initial import
Change-Id: I46029b5f36a4711c32b8bcd14a7255cd9f495853
diff --git a/R/icc_stats.R b/R/icc_stats.R
new file mode 100644
index 0000000..d775964
--- /dev/null
+++ b/R/icc_stats.R
@@ -0,0 +1,91 @@
+library(RKorAPClient)
+library(httr)
+library(httpuv)
+library(tidyverse)
+library(scales)
+library(idsThemeR)
+
+icc_base_url = "https://korap.ids-mannheim.de/instance/icc";
+
+connections = list()
+
+icc <- tibble(
+ lang = c("eng", "ger", "nor"),
+ app_id = c(
+ "mTTTnJ6f6hGrPh6dRhbJhJ",
+ "TMLPTJfP7rHb93bpFp39mL",
+ "TMFtPJnbb7f4MRmd76Rb34"
+ )
+)
+
+
+icc_token <- function(lang, app_id, url = paste0(icc_base_url, '/', lang)) {
+ token_key = paste0("KORAP_ICC_TOKEN_", lang)
+ token = Sys.getenv(token_key)
+ if (token != "")
+ return(token)
+
+ korap_app <- oauth_app("icc-iclc10-contribution", key = app_id, secret = NULL)
+
+ korap_endpoint <- oauth_endpoint(NULL,
+ "settings/oauth/authorize",
+ "api/v1.0/oauth2/token",
+ base_url = url)
+ token_bundle = oauth2.0_token(korap_endpoint, korap_app, scope = "search match_info", cache = FALSE)
+ token = token_bundle[["credentials"]][["access_token"]]
+ do.call(Sys.setenv, as.list(setNames(token, token_key)))
+ return(token)
+}
+
+icc_con <- function(lang, token = Sys.getenv(paste0("KORAP_ICC_TOKEN_", lang))) {
+ if ((! lang %in% names(connections)) || is_empty(connections[[lang]])) {
+ url <- paste0(icc_base_url, '/', lang)
+ connections[[lang]] <<- new("KorAPConnection", KorAPUrl = url, accessToken = token, cache = F)
+ }
+ return(connections[[lang]])
+}
+
+icc <- icc %>%
+ rowwise() %>%
+ mutate(token = icc_token(lang, app_id)) %>%
+ add_column(tokens=0)
+
+
+genre <- c("Blog",
+ "Creative:Novels_ShortStories",
+ "Informational:Learned:Humanities",
+ "Informational:Learned:NaturalSciences",
+ "Informational:Learned:SocialSciences",
+ "Informational:Learned:Technology",
+ "Informational:Popular:Humanities",
+ "Informational:Popular:NaturalSciences",
+ "Informational:Popular:SocialSciences",
+ "Informational:Popular:Technology",
+ "Informational:Reportage",
+ "Instructional:AdministrativeRegulatoryProse",
+ "Instructional:Skills_Hobbies",
+ "Persuasive"
+)
+
+icc_genre <- icc %>%
+ expand_grid(genre) %>%
+ mutate(vc = paste0("iccGenre=", genre)) %>%
+ rowwise() %>%
+ mutate(tokens= corpusStats(icc_con(lang, token), vc = vc)@tokens)
+
+icc_genre %>% ggplot(aes(x=lang, fill=genre, y=tokens)) +
+ geom_col() + scale_y_continuous(labels = label_number(scale_cut = cut_short_scale())) +
+ theme_ids() +
+ geom_text(aes(label=if_else(tokens > 0, as.character(tokens), ""), y=tokens), position= position_stack(reverse = F, vjust = 0.5), color="white", size=3.2, family="Fira Sans Condensed")
+
+year <- c(1988:2022)
+
+icc_year <- icc %>%
+ expand_grid(year) %>%
+ mutate(vc = paste0("pubDate in ", year)) %>%
+ rowwise() %>%
+ mutate(tokens= corpusStats(icc_con(lang, token), vc = vc)@tokens)
+
+icc_year %>% ggplot(aes(x=year, fill=lang, color=lang, y=tokens)) +
+ geom_line() + geom_point() + geom_area() + scale_y_continuous(labels = label_number(scale_cut = cut_short_scale())) +
+ theme_ids()
diff --git a/icc-iclc10.Rproj b/icc-iclc10.Rproj
new file mode 100644
index 0000000..b42a3cc
--- /dev/null
+++ b/icc-iclc10.Rproj
@@ -0,0 +1,16 @@
+Version: 1.0
+
+RestoreWorkspace: Default
+SaveWorkspace: Default
+AlwaysSaveHistory: Default
+
+EnableCodeIndexing: Yes
+UseSpacesForTab: Yes
+NumSpacesForTab: 2
+Encoding: UTF-8
+
+RnwWeave: knitr
+LaTeX: XeLaTeX
+
+AutoAppendNewline: Yes
+StripTrailingWhitespace: Yes