refactor(dsp): do not export sfilt, sfiltfilt and sresample, but re-e…

…xport DSP.stft

Project.toml

@@ -1,7 +1,7 @@
 name = "SignalAnalysis"
 uuid = "df1fea92-c066-49dd-8b36-eace3378ea47"
 authors = ["Mandar Chitre <[email protected]>"]
-version = "0.8.0"
+version = "0.9.0"
 
 [deps]
 DSP = "717857b8-e6f2-59f4-9121-6e50c889abd2"

src/SignalAnalysis.jl

@@ -7,7 +7,7 @@ using SignalBase
 using SignalBase.Units
 
 export nframes, nchannels, sampletype, framerate, duration
-export db2amp, amp2db, pow2db, db2pow
+export db2amp, amp2db, pow2db, db2pow, stft
 export 𝓈, ms, Hz, kHz
 
 const 𝓈 = Units.s

src/dsp.jl

@@ -6,7 +6,7 @@ import Optim: optimize, minimizer, BFGS
 export fir, removedc, removedc!, demon
 export upconvert, downconvert, rrcosfir, rcosfir
 export mseq, gmseq, circconv, goertzel, pll, hadamard
-export sfilt, sfiltfilt, sresample, mfilter, findsignal
+export mfilter, findsignal
 export istft, whiten, filt, filtfilt, resample, delay!, compose
 
 """
@@ -410,45 +410,40 @@ function pll(x::AbstractVecOrMat, fc=0.0, bandwidth=1e-3; fs=framerate(x))
   signal(y, fs)
 end
 
+sfilt(f::AbstractVector{<:Number}, x::AbstractVector) = signal(filt(f, samples(x)), framerate(x))
+sfilt(f::AbstractVector{<:Number}, x::AbstractVector, si) = signal(filt(f, samples(x), si), framerate(x))
+sfilt(b::AbstractVector{<:Number}, a::Union{Number,AbstractVector}, x::AbstractVector) = signal(filt(b, a, samples(x)), framerate(x))
+sfilt(b::AbstractVector{<:Number}, a::Union{Number,AbstractVector}, x::AbstractVector, si) = signal(filt(b, a, samples(x), si), framerate(x))
+sfiltfilt(coef, x) = signal(filtfilt(coef, samples(x)), framerate(x))
+sresample(x, rate) = signal(resample(samples(x), rate; dims=1), rate * framerate(x))
+sresample(x, rate, coef) = signal(resample(samples(x), rate, coef; dims=1), rate * framerate(x))
+
 """
-    sfilt(f, x[, si])
     filt(f, x::SampledSignal[, si])
-    sfilt(b, a, x[, si])
     filt(b, a, x::SampledSignal[, si])
 
 Same as [`filt`](https://docs.juliadsp.org/stable/filters/#DSP.filt),
 but retains sampling rate information.
 """
-sfilt(f::AbstractVector{<:Number}, x::AbstractVector) = signal(filt(f, samples(x)), framerate(x))
-sfilt(f::AbstractVector{<:Number}, x::AbstractVector, si) = signal(filt(f, samples(x), si), framerate(x))
-sfilt(b::AbstractVector{<:Number}, a::Union{Number,AbstractVector}, x::AbstractVector) = signal(filt(b, a, samples(x)), framerate(x))
-sfilt(b::AbstractVector{<:Number}, a::Union{Number,AbstractVector}, x::AbstractVector, si) = signal(filt(b, a, samples(x), si), framerate(x))
+DSP.filt(f::AbstractVector{<:Number}, x::SampledSignal) = sfilt(f, x)
+DSP.filt(f::AbstractVector{<:Number}, x::SampledSignal, si) = sfilt(f, x, si)
+DSP.filt(b::AbstractVector{<:Number}, a::Union{Number,AbstractVector}, x::SampledSignal) = sfilt(b, a, x)
+DSP.filt(b::AbstractVector{<:Number}, a::Union{Number,AbstractVector}, x::SampledSignal, si) = sfilt(b, a, x, si)
 
 """
-    sfiltfilt(coef, x)
     filtfilt(coef, x::SampledSignal)
 
 Same as [`filtfilt`](https://docs.juliadsp.org/stable/filters/#DSP.Filters.filtfilt),
 but retains sampling rate information.
 """
-sfiltfilt(coef, x) = signal(filtfilt(coef, samples(x)), framerate(x))
+DSP.Filters.filtfilt(coef::AbstractVector{<:Number}, x::SampledSignal) = sfiltfilt(coef, x)
 
 """
-    sresample(x, rate[, coef])
     resample(x::SampledSignal, rate[, coef])
 
 Same as [`resample`](https://docs.juliadsp.org/stable/filters/#DSP.Filters.resample),
 but correctly handles sampling rate conversion.
 """
-sresample(x, rate) = signal(resample(samples(x), rate; dims=1), rate * framerate(x))
-sresample(x, rate, coef) = signal(resample(samples(x), rate, coef; dims=1), rate * framerate(x))
-
-# overload DSP versions of the above functions
-DSP.filt(f::AbstractVector{<:Number}, x::SampledSignal) = sfilt(f, x)
-DSP.filt(f::AbstractVector{<:Number}, x::SampledSignal, si) = sfilt(f, x, si)
-DSP.filt(b::AbstractVector{<:Number}, a::Union{Number,AbstractVector}, x::SampledSignal) = sfilt(b, a, x)
-DSP.filt(b::AbstractVector{<:Number}, a::Union{Number,AbstractVector}, x::SampledSignal, si) = sfilt(b, a, x, si)
-DSP.Filters.filtfilt(coef::AbstractVector{<:Number}, x::SampledSignal) = sfiltfilt(coef, x)
 DSP.Filters.resample(x::SampledSignal, rate::Real) = sresample(x, rate)
 DSP.Filters.resample(x::SampledSignal, rate::Real, coef::Vector) = sresample(x, rate, coef)
 

test/tests-core.jl

@@ -528,28 +528,16 @@ function test_dsp()
 
   x = cw(7kHz, 1s, 44.1kHz)
   f = fir(127, 5kHz, 10kHz; fs=44.1kHz)
-  x1 = sfilt(f, x)
-  @test framerate(x1) == framerate(x)
   x1 = filt(f, x)
   @test framerate(x1) == framerate(x)
-  x1 = sfilt(f, 1, x)
-  @test framerate(x1) == framerate(x)
   x1 = filt(f, 1, x)
   @test framerate(x1) == framerate(x)
-  x1 = sfiltfilt(f, x)
-  @test framerate(x1) == framerate(x)
   x1 = filtfilt(f, x)
   @test framerate(x1) == framerate(x)
-  x1 = sresample(x, 3//2)
-  @test framerate(x1) == 3 * framerate(x) / 2
-  x1 = sresample(x, 3//2, [1,1,1])
-  @test framerate(x1) == 3 * framerate(x) / 2
-  x1 = sresample([x x], 3//2)
-  @test framerate(x1) == 3 * framerate(x) / 2
-  x1 = sresample([x x], 3//2, [1,1,1])
-  @test framerate(x1) == 3 * framerate(x) / 2
   x1 = resample(x, 3//2)
   @test framerate(x1) == 3 * framerate(x) / 2
+  x1 = resample(x, 3//2, [1,1,1])
+  @test framerate(x1) == 3 * framerate(x) / 2
   x1 = resample([x x], 3//2)
   @test framerate(x1) == 3 * framerate(x) / 2
   x1 = resample([x x], 3//2, [1,1,1])