Can we together compile a list of accepted abbreviations, so they can be
consistently applied?

I'll begin with the most basic ones:

    aref        array reference
    bool        boolean
    const       constant
    elem        element
    err         error
    fh          filehandle
    func        function
    href        hash reference
    int         integer
    kv          key/value
    num         number
    obj         object
    op          operator, operation
    r           read(able)
    ref         reference
    regex       regular expression
    rw          read/write
    str         string
    sub         subroutine
    w           write(able)

Some may be source of discussion, in which case I think it's better to
pick one -perhaps against someone's preference- than to use a dozen
abbreviations for the same thing. One such example is "regex", which
used to be spelled by most as "regexp". (Yes, I know we're using rules
now... :).)

Juerd
--
http://convolution.nl/maak_juerd_blij.html
http://convolution.nl/make_juerd_happy.html
http://convolution.nl/gajigu_juerd_n.html

Juerd skribis 2005-04-22 16:11 (+0200):

Although readline should not be made rline, and I still think both :r
and :read should work!

    Bool +$read is short<r>

or preferrably with even simpler syntax:

    Bool +$read|r

Juerd
--
http://convolution.nl/maak_juerd_blij.html
http://convolution.nl/make_juerd_happy.html
http://convolution.nl/gajigu_juerd_n.html

Are you suggesting that these are accepted by the compiler or by us (for
discussion)?

This makes it sound like you mean us, not the compiler. I'll work with
that assumption.

You can decide to use whatever you like, but if you're honestly
expecting that people participating in this mailing list are going to
use your favorite abbreviations, you're the kind of person I think we
should have more of in the world: optimists (opmst for short). ;-)

Seriously, do any of us have a hard time making out what someone means
if they say "regexp" vs "regex"?

What's more, I'd rather you didn't w comments with single-letter
abbreviations, as it would make it much harder for me to r.

--
Aaron Sherman <a...@ajs.com>
Senior Systems Engineer and Toolsmith
"It's the sound of a satellite saying, 'get me down!'" -Shriekback

Aaron Sherman skribis 2005-04-22 10:00 (-0400):

By us. Mostly for the purpose of choosing identifiers (str, and thus
substr, not substring), but also for discussion (we talk about subs, not
subroutines) and perhaps documentation.

See below for further clarification.

No, it's not to change the world. But whatever ends up in the official
thing, should be consistent throughout. And for that, we need consensus.

Because it can take a long time before people agree, I think we should
begin as early as possible.

Using identifiers consistently also helps when reading example code. For
example, $fh is immediately clear, as are $dbh and $sth. But when
someone uses $handle, $connection and $statement for these, it gets a
bit harder.

Learning something is easiest when in the explanation of one thing, you
recognise parts of something you've already seen. And for that, you need
consistent documentation, along with consistent usage in the language
itself.

To provide one example: a few years ago I had a very hard time trying to
figure out what 'ary' stood for, simply because in the lots of Perl
documentation and code, I had not seen it before.

Things that come naturally for experienced programmers can be very
intimidating for beginners, and wherever we can, I think we should
provide consistency.

Of course I'm NOT saying that we should start considering one of regex
and regexp wrong. When it comes to reading, both are equally good and
recogniseable to everyone subscribed to this mailinglist. But many
beginners wonder what the p in regexp stands for!

Those are for identifiers, so we don't end up with one function using :r
and another using :read. That'd be inconsistent.

I should have been clearer when I sent the initial list. I'm not trying
to change the world. Sorry about that.

Juerd
--
http://convolution.nl/maak_juerd_blij.html
http://convolution.nl/make_juerd_happy.html
http://convolution.nl/gajigu_juerd_n.html

interp   interpreter
i   iterator?

expr   expression

ro   read-only

in   input
out   output
io   input/output
mem   memory
os   operating system
anon   anonymous

interestingly, among references there are aref and href, but scalar and
subroutine references are not presented.

Vadim

konovalo skribis 2005-04-24 10:24 (+0400):

I don't recall having seen that used.

Used as such, but I think discouraging i in favour of iter is a better
idea, because i is also often used as index.

Is this used?

Because I have never seen them used as <singleletter>ref yet. Probably
because sref would be highly ambiguous.

Juerd
--
http://convolution.nl/maak_juerd_blij.html
http://convolution.nl/make_juerd_happy.html
http://convolution.nl/gajigu_juerd_n.html