anastasia-psy/statsandbooze

Find the best day to have a beer!

statsAndBooze

The goal of statsAndBooze is to find the optimal date for a happy hour, i.e. the closest date when all the participants are available.

Installation

You can install the development version of statsAndBooze from GitHub with:

# install.packages("devtools")
devtools::install_github("Anastasia-psy/statsAndBooze")

Usage and Examples

To load the package:

library(statsAndBooze)

The input must be a list of names with the available dates for each participant and must be transformed from string to data format. Use the function parse_dates() for this task.

In the list, you must specify the dates in YYYY-MM-DD format like in the example below:

dates<- list(
  andrea = "2023-03-29", # exactly with this format
  federico = "2023-03-30" # exactly with this format
)

parse_dates(dates)
#> $andrea
#> [1] "2023-03-29"
#> 
#> $federico
#> [1] "2023-03-30"

You can also specify a time interval when a participant is available. This must be done in the following format: YYYY-MM-DD / YYYY-MM-DD (see example).

dates <- list(
  andrea = c("2023-04-01", "2023-04-03 / 2023-04-05"),
  federico = c("2023-04-02", "2023-04-03"),
  chiara = "2023-04-02 / 2023-04-03"
)

parse_dates(dates)
#> $andrea
#> [1] "2023-04-01" "2023-04-03" "2023-04-04" "2023-04-05"
#> 
#> $federico
#> [1] "2023-04-02" "2023-04-03"
#> 
#> $chiara
#> [1] "2023-04-02" "2023-04-03"

You can also specify the dates using the workdays names in your local language in the abbreviated form. To see the available values for days of week, run the function lubridate::wday(x, label = TRUE).The names of the weekdays refer to the subsequent 7 days with respect to the current day (included) when the function is run. For each person the dates must be specified or in date form or as days of the week, but not in a mixed form. See the following example:

lubridate::wday("2023-04-28", label = TRUE)
#> [1] ven
#> Levels: dom < lun < mar < mer < gio < ven < sab

dates <- list(
  andrea = c("sab", "gio", "dom"),
  federico = c("2023-04-27", "2023-04-30"),
  chiara = "2023-04-28 / 2023-04-30"
)

parse_dates(dates)
#> $andrea
#> [1] "2023-04-29" "2023-04-30" "2023-05-04"
#> 
#> $federico
#> [1] "2023-04-27" "2023-04-30"
#> 
#> $chiara
#> [1] "2023-04-28" "2023-04-29" "2023-04-30"

Finally, to define the date for a happy hour, you need the function decide_happy_hour(), which takes the output of the function parse_dates as input. In case there are several dates when all the participants are available, the function returns the nearest suitable date. See the example below.

dates <- list(
  andrea = c("sab", "gio", "dom"),
  federico = c("2023-04-27", "2023-04-30"),
  chiara = "2023-04-28 / 2023-04-30"
)

parsed_dates <- parse_dates(dates)

decide_happy_hour(parsed_dates)
#> [1] "2023-04-30"

The function returns Date of length 0 if there is no date when all the participants are available.

statsandbooze's Issues

Package review

Dear all, I just checked your package and I think that you did a great job, congratulations 🎉 I think there are just a few minor points that are not correct.

1. It is not helpful to test a condition like if (x == TRUE) since the object x can immediately be evaluated as TRUE or FALSE without that extra step

statsAndBooze/R/parse.R

Line 31 in cf1bb91

if (all(!stringr::str_detect(unlist(x), paste(c("lun", "mar", "mer", "gio", "ven", "sab", "dom"),collapse = '|')) == TRUE)) {

2. Looking at the README, the documentation, and the R code, it looks like you decided that the weekdays must be coded using the local language. This is fine but, unfortunately, you also need to properly adjust the lubridate::wday function to accommodate your decision, otherwise the following example fails:

remotes::install_github("Anastasia-psy/statsAndBooze", quiet = TRUE)
library(statsAndBooze)
dates <- list(a = "dom")
parse_dates(dates)
#> $a
#> Date of length 0

# since
lubridate::wday("2023-05-07", label = TRUE)
#> [1] Sun
#> Levels: Sun < Mon < Tue < Wed < Thu < Fri < Sat

^{Created on 2023-05-05 with reprex v2.0.2}

In particular, you need to set lubridate::wday(next_week, label = TRUE, locale = "italian") instead of the following piece of code:

statsAndBooze/R/parse.R

Line 42 in cf1bb91

weekday = lubridate::wday(next_week, label = TRUE))

3. The documentation is fine but you forgot to add extra tests regarding the new functionalities.

I will send you a short report about the remaining part of the exam in a few minutes.

Andrea