This is an archive of the discontinued Mercurial Phabricator instance.

Differential D2576

xdiff: add a python wrapper
AbandonedPublic

Authored by quark on Mar 2 2018, 7:25 PM.

Download Raw Diff

Details

Reviewers

ryanmce

Group Reviewers

hg-reviewers

Summary

Implement a mercurial.cext.xdiff module that exposes the xdiff algorithm.

xdiff.blocks should be a drop-in replacement for bdiff.blocks.

In theory we can change the pure C version of bdiff.c directly. However
that means we lose bdiff entirely. It seems more flexible to have both at
the same time so they can be easily switched via Python code. Hence the
Python module approach.

Test Plan

make local. And test from hg debugshell:

In [1]: from mercurial.cext import bdiff, xdiff

In [2]: bdiff.blocks('b\nc\nd\n', 'a\nb\nc\ne\nf\n')
Out[2]: [(0, 2, 1, 3), (3, 3, 5, 5)]

In [3]: xdiff.blocks('b\nc\nd\n', 'a\nb\nc\ne\nf\n')
Out[3]: [(0, 2, 1, 3), (3, 3, 5, 5)]

In [4]: bdiff.blocks('a\nb', '')
Out[4]: [(2, 2, 0, 0)]

In [5]: xdiff.blocks('a\nb', '')
Out[5]: [(2, 2, 0, 0)]

In [6]: bdiff.blocks('a\nb', 'a\nb\n')
Out[6]: [(0, 1, 0, 1), (2, 2, 2, 2)]

In [7]: xdiff.blocks('a\nb', 'a\nb\n')
Out[7]: [(0, 1, 0, 1), (2, 2, 2, 2)]

In [8]: xdiff.blocks('', '')
Out[8]: [(0, 0, 0, 0)]

In [9]: bdiff.blocks('', '')
Out[9]: [(0, 0, 0, 0)]

In [10]: xdiff.blocks('', '\n')
Out[10]: [(0, 0, 1, 1)]

Diff Detail

Repository

rHG Mercurial

Lint

Lint Skipped

Unit

Unit Tests Skipped

Event Timeline

ryanmce created this revision.Mar 2 2018, 7:25 PM

Herald added a reviewer: hg-reviewers. · View Herald TranscriptMar 2 2018, 7:25 PM

Herald added a subscriber: mercurial-devel. · View Herald Transcript

ryanmce added a child revision: D2577: mdiff: prefer xdiff for diff calculation.Mar 2 2018, 7:25 PM

Traditionally things under mercurial.cext has a pure equivalent. So it's cleaner to either make the changes directly in mercurial.cext.bdiff, or mercurial/bdiff.c. Or add a xdiff pure shim.

durin42 added a subscriber: durin42.Mar 3 2018, 9:49 AM

durin42 added inline comments.

setup.py
223	wat?

Not going to add another module.

quark removed a parent revision: D2575: xdiff: add a bdiff hunk mode.Mar 3 2018, 3:08 PM

quark abandoned this revision.

Revision Contents
Changeset List

		Path
M		contrib/check-code.py (3 lines)
A	M	mercurial/thirdparty/xdiff.c (96 lines)
M		setup.py (24 lines)

Status	Author	Revision
Abandoned	quark	D2578 xdiff: enable indent heuristic unconditionally
Abandoned	quark	D2577 mdiff: prefer xdiff for diff calculation
Abandoned	quark	D2576 xdiff: add a python wrapper

Diff 6419

contrib/check-code.py

	(r'\.has_key\b', "dict.has_key is not available in Python 3+"),			(r'\.has_key\b', "dict.has_key is not available in Python 3+"),
	(r'\s<>\s', '<> operator is not available in Python 3+, use !='),			(r'\s<>\s', '<> operator is not available in Python 3+, use !='),
	(r'^\s*\t', "don't use tabs"),			(r'^\s*\t', "don't use tabs"),
	(r'\S;\s*\n', "semicolon"),			(r'\S;\s*\n', "semicolon"),
	(r'[^_]_\([ \t\n](?:"[^"]+"[ \t\n+])+%', "don't use % inside _()"),			(r'[^_]_\([ \t\n](?:"[^"]+"[ \t\n+])+%', "don't use % inside _()"),
	(r"[^_]_\([ \t\n](?:'[^']+'[ \t\n+])+%", "don't use % inside _()"),			(r"[^_]_\([ \t\n](?:'[^']+'[ \t\n+])+%", "don't use % inside _()"),
	(r'(\w\|\)),\w', "missing whitespace after ,"),			(r'(\w\|\)),\w', "missing whitespace after ,"),
	(r'(\w\|\))[+/*\-<>]\w', "missing whitespace in expression"),			(r'(\w\|\))[+/*\-<>]\w', "missing whitespace in expression"),
	(r'^\s+(\w\|\.)+=\w[^,()\n]*$', "missing whitespace in assignment"),			# false positive on multi-line keyword arguments
				# (r'^\s+(\w\|\.)+=\w[^,()\n]*$', "missing whitespace in assignment"),
	(r'\w\s=\s\s+\w', "gratuitous whitespace after ="),			(r'\w\s=\s\s+\w', "gratuitous whitespace after ="),
	((			((
	# a line ending with a colon, potentially with trailing comments			# a line ending with a colon, potentially with trailing comments
	r':([ \t]#[^\n])?\n'			r':([ \t]#[^\n])?\n'
	# one that is not a pass and not only a comment			# one that is not a pass and not only a comment
	r'(?P<indent>[ \t]+)[^#][^\n]+\n'			r'(?P<indent>[ \t]+)[^#][^\n]+\n'
	# more lines at the same indent level			# more lines at the same indent level
	r'((?P=indent)[^\n]+\n)*'			r'((?P=indent)[^\n]+\n)*'

mercurial/thirdparty/xdiff.c

This file was added.

				/*
				xdiff.c: simple Python wrapper for xdiff library

				Copyright (c) 2018 Facebook, Inc.

				This software may be used and distributed according to the terms of the
				GNU General Public License version 2 or any later version.
				*/

				#include "xdiff.h"
				#include "Python.h"

				static int hunk_consumer(long a1, long a2, long b1, long b2, void *priv)
				{
				PyObject rl = (PyObject )priv;
				PyObject *m = Py_BuildValue("llll", a1, a2, b1, b2);
				if (!m)
				return -1;
				if (PyList_Append(rl, m) != 0) {
				Py_DECREF(m);
				return -1;
				}
				return 0;
				}

				static PyObject blocks(PyObject self, PyObject *args)
				{
				char sa, sb;
				int na, nb;

				if (!PyArg_ParseTuple(args, "s#s#", &sa, &na, &sb, &nb))
				return NULL;

				mmfile_t a = {sa, na}, b = {sb, nb};

				PyObject *rl = PyList_New(0);
				if (!rl)
				return PyErr_NoMemory();

				xpparam_t xpp = {
				0, /* flags */
				NULL, /* anchors */
				0, /* anchors_nr */
				};
				xdemitconf_t xecfg = {
				0, /* ctxlen */
				0, /* interhunkctxlen */
				XDL_EMIT_BDIFFHUNK, /* flags */
				NULL, /* find_func */
				NULL, /* find_func_priv */
				hunk_consumer, /* hunk_consume_func */
				};
				xdemitcb_t ecb = {
				rl, /* priv */
				NULL, /* outf */
				};

				if (xdl_diff(&a, &b, &xpp, &xecfg, &ecb) != 0) {
				Py_DECREF(rl);
				return PyErr_NoMemory();
				}

				return rl;
				}

				static char xdiff_doc[] = "xdiff wrapper";

				static PyMethodDef methods[] = {
				{"blocks", blocks, METH_VARARGS,
				"(a: str, b: str) -> List[(a1, a2, b1, b2)].\n"
				"Yield matched blocks. (a1, a2, b1, b2) are line numbers.\n"},
				{NULL, NULL},
				};

				static const int version = 1;

				#ifdef IS_PY3K
				static struct PyModuleDef bdiff_module = {
				PyModuleDef_HEAD_INIT, "bdiff", xdiff_doc, -1, methods,
				};

				PyMODINIT_FUNC PyInit_xdiff(void)
				{
				PyObject *m;
				m = PyModule_Create(&xdiff_module);
				PyModule_AddIntConstant(m, "version", version);
				return m;
				}
				#else
				PyMODINIT_FUNC initxdiff(void)
				{
				PyObject *m;
				m = Py_InitModule3("xdiff", methods, xdiff_doc);
				PyModule_AddIntConstant(m, "version", version);
				}
				#endif

setup.py

	return cancompile(cc, code)			return cancompile(cc, code)

	# py2exe needs to be installed to work			# py2exe needs to be installed to work
	try:			try:
	import py2exe			import py2exe
	py2exe.Distribution # silence unused import warning			py2exe.Distribution # silence unused import warning
	py2exeloaded = True			py2exeloaded = True
	# import py2exe's patched Distribution class			# import py2exe's patched Distribution class
	from distutils.core import Distribution			from distutils.core import Distribution # noqa: F811
				durin42Unsubmitted Not Done wat? durin42: wat?
	except ImportError:			except ImportError:
	py2exeloaded = False			py2exeloaded = False

	def runcmd(cmd, env):			def runcmd(cmd, env):
	p = subprocess.Popen(cmd, stdout=subprocess.PIPE,			p = subprocess.Popen(cmd, stdout=subprocess.PIPE,
	stderr=subprocess.PIPE, env=env)			stderr=subprocess.PIPE, env=env)
	out, err = p.communicate()			out, err = p.communicate()
	return p.returncode, out, err			return p.returncode, out, err
	'mercurial/cext/revlog.c'],			'mercurial/cext/revlog.c'],
	include_dirs=common_include_dirs,			include_dirs=common_include_dirs,
	depends=common_depends + ['mercurial/cext/charencode.h']),			depends=common_depends + ['mercurial/cext/charencode.h']),
	Extension('mercurial.cext.osutil', ['mercurial/cext/osutil.c'],			Extension('mercurial.cext.osutil', ['mercurial/cext/osutil.c'],
	include_dirs=common_include_dirs,			include_dirs=common_include_dirs,
	extra_compile_args=osutil_cflags,			extra_compile_args=osutil_cflags,
	extra_link_args=osutil_ldflags,			extra_link_args=osutil_ldflags,
	depends=common_depends),			depends=common_depends),
				Extension('mercurial.cext.xdiff',
				sources=[
				'mercurial/thirdparty/xdiff/xdiffi.c',
				'mercurial/thirdparty/xdiff/xemit.c',
				'mercurial/thirdparty/xdiff/xmerge.c',
				'mercurial/thirdparty/xdiff/xprepare.c',
				'mercurial/thirdparty/xdiff/xutils.c',
				'mercurial/thirdparty/xdiff.c',
				],
				include_dirs=common_include_dirs + [
				'mercurial/thirdparty/xdiff',
				],
				depends=common_depends + [
				'mercurial/thirdparty/xdiff/xdiff.h',
				'mercurial/thirdparty/xdiff/xdiffi.h',
				'mercurial/thirdparty/xdiff/xemit.h',
				'mercurial/thirdparty/xdiff/xinclude.h',
				'mercurial/thirdparty/xdiff/xmacros.h',
				'mercurial/thirdparty/xdiff/xprepare.h',
				'mercurial/thirdparty/xdiff/xtypes.h',
				'mercurial/thirdparty/xdiff/xutils.h',
				]),
	Extension('hgext.fsmonitor.pywatchman.bser',			Extension('hgext.fsmonitor.pywatchman.bser',
	['hgext/fsmonitor/pywatchman/bser.c']),			['hgext/fsmonitor/pywatchman/bser.c']),
	]			]

	sys.path.insert(0, 'contrib/python-zstandard')			sys.path.insert(0, 'contrib/python-zstandard')
	import setup_zstd			import setup_zstd
	extmodules.append(setup_zstd.get_c_extension(name='mercurial.zstd'))			extmodules.append(setup_zstd.get_c_extension(name='mercurial.zstd'))

Diff	ID	Base	Description	Created	Lint	Unit
Base			Base
Diff 1	6419			Mar 2 2018, 7:25 PM	★	★