Preserve encoding while copy and pasting in maintable

[stefan-kolb]

Fixes #5043.

Encodes and decodes data with UTF-8.

[matthiasgeiger]

Hi Stefan,
Nice to see some commits from you ;-)
I'm not sure whether this is not breaking then the parsing from Strings in other encodings... Have you Hi checked it with some ISO encoded special characters?

[stefan-kolb]

ISO special characters should be mappable in UTF-8, right?! I guess currently the default encoding for copying to the clipboard is UTF-8, so pasting with UTF-8 made sense to me. If we want to preserve the real encoding, we probably need to pass the encoding with the DataFlavor somehow? Like https://stackoverflow.com/questions/29651301/text-copied-from-jtextarea-have-broken-encoding-after-paste-in-foxpro-applicatio

[Siedlerchr]

We always use UTF-8 internally, so this makes indeed sense. Only use case: What if the users has selected a different encoding for the database in the prefs?

[stefan-kolb]

If we want to preserve the real encoding, we probably need to pass the encoding with the DataFlavor somehow? Like https://stackoverflow.com/questions/29651301/text-copied-from-jtextarea-have-broken-encoding-after-paste-in-foxpro-applicatio

I guess we need to do something like this then.

[Siedlerchr]

This is old swing stuff, don't do this.
It should be sufficient to use the encoding in this method and apply the same encoding when pasting.

jabref/src/main/java/org/jabref/gui/ClipBoardManager.java

Lines 83 to 89 in eb42850

	public void setContent(List<BibEntry> entries) throws IOException {
	final ClipboardContent content = new ClipboardContent();
	BibEntryWriter writer = new BibEntryWriter(new LatexFieldFormatter(Globals.prefs.getLatexFieldFormatterPreferences()), false);
	String serializedEntries = writer.serializeAll(entries, BibDatabaseMode.BIBTEX);
	content.put(DragAndDropDataFormats.ENTRIES, serializedEntries);
	content.putString(serializedEntries);
	clipboard.setContent(content);

[stefan-kolb]

I cannot spot any place during serialization where we use an encoding different to the default encoding which seems to be UTF-8 then.

[tobiasdiez]

I don't we need to pay attention to the database encoding since this option only controls the encoding when the database is written to the file (and does not affect the entries in itself).

It may be worth a try to use parser.parseEntries(new StringReader(entries)) which would circumvent the need to specify an encoding. But whether this works depends on if the JavaFX clipboard is able to correctly encode/decode the string (I tried to dig into the javafx code but it seems that they let the os handle the encoding). Otherwise your solution looks good to me.

[Siedlerchr]

Merging this now!