MIME-Version: 1.0
References: 
 <CA+hUKGLzfb0cBYaKybD7q+N8iypLc=ENh+7XxQqEmufzcTv3sA@mail.gmail.com>
In-Reply-To: 
 <CA+hUKGLzfb0cBYaKybD7q+N8iypLc=ENh+7XxQqEmufzcTv3sA@mail.gmail.com>
From: Julien Rouhaud <rjuju123@gmail.com>
Date: Fri, 28 May 2021 13:00:36 +0800
Message-ID: 
 <CAOBaU_Z_7A0M3khUbPaS1L8uq7dAF4UQcXDoyHk7ytk7rvZ1cw@mail.gmail.com>
Subject: Re: French translation typos
To: Thomas Munro <thomas.munro@gmail.com>
Cc: pgsql-translators@postgresql.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: 
 <https://www.postgresql.org/message-id/CAOBaU_Z_7A0M3khUbPaS1L8uq7dAF4UQcXDoyHk7ytk7rvZ1cw%40mail.gmail.com>
Precedence: bulk

On Fri, May 28, 2021 at 12:31 PM Thomas Munro <thomas.munro@gmail.com> wrot=
e:
>
> It's not my language so I could be totally wrong in some of these
> cases, but I noticed a few typos while grepping for an error message,
> and then I couldn't help looking for more (from my recollection of the
> rules):
>
> msgstr "contrainte %s non suppor=C3=A9e avec la d=C3=A9finition de cl=C3=
=A9 de partitionnement"
>
>   suppor=C3=A9e -> support=C3=A9e

Indeed.

> msgstr "La table =C2=AB %s =C2=BB contient des partitionso qui ne sont pa=
s des
> tables distantes."
>
>   partitionso -> partitions

+1

> Several like this:
>
> msgstr "la contrainte =C2=AB %s =C2=BB entre en conflit avec la constrain=
te non
> h=C3=A9rit=C3=A9e sur la relation =C2=AB %s =C2=BB"
> msgstr ""
> "les constraintes sur des tables temporaires ne peuvent r=C3=A9f=C3=A9ren=
cer que des\n"
> "tables temporaires"
> ... (a few more like that) ...
>
>   constrainte -> contrainte
>   constraintes -> contraintes

+1

> msgstr "les triggers sur =C3=A9v=C3=A9nemenr ne sont pas support=C3=A9s p=
our %s"
>
>   =C3=A9v=C3=A9nemenr -> =C3=A9v=C3=A9nement(s?)

It should be =C3=A9v=C3=A9nement (singular)

> That's clearly a typo at the end, but separately:  In many places we
> have =C3=A9v=C3=A9nement, but supposedly =C3=A9v=C3=A8nement is the corre=
ct modern
> spelling (?); I think it's the same for cha=C3=AEne (old, we're using tha=
t)
> vs chaine (modern), but I have no idea if anyone cares about spelling
> reforms.

Indeed, in both cases the 2nd version was introduced in the 1990
reform [1].  It's a matter of taste, but I personally dislike most of
the changes that it introduced.  I'm assuming that most of younger
people will disagree :)

> msgstr "ne peut pas pu envoyer un message de taille %zu via la queue
> en m=C3=A9moire partag=C3=A9e"
>
>   ne peut pas pu -> ne peut pas

Indeed, probably the translator started with "n'a pas pu" and switched
to present tense.

> msgstr "la colonne =C2=AB %s =C2=BB ne peut pas =C3=AAtre utilis=C3=A9 da=
ns des
> statistiques parce que son type %s n'a pas de classe d'op=C3=A9rateur btr=
ee
> par d=C3=A9faut"
>
>   utilis=C3=A9 -> utilis=C3=A9e
>
> msgstr "l'expression ne peut pas =C3=AAtre utilis=C3=A9 dans des statisti=
ques
> multivariates parce que son type %s n'a pas de classe d'op=C3=A9rateur
> btree par d=C3=A9faut"
>
>   utilis=C3=A9 -> utilis=C3=A9e
>
> msgstr "la contrainte doit aussi =C3=AAtre valid=C3=A9es sur les tables e=
nfants"
>
>   valid=C3=A9es -> valid=C3=A9e
>
> msgstr "les fonctions de distance =C3=A0 perte ne sont pas support=C3=A9s=
 dans
> les parcours d'index seul"
>
>   support=C3=A9s -> support=C3=A9es
>
> Several like this:
>
> msgstr "option de CLUSTER =C2=AB %s =C2=BB non reconnu"
> msgstr "option =C2=AB %s =C2=BB non reconnu"
> msgstr "variable =C2=AB %s =C2=BB du filtre non reconnu"
> msgstr "option d'EXPLAIN =C2=AB %s =C2=BB non reconnu"
> msgstr "option de REINDEX =C2=AB %s =C2=BB non reconnu"
>
>   reconnu -> reconnue

Indeed, good catch :)

> msgstr ""
> "Initialise le niveau (=C2=AB facility =C2=BB) de syslog =C3=A0 utilis=C3=
=A9 lors de l'activation\n"
> "de syslog."
>
>   utilis=C3=A9 -> utiliser
>
> msgstr "la ligne =C3=A0 verrouill=C3=A9e =C3=A9tait d=C3=A9j=C3=A0 d=C3=
=A9plac=C3=A9e dans une autre
> partition du fait d'une mise =C3=A0 jour concurrente"
>
>   verrouill=C3=A9e -> verrouiller
>
> msgstr "=C3=89checs multiples --- l'erreur d'=C3=A9criture pourrait =C3=
=AAtre permanent."
>
>   permanent -> permanente
>
> msgstr "nombre de descripteurs de fichier insuffisants pour lancer le
> processus serveur"
>
>   insuffisants -> insuffisant
>
> msgstr "longueurs d'=C3=A9l=C3=A9ment corrompus : total %u, espace dispon=
ible %u"
>
>   corrompus -> corrompues
>
> msgstr "L'utilisateur =C3=A9tait connect=C3=A9 =C3=A0 une base de donn=C3=
=A9e qui doit
> =C3=AAtre supprim=C3=A9."
>
>   supprim=C3=A9 -> supprim=C3=A9e

All good!

> msgid "floating-point exception"
> msgstr "exception d=C3=BB =C3=A0 une virgule flottante"
>
>   Should be "d=C3=BBe", or would this be better as "exception en virgule =
flottante"?

"d=C3=BBe" indeed.  I'm not sure which one would be better actually, but
most of the French translation is probably using a style similar to
the 1st one so keeping it may be better for consistency.  Guillaume
may have a better opinion :)

> msgstr "trop de lex=C3=A8mes dans l'entre du th=C3=A9saurus"
>
>   entre -> entr=C3=A9e (?)
>
> msgstr "le dictionnaire Ispell supporte seulement les valeurs de
> drapeau =C2=AB default =C2=BB, =C2=AB long =C2=BBet =C2=AB num =C2=BB"
>
>   (missing space)
>
> msgstr "Valeur  manquante de la dimension du tableau."
>
>   (extra space)
>
> msgid "invalid number of dimensions: %d"
> msgstr "nombre de dimensions invalides : %d"
>
>   invalides -> invalide
>
> msgstr "la fonctions de hashage 1 doit avoir un argument"
> ... several like this ...
>
>   fonctions -> fonction

all good!

>   hashage -> hachage (that's how it's written elsewhere...), or maybe
> h=C3=A2chage (hash joins are called jointures h=C3=A2ch=C3=A9es)?

It should be "hachage".  I don't think that "h=C3=A2che" is correct in Fren=
ch.

> msgstr "%s =3D %d est un param=C3=A8trage plus bas que celui du serveur
> primaire, o=C3=B9 sa valeur =C3=A9tait %d."
>
>   param=C3=A8trage -> param=C3=A9trage (3 examples)
>
>   msgstr "Les param=C3=A8tres jsonpath doivent =C3=AAtre encod=C3=A9s en =
paires
> cl=C3=A9-valeur d'objets =C2=AB vars=C2=BB"
>
>   (missing space)
>
> msgstr "la cible du lien symbolique =C2=AB %s =C2=BB est trop long"
> msgstr "cible du lien symbolique trop long pour le format tar  : nom
> de fichier =C2=AB %s =C2=BB, cible =C2=AB %s =C2=BB"
>
>   long -> longue
>
> "la fonction avec set-value a =C3=A9t=C3=A9 appel=C3=A9 dans un contexte =
qui n'accepte pas\n"
> "un ensemble"
>
>   appel=C3=A9 -> appel=C3=A9e
>
> msgstr "certains droits n'ont pu =C3=AAtre r=C3=A9voqu=C3=A9 pour la colo=
nne =C2=AB %s =C2=BB
> de la relation =C2=AB %s =C2=BB"
> msgstr "certains droits n'ont pu =C3=AAtre r=C3=A9voqu=C3=A9 pour =C2=AB =
%s =C2=BB"
>
>   r=C3=A9voqu=C3=A9 -> r=C3=A9voqu=C3=A9s
>
> msgstr "unit=C3=A9s d'intervalle =C2=AB %s =C2=BB non support=C3=A9 car l=
es mois ont
> g=C3=A9n=C3=A9ralement des semaines fractionnaires"
>
>   support=C3=A9 -> support=C3=A9es (? or somehow this agrees with %s?)

I think it's "support=C3=A9es".

> msgid "distance in phrase operator should be non-negative and less than %=
d"
> msgstr "la distance dans l'op=C3=A9rateur de phrase devrait pas positif e=
t
> inf=C3=A9rieur =C3=A0 %d"
>
>   devrait pas positif et inf=C3=A9rieur =C3=A0 -> devrait =C3=AAtre non n=
=C3=A9gative et
> inf=C3=A9rieure =C3=A0 (?)

Agreed

> Sorry if I'm wrong about any of these, I'll be happy to learn...

Your skills are impressive!

[1]: https://fr.wiktionary.org/wiki/Annexe:Rectifications_orthographiques_d=
u_fran%C3%A7ais_en_1990