Path: csiph.com!usenet.pasdenom.info!weretis.net!feeder1.news.weretis.net!feeder.erje.net!newsfeed.xs4all.nl!newsfeed6.news.xs4all.nl!xs4all!post.news.xs4all.nl!not-for-mail
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=wtOdr3VkjnSqCf6ohkqOt0Fy+3hfsWj2ILbMO02xpAOAASpM6CClGPQQnF0aOH9aBkPhLvyByHERiWgWsJiyqCxfKYGuBkJa1zgPdHaAk/i6EyDQapqbwfEIyWcOjODoqqYBkb2BsTon75207nMRsHQOUeQPmhGj58subRR8gu8=;
References: <CABgq=FyhR+Ldujj3YKRBpVXRVoeoayuXZviUUcNx-gXnFiHLSw@mail.gmail.com> <CACoeR0xV=28j6rFHm1ep3gzH9C0sbDqyxRRwX-JrVm3G+rrXww@mail.gmail.com> <1333174946.18436.YahooMailNeo@web121506.mail.ne1.yahoo.com> <jl6ma0$ptg$1@dough.gmane.org> <mailman.1200.1333352587.3037.python-list@python.org> <4f7962b0$0$29981$c3e8da3$5496439d@news.astraweb.com>
Date: Mon, 2 Apr 2012 02:11:46 -0700 (PDT)
From: Yingjie Lan <lanyjie@yahoo.com>
Subject: Re: string interpolation for python
To: "python-list@python.org" <python-list@python.org>
In-Reply-To: <4f7962b0$0$29981$c3e8da3$5496439d@news.astraweb.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
Precedence: list
Reply-To: Yingjie Lan <lanyjie@yahoo.com>
Newsgroups: comp.lang.python
Message-ID: <mailman.1205.1333358032.3037.python-list@python.org>
Lines: 46
NNTP-Posting-Host: 2001:888:2000:d::a6
Xref: csiph.com comp.lang.python:22451

=0A=0A=0A=0A----- Original Message -----=0A> From: Steven D'Aprano <steve+c=
omp.lang.python@pearwood.info>=0A> To: python-list@python.org=0A> Cc: =0A> =
Sent: Monday, April 2, 2012 4:26 PM=0A> Subject: Re: string interpolation f=
or python=0A> =0A> On Mon, 02 Apr 2012 00:39:42 -0700, Yingjie Lan wrote:=
=0A> =0A>>>  You can already do essentially that without adding a special-c=
ase=0A>>>  string=0A>> =0A>>>  formatting method to the general methods we =
already have.=0A>>> =0A>>>>>> =A0 balls =3D 5=0A>>>>>> =A0 people =3D 3=0A>=
>>>>> =A0 'The {people} people have {balls}=0A>>>  balls.'.format(**locals(=
))=0A>>>  'The 3 people have 5 balls.'=0A>> =0A>> =0A>>  Clearly dynamic st=
rings are much more powerful, allowing arbitrary=0A>>  expressions inside.=
=0A> =0A> And so it may be a security risk, if user-input somehow ends up t=
reated =0A> as a dynamic string.=0A> =0A> We already have three ways to eva=
luate arbitrary expressions:=0A> =0A> * Python code=0A> * eval=0A> * exec=
=0A> =0A> Why do we need yet another one?=0A> =0A> =0A>>  It is also more t=
erse and readable, since we need no=0A>>  dictionary.=0A> =0A> I think you =
mean terse and unreadable, since we need no dictionary. That =0A> means tha=
t variables will be evaluated by magic from... where? Globals? =0A> Local s=
cope? Non-local scope? All of the above?=0A> =0A> We already have one way o=
f evaluating implicit variables using implicit =0A> rules, namely regular P=
ython code. Why do we need a second one?=0A> =0A> =0A>>  I would probably r=
ather liken dynamic expressions as a little brother of=0A>>  computable doc=
uments=A0in Mathematica. It is a new kind of expression,=0A>>  rather than =
formatting -- though it has formatting connections.=0A> =0A> Why do we need=
 a new kind of expression?=0A> =0A> =0A>>  Dynamic strings are mainly usefu=
l at time of writing readable code=0A>>  before compilation.=0A> =0A> What =
does that mean?=0A> =0A> =0A>>  The compiler can choose to convert it into =
a string=0A>>  formatting expression, of course. To efficiently format=A0st=
rings at=0A>>  runtime, the best choice (especially=0A>>  for safty reasons=
) is string formatting, not evaluating a dynamic=0A>>  string.=0A> =0A> So =
you're suggesting that we should have dynamic strings, but not =0A> actuall=
y use dynamic strings. The compiler should just convert them to =0A> regula=
r string formatting.=0A> =0A> Why not cut out the middle-man and just use r=
egular string formatting?=0A> =0A=0A=0AI believe non of the other three alt=
ernatives are as terse and readable.=0AWe've got template based, formatting=
 with dict, formatting with tuple.=0AThey all require the coder extra effor=
t:=0A=0ABoth template based and dict-based formatting require writing the=
=0Aidentifier three times:=0A=0A>>> name =3D 'Peter'=0A>>> "Are you %(name)=
s"%{'name':name}=0A=A0=0AIf dynamic string is used:=0A>>> "Are you $name$?"=
=0A=0ATemplate:=0A>>> Template("Are you $name?").substitute(name=3Dname)=0A=
=0AIt is three to one in compactness, what a magic 3!=0A=0AOf course, the o=
ld C style way:=0A=0A>>> "Are you %s?"%name=0A=0AAlmost as terse, but not a=
s readable, especially=0Awhen there are many=A0parts to substitute --=0Athe=
 coder and reader need to be careful=A0=0Ato=A0make sure the sequence is co=
rrect.=0A=0AWhy the Python community is so=0Ahostile to new things now?=A0=
=0APython has merits,=0Abut it is far from being perfect.=0A=0ACheers,=0AYi=
ngjie