##// END OF EJS Templates
repository: formalize interfaces for revision deltas and requests...
repository: formalize interfaces for revision deltas and requests Now that we've sufficiently abstracted how revision deltas are produced in changegroup code, we can now start the process of formalizing that as part of the interfaces defined in the repository module. This commit essentially converts the revisiondelta and revisiondeltarequest classes into well-defined interfaces. This is not strictly necessary. But I want all types formalized by the storage interface to have interfaces. This makes it much easier to test for interface conformance and for implementing new storage backends. Because the interface is documented, comments and docstrings from changegroup.py have been dropped. Differential Revision: https://phab.mercurial-scm.org/D4225

File last commit:

r39251:5961517f default
r39266:b518d495 default
Show More
parsers.py
177 lines | 5.4 KiB | text/x-python | PythonLexer
Martin Geisler
pure Python implementation of parsers.c
r7700 # parsers.py - Python implementation of parsers.c
#
# Copyright 2009 Matt Mackall <mpm@selenic.com> and others
#
Martin Geisler
updated license to be explicit about GPL version 2
r8225 # This software may be used and distributed according to the terms of the
Matt Mackall
Update license to GPLv2+
r10263 # GNU General Public License version 2 or any later version.
Martin Geisler
pure Python implementation of parsers.c
r7700
Gregory Szorc
parsers: use absolute_import
r27339 from __future__ import absolute_import
import struct
import zlib
Yuya Nishihara
parsers: switch to policy importer...
r32372 from ..node import nullid
from .. import pycompat
Gregory Szorc
util: prefer "bytesio" to "stringio"...
r36976 stringio = pycompat.bytesio
Martin Geisler
pure Python implementation of parsers.c
r7700
Pulkit Goyal
parsers: alias long to int on Python 3
r31220
Martin Geisler
pure Python implementation of parsers.c
r7700 _pack = struct.pack
_unpack = struct.unpack
_compress = zlib.compress
_decompress = zlib.decompress
Siddharth Agarwal
parsers: inline fields of dirstate values in C version...
r21809 # Some code below makes tuples directly because it's more convenient. However,
# code outside this module should always use dirstatetuple.
def dirstatetuple(*x):
# x is a tuple
return x
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 indexformatng = ">Qiiiiii20s12x"
indexfirst = struct.calcsize('Q')
sizeint = struct.calcsize('i')
indexsize = struct.calcsize(indexformatng)
def gettype(q):
return int(q & 0xFFFF)
Matt Mackall
pure/parsers: fix circular imports, import mercurial modules properly
r7945
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 def offset_type(offset, type):
Martin von Zweigbergk
pure: use int instead of long...
r31529 return int(int(offset) << 16 | type)
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133
class BaseIndexObject(object):
def __len__(self):
Martin von Zweigbergk
index: don't include nullid in len()...
r38887 return self._lgt + len(self._extra)
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133
Martin von Zweigbergk
index: replace insert(-1, e) method by append(e) method...
r38886 def append(self, tup):
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 self._extra.append(tup)
Matt Mackall
pure/parsers: fix circular imports, import mercurial modules properly
r7945
Martin von Zweigbergk
index: rename _fix_index() since it no longer fixes the index...
r39251 def _check_index(self, i):
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 if not isinstance(i, int):
raise TypeError("expecting int indexes")
Martin von Zweigbergk
index: don't include nullid in boundary check in pure code...
r39250 if i < 0 or i >= len(self):
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 raise IndexError
Matt Mackall
pure/parsers: fix circular imports, import mercurial modules properly
r7945
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 def __getitem__(self, i):
Augie Fackler
parsers: adjust pure-python version to mimic a3dacabd476b...
r39082 if i == -1:
Martin von Zweigbergk
index: handle index[-1] as nullid more explicitly...
r38883 return (0, 0, 0, -1, -1, -1, -1, nullid)
Martin von Zweigbergk
index: rename _fix_index() since it no longer fixes the index...
r39251 self._check_index(i)
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 if i >= self._lgt:
return self._extra[i - self._lgt]
index = self._calculate_index(i)
r = struct.unpack(indexformatng, self._data[index:index + indexsize])
if i == 0:
e = list(r)
type = gettype(e[0])
e[0] = offset_type(0, type)
return tuple(e)
return r
class IndexObject(BaseIndexObject):
def __init__(self, data):
assert len(data) % indexsize == 0
self._data = data
self._lgt = len(data) // indexsize
self._extra = []
def _calculate_index(self, i):
return i * indexsize
Matt Mackall
revlog: remove lazy index
r13253
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 def __delitem__(self, i):
Alex Gaynor
style: always use `x is not None` instead of `not x is None`...
r34332 if not isinstance(i, slice) or not i.stop == -1 or i.step is not None:
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 raise ValueError("deleting slices only supports a:-1 with step 1")
Martin von Zweigbergk
index: rename _fix_index() since it no longer fixes the index...
r39251 i = i.start
self._check_index(i)
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 if i < self._lgt:
self._data = self._data[:i * indexsize]
self._lgt = i
self._extra = []
else:
self._extra = self._extra[:i - self._lgt]
class InlinedIndexObject(BaseIndexObject):
def __init__(self, data, inline=0):
self._data = data
self._lgt = self._inline_scan(None)
self._inline_scan(self._lgt)
self._extra = []
Martin Geisler
pure Python implementation of parsers.c
r7700
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 def _inline_scan(self, lgt):
off = 0
if lgt is not None:
self._offsets = [0] * lgt
count = 0
while off <= len(self._data) - indexsize:
s, = struct.unpack('>i',
self._data[off + indexfirst:off + sizeint + indexfirst])
if lgt is not None:
self._offsets[count] = off
count += 1
off += indexsize + s
if off != len(self._data):
raise ValueError("corrupted data")
return count
Augie Fackler
pure parsers: properly detect corrupt index files...
r14421
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 def __delitem__(self, i):
Alex Gaynor
style: always use `x is not None` instead of `not x is None`...
r34332 if not isinstance(i, slice) or not i.stop == -1 or i.step is not None:
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 raise ValueError("deleting slices only supports a:-1 with step 1")
Martin von Zweigbergk
index: rename _fix_index() since it no longer fixes the index...
r39251 i = i.start
self._check_index(i)
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 if i < self._lgt:
self._offsets = self._offsets[:i]
self._lgt = i
self._extra = []
else:
self._extra = self._extra[:i - self._lgt]
Martin Geisler
pure Python implementation of parsers.c
r7700
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 def _calculate_index(self, i):
return self._offsets[i]
Martin Geisler
pure Python implementation of parsers.c
r7700
Maciej Fijalkowski
pure: write a really lazy version of pure indexObject...
r29133 def parse_index2(data, inline):
if not inline:
return IndexObject(data), None
return InlinedIndexObject(data, inline), (0, data)
Martin Geisler
pure Python implementation of parsers.c
r7700
def parse_dirstate(dmap, copymap, st):
parents = [st[:20], st[20: 40]]
Mads Kiilerich
fix wording and not-completely-trivial spelling errors and bad docstrings
r17425 # dereference fields so they will be local in loop
Matt Mackall
pure/parsers: fix circular imports, import mercurial modules properly
r7945 format = ">cllll"
e_size = struct.calcsize(format)
Martin Geisler
pure Python implementation of parsers.c
r7700 pos1 = 40
l = len(st)
# the inner loop
while pos1 < l:
pos2 = pos1 + e_size
e = _unpack(">cllll", st[pos1:pos2]) # a literal here is faster
pos1 = pos2 + e[4]
f = st[pos2:pos1]
if '\0' in f:
f, c = f.split('\0')
copymap[f] = c
dmap[f] = e[:4]
return parents
Siddharth Agarwal
dirstate: move pure python dirstate packing to pure/parsers.py
r18567
def pack_dirstate(dmap, copymap, pl, now):
now = int(now)
timeless
pycompat: switch to util.stringio for py3 compat
r28861 cs = stringio()
Siddharth Agarwal
dirstate: move pure python dirstate packing to pure/parsers.py
r18567 write = cs.write
write("".join(pl))
for f, e in dmap.iteritems():
if e[0] == 'n' and e[3] == now:
# The file was last modified "simultaneously" with the current
# write to dirstate (i.e. within the same second for file-
# systems with a granularity of 1 sec). This commonly happens
# for at least a couple of files on 'update'.
# The user could change the file without changing its size
Siddharth Agarwal
pack_dirstate: only invalidate mtime for files written in the last second...
r19652 # within the same second. Invalidate the file's mtime in
Siddharth Agarwal
dirstate: move pure python dirstate packing to pure/parsers.py
r18567 # dirstate, forcing future 'status' calls to compare the
Siddharth Agarwal
pack_dirstate: only invalidate mtime for files written in the last second...
r19652 # contents of the file if the size is the same. This prevents
# mistakenly treating such files as clean.
Siddharth Agarwal
parsers: inline fields of dirstate values in C version...
r21809 e = dirstatetuple(e[0], e[1], e[2], -1)
Siddharth Agarwal
dirstate: move pure python dirstate packing to pure/parsers.py
r18567 dmap[f] = e
if f in copymap:
f = "%s\0%s" % (f, copymap[f])
e = _pack(">cllll", e[0], e[1], e[2], e[3], len(f))
write(e)
write(f)
return cs.getvalue()