Allow always matching extractor filters

That is, allowing to list all nodes of a specific type without any
further constraints.

autotests/extractorfiltertest.cpp

@@ -4,6 +4,8 @@
     SPDX-License-Identifier: LGPL-2.0-or-later
 */
 
+#include "config-kitinerary.h"
+
 #include <KItinerary/ExtractorDocumentNode>
 #include <KItinerary/ExtractorDocumentNodeFactory>
 #include <KItinerary/ExtractorDocumentProcessor>
@@ -124,6 +126,29 @@ private Q_SLOTS:
         filter.setPattern(s("Milan"));
         QVERIFY(filter.matches(root));
     }
+
+    void testIataBcbpFilter()
+    {
+        QFile f(s(SOURCE_DIR "/extractordata/synthetic/iata-bcbp-demo.pdf"));
+        QVERIFY(f.open(QFile::ReadOnly));
+
+        ExtractorEngine engine;
+        engine.setData(f.readAll());
+        engine.extract();
+        const auto root = engine.rootDocumentNode();
+        QVERIFY(!root.isNull());
+
+#ifdef HAVE_ZXING
+        ExtractorFilter filter;
+        filter.setMimeType(s("internal/iata-bcbp"));
+        filter.setScope(ExtractorFilter::Descendants);
+        QVERIFY(filter.matches(root));
+
+        std::vector<ExtractorDocumentNode> matches;
+        filter.allMatches(root, matches);
+        QCOMPARE(matches.size(), 1);
+#endif
+    }
 };
 
 QTEST_GUILESS_MAIN(ExtractorFilterTest)

src/lib/engine/extractorfilter.cpp

@@ -156,7 +156,8 @@ static bool filterMachesNode(const ExtractorFilter &filter, ExtractorFilter::Sco
         return false;
     }
 
-    if (filter.mimeType() == node.mimeType() && node.processor()->matches(filter, node)) {
+    // filter without field/pattern always match, if the mimetype does
+    if (filter.mimeType() == node.mimeType() && ((filter.fieldName().isEmpty() && filter.pattern().isEmpty()) || node.processor()->matches(filter, node))) {
         if (matchMode == All) {
             matches.push_back(node);
         }
@@ -249,8 +250,14 @@ ExtractorFilter ExtractorFilter::fromJSValue(const QJSValue &js)
 {
     ExtractorFilter f;
     f.setMimeType(js.property(QLatin1String("mimeType")).toString());
-    f.setFieldName(js.property(QLatin1String("field")).toString());
-    f.setPattern(js.property(QLatin1String("match")).toString());
+    const auto fieldName = js.property(QLatin1String("field"));
+    if (fieldName.isString()) {
+        f.setFieldName(fieldName.toString());
+    }
+    const auto match = js.property(QLatin1String("match"));
+    if (match.isString()) {
+        f.setPattern(match.toString());
+    }
     f.setScope(readEnum<ExtractorFilter::Scope>(js.property(QLatin1String("scope")).toString(), ExtractorFilter::Current));
     return f;
 }

src/lib/scripts/americanairlines.js

@@ -6,7 +6,7 @@
 
 function main(pdf, node) {
     var res = [];
-    for (barcode of node.findChildNodes({ scope: "Descendants", mimeType: "text/plain", match: "M.*" })) {
+    for (barcode of node.findChildNodes({ scope: "Descendants", mimeType: "internal/iata-bcbp" })) {
         if (barcode.location == undefined)
             continue;
         res.push(parsePage(pdf.pages[barcode.location], barcode));