Working with Your Data in R

Once zc_get_readings() (or zc_sync()) has given you a data frame of readings, here are the everyday R commands for looking at it, filtering it, summarising it, and saving it. You need very little R to be productive — this covers the essentials, with links to fuller R guides at the end.

library(zentraR)
readings <- zc_get_readings("z6-00930", start = Sys.Date() - 7)

Save results so you can reuse them

The single most useful habit in R: store a result in a named object with <-, or it prints once and is gone.

# Prints once, then lost:
zc_pivot_wider(readings)

# Saved as `wide` — now you can view it, filter it, plot it, or export it:
wide <- zc_pivot_wider(readings)
wide

You choose the name (wide, daily, air_temp, …). The <- is R’s assignment arrow: name on the left, value on the right.

Look at your data

readings                 # a tibble: prints the first 10 rows and the column types
View(readings)           # open the RStudio spreadsheet viewer (capital V)
head(readings, 20)       # first 20 rows;  tail(readings) for the last few
str(readings)            # structure: every column and its type
dplyr::glimpse(readings) # a tidy, transposed overview
summary(readings)        # quick per-column statistics
dim(readings)            # number of rows and columns;  nrow() / ncol()
names(readings)          # the column names

View() is interactive (RStudio only) — in a script or a scheduled job use print(), head(), or str() instead.

See what’s inside

readings$column pulls out a single column by name. Combine that with a few base functions to get your bearings:

unique(readings$measurement)   # which measurements are present
table(readings$measurement)    # how many readings of each
unique(readings$device_id)     # which devices
range(readings$datetime)       # earliest and latest timestamp

Filter and sort

The dplyr package (part of the tidyverse) reads almost like English:

library(dplyr)

# Keep only valid air-temperature readings:
readings |> filter(measurement == "Air Temperature", error_code == 0)

# Highest values first:
readings |> arrange(desc(value))

The |> is R’s pipe: it feeds the value on its left into the function on its right. Base R does the same with square brackets, if you prefer:

readings[readings$measurement == "Air Temperature", ]

Quick summaries

mean(readings$value, na.rm = TRUE)   # na.rm = TRUE ignores missing values

# Average and count per measurement:
readings |>
  group_by(measurement) |>
  summarise(avg = mean(value, na.rm = TRUE), n = n())

A quick plot

plot(readings$datetime, readings$value, type = "l")

For publication-quality graphics with ggplot2, see the plotting example in the Getting Started with zentraR vignette.

Save and export

# CSV — opens in Excel / Google Sheets, easy to share:
write.csv(readings, "readings.csv", row.names = FALSE)

# RDS — an exact copy of the R object (types preserved); reload with readRDS():
saveRDS(readings, "readings.rds")
readings <- readRDS("readings.rds")

For an automated, incremental local archive, use zentraR’s own stores (zc_store_csv(), zc_store_rds()) with zc_sync() — see the Scheduling Automatic Syncs vignette.

Getting help

?zc_get_readings                 # the help page for any function
vignette(package = "zentraR")    # list this package's guides

Learn more R

These free resources cover R itself, well beyond what you need for zentraR:

R for Data Science (2nd ed.) — the standard introduction to the tidyverse.
Posit cheatsheets — one-page references for dplyr, ggplot2, RStudio, and more.
RStudio beginner resources — guided starting points.

Related zentraR guides:

vignette("getting-started", package = "zentraR")
vignette("scheduling", package = "zentraR")