From a7c2d469cae543e200d8f38b74bbe0257b7ea74b Mon Sep 17 00:00:00 2001 From: Barbara Miller <3253863+galgeek@users.noreply.github.com> Date: Wed, 24 Apr 2024 10:42:59 -0700 Subject: [PATCH 1/2] bump version --- setup.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/setup.py b/setup.py index ca26426..e095162 100644 --- a/setup.py +++ b/setup.py @@ -34,7 +34,7 @@ def find_package_data(package): setuptools.setup( name="brozzler", - version="1.5.46", + version="1.5.47", description="Distributed web crawling with browsers", url="https://github.com/internetarchive/brozzler", author="Noah Levitt", From 5432844994c5a15e47f89a5405f0a7e1746158f8 Mon Sep 17 00:00:00 2001 From: Barbara Miller Date: Thu, 25 Apr 2024 14:18:00 -0700 Subject: [PATCH 2/2] nsw.gov pubs behavior --- brozzler/behaviors.yaml | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/brozzler/behaviors.yaml b/brozzler/behaviors.yaml index 39aad2b..fcaf5dd 100644 --- a/brozzler/behaviors.yaml +++ b/brozzler/behaviors.yaml @@ -49,6 +49,14 @@ - selector: a.coreSpriteRightPaginationArrow repeatSameElement: true rmSelector: '.RnEpo' +- + url_regex: '^https?://www.opengov\.nsw\.gov\.au/publications.*$' + behavior_js_template: umbraBehavior.js.j2 + default_parameters: + interval: 1000 + actions: + - selector: span.DV-trigger.DV-next + repeatSameElement: true - url_regex: '^https?://americaspresidents\.si\.edu/gallery.*$' behavior_js_template: umbraBehavior.js.j2