test: require success in cli workflows

archivebox/cli/archivebox_shell.py

@@ -12,9 +12,10 @@ from archivebox.misc.util import docstring
 def shell(args: Iterable[str] = ()) -> None:
     """Enter an interactive ArchiveBox Django shell"""
 
-    from django.core.management import call_command
+    from django.core.management import call_command, get_commands
 
-    call_command("shell_plus", *args)
+    shell_command = "shell_plus" if "shell_plus" in get_commands() else "shell"
+    call_command(shell_command, *args)
 
 
 @click.command(add_help_option=False, context_settings=dict(ignore_unknown_options=True))

archivebox/tests/test_cli_add.py

@@ -614,10 +614,12 @@ def test_cli_add_real_urls_with_options_writes_inspectable_outputs(initialized_a
     by_url_plugin = {(url, plugin): status for url, plugin, status, _files, _size, _output in archive_results}
     assert by_url_plugin[("https://example.com", "wget")] == "succeeded"
     assert by_url_plugin[("https://pirate.github.io/stress-tests/challenge.html", "wget")] == "succeeded"
-    assert (chrome_url, "headers") in by_url_plugin
-    assert (chrome_url, "title") in by_url_plugin
-    failed_results = [(url, plugin, output) for url, plugin, status, _files, _size, output in archive_results if status == "failed"]
-    assert len(failed_results) <= 2, failed_results
+    assert by_url_plugin[(chrome_url, "headers")] == "succeeded"
+    assert by_url_plugin[(chrome_url, "title")] == "succeeded"
+    unexpected_results = [
+        (url, plugin, status, output) for url, plugin, status, _files, _size, output in archive_results if status != "succeeded"
+    ]
+    assert not unexpected_results
 
     snapshot_root = initialized_archive / "archive/users/system/snapshots"
     html_outputs = [path for path in snapshot_root.rglob("wget/**/*.html") if path.is_file()]
@@ -625,11 +627,10 @@ def test_cli_add_real_urls_with_options_writes_inspectable_outputs(initialized_a
     title_outputs = [path for path in snapshot_root.rglob("title/title.txt") if path.is_file() and path.stat().st_size > 0]
     index_outputs = [path for path in snapshot_root.rglob("index.jsonl") if path.is_file()]
     assert html_outputs
-    if by_url_plugin[(chrome_url, "headers")] == "succeeded":
-        assert header_outputs
-    if by_url_plugin[(chrome_url, "title")] == "succeeded":
-        assert title_outputs
-        assert any("Example Domain" in path.read_text(errors="ignore") for path in title_outputs)
+    assert header_outputs
+    assert any("example.com" in path.read_text(errors="ignore").lower() for path in header_outputs)
+    assert title_outputs
+    assert any("Example Domain" in path.read_text(errors="ignore") for path in title_outputs)
     assert len(index_outputs) >= len(wget_urls) + 1
 
     combined_html = "\n".join(path.read_text(errors="ignore") for path in html_outputs)

archivebox/tests/test_cli_install.py

@@ -6,6 +6,7 @@ Verify install detects and records binary dependencies in DB.
 
 import os
 from pathlib import Path
+
 from archivebox.tests.conftest import run_archivebox_cmd
 
 import pytest
@@ -25,20 +26,21 @@ def test_install_runs_successfully(initialized_archive):
         timeout=60,
     )
 
-    # Dry run should complete quickly
-    assert result.returncode in [0, 1]  # May return 1 if binaries missing
+    assert result.returncode == 0, result.stderr or result.stdout
+    assert "Dry run - would detect ArchiveBox dependencies" in result.stdout
 
 
 def test_install_creates_binary_records_in_db(initialized_archive):
-    """Test that install creates Binary records in database."""
+    """Test that install --dry-run does not create Binary records in database."""
 
-    run_archivebox_cmd(
+    result = run_archivebox_cmd(
         ["install", "--dry-run"],
         timeout=60,
     )
+    assert result.returncode == 0, result.stderr or result.stdout
 
     with use_archivebox_db(initialized_archive):
-        Binary.objects.count()
+        assert Binary.objects.count() == 0
 
 
 def test_install_dry_run_does_not_install(initialized_archive):
@@ -49,8 +51,8 @@ def test_install_dry_run_does_not_install(initialized_archive):
         timeout=60,
     )
 
-    # Should complete without actually installing
-    assert "dry" in result.stdout.lower() or result.returncode in [0, 1]
+    assert result.returncode == 0, result.stderr or result.stdout
+    assert result.stdout.strip() == "Dry run - would detect ArchiveBox dependencies and run the abx-dl install flow"
 
 
 def test_install_detects_system_binaries(initialized_archive):
@@ -61,8 +63,8 @@ def test_install_detects_system_binaries(initialized_archive):
         timeout=60,
     )
 
-    # Should detect at least some common binaries (python, curl, etc)
-    assert result.returncode in [0, 1]
+    assert result.returncode == 0, result.stderr or result.stdout
+    assert "ArchiveBox dependencies" in result.stdout
 
 
 def test_install_shows_binary_status(initialized_archive):
@@ -73,9 +75,8 @@ def test_install_shows_binary_status(initialized_archive):
         timeout=60,
     )
 
-    output = result.stdout + result.stderr
-    # Should show some binary information
-    assert len(output) > 50
+    assert result.returncode == 0, result.stderr or result.stdout
+    assert result.stdout.strip() == "Dry run - would detect ArchiveBox dependencies and run the abx-dl install flow"
 
 
 def test_install_dry_run_prints_dry_run_message(initialized_archive):

archivebox/tests/test_cli_manage.py

@@ -53,5 +53,5 @@ def test_manage_check_works(initialized_archive):
         timeout=30,
     )
 
-    # Check should complete
-    assert result.returncode in [0, 1]
+    assert result.returncode == 0, result.stderr or result.stdout
+    assert "System check identified no issues" in result.stdout

archivebox/tests/test_cli_remove.py

@@ -219,20 +219,29 @@ def test_remove_reports_remaining_link_count_correctly(initialized_archive):
 
 def test_remove_after_flag(initialized_archive):
     """Test remove --after flag removes snapshots after date."""
-    env = cli_env(disable_extractors=True)
+    env = cli_env()
 
     run_archivebox_cmd(
         ["add", "--index-only", "--depth=0", "https://example.com"],
         env=env,
+        check=True,
     )
     run_queued_crawls(initialized_archive, env)
 
-    # Try remove with --after flag (should work or show usage)
+    rows = _snapshot_rows(initialized_archive, env)
+    assert len(rows) == 1
+    snapshot_dir = find_snapshot_dir(initialized_archive, rows[0]["id"])
+    assert snapshot_dir is not None, f"Snapshot output directory not found for {rows[0]['id']}"
+
     result = run_archivebox_cmd(
-        ["remove", "--after=2020-01-01", "--yes"],
+        ["remove", "--after=1577836800", "--yes"],
         env=env,
         timeout=30,
+        check=True,
     )
 
-    # Should complete
-    assert result.returncode in [0, 1, 2]
+    output = result.stdout + result.stderr
+    assert "Removed 1 out of 1 links" in output
+    assert "Index now contains 0 links." in output
+    assert len(_snapshot_rows(initialized_archive, env)) == 0
+    assert not snapshot_dir.exists()

archivebox/tests/test_cli_shell.py

@@ -10,14 +10,14 @@ from archivebox.tests.conftest import run_archivebox_cmd
 def test_shell_command_exists(initialized_archive):
     """Test that shell command is recognized."""
 
-    # Test that the command exists (will fail without input but should recognize command)
     result = run_archivebox_cmd(
         ["shell", "--help"],
         timeout=10,
     )
 
-    # Should show shell help or recognize command
-    assert result.returncode in [0, 1, 2]
+    assert result.returncode == 0, result.stderr or result.stdout
+    assert "usage:" in result.stdout
+    assert "shell" in result.stdout
 
 
 def test_shell_c_executes_python(initialized_archive):