Fix online link detection in entry editor (#8514)

* Fix online link detection in entry editor * add changelog * Fix typo * update linked file * fix test * remove check for colon * fix checkstyle * fix checkstyle javadoc Co-authored-by: Oliver Kopp <[email protected]>
JabRef · Mar 7, 2022 · 0c439de · 0c439de
1 parent 3c247ff
commit 0c439de
Show file tree

Hide file tree

Showing 4 changed files with 26 additions and 4 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -21,6 +21,7 @@ Note that this project **does not** adhere to [Semantic Versioning](http://semve
 
 ### Fixed
 
+- We fixed an issue where online links in the file field were not detected correctly and could produce an exception [#8150](https://github.com/JabRef/jabref/issues/8510)
 - We fixed an issue where an exception could occur when saving the preferences [#7614](https://github.com/JabRef/jabref/issues/7614)
 - We fixed an issue where "Copy DOI url" in the right-click menu of the Entry List would just copy the DOI and not the DOI url. [#8389](https://github.com/JabRef/jabref/issues/8389)
 - We fixed an issue where opening the console from the drop-down menu would cause an exception. [#8466](https://github.com/JabRef/jabref/issues/8466)

diff --git a/src/main/java/org/jabref/logic/importer/SearchBasedParserFetcher.java b/src/main/java/org/jabref/logic/importer/SearchBasedParserFetcher.java
@@ -20,7 +20,7 @@
  *     <li>Post-process fetched entries</li>
  * </ol>
  * <p>
- *     This interface is used for web resources which do NOT provide BibTeX data (@link {@link BibEntry)}.
+ *     This interface is used for web resources which do NOT provide BibTeX data {@link BibEntry}.
  *     JabRef's infrastructure to convert arbitrary input data to BibTeX is {@link Parser}.
  * </p>
  * <p>
@@ -30,7 +30,7 @@
  * <p>
  *     Note that this interface "should" be an abstract class.
  *     However, Java does not support multi inheritance with classes (but with interfaces).
- *     We need multi inheritance, because a fetcher might implement multiple query types (such as id fetching {@link IdBasedFetcher)}, complete entry {@link EntryBasedFetcher}, and search-based fetcher (this class).
+ *     We need multi inheritance, because a fetcher might implement multiple query types (such as id fetching {@link IdBasedFetcher}), complete entry {@link EntryBasedFetcher}, and search-based fetcher (this class).
  * </p>
  */
 public interface SearchBasedParserFetcher extends SearchBasedFetcher {

diff --git a/src/main/java/org/jabref/model/entry/LinkedFile.java b/src/main/java/org/jabref/model/entry/LinkedFile.java
@@ -11,6 +11,7 @@
 import java.util.List;
 import java.util.Objects;
 import java.util.Optional;
+import java.util.regex.Pattern;
 
 import javafx.beans.Observable;
 import javafx.beans.property.SimpleStringProperty;
@@ -26,6 +27,9 @@
  */
 public class LinkedFile implements Serializable {
 
+    private static final String REGEX_URL = "^((?:https?\\:\\/\\/|www\\.)(?:[-a-z0-9]+\\.)*[-a-z0-9]+.*)";
+    private static final Pattern URL_PATTERN = Pattern.compile(REGEX_URL);
+
     private static final LinkedFile NULL_OBJECT = new LinkedFile("", Path.of(""), "");
 
     // We have to mark these properties as transient because they can't be serialized directly
@@ -145,7 +149,7 @@ private void readObject(ObjectInputStream in) throws IOException {
      */
     public static boolean isOnlineLink(String toCheck) {
         String normalizedFilePath = toCheck.trim().toLowerCase();
-        return normalizedFilePath.startsWith("http://") || normalizedFilePath.startsWith("https://") || normalizedFilePath.contains("www.");
+        return URL_PATTERN.matcher(normalizedFilePath).matches();
     }
 
     @Override

diff --git a/src/test/java/org/jabref/logic/importer/util/FileFieldParserTest.java b/src/test/java/org/jabref/logic/importer/util/FileFieldParserTest.java
@@ -138,17 +138,34 @@ private static Stream<Arguments> stringsToParseTestData() throws Exception {
                         "desc:file.pdf:PDF:asdf"
                 ),
 
+                // www inside filename
+                Arguments.of(
+                        Collections.singletonList(new LinkedFile("", Path.of("/home/www.google.de.pdf"), "")),
+                        ":/home/www.google.de.pdf"
+                ),
+
                 // url
                 Arguments.of(
                          Collections.singletonList(new LinkedFile(new URL("https://books.google.de/"), "")),
                          "https://books.google.de/"
                 ),
 
+                // url with www
+                Arguments.of(
+                             Collections.singletonList(new LinkedFile(new URL("https://www.google.de/"), "")),
+                             "https://www.google.de/"
+                ),
+
                 // url as file
                 Arguments.of(
                              Collections.singletonList(new LinkedFile("", new URL("http://ceur-ws.org/Vol-438"), "URL")),
                              ":http\\://ceur-ws.org/Vol-438:URL"
-                )
+                ),
+                // url as file with desc
+                Arguments.of(
+                             Collections.singletonList(new LinkedFile("desc", new URL("http://ceur-ws.org/Vol-438"), "URL")),
+                             "desc:http\\://ceur-ws.org/Vol-438:URL"
+               )
         );
     }