[PATCH 3 of 9] pure Python implementation of parsers.c

[Martin Geisler]

# HG changeset patch
# User Martin Geisler <mg at daimi.au.dk>
# Date 1232752338 -3600
# Node ID fbd0c5b0461663203c822c0fb65a65ab98dfb248
# Parent  505d542bda1836783f5c021d0dceb49442e749fd
pure Python implementation of parsers.c

diff -r 505d542bda18 -r fbd0c5b04616 mercurial/pure/parsers.py
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/mercurial/pure/parsers.py	Sat Jan 24 00:12:18 2009 +0100
@@ -0,0 +1,81 @@
+# parsers.py - Python implementation of parsers.c
+#
+# Copyright 2009 Matt Mackall <mpm at selenic.com> and others
+#
+# This software may be used and distributed according to the terms
+# of the GNU General Public License, incorporated herein by reference.
+
+from node import bin, hex, nullid, nullrev
+import revlog, dirstate, struct, util, zlib
+
+_pack = struct.pack
+_unpack = struct.unpack
+_compress = zlib.compress
+_decompress = zlib.decompress
+_sha = util.sha1
+
+def parse_manifest(mfdict, fdict, lines):
+    for l in lines.splitlines():
+        f, n = l.split('\0')
+        if len(n) > 40:
+            fdict[f] = n[40:]
+            mfdict[f] = bin(n[:40])
+        else:
+            mfdict[f] = bin(n)
+
+def parse_index(data, inline):
+    indexformatng = revlog.indexformatng
+    s = struct.calcsize(indexformatng)
+    index = []
+    cache = None
+    nodemap = {nullid: nullrev}
+    n = off = 0
+    # if we're not using lazymap, always read the whole index
+    l = len(data) - s
+    append = index.append
+    if inline:
+        cache = (0, data)
+        while off <= l:
+            e = _unpack(indexformatng, data[off:off + s])
+            nodemap[e[7]] = n
+            append(e)
+            n += 1
+            if e[1] < 0:
+                break
+            off += e[1] + s
+    else:
+        while off <= l:
+            e = _unpack(indexformatng, data[off:off + s])
+            nodemap[e[7]] = n
+            append(e)
+            n += 1
+            off += s
+
+    e = list(index[0])
+    type = revlog.gettype(e[0])
+    e[0] = revlog.offset_type(0, type)
+    index[0] = tuple(e)
+
+    # add the magic null revision at -1
+    index.append((0, 0, 0, -1, -1, -1, -1, nullid))
+
+    return index, nodemap, cache
+
+def parse_dirstate(dmap, copymap, st):
+    parents = [st[:20], st[20: 40]]
+    # deref fields so they will be local in loop
+    e_size = struct.calcsize(dirstate._format)
+    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