feat(cli): update argument handler to enable local analysis with docker

bases/ecoindex/cli/arguments_handler.py

@@ -1,9 +1,11 @@
 from tempfile import NamedTemporaryFile
 from typing import Set
-from urllib.parse import urlparse
+from urllib.parse import urlparse, urlunparse
 
 from ecoindex.cli.crawl import EcoindexSpider
+from ecoindex.cli.helper import replace_localhost_with_hostdocker
 from ecoindex.models import WindowSize
+
 from pydantic import AnyHttpUrl, validate_call
 from pydantic.types import FilePath
 from scrapy.crawler import CrawlerProcess
@@ -36,8 +38,10 @@ def get_urls_from_file(urls_file: FilePath) -> Set[str]:
 
 def get_urls_recursive(main_url: str) -> Set[str]:
     parsed_url = urlparse(main_url)
-    domain = parsed_url.netloc
-    main_url = f"{parsed_url.scheme}://{domain}"
+    host_infos = replace_localhost_with_hostdocker(parsed_url.netloc)
+    netloc = host_infos.netloc
+    domain = host_infos.domain
+    main_url = f"{parsed_url.scheme}://{netloc}"
     process = CrawlerProcess()
 
     with NamedTemporaryFile(mode="w+t") as temp_file:
@@ -58,6 +62,9 @@ def get_urls_recursive(main_url: str) -> Set[str]:
 def get_url_from_args(urls_arg: list[AnyHttpUrl]) -> set[AnyHttpUrl]:
     urls_from_args = set()
     for url in urls_arg:
+        parsed_url = urlparse(str(url))
+        host_infos = replace_localhost_with_hostdocker(parsed_url.netloc)
+        url = AnyHttpUrl(urlunparse((parsed_url.scheme, host_infos.netloc, parsed_url.path, parsed_url.params, parsed_url.query, parsed_url.fragment)))
         urls_from_args.add(url)
 
     return urls_from_args

bases/ecoindex/cli/helper.py

@@ -1,6 +1,7 @@
 from asyncio import run
+from ecoindex.config import Settings
 
-from ecoindex.models import Result, WindowSize
+from ecoindex.models import Result, WindowSize, CliHost
 from ecoindex.scraper import EcoindexScraper
 
 
@@ -36,3 +37,15 @@ def run_page_analysis(
             ),
             False,
         )
+
+
+def replace_localhost_with_hostdocker(netloc: str) -> CliHost:
+    if Settings().DOCKER_CONTAINER and "localhost" in netloc:
+        domain = "host.docker.internal"
+        netloc = netloc.replace("localhost", domain)
+    elif "localhost" in netloc :
+        domain = "localhost"
+    else :
+        domain = netloc
+
+    return CliHost(domain=domain, netloc=netloc)

components/ecoindex/config/settings.py

@@ -1,13 +1,13 @@
 from pydantic_settings import BaseSettings, SettingsConfigDict
 
-
 class Settings(BaseSettings):
     CORS_ALLOWED_CREDENTIALS: bool = True
     CORS_ALLOWED_HEADERS: list = ["*"]
     CORS_ALLOWED_METHODS: list = ["*"]
     CORS_ALLOWED_ORIGINS: list = ["*"]
     DAILY_LIMIT_PER_HOST: int = 0
     DATABASE_URL: str = "sqlite+aiosqlite:///db.sqlite3"
+    DOCKER_CONTAINER: bool = False
     DEBUG: bool = False
     ENABLE_SCREENSHOT: bool = False
     EXCLUDED_HOSTS: list[str] = ["localhost", "127.0.0.1"]

components/ecoindex/models/__init__.py

@@ -1,3 +1,6 @@
+from ecoindex.models.cli import (
+    CliHost,
+)
 from ecoindex.models.compute import (
     Ecoindex,
     PageMetrics,
@@ -19,6 +22,7 @@
 from ecoindex.models.sort import Sort
 
 __all__ = [
+    "CliHost",
     "Ecoindex",
     "example_daily_limit_response",
     "example_ecoindex_not_found",

components/ecoindex/models/cli.py

@@ -0,0 +1,6 @@
+from pydantic import BaseModel
+
+class CliHost(BaseModel):
+    domain: str
+    netloc: str
+

projects/ecoindex_cli/README.md

@@ -25,16 +25,27 @@ The output is a CSV or JSON file with the results of the analysis.
 
 The simplest way to start with ecoindex-cli is to install docker and then create an alias in your .bashrc or .zshrc file:
 
+For MacOS and Windows :
+
 ```bash
 alias ecoindex-cli="docker run -it --rm -v /tmp/ecoindex-cli:/tmp/ecoindex-cli vvatelot/ecoindex-cli:latest ecoindex-cli"
 ```
 
+For Linux :
+
+```bash
+alias ecoindex-cli="docker run -it --rm --add-host=host.docker.internal:host-gateway -v /tmp/ecoindex-cli:/tmp/ecoindex-cli vvatelot/ecoindex-cli:latest ecoindex-cli"
+```
+
 Then you can use the cli as if it was installed on your computer:
 
 ```bash
 ecoindex-cli --help
 ```
 
+To enable local testing we use the docker alias `host.docker.internal` to access your machine localhost.
+Currently, `host.docker.internal` is mapped to your IPv4 address. To facilitate local testing, ensure that your localhost is accessible via your IPv4 address (127.0.0.1).
+
 ## Use case
 
 The docker image [vvatelot/ecoindex-cli](https://hub.docker.com/r/vvatelot/ecoindex-cli) is available for `linux/amd64` and `linux/arm64` platforms and provides you an easy way to use this CLI on your environment.

projects/ecoindex_cli/dockerfile

@@ -8,8 +8,9 @@ RUN poetry export --output=requirements.txt --without-hashes
 
 
 FROM python:3.12-slim
-
-ARG wheel=ecoindex_cli-2.23.0-py3-none-any.whl
+
+ARG wheel=ecoindex_cli-2.26.0a0-py3-none-any.whl
+ENV DOCKER_CONTAINER=True
 
 WORKDIR /code