Repository navigation
A complex constant does not make the two sides of a limit differ - #2376
Merged
Rafael-SOWNet merged 1 commit intoNov 6, 2026
Merged
Conversation
Where the descent cannot compare the two sides of a limit, each side is asked through the whole of the machinery, but only of an expression that IsRealOnBothSides passes, and that check required every constant to be real. pi^2 / log(x, -2) is pi^2 ln(x) / ln(-2), 0 from each side of 1, and its two-sided limit was NaN. A constant is the same number on both sides, so a complex one cannot make them differ; it is now accepted, and what depends on x is read as before. x^x, x ln(x), x^i, i sgn(x) and (2 + i)/x at 0 stay NaN. limitcheck at DEPTH=3 over 39 seeds: eight two-sided NaNs answered (0, -oo, pi/2 among them), no other move. tanh(log(x, -1) - tan(x)) at 1 is still NaN; asking both sides now takes 14 s where it took 3 s, since each one-sided limit of it takes 4-6 s on master already. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #2376 +/- ##
=======================================
Coverage 86.77% 86.77%
=======================================
Files 312 312
Lines 68987 68987
Branches 20112 20112
=======================================
Hits 59862 59862
+ Misses 4920 4919 -1
- Partials 4205 4206 +1
Flags with carried forward coverage won't be shown. Click here to find out more.
🚀 New features to boost your workflow:
|
Rafael-SOWNet
deleted the
feature/a-complex-constant-is-the-same-on-both-sides
branch
November 6, 2026 21:49
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Where the descent cannot compare the two sides of a limit, each side is asked through the whole of the machinery, but only of an expression
IsRealOnBothSidespasses, and that check required every constant to be real.pi^2 / log(x, -2)ispi^2 ln(x) / ln(-2), 0 from each side of 1, and its two-sided limit was NaN.A constant is the same number on both sides, so a complex one cannot make them differ. It is now accepted, and what depends on
xis read as before.pi^2 / log(x, -2)at 10(log(1/2, 1) + pi^2) / log(x, -2)at 10log(x ^ (-1), -1) ^ 2at 1-oox^x,x ln(x),x^i,i sgn(x),(2 + i)/xat 0;ln(-2)/ln(x)at 1Measured
DEPTH=3, across 39 seeds against master: eight two-sided NaNs are answered, all correct, and nothing else moved.tanh(log(x, -1) - tan(x))at 1 is still NaN, but asking both sides now takes 14 s where it took 3 s. On master each one-sided limit of it already takes 4–6 s, and that slowness is a separate defect.Found by limitcheck at
SEED=69; no issue was filed.🤖 Generated with Claude Code