feat: unfiltered product export for the website's catalogue import

src/BusProNet/XmlParser/ProductExportParser.php

@@ -0,0 +1,161 @@
+<?php
+
+declare(strict_types=1);
+
+namespace App\BusProNet\XmlParser;
+
+use App\BusProNet\Constants;
+use Symfony\Component\DomCrawler\Crawler;
+
+/**
+ * Reads the parts of a `Ziel_*.xml` travel date that the product export carries, verbatim.
+ *
+ * Deliberately separate from TravelParser, which builds the booking path's models and
+ * applies the booking path's rules: it drops the PDGS pseudo room, excludes the
+ * hotel-nested services and levels boarding-point prices. The product export filters and
+ * rewrites nothing -- its consumer applies its own rules -- so it reads the nodes directly
+ * and returns plain arrays in source order.
+ *
+ * @see \App\Service\ProductExportBuilder
+ */
+class ProductExportParser extends AbstractParser
+{
+    /**
+     * Every selection group on the date, not only the `Leistungen *` ones.
+     *
+     * @return list<array{busProId: int, label: ?string, selections: list<array{busProId: int, label: ?string}>}>
+     */
+    public function getSelectionGroups(Crawler $dateNode): array
+    {
+        return $this->children($dateNode, 'selektiongruppe')->each(fn (Crawler $group): array => [
+            'busProId' => (int) $group->attr('idbuspro'),
+            'label' => $group->attr('bezeichnung'),
+            'selections' => $this->children($group, 'selektion')->each(fn (Crawler $selection): array => [
+                'busProId' => (int) $selection->attr('idbuspro'),
+                'label' => $selection->attr('bezeichnung'),
+            ]),
+        ]);
+    }
+
+    /**
+     * Services of one `lei_*` list, keyed by nothing -- the consumer keys them by `busProId`.
+     *
+     * The per-list `id` attribute is not read: it restarts at 1 in every list, so it
+     * identifies nothing once a date-level and a hotel-level list are merged.
+     *
+     * @return list<array{busProId: int, subType: ?string, label: ?string, price: ?float, status: ?string, direction: ?string}>
+     */
+    public function getServices(Crawler $serviceNodes): array
+    {
+        return $serviceNodes->each(fn (Crawler $service): array => [
+            'busProId' => (int) $service->attr('idbuspro'),
+            'subType' => $service->attr('unterart'),
+            'label' => $this->getStringOrNullValue($this->children($service, 'text')),
+            'price' => $this->getFloatOrNullValue($this->children($service, 'preis')),
+            'status' => $this->getStringOrNullValue($this->children($service, 'status')),
+            'direction' => $this->getStringOrNullValue($this->children($service, 'richtung')),
+        ]);
+    }
+
+    /**
+     * The hotel's rooms, without the PDGS pseudo room.
+     *
+     * `available` is a count of rooms, not persons. Occupancy is read case-insensitively:
+     * the travel export writes `MinPax`/`MaxPax`.
+     *
+     * @return list<array<string, mixed>>
+     */
+    public function getRooms(Crawler $hotelNode): array
+    {
+        $rooms = [];
+
+        $this->children($hotelNode, 'zimmer/preis')->each(function (Crawler $room) use (&$rooms) {
+            $code = $this->trimOrNull($this->getAttrOrNullValue($room, 'zimmercode'));
+
+            if (Constants::PSEUDO_PRICE_ROOM_CODE === $code) {
+                return;
+            }
+
+            $rooms[] = [
+                'busProId' => $this->getIntAttrOrNullValue($room, 'idbuspro_zimmer'),
+                'code' => $code,
+                'label' => $this->getAttrOrNullValue($room, 'zimmertext'),
+                'category' => $this->getAttrOrNullValue($room, 'kat'),
+                'minPax' => $this->getIntAttrOrNullValue($room, 'minpax'),
+                'maxPax' => $this->getIntAttrOrNullValue($room, 'maxpax'),
+                'nights' => $this->getIntAttrOrNullValue($room, 'naechte'),
+                'price' => $this->stringToFloat($this->getAttrOrNullValue($room, 'preis')),
+                'available' => $this->getIntAttrOrNullValue($room, 'verfuegbar'),
+                'status' => $this->getAttrOrNullValue($room, 'status'),
+                'boardCode' => $this->getAttrOrNullValue($room, 'vpcode'),
+                'boardBusProId' => $this->getIntAttrOrNullValue($room, 'idbuspro_vp'),
+            ];
+        });
+
+        return $rooms;
+    }
+
+    /**
+     * The struck-through comparison price, carried by the PDGS pseudo room.
+     */
+    public function getPseudoPrice(Crawler $hotelNode): ?float
+    {
+        foreach ($this->children($hotelNode, 'zimmer/preis') as $room) {
+            $room = new Crawler($room);
+
+            if (Constants::PSEUDO_PRICE_ROOM_CODE === $this->trimOrNull($this->getAttrOrNullValue($room, 'zimmercode'))) {
+                return $this->stringToFloat($this->getAttrOrNullValue($room, 'preis'));
+            }
+        }
+
+        return null;
+    }
+
+    /**
+     * The guide of the first outbound bus that has one.
+     *
+     * Not TravelParser::getGuide(), which takes the first guide of any bus, return leg
+     * included.
+     *
+     * @return array{name: ?string, phone: ?string}|null
+     */
+    public function getGuide(Crawler $dateNode): ?array
+    {
+        $guides = $this->children($dateNode, 'lei_befoerderung/leistung[@unterart="BUS"][not(richtung="RUECK")]/zustiegsplanung/reiseleiter');
+
+        if (0 === $guides->count()) {
+            return null;
+        }
+
+        $guide = $guides->first();
+
+        return [
+            'name' => $this->getStringOrNullValue($this->children($guide, 'name')),
+            'phone' => $this->getStringOrNullValue($this->children($guide, 'telefon')),
+        ];
+    }
+
+    /**
+     * A path below the node's own children, never deeper.
+     *
+     * DomCrawler evaluates filterXPath() against a fake parent of the node, so `./x` finds
+     * nothing and `.//x` searches every descendant -- which would pull a hotel's
+     * `lei_sonstiges` into the date's list. A leading `*` step anchors the path at the
+     * node itself, whatever its name.
+     */
+    public function children(Crawler $node, string $path): Crawler
+    {
+        return $node->filterXPath('*/'.$path);
+    }
+
+    private function trimOrNull(?string $value): ?string
+    {
+        if (null === $value) {
+            return null;
+        }
+
+        $trimmed = trim($value);
+
+        return '' === $trimmed ? null : $trimmed;
+    }
+}