You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Many bugs occur when printing float values containing integer values. The successive-scaling approach in the original algorithm would quickly make these values depart from whole numbers.
This modification attempts to avoid scaling the original value for numbers in the range 1.0 to 1e10, and instead compares with other floats constructed only by multiplying whole numbers, so guaranteed to remain exact until the mantissa runs out.
I added a test that verifies consistency with CPython for a range of values. In fact, CPython also struggles with some values (see, e.g. print(13 * .001)), but at least we have agreement for the values in the test (on my machine).
Thanks for the contribution, this looks really good! So far my testing of this PR has shown that it does do a better job than before. One test I used was micropython-ulab's tests suite and it fixed some printing issues there.
After looking at @goatchurchprime’s earlier proposed fix, I changed mine to use integer math for the whole-number part. I also changed the test to only test whole-numbers (since that is the focus of the fix) and added a couple of specific cases.
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This is to fix issue #4212
Many bugs occur when printing float values containing integer values. The successive-scaling approach in the original algorithm would quickly make these values depart from whole numbers.
This modification attempts to avoid scaling the original value for numbers in the range 1.0 to 1e10, and instead compares with other floats constructed only by multiplying whole numbers, so guaranteed to remain exact until the mantissa runs out.
I added a test that verifies consistency with CPython for a range of values. In fact, CPython also struggles with some values (see, e.g.
print(13 * .001)), but at least we have agreement for the values in the test (on my machine).Signed-off-by: Dan Ellis dan.ellis@gmail.com