From 394a1ec35c01a15b6947646cc7c8d0f9dbd844e2 Mon Sep 17 00:00:00 2001
From: mynttt <marc@herschel.io>
Date: Tue, 25 Aug 2020 18:06:30 +0200
Subject: [PATCH] 1.4.3 <=> IMDB ID cleaning to minimize amount of non-matches

---
 CHANGELOG.md                                              | 3 +++
 VERSION                                                   | 2 +-
 build.gradle                                              | 2 +-
 src/main/java/updatetool/imdb/ImdbPipeline.java           | 2 +-
 .../java/updatetool/imdb/ImdbRatingDatasetFactory.java    | 2 +-
 src/main/java/updatetool/imdb/ImdbTransformer.java        | 8 ++++++++
 src/main/resources/VERSION                                | 2 +-
 7 files changed, 16 insertions(+), 5 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 749defe..e438f0d 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,3 +1,6 @@
+## 1.4.3
+- IMDB id cleaning to minimize the amount of non matches after having an id resolved
+
 ## 1.4.2
 - tmdb resolve bug fixed
 
diff --git a/VERSION b/VERSION
index 13175fd..3c80e4f 100644
--- a/VERSION
+++ b/VERSION
@@ -1 +1 @@
-1.4.1
\ No newline at end of file
+1.4.3
\ No newline at end of file
diff --git a/build.gradle b/build.gradle
index 64a5e75..2037a60 100644
--- a/build.gradle
+++ b/build.gradle
@@ -5,7 +5,7 @@ plugins {
     id 'com.github.spotbugs' version '2.0.1'
 }
 
-version = '1.4.2'
+version = '1.4.3'
 sourceCompatibility = '11'
 
 new File(projectDir, "VERSION").text = version;
diff --git a/src/main/java/updatetool/imdb/ImdbPipeline.java b/src/main/java/updatetool/imdb/ImdbPipeline.java
index a81c615..d7a2f1c 100644
--- a/src/main/java/updatetool/imdb/ImdbPipeline.java
+++ b/src/main/java/updatetool/imdb/ImdbPipeline.java
@@ -147,7 +147,7 @@ public void accumulateMetadata(ImdbJob job) throws Exception {
     @Override
     public void transformMetadata(ImdbJob job) throws Exception {
         var map = new HashMap<ImdbMetadataResult, ExportedRating>();
-        job.items.forEach(i -> map.put(i, dataset.getRatingFor(i.imdbId)));
+        job.items.forEach(i -> map.put(i, dataset.getRatingFor(ImdbTransformer.clean(i.imdbId))));
         var noUpdate = map.entrySet().stream().filter(Predicate.not(ImdbTransformer::needsUpdate)).collect(Collectors.toSet());
         if(!noUpdate.isEmpty()) {
             Logger.info(noUpdate.size() + " item(s) need no update.");
diff --git a/src/main/java/updatetool/imdb/ImdbRatingDatasetFactory.java b/src/main/java/updatetool/imdb/ImdbRatingDatasetFactory.java
index c95eca4..ee2d635 100644
--- a/src/main/java/updatetool/imdb/ImdbRatingDatasetFactory.java
+++ b/src/main/java/updatetool/imdb/ImdbRatingDatasetFactory.java
@@ -37,7 +37,7 @@ public static class ImdbRatingDataset {
         private HashMap<String, String> data = new HashMap<>();
 
         public ExportedRating getRatingFor(String imdbId) {
-            final String rating = data.get(imdbId);
+            String rating = data.get(imdbId);
             return () -> rating;
         }
     }
diff --git a/src/main/java/updatetool/imdb/ImdbTransformer.java b/src/main/java/updatetool/imdb/ImdbTransformer.java
index 9bfd4f2..2bfac6e 100644
--- a/src/main/java/updatetool/imdb/ImdbTransformer.java
+++ b/src/main/java/updatetool/imdb/ImdbTransformer.java
@@ -77,4 +77,12 @@ public static void updateMetadata(Map.Entry<ImdbMetadataResult, ExportedRating>
         
         meta.extraData = extra.toURI();
     }
+
+    // Cleaning required because of interesting data from TVDB
+    public static String clean(String imdbId) {
+        if(imdbId == null)
+            return null;
+        String numbers = ImdbUtility.extractId(ImdbUtility.NUMERIC, imdbId);
+        return "tt"+numbers;
+    }
 }
diff --git a/src/main/resources/VERSION b/src/main/resources/VERSION
index 13175fd..3c80e4f 100644
--- a/src/main/resources/VERSION
+++ b/src/main/resources/VERSION
@@ -1 +1 @@
-1.4.1
\ No newline at end of file
+1.4.3
\ No newline at end of file