Mailing-List: contact dev-help@commons.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Commons Developers List" <dev@commons.apache.org>
Received-SPF: pass (nike.apache.org: domain of flamefew@gmail.com designates
 74.125.82.171 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type:content-transfer-encoding;
        b=YGfUYUAI9Y1zR6p8NoO6JsNv/sYO0tRJrUZjyft1zcIXNAiSdN7Wh/O9G4OpDuYez4
         uB88zAPsVxzDErkWQ77DBmhmPRnAunuLdrokw7Nxorendg44RRxnWbX/P3n0/S9itMat
         2PEC4nQAlPFDRZ8Eia4v9hm9bpSZoVhfzoWdw=
MIME-Version: 1.0
In-Reply-To: <BANLkTikOiX_tVchgf77gT5v1ytMb1H8EWg@mail.gmail.com>
References: <BANLkTikOiX_tVchgf77gT5v1ytMb1H8EWg@mail.gmail.com>
Date: Fri, 8 Apr 2011 01:36:18 -0700
Message-ID: <BANLkTinazvDyG5eDrr37GpMrBKCX0N==uw@mail.gmail.com>
Subject: Re: Translators and Range [Was: [VOTE] Release Commons Lang 3.0
 (RC1)]
From: Henri Yandell <flamefew@gmail.com>
To: Commons Developers List <dev@commons.apache.org>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

On Thu, Apr 7, 2011 at 2:55 PM, Henri Yandell <flamefew@gmail.com> wrote:
> On Mon, Mar 7, 2011 at 10:05 PM, Henri Yandell <flamefew@gmail.com> wrote=
:
>> On Sat, Mar 5, 2011 at 8:46 AM, Oliver Heger
>> <oliver.heger@oliver-heger.de> wrote:
>>> Two minor points from my side:
>>>
>>
>>> - Just a proposal: There are some translators in the new translate pack=
age
>>> which can be configured with a range of the codes to be processed. Woul=
d it
>>> make sense to use the Range class for this purpose? Then configuration =
could
>>> be more flexible (because multiple ranges could be specified), and ther=
e
>>> would probably be less duplicate code for checking the ranges.
>>
>> Very interesting.
>>
>> Also makes me wonder if Range should support the notion of
>> Range.above, Range.below and Range.outside in addition to
>> Range.between and Range.is. That change the API from:
>>
>> =A0UnicodeEscaper.between(x, y)
>>
>> to:
>>
>> =A0new UnicodeEscaper(Range.between(x,y))
>> =A0new UnicodeEscaper(Range.above(y))
>> =A0new UnicodeEscaper(Range.below(x))
>> =A0new UnicodeEscaper(Range.outsideOf(x,y))
>>
>> For the translators; sounds great. I'm trying to remember if I hit
>> problems introducing the feature of either an open-ended Range, or an
>> inverted Range. Looking at LANG-551, it looks like I tried to
>> introduce the inverted Range notion (outsideOf) so that I could merge
>> in CharRange, and it never really worked.
>>
>> Worth trying again I think.
>
> I've begun by implementing the code in the translators as a new
> CodepointRange. Stunningly (actually I was a bit surprised) I've
> rewritten CharRange :) So at a minimum I think:
>
> a) CharRange should be updated for codepoints.
> b) Translators should use CharRange.
>
> I also tried to update Range to support the necessary features, and
> the two APIs don't fit too well together. I'm going to go ahead and do
> a) and b); then ponder the CharRange into Range question again.

Scratch that, as you'll see from svn I was able to move the translate
package classes over to the Range class without too much pain. The
lack of being able to say 'minimum excluded' is a bit annoying; and
you have to define 'outsideOf' as two separate ranges. I think the
latter is the right thing and I'm working on the former.

I'm working on moving CharSet over to Range (and dumping CharRange).
All very possible; it's the additional APIs of toString and having
tests that check the input was parsed correctly that are a pain as
both of them are reflections on CharSet's private mechanics.

Hen

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org