|
| 1 | +Changes that were made in August 2002. |
| 2 | + |
| 3 | +Note that this was based on a 7.3 development version and changes may not |
| 4 | +directly work with earlier versions. |
| 5 | + |
| 6 | +I fixed a bug in cubescan.pl that prevented signed numbers with no digits |
| 7 | +before a decimal point from being accepted. This was submitted as a separate |
| 8 | +patch and may already be applied. |
| 9 | + |
| 10 | +I reported but did not fix a potential buffer overrun problem in cube_yyerror |
| 11 | +in cubeparse.y. |
| 12 | + |
| 13 | +cube_inter should really return NULL if the two cubes don't overlap. However |
| 14 | +this requires changing to the new calling sequence and I don't know enough |
| 15 | +about how to do it to make the change. |
| 16 | + |
| 17 | +I changed all floats to doubles except for g_cube_penalty which I don't |
| 18 | +think can be changed to return double. This might cause the penalty to |
| 19 | +overflow sooner than one might expect, but overflow could have happened |
| 20 | +even with floats. |
| 21 | + |
| 22 | +I changed the output format (in cube_out) to use %.16g instead of %g, since the |
| 23 | +default is only 6 digits of precision. |
| 24 | + |
| 25 | +I changed all of the functions declared with (isstrict) to use the current |
| 26 | +method of declaring this. |
| 27 | + |
| 28 | +I changed all of the externally visible functions to be immutable which |
| 29 | +they are. I don't think this matters for the gist functions and didn't |
| 30 | +try to declare them immutable in case there was something tricky about them |
| 31 | +that I don't understand. |
| 32 | + |
| 33 | +I changed the regression tests to use some larger exponents to test output |
| 34 | +in exponential form. 1e7 was too small for this. |
| 35 | + |
| 36 | +I added some regression tests to check for 16 digits of precision. This |
| 37 | +may or may not be a good idea. |
| 38 | + |
| 39 | +I got rid of the swap_corners function. It created scratch boxes that |
| 40 | +were iterated through and deleted. This is slower than just getting the |
| 41 | +larger or smaller coordinate as needed, since swap_corners was doing the |
| 42 | +same thing with the overhead of a function call and memory allocation. |
| 43 | + |
| 44 | +I added memset calls to zero out newly allocated NDBOXes as the documentation |
| 45 | +on functions indicates should be done. |
| 46 | + |
| 47 | +I got rid of a call to cube_same in cube_lt and cube_gt since the test |
| 48 | +was redundant with other checks being made. The call to cube_same would |
| 49 | +only be faster if most of the time you were comparing equivalent cubes. |
| 50 | + |
| 51 | +In cube_lt and cube_gt, the second (UR) for loop for comparing |
| 52 | +extra coordinates to 0 had the wrong range. |
| 53 | + |
| 54 | +Note that the cube_distance function wasn't mentioned in the README.cube file. |
| 55 | + |
| 56 | +I added regression tests for the cube_distance function. |
| 57 | + |
| 58 | +I added the following new functions: |
| 59 | +cube |
| 60 | +cube_dim |
| 61 | +cube_ll_coord |
| 62 | +cube_ur_coord |
| 63 | +cube_is_point |
| 64 | +cube_enlarge |
| 65 | + |
| 66 | +cube takes text input and returns a cube. This is useful for making cubes |
| 67 | +from computed strings. |
| 68 | + |
| 69 | +cube_dim returns the number of dimensions stored in the the data structure |
| 70 | +for a cube. This is useful for constraints on the dimensions of a cube. |
| 71 | + |
| 72 | +cube_ll_coord returns the nth coordinate value for the lower left corner |
| 73 | +of a cube. This is useful for doing coordinate transformations. |
| 74 | + |
| 75 | +cube_ur_coord returns the nth coordinate value for the upper right corner |
| 76 | +of a cube. This is useful for doing coordinate transformations. |
| 77 | + |
| 78 | +cube_is_point returns true if a cube is also a point. This is true when the |
| 79 | +two defining corners are the same. |
| 80 | + |
| 81 | +cube_enlarge increases the size of a cube by a specified radius in at least |
| 82 | +n dimensions. If the radius is negative the box is shrunk instead. This |
| 83 | +is useful for creating bounding boxes around a point for searching for |
| 84 | +nearby points. All defined dimensions are changed by the radius. If n |
| 85 | +is greater than the number of defined dimensions and the cube is being |
| 86 | +increased (r >= 0) then 0 is used as the base for the extra coordinates. |
| 87 | +LL coordinates are decreased by r and UR coordinates are increased by r. If a |
| 88 | +LL coordinate is increased to larger than the corresponding UR coordinate |
| 89 | +(this can only happen when r < 0) than both coordinates are set to their |
| 90 | +average. |
| 91 | + |
| 92 | +I added regression tests for the new functions. |
| 93 | + |
| 94 | +I added documentation for cube_distance and the new functions to README.cube |
| 95 | +as well as making a few other minor changes. |
| 96 | + |
| 97 | +Bruno Wolff III <bruno@wolff.to> |
0 commit comments