Hi Thomas.
> ==============================================================================
> ---
> commons/proper/math/trunk/src/main/java/org/apache/commons/math/exception/util/LocalizedFormats.java
> (original)
> +++
> commons/proper/math/trunk/src/main/java/org/apache/commons/math/exception/util/LocalizedFormats.java
> Mon Jan 23 22:26:37 2012
> @@ -80,6 +80,7 @@ public enum LocalizedFormats implements
> CROSSING_BOUNDARY_LOOPS("some outline boundary loops cross each other"),
> CUMULATIVE_PROBABILITY_RETURNED_NAN("Cumulative probability function
> returned NaN for argument {0} p = {1}"),
> DIFFERENT_ROWS_LENGTHS("some rows have length {0} while others have
> length {1}"),
> + DIFFERENT_ORIG_AND_PERMUTED_DATA("original and permuted data must
> contain the same elements"),
OK (although, I'd have preferred to use a "NoDataException").
> DIGEST_NOT_INITIALIZED("digest not initialized"),
> DIMENSIONS_MISMATCH_2x2("got {0}x{1} but expected {2}x{3}"), /* keep */
> DIMENSIONS_MISMATCH_SIMPLE("{0} != {1}"), /* keep */
> @@ -133,7 +134,10 @@ public enum LocalizedFormats implements
> INTEGRATION_METHOD_NEEDS_AT_LEAST_TWO_PREVIOUS_POINTS("{0} method needs
> at least two previous points"),
> INTERNAL_ERROR("internal error, please fill a bug report at {0}"),
> INVALID_BINARY_DIGIT("invalid binary digit: {0}"),
> + INVALID_BINARY_CHROMOSOME("binary mutation works on BinaryChromosome
> only"),
OK.
> INVALID_BRACKETING_PARAMETERS("invalid bracketing parameters: lower
> bound={0}, initial={1}, upper bound={2}"),
> + INVALID_CROSSOVER_CHROMOSOME_LENGTH("both chromosomes must have same
> lengths: ({0}) ({1})"),
No: please use "DimensionMismatchException". No need to have a particular
message just because the "list" happens to be a "chromosome".
> + INVALID_FIXED_LENGTH_CHROMOSOME("one-point crossover works only on
> instances of AbstractListChromosome"),
I don't see how the "FIXED_LENGTH" fits with the message.
> INVALID_INTERVAL_INITIAL_VALUE_PARAMETERS("invalid interval, initial
> value parameters: lower={0}, initial={1}, upper={2}"),
> INVALID_ITERATIONS_LIMITS("invalid iteration limits: min={0}, max={1}"),
> INVALID_MAX_ITERATIONS("bad value for maximum iterations number: {0}"),
> @@ -270,6 +274,9 @@ public enum LocalizedFormats implements
> OUT_OF_BOUND_SIGNIFICANCE_LEVEL("out of bounds significance level {0},
> must be between {1} and {2}"),
> SIGNIFICANCE_LEVEL("significance level ({0})"), /* keep */
> OUT_OF_ORDER_ABSCISSA_ARRAY("the abscissae array must be sorted in a
> strictly increasing order, but the {0}-th element is {1} whereas {2}-th is
> {3}"),
> + OUT_OF_RANGE_CROSSOVER_RATE("crossover rate ({0}) out of [{1}, {2}]
> range"),
No. Please combine the simple "OUT_OF_RANGE" key with a "CROSSOVER_RATE" key
to be defined as:
---CUT---
CROSSOVER_RATE("crossover rate ({0})"),
---CUT---
[See the other constructor of "OutOfRangeException".]
> + OUT_OF_RANGE_ELITISM_RATE("elitism rate ({0}) out of [{1}, {2}] range"),
Ditto.
> + OUT_OF_RANGE_MUTATION_RATE("mutation rate ({0}) out of [{1}, {2}]
> range"),
Ditto.
> OUT_OF_RANGE_ROOT_OF_UNITY_INDEX("out of range root of unity index {0}
> (must be in [{1};{2}])"),
> OUT_OF_RANGE("out of range"), /* keep */
> OUT_OF_RANGE_SIMPLE("{0} out of [{1}, {2}] range"), /* keep */
> @@ -303,10 +310,12 @@ public enum LocalizedFormats implements
> SINGULAR_OPERATOR("operator is singular"),
> SUBARRAY_ENDS_AFTER_ARRAY_END("subarray ends after array end"),
> TOO_LARGE_CUTOFF_SINGULAR_VALUE("cutoff singular value is {0}, should be
> at most {1}"),
> + TOO_LARGE_TOURNAMENT_ARITY("tournament arity ({0}) cannot be bigger than
> population size ({1})"),
Please use the more specific "NumberIsTooLargeException", and combine it
with a "TOURNAMENT_ARITY" (if really necessary).
> TOO_MANY_ELEMENTS_TO_DISCARD_FROM_ARRAY("cannot discard {0} elements
> from a {1} elements array"),
> TOO_MUCH_CANCELLATION("too much cancellation in a denominator"),
> TOO_MANY_REGRESSORS("too many regressors ({0}) specified, only {1} in
> the model"),
> TOO_SMALL_COST_RELATIVE_TOLERANCE("cost relative tolerance is too small
> ({0}), no further reduction in the sum of squares is possible"),
> + TOO_SMALL_GENERATION_COUNT("number of generations ({0}) must be >= {1}"),
Please use the more specific "NumberIsTooSmallException".
> TOO_SMALL_INTEGRATION_INTERVAL("too small integration interval: length =
> {0}"),
> TOO_SMALL_ORTHOGONALITY_TOLERANCE("orthogonality tolerance is too small
> ({0}), solution is orthogonal to the jacobian"),
> TOO_SMALL_PARAMETERS_RELATIVE_TOLERANCE("parameters relative tolerance
> is too small ({0}), no further improvement in the approximate solution is
> possible"),
> @@ -340,6 +349,9 @@ public enum LocalizedFormats implements
> WEIGHT_AT_LEAST_ONE_NON_ZERO("weigth array must contain at least one
> non-zero value"),
> WRONG_BLOCK_LENGTH("wrong array shape (block length = {0}, expected
> {1})"),
> WRONG_NUMBER_OF_POINTS("{0} points are required, got only {1}"),
> + WRONG_SEQUENCE_LENGTH_RANDOMKEY("length of sequence for decoding ({0})
> has to be equal to the RandomKey length ({1})"),
Please use the more specific "DimensionMismatchException". I don't think
that pecific messages are necessary: those seems not to be "user" methods.
> + WRONG_REPR_AND_SREPR_SIZE("size of representation and sorted
> representation must be equals: {0} != {1}"),
Ditto.
> + WRONG_ORIG_AND_PERMUTED_SIZE("original and permuted data must have same
> length: {0} != {1}"),
Ditto.
> NUMBER_OF_POINTS("number of points ({0})"), /* keep */
> ZERO_DENOMINATOR("denominator must be different from 0"),
> ZERO_DENOMINATOR_IN_FRACTION("zero denominator in fraction {0}/{1}"),
> [...]
Thanks for helping with this,
Gilles
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org
