Tag List Report

The following document contains the listing of user tags found in the code. Below is the summary of the occurrences per tag.

Tag ClassTotal number of occurrencesTag strings used by tag class
TODO186@todo

Each tag is detailed below:

TODO

Number of occurrences found in the code: 186

org.geotoolkit.GeoapiTestLine
For a mysterious reason, some GIGS tests are skipped by this test suite (see the XML files in the {@code target/surefire-report} directory). We need more investigation about what is going on. Even when no tests is skipped, some tests are incomplete. For example GIGS 2008 should fail when "gravity-related depth" is not on the list of allowed axis names - this is not currently the case.79
org.geotoolkit.console.InteractiveConsoleLine
Localize495
org.geotoolkit.console.ReferencingCommandsLine
Localize messages.323
org.geotoolkit.coverage.AbstractCoverageLine
Proposed default implementation: invokes {@link #getEnvelope}, extract the spatial and temporal parts and put them in a {@link Extent} object.263
Proposed default implementation: invokes {@link #getEnvelope}, extract the spatial and temporal parts, get the grid geometry and create on-the-fly a {@link DomainObject} for each cell.279
org.geotoolkit.coverage.CategoryLine
The algorithm for finding minimum and maximum values is very simple for now and will not work if the transformation has local extrema. We would need some more sophisticated algorithm for the most general cases. Such a general algorithm would be useful in {@link GeophysicsCategory#getRange} as well.508
org.geotoolkit.coverage.CategoryListLine
Returns an instance of {@link MeasurementRange} if we are a geophysics category list.526
We could add an optimization after the loops checking for category change: if we were allowed to search for nearest category (overflowFallback!=null), then make sure that the category really changed. There is already a slight optimization for the most common cases, but maybe we could go a little bit further.962
Not yet implemented.1228
org.geotoolkit.coverage.ColorModelFactoryLine
The user may want to set explicitly the number of bits each pixel occupied. We need to think about an API to allows that.91
org.geotoolkit.coverage.CoverageStackLine
provides a better formatting of the point coordinate.1023
org.geotoolkit.coverage.GeophysicsCategoryLine
The algorithm for finding minimum and maximum values is very simple for now and will not work if the transformation has local extremas. We would need some more sophesticated algorithm for the most general cases. Such a general algorithm would be useful in the super-class constructor as well.86
org.geotoolkit.coverage.GridSampleDimensionLine
We should do a better job in {@code CategoryList.getRange()} when selecting the appropriate data type. {@link TypeMap#getSampleDimensionType(Range)} may be of some help.1008
What should we do when the value can't be formatted? {@code GridSampleDimension} returns {@code null} if there is no category or if an exception is thrown, but {@code CategoryList} returns "Untitled" if the value is an unknown NaN, and try to format the number anyway in other cases.1068
This method may be deprecated in a future version. It it strange to use only one {@code SampleDimension} object for creating a multi-bands color model. Logically, we would expect as many {@code SampleDimension}s as bands.1372
This method may be deprecated in a future version. It it strange to use only one {@code SampleDimension} object for creating a multi-bands color model. Logically, we would expect as many {@code SampleDimension}s as bands.1402
This static initializer will imply immediate class loading of a lot of JAI dependencies. This is a pretty high overhead if JAI is not wanted right now. The correct approach is to declare the image operation into the {@code META-INF/registryFile.jai} file, which is automatically parsed during JAI initialization. Unfortunately, it can't access private classes and we don't want to make our registration classes public. We can't move our registration classes into a hidden "resources" package neither because we need package-private access to {@code CategoryList}. For now, we assume that people using the GC package probably want to work with {@link org.geotoolkit.coverage.grid.GridCoverage2D}, which make extensive use of JAI. Peoples just working with {@link org.geotoolkit.coverage.Coverage} are stuck with the overhead. Note that we register the image operation here because the only operation argument is of type {@code GridSampleDimension[]}. Consequently, the image operation may be invoked at any time after class loading of {@link GridSampleDimension}. <p> Additional note: moving the initialization into the {@code META-INF/registryFile.jai} file may not be the best idea neither, since peoples using JAI without the GCS module may be stuck with the overhead of loading GC classes.1502
org.geotoolkit.coverage.SampleTranscoderLine
If the destination image is a single-banded, non-interleaved sample model, we could apply the transform directly in the {@link java.awt.image.DataBuffer}. We can even avoid to copy sample value if source and destination raster are the same.126
org.geotoolkit.coverage.TransformedCoverageLine
Provides a localized message.151
org.geotoolkit.coverage.grid.GridCoverage2DLine
Override {@link #createRendering} and use the affine transform operation. Also uses the JAI's "Transpose" operation is x and y axis are interchanged.790
org.geotoolkit.coverage.grid.GridCoverageFactoryLine
Put there only the hints we need.122
org.geotoolkit.coverage.grid.ViewsManagerLine
The "Piecewise" operation is disabled because javac 1.4.1_01 generate illegal bytecode. This bug is fixed in javac 1.4.2-beta. However, we still have an ArrayIndexOutOfBoundsException in JAI code...327
Not yet implemented. For now we use the packed view as a close match. Future version will needs to make sure that we returns the same instance than PACKED when suitable.599
org.geotoolkit.coverage.io.GridCoverageStorePoolLine
Needs a background task for disposing the readers or writers after some amount of inactivity.57
org.geotoolkit.coverage.io.ImageCoverageWriterTestLine
Platforms other than MacOS produce random numbers in the area where pad values are expected. This is related to the JAI issue documented in the Resample2D class (nearest neighbor interpolation of floating point values).333
org.geotoolkit.coverage.processing.AbstractOperationLine
The return type will be changed from {@link String} to {@code Identifier}.94
org.geotoolkit.coverage.processing.CachingCoverageProcessorLine
Use the capability of {@link Cache} to evict entries based on cost calculation.66
org.geotoolkit.coverage.processing.ColorMapLine
We need to investigate if this object should be defined as an implementation of {@link org.geotoolkit.styling.ColorMap}.59
org.geotoolkit.coverage.processing.OperationsTestLine
Investigate why the color palette is lost.65
Investigate why the geophysics view is much more visible than the non-geophysics one.157
org.geotoolkit.coverage.processing.ResampleTestLine
Investigate why we get a Lookup operation on the first coverage.115
org.geotoolkit.coverage.processing.operation.AddConstLine
Should operates on {@code sampleToGeophysics} transform when possible. See <A HREF="http://jira.codehaus.org/browse/GEOT-610">GEOT-610</A>.69
org.geotoolkit.coverage.processing.operation.BandSelector2DLine
It would be nice if we could use always the "BandSelect" operation without the "Null" one. But as of JAI-1.1.1, "BandSelect" does not detect by itself the case were no copy is required.75
org.geotoolkit.coverage.processing.operation.DivideByConstLine
Should operates on {@code sampleToGeophysics} transform when possible. See <A HREF="http://jira.codehaus.org/browse/GEOT-610">GEOT-610</A>.69
org.geotoolkit.coverage.processing.operation.GradientMagnitudeLine
When the masks are symmetric (e.g. Sobel, Prewitt (or Smoothed), isotropic, etc.), then this algorithm matches the "normalization factor" times "spatial factor" provided by249
org.geotoolkit.coverage.processing.operation.MultiplyConstLine
Should operates on {@code sampleToGeophysics} transform when possible. See <A HREF="http://jira.codehaus.org/browse/GEOT-610">GEOT-610</A>.69
org.geotoolkit.coverage.processing.operation.ResampleLine
Needs to migrate here the relevant code from {@link Resampler2D}.222
org.geotoolkit.coverage.processing.operation.RescaleLine
Should operates on {@code sampleToGeophysics} transform when possible. See <A HREF="http://jira.codehaus.org/browse/GEOT-610">GEOT-610</A>.77
org.geotoolkit.coverage.processing.operation.SubtractConstLine
Should operates on {@code sampleToGeophysics} transform when possible. See <A HREF="http://jira.codehaus.org/browse/GEOT-610">GEOT-610</A>.66
org.geotoolkit.coverage.processing.operation.SubtractFromConstLine
Should operates on {@code sampleToGeophysics} transform when possible. See <A HREF="http://jira.codehaus.org/browse/GEOT-610">GEOT-610</A>.66
org.geotoolkit.coverage.sql.LayerEntryLine
We should compute it automatically instead.100
org.geotoolkit.coverage.sql.NewGridCoverageLine
This is an ugly hack. Needs to do something better.39
org.geotoolkit.display.axis.AbstractGraduationLine
Add localization support.116
org.geotoolkit.display.shape.XRectangle2DLine
Check for negative width or height (should returns {@code false}).475
org.geotoolkit.gui.swing.FormatChooserLine
Need a way to find the format locale.163
org.geotoolkit.gui.swing.MouseReshapeTrackerLine
Consider renaming get/setClip into something else.109
Be more consistent on "visor", "rectangle" and "rectangular shape" wording. The later don't said enough what there are about.110
org.geotoolkit.gui.swing.ParameterEditorLine
This class do not yet support the edition of parameter value.90
This interface should have a {@code setEditable(boolean)} method.605
This editor should use {@code JSpinner}, but we need to gets the minimum and maximum values first since spinner needs bounds.629
org.geotoolkit.gui.swing.coverage.CategoryRecordLine
Current implementation does not yet recognize logarithmic transfer function.149
org.geotoolkit.gui.swing.coverage.CoverageListLine
Use switch(String) with Java 7.284
Current implementation remove only the row from the JTable. It does not yet update the database.606
org.geotoolkit.gui.swing.coverage.NewGridCoverageDetailsLine
switch(String) with Java 7.266
org.geotoolkit.gui.swing.image.GradientKernelEditorLine
Why positives numbers are on the first row? This is the opposite of other vertical gradient masks. Need to verify in J.J. Simpson (1990).130
org.geotoolkit.gui.swing.image.ImagePropertiesLine
Check for {@code WritablePropertySource} and make cells editable accordingly.700
org.geotoolkit.gui.swing.image.ImageTableModelLine
Should supports deferred execution: request for a new tile should wait some maximal amount of time (e.g. 0.1 seconds). If the tile is not yet available after that time, the model should returns {@code null} at this time and send a "data changed" event later when the tile is finally available.47
org.geotoolkit.gui.swing.referencing.AuthorityCodesComboBoxLine
switch(String) when allowed to use Java 7.271
org.geotoolkit.gui.swing.referencing.CoordinateChooserLine
Provide a better error message.411
org.geotoolkit.image.DeferredPlanarImageLine
Should be inlined in the constructor if only Sun was to fix RFE #4093999 ("Relax constraint on placement of this()/super() call in constructors").211
org.geotoolkit.image.TransfertRectIterLine
Localize.58
org.geotoolkit.image.io.PaletteFactoryLine
Current implementation ignores the name and builds a gray scale in all cases. Future version may improve on that.909
org.geotoolkit.image.io.metadata.ReferencingBuilderLine
The base CRS is not yet declared for the {@code DerivedCRS} case.379
{@code VerticalDatum}, {@code TemporalDatum} and {@code ImageDatum} are not yet implemented.643
org.geotoolkit.image.io.mosaic.GridNodeLine
In its current form, this method is not quite useful since it does its job only in the cases where <code>GridTileManager</code> would have been used instead than the <code>TreeTileManager</code>. It still useful for assertions since this method is used in the context of <code>ComparedTileManager</code>. <p> This method could be made more useful by extending its scope beyond the cases handled by <code>GridTileManager</code>. We could accept larger tiles having a size which is a multiple of "normal" tiles.491
org.geotoolkit.image.io.mosaic.MosaicControllerLine
Handle the {@link MosaicImageWriteParam} case.93
org.geotoolkit.image.io.mosaic.MosaicImageReaderLine
Localize.1487
org.geotoolkit.image.io.mosaic.MosaicImageWriterLine
Current implementation do not yet supports source region and subsampling settings. An exception will be thrown if any of those parameters are set.328
Localize.1776
org.geotoolkit.image.io.mosaic.MosaicTestBaseLine
The commented-out values are the ones for a smaller tile size (480 instead of 960). It may be better to use them since it use different values for X and Y axis.65
org.geotoolkit.image.io.mosaic.TileLine
Localize the exception message.552
org.geotoolkit.image.io.plugin.AsciiGridReaderLine
The current implementation ignores the <code>seekForwardOnly</code> parameter. It processes as if that parameter was always set to <code>true</code>.167
org.geotoolkit.image.io.plugin.AsciiGridWriterLine
Overriding not yet allowed. We are waiting to see if this API is really appropriate.256
org.geotoolkit.image.io.plugin.AsciiGridWriterTestLine
Can not run because spatial metadata are missing.61
Can not run because spatial metadata are missing.69
Can not run because spatial metadata are missing.77
Can not run because spatial metadata are missing.85
Can not run because spatial metadata are missing.93
Can not run because spatial metadata are missing.101
Can not run because spatial metadata are missing.109
org.geotoolkit.image.io.plugin.NetcdfDimensionLine
Define a 'sourceToTargetDimension' method somewhere based on the value of the derivative at the center position.142
org.geotoolkit.image.io.plugin.NetcdfImageReaderLine
Current implementation delegates to {@code read(int, param)}. Futures versions should do a more efficient work.1252
Localize1418
org.geotoolkit.image.io.plugin.NetcdfImageWriterLine
Localize330
org.geotoolkit.image.io.plugin.RawTiffImageReaderLine
Localize.1191
org.geotoolkit.image.io.plugin.TextMatrixImageWriterTestLine
The constructor uses a quite large tolerance threshold for floating point values because of the precision lost when formating the numbers. We should probably increase the writer accuracy, then revisit this threshold.45
org.geotoolkit.image.io.plugin.WorldFileImageReaderLine
Current implementation checks only {@link File} object. We should check URL as well.173
org.geotoolkit.image.jai.FloodFillLine
This class is abstract for now because not yet implemented as a JAI operation. However the static methods are ready for use.59
org.geotoolkit.image.jai.FloodFillTestLine
The "Flood fill" operation is not yet a registered JAI operation.55
org.geotoolkit.internal.coverage.CoverageUtilitiesLine
Move this method in {@link org.geotoolkit.coverage.processing.Operation2D}.319
org.geotoolkit.internal.coverage.TransferFunctionLine
Needs to handle the logarithmic case.87
org.geotoolkit.internal.image.io.DataTypesLine
Use switch on String.104
org.geotoolkit.internal.jaxb.code.CodeListProxyLine
JAXB BUG: Properties must be declared in reverse order. This is fixed in a JAXB version more recent than the one provided in the JDK. A Geotk test case should fail when the bug will be fixed, which will remind us to restore the correct order.51
Replace the argument type by {@link Enum} if we fix the type of ISO 19115 code lists which are supposed to be enum.116
org.geotoolkit.internal.jaxb.gco.MeasureLine
We should annotate {@link org.geotoolkit.measure.Measure} directly if we can find some way to use {@code @XmlValue} with that class.48
org.geotoolkit.internal.jaxb.metadata.RS_ReferenceSystemLine
We need a better plugin mechanism.106
We need a better plugin mechanism.128
org.geotoolkit.internal.jaxb.referencing.TimeInstantLine
The namespace of this class is set to {@link Namespaces#GMD} as a workaround. Actually we do that because we already have an other class in the GML binding of Constellation, and it falls on conflict. Remove the namespace, in order to fallback on GML, when the temporal implementation will have a floor in Geotk.47
org.geotoolkit.internal.jaxb.referencing.TimePeriodLine
The namespace of this class is set to {@link Namespaces#GMD} as a workaround. Actually we do that because we already have an other class in the GML binding of Constellation, and it falls on conflict. Remove the namespace, in order to fallback on GML, when the temporal implementation will have a floor in Geotk.43
A time period can also be expressed as a begin position and a period or duration. This is not yet supported in the current implementation.48
org.geotoolkit.internal.jaxb.referencing.TimePeriodBoundLine
The namespace of this class is set to {@link Namespaces#GMD} as a workaround. Actually we do that because we already have an other class in the GML binding of Constellation, and it falls on conflict. Remove the namespace, in order to fallback on GML, when the temporal implementation will have a floor in Geotk.44
org.geotoolkit.internal.referencing.AxisDirectionsLine
Consider moving to a public package.41
org.geotoolkit.internal.referencing.SeparableTransformLine
This is a temporary interface. We should probably move the "separate" method as a AbstractMathTransform protected method, and change the argument type to DimensionFilter (we need to find a better name) in order to store the target dimensions resulting from the separation. We don't do that now because it would probably be better to refactor DimensionFilter in order to implement its work in the various AbstractMathTransform sub-classes.26
org.geotoolkit.internal.referencing.factory.ImplementationHintsLine
Maybe we should provide a generalized form of this annotation in a public package. An alternative would be to use reflection for inspecting some static flavor of <code>getImplementationHints()</code>.32
org.geotoolkit.internal.swing.table.IdentifiedObjectRendererLine
Future version should provides more sophisticated rendering, for example a button for showing a popup with more properties.36
org.geotoolkit.internal.wizard.frame.MainLine
Use "switch in strings" when we will be allowed to compile for Java 7.165
org.geotoolkit.io.wkt.ParserLine
See if a future version of J2SE allows us to get ride of this ugly hack.55
org.geotoolkit.io.wkt.ReferencingParserLine
Authority code is currently ignored. We may consider to create a subclass of {@link Unit} which implements {@link IdentifiedObject} in a future version.526
The vertical datum type is currently ignored.886
The coordinate system used is always a Geotk implementation, since we don't know which method to invokes in the {@link CSFactory} (is it a Cartesian coordinate system? a spherical one? etc.).913
org.geotoolkit.measure.CoordinateFormatLine
parsing is not yet implemented in this version.77
org.geotoolkit.measure.UnitsLine
<a href="http://kenai.com/jira/browse/JSR_275-41">JSR-275 bug</a>68
<a href="http://kenai.com/jira/browse/JSR_275-41">JSR-275 bug</a>84
<a href="http://kenai.com/jira/browse/JSR_275-41">JSR-275 bug</a>92
The <code>org.geotoolkit.image.io.plugin.NetcdfMetadataTranscoder</code> class needs that we parse the {@code "degrees_west"} symbol in such a way that the conversion from {@code "degrees_west"} to {@code "degrees_east"} reverse the sign of the values. This is not yet supported by current implementation.325
org.geotoolkit.metadata.ISOTestLine
Current implementation relies on {@link Metadata} dependencies. This is probably not enough; we should provide an explicit list of metadata interface.51
org.geotoolkit.metadata.MetadataAnnotationsTestLine
Use string switch with JDK 7.234
org.geotoolkit.metadata.NameMapTestLine
Apparently, String literals in annotation are not interned. We should check if this behavior changes in a future JDK version, and if so enable the commented-out test.35
org.geotoolkit.metadata.iso.acquisition.DefaultPlatformPassLine
annotate an implementation of {@link Geometry} in order to annotate this method.127
org.geotoolkit.metadata.iso.citation.DefaultCitationLine
Use an unmodifiable {@link Date} here.339
org.geotoolkit.metadata.iso.citation.DefaultCitationTestLine
Disabled for now, because it doesn't seem to work.122
org.geotoolkit.metadata.iso.content.DefaultRangeElementDescriptionLine
implements {@link Record} in order to use the annotation.152
org.geotoolkit.metadata.iso.lineage.DefaultSourceLine
needs to annotate the referencing module before.202
org.geotoolkit.metadata.iso.maintenance.DefaultMaintenanceInformationLine
needs an implementation of org.opengis.temporal modules to anntote this parameter.202
org.geotoolkit.metadata.iso.quality.DefaultQuantitativeResultLine
Find an implementation of {@link Record}. The one in this class is deprecated.129
org.geotoolkit.metadata.iso.spatial.DefaultGCPLine
finish the annotation on the referencing module before107
org.geotoolkit.metadata.iso.spatial.DefaultGeoreferenceableLine
This attribute is declared as mandatory in ISO 19115-2. However metadata compliant with ISO 19115 (without the -2 part) do not contains this attribute. How should we handle the XML formatting for this one?240
org.geotoolkit.metadata.netcdf.NetcdfMetadataReaderLine
Current Units.valueOf(String) implementation ignore direction in "degrees_east" or "degrees_west". We need to take that in account (with "degrees_west" to "degrees_east" converter that reverse the sign).375
org.geotoolkit.nature.SeaWaterLine
What to do with pression!?! Check the equation of state.346
org.geotoolkit.openoffice.FormulasLine
localize.390
org.geotoolkit.parameter.ImagingParameterLine
Add automatic conversions, if it appears useful for JAI parameters.211
org.geotoolkit.referencing.CRS_TestLine
Not yet working.144
Not yet fixed.203
org.geotoolkit.referencing.GeodeticCalculatorLine
Current implementation will provides an inaccurate value for antipodal points. For now a warning is logged in such case. In a future version (if we have volunter time) we should provides a solution (search Internet for "<cite>azimuth antipodal points</cite>").770
We should check for cases where the path cross the 90°N, 90°S, 90°E or 90°W boundaries.1136
org.geotoolkit.referencing.PropertiesLine
Add properties for {@link IdentifiedObject} sub-interfaces.151
org.geotoolkit.referencing.adapters.NetcdfAxisLine
This method is currently implemented only for the 1D-case. Generalization to the 2D case would probably require a change in the method signature.377
org.geotoolkit.referencing.crs.DefaultDerivedCRSLine
Provides an API for specifying the datum at construction time which is different than the datum of the base CRS.54
org.geotoolkit.referencing.datum.DefaultVerticalDatumLine
Disabled for now - we need to define an adapter.255
org.geotoolkit.referencing.factory.AbstractAuthorityFactoryLine
We need a better way to declare some kind of "default" authority. At the very least, we should not accept hard-coded heuristic rules like this one... A possible approach may be to declare a "public static final Citation AUTHORITY" constant in the classes, to be fetched by reflection.133
org.geotoolkit.referencing.factory.AuthorityFactoryAdapterLine
We should returns the union of authority codes from all underlying factories.785
org.geotoolkit.referencing.factory.FallbackAuthorityFactoryLine
Needs a mechanism for avoiding to query the same factory twice when the fallback is the same instance than the primary factory for some {@link AuthorityFactory} interfaces.62
org.geotoolkit.referencing.factory.ReferencingFactoryContainerLine
The current implementation does not break a 3D Geographic CRS or a 3D Projected CRS into its components. The capability may be added in a future release (see <a href="http://jira.geotoolkit.org/browse/GEOTK-129">GEOTK-129</a>).533
org.geotoolkit.referencing.factory.ReferencingObjectFactoryLine
Not yet implemented.1054
org.geotoolkit.referencing.factory.epsg.AuthorityCodesLine
Not yet implemented.491
org.geotoolkit.referencing.factory.epsg.DirectEpsgFactoryLine
Current implementation maps all "vertical" datum to {@link VerticalDatumType#GEOIDAL}. We don't know yet how to maps the exact vertical datum type from the EPSG database.1637
The ordering is not consistent among all database software, because the "accuracy" column may contains null values. When used in an "ORDER BY" clause, PostgreSQL put null values last, while Access and HSQL put them first. The PostgreSQL's behavior is better for what we want (put operations with unknown accuracy last). Unfortunately, I don't know yet how to instruct Access to put null values last using standard SQL ("IIF" is not standard, and Access doesn't seem to understand "CASE ... THEN" clauses).2930
org.geotoolkit.referencing.factory.web.AutoCRSFactoryLine
Replace this with full FactorySPI system.118
org.geotoolkit.referencing.factory.web.URN_ParserLine
Implementation should be replaced by some mechanism using {@code GenericName} (at least the call to {@code String.regionMatches}) otherwise this method will fails if there is spaces around the separator.95
org.geotoolkit.referencing.operation.COFactoryUsingMolodenskyTestLine
We should not need to specify any {@code Hints.DATUM_SHIFT_METHOD} in order to ensure that the default is really {@code Molodensky}. Unfortunately letting the default factory cause random test failure if a factory using a different method has been used previously and still in the cache. We need to see if this problem can be avoid after we switch to JSR-330.117
org.geotoolkit.referencing.operation.DefaultConcatenatedOperationLine
We should use a Map and merge only one accuracy for each specification.278
org.geotoolkit.referencing.operation.DefaultCoordinateOperationFactoryLine
Implement GEOT-352 here.701
When rotating the prime meridian, we should ensure that transformed longitudes stay in the range [-180..+180°].726
Provides a non-null method.938
Rotation of prime meridian not yet implemented.966
Transformation version set to "(unknown)". We should search this information somewhere.967
(<a href="http://jira.geotoolkit.org/browse/GEOTK-83">GEOTK-83</a>) This method works for some simple cases (e.g. no datum change), and give up otherwise. Before to give up at the end of this method, we should try the following: <ul> <li>Maybe {@code sourceCRS} uses a non-ellipsoidal height. We should replace the non-ellipsoidal height by an ellipsoidal one, create a transformation step for that (to be concatenated), and then try again this operation step.</li>1123
org.geotoolkit.referencing.operation.DefaultOperationMethodLine
The check for {@link ConcatenatedTransform} and {@link PassThroughTransform} works only for Geotk implementations.418
org.geotoolkit.referencing.operation.DefaultPassThroughOperationLine
Current version works only with Geotk implementation.142
org.geotoolkit.referencing.operation.SamplePointsLine
The coordinate values in the EPSG case (the last row in each constructor call) is for Coordinate operation "NAD27 to WGS 84 (88)" for Cuba. May not be the most appropriate operation, but this is the one selected by the current ordering criterion in the SQL statements (GEOTK-80).56
org.geotoolkit.referencing.operation.Script3DTestLine
Not yet enabled because the hook to {@code DefaultCoordinateOperationFactory} are not yet there.38
org.geotoolkit.referencing.operation.builder.LocalizationGridLine
<code>LocalizationGridBuilder</code> would be a better name.116
org.geotoolkit.referencing.operation.builder.LocalizationGridTransform2DLine
Current implementation projects an inside point on the nearest border. Could we do something better?358
org.geotoolkit.referencing.operation.provider.AlbersEqualAreaLine
According the <cite>remote-sensing</cite> web site, the OGC name for this parameter is <code>"longitude_of_center"</code>. However the <cite>spatial-reference</cite> web site said <code>"central_meridian"</code>, which was also the usage in GeoTools 2.x and is preserved for now.90
According the <cite>remote-sensing</cite> web site, the OGC name for this parameter is <code>"latitude_of_center"</code>. However the <cite>spatial-reference</cite> web site said <code>"latitude_of_origin"</code>, which was also the usage in GeoTools 2.x and is preserved for now.115
org.geotoolkit.referencing.operation.transform.DimensionFilterLine
This class contains a set of static methods that could be factored out in some kind of {@code org.geotoolkit.util.SortedIntegerSet} implementation.59
Consider providing a {@code subTransform(DimensionFilter)} method in {@link AbstractMathTransform}, and move some {@code DimensionFilter} code in {@code AbstractMathTransform} sub-classes. This would allow us to separate transforms that are defined in downstream modules, like NetCDF.62
org.geotoolkit.referencing.operation.transform.PassThroughTransformLine
The {@link #numTrailingOrdinates} parameter is not part of OpenGIS specification. We should returns a more complex WKT when {@code numTrailingOrdinates != 0}, using an affine transform to change the coordinates order.573
org.geotoolkit.referencing.operation.transform.TransformTestBaseLine
We may need to allow this method to use the {@linkplain #tolerance} value when comparing floating point numbers, but that would require a kind of WKT parsing in this method.371
org.geotoolkit.resources.DescriptionsLine
Current implementation just invokes {@link #format}. Need to format only when {@code toString(Locale)} is invoked.207
Current implementation just invokes {@link #format}. Need to format only when {@code toString(Locale)} is invoked.225
org.geotoolkit.resources.ErrorsLine
Current implementation just invokes {@link #format}. Need to format only when {@code toString(Locale)} is invoked.1484
Current implementation just invokes {@link #format}. Need to format only when {@code toString(Locale)} is invoked.1502
org.geotoolkit.resources.LocalesLine
Current implementation returns a hard-coded list. Future implementations may perform a more intelligent work.58
org.geotoolkit.resources.VocabularyLine
Current implementation just invokes {@link #format}. Need to format only when {@code toString(Locale)} is invoked.1898
Current implementation just invokes {@link #format}. Need to format only when {@code toString(Locale)} is invoked.1914
org.geotoolkit.util.DefaultInternationalStringLine
Current implementation does not return a synchronized set. We should synchronize on the same lock than the one used for accessing the internal locale map.209
org.geotoolkit.util.collection.CheckedHashMapLine
Current implementation do not synchronize the {@linkplain #entrySet entry set}, {@linkplain #keySet key set} and {@linkplain #values values} collection.40
org.geotoolkit.util.collection.FrequencySortedSetLine
Not yet implemented.274
Not yet implemented.282
Not yet implemented.290
Move to a XCollections constant.376
org.geotoolkit.util.collection.RangeSetLine
Support open intervals.364
Support closed intervals.614
org.geotoolkit.util.collection.WeakValueHashMapLine
Use the commented line instead if a future Java version supports generic arrays.255
org.geotoolkit.xml.MarshallerPoolLine
Need a timeout for disposing marshallers that have been unused for a while.63