Re: [RFC/PATCH] l10n: de.po: translate 825 new messages

Hi Ralf,

This is the middle third of my review.  Sorry for the long wait!  I hope
it can still be useful.

Also, sorry for the other mail that I just sent out seconds ago.  That
was what I wanted to save as "to be done in the last third", but
I accidentally hit the wrong key and sent it out.

- Thomas

I don't really share your apparent aversion to just using "<n>" :-)

 #: builtin/for-each-ref.c:994
 msgid "quote placeholders suitably for shells"
-msgstr ""
+msgstr "quotiert Platzhalter geeignet für Shells"

[...]

 #: builtin/for-each-ref.c:996
 msgid "quote placeholders suitably for perl"
+msgstr "quotiert Platzhalter geeignet für Perl"

[etc]

"quote" is a hard one, and "quotiert" sounds very wrong to me.  How
about weaseling out along the lines of

  Formatiert Platzhalter als {Shell|Perl|Python|TCL} Zeichenfolge

or even "Shell-String".

 #: builtin/for-each-ref.c:1003
 msgid "show only <n> matched refs"
-msgstr ""
+msgstr "zeigt nur <Nummer> passende Referenzen"

This should be <Anzahl> if you insist on translating the 'n'.

 #: builtin/for-each-ref.c:1006
 msgid "field name to sort on"
-msgstr ""
+msgstr "Name des Felder zum Sortieren"

Feld_s_?

Or just "sortiere nach diesem Feld".

 #: builtin/fsck.c:618
 msgid "make index objects head nodes"
-msgstr ""
+msgstr "erzeugt Kopfknoten der Bereitstellungsobjekte"

The English version is equally obscure, but this refers to (also)
walking the objects listed in the index.  Furthermore, --cache is
actually the default if you don't list any objects to walk.

 #: builtin/fsck.c:620
 msgid "also consider packs and alternate objects"
-msgstr ""
+msgstr "betrachtet auch Pakete und wechselnde Objekte"

The glossary tells me pack -> Archiv.

An alternate object [store] is essentially a pointer to the object store
of another repository.  It may best be left untranslated, as it's a
pretty advanced feature.

 #: builtin/grep.c:741
 msgid "match patterns only at word boundaries"
-msgstr ""
+msgstr "findet Muster nur innerhalb von Wörtern"

The translation gets the purpose wrong: the option forces a
word-boundary match on either side of the pattern, so that you will find
only one or more words matching the pattern.  The manpage has a precise
definition:

	Match the pattern only at word boundary (either begin at the
	beginning of a line, or preceded by a non-word character; end at
	the end of a line or followed by a non-word character).

Maybe "suche nur ganze Wörter" would suffice as a concise description.

 #: builtin/grep.c:817
 msgid "indicate hit with exit status without output"
-msgstr ""
+msgstr "kennzeichnet Übereinstimmungen mit Beendigungsstatus, ohne Ausgabe"

The "indicate" should be "zeigen" or some such, since it only tells
whether there were any matches at all.

I spent some time looking for a better translation of "exit status", but
couldn't find anything helpful (such as a German translation of diff(1)
or a Wikipedia page on "exit status").

 #: builtin/hash-object.c:77
 msgid "process file as it were from this path"
-msgstr ""
+msgstr "verarbeitet Datei, wie es von diesem Pfad wäre"

als ob sie von diesem Pfad wäre?

 #: builtin/help.c:47
 msgid "show info page"
-msgstr ""
+msgstr "zeigt Informationsseite"

info here refers to GNU info, presumably it should be left as
"info-Seite" or some such.

*vervollständigt* mit %d lokalen Objekten

 #: builtin/log.c:102
 msgid "decorate options"
-msgstr ""
+msgstr "Ausgabeoptionen"

'decorate' is the option to mark commits by the refs pointing at them,
and should probably not be translated.

The English version is quite tautological though, since it just turns
into

    --decorate[=...]      decorate options

 #: builtin/log.c:1091
 msgid "don't include a patch matching a commit upstream"
 msgstr ""
+"schließt keine Patches ein, die einer Version im externen Projektarchiv "
+"entsprechen"

Upstream refers to the notional upstream state, which does not even have
to be a branch here.  I.e., if you say

  git format-patch --ignore-if-in-upstream <commit>

then it will format the range 'commit..HEAD', ignoring patches that are
already in 'HEAD..commit'.

So it's not an upstream repo, and in fact doesn't even have to be a
branch.  I don't really have a good suggestion though.  Perhaps you can
still call it "im Übernahmezweig", since the original makes the same
mistake.

 #: builtin/log.c:1095
-#, fuzzy
 msgid "Messaging"
-msgstr "Zusammenführung:"
+msgstr "Datentransfer"

"Email-Einstellungen"?

 #: builtin/log.c:1096
-#, fuzzy
 msgid "header"
-msgstr "voraus "
+msgstr "Kopfteil"
 
 #: builtin/log.c:1097
 msgid "add email header"
-msgstr ""
+msgstr "fügt Kopfteil für eMail hinzu"

Does anyone call them anything except "headers" in german?

I can even quote a (non)reference:

  http://de.wikipedia.org/wiki/E-Mail#Header_.E2.80.93_der_Kopf_der_E-Mail

 #: builtin/log.c:1108
-#, fuzzy
 msgid "inline the patch"
-msgstr "ignoriert hinzugefügte Zeilen des Patches"
+msgstr "fügt einen Patch ein"

fügt den Patch direkt in die Nachricht ein

 #: builtin/ls-files.c:463
 msgid "identify the file status with tags"
-msgstr ""
+msgstr "bestimmt den Dateistatus mit Markierungen"

*zeigt* den Dateistatus mit Markierungen

 #: builtin/merge.c:189
 msgid "add (at most <n>) entries from shortlog to merge commit message"
 msgstr ""
+"fügt (höchstens <Nummber>) Einträge von \"short\" zur Beschreibung der "

                        ^^^
typo  -------------------´

+"Zusammenführung hinzu"

Furthermore, you changed 'shortlog' to '"short"'.  Presumably you meant
to keep the 'log' too?

 #: builtin/notes.c:617
 msgid "load rewriting config for <command> (implies --stdin)"
 msgstr ""
+"lädt Konfiguration für <Kommando> beim Überschreiben von Versionen "
+"(impliziert --stdin)"

This would be a good time to settle on a good translation for
"rewriting".  Perhaps "neu schreiben".  "Überschreiben" to me implies
that it's done in-place, but the whole point of calling it "rewriting"
instead of "changing" is that it is *not* done (or indeed possible to
do) in-place.

 #: builtin/pack-objects.c:2502
 msgid "create thin packs"
-msgstr ""
+msgstr "erzeugt kleinere Pakete"

Smaller is not really the point: they are packs that do not have the
delta bases.  It can only be used in the context of a repository that
has them.  A thin pack is used only(?) in network transfer, and then
"completed" by adding the delta bases to it, which lets it stand on its
own.

You could call them "abgespeckt" ;-)


-- 
Thomas Rast
trast@{inf,student}.ethz.ch