v2.11.1

QuadGK v2.11.1

Diff since v2.11.0

Merged pull requests:

• Adapt to Enzyme breaking change (#118) (@wsmoses)
• Fix in-place broadcast fusion in evalrule (#121) (@danielwe)

v2.11.0

QuadGK v2.11.0

Diff since v2.10.1

• gauss speedups for symmetric rules (#113)
• add reverse rules to make quadgk differentiable by Enzyme (#110), though Enzyme.jl#1692 needs to be fixed for this to be fully usable
• thread safety (#116)

Merged pull requests:

• Add Enzyme reverse rules (#110) (@wsmoses)
• speedup gauss for symmetric rules (#113) (@stevengj)
• CompatHelper: add new compat entry for Enzyme in [weakdeps] at version 0.12, (keep existing compat) (#114) (@github-actions[bot])
• add Enzyme to "extras" for testing, docs (#115) (@stevengj)
• make quadgk thread-safe (#116) (@stevengj)

v2.10.1

QuadGK v2.10.1

Diff since v2.10.0

Merged pull requests:

• omit pointless extra Newton iteration for λ (#112) (@stevengj)

v2.10.0

QuadGK v2.10.0

Diff since v2.9.4

Merged pull requests:

• Create Invalidations.yml (#60) (@ranocha)
• fix spurious underflow in kronrod (#94) (@stevengj)
• Bump actions/cache from 3 to 4 (#100) (@dependabot[bot])
• [CI] Some improvements (#101) (@giordano)
• Bump codecov/codecov-action from 3 to 4 (#102) (@dependabot[bot])
• Bump julia-actions/setup-julia from 1 to 2 (#103) (@dependabot[bot])
• Add language tags to code samples (#104) (@abhro)
• Bump julia-actions/cache from 1 to 2 (#105) (@dependabot[bot])
• rename gw -> wg (#107) (@stevengj)
• add eval_segbuf and quadgk_segbuf options (#108) (@stevengj)
• bugfix in gauss for large n (#111) (@stevengj)

Closed issues:

• spurious underflow in kronrod for large n (#93)
• confusion in kronrod() documentation? (#106)
• gauss incorrect for large n (#109)

v2.9.4

QuadGK v2.9.4

Diff since v2.9.3

Merged pull requests:

• allow endpoints evaluation for tiny initial intervals (#99) (@stevengj)

v2.9.3

QuadGK v2.9.3

Diff since v2.9.2

Merged pull requests:

• fix empty intervals (#98) (@stevengj)

Closed issues:

• quadgk with identical lower and upper bound is broken (#97)

v2.9.2

QuadGK v2.9.2

Diff since v2.9.1

Merged pull requests:

• avoid integrand evaluations at endpoints (#91) (@lxvm)

v2.9.1

QuadGK v2.9.1

Diff since v2.9.0

Merged pull requests:

• handle integrands <: AbstractVector (#90) (@stevengj)

Closed issues:

• Regression with QuadGK@2.9 (#89)

v2.9.0

QuadGK v2.9.0

Diff since v2.8.2

Merged pull requests:

• enable dependabot for GitHub actions (#75) (@ranocha)
• Bump actions/checkout from 2 to 3 (#76) (@dependabot[bot])
• Bump actions/cache from 1 to 3 (#77) (@dependabot[bot])
• Bump codecov/codecov-action from 1 to 3 (#78) (@dependabot[bot])
• Parallelization of integrand evaluations (#80) (@lxvm)
• Gauss–Kronrod weights for arbitrary weight functions (#83) (@stevengj)
• ensure int keys in rulecache (#84) (@stevengj)
• don't over-promote weighted-Gauss type to Float64 (#85) (@stevengj)
• Bump actions/checkout from 3 to 4 (#87) (@dependabot[bot])

Closed issues:

• Can not catch a JuliaError caused by function blowing up to infinity (#81)
• No cachedrule for DynamicQuantities types (#88)

v2.8.2

QuadGK v2.8.2

Diff since v2.8.1

Closed issues:

• stack overflow for integrating other Number types (#13)

Merged pull requests:

• Update adapt.jl (#74) (@Hmegh)