Mailing-List: contact derby-dev-help@db.apache.org; run by ezmlm
Precedence: bulk
Reply-To: <derby-dev@db.apache.org>
Received-SPF: neutral (nike.apache.org: local policy)
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
  s=s1024; d=yahoo.co.uk;
  h=Message-ID:X-YMail-OSG:Received:X-Mailer:References:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding;
  b=aGZQ1TJSz6/b9xUOW36fMTMLDIkSw+ZYLRBZYtFTkOR5AEU223D1+sEJgznPTNSjXQ6bknrAyWFSi7U6C7ouLCn15PQ9+xynJmVPqx8S3l62ozUCdlYgEk26ho5okOB+a7/u2AWAdArcP9eRB1z/PVre6Ia1oA/N04iQAPp8YtM=;
Message-ID: <156385.56364.qm@web29107.mail.ird.yahoo.com>
References: <866536.18935.qm@web29109.mail.ird.yahoo.com>
 <wjo8occ2lzzs.fsf@oracle.com> <215495.51344.qm@web29101.mail.ird.yahoo.com>
 <wjo87hipn9lk.fsf@oracle.com>
Date: Mon, 13 Sep 2010 20:39:37 +0000 (GMT)
From: Tiago Espinha <tiago.derby@yahoo.co.uk>
Subject: Re: Database name length
To: derby-dev@db.apache.org
In-Reply-To: <wjo87hipn9lk.fsf@oracle.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

Aha, we do indeed Knut, but I just found a bug that is present at least on =
=0A10.5.3.0. If you try to create a database using the client driver with 2=
43 =0A[excluding ;create=3Dtrue] Latin characters (e.g. '=E7'), the server =
will throw the =0Afollowing exception: =0A=0A--------------8<--------------=
--=0A2010-09-13 20:26:11.052 GMT : fromIndex(518) > toIndex(275)=0Ajava.lan=
g.IllegalArgumentException: fromIndex(518) > toIndex(275)=0A        at java=
.util.Arrays.rangeCheck(Arrays.java:1306)=0A        at java.util.Arrays.fil=
l(Arrays.java:2567)=0A        at org.apache.derby.impl.drda.DDMWriter.padBy=
tes(Unknown Source)=0A        at org.apache.derby.impl.drda.DDMWriter.write=
ScalarPaddedBytes(Unknown =0ASource)=0A        at org.apache.derby.impl.drd=
a.DRDAConnThread.writeRDBNAM(Unknown Source)=0A------------8<--------------=
---------=0A=0AThis causes havoc as probably '=E7' is indeed requiring more=
 than one byte to be =0Aencoded. I'm actually surprised this works at all b=
ut it's probably some obscure =0Aaddition of EBCDIC on top of just regular =
US-ASCII. When I try to get the bytes =0Afor '=E7' in plain US-ASCII, I get=
 63 which is ASCII for the question mark.=0A=0AThis means that as it is, on=
 the client driver, the limit will be shorter than =0A255 characters if the=
se special characters are used. The only matter for concern =0Ais that in p=
revious releases this limit isn't properly handled and we just let =0Athe s=
erver crash with a serious unhandled exception.=0A=0AI'm all for raising th=
is limit if the community agrees that this does not =0Aviolate the DRDA pro=
tocol. I think Kathey also expressed some concern about the =0Alimit on JIR=
A, so it would be interesting to hear her thoughts on this.=0A=0ATiago=0A=
=0A----- Original Message ----=0AFrom: Knut Anders Hatlen <knut.hatlen@orac=
le.com>=0ATo: derby-dev@db.apache.org=0ASent: Mon, 13 September, 2010 12:53=
:11=0ASubject: Re: Database name length=0A=0ATiago Espinha <tiago.derby@yah=
oo.co.uk> writes:=0A=0A> Thank you Knut, for your reply.=0A>=0A> Your point=
 #1 is correct. As for point #2 and #3 just a small correction. It is =0A>=
=0A> all characters falling outside the *US-ASCII* encoding that will get a=
 length =0A> lower than 255 characters as anything other than ASCII require=
s more than just =0A>1 =0A>=0A> byte to encode. I'm fairly sure that at thi=
s point we do not support ISO-8859-1 =0A>=0A> through the client driver as =
these characters (the extended ones like =E1=E9=F3 etc) =0A=0A> fall outsid=
e US-ASCII. So hopefully this won't break anything as we didn't =0A> suppor=
t these characters previously.=0A=0AI think we do support this. At least, t=
his code ran fine on my machine=0Awith Derby 10.6.1.0 and created a databas=
e named =E1=E9=F3:=0A=0A    Connection c =3D DriverManager.getConnection(=
=0A            "jdbc:derby://localhost/\u00E1\u00E9\u00F3;create=3Dtrue");=
=0A=0A> As for your suggestion of increasing the length of the field, I'm n=
ot sure =0A> that's an option. This length limitation is imposed by the DRD=
A specification =0A> and the ACR unfortunately didn't change this. On the A=
CR it reads "As of DDM =0A> Level 7, the RDBNAM can accommodate an RDB name=
 of up to 255 bytes in length, =0A> and its format will vary depending on t=
he length of the RDB name". So =0A> essentially, we could easily support a =
much larger RDB name on Derby but the =0A> specification forbids it.=0A=0AD=
RDA does allow product-unique extensions. I have a hard time seeing any=0Ad=
ownside with extending the protocol here. Sending a longer string than=0Awh=
at a strict reading of the spec permits sounds like a lesser evil to=0Ame t=
han disallowing network access to the database.=0A=0A-- =0AKnut Anders=0A=
=0A=0A=0A