Update Bitbucket Pipeline instructions in INSTALL.md, remove redundant functionality

Get previous incremental review

Dockerfile.bitbucket_pipeline

@@ -1,18 +0,0 @@
-FROM python:3.10 as base
-
-ENV OPENAI_API_KEY=${OPENAI_API_KEY} \
-    BITBUCKET_BEARER_TOKEN=${BITBUCKET_BEARER_TOKEN} \
-    BITBUCKET_PR_ID=${BITBUCKET_PR_ID} \
-    BITBUCKET_REPO_SLUG=${BITBUCKET_REPO_SLUG} \
-    BITBUCKET_WORKSPACE=${BITBUCKET_WORKSPACE}
-
-
-WORKDIR /app
-ADD pyproject.toml .
-ADD requirements.txt .
-RUN pip install . && rm pyproject.toml requirements.txt
-ENV PYTHONPATH=/app
-ADD pr_agent pr_agent
-ADD bitbucket_pipeline/entrypoint.sh /
-RUN chmod +x /entrypoint.sh
-ENTRYPOINT ["/entrypoint.sh"]

INSTALL.md

@@ -375,59 +375,28 @@ In the "Trigger" section, check the ‘comments’ and ‘merge request events
 ### Method 9: Run as a Bitbucket Pipeline
 
 
-You can use our pre-build Bitbucket-Pipeline docker image to run as Bitbucket-Pipeline.
+You can use the Bitbucket Pipeline system to run PR-Agent on every pull request open or update.
 
 1. Add the following file in your repository bitbucket_pipelines.yml
 
 ```yaml
-  pipelines:
+pipelines:
     pull-requests:
       '**':
         - step:
-            name: PR Agent Pipeline
+            name: PR Agent Review
-            caches:
+            image: python:3.10
-              - pip
-            image: python:3.8
             services:
               - docker
             script:
-              - git clone https://github.com/Codium-ai/pr-agent.git
+              - docker run -e CONFIG.GIT_PROVIDER=bitbucket -e OPENAI.KEY=$OPENAI_API_KEY -e BITBUCKET.BEARER_TOKEN=$BITBUCKET_BEARER_TOKEN codiumai/pr-agent:latest --pr_url=https://bitbucket.org/$BITBUCKET_WORKSPACE/$BITBUCKET_REPO_SLUG/pull-requests/$BITBUCKET_PR_ID review
-              - cd pr-agent
-              - docker build -t bitbucket_runner:latest -f Dockerfile.bitbucket_pipeline .
-              - docker run -e OPENAI_API_KEY=$OPENAI_API_KEY -e BITBUCKET_BEARER_TOKEN=$BITBUCKET_BEARER_TOKEN -e BITBUCKET_PR_ID=$BITBUCKET_PR_ID -e BITBUCKET_REPO_SLUG=$BITBUCKET_REPO_SLUG -e BITBUCKET_WORKSPACE=$BITBUCKET_WORKSPACE bitbucket_runner:latest
 ```
 
-2. Add the following secret to your repository under Repository settings > Pipelines > Repository variables.
+2. Add the following secure variables to your repository under Repository settings > Pipelines > Repository variables.
 OPENAI_API_KEY: <your key>
 BITBUCKET_BEARER_TOKEN: <your token>
 
-3. To get BITBUCKET_BEARER_TOKEN follow these steps
+You can get a Bitbucket token for your repository by following Repository Settings -> Security -> Access Tokens
-  So here is my step by step tutorial
-  i) Insert your workspace name instead of {workspace_name} and go to the following link in order to create an OAuth consumer.
-
-      https://bitbucket.org/{workspace_name}/workspace/settings/api
-
-      set callback URL to http://localhost:8976 (doesn't need to be a real server there)
-      select permissions: repository -> read
-
-  ii) use consumer's Key as a {client_id} and open the following URL in the browser
-
-      https://bitbucket.org/site/oauth2/authorize?client_id={client_id}&response_type=code
-
-  iii)
-      after you press "Grant access" in the browser it will redirect you to
-
-      http://localhost:8976?code=<CODE>
-
-  iv) use the code from the previous step and consumer's Key as a {client_id}, and consumer's Secret as {client_secret}
-
-      curl -X POST -u "{client_id}:{client_secret}" \
-          https://bitbucket.org/site/oauth2/access_token \
-          -d grant_type=authorization_code \
-          -d code={code} \
-
-
-After completing this steps, you just to place this access token in the repository varibles.
 
 
 =======

bitbucket_pipeline/entrypoint.sh

@@ -1,2 +0,0 @@
-#!/bin/bash
-python /app/pr_agent/servers/bitbucket_pipeline_runner.py

pr_agent/git_providers/azuredevops_provider.py

@@ -236,9 +236,6 @@ class AzureDevopsProvider:
     def _parse_pr_url(pr_url: str) -> Tuple[str, int]:
         parsed_url = urlparse(pr_url)
 
-        if 'azure.com' not in parsed_url.netloc:
-            raise ValueError("The provided URL is not a valid Azure DevOps URL")
-
         path_parts = parsed_url.path.strip('/').split('/')
 
         if len(path_parts) < 6 or path_parts[4] != 'pullrequest':

pr_agent/git_providers/github_provider.py

@@ -77,7 +77,7 @@ class GithubProvider(GitProvider):
         self.previous_review = None
         self.comments = list(self.pr.get_issue_comments())
         for index in range(len(self.comments) - 1, -1, -1):
-            if self.comments[index].body.startswith("## PR Analysis"):
+            if self.comments[index].body.startswith("## PR Analysis") or self.comments[index].body.startswith("## Incremental PR Review"):
                 self.previous_review = self.comments[index]
                 break
 

pr_agent/servers/bitbucket_pipeline_runner.py

@@ -1,34 +0,0 @@
-import os
-from pr_agent.agent.pr_agent import PRAgent
-from pr_agent.config_loader import get_settings
-from pr_agent.tools.pr_reviewer import PRReviewer
-import asyncio
-
-async def run_action():
-    try:
-        pull_request_id = os.environ.get("BITBUCKET_PR_ID", '')
-        slug = os.environ.get("BITBUCKET_REPO_SLUG", '')
-        workspace = os.environ.get("BITBUCKET_WORKSPACE", '')
-        bearer_token = os.environ.get('BITBUCKET_BEARER_TOKEN', None)
-        OPENAI_KEY = os.environ.get('OPENAI_API_KEY') or os.environ.get('OPENAI.KEY')
-        OPENAI_ORG = os.environ.get('OPENAI_ORG') or os.environ.get('OPENAI.ORG')
-        # Check if required environment variables are set
-        if not bearer_token:
-            print("BITBUCKET_BEARER_TOKEN not set")
-            return
-        
-        if not OPENAI_KEY:
-            print("OPENAI_KEY not set")
-            return
-        # Set the environment variables in the settings
-        get_settings().set("BITBUCKET.BEARER_TOKEN", bearer_token)
-        get_settings().set("OPENAI.KEY", OPENAI_KEY)
-        if OPENAI_ORG:
-            get_settings().set("OPENAI.ORG", OPENAI_ORG)
-        if pull_request_id and slug and workspace:
-            pr_url = f"https://bitbucket.org/{workspace}/{slug}/pull-requests/{pull_request_id}"
-            await PRReviewer(pr_url).run()
-    except Exception as e:
-        print(f"An error occurred: {e}")
-if __name__ == "__main__":
-    asyncio.run(run_action())

pr_agent/tools/pr_similar_issue.py

@@ -1,3 +1,4 @@
+import time
 from enum import Enum
 from typing import List
 
@@ -46,6 +47,13 @@ class PRSimilarIssue:
 
         # check if index exists, and if repo is already indexed
         run_from_scratch = False
+        if run_from_scratch:  # for debugging
+            if not index_name in pinecone.list_indexes():
+                get_logger().info('Removing index...')
+                pinecone.init(api_key=api_key, environment=environment)
+                pinecone.delete_index(index_name)
+                get_logger().info('Done')
+
         upsert = True
         pinecone.init(api_key=api_key, environment=environment)
         if not index_name in pinecone.list_indexes():
@@ -116,7 +124,16 @@ class PRSimilarIssue:
         relevant_comment_number_list = []
         score_list = []
         for r in res['matches']:
-            issue_number = int(r["id"].split('.')[0].split('_')[-1])
+            # skip example issue
+            if 'example_issue_' in r["id"]:
+                continue
+
+            try:
+                issue_number = int(r["id"].split('.')[0].split('_')[-1])
+            except:
+                get_logger().debug(f"Failed to parse issue number from {r['id']}")
+                continue
+
             if original_issue_number == issue_number:
                 continue
             if issue_number not in relevant_issues_number_list:
@@ -237,6 +254,7 @@ class PRSimilarIssue:
         if not upsert:
             get_logger().info('Creating index from scratch...')
             ds.to_pinecone_index(self.index_name, api_key=api_key, environment=environment)
+            time.sleep(15)  # wait for pinecone to finalize indexing before querying
         else:
             get_logger().info('Upserting index...')
             namespace = ""
@@ -244,6 +262,7 @@ class PRSimilarIssue:
             concurrency: int = 10
             pinecone.init(api_key=api_key, environment=environment)
             ds._upsert_to_index(self.index_name, namespace, batch_size, concurrency)
+            time.sleep(5)  # wait for pinecone to finalize upserting before querying
         get_logger().info('Done')