object-snapper.h revision 75bbef79b0ac9fed0049660182ff6ef9e96d1a73
#ifndef SEEN_OBJECT_SNAPPER_H
#define SEEN_OBJECT_SNAPPER_H
/**
* \file object-snapper.h
* \brief Snapping things to objects.
*
* Authors:
* Carl Hetherington <inkscape@carlh.net>
* Diederik van Lierop <mail@diedenrezi.nl>
*
* Copyright (C) 2005 - 2008 Authors
*
* Released under GNU GPL, read the file 'COPYING' for more information
*/
#include "snapper.h"
#include "sp-path.h"
#include "splivarot.h"
#include "snap-candidate.h"
struct SPNamedView;
struct SPItem;
struct SPObject;
namespace Inkscape
{
class ObjectSnapper : public Snapper
{
public:
ObjectSnapper(SnapManager *sm, Geom::Coord const d);
~ObjectSnapper();
enum DimensionToSnap {
GUIDE_TRANSL_SNAP_X, // For snapping a vertical guide (normal in the X-direction) to objects,
GUIDE_TRANSL_SNAP_Y, // For snapping a horizontal guide (normal in the Y-direction) to objects
ANGLED_GUIDE_TRANSL_SNAP, // For snapping an angled guide, while translating it accross the desktop
TRANSL_SNAP_XY}; // All other cases; for snapping to objects, other than guides
void guideFreeSnap(SnappedConstraints &sc,
Geom::Point const &p,
Geom::Point const &guide_normal) const;
void guideConstrainedSnap(SnappedConstraints &sc,
Geom::Point const &p,
Geom::Point const &guide_normal,
SnapConstraint const &c) const;
bool ThisSnapperMightSnap() const;
Geom::Coord getSnapperTolerance() const; //returns the tolerance of the snapper in screen pixels (i.e. independent of zoom)
bool getSnapperAlwaysSnap() const; //if true, then the snapper will always snap, regardless of its tolerance
void freeSnap(SnappedConstraints &sc,
Inkscape::SnapCandidatePoint const &p,
Geom::OptRect const &bbox_to_snap,
std::vector<SPItem const *> const *it,
std::vector<SnapCandidatePoint> *unselected_nodes) const;
void constrainedSnap(SnappedConstraints &sc,
Inkscape::SnapCandidatePoint const &p,
Geom::OptRect const &bbox_to_snap,
SnapConstraint const &c,
std::vector<SPItem const *> const *it) const;
private:
//store some lists of candidates, points and paths, so we don't have to rebuild them for each point we want to snap
std::vector<SnapCandidateItem> *_candidates;
std::vector<SnapCandidatePoint> *_points_to_snap_to;
std::vector<SnapCandidatePath > *_paths_to_snap_to;
void _findCandidates(SPObject* parent,
std::vector<SPItem const *> const *it,
bool const &first_point,
Geom::Rect const &bbox_to_snap,
DimensionToSnap snap_dim,
bool const _clip_or_mask,
Geom::Matrix const additional_affine) const;
void _snapNodes(SnappedConstraints &sc,
Inkscape::SnapCandidatePoint const &p,
std::vector<SnapCandidatePoint> *unselected_nodes) const; // in desktop coordinates
void _snapTranslatingGuide(SnappedConstraints &sc,
Geom::Point const &p,
Geom::Point const &guide_normal) const;
void _collectNodes(Inkscape::SnapSourceType const &t,
bool const &first_point) const;
void _snapPaths(SnappedConstraints &sc,
Inkscape::SnapCandidatePoint const &p, // in desktop coordinates
std::vector<Inkscape::SnapCandidatePoint> *unselected_nodes, // in desktop coordinates
SPPath const *selected_path) const;
void _snapPathsConstrained(SnappedConstraints &sc,
Inkscape::SnapCandidatePoint const &p, // in desktop coordinates
SnapConstraint const &c) const;
bool isUnselectedNode(Geom::Point const &point, std::vector<Inkscape::SnapCandidatePoint> const *unselected_nodes) const;
void _collectPaths(Inkscape::SnapCandidatePoint const &p,
bool const &first_point) const;
void _clear_paths() const;
Geom::PathVector* _getBorderPathv() const;
Geom::PathVector* _getPathvFromRect(Geom::Rect const rect) const;
void _getBorderNodes(std::vector<SnapCandidatePoint> *points) const;
}; // end of ObjectSnapper class
void getBBoxPoints(Geom::OptRect const bbox, std::vector<SnapCandidatePoint> *points, bool const isTarget, bool const includeCorners, bool const includeLineMidpoints, bool const includeObjectMidpoints);
} // end of namespace Inkscape
#endif