library(tidyverse)
#library(dsl)
library(dplyr)
dsl_csv <-"~/dsl/111725_DSL_frame.csv"
dsl_df <- read.csv(dsl_csv, header = TRUE) 


ggplot(
  dsl_df,
  aes(
    x=as.factor(week_index),
    y= (TTR/168), 
    fill=isAuthorWMF
  )
) + 
  facet_grid(source ~ ., scales = "free_y") + 
  geom_boxplot() +
  theme_minimal() + 
  scale_fill_viridis_d() + 
  labs(x = "Weeks from Release", y = "Time to Resolution (weeks)", title = "TTR by Task Creation Date")

weekly_summary <- dsl_df |>
  group_by(week_index, source, isAuthorWMF)|>
  summarise(
    tasks_made = sum(!is.na(resolution_outcome)),
    count_resolution_outcome = sum(dsl_score),
    author_closer_sum  = sum(author_closer == TRUE),
    median_olmo_EP_prop_adac = median(olmo_EP_prop_adac),
    median_olmo_TSOL_prop_adac = median(olmo_TSOL_prop_adac),
    median_comments_before_resolution = median(n_comments_before)
  )

ggplot(
  weekly_summary,
  aes(
    x=week_index,
    y=tasks_made, 
    fill=isAuthorWMF
  )
) + 
  facet_grid(source ~ ., scales = "free_y") +  
  geom_col(position = position_dodge(width = 0.9), width = 0.8) +
  geom_vline(data = weekly_summary |> filter(source == "c1"), 
             aes(xintercept = -29), 
             linetype = "dotted", color = "black", linewidth = 0.5) +
  geom_vline(data = weekly_summary |> filter(source == "c1"), 
             aes(xintercept = -9), 
             linetype = "dotted", color = "black", linewidth = 0.5) +
  geom_vline(data = weekly_summary |> filter(source == "c1"), 
             aes(xintercept = -4), 
             linetype = "3313", color = "black", linewidth = 0.5) +
  geom_vline(data = weekly_summary |> filter(source == "c2"), 
             aes(xintercept = -99), 
             linetype = "dotted", color = "black", linewidth = 0.5) +
  geom_vline(data = weekly_summary |> filter(source == "c2"), 
             aes(xintercept = -4),   
             linetype = "3313", color = "black", linewidth = 0.5) +
  geom_vline(data = weekly_summary |> filter(source == "c3"), 
             aes(xintercept = -97), 
             linetype = "dotted", color = "black", linewidth = 0.5) +
  geom_vline(data = weekly_summary |> filter(source == "c3"), 
             aes(xintercept = -3), 
             linetype = "3313", color = "black", linewidth = 0.5) +
  geom_vline(xintercept = 0, linetype = "dashed", color = "black", linewidth = 0.5) +
  theme_minimal() + 
  scale_fill_viridis_d() 


outcome_summary <- dsl_df |>
  group_by(source, isAuthorWMF)|>
  summarise(
    total_sum = sum(!is.na(resolution_outcome)),
    count_resolution_outcome = sum(resolution_outcome),
    success_prop = count_resolution_outcome / total_sum,
    median_ttr_days = median(TTR, na.rm = TRUE) / 24,
    median_comments_before_resolution = median(n_comments_before)
    )


library(ggplot2)
library(ggdist)


ggplot(
  dsl_df, 
  aes(
    x=week_index,
    y=olmo_EP_prop_adac,
    color=isAuthorWMF
  )
) +
  facet_grid(source ~ .) +
  geom_point() +
  geom_smooth() + 
  scale_color_viridis_d() +
  theme_minimal() + 
  labs(x = "Weeks from Release", y = "% of sentences machine-tagged as'Existent Problems'", title = "Proportion of 'Existent Problems' tags over time")

dsl_df <- dsl_df |>
  mutate(priority = factor(priority, 
         levels = c("Unbreak Now!", "High", "Medium", "Low", "Lowest", "Needs Triage")))

ggplot(dsl_df, 
       aes(
         fill=resolution_outcome, 
         x=priority
       )) + 
  facet_grid(~source) +
  geom_bar() +
theme_minimal() 


signed_power <- function(x, p) {
  sign(x) * abs(x) ^ p
}

signed_log <- function(x) sign(x) * log1p(abs(x))  
dsl_df <- dsl_df |>
  mutate(
    sp_med_pc3_adac = signed_power(median_PC3_adac, 0.2),
    sp_med_pc4_adac = signed_power(median_PC4_adac, 0.2),
    sl_med_pc4_adac = signed_log(median_PC4_adac),
    sl_med_pc3_adac = signed_log(median_PC3_adac)
  )


ggplot(dsl_df, aes(
  y= log1p(TTR/24),
  x=sl_med_pc4_adac,
  shape=isAuthorWMF,
  color=isAuthorWMF
                   )) +
  facet_grid(~source) +
  theme_minimal() + 
  geom_smooth(method="loess", span=0.5) + 
  geom_point() + 
  scale_color_viridis_d()