s434730/RandomSec
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Use UrlValidator to detect urls in tabular exporter

Browse Source
This commit is contained in:
Antonin Delpeuch 2018-09-03 19:48:11 +01:00
parent 18c489855b
commit 6aa93bbc0c
1 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
main/src/com/google/refine/exporters/CustomizableTabularExporterUtilities.java
View File

@ -47,6 +47,7 @@ import java.util.Properties;
import java.util.TimeZone; import java.util.TimeZone;
import org.apache.commons.lang3.StringUtils; import org.apache.commons.lang3.StringUtils;
import org.apache.commons.validator.routines.UrlValidator;
import org.json.JSONArray; import org.json.JSONArray;
import org.json.JSONException; import org.json.JSONException;
import org.json.JSONObject; import org.json.JSONObject;
@ -228,6 +229,8 @@ abstract public class CustomizableTabularExporterUtilities {
boolean date_omitTime = false; boolean date_omitTime = false;
DateFormat dateFormatter; DateFormat dateFormatter;
String[] urlSchemes = {"http","https", "ftp"};
UrlValidator urlValidator = new UrlValidator(urlSchemes);
Map<String, String> identifierSpaceToUrl = null; Map<String, String> identifierSpaceToUrl = null;
@ -352,9 +355,8 @@ abstract public class CustomizableTabularExporterUtilities {
text = (String) value; text = (String) value;
if(text.contains(":")) { if(text.contains(":")) {
try { if(urlValidator.isValid(text)) {
link = new URL(text).toString(); link = text;
} catch (MalformedURLException e) {
} }
} }
} else if (value instanceof OffsetDateTime) { } else if (value instanceof OffsetDateTime) {