Kevacoin source tree
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

621 lines
20 KiB

// Copyright (c) 2009-2010 Satoshi Nakamoto
// Copyright (c) 2009-2014 The Bitcoin Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include "chain.h"
#include "primitives/block.h"
#include "primitives/transaction.h"
#include "main.h"
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
#include "httpserver.h"
#include "rpcserver.h"
#include "streams.h"
#include "sync.h"
#include "txmempool.h"
#include "utilstrencodings.h"
#include "version.h"
#include <boost/algorithm/string.hpp>
#include <boost/dynamic_bitset.hpp>
#include "univalue/univalue.h"
using namespace std;
static const size_t MAX_GETUTXOS_OUTPOINTS = 15; //allow a max of 15 outpoints to be queried at once
enum RetFormat {
RF_UNDEF,
RF_BINARY,
RF_HEX,
RF_JSON,
};
static const struct {
enum RetFormat rf;
const char* name;
} rf_names[] = {
{RF_UNDEF, ""},
{RF_BINARY, "bin"},
{RF_HEX, "hex"},
{RF_JSON, "json"},
};
struct CCoin {
uint32_t nTxVer; // Don't call this nVersion, that name has a special meaning inside IMPLEMENT_SERIALIZE
uint32_t nHeight;
CTxOut out;
ADD_SERIALIZE_METHODS;
template <typename Stream, typename Operation>
inline void SerializationOp(Stream& s, Operation ser_action, int nType, int nVersion)
{
READWRITE(nTxVer);
READWRITE(nHeight);
READWRITE(out);
}
};
extern void TxToJSON(const CTransaction& tx, const uint256 hashBlock, UniValue& entry);
extern UniValue blockToJSON(const CBlock& block, const CBlockIndex* blockindex, bool txDetails = false);
extern UniValue mempoolInfoToJSON();
extern UniValue mempoolToJSON(bool fVerbose = false);
extern void ScriptPubKeyToJSON(const CScript& scriptPubKey, UniValue& out, bool fIncludeHex);
extern UniValue blockheaderToJSON(const CBlockIndex* blockindex);
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
static bool RESTERR(HTTPRequest* req, enum HTTPStatusCode status, string message)
{
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "text/plain");
req->WriteReply(status, message + "\r\n");
return false;
}
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
static enum RetFormat ParseDataFormat(vector<string>& params, const string& strReq)
{
boost::split(params, strReq, boost::is_any_of("."));
if (params.size() > 1) {
for (unsigned int i = 0; i < ARRAYLEN(rf_names); i++)
if (params[1] == rf_names[i].name)
return rf_names[i].rf;
}
return rf_names[0].rf;
}
static string AvailableDataFormatsString()
{
string formats = "";
for (unsigned int i = 0; i < ARRAYLEN(rf_names); i++)
if (strlen(rf_names[i].name) > 0) {
formats.append(".");
formats.append(rf_names[i].name);
formats.append(", ");
}
if (formats.length() > 0)
return formats.substr(0, formats.length() - 2);
return formats;
}
static bool ParseHashStr(const string& strReq, uint256& v)
{
if (!IsHex(strReq) || (strReq.size() != 64))
return false;
v.SetHex(strReq);
return true;
}
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
static bool CheckWarmup(HTTPRequest* req)
{
std::string statusmessage;
if (RPCIsInWarmup(&statusmessage))
return RESTERR(req, HTTP_SERVICE_UNAVAILABLE, "Service temporarily unavailable: " + statusmessage);
return true;
}
static bool rest_headers(HTTPRequest* req,
const std::string& strURIPart)
10 years ago
{
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
if (!CheckWarmup(req))
return false;
10 years ago
vector<string> params;
const RetFormat rf = ParseDataFormat(params, strURIPart);
10 years ago
vector<string> path;
boost::split(path, params[0], boost::is_any_of("/"));
if (path.size() != 2)
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_BAD_REQUEST, "No header count specified. Use /rest/headers/<count>/<hash>.<ext>.");
10 years ago
long count = strtol(path[0].c_str(), NULL, 10);
if (count < 1 || count > 2000)
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_BAD_REQUEST, "Header count out of range: " + path[0]);
10 years ago
string hashStr = path[1];
uint256 hash;
if (!ParseHashStr(hashStr, hash))
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_BAD_REQUEST, "Invalid hash: " + hashStr);
10 years ago
std::vector<const CBlockIndex *> headers;
10 years ago
headers.reserve(count);
{
LOCK(cs_main);
BlockMap::const_iterator it = mapBlockIndex.find(hash);
const CBlockIndex *pindex = (it != mapBlockIndex.end()) ? it->second : NULL;
while (pindex != NULL && chainActive.Contains(pindex)) {
headers.push_back(pindex);
10 years ago
if (headers.size() == (unsigned long)count)
break;
pindex = chainActive.Next(pindex);
}
}
CDataStream ssHeader(SER_NETWORK, PROTOCOL_VERSION);
BOOST_FOREACH(const CBlockIndex *pindex, headers) {
ssHeader << pindex->GetBlockHeader();
10 years ago
}
switch (rf) {
case RF_BINARY: {
string binaryHeader = ssHeader.str();
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "application/octet-stream");
req->WriteReply(HTTP_OK, binaryHeader);
10 years ago
return true;
}
case RF_HEX: {
string strHex = HexStr(ssHeader.begin(), ssHeader.end()) + "\n";
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "text/plain");
req->WriteReply(HTTP_OK, strHex);
10 years ago
return true;
}
case RF_JSON: {
UniValue jsonHeaders(UniValue::VARR);
BOOST_FOREACH(const CBlockIndex *pindex, headers) {
jsonHeaders.push_back(blockheaderToJSON(pindex));
}
string strJSON = jsonHeaders.write() + "\n";
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "application/json");
req->WriteReply(HTTP_OK, strJSON);
return true;
}
10 years ago
default: {
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_NOT_FOUND, "output format not found (available: .bin, .hex)");
10 years ago
}
}
// not reached
return true; // continue to process further HTTP reqs on this cxn
}
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
static bool rest_block(HTTPRequest* req,
const std::string& strURIPart,
bool showTxDetails)
{
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
if (!CheckWarmup(req))
return false;
vector<string> params;
const RetFormat rf = ParseDataFormat(params, strURIPart);
string hashStr = params[0];
uint256 hash;
if (!ParseHashStr(hashStr, hash))
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_BAD_REQUEST, "Invalid hash: " + hashStr);
CBlock block;
CBlockIndex* pblockindex = NULL;
{
LOCK(cs_main);
if (mapBlockIndex.count(hash) == 0)
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_NOT_FOUND, hashStr + " not found");
pblockindex = mapBlockIndex[hash];
if (fHavePruned && !(pblockindex->nStatus & BLOCK_HAVE_DATA) && pblockindex->nTx > 0)
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_NOT_FOUND, hashStr + " not available (pruned data)");
if (!ReadBlockFromDisk(block, pblockindex))
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_NOT_FOUND, hashStr + " not found");
}
CDataStream ssBlock(SER_NETWORK, PROTOCOL_VERSION);
ssBlock << block;
switch (rf) {
case RF_BINARY: {
string binaryBlock = ssBlock.str();
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "application/octet-stream");
req->WriteReply(HTTP_OK, binaryBlock);
return true;
}
case RF_HEX: {
string strHex = HexStr(ssBlock.begin(), ssBlock.end()) + "\n";
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "text/plain");
req->WriteReply(HTTP_OK, strHex);
return true;
}
case RF_JSON: {
UniValue objBlock = blockToJSON(block, pblockindex, showTxDetails);
string strJSON = objBlock.write() + "\n";
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "application/json");
req->WriteReply(HTTP_OK, strJSON);
return true;
}
default: {
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_NOT_FOUND, "output format not found (available: " + AvailableDataFormatsString() + ")");
}
}
// not reached
return true; // continue to process further HTTP reqs on this cxn
}
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
static bool rest_block_extended(HTTPRequest* req, const std::string& strURIPart)
{
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return rest_block(req, strURIPart, true);
}
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
static bool rest_block_notxdetails(HTTPRequest* req, const std::string& strURIPart)
{
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return rest_block(req, strURIPart, false);
}
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
static bool rest_chaininfo(HTTPRequest* req, const std::string& strURIPart)
{
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
if (!CheckWarmup(req))
return false;
vector<string> params;
const RetFormat rf = ParseDataFormat(params, strURIPart);
switch (rf) {
case RF_JSON: {
UniValue rpcParams(UniValue::VARR);
UniValue chainInfoObject = getblockchaininfo(rpcParams, false);
string strJSON = chainInfoObject.write() + "\n";
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "application/json");
req->WriteReply(HTTP_OK, strJSON);
return true;
}
default: {
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_NOT_FOUND, "output format not found (available: json)");
}
}
// not reached
return true; // continue to process further HTTP reqs on this cxn
}
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
static bool rest_mempool_info(HTTPRequest* req, const std::string& strURIPart)
{
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
if (!CheckWarmup(req))
return false;
vector<string> params;
const RetFormat rf = ParseDataFormat(params, strURIPart);
switch (rf) {
case RF_JSON: {
UniValue mempoolInfoObject = mempoolInfoToJSON();
string strJSON = mempoolInfoObject.write() + "\n";
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "application/json");
req->WriteReply(HTTP_OK, strJSON);
return true;
}
default: {
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_NOT_FOUND, "output format not found (available: json)");
}
}
// not reached
return true; // continue to process further HTTP reqs on this cxn
}
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
static bool rest_mempool_contents(HTTPRequest* req, const std::string& strURIPart)
{
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
if (!CheckWarmup(req))
return false;
vector<string> params;
const RetFormat rf = ParseDataFormat(params, strURIPart);
switch (rf) {
case RF_JSON: {
UniValue mempoolObject = mempoolToJSON(true);
string strJSON = mempoolObject.write() + "\n";
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "application/json");
req->WriteReply(HTTP_OK, strJSON);
return true;
}
default: {
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_NOT_FOUND, "output format not found (available: json)");
}
}
// not reached
return true; // continue to process further HTTP reqs on this cxn
}
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
static bool rest_tx(HTTPRequest* req, const std::string& strURIPart)
{
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
if (!CheckWarmup(req))
return false;
vector<string> params;
const RetFormat rf = ParseDataFormat(params, strURIPart);
string hashStr = params[0];
uint256 hash;
if (!ParseHashStr(hashStr, hash))
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_BAD_REQUEST, "Invalid hash: " + hashStr);
CTransaction tx;
uint256 hashBlock = uint256();
if (!GetTransaction(hash, tx, hashBlock, true))
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_NOT_FOUND, hashStr + " not found");
CDataStream ssTx(SER_NETWORK, PROTOCOL_VERSION);
ssTx << tx;
switch (rf) {
case RF_BINARY: {
string binaryTx = ssTx.str();
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "application/octet-stream");
req->WriteReply(HTTP_OK, binaryTx);
return true;
}
case RF_HEX: {
string strHex = HexStr(ssTx.begin(), ssTx.end()) + "\n";
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "text/plain");
req->WriteReply(HTTP_OK, strHex);
return true;
}
case RF_JSON: {
UniValue objTx(UniValue::VOBJ);
TxToJSON(tx, hashBlock, objTx);
string strJSON = objTx.write() + "\n";
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "application/json");
req->WriteReply(HTTP_OK, strJSON);
return true;
}
default: {
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_NOT_FOUND, "output format not found (available: " + AvailableDataFormatsString() + ")");
}
}
// not reached
return true; // continue to process further HTTP reqs on this cxn
}
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
static bool rest_getutxos(HTTPRequest* req, const std::string& strURIPart)
{
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
if (!CheckWarmup(req))
return false;
vector<string> params;
enum RetFormat rf = ParseDataFormat(params, strURIPart);
vector<string> uriParts;
if (params.size() > 0 && params[0].length() > 1)
{
std::string strUriParams = params[0].substr(1);
boost::split(uriParts, strUriParams, boost::is_any_of("/"));
}
// throw exception in case of a empty request
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
std::string strRequestMutable = req->ReadBody();
if (strRequestMutable.length() == 0 && uriParts.size() == 0)
return RESTERR(req, HTTP_INTERNAL_SERVER_ERROR, "Error: empty request");
bool fInputParsed = false;
bool fCheckMemPool = false;
vector<COutPoint> vOutPoints;
// parse/deserialize input
// input-format = output-format, rest/getutxos/bin requires binary input, gives binary output, ...
if (uriParts.size() > 0)
{
//inputs is sent over URI scheme (/rest/getutxos/checkmempool/txid1-n/txid2-n/...)
if (uriParts.size() > 0 && uriParts[0] == "checkmempool")
fCheckMemPool = true;
for (size_t i = (fCheckMemPool) ? 1 : 0; i < uriParts.size(); i++)
{
uint256 txid;
int32_t nOutput;
std::string strTxid = uriParts[i].substr(0, uriParts[i].find("-"));
std::string strOutput = uriParts[i].substr(uriParts[i].find("-")+1);
if (!ParseInt32(strOutput, &nOutput) || !IsHex(strTxid))
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_INTERNAL_SERVER_ERROR, "Parse error");
txid.SetHex(strTxid);
vOutPoints.push_back(COutPoint(txid, (uint32_t)nOutput));
}
if (vOutPoints.size() > 0)
fInputParsed = true;
else
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_INTERNAL_SERVER_ERROR, "Error: empty request");
}
switch (rf) {
case RF_HEX: {
// convert hex to bin, continue then with bin part
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
std::vector<unsigned char> strRequestV = ParseHex(strRequestMutable);
strRequestMutable.assign(strRequestV.begin(), strRequestV.end());
}
case RF_BINARY: {
try {
//deserialize only if user sent a request
if (strRequestMutable.size() > 0)
{
if (fInputParsed) //don't allow sending input over URI and HTTP RAW DATA
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_INTERNAL_SERVER_ERROR, "Combination of URI scheme inputs and raw post data is not allowed");
CDataStream oss(SER_NETWORK, PROTOCOL_VERSION);
oss << strRequestMutable;
oss >> fCheckMemPool;
oss >> vOutPoints;
}
} catch (const std::ios_base::failure& e) {
// abort in case of unreadable binary data
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_INTERNAL_SERVER_ERROR, "Parse error");
}
break;
}
case RF_JSON: {
if (!fInputParsed)
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_INTERNAL_SERVER_ERROR, "Error: empty request");
break;
}
default: {
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_NOT_FOUND, "output format not found (available: " + AvailableDataFormatsString() + ")");
}
}
// limit max outpoints
if (vOutPoints.size() > MAX_GETUTXOS_OUTPOINTS)
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_INTERNAL_SERVER_ERROR, strprintf("Error: max outpoints exceeded (max: %d, tried: %d)", MAX_GETUTXOS_OUTPOINTS, vOutPoints.size()));
// check spentness and form a bitmap (as well as a JSON capable human-readble string representation)
vector<unsigned char> bitmap;
vector<CCoin> outs;
std::string bitmapStringRepresentation;
boost::dynamic_bitset<unsigned char> hits(vOutPoints.size());
{
LOCK2(cs_main, mempool.cs);
CCoinsView viewDummy;
CCoinsViewCache view(&viewDummy);
CCoinsViewCache& viewChain = *pcoinsTip;
CCoinsViewMemPool viewMempool(&viewChain, mempool);
if (fCheckMemPool)
view.SetBackend(viewMempool); // switch cache backend to db+mempool in case user likes to query mempool
for (size_t i = 0; i < vOutPoints.size(); i++) {
CCoins coins;
uint256 hash = vOutPoints[i].hash;
if (view.GetCoins(hash, coins)) {
mempool.pruneSpent(hash, coins);
if (coins.IsAvailable(vOutPoints[i].n)) {
hits[i] = true;
// Safe to index into vout here because IsAvailable checked if it's off the end of the array, or if
// n is valid but points to an already spent output (IsNull).
CCoin coin;
coin.nTxVer = coins.nVersion;
coin.nHeight = coins.nHeight;
coin.out = coins.vout.at(vOutPoints[i].n);
assert(!coin.out.IsNull());
outs.push_back(coin);
}
}
bitmapStringRepresentation.append(hits[i] ? "1" : "0"); // form a binary string representation (human-readable for json output)
}
}
boost::to_block_range(hits, std::back_inserter(bitmap));
switch (rf) {
case RF_BINARY: {
// serialize data
// use exact same output as mentioned in Bip64
CDataStream ssGetUTXOResponse(SER_NETWORK, PROTOCOL_VERSION);
ssGetUTXOResponse << chainActive.Height() << chainActive.Tip()->GetBlockHash() << bitmap << outs;
string ssGetUTXOResponseString = ssGetUTXOResponse.str();
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "application/octet-stream");
req->WriteReply(HTTP_OK, ssGetUTXOResponseString);
return true;
}
case RF_HEX: {
CDataStream ssGetUTXOResponse(SER_NETWORK, PROTOCOL_VERSION);
ssGetUTXOResponse << chainActive.Height() << chainActive.Tip()->GetBlockHash() << bitmap << outs;
string strHex = HexStr(ssGetUTXOResponse.begin(), ssGetUTXOResponse.end()) + "\n";
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "text/plain");
req->WriteReply(HTTP_OK, strHex);
return true;
}
case RF_JSON: {
UniValue objGetUTXOResponse(UniValue::VOBJ);
// pack in some essentials
// use more or less the same output as mentioned in Bip64
objGetUTXOResponse.push_back(Pair("chainHeight", chainActive.Height()));
objGetUTXOResponse.push_back(Pair("chaintipHash", chainActive.Tip()->GetBlockHash().GetHex()));
objGetUTXOResponse.push_back(Pair("bitmap", bitmapStringRepresentation));
UniValue utxos(UniValue::VARR);
BOOST_FOREACH (const CCoin& coin, outs) {
UniValue utxo(UniValue::VOBJ);
utxo.push_back(Pair("txvers", (int32_t)coin.nTxVer));
utxo.push_back(Pair("height", (int32_t)coin.nHeight));
utxo.push_back(Pair("value", ValueFromAmount(coin.out.nValue)));
// include the script in a json output
UniValue o(UniValue::VOBJ);
ScriptPubKeyToJSON(coin.out.scriptPubKey, o, true);
utxo.push_back(Pair("scriptPubKey", o));
utxos.push_back(utxo);
}
objGetUTXOResponse.push_back(Pair("utxos", utxos));
// return json string
string strJSON = objGetUTXOResponse.write() + "\n";
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
req->WriteHeader("Content-Type", "application/json");
req->WriteReply(HTTP_OK, strJSON);
return true;
}
default: {
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
return RESTERR(req, HTTP_NOT_FOUND, "output format not found (available: " + AvailableDataFormatsString() + ")");
}
}
// not reached
return true; // continue to process further HTTP reqs on this cxn
}
static const struct {
const char* prefix;
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
bool (*handler)(HTTPRequest* req, const std::string& strReq);
} uri_prefixes[] = {
{"/rest/tx/", rest_tx},
{"/rest/block/notxdetails/", rest_block_notxdetails},
{"/rest/block/", rest_block_extended},
{"/rest/chaininfo", rest_chaininfo},
{"/rest/mempool/info", rest_mempool_info},
{"/rest/mempool/contents", rest_mempool_contents},
10 years ago
{"/rest/headers/", rest_headers},
{"/rest/getutxos", rest_getutxos},
};
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
bool StartREST()
{
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
for (unsigned int i = 0; i < ARRAYLEN(uri_prefixes); i++)
RegisterHTTPHandler(uri_prefixes[i].prefix, false, uri_prefixes[i].handler);
return true;
}
evhttpd implementation - *Replace usage of boost::asio with [libevent2](http://libevent.org/)*. boost::asio is not part of C++11, so unlike other boost there is no forwards-compatibility reason to stick with it. Together with #4738 (convert json_spirit to UniValue), this rids Bitcoin Core of the worst offenders with regard to compile-time slowness. - *Replace spit-and-duct-tape http server with evhttp*. Front-end http handling is handled by libevent, a work queue (with configurable depth and parallelism) is used to handle application requests. - *Wrap HTTP request in C++ class*; this makes the application code mostly HTTP-server-neutral - *Refactor RPC to move all http-specific code to a separate file*. Theoreticaly this can allow building without HTTP server but with another RPC backend, e.g. Qt's debug console (currently not implemented) or future RPC mechanisms people may want to use. - *HTTP dispatch mechanism*; services (e.g., RPC, REST) register which URL paths they want to handle. By using a proven, high-performance asynchronous networking library (also used by Tor) and HTTP server, problems such as #5674, #5655, #344 should be avoided. What works? bitcoind, bitcoin-cli, bitcoin-qt. Unit tests and RPC/REST tests pass. The aim for now is everything but SSL support. Configuration options: - `-rpcthreads`: repurposed as "number of work handler threads". Still defaults to 4. - `-rpcworkqueue`: maximum depth of work queue. When this is reached, new requests will return a 500 Internal Error. - `-rpctimeout`: inactivity time, in seconds, after which to disconnect a client. - `-debug=http`: low-level http activity logging
10 years ago
void InterruptREST()
{
}
void StopREST()
{
for (unsigned int i = 0; i < ARRAYLEN(uri_prefixes); i++)
UnregisterHTTPHandler(uri_prefixes[i].prefix, false);
}