Mirror/MarginaliaSearch
1
0
Fork 0
mirror of https://github.com/MarginaliaSearch/MarginaliaSearch.git synced 2025-02-24 05:18:58 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Use a default tag for unset or invalid generators.

Browse Source
This commit is contained in:
Viktor Lofgren 2023-06-21 17:30:14 +02:00
parent a9a2960e86
commit f140e7d7c7
1 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
code/processes/converting-process/src/main/java/nu/marginalia/converting/processor/logic/DocumentGeneratorExtractor.java
View File

@ -9,6 +9,8 @@ import java.util.List;
/** Extract keywords for the document meta generator tag */ /** Extract keywords for the document meta generator tag */
public class DocumentGeneratorExtractor { public class DocumentGeneratorExtractor {
private final String defaultValue = "unset";
public List<String> generatorCleaned(Document doc) { public List<String> generatorCleaned(Document doc) {
String generator = doc String generator = doc
@ -19,11 +21,11 @@ public class DocumentGeneratorExtractor {
generator = trim(generator); generator = trim(generator);
if (generator.isBlank()) if (generator.isBlank())
return Collections.emptyList(); return List.of(defaultValue);
String[] parts = StringUtils.split(generator, " ,:!"); String[] parts = StringUtils.split(generator, " ,:!");
if (parts.length == 0) if (parts.length == 0)
return Collections.emptyList(); return List.of(defaultValue);
int slashIdx = parts[0].indexOf('/'); int slashIdx = parts[0].indexOf('/');
if (slashIdx >= 0) { if (slashIdx >= 0) {
@ -32,7 +34,7 @@ public class DocumentGeneratorExtractor {
} }
if (parts.length > 3) { if (parts.length > 3) {
return Collections.emptyList(); // if it's still very long after trim(), it's probably a custom hand written message return List.of(defaultValue); // if it's still very long after trim(), it's probably a custom hand written message
} }
switch (parts[0]) { switch (parts[0]) {