Date: Fri, 01 Aug 2014 08:40:37 +0200 From: John Marino <freebsd.contact@marino.st> To: Alexey Dokuchaev <danfe@FreeBSD.org>, marino@freebsd.org Cc: "svn-ports-head@freebsd.org" <svn-ports-head@freebsd.org>, Max Brazhnikov <makc@freebsd.org>, Antoine Brodin <antoine@FreeBSD.org>, "svn-ports-all@freebsd.org" <svn-ports-all@freebsd.org>, William Grzybowski <wg@freebsd.org>, "ports-committers@freebsd.org" <ports-committers@freebsd.org>, Mathieu Arnold <mat@freebsd.org>, Adam Weinberger <adamw@adamw.org> Subject: Re: svn commit: r363361 - in head/editors/fte: . files Message-ID: <53DB3665.6020201@marino.st> In-Reply-To: <20140801063243.GE71116@FreeBSD.org> References: <6270029E710D3C52B60B6224@atuin.in.mat.cc> <20140730081413.GA29876@FreeBSD.org> <20140730160843.GA2688@FreeBSD.org> <53D919BC.6010307@marino.st> <20140801051652.GA59625@FreeBSD.org> <CAALwa8=R2i9PA7R8VXHStZCDGy_ZFaYcBtRmY-_ZFq9k=toJ%2Bw@mail.gmail.com> <20140801060506.GA71116@FreeBSD.org> <53DB3021.7010909@marino.st> <20140801062047.GC71116@FreeBSD.org> <53DB32E8.7020700@marino.st> <20140801063243.GE71116@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 8/1/2014 08:32, Alexey Dokuchaev wrote:
> On Fri, Aug 01, 2014 at 08:25:44AM +0200, John Marino wrote:
>> On 8/1/2014 08:20, Alexey Dokuchaev wrote:
>>> Basically right now one patch will overwrite another conflicting one, which
>>> ever comes first. I think I can add a check beforea > $${OUT} and issue an
>>> error and refuse to overwrite a file, or maybe just warn. Which you prefer?
>>
>> Neither. I prefer the original character escaped with itself You may
>> not like this, but something like: [...]
>
> You're right, I don't. :-) However, I will implement autoexpantion (char
> -> charchar) since it doesn't add too much complexity.
>
>> Thoughts? any potential of overwritting is bad.
>
> Well, my original idea (bail out with an error) is how I saw we prevent
> overwriting: by asserting that output file does not exist; you think it is
> not robust enough?
No, because you don't cover the case where both files legitimately need
patching. Why bail out with error when I *want* both patches to exist?
You are forcing me to rename one? to what?
I've been using "_" => "__" for a while now, it's not bad. Yes, double
underscore remains but a very low frequency. The main issue is that it
represents a different character ("_") than it does now ("\").
John
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?53DB3665.6020201>