Tweak WACZ enricher docs + add comment on WACZ_ENABLE_DOCKER

src/auto_archiver/core/module.py

@@ -5,6 +5,7 @@ by handling user configuration, validating the steps properties, and implementin
 """
 
 from __future__ import annotations
+import subprocess
 
 from dataclasses import dataclass
 from typing import List, TYPE_CHECKING, Type
@@ -17,7 +18,7 @@ import os
 from os.path import join
 from loguru import logger
 import auto_archiver
-from auto_archiver.core.consts import DEFAULT_MANIFEST, MANIFEST_FILE
+from auto_archiver.core.consts import DEFAULT_MANIFEST, MANIFEST_FILE, SetupError
 
 if TYPE_CHECKING:
     from .base_module import BaseModule
@@ -216,9 +217,9 @@ class LazyBaseModule:
                 if not check(dep):
                     logger.error(
                         f"Module '{self.name}' requires external dependency '{dep}' which is not available/setup. \
-                                 Have you installed the required dependencies for the '{self.name}' module? See the README for more information."
+                                 Have you installed the required dependencies for the '{self.name}' module? See the documentation for more information."
                     )
-                    exit(1)
+                    raise SetupError()
 
         def check_python_dep(dep):
             # first check if it's a module:
@@ -238,9 +239,18 @@ class LazyBaseModule:
             return find_spec(dep)
 
         def check_bin_dep(dep):
-            if dep == "docker" and os.environ.get("RUNNING_IN_DOCKER"):
-                return True
-            return shutil.which(dep)
+            dep_exists = shutil.which(dep)
+
+            if dep == "docker":
+                if os.environ.get("RUNNING_IN_DOCKER"):
+                    # this is only for the WACZ enricher, which requires docker
+                    # if we're already running in docker then we don't need docker
+                    return True
+
+                # check if docker daemon is running
+                return dep_exists and subprocess.run(["docker", "ps", "-q"]).returncode == 0
+
+            return dep_exists
 
         check_deps(self.dependencies.get("python", []), check_python_dep)
         check_deps(self.dependencies.get("bin", []), check_bin_dep)

src/auto_archiver/modules/wacz_extractor_enricher/__manifest__.py

@@ -11,7 +11,7 @@
     "configs": {
         "profile": {
             "default": None,
-            "help": "browsertrix-profile (for profile generation see https://github.com/webrecorder/browsertrix-crawler#creating-and-using-browser-profiles).",
+            "help": "browsertrix-profile (for profile generation see https://crawler.docs.browsertrix.com/user-guide/browser-profiles/).",
         },
         "docker_commands": {"default": None, "help": "if a custom docker invocation is needed"},
         "timeout": {"default": 120, "help": "timeout for WACZ generation in seconds", "type": "int"},
@@ -40,14 +40,27 @@
     Creates .WACZ archives of web pages using the `browsertrix-crawler` tool, with options for media extraction and screenshot saving.
     [Browsertrix-crawler](https://crawler.docs.browsertrix.com/user-guide/) is a headless browser-based crawler that archives web pages in WACZ format.
 
-    ### Features
+    ## Setup
+
+    **Docker**
+    If you are using the Docker file to run Auto Archiver (recommended), then everything is set up and you can use WACZ out of the box!
+    Otherwise, if you are using a local install of Auto Archiver (e.g. pip or dev install), then you will need to install Docker and run 
+    the docker daemon to be able to run the `browsertrix-crawler` tool.
+
+    **Browsertrix Profiles**
+    A browsertrix profile is a custom browser profile (login information, browser extensions, etc.) that can be used to archive private or dynamic content.
+    You can run the WACZ Enricher without a profile, but for more resilient archiving, it is recommended to create a profile. See the [Browsertrix documentation](https://crawler.docs.browsertrix.com/user-guide/browser-profiles/)
+    for more information.
+
+    ** Docker in Docker **
+    If you are running Auto Archiver within a Docker container, you will need to enable Docker in Docker to run the `browsertrix-crawler` tool.
+    This can be done by setting the `WACZ_ENABLE_DOCKER` environment variable to `1`.
+
+    ## Features
     - Archives web pages into .WACZ format using Docker or direct invocation of `browsertrix-crawler`.
     - Supports custom profiles for archiving private or dynamic content.
     - Extracts media (images, videos, audio) and screenshots from the archive, optionally adding them to the enrichment pipeline.
     - Generates metadata from the archived page's content and structure (e.g., titles, text).
 
-    ### Notes
-    - Requires Docker for running `browsertrix-crawler` .
-    - Configurable via parameters for timeout, media extraction, screenshots, and proxy settings.
     """,
 }

tests/enrichers/test_wacz_enricher.py

@@ -4,6 +4,7 @@ from zipfile import ZipFile
 import pytest
 
 from auto_archiver.core import Metadata, Media
+from auto_archiver.core.consts import SetupError
 
 
 @pytest.fixture
@@ -22,6 +23,15 @@ def wacz_enricher(setup_module, mock_binary_dependencies):
     return wacz
 
 
+def test_raises_error_without_docker_installed(setup_module, mocker, caplog):
+    # pretend that docker isn't installed
+    mocker.patch("shutil.which").return_value = None
+    with pytest.raises(SetupError):
+        setup_module("wacz_extractor_enricher", {})
+
+    assert "requires external dependency 'docker' which is not available/setup" in caplog.text
+
+
 def test_setup_without_docker(wacz_enricher, mocker):
     mocker.patch.dict(os.environ, {"RUNNING_IN_DOCKER": "1"}, clear=True)
     wacz_enricher.setup()