Merge branch 'release-0.7.1'

README.md

@@ -1,7 +1,7 @@
 # ACME.jl - Analog Circuit Modeling and Emulation for Julia
 
 [![Join the chat at https://gitter.im/HSU-ANT/ACME.jl](https://badges.gitter.im/HSU-ANT/ACME.jl.svg)](https://gitter.im/HSU-ANT/ACME.jl?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)
-[![Documentation](https://img.shields.io/badge/docs-v0.7.0-blue.svg)](https://hsu-ant.github.io/ACME.jl/v0.7.0)
+[![Documentation](https://img.shields.io/badge/docs-v0.7.1-blue.svg)](https://hsu-ant.github.io/ACME.jl/v0.7.1)
 
 ACME is a [Julia](http://julialang.org/) package for the simulation of
 electrical circuits, focusing on audio effect circuits. It allows to
@@ -122,7 +122,7 @@ fail to run altogether.
 
 ## Moving on
 
-There is some [documentation](https://hsu-ant.github.io/ACME.jl/v0.7.0)
+There is some [documentation](https://hsu-ant.github.io/ACME.jl/v0.7.1)
 available for how
 to use ACME. Additionally, you can take a look at the examples that can be found
 in the `examples` directory below `Pkg.dir("ACME")`.

REQUIRE

@@ -1,5 +1,5 @@
 julia 0.6
-Compat 0.61.0
+Compat 0.64.0
 DataStructures 0.2.9
 IterTools 0.1.0
 ProgressMeter 0.2.1

src/ACME.jl

@@ -298,7 +298,7 @@ function model_matrices(circ::Circuit, t::Rational{BigInt})
     merge!(res, Dict(zip([:v0 :ev :dv; :i0 :ei :di; :x0 :b :a; :q0 :eq_full :dq_full],
                          matsplit(x, rowsizes, [1; nu(circ); nx(circ)]))))
     for v in (:v0, :i0, :x0, :q0)
-        res[v] = squeeze(res[v], 2)
+        res[v] = squeeze(res[v], dims=2)
     end
 
     p = [pv(circ) pi(circ) px(circ)//2+pxd(circ)//t pq(circ)]
@@ -741,7 +741,7 @@ gensolve(a, b, thresh=0.1) =
 
 function rank_factorize(a::SparseMatrixCSC)
     f = a
-    nullspace = gensolve(a', spzeros(size(a, 2), 0))[2]
+    nullspace = gensolve(a', spzeros(eltype(a), size(a, 2), 0))[2]
     c = Matrix{eltype(a)}(I, size(a, 1), size(a, 1))
     while size(nullspace, 2) > 0
         i, j = argmax(abs.(nullspace)).I

src/circuit.jl

@@ -286,8 +286,9 @@ function topomat!(incidence::SparseMatrixCSC{T}) where {T<:Integer}
 
     dl = ti[:, (!).(t)]
     tv = spzeros(T, size(dl, 2), size(incidence, 2))
-    tv[:, t] = -dl'
-    tv[:, (!).(t)] = SparseMatrixCSC{T}(I, size(dl, 2), size(dl, 2))
+    # findall here works around JuliaLang/julia#27013 (introdcued in 0.7.0-DEV.4983)
+    tv[:, findall(t)] = -dl'
+    tv[:, findall((!).(t))] = SparseMatrixCSC{T}(I, size(dl, 2), size(dl, 2))
 
     tv, ti
 end

test/runtests.jl

@@ -203,7 +203,7 @@ end
 @testset "gensolve/rank_factorize" begin
     a = Rational{BigInt}[1 1 1; 1 1 2; 1 2 1; 1 2 2; 2 1 1; 2 1 2]
     b = Rational{BigInt}[1 2 3 4 5 6; 6 5 4 3 2 1; 1 0 1 0 1 0]
-    nullspace = ACME.gensolve(sparse(a'), spzeros(size(a, 2), 0))[2]
+    nullspace = ACME.gensolve(sparse(a'), spzeros(Rational{BigInt}, size(a, 2), 0))[2]
     @test nullspace'*a == spzeros(3, 3)
     c, f = ACME.rank_factorize(sparse(a * b))
     @test c*f == a*b