Add CI/CD and phpstan-related fixes

.github/workflows/test.yaml

@@ -0,0 +1,40 @@
+name: test and deploy
+on:
+  workflow_dispatch: ~
+  push: ~
+  release:
+    types: [published]
+jobs:
+  phpunit:
+    strategy:
+      matrix:
+        php: [8.1, 8.2, 8.3, 8.4]
+        imageTag: ['latest']
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+    - uses: shivammathur/setup-php@v2
+      with:
+        php-version: ${{ matrix.php }}
+        extensions: json,mbstring,pcre,yaml
+        coverage: xdebug
+      env:
+        fail-fast: true
+    - name: composer update
+      run: |
+        composer update
+    - name: phpstan
+      run: |
+        vendor/bin/phpstan analyze -l 6 src tests
+    - name: phpunit
+      run: |
+        XDEBUG_MODE=coverage vendor/bin/phpunit
+    - name: coveralls
+      run: |
+        export COVERALLS_RUN_LOCALLY=1
+        export COVERALLS_REPO_TOKEN=${{ secrets.coverallsToken }}
+        rm -fR composer.json composer.lock vendor
+        composer require php-coveralls/php-coveralls
+        php vendor/bin/php-coveralls -v
+
+

.gitignore

@@ -4,3 +4,4 @@
 /config.yaml
 /*.ttl*
 /.phpunit.cache
+/build

README.md

@@ -1,5 +1,8 @@
 # arche-ref-sources
 
+[![Build status](https://github.com/acdh-oeaw/arche-ref-sources/actions/workflows/test.yaml/badge.svg)](https://github.com/acdh-oeaw/arche-ref-sources/actions/workflows/test.yaml)
+[![Coverage Status](https://coveralls.io/repos/github/acdh-oeaw/arche-ref-sources/badge.svg?branch=master)](https://coveralls.io/github/acdh-oeaw/arche-ref-sources?branch=master)
+
 A script for enhancing an ARCHE Suite repository data with information gathered from external reference sources (Geonames, GND, wikidata, etc.).
 
 It works by resolving existing identifiers that are external reference source URIs, fetching information from there and updating a corresponding ARCHE Suite repository resource.

phpunit.xml

@@ -1,6 +1,10 @@
 <?xml version="1.0"?>
 <phpunit xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" colors="true" bootstrap="vendor/autoload.php" xsi:noNamespaceSchemaLocation="https://schema.phpunit.de/10.3/phpunit.xsd" cacheDirectory=".phpunit.cache">
-  <coverage/>
+  <coverage>
+    <report>
+      <clover outputFile="build/logs/clover.xml"/>
+    </report>
+  </coverage>
   <testsuite name="tests">
     <directory>tests</directory>
   </testsuite>

src/acdhOeaw/arche/refSources/Crawler.php

@@ -28,9 +28,11 @@
 
 use Generator;
 use Psr\Log\LoggerInterface;
-use rdfInterface\DatasetNode;
+use rdfInterface\DatasetInterface;
+use rdfInterface\DatasetNodeInterface;
 use acdhOeaw\UriNormalizer;
 use acdhOeaw\UriNormalizerCache;
+use acdhOeaw\UriNormalizerException;
 use acdhOeaw\arche\lib\Schema;
 
 /**
@@ -40,6 +42,7 @@
  */
 class Crawler {
 
+    /** @phpstan-ignore property.onlyWritten */
     private Schema $schema;
     private PropertyMappings $mappings;
     private UriNormalizer $normalizer;
@@ -57,8 +60,7 @@ public function __construct(object $refSrcsCfg, Schema $schema,
     /**
      * 
      * @param NamedEntityIteratorInterface $source
-     * @param bool $verbose
-     * @return Generator<array<DatasetNode>>
+     * @return Generator<array<DatasetInterface>>
      */
     public function crawl(NamedEntityIteratorInterface $source,
                           string $dateFilter = '',
@@ -90,6 +92,10 @@ public function crawl(NamedEntityIteratorInterface $source,
         }
     }
 
+    /**
+     * 
+     * @return array<DatasetInterface>
+     */
     private function processEntity(NamedEntityInterface $entity): array {
         // collect data from all external databases reachable from this entity
         $entityExtMeta = [];

src/acdhOeaw/arche/refSources/NamedEntityIteratorInterface.php

@@ -37,6 +37,10 @@ interface NamedEntityIteratorInterface {
     public function setFilter(?string $class = null, ?string $idMatch = null,
                               ?string $minModDate = null, ?int $limit = null): void;
 
+    /**
+     * 
+     * @return Generator<NamedEntityInterface>
+     */
     public function getNamedEntities(): Generator;
 
     public function getCount(): int;

src/acdhOeaw/arche/refSources/NamedEntityRepo.php

@@ -29,7 +29,7 @@
 use termTemplates\PredicateTemplate as PT;
 use acdhOeaw\UriNormalizer;
 use acdhOeaw\UriNormalizerException;
-use acdhOeaw\arche\lib\Repo;
+use acdhOeaw\arche\lib\RepoInterface;
 use acdhOeaw\arche\lib\RepoResource;
 use rdfInterface\DatasetNodeInterface;
 
@@ -43,7 +43,7 @@ class NamedEntityRepo implements NamedEntityInterface {
     use NamedEntityTrait;
 
     private RepoResource $res;
-    private Repo $repo;
+    private RepoInterface $repo;
 
     public function __construct(RepoResource $res) {
         $this->res  = $res;

src/acdhOeaw/arche/refSources/PropertyMapping.php

@@ -90,6 +90,7 @@ public function __construct(object $cfg) {
         $this->maxPerLang     = $cfg->maxPerLang ?? PHP_INT_MAX;
         $this->preferredLangs = $cfg->preferredLangs ?? [];
         if (!empty($cfg->value)) {
+            /** @phpstan-ignore property.notFound */
             $this->value = $cfg->type === 'resource' ? DF::namedNode($cfg->value) : DF::literal($cfg->value);
         } else {
             $this->path = array_map(fn($x) => DF::namedNode($x), $cfg->path);
@@ -213,13 +214,16 @@ private function processLang(DatasetInterface $meta): void {
         // assure all values are literals
         $meta->forEach(fn(QuadInterface $q) => $q->getObject() instanceof LiteralInterface ? $q : $q->withObject(DF::literal($q->getObject()->getValue())));
         if (count($this->preferredLangs) > 0) {
+            /** @phpstan-ignore method.notFound */
             $meta->deleteExcept(fn(QuadInterface $q) => in_array($q->getObject()->getLang(), $this->preferredLangs));
         }
         // sort according to language preferences if needed
         $triples = iterator_to_array($meta->getIterator());
         if ($this->maxPerLang < PHP_INT_MAX) {
             uasort($triples, function ($a, $b) {
+                /** @phpstan-ignore method.notFound */
                 $aPref = array_search($a->getObject()->getLang(), $this->preferredLangs) ?: PHP_INT_MAX;
+                /** @phpstan-ignore method.notFound */
                 $bPref = array_search($b->getObject()->getLang(), $this->preferredLangs) ?: PHP_INT_MAX;
                 $comp  = $aPref <=> $bPref;
                 if ($comp === 0) {
@@ -235,15 +239,18 @@ private function processLang(DatasetInterface $meta): void {
             $literal       = $triple->getObject();
             /* @var $literal LiteralInterface */
             $lang          = match ($this->langProcess) {
+            /** @phpstan-ignore method.notFound */
                 PropertyMapping::LANG_PASS => $literal->getLang(),
+                /** @phpstan-ignore method.notFound */
                 PropertyMapping::LANG_ASSURE => $literal->getLang() ?? $this->langValue,
                 PropertyMapping::LANG_OVERWRITE => $this->langValue,
                 PropertyMapping::LANG_REMOVE => null,
-                default => throw new RuntimeException("Unsupported lang tag processing mode $process"),
+                default => throw new RuntimeException("Unsupported lang tag processing mode $this->langProcess"),
             };
             $lang          ??= '';
             $counts[$lang] = ($counts[$lang] ?? 0) + 1;
             if ($counts[$lang] <= $this->maxPerLang) {
+                /** @phpstan-ignore method.notFound */
                 $toAdd = $literal->withLang($lang);
                 if (!empty($this->datatype)) {
                     $toAdd = $toAdd->withDatatype($this->datatype);
@@ -275,11 +282,7 @@ private function normalize(DatasetInterface $meta, UriNormalizer $normalizer): v
 
     /**
      * 
-     * @param DatasetInterface $meta
-     * @param NamedNodeInterface $sbj
-     * @param UriNormalizer $normalizer
      * @param array<NamedNodeInterface> $path
-     * @return DatasetInterface
      */
     private function resolveRecursively(DatasetInterface $meta,
                                         TermInterface $sbj,