projects / epclust.git / blame
| Commit | Line | Data |
|---|---|---|
| 56857861 | 1 | context("de_serialize") |
| 62deb424 | 2 | |
| 56857861 | 3 | test_that("serialization + getDataInFile retrieve original data / from matrix", |
| 62deb424 | 4 | { |
| 4bcfdbee BA |
5 | data_bin_file = "/tmp/epclust_test_m.bin" |
| 6 | unlink(data_bin_file) | |
| 7 | ||
| 56857861 BA |
8 | #dataset 200 lignes / 30 columns |
| 9 | data_ascii = matrix(runif(200*30,-10,10),ncol=30) | |
| 10 | nbytes = 4 #lead to a precision of 1e-7 / 1e-8 | |
| 11 | endian = "little" | |
| 12 | ||
| 13 | #Simulate serialization in one single call | |
| 4bcfdbee | 14 | binarize(data_ascii, data_bin_file, 500, ",", nbytes, endian) |
| 56857861 BA |
15 | expect_equal(file.info(data_bin_file)$size, length(data_ascii)*nbytes+8) |
| 16 | for (indices in list(c(1,3,5), 3:13, c(5,20,50), c(75,130:135), 196:200)) | |
| 62deb424 | 17 | { |
| 56857861 BA |
18 | data_lines = getDataInFile(indices, data_bin_file, nbytes, endian) |
| 19 | expect_equal(data_lines, data_ascii[indices,], tolerance=1e-6) | |
| 62deb424 | 20 | } |
| 56857861 BA |
21 | unlink(data_bin_file) |
| 22 | ||
| 23 | #...in several calls (last call complete, next call NULL) | |
| 24 | for (i in 1:20) | |
| 4bcfdbee | 25 | binarize(data_ascii[((i-1)*10+1):(i*10),], data_bin_file, 20, ",", nbytes, endian) |
| 56857861 BA |
26 | expect_equal(file.info(data_bin_file)$size, length(data_ascii)*nbytes+8) |
| 27 | for (indices in list(c(1,3,5), 3:13, c(5,20,50), c(75,130:135), 196:200)) | |
| 62deb424 | 28 | { |
| 56857861 BA |
29 | data_lines = getDataInFile(indices, data_bin_file, nbytes, endian) |
| 30 | expect_equal(data_lines, data_ascii[indices,], tolerance=1e-6) | |
| 62deb424 | 31 | } |
| 56857861 BA |
32 | unlink(data_bin_file) |
| 33 | }) | |
| 34 | ||
| 35 | test_that("serialization + getDataInFile retrieve original data / from connection", | |
| 36 | { | |
| 4bcfdbee BA |
37 | data_bin_file = "/tmp/epclust_test_c.bin" |
| 38 | unlink(data_bin_file) | |
| 39 | ||
| 56857861 BA |
40 | #dataset 300 lignes / 50 columns |
| 41 | data_csv = system.file("testdata","de_serialize.csv",package="epclust") | |
| 42 | nbytes = 8 | |
| 43 | endian = "big" | |
| 4bcfdbee | 44 | data_ascii = as.matrix(read.csv(data_csv, sep=";", header=FALSE)) #for ref |
| 56857861 BA |
45 | |
| 46 | #Simulate serialization in one single call | |
| 4bcfdbee | 47 | binarize(data_csv, data_bin_file, 350, ";", nbytes, endian) |
| 56857861 BA |
48 | expect_equal(file.info(data_bin_file)$size, 300*50*8+8) |
| 49 | for (indices in list(c(1,3,5), 3:13, c(5,20,50), c(75,130:135), 196:200)) | |
| 62deb424 | 50 | { |
| 56857861 BA |
51 | #HACK: as.matrix(as.data.frame( )) required to match (ref) data structure |
| 52 | data_lines = as.matrix(as.data.frame( getDataInFile(indices,data_bin_file,nbytes,endian) )) | |
| 53 | expect_equal(data_lines, data_ascii[indices,]) | |
| 62deb424 | 54 | } |
| 56857861 BA |
55 | unlink(data_bin_file) |
| 56 | ||
| 57 | #...in several calls / chunks of 29 --> 29*10 + 10, incomplete last | |
| 58 | data_con = file(data_csv, "r") | |
| 4bcfdbee | 59 | binarize(data_con, data_bin_file, 29, ";", nbytes, endian) |
| 56857861 BA |
60 | expect_equal(file.info(data_bin_file)$size, 300*50*8+8) |
| 61 | for (indices in list(c(1,3,5), 3:13, c(5,20,50), c(75,130:135), 196:200)) | |
| 62deb424 | 62 | { |
| 56857861 BA |
63 | data_lines = as.matrix(as.data.frame( getDataInFile(indices,data_bin_file,nbytes,endian) )) |
| 64 | expect_equal(data_lines, data_ascii[indices,]) | |
| 62deb424 | 65 | } |
| 56857861 | 66 | unlink(data_bin_file) |
| 4bcfdbee | 67 | #close(data_con) --> done in binarize() |
| 56857861 | 68 | }) |