##// END OF EJS Templates
emitrevision: consider ancestors revision to emit as available base...
emitrevision: consider ancestors revision to emit as available base This should make more delta base valid. This notably affects: * case where we skipped some parent with empty delta to directly delta against an ancestors * case where an intermediate snapshots is stored. This change means we could sent largish intermediate snapshots over the wire. However this is actually a sub goal here. Sending snapshots over the wire means the client have a high odd of simply storing the pre-computed delta instead of doing a lengthy process that will… end up doing the same intermediate snapshot. In addition the overall size of snapshot (or any level) is "only" some or the overall delta size. (0.17% for my mercurial clone, 20% for my clone of Mozilla try). So Sending them other the wire is unlikely to change large impact on the bandwidth used. If we decide that minimising the bandwidth is an explicit goal, we should introduce new logic to filter-out snapshot as delta. The current code has no notion explicite of snapshot so far, they just tended to fall into the wobbly filtering options. In some cases, this patch can yield large improvement to the bundling time: ### data-env-vars.name = mozilla-try-2019-02-18-zstd-sparse-revlog # benchmark.name = perf-bundle # benchmark.variants.revs = last-100000 before: 68.787066 seconds after: 47.552677 seconds (-30.87%) That translate to large improvement to the pull time : ### data-env-vars.name = mozilla-try-2019-02-18-zstd-sparse-revlog # benchmark.name = pull # benchmark.variants.issue6528 = disabled # benchmark.variants.revs = last-100000 before: 142.186625 seconds after: 75.897745 seconds (-46.62%) No significant negative impact have been observed.

File last commit:

r49676:b0dd39b9 default
r50685:f463eb67 stable
Show More
mpatch.c
203 lines | 4.4 KiB | text/x-c | CLexer
/*
mpatch.c - efficient binary patching for Mercurial
This implements a patch algorithm that's O(m + nlog n) where m is the
size of the output and n is the number of patches.
Given a list of binary patches, it unpacks each into a hunk list,
then combines the hunk lists with a treewise recursion to form a
single hunk list. This hunk list is then applied to the original
text.
The text (or binary) fragments are copied directly from their source
Python objects into a preallocated output string to avoid the
allocation of intermediate Python objects. Working memory is about 2x
the total number of hunks.
Copyright 2005, 2006 Olivia Mackall <olivia@selenic.com>
This software may be used and distributed according to the terms
of the GNU General Public License, incorporated herein by reference.
*/
#define PY_SSIZE_T_CLEAN
#include <Python.h>
#include <stdlib.h>
#include <string.h>
#include "bitmanipulation.h"
#include "compat.h"
#include "mpatch.h"
#include "util.h"
static char mpatch_doc[] = "Efficient binary patching.";
static PyObject *mpatch_Error;
static void setpyerr(int r)
{
switch (r) {
case MPATCH_ERR_NO_MEM:
PyErr_NoMemory();
break;
case MPATCH_ERR_CANNOT_BE_DECODED:
PyErr_SetString(mpatch_Error, "patch cannot be decoded");
break;
case MPATCH_ERR_INVALID_PATCH:
PyErr_SetString(mpatch_Error, "invalid patch");
break;
}
}
struct mpatch_flist *cpygetitem(void *bins, ssize_t pos)
{
Py_buffer buffer;
struct mpatch_flist *res = NULL;
int r;
PyObject *tmp = PyList_GetItem((PyObject *)bins, pos);
if (!tmp) {
return NULL;
}
if (PyObject_GetBuffer(tmp, &buffer, PyBUF_CONTIG_RO)) {
return NULL;
}
if ((r = mpatch_decode(buffer.buf, buffer.len, &res)) < 0) {
if (!PyErr_Occurred()) {
setpyerr(r);
}
res = NULL;
}
PyBuffer_Release(&buffer);
return res;
}
static PyObject *patches(PyObject *self, PyObject *args)
{
PyObject *text, *bins, *result;
struct mpatch_flist *patch;
Py_buffer buffer;
int r = 0;
char *out;
Py_ssize_t len, outlen;
if (!PyArg_ParseTuple(args, "OO:mpatch", &text, &bins)) {
return NULL;
}
len = PyList_Size(bins);
if (!len) {
/* nothing to do */
Py_INCREF(text);
return text;
}
if (PyObject_GetBuffer(text, &buffer, PyBUF_CONTIG_RO)) {
return NULL;
}
patch = mpatch_fold(bins, cpygetitem, 0, len);
if (!patch) { /* error already set or memory error */
if (!PyErr_Occurred()) {
PyErr_NoMemory();
}
result = NULL;
goto cleanup;
}
outlen = mpatch_calcsize(buffer.len, patch);
if (outlen < 0) {
r = (int)outlen;
result = NULL;
goto cleanup;
}
result = PyBytes_FromStringAndSize(NULL, outlen);
if (!result) {
result = NULL;
goto cleanup;
}
out = PyBytes_AsString(result);
/* clang-format off */
{
Py_BEGIN_ALLOW_THREADS
r = mpatch_apply(out, buffer.buf, buffer.len, patch);
Py_END_ALLOW_THREADS
}
/* clang-format on */
if (r < 0) {
Py_DECREF(result);
result = NULL;
}
cleanup:
mpatch_lfree(patch);
PyBuffer_Release(&buffer);
if (!result && !PyErr_Occurred()) {
setpyerr(r);
}
return result;
}
/* calculate size of a patched file directly */
static PyObject *patchedsize(PyObject *self, PyObject *args)
{
long orig, start, end, len, outlen = 0, last = 0, pos = 0;
Py_ssize_t patchlen;
char *bin;
if (!PyArg_ParseTuple(args, "ly#", &orig, &bin, &patchlen)) {
return NULL;
}
while (pos >= 0 && pos < patchlen) {
start = getbe32(bin + pos);
end = getbe32(bin + pos + 4);
len = getbe32(bin + pos + 8);
if (start > end) {
break; /* sanity check */
}
pos += 12 + len;
outlen += start - last;
last = end;
outlen += len;
}
if (pos != patchlen) {
if (!PyErr_Occurred()) {
PyErr_SetString(mpatch_Error,
"patch cannot be decoded");
}
return NULL;
}
outlen += orig - last;
return Py_BuildValue("l", outlen);
}
static PyMethodDef methods[] = {
{"patches", patches, METH_VARARGS, "apply a series of patches\n"},
{"patchedsize", patchedsize, METH_VARARGS, "calculed patched size\n"},
{NULL, NULL},
};
static const int version = 1;
static struct PyModuleDef mpatch_module = {
PyModuleDef_HEAD_INIT, "mpatch", mpatch_doc, -1, methods,
};
PyMODINIT_FUNC PyInit_mpatch(void)
{
PyObject *m;
m = PyModule_Create(&mpatch_module);
if (m == NULL)
return NULL;
mpatch_Error =
PyErr_NewException("mercurial.cext.mpatch.mpatchError", NULL, NULL);
Py_INCREF(mpatch_Error);
PyModule_AddObject(m, "mpatchError", mpatch_Error);
PyModule_AddIntConstant(m, "version", version);
return m;
}