refactor: update code to get site parsers from database.

BREAKING CHANGE: site parsers from application.properties are now ignored. It's supposed that they
have been migrated by running application with the code from previous commit. Otherwise,
administrator can migrate them manually.

Addressed to #975

Author: php-coder

docs/diagrams/parse-page.png

@@ -1,12 +1,15 @@
 title Series import flow: stage 3 (parse page)
 
 participant DownloadingSucceededEventListener
+participant SiteParserService
 participant SeriesImportService
 participant SiteParser
 participant ExtractorService
 participant SeriesSalesImportService
 participant EventPublisher
 
+DownloadingSucceededEventListener->+SiteParserService: page URL
+SiteParserService->-DownloadingSucceededEventListener: site parser
 DownloadingSucceededEventListener->+SeriesImportService: requestId
 SeriesImportService->-DownloadingSucceededEventListener: content of downloaded page
 DownloadingSucceededEventListener->+SiteParser: content of downloaded page

docs/diagrams/parse-page.txt

@@ -59,6 +59,7 @@ public class DownloadingSucceededEventListener
 
 	@PostConstruct
 	public void init() {
+		// TODO: get all parser names from database
 		log.info("Registered site parsers: {}", siteParsers);
 
 		// TODO: remove migration logic after finishing migration
@@ -71,26 +72,17 @@ public void onApplicationEvent(DownloadingSucceeded event) {
 
 		log.info("Request #{}: downloading succeeded", requestId);
 
-		String content = seriesImportService.getDownloadedContent(requestId);
-		if (content == null) {
+		SiteParser parser = siteParserService.findForUrl(event.getUrl());
+		if (parser == null) {
 			// TODO: how to handle error? maybe publish UnexpectedErrorEvent?
-			log.error("Request #{}: could not load a content from database", requestId);
+			log.error("Request #{}: could not find appropriate parser", requestId);
 			return;
 		}
 
-		// TODO: replace with siteParserService.findForUrl(url) and update diagrams
-		String url = event.getUrl();
-		SiteParser parser = null;
-		for (SiteParser candidate : siteParsers) {
-			if (candidate.canParse(url)) {
-				parser = candidate;
-				break;
-			}
-		}
-		
-		if (parser == null) {
+		String content = seriesImportService.getDownloadedContent(requestId);
+		if (content == null) {
 			// TODO: how to handle error? maybe publish UnexpectedErrorEvent?
-			log.error("Request #{}: could not find appropriate parser", requestId);
+			log.error("Request #{}: could not load a content from database", requestId);
 			return;
 		}
 

src/main/java/ru/mystamps/web/feature/series/importing/event/DownloadingSucceededEventListener.java

@@ -30,6 +30,7 @@
 
 import lombok.AccessLevel;
 import lombok.Getter;
+import lombok.NoArgsConstructor;
 import lombok.Setter;
 
 // Getters/setters are being used in unit tests
@@ -41,13 +42,16 @@
 	// false positive because setField() modifies the fields
 	"PMD.ImmutableField"
 })
+// TODO: consider to remove no-arg constructor after removing code for creating parsers from file
+@NoArgsConstructor
 public class JsoupSiteParser implements SiteParser {
 	private static final Logger LOG = LoggerFactory.getLogger(JsoupSiteParser.class);
 
 	// When you're adding a new field don't forget to also update:
 	// - JsoupSiteParser.setField()
 	// - JsoupSiteParser.isFullyInitialized() (optionally)
 	// - JsoupSiteParserTest.describe()
+	// - JsoupSiteParser constructor
 	// - SiteParserConfiguration
 	private String name;
 	private String matchedUrl;
@@ -61,6 +65,21 @@ public class JsoupSiteParser implements SiteParser {
 	private String priceLocator;
 	private String currencyValue;
 
+	// @todo #975 SiteParserServiceImpl: add unit tests for constructor
+	public JsoupSiteParser(SiteParserConfiguration cfg) {
+		name                    = cfg.getName();
+		matchedUrl              = cfg.getMatchedUrl();
+		categoryLocator         = cfg.getCategoryLocator();
+		countryLocator          = cfg.getCountryLocator();
+		shortDescriptionLocator = cfg.getShortDescriptionLocator();
+		imageUrlLocator         = cfg.getImageUrlLocator();
+		imageUrlAttribute       = cfg.getImageUrlAttribute();
+		issueDateLocator        = cfg.getIssueDateLocator();
+		sellerLocator           = cfg.getSellerLocator();
+		priceLocator            = cfg.getPriceLocator();
+		currencyValue           = cfg.getCurrencyValue();
+	}
+	
 	// TODO: remove because it's only needed for migrating configuration from file to database
 	public SiteParserConfiguration toConfiguration() {
 		SiteParserConfiguration cfg = new SiteParserConfiguration(name, matchedUrl);
@@ -153,14 +172,6 @@ public boolean isFullyInitialized() {
 			);
 	}
 
-	@Override
-	public boolean canParse(String url) {
-		Validate.validState(url != null, "Site URL must be non-null");
-		Validate.validState(matchedUrl != null, "Matched URL must be set");
-		
-		return url.startsWith(matchedUrl);
-	}
-	
 	/**
 	 * Parse HTML document to get info about series.
 	 *

src/main/java/ru/mystamps/web/feature/series/importing/extractor/JsoupSiteParser.java

@@ -23,8 +23,5 @@ public interface SiteParser {
 
 	boolean isFullyInitialized();
 
-	// TODO: remove because it's only needed for migrating configuration from file to database
-	boolean canParse(String url);
-	
 	SeriesInfo parse(String htmlPage);
 }

src/main/java/ru/mystamps/web/feature/series/importing/extractor/SiteParser.java

@@ -19,5 +19,5 @@
 
 public interface SiteParserService {
 	void add(SiteParserConfiguration cfg);
-	SiteParserConfiguration findForUrl(String url);
+	SiteParser findForUrl(String url);
 }

src/main/java/ru/mystamps/web/feature/series/importing/extractor/SiteParserService.java

@@ -62,7 +62,7 @@ public void add(SiteParserConfiguration cfg) {
 	// @todo #975 SiteParserServiceImpl.findForUrl(): add unit tests
 	@Override
 	@Transactional(readOnly = true)
-	public SiteParserConfiguration findForUrl(String url) {
+	public SiteParser findForUrl(String url) {
 		Validate.isTrue(url != null, "Url must be non null");
 
 		Integer parserId = siteParserDao.findParserIdForUrl(url);
@@ -71,7 +71,13 @@ public SiteParserConfiguration findForUrl(String url) {
 			return null;
 		}
 
-		return siteParserDao.findConfigurationForParser(parserId);
+		SiteParserConfiguration cfg = siteParserDao.findConfigurationForParser(parserId);
+		if (cfg == null) {
+			log.warn("Could not find configuration for parser #{}", parserId);
+			return null;
+		}
+		
+		return new JsoupSiteParser(cfg);
 	}
 
 }

src/main/java/ru/mystamps/web/feature/series/importing/extractor/SiteParserServiceImpl.java

@@ -46,11 +46,6 @@ public boolean isFullyInitialized() {
 		return parser.isFullyInitialized();
 	}
 
-	@Override
-	public boolean canParse(String url) {
-		return parser.canParse(url);
-	}
-	
 	@Override
 	public SeriesInfo parse(String htmlPage) {
 		// Why we don't use Spring's StopWatch?

src/main/java/ru/mystamps/web/feature/series/importing/extractor/TimedSiteParser.java

@@ -286,42 +286,6 @@ public void isFullyInitializedWhenAllMandatoryFieldsAreSet() {
 		assertThat(msg, parser.isFullyInitialized(), is(true));
 	}
 
-	//
-	// Tests for canParse()
-	//
-	
-	@Test
-	public void canParseShouldRequireNonNullUrl() {
-		thrown.expect(IllegalStateException.class);
-		thrown.expectMessage("Site URL must be non-null");
-		
-		parser.canParse(null);
-	}
-	
-	@Test
-	public void canParseShouldRequireNonNullMatchedUrl() {
-		parser.setMatchedUrl(null);
-		
-		thrown.expect(IllegalStateException.class);
-		thrown.expectMessage("Matched URL must be set");
-		
-		parser.canParse(Random.url());
-	}
-	
-	@Test
-	public void canParseWithUnsupportedUrl() {
-		parser.setMatchedUrl("http://example.org");
-		
-		assertThat(parser.canParse("http://example.com/test/fail"), is(false));
-	}
-	
-	@Test
-	public void canParseWithSupportedUrl() {
-		parser.setMatchedUrl("http://example.org");
-		
-		assertThat(parser.canParse("http://example.org/test/success"), is(true));
-	}
-	
 	//
 	// Tests for parse()
 	//