library(tidyverse)
library(skimr)

dir_path <- "../../slides/datamanaging/data/"
list.files(dir_path)
file_path <- paste0(dir_path, "hotels.csv")

hotels <- read_csv(file_path)

#ESERCIZIO 1
skim(hotel)

#ESERCIZIO 2
last_minute <- hotels %>%
              filter(
                    country != "USA", 
                    lead_time < 1
                )
# la 'virgola' potrebbe essere convertita in '&' (AND logico)
hotels %>% filter(country != "USA" & lead_time < 1)

# last_minute: 6,174 rows

#ESERCIZIO 3
no0_little_one <- hotels %>%
                filter(
            children >= 1 | babies >= 1
                )
# 9,332

#ESERCIZIO 4

no0_little_one_grouped <- no0_little_one %>% 
                    group_by(hotel) %>% summarize(n = n())

# A tibble: 2 × 2
# hotel            n
# <chr>        <int>
# 1 City Hotel    5403
# 2 Resort Hotel  3929

# check by an alternative code (used by most of you)
nrow(hotels %>% filter(children >= 1 | babies >= 1, hotel=="City Hotel"))
nrow(hotels %>% filter(children >= 1 | babies >= 1, hotel!="City Hotel"))

#ESERCIZIO 5

hotels %>% count(adults, sort=T) # or
hotels %>% count(adults) %>% arrange(desc(n))

#ESERCIZIO 6

hotels %>% count(is_canceled, adults) %>% 
  arrange(is_canceled, desc(n)) # or
hotels %>% group_by(is_canceled) %>% 
  count(adults) %>% arrange(is_canceled, desc(n)) 

# an alternative code used by some of you
hotels %>% filter(is_canceled==0) %>% count(adults, sort=T) 
hotels %>% filter(is_canceled==1) %>% count(adults, sort=T) 

#ESERCIZIO 7

hotels %>% group_by(hotel) %>% 
  summarise(min_adr=min(adr), mean_adr=mean(adr), 
            median_adr=median(adr), max_adr=max(adr))

# A tibble: 2 × 5
# hotel          Min  Mean Median   Max
# <chr>        <dbl> <dbl>  <dbl> <dbl>
# 1 City Hotel    0    105.    99.9  5400
# 2 Resort Hotel -6.38  95.0   75     508

#ESERCIZIO 8

anomalous_adr <- hotels %>%  
  filter(adr %in% c(-6.38, 5400)) %>% 
  select(adr, everything())  

anomalous_adr %>% select(
  adr, hotel, adults, children, babies, 
         starts_with("arrival")) %>% 
  select(-contains(c("week", "day")))

hotels %>% slice_min(adr) %>% select(adr, everything())
hotels %>% slice_max(adr) %>% select(adr, everything())

hotels %>% slice_min(adr) %>% select(
  adr, hotel, adults, children, babies, 
            starts_with("arrival")))
hotels %>% slice_max(adr) %>% select(
  adr, hotel, adults, children, babies, 
  starts_with("arrival")))