NBE
Reputation: 651
Filtering dates for time series plot using dplyr
I'm trying to filter my dataset to only have rows from the year 1996-2015 and only the columns orgid,stdate,locid,charnam. The entire dataset contains years from 1988 to 2015 and a great deal of different columns. I just recently learned about the dplyr package and figured this would be the way to go. However I keep getting this error and I don't understand why I'm getting it. The error I keep getting is Error in is_character(x) : object 'rlang_is_character' not found
This is my code so far:
########## download necessary packages to make script run #########################################################################
if (!require(pacman)) {
install.packages('pacman')
}
pacman::p_load("tidyverse")
#### Read in the necessary data ######
roadsalt_data<-read.table("QADportaldata_1988-2015.tsv",header=T,sep="\t",fill=T,stringsAsFactors = F)
#Convert date column from a character class to a date class so ggplot can display as a continuous variable ###
roadsalt_data$stdate <- as.Date(roadsalt_data$stdate)
## Filter dataset to only contain years 1996-2015 ########
roadsalt_data_sub<-roadsalt_data %>%
select(orgid,stdate,locid,charnam) %>%
filter(between(roadsalt_data, as.Date("1996-01-01"), as.Date("2015-07-01")))
Heres a preview of the dataset:
structure(list(orgid = c("USGS-NJ", "USGS-NJ", "USGS-NJ", "USGS-NJ",
"USGS-NJ", "USGS-NJ", "USGS-NJ", "USGS-NJ", "USGS-NJ", "USGS-NJ"
), actid = c("nwisnj.01.01300274", "nwisnj.01.01300273", "nwisnj.01.01300247",
"nwisnj.01.01300242", "nwisnj.01.01300238", "nwisnj.01.01300237",
"nwisnj.01.01300189", "nwisnj.01.01300189", "nwisnj.01.01300189",
"nwisnj.01.01300190"), actyp = c("Sample-Routine", "Sample-Routine",
"Sample-Routine", "Sample-Routine", "Sample-Routine", "Sample-Routine",
"Sample-Routine", "Sample-Routine", "Sample-Routine", "Sample-Routine"
), stdate = structure(c(15755, 15755, 15748, 15748, 15748, 15748,
15740, 15740, 15740, 15740), class = "Date"), sttime = c("09:30:00",
"11:00:00", "10:30:00", "12:00:00", "11:00:00", "11:30:00", "09:25:00",
"09:25:00", "09:25:00", "09:30:00"), actdep = c(NA_real_, NA_real_,
NA_real_, NA_real_, NA_real_, NA_real_, NA_real_, NA_real_, NA_real_,
NA_real_), actdepun = c("", "", "", "", "", "", "", "", "", ""
), locid = c("USGS-01407760", "USGS-01445030", "USGS-01380075",
"USGS-01368820", "USGS-01409815", "USGS-01411400", "USGS-01458570",
"USGS-01458570", "USGS-01458570", "USGS-01445160"), actcom = c("A-0520044 TPCN = 64mL filtered",
"A-0520046 TPCN = 124mL filtered", "A-0460036 TPCN = 124mL filtered L-0460036 Received February 14, 2013",
"A-0460025 TPCN = 125mL filtered L-0460025 Received February 14, 2013",
"A-0460027 TPCN = 64mL filtered. ATTN: H.Ardourel, LL ANC and LL pH L-0460027 Received February 14, 2013",
"A-0460028 TPCN = 125mL filtered. L-0460028 Received February 14, 2013",
"A-0370012 TPCN = 125mL filtered", "A-0370012 TPCN = 125mL filtered",
"A-0370012 TPCN = 125mL filtered", "A-0370011 TPCN = 125mL filtered"
), hydcond = c("Stable, normal stage", "Stable, normal stage",
"Stable, normal stage", "Stable, normal stage", "Stable, high stage",
"Falling stage", "Stable, normal stage", "Stable, normal stage",
"Stable, normal stage", "Stable, high stage"), hydev = c("Routine sample",
"Routine sample", "Routine sample", "Routine sample", "Routine sample",
"Routine sample", "Routine sample", "Routine sample", "Routine sample",
"Routine sample"), metcont = c("USGS parameter code 82398", "USGS parameter code 82398",
"USGS parameter code 82398", "USGS parameter code 82398", "USGS parameter code 82398",
"USGS parameter code 82398", "USGS parameter code 82398", "USGS parameter code 82398",
"USGS parameter code 82398", "USGS parameter code 82398"), metnam = c("Multiple verticals",
"Multiple verticals", "Multiple verticals", "Multiple verticals",
"Single vertical", "Multiple verticals", "Grab sample (dip)",
"Grab sample (dip)", "Grab sample (dip)", "Multiple verticals"
), detcond = c("", "", "", "", "Not Detected", "", "", "", "",
""), charnam = c("Nitrogen, mixed forms (NH3), (NH4), organic, (NO2) and (NO3)",
"Nitrogen, mixed forms (NH3), (NH4), organic, (NO2) and (NO3)",
"Nitrogen, mixed forms (NH3), (NH4), organic, (NO2) and (NO3)",
"Nitrogen, mixed forms (NH3), (NH4), organic, (NO2) and (NO3)",
"Nitrogen, mixed forms (NH3), (NH4), organic, (NO2) and (NO3)",
"Nitrogen, mixed forms (NH3), (NH4), organic, (NO2) and (NO3)",
"Inorganic nitrogen (nitrate and nitrite)", "Phosphorus", "Phosphorus",
"Kjeldahl nitrogen"), samfrac = c("Dissolved", "Dissolved", "Dissolved",
"Dissolved", "Dissolved", "Dissolved", "Dissolved", "Total",
"Dissolved", "Dissolved"), val = c("0.84", "1.2", "0.46", "0.28",
"", "0.66", "3.10", "0.032", "0.028", "0.21"), valunit = c("mg/l",
"mg/l", "mg/l", "mg/l", "", "mg/l", "mg/l as N", "mg/l as P",
"mg/l as P", "mg/l as N"), valqual = c("", "", "", "", "", "",
"", "", "", ""), valstat = c("Accepted", "Accepted", "Accepted",
"Accepted", "Accepted", "Accepted", "Accepted", "Accepted", "Accepted",
"Accepted"), statcode = c("", "", "", "", "", "", "", "", "",
""), valtype = c("Actual", "Actual", "Actual", "Actual", "Actual",
"Actual", "Actual", "Actual", "Actual", "Actual"), precval = c(NA_real_,
NA_real_, NA_real_, NA_real_, NA_real_, NA_real_, NA_real_, NA_real_,
NA_real_, NA_real_), valcom = c("", "", "", "", "", "", "Report level code updated Oct., Nov. 2015. Reference: NWQL Rapi-Note 2011-21 (RLC: IRL => LT-MDL)",
"", "", ""), valdep = c(NA_real_, NA_real_, NA_real_, NA_real_,
NA_real_, NA_real_, NA_real_, NA_real_, NA_real_, NA_real_),
valdepun = c("", "", "", "", "", "", "", "", "", ""), valmetnam = c("Computation by NWIS algorithm",
"Computation by NWIS algorithm", "Computation by NWIS algorithm",
"Computation by NWIS algorithm", "Computation by NWIS algorithm",
"Computation by NWIS algorithm", "NO3+NO2, wf, FCC,NaR, DA",
"P, wu, WCA, persulfate CF color", "P, wf, FCA, persulfate CF color",
"NH3+org-N, wf, FCA, Kjeldahl, CF"), metdesc = c("NWIS User's Manual, QW System, Section 3.6.7",
"NWIS User's Manual, QW System, Section 3.6.7", "NWIS User's Manual, QW System, Section 3.6.7",
"NWIS User's Manual, QW System, Section 3.6.7", "NWIS User's Manual, QW System, Section 3.6.7",
"NWIS User's Manual, QW System, Section 3.6.7", "", "", "",
"USGS OF 00-170"), labcom = c("", "", "", "", "", "", "",
"", "", ""), detlimnam = c("", "", "", "", "Historical Lower Reporting Limit",
"", "Long Term Method Detection Level", "Long Term Method Detection Level",
"Long Term Method Detection Level", "Long Term Method Detection Level"
), detlimval = c("", "", "", "", "0.23", "", "0.04", "0.004",
"0.0040", "0.07"), detlimun = c("", "", "", "", "mg/l", "",
"mg/l as N", "mg/l as P", "mg/l as P", "mg/l as N"), V63 = c("NWIS",
"NWIS", "NWIS", "NWIS", "NWIS", "NWIS", "NWIS", "NWIS", "NWIS",
"NWIS")), .Names = c("orgid", "actid", "actyp", "stdate",
"sttime", "actdep", "actdepun", "locid", "actcom", "hydcond",
"hydev", "metcont", "metnam", "detcond", "charnam", "samfrac",
"val", "valunit", "valqual", "valstat", "statcode", "valtype",
"precval", "valcom", "valdep", "valdepun", "valmetnam", "metdesc",
"labcom", "detlimnam", "detlimval", "detlimun", "V63"), row.names = c(NA,
10L), class = "data.frame")
Any help would be greatly appreciated! Thanks in advance!
Upvotes: 1
Views: 2900
Answers (2)
knapply
Reputation: 667
If I remember correctly, between() didn't work with Dates at one point, even when the left and right arguments were converted with as.Date().
Here are some alternatives. Since all of your sample data fall between the specified years, these all filter for dates between 2013-02-04 and 2013-02-12. Adjust accordingly.
library(dplyr)
roadsalt_data <- as_tibble(roadsalt_data) # not necessary, just convenient console output
roadsalt_data %>%
select(orgid, stdate, locid, charnam) %>%
filter(stdate >= "2013-02-04", stdate <= "2013-02-12")
#> # A tibble: 8 x 4
#> orgid stdate locid charnam
#> <chr> <date> <chr> <chr>
#> 1 USGS-NJ 2013-02-12 USGS-01380075 Nitrogen, mixed forms (NH3), (NH4), or~
#> 2 USGS-NJ 2013-02-12 USGS-01368820 Nitrogen, mixed forms (NH3), (NH4), or~
#> 3 USGS-NJ 2013-02-12 USGS-01409815 Nitrogen, mixed forms (NH3), (NH4), or~
#> 4 USGS-NJ 2013-02-12 USGS-01411400 Nitrogen, mixed forms (NH3), (NH4), or~
#> 5 USGS-NJ 2013-02-04 USGS-01458570 Inorganic nitrogen (nitrate and nitrit~
#> 6 USGS-NJ 2013-02-04 USGS-01458570 Phosphorus
#> 7 USGS-NJ 2013-02-04 USGS-01458570 Phosphorus
#> 8 USGS-NJ 2013-02-04 USGS-01445160 Kjeldahl nitrogen
roadsalt_data %>%
select(orgid, stdate, locid, charnam) %>%
filter(between(stdate, as.Date("2013-02-04"), as.Date("2013-02-12")))
#> # A tibble: 8 x 4
#> orgid stdate locid charnam
#> <chr> <date> <chr> <chr>
#> 1 USGS-NJ 2013-02-12 USGS-01380075 Nitrogen, mixed forms (NH3), (NH4), or~
#> 2 USGS-NJ 2013-02-12 USGS-01368820 Nitrogen, mixed forms (NH3), (NH4), or~
#> 3 USGS-NJ 2013-02-12 USGS-01409815 Nitrogen, mixed forms (NH3), (NH4), or~
#> 4 USGS-NJ 2013-02-12 USGS-01411400 Nitrogen, mixed forms (NH3), (NH4), or~
#> 5 USGS-NJ 2013-02-04 USGS-01458570 Inorganic nitrogen (nitrate and nitrit~
#> 6 USGS-NJ 2013-02-04 USGS-01458570 Phosphorus
#> 7 USGS-NJ 2013-02-04 USGS-01458570 Phosphorus
#> 8 USGS-NJ 2013-02-04 USGS-01445160 Kjeldahl nitrogen
# How I would've done it
library(lubridate)
roadsalt_data %>%
select(orgid, stdate, locid, charnam) %>%
# filter(between(year(stdate), 1996, 2015)) # for years instead of days
filter(between(day(stdate), 4, 12))
#> # A tibble: 8 x 4
#> orgid stdate locid charnam
#> <chr> <date> <chr> <chr>
#> 1 USGS-NJ 2013-02-12 USGS-01380075 Nitrogen, mixed forms (NH3), (NH4), or~
#> 2 USGS-NJ 2013-02-12 USGS-01368820 Nitrogen, mixed forms (NH3), (NH4), or~
#> 3 USGS-NJ 2013-02-12 USGS-01409815 Nitrogen, mixed forms (NH3), (NH4), or~
#> 4 USGS-NJ 2013-02-12 USGS-01411400 Nitrogen, mixed forms (NH3), (NH4), or~
#> 5 USGS-NJ 2013-02-04 USGS-01458570 Inorganic nitrogen (nitrate and nitrit~
#> 6 USGS-NJ 2013-02-04 USGS-01458570 Phosphorus
#> 7 USGS-NJ 2013-02-04 USGS-01458570 Phosphorus
#> 8 USGS-NJ 2013-02-04 USGS-01445160 Kjeldahl nitrogen
# If {lubridate} isn't installed, this is all year() and day() do...
get_day <- function(date) as.POSIXlt(date, tz = tz(date))$mday
# get_year <- function(date) as.POSIXlt(date, tz = tz(date))$year + 1900 # for years instead of days
roadsalt_data %>%
select(orgid, stdate, locid, charnam) %>%
# filter(between(get_year(stdate), 1996, 2015)) # for years instead of days
filter(between(get_day(stdate), 4, 12))
#> # A tibble: 8 x 4
#> orgid stdate locid charnam
#> <chr> <date> <chr> <chr>
#> 1 USGS-NJ 2013-02-12 USGS-01380075 Nitrogen, mixed forms (NH3), (NH4), or~
#> 2 USGS-NJ 2013-02-12 USGS-01368820 Nitrogen, mixed forms (NH3), (NH4), or~
#> 3 USGS-NJ 2013-02-12 USGS-01409815 Nitrogen, mixed forms (NH3), (NH4), or~
#> 4 USGS-NJ 2013-02-12 USGS-01411400 Nitrogen, mixed forms (NH3), (NH4), or~
#> 5 USGS-NJ 2013-02-04 USGS-01458570 Inorganic nitrogen (nitrate and nitrit~
#> 6 USGS-NJ 2013-02-04 USGS-01458570 Phosphorus
#> 7 USGS-NJ 2013-02-04 USGS-01458570 Phosphorus
#> 8 USGS-NJ 2013-02-04 USGS-01445160 Kjeldahl nitrogen
# Base R
roadsalt_data <- roadsalt_data[, c("orgid", "stdate", "locid", "charnam")]
roadsalt_data[roadsalt_data$stdate >= as.Date("2013-02-04") & roadsalt_data$stdate <= as.Date("2013-02-12") ,]
#> # A tibble: 8 x 4
#> orgid stdate locid charnam
#> <chr> <date> <chr> <chr>
#> 1 USGS-NJ 2013-02-12 USGS-01380075 Nitrogen, mixed forms (NH3), (NH4), or~
#> 2 USGS-NJ 2013-02-12 USGS-01368820 Nitrogen, mixed forms (NH3), (NH4), or~
#> 3 USGS-NJ 2013-02-12 USGS-01409815 Nitrogen, mixed forms (NH3), (NH4), or~
#> 4 USGS-NJ 2013-02-12 USGS-01411400 Nitrogen, mixed forms (NH3), (NH4), or~
#> 5 USGS-NJ 2013-02-04 USGS-01458570 Inorganic nitrogen (nitrate and nitrit~
#> 6 USGS-NJ 2013-02-04 USGS-01458570 Phosphorus
#> 7 USGS-NJ 2013-02-04 USGS-01458570 Phosphorus
#> 8 USGS-NJ 2013-02-04 USGS-01445160 Kjeldahl nitrogen
Created on 2018-05-23 by the reprex package (v0.2.0).
==============================================================
If none of these work, there's something else entirely going on.
Upvotes: 2
Matt Summersgill
Reputation: 4242
If you get tired of battling tidyverse dependency hell, you could try data.table instead:
library(data.table)
## Convert to a data.table by reference
setDT(roadsalt_data)
## Filter on date and only return a subset of columns
roadsalt_data[between(stdate, as.Date("1996-01-01"), as.Date("2015-07-01")), .(orgid,stdate,locid,charnam)]
# orgid stdate locid charnam
# 1: USGS-NJ 2013-02-19 USGS-01407760 Nitrogen, mixed forms (NH3), (NH4), organic, (NO2) and (NO3)
# 2: USGS-NJ 2013-02-19 USGS-01445030 Nitrogen, mixed forms (NH3), (NH4), organic, (NO2) and (NO3)
# 3: USGS-NJ 2013-02-12 USGS-01380075 Nitrogen, mixed forms (NH3), (NH4), organic, (NO2) and (NO3)
# 4: USGS-NJ 2013-02-12 USGS-01368820 Nitrogen, mixed forms (NH3), (NH4), organic, (NO2) and (NO3)
# 5: USGS-NJ 2013-02-12 USGS-01409815 Nitrogen, mixed forms (NH3), (NH4), organic, (NO2) and (NO3)
# 6: USGS-NJ 2013-02-12 USGS-01411400 Nitrogen, mixed forms (NH3), (NH4), organic, (NO2) and (NO3)
# 7: USGS-NJ 2013-02-04 USGS-01458570 Inorganic nitrogen (nitrate and nitrite)
# 8: USGS-NJ 2013-02-04 USGS-01458570 Phosphorus
# 9: USGS-NJ 2013-02-04 USGS-01458570 Phosphorus
# 10: USGS-NJ 2013-02-04 USGS-01445160 Kjeldahl nitrogen
Upvotes: 1
Related Questions
- How to filter data.frame based on date in R
- Filter in dplyr interval of dates
- How to filter by multiple range of dates in R?
- dplyr filter by multiple conditions including date
- Filtering dates in dplyr
- How to filter dates in different rows?
- Filter within a column by date in R
- How to properly filter date in R
- Filtering individual dates in R, dplyr
- Using Date/Time to Filter in R