Rewrite ATSB feed generator, closes #119

itsb.xml

@@ -298,12 +298,7 @@
       <frequency>100 reports/year</frequency>
       <url>https://www.atsb.gov.au/</url>
       <feed format="rss" lang="English" id="atsb-rss">
-        <curl>
-          <url>https://www.atsb.gov.au/publications/safety-investigation-reports/?s=1&amp;sort=OccurrenceReleaseDate&amp;sortAscending=descending&amp;investigationStatus=Completed,Discontinued&amp;printAll=true</url>
-        </curl>
-        <pup>table.selectable_grid tr:not(.header)</pup>
-        <jq path="atsb.jq" />
-        <json2xml />
+        <xquery path="atsb.xqy" user-agent="a/1" />
         <output>atsb.xml</output>
       </feed>
     </source>

xquery/atsb.xqy

@@ -0,0 +1,60 @@
+<rss
+  version="2.0"
+  xmlns:admin="http://webns.net/mvcb/"
+  xmlns:atom="http://www.w3.org/2005/Atom"
+  xmlns:dc="http://purl.org/dc/elements/1.1/"
+  xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+  xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
+  xmlns:webfeeds="http://webfeeds.org/rss/1.0"
+  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+  xsi:noNamespaceSchemaLocation="xsd/rss.xsd"
+  xsi:schemaLocation="
+    http://purl.org/rss/1.0/modules/syndication/ xsd/syndication.xsd
+    http://webfeeds.org/rss/1.0 xsd/webfeeds.xsd
+  "
+>
+  <channel>
+    <title>ATSB</title>
+    <description>Australian Transport Safety Bureau accident reports</description>
+    <link>https://www.atsb.gov.au/marine-investigation-reports?field_mode_of_transport_target_id=All</link>
+    <language>en-au</language>
+    <pubDate>{format-dateTime(adjust-dateTime-to-timezone(current-dateTime(), xs:dayTimeDuration("PT0S")), "[FNn,*-3], [D01] [MNn,*-3] [Y0001] [H01]:[m01]:[s01] GMT")}</pubDate>
+    <webMaster>lucidiot@envs.net (lucidiot)</webMaster>
+    <docs>http://www.rssboard.org/rss-specification</docs>
+    <ttl>1440</ttl>
+
+    <admin:errorReportsTo rdf:resource="mailto:lucidiot@envs.net" />
+    <admin:generatorAgent rdf:resource="https://tildegit.org/lucidiot/itsb/" />
+
+    <atom:link href="https://tilde.town/~lucidiot/itsb/feeds/atsb.xml" rel="self" type="application/rss+xml" />
+
+    <dc:format>application/rss+xml</dc:format>
+
+    <sy:updatePeriod>daily</sy:updatePeriod>
+    <sy:updateFrequency>1</sy:updateFrequency>
+    <sy:updateBase>1990-01-01T15:00+00:00</sy:updateBase>
+
+    <webfeeds:partial>true</webfeeds:partial>
+    <webfeeds:deprecated>false</webfeeds:deprecated>
+
+    {
+      (
+        163, (: Completed :)
+        168 (: Discontinued :)
+      ) ! x:parse-html(x:request({
+        "url": concat("https://www.atsb.gov.au/marine-investigation-reports?field_investigation_status_target_id=", ., "&amp;field_mode_of_transport_target_id=All"),
+        "headers": "Accept-Encoding: lol
+Accept-Language: zz,
+Cache-Control: lol"
+      })/raw)//table[contains(@class, "views-table")]/tbody/tr
+      ! <item>
+        <title>{./td[2]/a/text()}</title>
+        <link>{fn:resolve-uri(./td[2]/a/@href)}</link>
+        <guid>{fn:resolve-uri(./td[2]/a/@href)}</guid>
+        <description>{./td[1]/a/text()}</description>
+        <pubDate>{format-dateTime(xs:dateTime(.//time/@datetime[1]), "[FNn,*-3], [D01] [MNn,*-3] [Y0001] [H01]:[m01]:[s01] GMT")}</pubDate>
+        <category domain="https://www.atsb.gov.au/">{./td[4]/text()}</category>
+      </item>
+    }
+  </channel>
+</rss>