adding new extract command to extract selector or main content as markdown, updateing version as 0.5.0

tests/test_api.py

@@ -182,6 +182,21 @@ class TestSearch:
         assert mock_send.call_args[0][1]["background"] is True
 
 
+class TestExtract:
+    def test_extract_markdown_default(self, b, mock_send):
+        mock_send.return_value = "# Title"
+
+        result = b.extract_markdown()
+
+        assert result == "# Title"
+        mock_send.assert_called_once_with("extract.markdown", {"selector": None}, profile=None)
+
+    def test_extract_markdown_selector(self, b, mock_send):
+        b.extract_markdown("article")
+
+        mock_send.assert_called_once_with("extract.markdown", {"selector": "article"}, profile=None)
+
+
 # ── Tabs ──────────────────────────────────────────────────────────────────────
 
 class TestTabs:

tests/test_cli.py

@@ -54,3 +54,19 @@ def test_clients_exits_cleanly_when_registry_is_missing():
 
     assert result.exit_code == 1
     assert "No browser clients found" in result.output
+
+def test_extract_markdown_command():
+    with patch("browser_cli.commands.extract.send_command", return_value="# Title\n") as send_command:
+        result = CliRunner().invoke(main, ["extract", "markdown"])
+
+    assert result.exit_code == 0
+    assert result.output == "# Title\n"
+    send_command.assert_called_once_with("extract.markdown", {"selector": None})
+
+def test_extract_markdown_command_with_selector():
+    with patch("browser_cli.commands.extract.send_command", return_value="## Post\n") as send_command:
+        result = CliRunner().invoke(main, ["extract", "markdown", "--selector", "article"])
+
+    assert result.exit_code == 0
+    assert result.output == "## Post\n"
+    send_command.assert_called_once_with("extract.markdown", {"selector": "article"})

tests/test_extract.py

@@ -7,18 +7,24 @@ def test_extract_links(browser, http_tab):
     browser("tabs.active", {"tabId": http_tab["id"]})
     links = browser("extract.links")
     assert isinstance(links, list)
+    hrefs = []
     for lnk in links:
         assert "href" in lnk
         assert "text" in lnk
+        hrefs.append(lnk["href"])
+    assert len(hrefs) == len(set(hrefs))
 
 
 def test_extract_images(browser, http_tab):
     browser("tabs.active", {"tabId": http_tab["id"]})
     images = browser("extract.images")
     assert isinstance(images, list)
+    sources = []
     for img in images:
         assert "src" in img
         assert img["src"] != ""
+        sources.append(img["src"])
+    assert len(sources) == len(set(sources))
 
 
 def test_extract_text(browser, http_tab):
@@ -35,6 +41,19 @@ def test_extract_html(browser, http_tab):
     assert "<" in html
 
 
+def test_extract_markdown(browser, http_tab):
+    browser("tabs.active", {"tabId": http_tab["id"]})
+    markdown = browser("extract.markdown")
+    assert isinstance(markdown, str)
+    assert len(markdown.strip()) > 0
+
+
+def test_extract_markdown_missing_selector_errors(browser, http_tab):
+    browser("tabs.active", {"tabId": http_tab["id"]})
+    with pytest.raises(RuntimeError, match="No element"):
+        browser("extract.markdown", {"selector": ".browser-cli-definitely-missing"})
+
+
 def test_dom_exists(browser, http_tab):
     browser("tabs.active", {"tabId": http_tab["id"]})
     result = browser("dom.exists", {"selector": "body"})