xts character 2nd subsetting a zero-width xts returns integer(0) instead of Date of length 0

[AndreMikulec]

Hello,

Given a a zero-width xts, xts character 2nd subsetting returns integer(0) instead of Date of length 0. For example:

library(xts)
testxts <- xts(1,as.Date('2018-07-18'))

# no dates in 2020

> testxts['2020']
     [,1]

# is zero width

> str(testxts['2020'])
An 'xts' object of zero-width

# is Date of length 0
# (no dates in 2020)

> index(testxts['2020'])
Date of length 0

# no dates in 2020 and 2021

> testxts['2020']['2021']
     [,1]

> str(testxts['2020']['2021'])
An 'xts' object of zero-width

# but index(.) should be 'Date of length 0'

> index(testxts['2020']['2021'])
integer(0)

Maybe a solution is this
In xts/R/index.R

fixInNamespace(x = "index.xts", ns = "xts")

replace
    if (length(x.index) == 0)
        return(integer())
by
    if (length(x.index) == 0)
        # return(integer())
        return(x.index)

SAVE notepad

> index(testxts['2020'])
Date of length 0

# So this works but the index class is wrong

> index(testxts['2020']['2021'])
POSIXct of length 0

fixInNamespace(x = "[.xts", ns = "xts")

replace
        x.tmp <- .xts(rep(NA,length(i)), .index(x)[i], dimnames=list(NULL, colnames(x)))

by ( replace .xts using xts and replace .index using index )
        # x.tmp <- .xts(rep(NA,length(i)), .index(x)[i], dimnames=list(NULL, colnames(x)))
          x.tmp <-  xts(rep(NA,length(i)),  index(x)[i], dimnames=list(NULL, colnames(x)))

SAVE notepad

So now this works

> index(testxts['2020'])
Date of length 0

> index(testxts['2020']['2021'])
Date of length 0

So now all of these cases work

> str(testxts['2018-07-18'])
An 'xts' object on 2018-07-18/2018-07-18 containing:
  Data: num [1, 1] 1
  Indexed by objects of class: [Date] TZ: UTC
  xts Attributes:
 NULL

> index(testxts['2018-07-18'])
[1] "2018-07-18"

> class(index(testxts['2018-07-18']))
[1] "Date"

> str(testxts[zoo::as.Date('2018-07-30')])
An 'xts' object of zero-width

> index(testxts[zoo::as.Date('2018-07-30')])
Date of length 0

> class(index(testxts[zoo::as.Date('2018-07-30')]))
[1] "Date"

> str(testxts[zoo::as.Date('2018-07-30')][zoo::as.Date('2018-07-18')])
An 'xts' object of zero-width

> index(testxts[zoo::as.Date('2018-07-30')][zoo::as.Date('2018-07-18')])
Date of length 0

> class(index(testxts[zoo::as.Date('2018-07-30')][zoo::as.Date('2018-07-18')]))
[1] "Date"

> str(testxts[zoo::as.Date('2018-07-18')][zoo::as.Date('2018-07-30')])
An 'xts' object of zero-width

> index(testxts[zoo::as.Date('2018-07-30')][zoo::as.Date('2018-07-18')])
Date of length 0

> class(index(testxts[zoo::as.Date('2018-07-30')][zoo::as.Date('2018-07-18')]))
[1] "Date"

> str(testxts[c(zoo::as.Date('2018-07-30'), zoo::as.Date('2018-07-18'))])
An 'xts' object on 2018-07-18/2018-07-18 containing:
  Data: num [1, 1] 1
  Indexed by objects of class: [Date] TZ: UTC
  xts Attributes:
 NULL

> index(testxts[c(zoo::as.Date('2018-07-30'), zoo::as.Date('2018-07-18'))])
[1] "2018-07-18"

> class(index(testxts[c(zoo::as.Date('2018-07-30'), zoo::as.Date('2018-07-18'))]))
[1] "Date"
>

> str(testxts['2018-07-30'])
An 'xts' object of zero-width

> index(testxts['2018-07-30'])
Date of length 0

> class(index(testxts['2018-07-30']))
[1] "Date"

> str(testxts['2018-07-18']['2018-07-30'])
An 'xts' object of zero-width

> index(testxts['2018-07-18']['2018-07-30'])
Date of length 0

> class(index(testxts['2018-07-18']['2018-07-30']))
[1] "Date"

> devtools::session_info()
Session info ------------------------------------------------------------------
 setting  value
 version  R version 3.5.1 (2018-07-02)
 system   x86_64, mingw32
 ui       RTerm
 language (EN)
 collate  English_United States.1252
 tz       America/Chicago
 date     2018-07-22

Packages ----------------------------------------------------------------------
 package    * version date       source
 base       * 3.5.1   2018-07-02 local
 compiler     3.5.1   2018-07-02 local
 datasets   * 3.5.1   2018-07-02 local
 devtools     1.13.6  2018-06-27 CRAN (R 3.5.1)
 digest       0.6.15  2018-01-28 CRAN (R 3.5.0)
 graphics   * 3.5.1   2018-07-02 local
 grDevices  * 3.5.1   2018-07-02 local
 grid         3.5.1   2018-07-02 local
 lattice      0.20-35 2017-03-25 CRAN (R 3.5.0)
 magrittr   * 1.5     2014-11-22 CRAN (R 3.5.0)
 memoise      1.1.0   2017-04-21 CRAN (R 3.5.0)
 methods    * 3.5.1   2018-07-02 local
 rstudioapi   0.7     2017-09-07 CRAN (R 3.5.0)
 stats      * 3.5.1   2018-07-02 local
 tools        3.5.1   2018-07-02 local
 utils      * 3.5.1   2018-07-02 local
 withr        2.1.2   2018-04-27 Github (jimhester/withr@79d7b0d)
 xts        * 0.11-0  2018-07-16 CRAN (R 3.5.1)
 zoo        * 1.8-3   2018-07-16 CRAN (R 3.5.1)
>

[joshuaulrich]

Please do not open a new issue to discuss a solution proposed in another issue. That splits the discussion into multiple places, which makes it harder to keep track of the relevant details.

[AndreMikulec]

This issue is a different problem than what is in
"
Can't subset empty xts by date anymore #252
#252
"

The problem presented in this issue is the following

 the index(.) result should be 'Date of length 0' and not integer(0) and not POSIXct of length 0.

> str(testxts['2020']['2021'])
An 'xts' object of zero-width

# but index(.) should be 'Date of length 0'

> index(testxts['2020']['2021'])
integer(0)

So, this issue, presents two(2) one line fixes( replace->by) (replace->by)
and issue
"
Can't subset empty xts by date anymore #252
#252
"
presents another one(1) line fix(replace->by).

[joshuaulrich]

My apologies. It wasn't clear to me that this issue did not depend on the suggested solution in the other issue you referenced.

I deleted the comments about the other issue from your initial post, so others (including future me) aren't confused by them.

[joshuaulrich]

Note that this behavior is consistent with v0.10-2:

library(xts)
packageVersion("xts")
# [1] '0.10.2'
testxts <- xts(1, as.Date("2018-07-18"))
index(testxts["2020"])
# Date of length 0
index(testxts["2020"]["2020"])
# integer(0)

[joshuaulrich]

I consider this a feature request, since the current behavior is consistent with v0.10-2. That makes this a lower priority than the bugs caused by regressions in v0.11-0.