summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorPeter Wu <peter@lekensteyn.nl>2016-11-04 12:34:48 +0100
committerPeter Wu <peter@lekensteyn.nl>2016-11-04 12:34:48 +0100
commit9513ce12e73d7e208766bd27f71ca05eccc522b4 (patch)
treed3891e7565fb295c80b2bfd968a982a85423724d
parent752d05b155254d4651031a39cf989454a30166d0 (diff)
downloadDatafiller-9513ce12e73d7e208766bd27f71ca05eccc522b4.tar.gz
Allow empty URLs
Retweets with the URL being cutoff by an ellipsis have an invalid URL in the entities field of the retweet: "urls": [{ "url": "", "expanded_url": null, "indices": [137, 137] }], Just skip such URLs then.
-rw-r--r--src/data/Tweet.java2
-rw-r--r--src/main/DataFiller.java6
2 files changed, 7 insertions, 1 deletions
diff --git a/src/data/Tweet.java b/src/data/Tweet.java
index d9b1de0..6ee89f4 100644
--- a/src/data/Tweet.java
+++ b/src/data/Tweet.java
@@ -57,7 +57,7 @@ public class Tweet {
@ValidatingJsonDeserializer.Validator
public Hashtag[] hashtags;
- @ValidatingJsonDeserializer.Validator
+ //@ValidatingJsonDeserializer.Validator
public Url[] urls;
@ValidatingJsonDeserializer.Validator
public Mention[] user_mentions;
diff --git a/src/main/DataFiller.java b/src/main/DataFiller.java
index a400975..2a06026 100644
--- a/src/main/DataFiller.java
+++ b/src/main/DataFiller.java
@@ -113,6 +113,9 @@ public class DataFiller {
m_insertHash.executeUpdate();
}
for (Tweet.Url url : tweet.entities.urls) {
+ if (url.expanded_url == null) {
+ continue;
+ }
m_insertTweetUrl.setLong("tweetid", tweet.id);
m_insertTweetUrl.setString("url", url.expanded_url);
m_insertTweetUrl.executeUpdate();
@@ -126,6 +129,9 @@ public class DataFiller {
User user = tweet.user;
if (user.entities != null) {
for (Tweet.Url url : tweet.entities.urls) {
+ if (url.expanded_url == null) {
+ continue;
+ }
m_insertUserUrl.setLong("userid", user.id);
m_insertUserUrl.setString("url", url.expanded_url);
m_insertUserUrl.executeUpdate();