#* topic_names_w_stats:
#*   attr:
#*     fillcolor: '3'
#*   desc: Generate more detailed topic names with usage, entropy, and z-scored coherence
#*     for use in cloud plots.
#*   ext: R
#*   inputs:
#*   - coherence_by_topic
#*   - topic_names
#*   - datapartner_topic_usage_entropy
#* 

library(dplyr)
library(stringr)

topic_names_w_stats <- function(topic_names, coherence_by_topic, datapartner_topic_usage_entropy) {
    topic_names <- topic_names %>%
    # remove trailing paren
    merge(datapartner_topic_usage_entropy, by = "topic_name") %>%
    merge(coherence_by_topic, by = "topic_name") %>%
    mutate(topic_name_stats = str_match(topic_name, "(T-\\d+)")[,2]) %>%
    mutate(U = round(topic_usage_percent * 100, 1),
           #H = round((entropy - mean(entropy))/sd(entropy), 2),
           H = round(relative_entropy, 2),
           C = round((sum_coherence - mean(sum_coherence))/sd(sum_coherence), 1)) %>%
    mutate(topic_name_stats = paste0(topic_name_stats, " (",
                               "U ", U, "%", 
                               ", H ", H, 
                               ", C ", C, ")"
                               ))

    return(topic_names)
}