##// END OF EJS Templates
httppeer: detect redirect to URL without query string (issue5860)...
httppeer: detect redirect to URL without query string (issue5860) 197d10e157ce subtly changed the HTTP peer's handling of HTTP redirects. Before that changeset, we instantiated an HTTP peer instance and performed the capabilities lookup with that instance. The old code had the following relevant properties: 1) The HTTP request layer would automatically follow HTTP redirects. 2) An encountered HTTP redirect would update a peer instance variable pointing to the repo URL. 3) The peer would automagically perform a "capabilities" command request if a caller requested capabilities but capabilities were not yet defined. The first HTTP request issued by a peer is for ?cmd=capabilities. If the server responds with an HTTP redirect to a ?cmd=capabilities URL, the HTTP request layer automatically followed it, retrieved a valid capabilities response, and the peer's base URL was updated automatically so subsequent requests used the proper URL. In other words, things "just worked." In the case where the server redirected to a URL without the ?cmd=capabilities query string, the HTTP request layer would follow the redirect and likely encounter HTML. The peer's base URL would be updated and the unexpected Content-Type would raise a RepoError. We would catch RepoError and immediately call between() (testing the case for pre 0.9.1 servers not supporting the "capabilities" command). e.g. try: inst._fetchcaps() except error.RepoError: inst.between([(nullid, nullid)]) between() would eventually call into _callstream(). And _callstream() made a call to self.capable('httpheader'). capable() would call self.capabilities(), which would see that no capabilities were set (because HTML was returned for that request) and call the "capabilities" command to fetch capabilities. Because the base URL had been updated from the redirect, this 2nd "capabilities" command would succeed and the client would immediately call "between," which would also succeed. The legacy handshake succeeded. Only because "capabilities" was successfully executed as a side effect did the peer recognize that it was talking to a modern server. In other words, this all appeared to work accidentally. After 197d10e157ce, we stopped calling the "capabilities" command on the peer instance. Instead, we made the request via a low-level opener, detected the redirect as part of response handling code, and passed the redirected URL into the constructed peer instance. For cases where the redirected URL included the query string, this "just worked." But for cases where the redirected URL stripped the query string, we threw RepoError and because we removed the "between" handshake fallback, we fell through to the "is a static HTTP repo" check and performed an HTTP request for .hg/requires. While 197d10e157ce was marked as backwards incompatible, the only intended backwards incompatible behavior was not performing the "between" fallback. It was not realized that the "between" command had the side-effect of recovering from an errant redirect that dropped the query string. This commit restores the previous behavior and allows clients to handle a redirect that drops the query string. In the case where the request is redirected and the query string is dropped, we raise a special case of RepoError. We then catch this special exception in the handshake code and perform another "capabilities" request against the redirected URL. If that works, all is well. Otherwise, we fall back to the "is a static HTTP repo" check. The new code is arguably better than before 197d10e157ce, as it is explicit about the expected behavior and we avoid performing a "between" request, saving a server round trip. Differential Revision: https://phab.mercurial-scm.org/D3433

File last commit:

r37513:b1fb341d default
r37851:6169d95d @24 stable
Show More
decompressionreader.c
459 lines | 10.9 KiB | text/x-c | CLexer
/**
* Copyright (c) 2017-present, Gregory Szorc
* All rights reserved.
*
* This software may be modified and distributed under the terms
* of the BSD license. See the LICENSE file for details.
*/
#include "python-zstandard.h"
extern PyObject* ZstdError;
static void set_unsupported_operation(void) {
PyObject* iomod;
PyObject* exc;
iomod = PyImport_ImportModule("io");
if (NULL == iomod) {
return;
}
exc = PyObject_GetAttrString(iomod, "UnsupportedOperation");
if (NULL == exc) {
Py_DECREF(iomod);
return;
}
PyErr_SetNone(exc);
Py_DECREF(exc);
Py_DECREF(iomod);
}
static void reader_dealloc(ZstdDecompressionReader* self) {
Py_XDECREF(self->decompressor);
Py_XDECREF(self->reader);
if (self->buffer.buf) {
PyBuffer_Release(&self->buffer);
}
PyObject_Del(self);
}
static ZstdDecompressionReader* reader_enter(ZstdDecompressionReader* self) {
if (self->entered) {
PyErr_SetString(PyExc_ValueError, "cannot __enter__ multiple times");
return NULL;
}
if (ensure_dctx(self->decompressor, 1)) {
return NULL;
}
self->entered = 1;
Py_INCREF(self);
return self;
}
static PyObject* reader_exit(ZstdDecompressionReader* self, PyObject* args) {
PyObject* exc_type;
PyObject* exc_value;
PyObject* exc_tb;
if (!PyArg_ParseTuple(args, "OOO:__exit__", &exc_type, &exc_value, &exc_tb)) {
return NULL;
}
self->entered = 0;
self->closed = 1;
/* Release resources. */
Py_CLEAR(self->reader);
if (self->buffer.buf) {
PyBuffer_Release(&self->buffer);
memset(&self->buffer, 0, sizeof(self->buffer));
}
Py_CLEAR(self->decompressor);
Py_RETURN_FALSE;
}
static PyObject* reader_readable(PyObject* self) {
Py_RETURN_TRUE;
}
static PyObject* reader_writable(PyObject* self) {
Py_RETURN_FALSE;
}
static PyObject* reader_seekable(PyObject* self) {
Py_RETURN_TRUE;
}
static PyObject* reader_close(ZstdDecompressionReader* self) {
self->closed = 1;
Py_RETURN_NONE;
}
static PyObject* reader_closed(ZstdDecompressionReader* self) {
if (self->closed) {
Py_RETURN_TRUE;
}
else {
Py_RETURN_FALSE;
}
}
static PyObject* reader_flush(PyObject* self) {
Py_RETURN_NONE;
}
static PyObject* reader_isatty(PyObject* self) {
Py_RETURN_FALSE;
}
static PyObject* reader_read(ZstdDecompressionReader* self, PyObject* args, PyObject* kwargs) {
static char* kwlist[] = {
"size",
NULL
};
Py_ssize_t size = -1;
PyObject* result = NULL;
char* resultBuffer;
Py_ssize_t resultSize;
ZSTD_outBuffer output;
size_t zresult;
if (!self->entered) {
PyErr_SetString(ZstdError, "read() must be called from an active context manager");
return NULL;
}
if (self->closed) {
PyErr_SetString(PyExc_ValueError, "stream is closed");
return NULL;
}
if (self->finishedOutput) {
return PyBytes_FromStringAndSize("", 0);
}
if (!PyArg_ParseTupleAndKeywords(args, kwargs, "n", kwlist, &size)) {
return NULL;
}
if (size < 1) {
PyErr_SetString(PyExc_ValueError, "cannot read negative or size 0 amounts");
return NULL;
}
result = PyBytes_FromStringAndSize(NULL, size);
if (NULL == result) {
return NULL;
}
PyBytes_AsStringAndSize(result, &resultBuffer, &resultSize);
output.dst = resultBuffer;
output.size = resultSize;
output.pos = 0;
readinput:
/* Consume input data left over from last time. */
if (self->input.pos < self->input.size) {
Py_BEGIN_ALLOW_THREADS
zresult = ZSTD_decompress_generic(self->decompressor->dctx,
&output, &self->input);
Py_END_ALLOW_THREADS
/* Input exhausted. Clear our state tracking. */
if (self->input.pos == self->input.size) {
memset(&self->input, 0, sizeof(self->input));
Py_CLEAR(self->readResult);
if (self->buffer.buf) {
self->finishedInput = 1;
}
}
if (ZSTD_isError(zresult)) {
PyErr_Format(ZstdError, "zstd decompress error: %s", ZSTD_getErrorName(zresult));
return NULL;
}
else if (0 == zresult) {
self->finishedOutput = 1;
}
/* We fulfilled the full read request. Emit it. */
if (output.pos && output.pos == output.size) {
self->bytesDecompressed += output.size;
return result;
}
/*
* There is more room in the output. Fall through to try to collect
* more data so we can try to fill the output.
*/
}
if (!self->finishedInput) {
if (self->reader) {
Py_buffer buffer;
assert(self->readResult == NULL);
self->readResult = PyObject_CallMethod(self->reader, "read",
"k", self->readSize);
if (NULL == self->readResult) {
return NULL;
}
memset(&buffer, 0, sizeof(buffer));
if (0 != PyObject_GetBuffer(self->readResult, &buffer, PyBUF_CONTIG_RO)) {
return NULL;
}
/* EOF */
if (0 == buffer.len) {
self->finishedInput = 1;
Py_CLEAR(self->readResult);
}
else {
self->input.src = buffer.buf;
self->input.size = buffer.len;
self->input.pos = 0;
}
PyBuffer_Release(&buffer);
}
else {
assert(self->buffer.buf);
/*
* We should only get here once since above block will exhaust
* source buffer until finishedInput is set.
*/
assert(self->input.src == NULL);
self->input.src = self->buffer.buf;
self->input.size = self->buffer.len;
self->input.pos = 0;
}
}
if (self->input.size) {
goto readinput;
}
/* EOF */
self->bytesDecompressed += output.pos;
if (safe_pybytes_resize(&result, output.pos)) {
Py_XDECREF(result);
return NULL;
}
return result;
}
static PyObject* reader_readall(PyObject* self) {
PyErr_SetNone(PyExc_NotImplementedError);
return NULL;
}
static PyObject* reader_readline(PyObject* self) {
PyErr_SetNone(PyExc_NotImplementedError);
return NULL;
}
static PyObject* reader_readlines(PyObject* self) {
PyErr_SetNone(PyExc_NotImplementedError);
return NULL;
}
static PyObject* reader_seek(ZstdDecompressionReader* self, PyObject* args) {
Py_ssize_t pos;
int whence = 0;
unsigned long long readAmount = 0;
size_t defaultOutSize = ZSTD_DStreamOutSize();
if (!self->entered) {
PyErr_SetString(ZstdError, "seek() must be called from an active context manager");
return NULL;
}
if (self->closed) {
PyErr_SetString(PyExc_ValueError, "stream is closed");
return NULL;
}
if (!PyArg_ParseTuple(args, "n|i:seek", &pos, &whence)) {
return NULL;
}
if (whence == SEEK_SET) {
if (pos < 0) {
PyErr_SetString(PyExc_ValueError,
"cannot seek to negative position with SEEK_SET");
return NULL;
}
if ((unsigned long long)pos < self->bytesDecompressed) {
PyErr_SetString(PyExc_ValueError,
"cannot seek zstd decompression stream backwards");
return NULL;
}
readAmount = pos - self->bytesDecompressed;
}
else if (whence == SEEK_CUR) {
if (pos < 0) {
PyErr_SetString(PyExc_ValueError,
"cannot seek zstd decompression stream backwards");
return NULL;
}
readAmount = pos;
}
else if (whence == SEEK_END) {
/* We /could/ support this with pos==0. But let's not do that until someone
needs it. */
PyErr_SetString(PyExc_ValueError,
"zstd decompression streams cannot be seeked with SEEK_END");
return NULL;
}
/* It is a bit inefficient to do this via the Python API. But since there
is a bit of state tracking involved to read from this type, it is the
easiest to implement. */
while (readAmount) {
Py_ssize_t readSize;
PyObject* readResult = PyObject_CallMethod((PyObject*)self, "read", "K",
readAmount < defaultOutSize ? readAmount : defaultOutSize);
if (!readResult) {
return NULL;
}
readSize = PyBytes_GET_SIZE(readResult);
/* Empty read means EOF. */
if (!readSize) {
break;
}
readAmount -= readSize;
}
return PyLong_FromUnsignedLongLong(self->bytesDecompressed);
}
static PyObject* reader_tell(ZstdDecompressionReader* self) {
/* TODO should this raise OSError since stream isn't seekable? */
return PyLong_FromUnsignedLongLong(self->bytesDecompressed);
}
static PyObject* reader_write(PyObject* self, PyObject* args) {
set_unsupported_operation();
return NULL;
}
static PyObject* reader_writelines(PyObject* self, PyObject* args) {
set_unsupported_operation();
return NULL;
}
static PyObject* reader_iter(PyObject* self) {
PyErr_SetNone(PyExc_NotImplementedError);
return NULL;
}
static PyObject* reader_iternext(PyObject* self) {
PyErr_SetNone(PyExc_NotImplementedError);
return NULL;
}
static PyMethodDef reader_methods[] = {
{ "__enter__", (PyCFunction)reader_enter, METH_NOARGS,
PyDoc_STR("Enter a compression context") },
{ "__exit__", (PyCFunction)reader_exit, METH_VARARGS,
PyDoc_STR("Exit a compression context") },
{ "close", (PyCFunction)reader_close, METH_NOARGS,
PyDoc_STR("Close the stream so it cannot perform any more operations") },
{ "closed", (PyCFunction)reader_closed, METH_NOARGS,
PyDoc_STR("Whether stream is closed") },
{ "flush", (PyCFunction)reader_flush, METH_NOARGS, PyDoc_STR("no-ops") },
{ "isatty", (PyCFunction)reader_isatty, METH_NOARGS, PyDoc_STR("Returns False") },
{ "readable", (PyCFunction)reader_readable, METH_NOARGS,
PyDoc_STR("Returns True") },
{ "read", (PyCFunction)reader_read, METH_VARARGS | METH_KEYWORDS,
PyDoc_STR("read compressed data") },
{ "readall", (PyCFunction)reader_readall, METH_NOARGS, PyDoc_STR("Not implemented") },
{ "readline", (PyCFunction)reader_readline, METH_NOARGS, PyDoc_STR("Not implemented") },
{ "readlines", (PyCFunction)reader_readlines, METH_NOARGS, PyDoc_STR("Not implemented") },
{ "seek", (PyCFunction)reader_seek, METH_VARARGS, PyDoc_STR("Seek the stream") },
{ "seekable", (PyCFunction)reader_seekable, METH_NOARGS,
PyDoc_STR("Returns True") },
{ "tell", (PyCFunction)reader_tell, METH_NOARGS,
PyDoc_STR("Returns current number of bytes compressed") },
{ "writable", (PyCFunction)reader_writable, METH_NOARGS,
PyDoc_STR("Returns False") },
{ "write", (PyCFunction)reader_write, METH_VARARGS, PyDoc_STR("unsupported operation") },
{ "writelines", (PyCFunction)reader_writelines, METH_VARARGS, PyDoc_STR("unsupported operation") },
{ NULL, NULL }
};
PyTypeObject ZstdDecompressionReaderType = {
PyVarObject_HEAD_INIT(NULL, 0)
"zstd.ZstdDecompressionReader", /* tp_name */
sizeof(ZstdDecompressionReader), /* tp_basicsize */
0, /* tp_itemsize */
(destructor)reader_dealloc, /* tp_dealloc */
0, /* tp_print */
0, /* tp_getattr */
0, /* tp_setattr */
0, /* tp_compare */
0, /* tp_repr */
0, /* tp_as_number */
0, /* tp_as_sequence */
0, /* tp_as_mapping */
0, /* tp_hash */
0, /* tp_call */
0, /* tp_str */
0, /* tp_getattro */
0, /* tp_setattro */
0, /* tp_as_buffer */
Py_TPFLAGS_DEFAULT, /* tp_flags */
0, /* tp_doc */
0, /* tp_traverse */
0, /* tp_clear */
0, /* tp_richcompare */
0, /* tp_weaklistoffset */
reader_iter, /* tp_iter */
reader_iternext, /* tp_iternext */
reader_methods, /* tp_methods */
0, /* tp_members */
0, /* tp_getset */
0, /* tp_base */
0, /* tp_dict */
0, /* tp_descr_get */
0, /* tp_descr_set */
0, /* tp_dictoffset */
0, /* tp_init */
0, /* tp_alloc */
PyType_GenericNew, /* tp_new */
};
void decompressionreader_module_init(PyObject* mod) {
/* TODO make reader a sub-class of io.RawIOBase */
Py_TYPE(&ZstdDecompressionReaderType) = &PyType_Type;
if (PyType_Ready(&ZstdDecompressionReaderType) < 0) {
return;
}
}