Improve some PDF-Importer to standardized tax lost adjustment transaction

[Nirus2000]

The tax lost adjustment transaction has been standardized for various PDF importers.

[Nirus2000]

I have no issues...

Hmm... AbstractPDFExtractor ?! The language?!
private final NumberFormat numberFormat = NumberFormat.getInstance(Locale.GERMANY);

[buchen]

The problem is the following.

• The tests run with the "en" locale.
• You write the shares to the context using String.format and hence the "en" locale (see line 177 in Direkt1822BankPDFExtractor), say "11.00000000"
• You then read the shares always with the German locale (that is hard coded in the AbstractPDFExtractor) which uses the comma as decimal separator

And, whoops, you have a mismatch because parsing does not see the dot as decimal separator.

It does not happen for you locally, because I assume you run with the German locale.

If you want to reproduce the problem locally, add this to your launch configuration

-Duser.language=en -Duser.country=

I think the important thing is to use the same formatter for reading and writing (and hence the same locale setup),
for example,

context.put("shares", getNumberFormat().format(item.getShares() / Values.Share.divider()));
[...]
asShares(context.get("shares"))

or

context.put("shares", Long.toString(item.getShares()));
[...]
t.setShares(Long.parseLong(context.get("shares")));

[Nirus2000]

Hello @buchen
Thanks for the hint.
I have changed it and moved it to the master.

Regards
Alex 👍🏻