Fix Japanese translation

[horiguti]

These are assorted patchs to fix Japanese translation on the master branch in response to a call by Dave Cramer in pgsql-translators . Plus additional cleanups of the whole Japanese translation.

This consists of the following fixes.

• 4b350ec (2nd) Fix messages pointed by feat: Implement Statement#getObject(int, Class) without JSR-310 #813
• 682fc4a (3rd) Fill up empty translation items. Fixes wrong translations and unifies wording. This might be better be reduced to just filling empty items and fixing apparently wrong ones.
• d5ed902 (4th) Fix a typo in PgConnection.java and related entry in ja.po.

And to put these changes on top of, 762cd4e (1st) is ja.po update made by mvn -Ptranslate compile.

regards.

[vlsi]

@horiguti , thanks for the quick update.

I've rebased your branch on top of current master, could you please recheck if it still looks good?

PS. @davecramer , we might need to review headers for po files as well: https://github.com/horiguti/pgjdbc/blob/d757321c09d773bfd21d572e22e126a36a19b489/pgjdbc/src/main/java/org/postgresql/translation/ja.po#L1-L3

[vlsi]

It looks like we have off-by-one errors like

pgjdbc/pgjdbc/src/main/java/org/postgresql/jdbc/PgCallableStatement.java

Line 686 in 178eecc

throw new PSQLException(GT.tr("Unsupported type conversion to {1}.", type),

Arguments are 0-indexed, so it should be {0}.

I think we can handle those issues later (and update po files accordingly)

[codecov-io]

Codecov Report

❗ No coverage uploaded for pull request base (master@21e126f). Click here to learn what that means.
The diff coverage is n/a.

@@            Coverage Diff            @@
##             master    #1275   +/-   ##
=========================================
  Coverage          ?   68.76%           
  Complexity        ?     3855           
=========================================
  Files             ?      174           
  Lines             ?    16029           
  Branches          ?     2612           
=========================================
  Hits              ?    11023           
  Misses            ?     3775           
  Partials          ?     1231

[davecramer]

@horiguti Thanks so much for this!
@vlsi yes, apparently some C&P errors exist

[horiguti]

(I'm not falimiar to this interface so replying here at once)

Sorry for the careless change of permission bits. (It's due to samba's setting).

@vlsi> I've rebased your branch on top of current master, could you please recheck if it still looks good?

I reviewed rebased file and found one difference at line 1608 (tried to call end..) which just changes line splitting position. In short, it looks still fine.

Thanks.

[vlsi]

just changes line splitting position

Yeah, that is one more "feature" of gettext that we cannot influence (there's --no-wrap command line option, however it just does not work)

I'll add a script to pgjdbc that would join multiline strings to a single line.