From c010e9f9cf25b943e4f4e9e46ad3728126f35bbb Mon Sep 17 00:00:00 2001
From: Matthew Gaughan <mjilg@klone-login03.hyak.local>
Date: Sun, 7 Dec 2025 13:32:35 -0800
Subject: [PATCH] updating rq1 plots

---
 dsl/rq1_plots.R                      | 154 +++++++++++++++++++++++++++
 mgaughan-rstudio-server_31713063.out |  17 +++
 2 files changed, 171 insertions(+)
 create mode 100644 dsl/rq1_plots.R
 create mode 100644 mgaughan-rstudio-server_31713063.out

diff --git a/dsl/rq1_plots.R b/dsl/rq1_plots.R
new file mode 100644
index 0000000..ca80b31
--- /dev/null
+++ b/dsl/rq1_plots.R
@@ -0,0 +1,154 @@
+library(tidyverse)
+#library(dsl)
+library(dplyr)
+dsl_csv <-"~/dsl/120725_DSL_frame.csv"
+dsl_df <- read.csv(dsl_csv, header = TRUE) 
+
+dsl_df <- dsl_df |>
+  mutate(priority = factor(priority, 
+                           levels = rev(c("Unbreak Now!", "High", "Medium", "Low", "Lowest", "Needs Triage"))))
+
+# n authors who also close their own
+ac_summary <- dsl_df %>%
+  group_by(week_index, author_closer, source) %>%
+  summarise(count = n(), .groups = 'drop') |>
+  group_by(week_index, source) %>%
+  mutate(proportion = count / sum(count)) %>%
+  ungroup()
+
+ac_summary|>
+  ggplot(aes(x = week_index, y = proportion, color = author_closer, fill=author_closer, group = author_closer)) +
+  geom_smooth()+
+  scale_color_viridis_d(option='turbo') + 
+  scale_fill_viridis_d(option='turbo') + 
+  facet_grid(source ~ ., scales = "free_y") + 
+  geom_point() +
+  labs(title = "Did the Task Author also close the task? by week created",
+       x = "Weeks from feature deployment",
+       y = "% of work-items",
+       color = "Did the Task Author also Close the Task?") +
+  theme_minimal()
+
+
+# % of EP 
+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")
+
+# TTR by affiliation and priority
+dsl_df |>
+  filter(priority == "Needs Triage" | 
+           priority == "Unbreak Now!" |
+           priority == "High") |>
+  filter(week_index >= -26) |>
+  ggplot(
+    aes(
+      x=as.factor(week_index),
+      y= TTR/168, 
+      color=priority,
+      fill=priority,
+      linetype=isAuthorWMF
+    )
+  ) + 
+  facet_grid(source ~ .) + 
+  geom_boxplot(outlier.shape = NA) +
+  theme_minimal() + 
+  coord_cartesian(ylim = c(0, 156)) + 
+  geom_vline(xintercept =27, linetype = "dashed", color = "black", linewidth = 0.5) +
+  scale_color_viridis_d() +
+  labs(x = "Weeks from Release", y = "Time to Resolution (weeks)", title = "TTR by Task Creation Date and Triage Priority")
+
+
+
+# % of tasks declined per week 
+declined_summary <- dsl_df %>%
+  group_by(week_index, task_status, source) %>%
+  summarise(count = n(), .groups = 'drop') |>
+  group_by(week_index, source) %>%
+  mutate(proportion = count / sum(count)) %>%
+  ungroup()
+
+declined_summary|>
+  ggplot(aes(x = week_index, y = proportion, color = task_status, fill=task_status, group = task_status)) +
+  geom_smooth()+
+  scale_color_viridis_d(option='turbo') + 
+  scale_fill_viridis_d(option='turbo') + 
+  facet_grid(source ~ ., scales = "free_y") + 
+  geom_point() +
+  labs(title = "Task status for new tasks by week created",
+       x = "Weeks from feature deployment",
+       y = "% of items in status",
+       color = "Task Status") +
+  theme_minimal()
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+# count of gerrit links to operations, extensions, or core
+gerrit_summary <- dsl_df |>
+  filter(!is.na(gerrit_repo)) |>
+  mutate(
+    focal_repo = case_when(
+      str_detect(gerrit_repo, "extensions") ~ "extensions",
+      str_detect(gerrit_repo, "operations") ~ "operations",
+      str_detect(gerrit_repo, "integration") ~ "operations",
+      TRUE ~ "core"
+    )
+  ) |>
+  group_by(week_index, focal_repo, source) %>%
+  summarise(count = n(), .groups = 'drop')
+
+gerrit_summary|>
+  ggplot(aes(x = week_index, y = count, color = focal_repo, fill=focal_repo, group = focal_repo)) +
+  geom_smooth()+
+  scale_color_viridis_d(option='turbo') + 
+  scale_fill_viridis_d(option='turbo') + 
+  facet_grid(source ~ ., scales = "free_y") + 
+  geom_point() +
+  labs(title = "Task status for new tasks by week created",
+       x = "Weeks from feature deployment",
+       y = "% of items in status",
+       color = "Task Status") +
+  theme_minimal()
diff --git a/mgaughan-rstudio-server_31713063.out b/mgaughan-rstudio-server_31713063.out
new file mode 100644
index 0000000..cc3b071
--- /dev/null
+++ b/mgaughan-rstudio-server_31713063.out
@@ -0,0 +1,17 @@
+1. SSH tunnel from your workstation using the following command:
+
+   ssh -N -L 8787:n3439:48979 mjilg@klone.hyak.uw.edu
+
+   and point your web browser to http://localhost:8787
+
+2. log in to RStudio Server using the following credentials:
+
+   user: mjilg
+   password: eTkReY42xI/86STHG+4f
+
+When done using RStudio Server, terminate the job by:
+
+1. Exit the RStudio Session ("power" button in the top right corner of the RStudio window)
+2. Issue the following command on the login node:
+
+      scancel -f 31713063