invalid polygon detected. Missing points to check" ) );
 
 return ( x_change && y_change );
 }
 
 Vertex* createList( const SHAPE_LINE_CHAIN& points )
 {
 Vertex* tail = nullptr;
 double sum = 0.0;
 
 // Check for winding order
 for( int i = 0; i < points.PointCount(); i++ )
 {
 VECTOR2D p1 = points.CPoint( i );
 VECTOR2D p2 = points.CPoint( i + 1 );
 
 sum += ( ( p2.x - p1.x ) * ( p2.y + p1.y ) );
 }
 
 if( sum > 0.0 )
 {
 for( int i = points.PointCount() - 1; i >= 0; i--)
 tail = insertVertex( i, points.CPoint( i ), tail );
 }
 else
 {
 for( int i = 0; i < points.PointCount(); i++ )
 tail = insertVertex( i, points.CPoint( i ), tail );
 }
 
 if( tail && ( *tail == *tail->next ) )
 {
 tail->next->remove();
 }
 
 return tail;
 }
 
 Vertex* getKink( Vertex* aPt ) const
 {
 // The point needs to be at a concave surface
 if( locallyInside( aPt->prev, aPt->next ) )
 return nullptr;
 
 // z-order range for the current point ± limit bounding box
 const int32_t maxZ = zOrder( aPt->x + m_limit, aPt->y + m_limit );
 const int32_t minZ = zOrder( aPt->x - m_limit, aPt->y - m_limit );
 
 // Subtract 1 to account for rounding inaccuracies in SquaredEuclideanNorm()
 // below. We would usually test for rounding in the final value but since we
 // are working in squared integers here, we allow the 1nm slop rather than
 // force a separate calculation
 const SEG::ecoord limit2 = SEG::Square( m_limit - 1 );
 
 // first look for points in increasing z-order
 Vertex* p = aPt->nextZ;
 SEG::ecoord min_dist = std::numeric_limits<SEG::ecoord>::max();
 Vertex* retval = nullptr;
 
 while( p && p->z <= maxZ )
 {
 int delta_i = std::abs( p->i - aPt->i );
 VECTOR2D diff( p->x - aPt->x, p->y - aPt->y );
 SEG::ecoord dist2 = diff.SquaredEuclideanNorm();
 
 if( delta_i > 1 && dist2 < limit2 && dist2 < min_dist && dist2 > 0.0
 && locallyInside( p, aPt ) && isSubstantial( p, aPt ) && isSubstantial( aPt, p ) )
 {
 min_dist = dist2;
 retval = p;
 }
 
 p = p->nextZ;
 }
 
 p = aPt->prevZ;
 
 while( p && p->z >= minZ )
 {
 int delta_i = std::abs( p->i - aPt->i );
 VECTOR2D diff( p->x - aPt->x, p->y - aPt->y );
 SEG::ecoord dist2 = diff.SquaredEuclideanNorm();
 
 if( delta_i > 1 && dist2 < limit2 && dist2 < min_dist && dist2 > 0.0
 && locallyInside( p, aPt ) && isSubstantial( p, aPt ) && isSubstantial( aPt, p ) )
 {
 min_dist = dist2;
 retval = p;
 }
 
 p = p->prevZ;
 }
 return retval;
 }
 
 
 double area( const Vertex* p, const Vertex* q, const Vertex* r ) const
 {
 return ( q->y - p->y ) * ( r->x - q->x ) - ( q->x - p->x ) * ( r->y - q->y );
 }
 
 
 bool locallyInside( const Vertex* a, const Vertex* b ) const
 {
 const Vertex* an = getNextOutlineVertex( a );
 const Vertex* ap = getPrevOutlineVertex( a );
 
 if( area( ap, a, an ) < 0 )
 return area( a, b, an ) >= 0 && area( a, ap, b ) >= 0;
 else
 return area( a, b, ap ) < 0 || area( a, an, b ) < 0;
 }
 
 Vertex* insertVertex( int aIndex, const VECTOR2I& pt, Vertex* last )
 {
 m_vertices.emplace_back( aIndex, pt.x, pt.y, this );
 
 Vertex* p = &m_vertices.back();
 
 if( !last )
 {
 p->prev = p;
 p->next = p;
 }
 else
 {
 p->next = last->next;
 p->prev = last;
 last->next->prev = p;
 last->next = p;
 }
 return p;
 }
 
private:
 int m_limit;
 BOX2I m_bbox;
 std::deque<Vertex> m_vertices;
 std::set<std::pair<int, int>> m_hits;
};
 
 
wxString DRC_TEST_PROVIDER_CONNECTION_WIDTH::layerDesc( PCB_LAYER_ID aLayer )
{
 return wxString::Format( wxT( "(%s)" ), m_drcEngine->GetBoard()->GetLayerName( aLayer ) );
}
 
 
bool DRC_TEST_PROVIDER_CONNECTION_WIDTH::Run()
{
 if( m_drcEngine->IsErrorLimitExceeded( DRCE_CONNECTION_WIDTH ) )
 return true; // Continue with other tests
 
 if( !reportPhase( _( "Checking nets for minimum connection width..." ) ) )
 return false; // DRC cancelled
 
 LSET copperLayerSet = m_drcEngine->GetBoard()->GetEnabledLayers() & LSET::AllCuMask();
 LSEQ copperLayers = copperLayerSet.Seq();
 BOARD* board = m_drcEngine->GetBoard();
 
 /*
 * Build a set of distinct minWidths specified by various DRC rules. We'll run a test for
 * each distinct minWidth, and then decide if any copper which failed that minWidth actually
 * was required to abide by it or not.
 */
 std::set<int> distinctMinWidths
 = m_drcEngine->QueryDistinctConstraints( CONNECTION_WIDTH_CONSTRAINT );
 
 if( m_drcEngine->IsCancelled() )
 return false; // DRC cancelled
 
 struct ITEMS_POLY
 {
 std::set<BOARD_ITEM*> Items;
 SHAPE_POLY_SET Poly;
 };
 
 std::unordered_map<NETCODE_LAYER_CACHE_KEY, ITEMS_POLY> dataset;
 std::atomic<size_t> done( 1 );
 
 auto calc_effort =
 [&]( const std::set<BOARD_ITEM*>& items, PCB_LAYER_ID aLayer ) -> size_t
 {
 size_t effort = 0;
 
 for( BOARD_ITEM* item : items )
 {
 if( item->Type() == PCB_ZONE_T )
 {
 ZONE* zone = static_cast<ZONE*>( item );
 effort += zone->GetFilledPolysList( aLayer )->FullPointCount();
 }
 else
 {
 effort += 4;
 }
 }
 
 return effort;
 };
 
 /*
 * For each net, on each layer, build a polygonSet which contains all the copper associated
 * with that net on that layer.
 */
 auto build_netlayer_polys =
 [&]( int aNetcode, const PCB_LAYER_ID aLayer ) -> size_t
 {
 if( m_drcEngine->IsCancelled() )
 return 0;
 
 ITEMS_POLY& itemsPoly = dataset[ { aNetcode, aLayer } ];
 
 for( BOARD_ITEM* item : itemsPoly.Items )
 {
 item->TransformShapeToPolygon( itemsPoly.Poly, aLayer, 0, ARC_HIGH_DEF,
 ERROR_OUTSIDE );
 }
 
 itemsPoly.Poly.Fracture( SHAPE_POLY_SET::PM_FAST );
 
 done.fetch_add( calc_effort( itemsPoly.Items, aLayer ) );
 
 return 1;
 };
 
 /*
 * Examine all necks in a given polygonSet which fail a given minWidth.
 */
 auto min_checker =
 [&]( const ITEMS_POLY& aItemsPoly, const PCB_LAYER_ID aLayer, int aMinWidth ) -> size_t
 {
 if( m_drcEngine->IsCancelled() )
 return 0;
 
 POLYGON_TEST test( aMinWidth );
 
 for( int ii = 0; ii < aItemsPoly.Poly.OutlineCount(); ++ii )
 {
 const SHAPE_LINE_CHAIN& chain = aItemsPoly.Poly.COutline( ii );
 
 test.FindPairs( chain );
 auto& ret = test.GetVertices();
 
 for( const std::pair<int, int>& pt : ret )
 {
 /*
 * We've found a neck that fails the given aMinWidth. We now need to know
 * if the objects the produced the copper at this location are required to
 * abide by said aMinWidth or not. (If so, we have a violation.)
 *
 * We find the contributingItems by hit-testing at the choke point (the
 * centre point of the neck), and then run the rules engine on those
 * contributingItems. If the reported constraint matches aMinWidth, then
 * we've got a violation.
 */
 SEG span( chain.CPoint( pt.first ), chain.CPoint( pt.second ) );
 VECTOR2I location = ( span.A + span.B ) / 2;
 int dist = ( span.A - span.B ).EuclideanNorm();
 
 std::vector<BOARD_ITEM*> contributingItems;
 
 for( auto* item : board->m_CopperItemRTreeCache->GetObjectsAt( location,
 aLayer,
 aMinWidth ) )
 {
 if( item->HitTest( location, aMinWidth ) )
 contributingItems.push_back( item );
 }
 
 for( auto& [ zone, rtree ] : board->m_CopperZoneRTreeCache )
 {
 if( !rtree.get() )
 continue;
 
 auto obj_list = rtree->GetObjectsAt( location, aLayer, aMinWidth );
 
 if( !obj_list.empty() && zone->HitTestFilledArea( aLayer, location, aMinWidth ) )
 {
 contributingItems.push_back( zone );
 }
 }
 
 if( !contributingItems.empty() )
 {
 BOARD_ITEM* item1 = contributingItems[0];
 BOARD_ITEM* item2 = contributingItems.size() > 1 ? contributingItems[1]
 : nullptr;
 DRC_CONSTRAINT c = m_drcEngine->EvalRules( CONNECTION_WIDTH_CONSTRAINT,
 item1, item2, aLayer );
 
 if( c.Value().Min() == aMinWidth )
 {
 auto drce = DRC_ITEM::Create( DRCE_CONNECTION_WIDTH );
 wxString msg;
 
 msg = formatMsg( _( "(%s minimum connection width %s; actual %s)" ),
 c.GetName(),
 aMinWidth,
 dist );
 
 msg += wxS( " " ) + layerDesc( aLayer );
 
 drce->SetErrorMessage( drce->GetErrorText() + wxS( " " ) + msg );
 drce->SetViolatingRule( c.GetParentRule() );
 
 for( BOARD_ITEM* item : contributingItems )
 drce->AddItem( item );
 
 reportViolation( drce, location, aLayer );
 }
 }
 }
 }
 
 done.fetch_add( calc_effort( aItemsPoly.Items, aLayer ) );
 
 return 1;
 };
 
 for( PCB_LAYER_ID layer : copperLayers )
 {
 for( ZONE* zone : board->m_DRCCopperZones )
 {
 if( !zone->GetIsRuleArea() && zone->IsOnLayer( layer ) )
 dataset[ { zone->GetNetCode(), layer } ].Items.emplace( zone );
 }
 
 for( PCB_TRACK* track : board->Tracks() )
 {
 if( PCB_VIA* via = dynamic_cast<PCB_VIA*>( track ) )
 {
 if( via->FlashLayer( static_cast<int>( layer ) ) )
 dataset[ { via->GetNetCode(), layer } ].Items.emplace( via );
 }
 else if( track->IsOnLayer( layer ) )
 {
 dataset[ { track->GetNetCode(), layer } ].Items.emplace( track );
 }
 }
 
 for( FOOTPRINT* fp : board->Footprints() )
 {
 for( PAD* pad : fp->Pads() )
 {
 if( pad->FlashLayer( static_cast<int>( layer ) ) )
 dataset[ { pad->GetNetCode(), layer } ].Items.emplace( pad );
 }
 
 // Footprint zones are also in the m_DRCCopperZones cache
 }
 }
 
 thread_pool& tp = GetKiCadThreadPool();
 std::vector<std::future<size_t>> returns;
 size_t total_effort = 0;
 
 for( const auto& [ netLayer, itemsPoly ] : dataset )
 total_effort += calc_effort( itemsPoly.Items, netLayer.Layer );
 
 total_effort += std::max( (size_t) 1, total_effort ) * distinctMinWidths.size();
 
 returns.reserve( dataset.size() );
 
 for( const auto& [ netLayer, itemsPoly ] : dataset )
 {
 returns.emplace_back( tp.submit( build_netlayer_polys, netLayer.Netcode, netLayer.Layer ) );
 }
 
 for( std::future<size_t>& ret : returns )
 {
 std::future_status status = ret.wait_for( std::chrono::milliseconds( 250 ) );
 
 while( status != std::future_status::ready )
 {
 m_drcEngine->ReportProgress( static_cast<double>( done ) / total_effort );
 status = ret.wait_for( std::chrono::milliseconds( 250 ) );
 }
 }
 
 returns.clear();
 returns.reserve( dataset.size() * distinctMinWidths.size() );
 
 for( const auto& [ netLayer, itemsPoly ] : dataset )
 {
 for( int minWidth : distinctMinWidths )
 returns.emplace_back( tp.submit( min_checker, itemsPoly, netLayer.Layer, minWidth ) );
 }
 
 for( std::future<size_t>& ret : returns )
 {
 std::future_status status = ret.wait_for( std::chrono::milliseconds( 250 ) );
 
 while( status != std::future_status::ready )
 {
 m_drcEngine->ReportProgress( static_cast<double>( done ) / total_effort );
 status = ret.wait_for( std::chrono::milliseconds( 250 ) );
 }
 }
 
 return true;
}
 
 
namespace detail
{
static DRC_REGISTER_TEST_PROVIDER<DRC_TEST_PROVIDER_CONNECTION_WIDTH> dummy;
}