Also parse Github URLs

This will replace valid Github URLs with tags.

bot.py

@@ -19,7 +19,8 @@ GITHUB_TOKEN=os.environ.get('GITHUB_TOKEN')
 DEFAULT_AVATAR_URL=os.environ.get('DEFAULT_AVATAR_URL')
 DEFAULT_REPOSITORY=os.environ.get('DEFAULT_REPOSITORY')
 
-RE_MATCH = re.compile('([A-Za-z0-9_.-]+)?#(\d+)')
+RE_TAG_PROG = re.compile('([A-Za-z0-9_.-]+)?#(\d+)')
+RE_URL_PROG = re.compile('github.com/([A-Za-z0-9_.-]+)/([A-Za-z0-9_.-]+)/(issues|pulls)/(\d+)\S*')
 
 def debug_print(msg):
     if DEBUG:
@@ -83,11 +84,147 @@ class Bot:
         }
         self.send(subscribe_msg)
 
-    def update_msg(self, msg):
+    def format_issue(self, repository, issue):
+        headers = { 'User-Agent': 'Godot Issuebot by hpvb', }
+        url = f"https://api.github.com/repos/{GITHUB_PROJECT}/{repository}/issues/{issue}"
+        debug_print(f"GitHub API request: {url}")
+
+        r = requests.get(url, headers=headers, auth=(GITHUB_USERNAME, GITHUB_TOKEN))
+        if r.status_code != 200:
+            debug_print(f"Github API returned an error {r.status_code}")
+            debug_print(r.content)
+            return None
+
+        issue = r.json()
+
+        avatar_url = DEFAULT_AVATAR_URL
+        if 'avatar_url' in issue['user'] and issue['user']['avatar_url']:
+            avatar_url = issue['user']['avatar_url']
+        if 'gravatar_id' in issue['user'] and issue['user']['gravatar_id']:
+            avatar_url = f"https://www.gravatar.com/avatar/{issue['user']['gravatar_id']}"
+
+        is_pr = False
+        pr_mergeable = None
+        pr_merged = None
+        pr_merged_by = None
+        pr_draft = False
+        pr_reviewers = None
+        status = None
+        closed_by = None
+
+        if 'pull_request' in issue and issue['pull_request']:
+            is_pr = True
+            debug_print(f"GitHub API request: {issue['pull_request']['url']}")
+
+            prr = requests.get(issue['pull_request']['url'], headers=headers, auth=(GITHUB_USERNAME, GITHUB_TOKEN))
+            if prr.status_code == 200:
+                pr = prr.json()
+                status = pr['state']
+
+                if 'merged_by' in pr and pr['merged_by']:
+                    pr_merged_by = pr['merged_by']['login']
+                if 'mergeable' in pr:
+                    pr_mergeable = pr['mergeable']
+                if 'merged' in pr:
+                    pr_merged = pr['merged']
+                if 'draft' in pr:
+                    pr_draft = pr['draft']
+                if 'requested_reviewers' in pr and pr['requested_reviewers']:
+                    reviewers = []
+                    for reviewer in pr['requested_reviewers']:
+                        reviewers.append(reviewer['login'])
+                    pr_reviewers = ', '.join(reviewers)
+                if 'requested_teams' in pr and pr['requested_teams']:
+                    teams = []
+                    for team in pr['requested_teams']:
+                        teams.append(f"team:{team['name']}")
+                    if pr_reviewers:
+                        pr_reviewers += ' and '
+                        pr_reviewers += ', '.join(teams)
+                    else:
+                        pr_reviewers = ', '.join(teams)
+
+        else:
+            status = issue['state']
+
+        if status == 'closed':
+            if 'closed_by' in issue and issue['closed_by']:
+                closed_by = issue['closed_by']['login']
+
+        issue_type = None
+
+        if is_pr:
+            issue_type = "Pull Request"
+            if pr_merged:
+                status = "PR merged"
+                if pr_merged_by:
+                    status += f" by {pr_merged_by}"
+            elif status == 'closed':
+                status = "PR closed"
+            elif not pr_merged:
+                status = "PR open"
+                if pr_draft:
+                    status += " [draft]"
+                if pr_mergeable != None:
+                    if pr_mergeable:
+                        status += " [mergeable]"
+                    else:
+                        status += " [needs rebase]"
+                if pr_reviewers:
+                    status += f" reviews required from {pr_reviewers}"
+                
+        else:
+            issue_type = "Issue"
+            status = f"Status: {status}"
+
+        if not pr_merged and closed_by and status == 'closed':
+            status += f" by {closed_by}"
+
+        return {
+            "author_icon": avatar_url,
+            "author_link": issue['html_url'],
+            "author_name": f"{repository.title()} [{issue_type}]: {issue['title']} #{issue['number']}",
+            "text": status,
+        }
+
+    def replace_issue_tags(self, msg):
         debug_print("Updating message!")
         links = []
 
-        for match in re.finditer(RE_MATCH, msg['msg']):
+        # First replace all the full links that rocket.chat has detected with tags
+        if 'urls' in msg and msg['urls']:
+            urls_to_keep = []
+            for url in msg['urls']:
+                debug_print(f"URL: {url['url']}")
+                if f'github.com/{GITHUB_PROJECT}' in url['url']:
+
+                    match = re.search(RE_URL_PROG, url['url'])
+                    repository = match.group(2)
+                    issue = int(match.group(4))
+
+                    tag = f'{repository}#{issue}'
+                    debug_print(f"Replacing url {url['url']} with {tag}")
+
+                    msg['msg'] = msg['msg'].replace(url['url'], tag)
+                    continue
+                urls_to_keep.append(url)
+
+            msg['urls'] = urls_to_keep
+
+        # Then we replace all of the part-urls with tags as well
+        for match in re.finditer(RE_URL_PROG, msg['msg']):
+            project = match.group(1)
+            repository = match.group(2)
+            issue = match.group(4)
+
+            tag = f'{repository}#{issue}'
+            if project == GITHUB_PROJECT:
+                print('DEBUG: ' + match.group(0))
+                msg['msg'] = msg['msg'].replace(match.group(0), tag)
+
+        # Then finally add the metadata for all our tags
+        debug_print("Scanning message for tags")
+        for match in re.finditer(RE_TAG_PROG, msg['msg']):
             repository = match.group(1)
             issue = int(match.group(2))
 
@@ -102,108 +239,10 @@ class Bot:
 
             debug_print(f"Message contains issue for {repository}")
 
-            headers = { 'User-Agent': 'Godot Issuebot by hpvb', }
+            link = self.format_issue(repository, issue)
-            url = f"https://api.github.com/repos/{GITHUB_PROJECT}/{repository}/issues/{issue}"
+            if link:
-            debug_print(f"GitHub API request: {url}")
+                links.append(link)
 
-            r = requests.get(url, headers=headers, auth=(GITHUB_USERNAME, GITHUB_TOKEN))
-            if r.status_code != 200:
-                debug_print(f"Github API returned an error {r.status_code}")
-                debug_print(r.content)
-                continue
-
-            issue = r.json()
-
-            avatar_url = DEFAULT_AVATAR_URL
-            if 'avatar_url' in issue['user'] and issue['user']['avatar_url']:
-                avatar_url = issue['user']['avatar_url']
-            if 'gravatar_id' in issue['user'] and issue['user']['gravatar_id']:
-                avatar_url = f"https://www.gravatar.com/avatar/{issue['user']['gravatar_id']}"
-
-            is_pr = False
-            pr_mergeable = None
-            pr_merged = None
-            pr_merged_by = None
-            pr_draft = False
-            pr_reviewers = None
-            status = None
-            closed_by = None
-
-            if 'pull_request' in issue and issue['pull_request']:
-                is_pr = True
-                debug_print(f"GitHub API request: {issue['pull_request']['url']}")
-
-                prr = requests.get(issue['pull_request']['url'], headers=headers, auth=(GITHUB_USERNAME, GITHUB_TOKEN))
-                if prr.status_code == 200:
-                    pr = prr.json()
-                    status = pr['state']
-
-                    if 'merged_by' in pr and pr['merged_by']:
-                        pr_merged_by = pr['merged_by']['login']
-                    if 'mergeable' in pr:
-                        pr_mergeable = pr['mergeable']
-                    if 'merged' in pr:
-                        pr_merged = pr['merged']
-                    if 'draft' in pr:
-                        pr_draft = pr['draft']
-                    if 'requested_reviewers' in pr and pr['requested_reviewers']:
-                        reviewers = []
-                        for reviewer in pr['requested_reviewers']:
-                            reviewers.append(reviewer['login'])
-                        pr_reviewers = ', '.join(reviewers)
-                    if 'requested_teams' in pr and pr['requested_teams']:
-                        teams = []
-                        for team in pr['requested_teams']:
-                            teams.append(f"team:{team['name']}")
-                        if pr_reviewers:
-                            pr_reviewers += ' and '
-                            pr_reviewers += ', '.join(teams)
-                        else:
-                            pr_reviewers = ', '.join(teams)
-
-            else:
-                status = issue['state']
-
-            if status == 'closed':
-                if 'closed_by' in issue and issue['closed_by']:
-                    closed_by = issue['closed_by']['login']
-
-            issue_type = None
-
-            if is_pr:
-                issue_type = "Pull Request"
-                if pr_merged:
-                    status = "PR merged"
-                    if pr_merged_by:
-                        status += f" by {pr_merged_by}"
-                elif status == 'closed':
-                    status = "PR closed"
-                elif not pr_merged:
-                    status = "PR open"
-                    if pr_draft:
-                        status += " [draft]"
-                    if pr_mergeable != None:
-                        if pr_mergeable:
-                            status += " [mergeable]"
-                        else:
-                            status += " [needs rebase]"
-                    if pr_reviewers:
-                        status += f" reviews required from {pr_reviewers}"
-                    
-            else:
-                issue_type = "Issue"
-                status = f"Status: {status}"
-
-            if not pr_merged and closed_by and status == 'closed':
-                status += f" by {closed_by}"
-
-            links.append({
-                "author_icon": avatar_url,
-                "author_link": issue['html_url'],
-                "author_name": f"{repository.title()} [{issue_type}]: {issue['title']} #{issue['number']}",
-                "text": status,
-                "ts": msg['ts'],
-            })
 
         if not len(links):
             return
@@ -222,6 +261,10 @@ class Bot:
         # Hack Hack, the clients won't update without a change to this field. Even if we add or remove attachments.
         msg['msg'] = msg['msg'] + " "
 
+        # Add timestamp to all attachments. These are visible in the mobile client.
+        for link in links:
+            link['ts'] = msg['ts'],
+
         [msg['attachments'].append(x) for x in links if x not in msg['attachments']] 
 
         update_msg = {
@@ -265,9 +308,9 @@ class Bot:
                 for chat_msg in decoded_msg['fields']['args']:
                     if 'editedBy' in chat_msg and chat_msg['editedBy']['_id'] == self.id:
                         continue
-                    if re.match(RE_MATCH, chat_msg['msg']):
+                    if re.search(RE_TAG_PROG, chat_msg['msg']) or re.search(RE_URL_PROG, chat_msg['msg']):
                         debug_print("Sending message to be update")
-                        self.update_msg(chat_msg)
+                        self.replace_issue_tags(chat_msg)
 
     def on_error(self, ws, error):
         debug_print(error)

test-regex.py

@@ -6,7 +6,7 @@ def makeurl(issue, repo = ''):
     if not repo:
         repo = 'godot'
 
-    return f'https://github.com/{repo}/issues/{issue}'
+    return f'https://github.com/godotengine/{repo}/issues/{issue}'
 
 tests = [
     { 'text': '#100', 'results' : [ makeurl(100) ] },
@@ -26,19 +26,28 @@ tests = [
     { 'text': '(#100) text', 'results' : [ makeurl(100) ] },
     { 'text': '(repo#100) text', 'results' : [ makeurl(100, 'repo') ] },
 
+    { 'text': 'https://github.com/godotengine/issue-bot/issues/2', 'results': [ makeurl(2, 'issue-bot') ] },
+    { 'text': 'https://github.com/godotengine/godot/pulls/100', 'results': [ makeurl(100) ] },
+    { 'text': 'https://github.com/godotengine/godot/pulls/100#issuecomment-1', 'results': [ makeurl(100) ] },
+
+    { 'text': 'a long line of text with an url https://github.com/godotengine/godot/issues/100 and some tags #102 repo#103', 'results': [ makeurl(102), makeurl(103, 'repo'), makeurl(100) ] },
+
     { 'text': 'just a bunch of text', 'results' : [  ] },
     { 'text': 'Bunch of ## nonsense ##sdf $$', 'results' : [  ] },
 ]
 
-prog = re.compile('([A-Za-z0-9_.-]+)?#(\d+)')
+tag_prog = re.compile('([A-Za-z0-9_.-]+)?#(\d+)')
+url_prog = re.compile('github.com/([A-Za-z0-9_.-]+)/([A-Za-z0-9_.-]+)/(issues|pulls)/(\d+)\S*')
 for test in tests:
     text = test['text']
     result = []
 
-    print(re.match(prog, text))
+    for match in re.finditer(tag_prog, text):
-    for match in re.finditer(prog, text):
         result.append(makeurl(match.group(2), match.group(1)))
 
+    for match in re.finditer(url_prog, text):
+        result.append(makeurl(match.group(4), match.group(2)))
+
     if test['results'] != result:
         print(f'FAILED for {text}: expected {test["results"]} got: {result}')