##// END OF EJS Templates
protocol: send application/mercurial-0.2 responses to capable clients...
protocol: send application/mercurial-0.2 responses to capable clients With this commit, the HTTP transport now parses the X-HgProto-<N> header to determine what media type and compression engine to use for responses. So far, we only compress responses that are already being compressed with zlib today (stream response types to specific commands). We can expand things to cover additional response types later. The practical side-effect of this commit is that non-zlib compression engines will be used if both ends support them. This means if both ends have zstd support, zstd - not zlib - will be used to compress data! When cloning the mozilla-unified repository between a local HTTP server and client, the benefits of non-zlib compression are quite noticeable: engine server CPU (s) client CPU (s) bundle size zlib (l=6) 174.1 283.2 1,148,547,026 zstd (l=1) 99.2 267.3 1,127,513,841 zstd (l=3) 103.1 266.9 1,018,861,363 zstd (l=7) 128.3 269.7 919,190,278 zstd (l=10) 162.0 - 894,547,179 none 95.3 277.2 4,097,566,064 The default zstd compression level is 3. So if you deploy zstd capable Mercurial to your clients and servers and CPU time on your server is dominated by "getbundle" requests (clients cloning and pulling) - and my experience at Mozilla tells me this is often the case - this commit could drastically reduce your server-side CPU usage *and* save on bandwidth costs! Another benefit of this change is that server operators can install *any* compression engine. While it isn't enabled by default, the "none" compression engine can now be used to disable wire protocol compression completely. Previously, commands like "getbundle" always zlib compressed output, adding considerable overhead to generating responses. If you are on a high speed network and your server is under high load, it might be advantageous to trade bandwidth for CPU. Although, zstd at level 1 doesn't use that much CPU, so I'm not convinced that disabling compression wholesale is worthwhile. And, my data seems to indicate a slow down on the client without compression. I suspect this is due to a lack of buffering resulting in an increase in socket read() calls and/or the fact we're transferring an extra 3 GB of data (parsing HTTP chunked transfer and processing extra TCP packets can add up). This is definitely worth investigating and optimizing. But since the "none" compressor isn't enabled by default, I'm inclined to punt on this issue. This commit introduces tons of tests. Some of these should arguably have been implemented on previous commits. But it was difficult to test without the server functionality in place.

File last commit:

r27060:4613a89b default
r30764:e75463e3 default
Show More
base85.c
182 lines | 3.4 KiB | text/x-c | CLexer
/*
base85 codec
Copyright 2006 Brendan Cully <brendan@kublai.com>
This software may be used and distributed according to the terms of
the GNU General Public License, incorporated herein by reference.
Largely based on git's implementation
*/
#define PY_SSIZE_T_CLEAN
#include <Python.h>
#include "util.h"
static const char b85chars[] = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ"
"abcdefghijklmnopqrstuvwxyz!#$%&()*+-;<=>?@^_`{|}~";
static char b85dec[256];
static void b85prep(void)
{
unsigned i;
memset(b85dec, 0, sizeof(b85dec));
for (i = 0; i < sizeof(b85chars); i++)
b85dec[(int)(b85chars[i])] = i + 1;
}
static PyObject *b85encode(PyObject *self, PyObject *args)
{
const unsigned char *text;
PyObject *out;
char *dst;
Py_ssize_t len, olen, i;
unsigned int acc, val, ch;
int pad = 0;
if (!PyArg_ParseTuple(args, "s#|i", &text, &len, &pad))
return NULL;
if (pad)
olen = ((len + 3) / 4 * 5) - 3;
else {
olen = len % 4;
if (olen)
olen++;
olen += len / 4 * 5;
}
if (!(out = PyBytes_FromStringAndSize(NULL, olen + 3)))
return NULL;
dst = PyBytes_AsString(out);
while (len) {
acc = 0;
for (i = 24; i >= 0; i -= 8) {
ch = *text++;
acc |= ch << i;
if (--len == 0)
break;
}
for (i = 4; i >= 0; i--) {
val = acc % 85;
acc /= 85;
dst[i] = b85chars[val];
}
dst += 5;
}
if (!pad)
_PyBytes_Resize(&out, olen);
return out;
}
static PyObject *b85decode(PyObject *self, PyObject *args)
{
PyObject *out;
const char *text;
char *dst;
Py_ssize_t len, i, j, olen, cap;
int c;
unsigned int acc;
if (!PyArg_ParseTuple(args, "s#", &text, &len))
return NULL;
olen = len / 5 * 4;
i = len % 5;
if (i)
olen += i - 1;
if (!(out = PyBytes_FromStringAndSize(NULL, olen)))
return NULL;
dst = PyBytes_AsString(out);
i = 0;
while (i < len)
{
acc = 0;
cap = len - i - 1;
if (cap > 4)
cap = 4;
for (j = 0; j < cap; i++, j++)
{
c = b85dec[(int)*text++] - 1;
if (c < 0)
return PyErr_Format(
PyExc_ValueError,
"bad base85 character at position %d",
(int)i);
acc = acc * 85 + c;
}
if (i++ < len)
{
c = b85dec[(int)*text++] - 1;
if (c < 0)
return PyErr_Format(
PyExc_ValueError,
"bad base85 character at position %d",
(int)i);
/* overflow detection: 0xffffffff == "|NsC0",
* "|NsC" == 0x03030303 */
if (acc > 0x03030303 || (acc *= 85) > 0xffffffff - c)
return PyErr_Format(
PyExc_ValueError,
"bad base85 sequence at position %d",
(int)i);
acc += c;
}
cap = olen < 4 ? olen : 4;
olen -= cap;
for (j = 0; j < 4 - cap; j++)
acc *= 85;
if (cap && cap < 4)
acc += 0xffffff >> (cap - 1) * 8;
for (j = 0; j < cap; j++)
{
acc = (acc << 8) | (acc >> 24);
*dst++ = acc;
}
}
return out;
}
static char base85_doc[] = "Base85 Data Encoding";
static PyMethodDef methods[] = {
{"b85encode", b85encode, METH_VARARGS,
"Encode text in base85.\n\n"
"If the second parameter is true, pad the result to a multiple of "
"five characters.\n"},
{"b85decode", b85decode, METH_VARARGS, "Decode base85 text.\n"},
{NULL, NULL}
};
#ifdef IS_PY3K
static struct PyModuleDef base85_module = {
PyModuleDef_HEAD_INIT,
"base85",
base85_doc,
-1,
methods
};
PyMODINIT_FUNC PyInit_base85(void)
{
b85prep();
return PyModule_Create(&base85_module);
}
#else
PyMODINIT_FUNC initbase85(void)
{
Py_InitModule3("base85", methods, base85_doc);
b85prep();
}
#endif