chore: Optional pillow requirement (#135)

* Optional pillow requirement

* Regen poetry lock files & add documentation

* Pass logger in EcoindexScraper class

Author: QuentiumYT

components/ecoindex/scraper/helper.py

@@ -20,11 +20,13 @@ def run_page_analysis(
         wait_after_scroll=wait_after_scroll,
         wait_before_scroll=wait_before_scroll,
         page_load_timeout=20,
+        logger=logger,
     )
     try:
         return (run(scraper.get_page_analysis()), True)
     except Exception as e:
-        logger.error(f"{url} -- {e.msg if hasattr(e, 'msg') else e}")
+        if logger:
+            logger.error(f"{url} -- {e.msg if hasattr(e, 'msg') else e}")
 
         return (
             Result(

components/ecoindex/scraper/scrap.py

@@ -32,6 +32,7 @@ def __init__(
         basic_auth: str | None = None,
         cookies: list[SetCookieParam] = [],
         custom_headers: dict[str, str] = {},
+        logger=None,
     ):
         self.url = url
         self.window_size = window_size
@@ -50,6 +51,7 @@ def __init__(
         self.basic_auth = basic_auth
         self.cookies = cookies
         self.custom_headers = custom_headers
+        self.logger = logger
 
     @staticmethod
     def get_user_agent() -> UserAgent:
@@ -129,7 +131,13 @@ async def scrap_page(self) -> PageMetrics:
     async def generate_screenshot(self) -> None:
         if self.screenshot and self.screenshot.folder and self.screenshot.id:
             await self.page.screenshot(path=self.screenshot.get_png())
-            await convert_screenshot_to_webp(self.screenshot)
+            try:
+                await convert_screenshot_to_webp(self.screenshot)
+            except ImportError:
+                if self.logger:
+                    self.logger.warning(
+                        "WebP conversion skipped: Pillow library is not installed."
+                    )
             await set_screenshot_rights(
                 screenshot=self.screenshot,
                 uid=self.screenshot_uid,
@@ -191,8 +199,7 @@ async def check_page_response(self, response) -> None:
                 {
                     "mimetype": content_type,
                     "message": (
-                        "This resource is not "
-                        "a standard page with mimeType 'text/html'"
+                        "This resource is not a standard page with mimeType 'text/html'"
                     ),
                 }
             )

components/ecoindex/utils/screenshots.py

@@ -1,10 +1,18 @@
 import os
 
 from ecoindex.models import ScreenShot
-from PIL import Image
+
+try:
+    from PIL import Image
+
+    _pillow_available = True
+except ImportError:
+    _pillow_available = False
 
 
 async def convert_screenshot_to_webp(screenshot: ScreenShot) -> None:
+    if not _pillow_available:
+        raise ImportError("Pillow is required for WebP conversion. Install it with: pip install ecoindex-scraper[webp]")
     image = Image.open(rf"{screenshot.get_png()}")
     width, height = image.size
     ratio = 800 / height if width > height else 600 / width
@@ -17,7 +25,9 @@ async def convert_screenshot_to_webp(screenshot: ScreenShot) -> None:
 
 
 async def set_screenshot_rights(
-    screenshot: ScreenShot, uid: int | None = None, gid: int | None = None
+    screenshot: ScreenShot,
+    uid: int | None = None,
+    gid: int | None = None,
 ) -> None:
     if uid and gid:
         os.chown(path=screenshot.get_webp(), uid=uid, gid=gid)

poetry.lock

@@ -27,7 +27,7 @@ aiofile = "^3.8.8"
 alembic = "^1.12.1"
 celery = "^5.3.4"
 fastapi = "^0.109.1"
-pillow = "^12.0.0"
+pillow = { version = "^12.2.0", optional = true }
 playwright = "^1.39.0"
 playwright-stealth = "^1.0.6"
 pydantic = { version = ">=2.1.1,<=2.4.2", extras = ["email"] }
@@ -46,7 +46,7 @@ ua-generator = "^2.0.5"
 uvicorn = "^0.23.2"
 
 [tool.poetry.group.worker.dependencies]
-pillow = "^12.0.0"
+pillow = { version = "^12.2.0", optional = true }
 playwright = "^1.39.0"
 playwright-stealth = "^1.0.6"
 

projects/ecoindex_api/poetry.lock

@@ -14,7 +14,13 @@ This module provides a simple interface to get the [Ecoindex](http://www.ecoinde
 ## Install
 
 ```shell
-pip install ecoindex-scraper
+pip install ecoindex_scraper
+```
+
+If you need to convert the screenshot to webp with the `generate_screenshot` method, you need to install the Pillow dependency.
+
+```shell
+pip install ecoindex_scraper[webp] # OR directly pillow
 ```
 
 ## Use
@@ -171,4 +177,4 @@ pprint(requests_by_category.model_dump())
 #  'javascript': {'total_count': 1, 'total_size': 9823.0},
 #  'other': {'total_count': 1, 'total_size': 892.0},
 #  'video': {'total_count': 0, 'total_size': 0.0}}
-```
+```

projects/ecoindex_api/pyproject.toml

@@ -24,10 +24,13 @@ pydantic = "^2.4.2"
 python = "^3.10"
 typing-extensions = "^4.8.0"
 pyyaml = "^6.0.1"
-pillow = ">=10.1,<13.0"
+pillow = { version = "^12.2.0", optional = true }
 setuptools = ">=69.5.1,<79.0.0"
 ua-generator = "^2.0.5"
 
+[tool.poetry.extras]
+webp = ["pillow"]
+
 [build-system]
 requires = ["poetry-core>=1.0.0"]
 build-backend = "poetry.core.masonry.api"

projects/ecoindex_scraper/README.md

@@ -36,10 +36,12 @@ setuptools = "^78.1.1"
 cryptography = "^44.0.2"
 
 [tool.poetry.group.scraper.dependencies]
-pillow = "^12.0.0"
 playwright = "^1.39.0"
 playwright-stealth = "^1.0.6"
 
+[tool.poetry.group.scraper-webp.dependencies]
+pillow = { version = "^12.2.0", optional = true }
+
 [tool.poetry.group.cli.dependencies]
 click-spinner = "^0.1.10"
 jinja2 = "^3.1.3"