First stab at figuring out the number of very active mentors (#206)

* First stab at figuring out the number of very active mentors in a project.

* Fix format and linter errors.

* Add flake to requirements.

* Adds call to mentor counter and output in markdown.

This adds the call to mentor counter and displays the results in markdown
including first tests for this functionality.

* Make mentor counting configurable.

This adds two configuration options: One to enable mentor counting and one for
configuring how many comments a user needs to leave in discussions, PRs. and
issues to be counted as an active mentor.

* Adds mentor counting to json output and adds missing config.

This adds mentor counting output to json format. In addition this change makes
max number of comments to evaluate configurable as well as the cutoff for
heavily involved mentors.

* Fix merge conflicts.

* Fix linting errors.

* fix: linting fixes

Signed-off-by: Zack Koppert <[email protected]>

* 8 is reasonable number of attrs

Signed-off-by: Zack Koppert <[email protected]>

* Update test_most_active_mentors.py

Co-authored-by: Jason Meridth <[email protected]>

* Update config.py

Co-authored-by: Jason Meridth <[email protected]>

* Update config.py

Remove merge residual

* Update requirements.txt

Remove lib only needed for testing.

* Update issue_metrics.py

Co-authored-by: Jason Meridth <[email protected]>

* Update config.py

* Update config.py

set type of `enable_mentor_count` to `bool`

* Update test_config.py

change tests to handle boolean change of enable_mentor_count

---------

Signed-off-by: Zack Koppert <[email protected]>
Co-authored-by: Drost-Fromm <[email protected]>
Co-authored-by: Zack Koppert <[email protected]>
Co-authored-by: Jason Meridth <[email protected]>
Co-authored-by: Jason Meridth <[email protected]>

Author: 4 people

README.md

@@ -145,6 +145,10 @@ This action can be configured to authenticate with GitHub App Installation or Pe
 | `HIDE_TIME_TO_CLOSE`          | False | False      | If set to `true`, the time to close will not be displayed in the generated Markdown file. |
 | `HIDE_TIME_TO_FIRST_RESPONSE` | False    | False   | If set to `true`, the time to first response will not be displayed in the generated Markdown file. |
 | `IGNORE_USERS`                | False | False      | A comma separated list of users to ignore when calculating metrics. (ie. `IGNORE_USERS: 'user1,user2'`). To ignore bots, append `[bot]` to the user (ie. `IGNORE_USERS: 'github-actions[bot]'`)  |
+| `ENABLE_MENTOR_COUNT` | False | False | If set to 'TRUE' count number of comments users left on discussions, issues and PRs and display number of active mentors |
+| `MIN_MENTOR_COMMENTS` | False | 10 | Minimum number of comments to count as a mentor |
+| `MAX_COMMENTS_EVAL` | False | 20 | Maximum number of comments per thread to evaluate for mentor stats |
+| `HEAVILY_INVOLVED_CUTOFF` | False | 3 | Cutoff after which a mentor's comments in one issue are no longer counted against their total score |
 | `LABELS_TO_MEASURE`           | False    | `""`    | A comma separated list of labels to measure how much time the label is applied. If not provided, no labels durations will be measured. Not compatible with discussions at this time. |
 | `SEARCH_QUERY`                | True     | `""`    | The query by which you can filter issues/PRs which must contain a `repo:`, `org:`, `owner:`, or a `user:` entry. For discussions, include `type:discussions` in the query. |
 

classes.py

@@ -19,9 +19,12 @@ class IssueWithMetrics:
         time_to_answer (timedelta, optional): The time it took to answer the
             discussions in the issue.
         label_metrics (dict, optional): A dictionary containing the label metrics
+        mentor_activity (dict, optional): A dictionary containing active mentors
 
     """
 
+    # pylint: disable=too-many-instance-attributes
+
     def __init__(
         self,
         title,
@@ -31,6 +34,7 @@ def __init__(
         time_to_close=None,
         time_to_answer=None,
         labels_metrics=None,
+        mentor_activity=None,
     ):
         self.title = title
         self.html_url = html_url
@@ -39,3 +43,4 @@ def __init__(
         self.time_to_close = time_to_close
         self.time_to_answer = time_to_answer
         self.label_metrics = labels_metrics
+        self.mentor_activity = mentor_activity

config.py

@@ -34,6 +34,10 @@ class EnvVars:
         hide_time_to_first_response (bool): If true, the time to first response metric is hidden in the output
         ignore_users (List[str]): List of usernames to ignore when calculating metrics
         labels_to_measure (List[str]): List of labels to measure how much time the lable is applied
+        enable_mentor_count (bool): If set to TRUE, compute number of mentors
+        min_mentor_comments (str): If set, defines the minimum number of comments for mentors
+        max_comments_eval (str): If set, defines the maximum number of comments to look at for mentor evaluation
+        heavily_involved_cutoff (str): If set, defines the cutoff after which heavily involved commentors in
         search_query (str): Search query used to filter issues/prs/discussions on GitHub
     """
 
@@ -51,6 +55,10 @@ def __init__(
         hide_time_to_first_response: bool,
         ignore_user: List[str],
         labels_to_measure: List[str],
+        enable_mentor_count: bool,
+        min_mentor_comments: str,
+        max_comments_eval: str,
+        heavily_involved_cutoff: str,
         search_query: str,
     ):
         self.gh_app_id = gh_app_id
@@ -65,6 +73,10 @@ def __init__(
         self.hide_time_to_answer = hide_time_to_answer
         self.hide_time_to_close = hide_time_to_close
         self.hide_time_to_first_response = hide_time_to_first_response
+        self.enable_mentor_count = enable_mentor_count
+        self.min_mentor_comments = min_mentor_comments
+        self.max_comments_eval = max_comments_eval
+        self.heavily_involved_cutoff = heavily_involved_cutoff
         self.search_query = search_query
 
     def __repr__(self):
@@ -82,6 +94,10 @@ def __repr__(self):
             f"{self.hide_time_to_first_response},"
             f"{self.ignore_users},"
             f"{self.labels_to_measure},"
+            f"{self.enable_mentor_count},"
+            f"{self.min_mentor_comments},"
+            f"{self.max_comments_eval},"
+            f"{self.heavily_involved_cutoff},"
             f"{self.search_query})"
         )
 
@@ -166,6 +182,10 @@ def get_env_vars(test: bool = False) -> EnvVars:
     hide_time_to_answer = get_bool_env_var("HIDE_TIME_TO_ANSWER")
     hide_time_to_close = get_bool_env_var("HIDE_TIME_TO_CLOSE")
     hide_time_to_first_response = get_bool_env_var("HIDE_TIME_TO_FIRST_RESPONSE")
+    enable_mentor_count = get_bool_env_var("ENABLE_MENTOR_COUNT")
+    min_mentor_comments = os.getenv("MIN_MENTOR_COMMENTS", "10")
+    max_comments_eval = os.getenv("MAX_COMMENTS_EVAL", "20")
+    heavily_involved_cutoff = os.getenv("HEAVILY_INVOLVED_CUTOFF", "3")
 
     return EnvVars(
         gh_app_id,
@@ -180,5 +200,9 @@ def get_env_vars(test: bool = False) -> EnvVars:
         hide_time_to_first_response,
         ignore_users_list,
         labels_to_measure_list,
+        enable_mentor_count,
+        min_mentor_comments,
+        max_comments_eval,
+        heavily_involved_cutoff,
         search_query,
     )

issue_metrics.py

@@ -31,6 +31,7 @@
 from json_writer import write_to_json
 from labels import get_label_metrics, get_stats_time_in_labels
 from markdown_writer import write_to_markdown
+from most_active_mentors import count_comments_per_user, get_mentor_count
 from time_to_answer import get_stats_time_to_answer, measure_time_to_answer
 from time_to_close import get_stats_time_to_close, measure_time_to_close
 from time_to_first_response import (
@@ -40,8 +41,6 @@
 from time_to_merge import measure_time_to_merge
 from time_to_ready_for_review import get_time_to_ready_for_review
 
-GITHUB_BASE_URL = "https://github.com"
-
 
 def search_issues(
     search_query: str, github_connection: github3.GitHub
@@ -126,6 +125,8 @@ def get_per_issue_metrics(
     discussions: bool = False,
     labels: Union[List[str], None] = None,
     ignore_users: Union[List[str], None] = None,
+    max_comments_to_eval: int = 20,
+    heavily_involved: int = 3,
 ) -> tuple[List, int, int]:
     """
     Calculate the metrics for each issue/pr/discussion in a list provided.
@@ -158,10 +159,20 @@ def get_per_issue_metrics(
                 None,
                 None,
                 None,
+                None,
             )
             issue_with_metrics.time_to_first_response = measure_time_to_first_response(
                 None, issue, ignore_users
             )
+            issue_with_metrics.mentor_activity = count_comments_per_user(
+                None,
+                issue,
+                ignore_users,
+                None,
+                None,
+                max_comments_to_eval,
+                heavily_involved,
+            )
             issue_with_metrics.time_to_answer = measure_time_to_answer(issue)
             if issue["closedAt"]:
                 issue_with_metrics.time_to_close = measure_time_to_close(None, issue)
@@ -188,6 +199,15 @@ def get_per_issue_metrics(
             issue_with_metrics.time_to_first_response = measure_time_to_first_response(
                 issue, None, pull_request, ready_for_review_at, ignore_users
             )
+            issue_with_metrics.mentor_activity = count_comments_per_user(
+                issue,
+                None,
+                pull_request,
+                ready_for_review_at,
+                ignore_users,
+                max_comments_to_eval,
+                heavily_involved,
+            )
             if labels:
                 issue_with_metrics.label_metrics = get_label_metrics(issue, labels)
             if issue.state == "closed":  # type: ignore
@@ -259,6 +279,10 @@ def main():
         token,
         env_vars.ghe,
     )
+    enable_mentor_count = env_vars.enable_mentor_count
+    min_mentor_count = int(env_vars.min_mentor_comments)
+    max_comments_eval = int(env_vars.max_comments_eval)
+    heavily_involved_cutoff = int(env_vars.heavily_involved_cutoff)
 
     # Get the repository owner and name from the search query
     owner = get_owner(search_query)
@@ -283,13 +307,13 @@ def main():
         issues = get_discussions(token, search_query)
         if len(issues) <= 0:
             print("No discussions found")
-            write_to_markdown(None, None, None, None, None, None, None)
+            write_to_markdown(None, None, None, None, None, None, None, None)
             return
     else:
         issues = search_issues(search_query, github_connection)
         if len(issues) <= 0:
             print("No issues found")
-            write_to_markdown(None, None, None, None, None, None, None)
+            write_to_markdown(None, None, None, None, None, None, None, None)
             return
 
     # Get all the metrics
@@ -298,6 +322,8 @@ def main():
         discussions="type:discussions" in search_query,
         labels=labels,
         ignore_users=ignore_users,
+        max_comments_to_eval=max_comments_eval,
+        heavily_involved=heavily_involved_cutoff,
     )
 
     stats_time_to_first_response = get_stats_time_to_first_response(issues_with_metrics)
@@ -307,6 +333,10 @@ def main():
 
     stats_time_to_answer = get_stats_time_to_answer(issues_with_metrics)
 
+    num_mentor_count = 0
+    if enable_mentor_count:
+        num_mentor_count = get_mentor_count(issues_with_metrics, min_mentor_count)
+
     # Get stats describing the time in label for each label and store it in a dictionary
     # where the key is the label and the value is the average time
     stats_time_in_labels = get_stats_time_in_labels(issues_with_metrics, labels)
@@ -320,6 +350,7 @@ def main():
         stats_time_in_labels,
         num_issues_open,
         num_issues_closed,
+        num_mentor_count,
         search_query,
     )
     write_to_markdown(
@@ -330,6 +361,7 @@ def main():
         stats_time_in_labels,
         num_issues_open,
         num_issues_closed,
+        num_mentor_count,
         labels,
         search_query,
     )

json_writer.py

@@ -30,6 +30,7 @@ def write_to_json(
     stats_time_in_labels: Union[dict[str, dict[str, timedelta]], None],
     num_issues_opened: Union[int, None],
     num_issues_closed: Union[int, None],
+    num_mentor_count: Union[int, None],
     search_query: str,
 ) -> str:
     """
@@ -42,6 +43,7 @@ def write_to_json(
         "average_time_to_answer": "1 day, 0:00:00",
         "num_items_opened": 2,
         "num_items_closed": 1,
+        "num_mentor_count": 5,
         "total_item_count": 2,
         "issues": [
             {
@@ -129,6 +131,7 @@ def write_to_json(
         "90_percentile_time_in_labels": p90_time_in_labels,
         "num_items_opened": num_issues_opened,
         "num_items_closed": num_issues_closed,
+        "num_mentor_count": num_mentor_count,
         "total_item_count": len(issues_with_metrics),
     }
 

markdown_writer.py

@@ -13,6 +13,7 @@
         average_time_to_answer: timedelta,
         num_issues_opened: int,
         num_issues_closed: int,
+        num_mentor_count: int,
         file: file object = None
     ) -> None:
         Write the issues with metrics to a markdown file.
@@ -79,6 +80,7 @@ def write_to_markdown(
     average_time_in_labels: Union[dict, None],
     num_issues_opened: Union[int, None],
     num_issues_closed: Union[int, None],
+    num_mentor_count: Union[int, None],
     labels=None,
     search_query=None,
     hide_label_metrics=False,
@@ -95,7 +97,8 @@ def write_to_markdown(
         file (file object, optional): The file object to write to. If not provided,
             a file named "issue_metrics.md" will be created.
         num_issues_opened (int): The Number of items that remain opened.
-        num_issues_closed (int): The number of issues that were closed.
+        num_issues_closed (int): The number of issues that were closedi.
+        num_mentor_count (int): The number of very active commentors.
         labels (List[str]): A list of the labels that are used in the issues.
         search_query (str): The search query used to find the issues.
         hide_label_metrics (bool): Represents whether the user has chosen to hide label metrics in the output
@@ -127,6 +130,7 @@ def write_to_markdown(
             average_time_in_labels,
             num_issues_opened,
             num_issues_closed,
+            num_mentor_count,
             labels,
             columns,
             file,
@@ -184,6 +188,7 @@ def write_overall_metrics_tables(
     stats_time_in_labels,
     num_issues_opened,
     num_issues_closed,
+    num_mentor_count,
     labels,
     columns,
     file,
@@ -246,4 +251,5 @@ def write_overall_metrics_tables(
     file.write("| --- | ---: |\n")
     file.write(f"| Number of items that remain open | {num_issues_opened} |\n")
     file.write(f"| Number of items closed | {num_issues_closed} |\n")
+    file.write(f"| Number of most active mentors | {num_mentor_count} |\n")
     file.write(f"| Total number of items created | {len(issues_with_metrics)} |\n\n")