Ability to run shot-craper multi --har and not take shots at all

Refs #166 (comment)

Author: simonw

shot_scraper/cli.py

@@ -549,6 +549,13 @@ def multi(
 
     scale_factor = normalize_scale_factor(retina, scale_factor)
     shots = yaml.safe_load(config)
+
+    # Special case: if we are recording a har_file output can be blank to skip a shot
+    if har_file:
+        for shot in shots:
+            if not shot.get("output"):
+                shot["skip_shot"] = True
+
     server_processes = []
     if shots is None:
         shots = []
@@ -1370,12 +1377,16 @@ def on_response(response):
                 ", ".join(list(selectors) + list(selectors_all)), url, output
             )
     else:
-        # Whole page
-        if return_bytes:
-            return page.screenshot(**screenshot_args)
+        if shot.get("skip_shot"):
+            message = "Skipping screenshot of '{}'".format(url)
         else:
-            page.screenshot(**screenshot_args)
-            message = "Screenshot of '{}' written to '{}'".format(url, output)
+            # Whole page
+            if return_bytes:
+                return page.screenshot(**screenshot_args)
+            else:
+                page.screenshot(**screenshot_args)
+                message = "Screenshot of '{}' written to '{}'".format(url, output)
+
     if not silent:
         click.echo(message, err=True)
 

tests/test_shot_scraper.py

@@ -262,31 +262,36 @@ def test_har(http_server, args, expect_zip):
 
 
 @pytest.mark.parametrize(
-    "args,expect_zip",
+    "args,expect_zip,record_shots",
     (
-        (["--har"], False),
-        (["--har-zip"], True),
-        (["--har-file", "output.har"], False),
-        (["--har-file", "output.har.zip"], True),
+        (["--har"], False, True),
+        (["--har-zip"], True, True),
+        (["--har-file", "output.har"], False, True),
+        (["--har-file", "output.har.zip"], True, True),
+        # And one where we don't record the shots:
+        (["--har"], False, False),
     ),
 )
-def test_multi_har(http_server, args, expect_zip):
+def test_multi_har(http_server, args, expect_zip, record_shots):
     runner = CliRunner()
     (http_server.base_dir / "two.html").write_text("<h1>Two</h1>")
     with runner.isolated_filesystem():
         pathlib.Path("shots.yml").write_text(
             f"- url: {http_server.base_url}/\n"
-            f"  output: index.png\n"
-            f"- url: {http_server.base_url}/two.html\n"
-            f"  output: two.png\n"
+            + (f"  output: index.png\n" if record_shots else "")
+            + f"- url: {http_server.base_url}/two.html\n"
+            + (f"  output: two.png\n" if record_shots else "")
         )
         # Should be no files
         here = pathlib.Path(".")
         files = [str(p) for p in here.glob("*.*")]
         assert files == ["shots.yml"]
         result = runner.invoke(cli, ["multi", "shots.yml"] + args)
         assert result.exit_code == 0
-        assert result.output.startswith("Screenshot of 'http://localhost")
+        if record_shots:
+            assert result.output.startswith("Screenshot of 'http://localhost")
+        else:
+            assert result.output.startswith("Skipping screenshot of 'http://localhost")
         assert "Wrote to HAR file:" in result.output
         assert (".har.zip" in result.output) == expect_zip
         # HAR file should have been created
@@ -298,5 +303,9 @@ def test_multi_har(http_server, args, expect_zip):
         # Should have created exactly one .har file
         assert len(har_files) == 1
         assert bool(zipfile.is_zipfile(har_files[0])) == expect_zip
-        # Should have taken shots
-        assert len(list(here.glob("*.png"))) == 2
+        shot_files = list(here.glob("*.png"))
+        num_shots = len(shot_files)
+        if record_shots:
+            assert num_shots == 2
+        else:
+            assert num_shots == 0