Merge 4c2edd4 into ebc1335

JuliaData · Nov 21, 2017 · 186d1d6 · 186d1d6
2 parents ebc1335 + 4c2edd4
commit 186d1d6
Show file tree

Hide file tree

Showing 17 changed files with 105 additions and 2 deletions.
diff --git a/src/RData.jl b/src/RData.jl
@@ -9,7 +9,8 @@ import FileIO: load
 export
     sexp2julia,
     DictoVec,
-    load # export FileIO.load()
+    load, # export FileIO.load()
+    readRDS
 
 include("config.jl")
 include("sxtypes.jl")
@@ -87,4 +88,28 @@ end
 
 load(s::Stream{format"RData"}; kwoptions...) = load(s, kwoptions)
 
+# TODO:
+# * maybe throw error instead of warning on conversion?
+# * tests
+# * load stuff (e.g. FileIO req on detect_rdata)
+# * maybe return tuple of (object, attribute_dict) for
+#   https://github.com/JuliaStats/RData.jl/issues/30
+function readRDS(f::AbstractString; kwoptions...)
+    io = open(f, "r")
+    try
+        gzipped = read(io, UInt8) == 0x1F && read(io, UInt8) == 0x8B # check GZip magic number
+        seekstart(io)
+        # if compressed, transcode gzipped stream
+        gzipped && (io = GzipDecompressorStream(io))
+        ctx = RDAContext(rdaio(io, chomp(readline(io))), kwoptions)
+        @assert ctx.fmtver == 2    # format version
+        convert2julia = get(ctx.kwdict,:convert,true)
+        return convert2julia ? sexp2julia(readitem(ctx)) : readitem(ctx)
+    catch
+        rethrow()
+    finally
+        close(io)
+    end
+end
+
 end # module
diff --git a/test/RDS.jl b/test/RDS.jl
@@ -0,0 +1,65 @@
+module TestRDS
+    using Base.Test
+    using DataFrames
+    using RData
+
+    testdir = dirname(@__FILE__)
+
+    @testset "RDS: Reading minimal rds" begin
+        df = DataFrame(num = [1.1, 2.2])
+        @test isequal(sexp2julia(readRDS("$testdir/data/minimal.rds",convert=false))["df"], df)
+        @test isequal(readRDS("$testdir/data/minimal.rds",convert=true)["df"], df)
+        @test isequal(readRDS("$testdir/data/minimal_ascii.rds")["df"], df)
+    end
+
+    @testset "RDS: Conversion to Julia types" begin
+        df = DataFrame(num = [1.1, 2.2],
+                       int = Int32[1, 2],
+                       logi = [true, false],
+                       chr = ["ab", "c"],
+                       factor = pool(["ab", "c"]),
+                       cplx = Complex128[1.1+0.5im, 1.0im])
+        rdf = sexp2julia(readRDS("$testdir/data/types.rds",convert=false))["df"]
+        @test eltypes(rdf) == eltypes(df)
+        @test isequal(rdf, df)
+        rdf_ascii = sexp2julia(readRDS("$testdir/data/types_ascii.rds",convert=false))["df"]
+        @test eltypes(rdf_ascii) == eltypes(df)
+        @test isequal(rdf_ascii, df)
+    end
+
+
+    @testset "RDS: NAs conversion" begin
+        df = DataFrame(num = [1.1, 2.2],
+                       int = Int32[1, 2],
+                       logi = [true, false],
+                       chr = ["ab", "c"],
+                       factor = pool(["ab", "c"]),
+                       cplx = Complex128[1.1+0.5im, 1.0im])
+
+        df[2, :] = NA
+        append!(df, df[2, :])
+        df[3, :num] = NaN
+        df[:, :cplx] = @data [NA, Complex128(1,NaN), NaN]
+        @test isequal(sexp2julia(readRDS("$testdir/data/NAs.rds",convert=false))["df"], df)
+        # ASCII format saves NaN as NA
+        df[3, :num] = NA
+        df[:, :cplx] = @data [NA, NA, NA]
+        @test isequal(sexp2julia(readRDS("$testdir/data/NAs_ascii.rds",convert=false))["df"], df)
+    end
+
+    @testset "RDS: Column names conversion" begin
+        rds_names = names(sexp2julia(readRDS("$testdir/data/names.rds",convert=false))["df"])
+        expected_names = [:_end, :x!, :x1, :_B_C_, :x, :x_1]
+        @test rds_names == expected_names
+        rds_names = names(sexp2julia(readRDS("$testdir/data/names_ascii.rds",convert=false))["df"])
+        @test rds_names == [:_end, :x!, :x1, :_B_C_, :x, :x_1]
+    end
+
+    @testset "RDS: Reading RDA with complex types (environments, closures etc)" begin
+        rds_envs = readRDS("$testdir/data/envs.rds",convert=false)
+        rds_pairlists = readRDS("$testdir/data/pairlists.rds",convert=false)
+        rds_closures = readRDS("$testdir/data/closures.rds",convert=false)
+        rds_cmpfuns = readRDS("$testdir/data/cmpfun.rds",convert=false)
+    end
+end
+
diff --git a/test/data/NAs.rds b/test/data/NAs.rds
diff --git a/test/data/NAs_ascii.rds b/test/data/NAs_ascii.rds
diff --git a/test/data/closures.rds b/test/data/closures.rds
diff --git a/test/data/cmpfun.rds b/test/data/cmpfun.rds
diff --git a/test/data/envs.rds b/test/data/envs.rds
diff --git a/test/data/envs_ascii.rds b/test/data/envs_ascii.rds
diff --git a/test/data/minimal.rds b/test/data/minimal.rds
diff --git a/test/data/minimal_ascii.rds b/test/data/minimal_ascii.rds
diff --git a/test/data/names.rds b/test/data/names.rds
diff --git a/test/data/names_ascii.rds b/test/data/names_ascii.rds
diff --git a/test/data/pairlists.rds b/test/data/pairlists.rds
diff --git a/test/data/types.rds b/test/data/types.rds
diff --git a/test/data/types_ascii.rds b/test/data/types_ascii.rds
diff --git a/test/generate_rda.R b/test/generate_rda.R
@@ -1,4 +1,4 @@
-# R script to generate test .rda files
+# R script to generate test .rda and .rds files
 
 df <- data.frame(num = c(1.1, 2.2))
 save(df, file = "data/minimal.rda")
@@ -48,3 +48,15 @@ test.cmpfun0 <- cmpfun( test.fun0 )
 test.cmpfun1 <- cmpfun( test.fun1 )
 test.cmpfun2 <- cmpfun( test.fun2 )
 save(test.cmpfun0, test.cmpfun1, test.cmpfun2, file = "data/cmpfun.rda")
+
+
+# for converting rda files to rds to test with readRDS
+rdafiles = list.files("data/", pattern="*.rda", full.names=T)
+for (rdafile in rdafiles) {
+    en = new.env()
+    load(rdafile, envir=en)
+    lst = as.list(en)
+    rdsfile = gsub("\\.rda$", ".rds", rdafile)
+    saveRDS(lst, file=rdsfile)
+}
+
diff --git a/test/runtests.jl b/test/runtests.jl
@@ -2,3 +2,4 @@ using RData
 using Base.Test
 
 include("RDA.jl")
+include("RDS.jl")