Re: Clarify `pcase' `rx' pattern doc

[Yuri Khan]

On Wed, Jun 13, 2018 at 3:22 PM Michael Heerdegen
<address@hidden> wrote:

> BTW this aspect concerns more or less all pcase pattern definitions: The
> first sentence either is not a sentence in the strict sense, or says
> "Build a pcase pattern matching..."  We should probably unify the
> language used.

The standard patterns are documented in this form:

    <pattern> matches <something in some conditions>.

It would be nice if extended patterns defined using pcase-defmacro
also followed this syntax:

    -- (seq &rest PATTERNS)

    Matches a SEQUENCE such that each element of PATTERNS matches a
corresponding element of SEQUENCE.

    Extra elements of the sequence are ignored if fewer PATTERNS are
given, and the match does not fail.

In other words, for the user, it is not important that underlying the
‘(seq …)’ is a ‘seq--pcase-macroexpander’ function that builds a pcase
pattern out of simpler constructs. From the user’s viewpoint, ‘(seq
…)’ *is* the pattern.