Scaling points for PRs based on size of the PR

app/api/leaderboard/functions.ts

@@ -15,7 +15,6 @@ export const getLeaderboardData = async (
   roles: ("core" | "intern" | "operations" | "contributor")[] = [],
 ) => {
   const contributors = await getContributors();
-
   const data = contributors
     .filter((a) => a.highlights.points)
     .map((contributor) => ({

lib/api.ts

@@ -115,9 +115,17 @@ export async function getContributorBySlug(file: string, detail = false) {
       return {
         activity: [
           ...acc.activity,
-          { ...activity, points: points[activity.type] || 0 },
+          {
+            ...activity,
+            points:
+              (points[activity.type] || 0) +
+              (activity.linked_issues?.length || 0),
+          },
         ],
-        points: acc.points + (points[activity.type] || 0),
+        points:
+          acc.points +
+          (points[activity.type] || 0) +
+          (activity.linked_issues?.length || 0),
         comment_created:
           acc.comment_created + (activity.type === "comment_created" ? 1 : 0),
         eod_update: acc.eod_update + (activity.type === "eod_update" ? 1 : 0),
@@ -131,6 +139,8 @@ export async function getContributorBySlug(file: string, detail = false) {
           acc.issue_assigned + (activity.type === "issue_assigned" ? 1 : 0),
         issue_opened:
           acc.issue_opened + (activity.type === "issue_opened" ? 1 : 0),
+        linked_issues:
+          acc.linked_issues + (activity.linked_issues?.length || 0),
       };
     },
     {
@@ -144,6 +154,7 @@ export async function getContributorBySlug(file: string, detail = false) {
       pr_reviewed: 0,
       issue_assigned: 0,
       issue_opened: 0,
+      linked_issues: 0,
     } as Highlights & { activity: Activity[] },
   );
 
@@ -181,12 +192,13 @@ export async function getContributorBySlug(file: string, detail = false) {
       pr_collaborated: weightedActivity.pr_collaborated,
       issue_assigned: weightedActivity.issue_assigned,
       issue_opened: weightedActivity.issue_opened,
+      linked_issues: weightedActivity.linked_issues,
     },
     weekSummary: getLastWeekHighlights(calendarData),
     summarize,
     calendarData: detail ? calendarData : [],
     ...data,
-  } as Contributor & { summarize: typeof summarize };
+  } as unknown as Contributor & { summarize: typeof summarize };
 }
 
 let contributors: Awaited<ReturnType<typeof getContributorBySlug>>[] | null =
@@ -218,6 +230,15 @@ function getCalendarData(activity: Activity[]) {
       } else {
         acc[date][activity.type] = 1;
       }
+
+      if (activity.type === "pr_merged") {
+        if (acc[date]["linked_issues"]) {
+          acc[date]["linked_issues"] += activity.linked_issues?.length || 0;
+        } else {
+          acc[date]["linked_issues"] = activity.linked_issues?.length || 0;
+        }
+      }
+
       if (!acc[date].types.includes(activity.type)) {
         acc[date].types.push(activity.type);
         // console.log(activity.type);
@@ -263,9 +284,11 @@ const computePoints = (
   calendarDataEntry: Highlights,
   initialPoints: number,
 ) => {
-  return HIGHLIGHT_KEYS.map(
+  let totalPoints = HIGHLIGHT_KEYS.map(
     (key) => points[key] * (calendarDataEntry[key] ?? 0),
   ).reduce((a, b) => a + b, initialPoints);
+  totalPoints += calendarDataEntry.linked_issues || 0;
+  return totalPoints;
 };
 
 const HighlightsReducer = (acc: Highlights, day: Highlights) => {
@@ -279,6 +302,7 @@ const HighlightsReducer = (acc: Highlights, day: Highlights) => {
     pr_collaborated: acc.pr_collaborated + (day.pr_collaborated ?? 0),
     issue_assigned: acc.issue_assigned + (day.issue_assigned ?? 0),
     issue_opened: acc.issue_opened + (day.issue_opened ?? 0),
+    linked_issues: acc.linked_issues + (day.linked_issues ?? 0),
   };
 };
 
@@ -292,6 +316,7 @@ const HighlightsInitialValue = {
   pr_collaborated: 0,
   issue_assigned: 0,
   issue_opened: 0,
+  linked_issues: 0,
 } as Highlights;
 
 const getLastWeekHighlights = (calendarData: Highlights[]) => {

lib/types.ts

@@ -42,6 +42,7 @@ export interface Highlights {
   pr_collaborated: number;
   issue_assigned: number;
   issue_opened: number;
+  linked_issues: number;
 }
 
 export interface WeekSummary {
@@ -76,6 +77,7 @@ export interface Activity {
   text: string;
   collaborated_with?: string[];
   turnaround_time?: number;
+  linked_issues?: string[]; //issues url, which are linked to merged prs
 }
 
 export interface OpenPr {

scraper/src/github.py

@@ -9,7 +9,7 @@
 from pathlib import Path
 from urllib.parse import parse_qsl, urlparse
 from zoneinfo import ZoneInfo
-
+from linked_issue_parser import LinkedIssueParser
 import requests
 
 logging.basicConfig(
@@ -71,6 +71,14 @@ def append(self, user, event):
                 "authored_issue_and_pr": [],
             }
 
+    def parse_linked_issues(self, pr_body):
+        if isinstance(pr_body, str):
+            pattern = r"#(\d+)"
+            matches = re.findall(pattern, pr_body)
+            return len(set(matches))
+        else:
+            return 0  
+
     def parse_event(self, event, event_time):
         user = event["actor"]["login"]
         try:
@@ -124,7 +132,7 @@ def parse_event(self, event, event_time):
                         "title": f'{event["repo"]["name"]}#{event["payload"]["pull_request"]["number"]}',
                         "time": event_time,
                         "link": event["payload"]["pull_request"]["html_url"],
-                        "text": event["payload"]["pull_request"]["title"],
+                        "text": event["payload"]["pull_request"]["title"]
                     },
                 )
 
@@ -133,6 +141,15 @@ def parse_event(self, event, event_time):
                 and event["payload"]["pull_request"]["merged"]
             ):
                 turnaround_time = self.caclculate_turnaround_time(event)
+                pr_body = event["payload"]["pull_request"]["body"]
+                repo = event["repo"]["name"]
+                parts = repo.split('/')
+                org_name = parts[0]
+                repo_name = parts[1]
+                pr_no = event['payload']['pull_request']['number']
+                linked_issue_parser = LinkedIssueParser(org=org_name,repo=repo_name,pr_no=pr_no,pr_body=pr_body)
+                linked_issues = linked_issue_parser.parse_linked_issues()
+                self.log.debug(f'linked_issues for pr {pr_no} are {linked_issues}')
                 self.append(
                     event["payload"]["pull_request"]["user"]["login"],
                     {
@@ -142,6 +159,7 @@ def parse_event(self, event, event_time):
                         "link": event["payload"]["pull_request"]["html_url"],
                         "text": event["payload"]["pull_request"]["title"],
                         "turnaround_time": turnaround_time,
+                        "linked_issues" : linked_issues
                     },
                 )
 

scraper/src/linked_issue_parser.py

@@ -0,0 +1,113 @@
+import re
+from os import getenv
+import requests
+
+class LinkedIssueParser:
+    def __init__(self,org:str, repo:str, pr_no:int, pr_body: str):
+        self.repo = repo
+        self.pr_no = pr_no
+        self.pr_body = pr_body
+        self.org = org
+
+    # The logic here is as follows:
+    #     Get a list of all events on a Pull Request of the type CONNECTED_EVENT and DISCONNECTED_EVENT
+    #     Create a map, keyed by Issue number and keep a count of how may times the issue is CONNECTED and DISCONNECTED
+    #     From that map, look for keys that have an odd-numbered count, as these are the events that have been CONNECTED that don't have a corresponding DISCONNECTED event.
+
+    def parse_ui_linked_issues(self):
+        query = """
+        {{
+        resource(url: "https://github.com/{org}/{repo}/pull/{pr_no}") {{
+            ... on PullRequest {{
+            timelineItems(itemTypes: [CONNECTED_EVENT, DISCONNECTED_EVENT], first: 100) {{
+                nodes {{
+                ... on ConnectedEvent {{
+                    id
+                    subject {{
+                    ... on Issue {{
+                        number
+                    }}
+                    }}
+                }}
+                ... on DisconnectedEvent {{
+                    id
+                    subject {{
+                    ... on Issue {{
+                        number
+                    }}
+                    }}
+                }}
+                }}
+            }}
+            }}
+        }}
+        }}
+        """.format(org = self.org, repo = self.repo, pr_no = self.pr_no)
+        gh_url = 'https://api.github.com/graphql'
+        token = getenv('GITHUB_TOKEN')
+        headers = {
+            'Authorization': f'Bearer {token}',
+            'Content-Type':'application/json'
+        }
+        response = requests.post(gh_url,headers=headers,json={'query':query})
+        if response.status_code != 200:
+            return []
+        data = response.json()
+        issues = {}
+        for node in data['data']['resource']['timelineItems']['nodes']:
+            issue_number = node['subject']['number'] 
+            if issue_number in issues:
+                issues[issue_number] +=1
+            else: 
+                issues[issue_number] = 1
+
+        linked_issues = []
+        for issue, count in issues.items():
+            if count % 2 != 0:
+                linked_issues.append(f'https://github.com/{self.org}/{self.repo}/issues/{issue}')
+        return linked_issues
+
+    def get_concat_commit_messages(self):
+        commit_url = f'https://api.github.com/repos/{self.org}/{self.repo}/pulls/{self.pr_no}/commits'
+        resposne = requests.get(commit_url)
+        if resposne.status_code != 200:
+            return ""
+        json_data = resposne.json()
+        result = ""
+        for commit in json_data:
+            message = commit['commit']['message']
+            result = f'{result} , {message}'
+        return result
+
+    def parse_desc_linked_issues(self):
+        pattern_same_repo = r'\b(?:close[sd]?|fix(?:e[sd])?|resolve[sd]?)\s+#(\d+)'
+        pattern_other_repo = r'\b(?:close[sd]?|fix(?:e[sd])?|resolve[sd]?)\s+(\S+\/\S+)#(\d+)'
+        commit_messages = self.get_concat_commit_messages()
+        text = f'{self.pr_body} {commit_messages}'
+        same_repo_linked_issues = re.findall(pattern_same_repo,text,re.IGNORECASE)
+        other_repo_linked_issues = re.findall(pattern_other_repo,text,re.IGNORECASE)
+        linked_issues = set([])
+        for issue in same_repo_linked_issues:
+            linked_issues.add(issue)
+        for issue in other_repo_linked_issues:
+            linked_issues.add(issue)
+        linked_issues_url = []
+        for issue in linked_issues:
+            if isinstance(issue, str):
+                linked_issues_url.append(f'https://github.com/{self.org}/{self.repo}/issues/{issue}')
+            elif isinstance(issue, tuple):
+                linked_issues_url.append(f'https://github.com/{issue[0]}/issues/{issue[1]}')
+            continue
+        return linked_issues_url
+
+
+    def parse_linked_issues(self):
+        linked_issues = []
+        ui_linked_issues = self.parse_ui_linked_issues()
+        desc_linked_issues = self.parse_desc_linked_issues()
+        for issue in ui_linked_issues:
+            linked_issues.append(issue)
+        for issue in desc_linked_issues:
+            linked_issues.append(issue)
+        return linked_issues
+