Path: csiph.com!eternal-september.org!feeder.eternal-september.org!mx02.eternal-september.org!.POSTED!not-for-mail
From: Jussi Piitulainen <jussi.piitulainen@helsinki.fi>
Newsgroups: comp.lang.python
Subject: Re: The Cost of Dynamism (was Re: Pyhon 2.x or 3.x, which is faster?)
Date: Tue, 22 Mar 2016 14:52:47 +0200
Organization: A noiseless patient Spider
Lines: 51
Message-ID: <lf5d1qmisio.fsf@ling.helsinki.fi>
References: <mailman.8.1457732171.12893.python-list@python.org> <nbvqg5$3cm$1@dont-email.me> <mailman.23.1457749230.12893.python-list@python.org> <56e44258$0$1598$c3e8da3$5496439d@news.astraweb.com> <mailman.46.1457819135.12893.python-list@python.org> <8737rvxs89.fsf@elektro.pacujo.net> <nc2a7s$6dv$1@dont-email.me> <mailman.48.1457831488.12893.python-list@python.org> <nc4ffn$f6s$1@dont-email.me> <56e7483d$0$1608$c3e8da3$5496439d@news.astraweb.com> <nc7kjr$jj8$1@dont-email.me> <ncnhp9$tt7$1@dont-email.me> <ncnip4$1pb$1@ger.gmane.org> <CAPTjJmqA4rs74JWXinvqHhELzjhhe3MtAruS_WnDB7OG_=FbVA@mail.gmail.com> <mailman.423.1458525784.12893.python-list@python.org> <ncpin0$4m9$1@dont-email.me> <56f09973$0$1601$c3e8da3$5496439d@news.astraweb.com> <ncr8n3$625$1@dont-email.me>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Injection-Info: mx02.eternal-september.org; posting-host="305c68510616a2e7ac08bcd2ff1598bd"; logging-data="30072"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX18SQgfbaMyJV5wGsgW4APhwvUiuXi+Z/cc="
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux)
Cancel-Lock: sha1:gR6alv3z5bw+4D8AcWmlPxd9c04= sha1:IcqfAtqZpptl+5piiKVQ0WzizAA=
Xref: csiph.com comp.lang.python:105474

BartC writes:

> Not everything fits into a for-loop you know! Why, take my own
> readtoken() function:
>
>   symbol = anything_other_than_skip_sym
>
>   while symbol != skip_sym:
>      symbol = readnextsymbol()
>
> Of course, a repeat-until or repeat-while would suit this better (but
> I don't know how it fits into Python syntax). So there's a case here
> for increasing the number of loop statements not reducing them.

Not sure why nobody seems to respond to this part. Perhaps I just missed
it? It's true that while has its uses, or at least I think I've used it
in Python once or twice. But there's more fun to be had by turning your
data into a stream-like object.

stream = iter('   /* this is C! */') # <-- produces a character at a time

Now you can ask for the next item that satisfies a condition using a
generator expression:

next(symbol for symbol in stream if not symbol.isspace())
---> '/'

next(symbol for symbol in stream if not symbol.isspace())
---> '*'

Or collect the remaining items:

list(symbol for symbol in stream if not symbol.isspace())
---> ['t', 'h', 'i', 's', 'i', 's', 'C', '!', '*', '/']

You could also say:

for symbol in stream:
   if symbol.isspace(): continue
   ...

But this particular stream is empty by now. I work with long streams of
tokenized and annotated sentences (which for me are streams of tokens)
that sometimes come packed in streams of paragraphs packed in streams of
texts. I build whatever stream I happen to want by nesting generator
functions and generator expressions and some related machinery. (You
could build on a character stream or a byte stream that you obtain by
opening a file for reading; I tend to read line by line through
itertools.groupby, because that's what I do.)

These things compose well.