This file is indexed.

/usr/include/ITK-4.9/itkSubsamplerBase.h is in libinsighttoolkit4-dev 4.9.0-4ubuntu1.

This file is owned by root:root, with mode 0o644.

The actual contents of the file can be viewed below.

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
/*=========================================================================
 *
 *  Copyright Insight Software Consortium
 *
 *  Licensed under the Apache License, Version 2.0 (the "License");
 *  you may not use this file except in compliance with the License.
 *  You may obtain a copy of the License at
 *
 *         http://www.apache.org/licenses/LICENSE-2.0.txt
 *
 *  Unless required by applicable law or agreed to in writing, software
 *  distributed under the License is distributed on an "AS IS" BASIS,
 *  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 *  See the License for the specific language governing permissions and
 *  limitations under the License.
 *
 *=========================================================================*/
#ifndef itkSubsamplerBase_h
#define itkSubsamplerBase_h

#include "itkObject.h"
#include "itkSample.h"
#include "itkSubsample.h"

namespace itk {
namespace Statistics {
/** \class SubsamplerBase
 * \brief This is the base subsampler class which defines the subsampler API.
 *
 * This class will search a Sample provided by SetSample and return a
 * Subsample that are related in some way to the queried value.
 * Some examples of subsampling strategies include uniform random selection,
 * selection based on KdTree, and selection based on spatial proximity.
 *
 * This is an Abstract class that can not be instantiated.
 * There are multiple subsamplers that derive from this class and
 * provide specific implementations of subsampling strategies.
 *
 * \sa RegionConstrainedSubsampler, SpatialNeighborSubsampler
 * \sa GaussianRandomSpatialNeighborSubsampler
 * \sa UniformRandomSpatialNeighborSubsampler
 * \ingroup ITKStatistics
 */

template < typename TSample >
class SubsamplerBase : public Object
{
public:
  /** Standard class typedefs */
  typedef SubsamplerBase                        Self;
  typedef Object                                Superclass;
  typedef Self                                  Baseclass;
  typedef SmartPointer<Self>                    Pointer;
  typedef SmartPointer<const Self>              ConstPointer;

  /** Run-time type information (and related methods) */
  itkTypeMacro(SubsamplerBase, Object);

  /** implement type-specific clone method */
  itkCloneMacro(Self);

  /** typedef alias for the source data container */
  typedef TSample                                          SampleType;
  typedef typename SampleType::ConstPointer                SampleConstPointer;
  typedef typename TSample::MeasurementVectorType          MeasurementVectorType;
  typedef typename TSample::InstanceIdentifier             InstanceIdentifier;

  typedef Subsample<TSample>                               SubsampleType;
  typedef typename SubsampleType::Pointer                  SubsamplePointer;
  typedef typename SubsampleType::ConstIterator            SubsampleConstIterator;
  typedef typename SubsampleType::InstanceIdentifierHolder InstanceIdentifierHolder;
  typedef unsigned int                                     SeedType;

  /** Plug in the actual sample data */
  itkSetConstObjectMacro(Sample, SampleType);
  itkGetConstObjectMacro(Sample, SampleType);

  /** Indicate whether the Search method can return the query point
   * as one element of the Subsample
   */
  itkSetMacro(CanSelectQuery, bool);
  itkGetConstReferenceMacro(CanSelectQuery, bool);
  itkBooleanMacro(CanSelectQuery);

  /** Provide an interface to set the seed.
   *  The seed value will be used by subclasses where appropriate.
   */
  itkSetMacro(Seed, SeedType);
  itkGetConstReferenceMacro(Seed, SeedType);


  /** Specify whether the subsampler should return all possible
   * matches. */
  virtual void RequestMaximumNumberOfResults()
  {
    if (!this->m_RequestMaximumNumberOfResults)
    {
      this->m_RequestMaximumNumberOfResults = true;
      this->Modified();
    }
  }

  /** Main Search method that MUST be implemented by each subclass
   * The Search method will find all points similar to query and return
   * them as a Subsample.  The definition of similar will be subclass-
   * specific.  And could mean spatial similarity or feature similarity
   * etc.  */
  virtual void Search(const InstanceIdentifier& query,
                      SubsamplePointer& results) = 0;

protected:
  /**
   * Clone the current subsampler.
   * This does a complete copy of the subsampler state
   * to the new subsampler
   */
  virtual typename LightObject::Pointer InternalClone() const ITK_OVERRIDE;

  SubsamplerBase();
  virtual ~SubsamplerBase() {};

  virtual void PrintSelf(std::ostream& os, Indent indent) const ITK_OVERRIDE;

  SampleConstPointer m_Sample;
  bool               m_RequestMaximumNumberOfResults;
  bool               m_CanSelectQuery;
  SeedType           m_Seed;

private:
  SubsamplerBase(const Self&) ITK_DELETE_FUNCTION;
  void operator=(const Self&) ITK_DELETE_FUNCTION;

}; // end of class SubsamplerBase

} // end of namespace Statistics
} // end of namespace itk

#ifndef ITK_MANUAL_INSTANTIATION
#include "itkSubsamplerBase.hxx"
#endif

#endif