Handling of NAs

[piccolbo]

Hi,
I have a data frame containing some NAs in one colum, When I write it out and read it back in, that column contains only NAs. What gives?

test case

df = 
structure(list(x = c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1), y = structure(1:10, .Label = c("1", 
"2", "3", "4", "5", "6", "7", "8", "9", "10"), class = "factor"), 
    fac = structure(c(1L, NA, NA, 2L, 2L, 1L, NA, 2L, 1L, 1L), .Label = c("b", 
    "c"), class = "factor")), .Names = c("x", "y", "fac"), row.names = c(NA, 
-10L), class = "data.frame")

write.avro(df, "/tmp/testxx.avro")
read.avro("/tmp/testxx.avro")
   x  y  fac
1  1  1 <NA>
2  1  2 <NA>
3  1  3 <NA>
4  1  4 <NA>
5  1  5 <NA>
6  1  6 <NA>
7  1  7 <NA>
8  1  8 <NA>
9  1  9 <NA>
10 1 10 <NA>

Thanks

[jamiefolson]

That is a very good question. An NA factor likely require serialization as
a union of an enum with null, which may be either serialized or serialized
incorrectly. I made some simplifying assumptions that were sufficient at
the time, but we may need to extend the logic around unions.
On Oct 20, 2014 7:24 PM, "Antonio Piccolboni" [email protected]
wrote:

Hi,
I have a data frame containing some NAs in one colum, When I write it out
and read it back in, that column contains only NAs. What gives?

test case

df =
structure(list(x = c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1), y = structure(1:10, .Label = c("1",
"2", "3", "4", "5", "6", "7", "8", "9", "10"), class = "factor"),
fac = structure(c(1L, NA, NA, 2L, 2L, 1L, NA, 2L, 1L, 1L), .Label = c("b",
"c"), class = "factor")), .Names = c("x", "y", "fac"), row.names = c(NA,
-10L), class = "data.frame")

write.avro(df, "/tmp/testxx.avro")
read.avro("/tmp/testxx.avro")

—
Reply to this email directly or view it on GitHub
#1.