References: <47df73f1-d26d-54ff-afab-b0237c6a933f@morningstar2.co.uk>
            <201707211607.v6LG7wsh007401@bilbo.localnet>
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101
            Thunderbird/52.2.1
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Language: en-GB
Content-Transfer-Encoding: 7bit
Message-ID:  <1158d236-9da2-cd46-6181-1ac0740e6d47@morningstar2.co.uk>
Date:         Fri, 21 Jul 2017 17:11:43 +0100
Reply-To:     Mailing list for the LaTeX3 project
              <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
Sender:       Mailing list for the LaTeX3 project <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
From:         Joseph Wright <joseph.wright@MORNINGSTAR2.CO.UK>
Subject: Re: \bool_if:n(TF) now *greedy*
To:           LATEX-L@LISTSERV.UNI-HEIDELBERG.DE
In-Reply-To:  <201707211607.v6LG7wsh007401@bilbo.localnet>
Precedence: list
Envelope-To: <rainer.schoepf@GMX.NET>
Status: R

On 21/07/2017 17:07, Boris Veytsman wrote:
> JW> Date:         Fri, 21 Jul 2017 08:59:33 +0100
> JW> From:         Joseph Wright <joseph.wright@MORNINGSTAR2.CO.UK>
> 
> 
> JW> In the latest CTAN update, \bool_if:n(TF) has been made *greedy*:
> JW> \bool_lazy_... functions are available for lazy evaluation. (The latter
> JW> have a different input syntax which does not suffer from the issue that
> JW> affects \bool_if:n(TF).) Any code which still assumes lazy evaluation
> JW> for \bool_if:n(TF) will therefore need adjustment.
> 
> Would not it be a little gentler to the developers to keep
> \bool_if... lazy and introduce a new family \bool_greedy...?

The problem would then have remained: there are certain forms of input
which with a lazy \bool_if:n(TF) break the parsing and lead to low-level
errors. Having a permanently-broken \bool_if:n(TF) was on-balance not
something we felt was the best solution.

Joseph