Age | Commit message (Collapse) | Author | Files | Lines |
|
* Changed JSONException to JsonParseException.
|
|
commit closes everything via safeClose, so ensure that tweetGenerated
re-opens it if necessary.
Fixes:
ƒ commit
€ Exception in thread "Thread-7" java.lang.NullPointerException
at io.DataWriter.writeObject(DataWriter.java:137)
at io.DataWriter.tweetGenerated(DataWriter.java:85)
at provider.CompositeResultListener.tweetGenerated(CompositeResultListener.java:51)
at io.StreamImpl$Poller.processObject(StreamImpl.java:255)
at io.StreamImpl$Poller.run(StreamImpl.java:245)
at java.lang.Thread.run(Thread.java:744)
|
|
|
|
These are always available from the tweets themselves...
|
|
Prevents occasional "Unexpected end of ZLIB input stream" errors.
|
|
Now print the faulty line and line number. Using BufferedReader instead
of Scanner also gave me a hint of the error, so let's stick to it.
|
|
|
|
Right now Store instances are created in the constructor of DataWriter.
This causes issues, namely that Store instances are unable to use
DataWriter in its constructor.
The real-world implication is that convertUncompressed in
CompressableDataWriter is unusable for the CompressedStore.
|
|
|
|
|
|
|
|
The input and output file are closely related, do not burden others with
tracking which output and input need to be matched.
|
|
Prepare for compressed file formats, this requires that the data writer
can handle compressed files. In theory an implementation can even
write stuff over the network instead of a file.
|
|
Files may not be accessible if they are read-only, or if the directory
is read-only, or ... whatever. So, let the caller handle IOException
in DataWriter.
Return a set if it can be read.
Do some misc cleanups (remove unused imports, fix docs, remove obsolete
TODOs).
|
|
* Implemented target file, which uses DataWriter as
RequestListener which writes the tweets and profiles.
* Added profiles-filename, tweets-filename to the config.
|
|
|
|
|