chore: many production changes

README.md

@@ -4,6 +4,7 @@
 ![GFI-Bot Tests](https://github.com/osslab-pku/gfi-bot/actions/workflows/test-gfi-bot.yml/badge.svg)
 ![GFI-Bot Coverage](https://img.shields.io/codecov/c/github/osslab-pku/gfi-bot?label=GFI-Bot%20Coverage)
 ![License](https://img.shields.io/github/license/osslab-pku/gfi-bot?label=License)
+[![GFI-Bot]https://gfibot.io/api/repo/badge?owner=osslab-pku&name=gfi-bot)](https://gfibot.io/?owner=osslab-pku&name=gfi-bot)
 
 ML-powered 🤖 for finding and labeling good first issues in your GitHub project!
 
@@ -78,6 +79,10 @@ python -m gfibot.data.dataset --since=2008.01.01 --nprocess=4
 
 ### Model Training
 
+Model training can be simply done by running the following script.
 
+```shell script
+
+```
 
 ### Backend Deployment

frontend/.env.development

@@ -0,0 +1,2 @@
+REACT_APP_ENV = 'development'
+REACT_APP_BASE_URL ?= undefined

frontend/.env.production

@@ -0,0 +1,2 @@
+REACT_APP_ENV = 'production'
+REACT_APP_BASE_URL ?= 'https://gfibot.io/'

frontend/.gitignore

@@ -22,4 +22,4 @@ npm-debug.log*
 yarn-debug.log*
 yarn-error.log*
 
-.env.*
+# .env.*

frontend/package.json

@@ -6,7 +6,7 @@
     "@ant-design/icons": "^4.7.0",
     "@babel/core": "^7.16.0",
     "@pmmmwh/react-refresh-webpack-plugin": "^0.5.3",
-    "@svgr/webpack": "^5.5.0",
+    "@svgr/webpack": "^6.2.1",
     "@testing-library/jest-dom": "^5.16.1",
     "@testing-library/react": "^12.1.2",
     "@testing-library/user-event": "^13.5.0",

gfibot/backend/gfi_backend.py

@@ -922,7 +922,7 @@ def get_repo_badge():
             ).count()
             print(gfi_num)
             img_src = "https://img.shields.io/badge/{}-{}".format(
-                "Using - GFIBot", "success"
+                f"GFIBot - {gfi_num} recommended good first issues", "success"
             )
             svg = requests.get(img_src).content
             return app.response_class(svg, mimetype="image/svg+xml")

gfibot/data/dataset.py

@@ -289,8 +289,6 @@ def get_dataset(issue: Union[OpenIssue, ResolvedIssue], before: datetime) -> Dat
         logger.error(f"{issue.owner}/{issue.name}#{issue.number}: Pull Request")
         return
 
-    logger.info(f"{issue.owner}/{issue.name}#{issue.number} (before {before}) start")
-
     repo: Repo = Repo.objects(owner=issue.owner, name=issue.name).first()
     contribs, n_closed, n_open, close_times = _get_background_data(
         issue.owner, issue.name, before
@@ -355,7 +353,6 @@ def get_dataset(issue: Union[OpenIssue, ResolvedIssue], before: datetime) -> Dat
     data.event_users = event_users
 
     data.save()
-    logger.info(f"{issue.owner}/{issue.name}#{issue.number} (before {before}) is done")
     return data
 
 
@@ -435,6 +432,14 @@ def get_dataset_for_repo(
         ResolvedIssue.objects(repo_query & Q(resolved_at__gte=since))
     )
     open_issues = list(OpenIssue.objects(repo_query & Q(updated_at__gte=since)))
+
+    logger.info(
+        "%s/%s: start building dataset (%d resolved, %d open)",
+        owner,
+        name,
+        len(resolved_issues),
+        len(open_issues),
+    )
     get_dataset_with_issues(resolved_issues, open_issues)
 
     log.updated_open_issues = len(open_issues)

production/docker-compose.yaml

@@ -20,12 +20,13 @@ services:
         depends_on:
             - mongodb
 
-    gfi-bot-frontend:
-        build:
-            context: ../
-            dockerfile: ./production/gfibot.frontend.dockerfile
-        ports:
-            - "80:80"
-            - "443:443"
-        depends_on:
-            - gfi-bot 
+    # TODO: fix strange dependency bugs before using frontend docker
+    #gfi-bot-frontend:
+    #    build:
+    #        context: ../
+    #        dockerfile: ./production/gfibot.frontend.dockerfile
+    #    ports:
+    #        - "80:80"
+    #        - "443:443"
+    #    depends_on:
+    #        - gfi-bot 

production/gfibot.frontend.dockerfile

@@ -1,20 +1,15 @@
 FROM node:16
+ENV NODE_ENV=production
+
+WORKDIR /
+COPY ./ /gfi-bot
+WORKDIR /gfi-bot/frontend
+RUN npm ci --force
+RUN npm run build:prod
+
+FROM caddy:2.5.1
 
 EXPOSE 80
 EXPOSE 443
-
-# Install caddy
-RUN apt install -y debian-keyring debian-archive-keyring apt-transport-https
-RUN curl -1sLf 'https://dl.cloudsmith.io/public/caddy/stable/gpg.key' | gpg --dearmor -o /usr/share/keyrings/caddy-stable-archive-keyring.gpg
-RUN curl -1sLf 'https://dl.cloudsmith.io/public/caddy/stable/debian.deb.txt' | tee /etc/apt/sources.list.d/caddy-stable.list
-RUN apt update
-RUN apt install caddy
-
-WORKDIR /
-COPY ./ /gfi-bot/
-WORKDIR /gfi-bot/frontend
-RUN npm install --force
-RUN npm run build:prod
-
-WORKDIR /gfi-bot/
-RUN caddy run --config=/gfibot/production/Caddyfile
+WORKDIR /gfi-bot
+RUN caddy run --config=production/Caddyfile

production/set_tokens_private_production.py

@@ -4,7 +4,7 @@ import json
 import logging
 import argparse
 
-MONGO_URI: Final = "mongodb://mongodb:27017/"
+MONGO_URI: Final = "mongodb://mongodb:27020/"
 SECRET_FILE_NAME: Final = "production/production_secrets.json"
 
 db_client = pymongo.MongoClient(MONGO_URI)

pyproject.toml

@@ -104,7 +104,7 @@ projects = [ # This list of 100 projects comes from RecGFI paper
 github_graphql_schema_path = "gfibot/data/github_graphql_schema.graphql"
 
 [mongodb]
-url = "mongodb://localhost:27017"
+url = "mongodb://localhost:27020"
 db = "gfibot"
 
 [tool.poetry]