[Python-ideas] Built-in parsing library

David Mertz mertz at gnosis.cx
Mon Apr 1 09:41:28 EDT 2019

On Mon, Apr 1, 2019 at 2:16 AM Stephen J. Turnbull <
turnbull.stephen.fw at u.tsukuba.ac.jp> wrote:

> In principle, no, one just needs to explain why this battery fits most
> of the toys encountered in practice.  That's good enough, and if
> during discussion somebody shows another one is better on a lot of
> fronts, sure, do that instead.

OK, I'll acknowledge my comment might have overstated the bar to overcome.
A parser added to the standard library doesn't need to be perfect for
everyone.  But adding to stdlib *does* provide a kind of endorsement of the
right default way to go about things.

Among the popular third party libraries, we have several very different
attitudes towards designing grammars.  On the one hand, there are formal
differences in the power of different grammars—some are LL, some LR, some
LALR, some Early.  Also, some libraries separate parser from lexer while
others are scannerless.

But most of these can all parse the simple cases fine, so that's good for
the 90% coverage.  However, cross-cutting that formal power issue, there
are two main programming styles used by different libraries.  Some
libraries use BNF definitions of a grammar as another mini-language inside
Python.  Exactly where those BNF definitions live varies, but using them is
largely similar (i.e. are they in a separate file, in docstrings, contents
of variables, etc).  And sure, EBNF vs. BNF proper.  But other libraries
instead use Python functions or classes to define the productions, where
each class/function is effectively one term of the grammar.  Typically this
latter style allows triggering events as soon as some production is
encountered—the event could be "accumulate a counter", or "write an output
string", or "perform a computation", or other things.

There are lots of good arguments for why to use different libraries along
the axes I mention, on all sides.  What is not possible is to reconcile the
very different decisions into a common denominator.  Something in the
standard library would have to be partisan in selecting one particular
approach as the "official" one.

Keeping medicines from the bloodstreams of the sick; food
from the bellies of the hungry; books from the hands of the
uneducated; technology from the underdeveloped; and putting
advocates of freedom in prisons.  Intellectual property is
to the 21st century what the slave trade was to the 16th.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-ideas/attachments/20190401/e99792b1/attachment-0001.html>

More information about the Python-ideas mailing list