Merge pull request #6 from JuliaStats/ast/fileio_integration

Replace read_rda() by FileIO integration

README.md

@@ -4,4 +4,27 @@
 [![Build Status](https://travis-ci.org/JuliaStats/RData.jl.svg?branch=master)](https://travis-ci.org/JuliaStats/RData.jl)
 [![Build status](https://ci.appveyor.com/api/projects/status/github/JuliaStats/RData.jl?svg=true&branch=master)](https://ci.appveyor.com/project/alyst/rdata-jl/branch/master)
 
-Read R data files (.rda, .RData) to native Julia objects.
+Read R data files (.rda, .RData) and optionally convert the contents into Julia equivalents.
+
+Can read any R data archive, although not all R types could be converted into Julia.
+
+Usage
+-----
+
+To read R objects from "example.rda" file:
+```julia
+using RData
+
+objs = load("path_to/example.rda")
+```
+
+The result is a dictionary of all R objects that are stored in "example.rda".
+
+If `convert=true` keyword option is specified, `load()` will try to automatically
+convert R objects into Julia equivalents:
+ * data frames into `DataFrames.DataFrame`
+ * named vectors into `DictoVec` objects that allow indexing both by element indices and by names
+ * ...
+
+If the conversion to Julia type is not supported (e.g. R closure or language expression),
+the internal RData representation of the object will be provided.

REQUIRE

@@ -1,5 +1,6 @@
 julia 0.4
 DataFrames
 DataArrays
+FileIO
 GZip
 Compat

src/RData.jl

@@ -1,14 +1,15 @@
 module RData
 
-using Compat, DataFrames, GZip
+using Compat, DataFrames, GZip, FileIO
 import DataArrays: data
 import DataFrames: identifier
 import Compat: UTF8String, unsafe_string
+import FileIO: load
 
 export
-    # read_rda,
     sexp2julia,
-    DictoVec
+    DictoVec,
+    load # export FileIO.load()
 
 include("config.jl")
 include("sxtypes.jl")
@@ -37,11 +38,32 @@ include("convert.jl")
 include("context.jl")
 include("readers.jl")
 
-function read_rda(io::IO, kwoptions::Vector{Any})
-    header = chomp(readline(io))
-    @assert header[1] == 'R' # readable header (or RDX2)
-    @assert header[2] == 'D'
-    @assert header[4] == '2'
+##############################################################################
+##
+## FileIO integration
+##
+##############################################################################
+
+# test for RD?2 magic sequence at the beginning of R data input stream
+function detect_rdata(io)
+    read(io, UInt8) == UInt8('R') &&
+    read(io, UInt8) == UInt8('D') &&
+    (fmt = read(io, UInt8); fmt == UInt8('A') || fmt == UInt8('B') || fmt == UInt8('X')) &&
+    read(io, UInt8) == UInt8('2') &&
+    read(io, UInt8) == 0x0A
+end
+
+add_format(format"RData", detect_rdata, [".rdata", ".rda"], [:RData])
+
+function load(f::File{format"RData"}; kwoptions...)
+    gzopen(filename(f)) do s
+        load(Stream(f, s), kwoptions)
+    end
+end
+
+function load(s::Stream{format"RData"}, kwoptions::Vector{Any})
+    io = stream(s)
+    @assert detect_rdata(io)
     ctx = RDAContext(rdaio(io, chomp(readline(io))), kwoptions)
     @assert ctx.fmtver == 2    # format version
 #    println("Written by R version $(ctx.Rver)")
@@ -69,9 +91,6 @@ function read_rda(io::IO, kwoptions::Vector{Any})
     return res
 end
 
-read_rda(io::IO; kwoptions...) = read_rda(io, kwoptions)
-
-read_rda(fnm::AbstractString; kwoptions...) = gzopen(fnm) do io read_rda(io, kwoptions) end
-
+load(s::Stream{format"RData"}; kwoptions...) = load(s, kwoptions)
 
 end # module

test/RDA.jl

@@ -14,7 +14,7 @@ module TestRDA
         # df['chr'] = c('ab', 'c')
         # df['factor'] = factor(df$chr)
         # df['cplx'] = complex( real=c(1.1,0.0), imaginary=c(0.5,1.0) )
-        # #utf=c('Ж', '∰')) R handles it, RData.read_rda doesn't.
+        # #utf=c('Ж', '∰')) R handles it, RData doesn't.
         # save(df, file='types.rda')
         # save(df, file='types_ascii.rda', ascii=TRUE)
 
@@ -57,39 +57,39 @@ module TestRDA
     testdir = dirname(@__FILE__)
 
     df = DataFrame(num = [1.1, 2.2])
-    @test isequal(sexp2julia(RData.read_rda("$testdir/data/minimal.rda",convert=false)["df"]), df)
-    @test isequal(RData.read_rda("$testdir/data/minimal.rda",convert=true)["df"], df)
-    @test isequal(open(RData.read_rda,"$testdir/data/minimal_ascii.rda")["df"], df)
+    @test isequal(sexp2julia(load("$testdir/data/minimal.rda",convert=false)["df"]), df)
+    @test isequal(load("$testdir/data/minimal.rda",convert=true)["df"], df)
+    @test isequal(load("$testdir/data/minimal_ascii.rda")["df"], df)
 
     df[:int] = Int32[1, 2]
     df[:logi] = [true, false]
     df[:chr] = ["ab", "c"]
     df[:factor] = pool(df[:chr])
     df[:cplx] = Complex128[1.1+0.5im, 1.0im]
-    @test isequal(sexp2julia(RData.read_rda("$testdir/data/types.rda",convert=false)["df"]), df)
-    @test isequal(sexp2julia(RData.read_rda("$testdir/data/types_ascii.rda",convert=false)["df"]), df)
+    @test isequal(sexp2julia(load("$testdir/data/types.rda",convert=false)["df"]), df)
+    @test isequal(sexp2julia(load("$testdir/data/types_ascii.rda",convert=false)["df"]), df)
 
     df[2, :] = NA
     append!(df, df[2, :])
     df[3, :num] = NaN
     df[:, :cplx] = @data [NA, @compat(Complex128(1,NaN)), NaN]
-    @test isequal(sexp2julia(RData.read_rda("$testdir/data/NAs.rda",convert=false)["df"]), df)
+    @test isequal(sexp2julia(load("$testdir/data/NAs.rda",convert=false)["df"]), df)
     # ASCII format saves NaN as NA
     df[3, :num] = NA
     df[:, :cplx] = @data [NA, NA, NA]
-    @test isequal(sexp2julia(RData.read_rda("$testdir/data/NAs_ascii.rda",convert=false)["df"]), df)
+    @test isequal(sexp2julia(load("$testdir/data/NAs_ascii.rda",convert=false)["df"]), df)
 
-    rda_names = names(sexp2julia(RData.read_rda("$testdir/data/names.rda",convert=false)["df"]))
+    rda_names = names(sexp2julia(load("$testdir/data/names.rda",convert=false)["df"]))
     expected_names = [:_end, :x!, :x1, :_B_C_, :x, :x_1]
     @test rda_names == expected_names
-    rda_names = names(sexp2julia(RData.read_rda("$testdir/data/names_ascii.rda",convert=false)["df"]))
+    rda_names = names(sexp2julia(load("$testdir/data/names_ascii.rda",convert=false)["df"]))
     @test rda_names == [:_end, :x!, :x1, :_B_C_, :x, :x_1]
 
-    rda_envs = RData.read_rda("$testdir/data/envs.rda",convert=false)
+    rda_envs = load("$testdir/data/envs.rda",convert=false)
 
-    rda_pairlists = RData.read_rda("$testdir/data/pairlists.rda",convert=false)
+    rda_pairlists = load("$testdir/data/pairlists.rda",convert=false)
 
-    rda_closures = RData.read_rda("$testdir/data/closures.rda",convert=false)
+    rda_closures = load("$testdir/data/closures.rda",convert=false)
 
-    rda_cmpfuns = RData.read_rda("$testdir/data/cmpfun.rda",convert=false)
+    rda_cmpfuns = load("$testdir/data/cmpfun.rda",convert=false)
 end