Merge pull request #44 from panukatan/dev

add 2021 typhoon data; fix #42

.gitignore

@@ -4,3 +4,4 @@
 .Ruserdata
 docs
 inst/doc
+data-raw/*.pdf

NEWS.md

@@ -1,5 +1,8 @@
 # bagyo v0.1.1.9000
 
+* added CRAN DOI badge
+* added 2021 typhoon data
+
 
 # bagyo 0.1.1
 

README.Rmd

@@ -43,7 +43,7 @@ Oceans and seas significantly impact continental weather, with evaporation from
 
 The Philippines frequently experiences tropical cyclones (called ***bagyo*** - pronounced /baɡˈjo/, [bɐɡˈjo] - in the Filipino language) because of its geographical position. These cyclones typically bring heavy rainfall, leading to widespread flooding, as well as strong winds that cause significant damage to human life, crops, and property. Data on cyclones are collected and curated by the [Philippine Atmospheric, Geophysical, and Astronomical Services Administration (PAGASA)](https://www.pagasa.dost.gov.ph/).
 
-This package contains Philippine tropical cyclones data from 2017 to 2020 in a machine-readable format. It is hoped that this data package provides an interesting and unique dataset for data exploration and visualisation as an adjunct to the traditional [`iris`](https://stat.ethz.ch/R-manual/R-devel/library/datasets/html/iris.html) dataset and to the current [`palmerpenguins`](https://allisonhorst.github.io/palmerpenguins/) dataset.
+This package contains Philippine tropical cyclones data from 2017 to 2021 in a machine-readable format. It is hoped that this data package provides an interesting and unique dataset for data exploration and visualisation as an adjunct to the traditional [`iris`](https://stat.ethz.ch/R-manual/R-devel/library/datasets/html/iris.html) dataset and to the current [`palmerpenguins`](https://allisonhorst.github.io/palmerpenguins/) dataset.
 
 ## Installation
 
@@ -140,7 +140,7 @@ bagyo |>
     x = "wind speed (km/h)",
     y = "central pressure (hPa)"
   ) +
-  facet_wrap(. ~ year, ncol = 4) +
+  facet_wrap(. ~ year, ncol = 5) +
   theme_bw() +
   theme(
     legend.position = "top",

README.md

@@ -82,7 +82,7 @@ library(bagyo)
 data(package = "bagyo")
 
 bagyo
-#> # A tibble: 86 × 9
+#> # A tibble: 101 × 9
 #>     year category_code category_name         name  rsmc_name start              
 #>    <dbl> <fct>         <fct>                 <chr> <chr>     <dttm>             
 #>  1  2017 TD            Tropical Depression   Auri… <NA>      2017-01-07 08:00:00
@@ -95,7 +95,7 @@ bagyo
 #>  8  2017 TS            Tropical Storm        Huan… Haitang   2017-07-30 02:00:00
 #>  9  2017 STS           Severe Tropical Storm Isang Hato      2017-08-20 08:00:00
 #> 10  2017 TS            Tropical Storm        Joli… Pakhar    2017-08-24 14:00:00
-#> # ℹ 76 more rows
+#> # ℹ 91 more rows
 #> # ℹ 3 more variables: end <dttm>, pressure <int>, speed <int>
 ```
 
@@ -117,11 +117,11 @@ bagyo |>
 #> # A tibble: 5 × 4
 #>   category_name             n mean_pressure mean_speed
 #>   <fct>                 <int>         <dbl>      <dbl>
-#> 1 Tropical Depression      23          996.       39.8
-#> 2 Tropical Storm           25          986.       61.6
-#> 3 Severe Tropical Storm    15          978.       75  
-#> 4 Typhoon                  21          941.      102. 
-#> 5 Super Typhoon             2          908.      112.
+#> 1 Tropical Depression      27          995.       39.3
+#> 2 Tropical Storm           29          987.       58.8
+#> 3 Severe Tropical Storm    17          979.       72.6
+#> 4 Typhoon                  23          943.       99.1
+#> 5 Super Typhoon             5          907       113
 ```
 
 ### `bagyo` is useful in learning how to work with dates
@@ -135,11 +135,11 @@ bagyo |>
 #> # A tibble: 5 × 2
 #>   category_name         mean_duration  
 #>   <fct>                 <drtn>         
-#> 1 Tropical Depression    46.69565 hours
-#> 2 Tropical Storm         57.48000 hours
-#> 3 Severe Tropical Storm  79.13333 hours
-#> 4 Typhoon               106.66667 hours
-#> 5 Super Typhoon          77.50000 hours
+#> 1 Tropical Depression    45.29630 hours
+#> 2 Tropical Storm         61.03448 hours
+#> 3 Severe Tropical Storm  81.29412 hours
+#> 4 Typhoon               110.04348 hours
+#> 5 Super Typhoon         115.60000 hours
 ```
 
 ### `bagyo` is great to visualise

data-raw/process_data.R

@@ -10,14 +10,14 @@ library(tibble)
 ## Download PDFs ----
 urls <- paste0(
   "https://pubfiles.pagasa.dost.gov.ph/pagasaweb/files/tamss/weather/tcsummary/PAGASA_ARTC_",
-  2017:2020, ".pdf"
+  2017:2021, ".pdf"
 )
 
 Map(
   f = download.file,
   url = as.list(urls),
   destfile = as.list(
-    paste0("data-raw/", 2017:2020, ".pdf")
+    paste0("data-raw/", 2017:2021, ".pdf")
   )
 )
 
@@ -372,8 +372,108 @@ df_2020 <- set1_2020 |>
   )
 
 
+## 2021 data ----
+
+#df_2021 <- pdftools::pdf_data("data-raw/2021.pdf")[[42]]
+
+set1_2021 <- pdftools::pdf_data("data-raw/2021.pdf")[[41]] |>
+  dplyr::filter(y %in% 409:616) |>
+  dplyr::pull(text) |>
+  (\(x) x[x != "to"])() |>
+  (\(x) c(
+    x[1:24], NA_character_, x[25:43], x[46:47], NA_character_,
+    x[48:88], x[91:112], x[115:149], NA_character_, x[150:168]
+  ))() |>
+  matrix(ncol = 11, byrow = TRUE) |>
+  data.frame() |>
+  tibble::tibble() |>
+  dplyr::rename_with(
+    .fn = function(x)
+      c("domestic_name", "international_name", "international_code",
+        "warning_start_date", "warning_start_time", "warning_end_date",
+        "warning_end_time", "peak_speed", "peak_pressure", "peak_date",
+        "peak_time")
+  ) |>
+  dplyr::mutate(
+    international_name = ifelse(
+      international_name == "Unnamed", NA_character_, international_name
+    ),
+    international_code = stringr::str_remove_all(
+      string = international_code, pattern = "\\(|\\)"
+    ),
+    warning_start = paste0(warning_start_date, "/2020 ", warning_start_time) |>
+      strptime(format = "%m/%d/%Y %H", tz = "UTC"),
+    warning_end = paste0(warning_end_date, "/2020 ", warning_end_time) |>
+      strptime(format = "%m/%d/%Y %H", tz = "UTC"),
+    dplyr::across(.cols = peak_pressure:peak_speed, .fns = ~as.integer(.x)),
+    peak_time = paste0(peak_date, "/2020 ", peak_time) |>
+      strptime(format = "%m/%d/%Y %H", tz = "UTC")
+  )
+
+set2_2021 <- pdftools::pdf_data("data-raw/2021.pdf")[[42]] |>
+  dplyr::filter(y %in% 168:328) |>
+  dplyr::pull(text) |>
+  (\(x) x[x != "to"])() |>
+  (\(x) c(
+    x[1:19], NA_character_, x[20:23], NA_character_,
+    x[24:28], NA_character_, x[29:43], NA_character_,
+    x[44:69], NA_character_, x[70:79], NA_character_,
+    x[80:112], paste(x[113:114], collapse = ""), x[115:140], NA_character_,
+    x[141:length(x)]
+  ))() |>
+  matrix(ncol = 11, byrow = TRUE) |>
+  data.frame() |>
+  tibble::tibble() |>
+  dplyr::rename_with(
+    .fn = function(x)
+      c("domestic_name", "international_name", "international_code",
+        "warning_start_date", "warning_start_time", "warning_end_date",
+        "warning_end_time", "duration_days", "duration_hours", "category_code",
+        "landfall")
+  ) |>
+  dplyr::mutate(
+    international_name = ifelse(
+      international_name == "Unnamed", NA_character_, international_name
+    ),
+    international_code = stringr::str_remove_all(
+      string = international_code, pattern = "\\(|\\)"
+    ),
+    warning_start = paste0(warning_start_date, "/2021 ", warning_start_time) |>
+      strptime(format = "%m/%d/%Y %H", tz = "UTC"),
+    warning_end = paste0(warning_end_date, "/2021 ", warning_end_time) |>
+      strptime(format = "%m/%d/%Y %H", tz = "UTC"),
+    category_code = factor(
+      category_code,
+      levels = c("TD", "TS", "STS", "TY", "STY")
+    ),
+    category_name = factor(
+      category_code,
+      levels = c("TD", "TS", "STS", "TY", "STY"),
+      labels = c(
+        "Tropical Depression", "Tropical Storm", "Severe Tropical Storm",
+        "Typhoon", "Super Typhoon"
+      )
+    )
+  )
+
+df_2021 <- set1_2021 |>
+  dplyr::mutate(
+    warning_start = set2_2021$warning_start,
+    warning_end = set2_2021$warning_end,
+    category_code = set2_2021$category_code,
+    category_name = set2_2021$category_name
+  ) |>
+  dplyr::select(
+    category_code, category_name, domestic_name, international_name,
+    warning_start, warning_end, peak_pressure, peak_speed
+  ) |>
+  dplyr::mutate(
+    domestic_name = stringr::str_to_title(domestic_name),
+    international_name = stringr::str_to_title(international_name)
+  )
+
 ## Concatenate ----
-bagyo <- rbind(df_2017, df_2018, df_2019, df_2020) |>
+bagyo <- rbind(df_2017, df_2018, df_2019, df_2020, df_2021) |>
   dplyr::mutate(
     year = lubridate::year(warning_start), .before = category_code
   ) |>

vignettes/bagyo.Rmd

@@ -49,9 +49,9 @@ The following information is available from the dataset:
 
 This metadata can be viewed in R through a call to `?bagyo` in the R console.
 
-Whilst tropical cyclones have affected the Philippines far earlier than 2017 and more currently than 2020, official and publicly available data for the information described above is only available in the reports for years 2017 to 2020. Earlier documents of this annual reporting pre-2017 have been produced but are not available on the [PAGASA](https://www.pagasa.dost.gov.ph/) website. These reports of the tropical cyclone season (re-started in 2019) are published within two years after the termination of the season. Hence, the most recent report is only up to 2020 for now.
+Whilst tropical cyclones have affected the Philippines far earlier than 2017 and more currently than 2021, official and publicly available data for the information described above is only available in the reports for years 2017 to 2020. Earlier documents of this annual reporting pre-2017 have been produced but are not available on the [PAGASA](https://www.pagasa.dost.gov.ph/) website. These reports of the tropical cyclone season (re-started in 2019) are published within two years after the termination of the season. Hence, the most recent report is only up to 2021 for now.
 
-It is expected that reports for 2021 onwards will continue to be published and made available by PAGASA. As such, the `bagyo` package and the `bagyo` dataset within it will be updated accordingly. Continued efforts are also being taken to find sources of information for years preceding 2017.
+It is expected that reports for 2022 onwards will continue to be published and made available by PAGASA. As such, the `bagyo` package and the `bagyo` dataset within it will be updated accordingly. Continued efforts are also being taken to find sources of information for years preceding 2017.
 
 <br>
 <br>

vignettes/visualisation.Rmd

@@ -151,12 +151,12 @@ bagyo |>
     mapping = aes(colour = year), method = "lm", se = FALSE, linewidth = 0.75
   ) +
   scale_colour_manual(
-    values = c("#9c5e60", "#4b876e", "#465b92", "#e5be72")
+    values = c("#9c5e60", "#4b876e", "#465b92", "#e5be72", "#5d0505")
   ) +
-  scale_shape_manual(values = 15:18) +
+  scale_shape_manual(values = 15:19) +
   labs(
     title = "Maximum sustained wind speed by duration of cyclones",
-    subtitle = "2017-2020",
+    subtitle = "2017-2021",
     x = "speed (km/h)", y = "duration (hours)",
     colour = "Year", shape = "Year"
   ) +
@@ -180,7 +180,7 @@ bagyo |>
   scale_y_continuous(breaks = seq(from = 0, to = 6, by = 1)) +
   labs(
     title = "Number of cyclones over time",
-    subtitle = "2017-2020",
+    subtitle = "2017-2021",
     x = NULL,
     y = "n"
   ) +
@@ -206,7 +206,7 @@ bagyo |>
   geom_boxplot(colour = "#4b876e", fill = "#4b876e", alpha = 0.5) +
   labs(
     title = "Distribution of tropical cyclone maximum sustained wind speed",
-    subtitle = "2017-2022",
+    subtitle = "2017-2021",
     x = NULL, y = "speed (km/h)"
   ) +
   theme_minimal() +
@@ -225,7 +225,7 @@ bagyo |>
   ) +
   labs(
     title = "Distribution of tropical cyclone maximum sustained wind speed",
-    subtitle = "2017-2022",
+    subtitle = "2017-2021",
     x = NULL, y = "speed (km/h)"
   ) +
   theme_minimal() +
@@ -240,7 +240,7 @@ bagyo |>
   geom_jitter(colour = "#4b876e", size = 3, width = 0.2) +
   labs(
     title = "Distribution of tropical cyclone maximum sustained wind speed",
-    subtitle = "2017-2022",
+    subtitle = "2017-2021",
     x = NULL, y = "speed (km/h)"
   ) +
   theme_minimal() +