[PATCH] getblockbycount, getblockbyhash RPCs

JoelKatz

legendary

Activity: 1596

Merit: 1012

Democracy is vulnerable to a 51% attack.

Doesn't this code access mapBlockIndex without holding the cs_main lock? If that's legal, what protects mapBlockIndex? If that's illegal, isn't the patch broken? (If this question is obsolete because there's a newer version of the patch, where can I find it?)

Furyan

full member

Activity: 175

Merit: 100

Quote from: TeraPool on July 21, 2011, 05:56:24 PM

Thank you incredibly Furyan.

Glad to help

Now if I can get Gavin's Monitor patch in there... much more invasive and stomps on some patches I've already applied..

TeraPool

newbie

Activity: 42

Merit: 0

Thank you incredibly Furyan.

Furyan

full member

Activity: 175

Merit: 100

~~Screw that, won't work. Need a clean diff.~~

Oh, ok, this is the latest:

https://github.com/jgarzik/bitcoin/commit/168f5a4b30d5307281eb5bc4447afb04348e9f5d.patch

But you'd need a unified diff of all changes to that file. There's a way to do that using git on the command line but I'm not familiar with it.

Edit: Here you go. This is against the 0.3.24 release version so if you've made other custom changes you'll need to apply the patch more carefully.

Code:

--- rpc.cpp       2011-07-21 11:37:38.000000000 -0700
+++ rpc.cpp     2011-07-21 11:39:21.000000000 -0700
@@ -168,6 +168,157 @@
 }


+Value BlockToValue(CBlock &block)
+{
+    Object obj;
+    obj.push_back(Pair("hash", block.GetHash().ToString().c_str()));
+    obj.push_back(Pair("version", block.nVersion));
+    obj.push_back(Pair("prev_block", block.hashPrevBlock.ToString().c_str()));
+    obj.push_back(Pair("mrkl_root", block.hashMerkleRoot.ToString().c_str()));
+    obj.push_back(Pair("time", (uint64_t)block.nTime));
+    obj.push_back(Pair("bits", (uint64_t)block.nBits));
+    obj.push_back(Pair("nonce", (uint64_t)block.nNonce));
+    obj.push_back(Pair("n_tx", (int)block.vtx.size()));
+    obj.push_back(Pair("size", (int)::GetSerializeSize(block, SER_NETWORK)));
+
+    Array tx;
+    for (int i = 0; i < block.vtx.size(); i++) {
+       Object txobj;
+
+   txobj.push_back(Pair("hash", block.vtx[i].GetHash().ToString().c_str()));
+   txobj.push_back(Pair("version", block.vtx[i].nVersion));
+   txobj.push_back(Pair("lock_time", (uint64_t)block.vtx[i].nLockTime));
+   txobj.push_back(Pair("size",
+       (int)::GetSerializeSize(block.vtx[i], SER_NETWORK)));
+
+   Array tx_vin;
+   for (int j = 0; j < block.vtx[i].vin.size(); j++) {
+       Object vino;
+
+       Object vino_outpt;
+
+       vino_outpt.push_back(Pair("hash",
+           block.vtx[i].vin[j].prevout.hash.ToString().c_str()));
+       vino_outpt.push_back(Pair("n", (uint64_t)block.vtx[i].vin[j].prevout.n));
+
+       vino.push_back(Pair("prev_out", vino_outpt));
+
+       if (block.vtx[i].vin[j].prevout.IsNull())
+           vino.push_back(Pair("coinbase", HexStr(
+           block.vtx[i].vin[j].scriptSig.begin(),
+           block.vtx[i].vin[j].scriptSig.end(), false).c_str()));
+       else
+           vino.push_back(Pair("scriptSig",
+           block.vtx[i].vin[j].scriptSig.ToString().c_str()));
+       if (block.vtx[i].vin[j].nSequence != UINT_MAX)
+           vino.push_back(Pair("sequence", (uint64_t)block.vtx[i].vin[j].nSequence));
+
+       tx_vin.push_back(vino);
+   }
+
+   Array tx_vout;
+   for (int j = 0; j < block.vtx[i].vout.size(); j++) {
+       Object vouto;
+
+       vouto.push_back(Pair("value",
+           (double)block.vtx[i].vout[j].nValue / (double)COIN));
+       vouto.push_back(Pair("scriptPubKey",
+       block.vtx[i].vout[j].scriptPubKey.ToString().c_str()));
+
+       tx_vout.push_back(vouto);
+   }
+
+   txobj.push_back(Pair("in", tx_vin));
+   txobj.push_back(Pair("out", tx_vout));
+
+   tx.push_back(txobj);
+    }
+
+    obj.push_back(Pair("tx", tx));
+
+    Array mrkl;
+    for (int i = 0; i < block.vMerkleTree.size(); i++)
+       mrkl.push_back(block.vMerkleTree[i].ToString().c_str());
+
+    obj.push_back(Pair("mrkl_tree", mrkl));
+
+    return obj;
+}
+
+
+Value getblockbycount(const Array& params, bool fHelp)
+{
+    if (fHelp || params.size() != 1)
+        throw runtime_error(
+            "getblockbycount height\n"
+            "Dumps the block existing at specified height");
+
+    int64 height = params[0].get_int64();
+    if (height > nBestHeight)
+        throw runtime_error(
+            "getblockbycount height\n"
+            "Dumps the block existing at specified height");
+
+    string blkname = strprintf("blk%d", height);
+
+    CBlock block;
+    CBlockIndex* pindex;
+    bool found = false;
+
+    CRITICAL_BLOCK(cs_main)
+    {
+
+    for (map::iterator mi = mapBlockIndex.begin();
+         mi != mapBlockIndex.end(); ++mi)
+    {
+       pindex = (*mi).second;
+   if ((pindex->nHeight == height) && (pindex->IsInMainChain())) {
+       found = true;
+       break;
+   }
+    }
+
+    if (!found)
+        throw runtime_error(
+            "getblockbycount height\n"
+            "Dumps the block existing at specified height");
+
+    block.ReadFromDisk(pindex);
+    block.BuildMerkleTree();
+    }
+
+    return BlockToValue(block);
+}
+
+
+Value getblockbyhash(const Array& params, bool fHelp)
+{
+    if (fHelp || params.size() != 1)
+        throw runtime_error(
+            "getblockbyhash hash\n"
+            "Dumps the block with specified hash");
+
+    uint256 hash;
+    hash.SetHex(params[0].get_str());
+    CBlock block;
+
+    CRITICAL_BLOCK(cs_main)
+    {
+
+    map::iterator mi = mapBlockIndex.find(hash);
+    if (mi == mapBlockIndex.end())
+        throw JSONRPCError(-18, "hash not found");
+
+    CBlockIndex* pindex = (*mi).second;
+
+    block.ReadFromDisk(pindex);
+    block.BuildMerkleTree();
+    }
+
+    return BlockToValue(block);
+}
+
+
 Value getblockcount(const Array& params, bool fHelp)
 {
     if (fHelp || params.size() != 0)
@@ -1434,6 +1585,8 @@
 {
     make_pair("help",                  &help),
     make_pair("stop",                  &stop),
+    make_pair("getblockbycount",       &getblockbycount),
+    make_pair("getblockbyhash",        &getblockbyhash),
     make_pair("getblockcount",         &getblockcount),
     make_pair("getblocknumber",        &getblocknumber),
     make_pair("getconnectioncount",    &getconnectioncount),
@@ -2130,6 +2283,7 @@
         if (strMethod == "listtransactions"       && n > 1) ConvertTo(params[1]);
         if (strMethod == "listtransactions"       && n > 2) ConvertTo(params[2]);
         if (strMethod == "listaccounts"           && n > 0) ConvertTo(params[0]);
+        if (strMethod == "getblockbycount"        && n > 0) ConvertTo(params[0]);
         if (strMethod == "sendmany"               && n > 1)
         {
             string s = params[1].get_str();

I suggest you use the -b flag to patch to make a backup of your rpc.cpp.

Works great!

Furyan

full member

Activity: 175

Merit: 100

Quote from: TeraPool on July 20, 2011, 03:08:27 AM

Quote from: error on May 22, 2011, 11:01:47 PM

Quote from: Luke-Jr on January 29, 2011, 11:07:49 AM

Any reason this shouldn't be merged with the main tree?

Seconded.

Thirded!

This is amazing stuff though, thanks jgarzik! Perfect for the novice pool operator.

Can somebody please explain to me how to add this patch to the main bitcoind client? I have never done a patch before :S

You need the diff to apply the patch. This should be against 0.3.24 source but I haven't checked.

You can get the diff at github (sort of) -

https://github.com/jgarzik/bitcoin/commit/168f5a4b30d5307281eb5bc4447afb04348e9f5d#diff-0

Copy that into a file (e.g. diff.txt), then run

Code:

patch -i diff.txt rpc.cpp

You may want to add

Code:

-r rejects.txt

to catch any rejections that may happen so you can fix them.

Will be doing this when I get home, I like this patch.

Edit:
Actually it looks like JoelKatz' pull request has the complete diff....

Unless jgarzik would be so kind as to generate one for us

(Reason I need a diff is that I have other patches applied to my local copy - most of JoelKatz' work).

TeraPool

newbie

Activity: 42

Merit: 0

Quote from: error on May 22, 2011, 11:01:47 PM

Quote from: Luke-Jr on January 29, 2011, 11:07:49 AM

Any reason this shouldn't be merged with the main tree?

Seconded.

Thirded!

This is amazing stuff though, thanks jgarzik! Perfect for the novice pool operator.

Can somebody please explain to me how to add this patch to the main bitcoind client? I have never done a patch before :S

Grouver (BtcBalance)

hero member

Activity: 530

Merit: 500

Right, got it to work. (Thanks to the bitcoin-dev channel)
Right now I get this:

Code:

array
  'hash' => string '000000000000053eb0059c82f86013cbf205033d82754b5252a04059e331c55d' (length=64)
  'version' => int 1
  'prev_block' => string '00000000000009e1e09847ee394f5394a3eed9d7c37949ac7ef0648760b0e488' (length=64)
  'mrkl_root' => string '53f1acacb93d62107d7ba62bdb50beefcc803dd4e3dfd33ed26857108901006d' (length=64)
  'time' => int 1309386843
  'bits' => int 437004818
  'nonce' => float 3446647446
  'n_tx' => int 81
  'size' => int 36104
  'tx' => 
    array
      0 => 
        array
          'hash' => string '1720c9d73186c87f31e9da0e3189fb67ce3252f8b300f603f8ae659b6f2bf60f' (length=64)
          'version' => int 1
          'lock_time' => int 0
          'size' => int 134
          'in' => 
            array
              ...
          'out' => 
            array
              ...
      1 => 
        array
          'hash' => string 'd3cd0851e088481afebd0178537cbb4cb9464ea03454b1c646643a1097727915' (length=64)
          'version' => int 1
          'lock_time' => int 0
          'size' => int 258
          'in' => 
            array
              ...
          'out' => 
            array
              ...
      2 => 
        array
          'hash' => string 'cb81abf818ad18c51c78c9724dbd9b74be49c7d7d748d9a3eafcd8f4b346b498' (length=64)
          'version' => int 1
          'lock_time' => int 0
          'size' => int 223
          'in' => 
            array
              ...
          'out' => 
            array
              ...

etc..

Why are the IN and OUT arrays empty?
When I check the startpost I see there is a another array in there?

EDIT:
Also I am not sure what I should do with Berkeley DB.
The readme told me to make but thats all.
How do I need to create and connect the DB to the Bitcoin server?
Or am I tottaly misunderstanding this.

jgarzik

legendary

Activity: 1596

Merit: 1091

Quote from: Grouver (BtcBalance) on June 29, 2011, 04:11:02 PM

Only when I restarted my Linux server and try for instance:

Code:

bitcoin getblock 71995

I get the error:

Code:

error: {"code":-32601,"message":"Method not found"}

That's because it is called 'getblockbycount' not 'getblock' these days.

Grouver (BtcBalance)

hero member

Activity: 530

Merit: 500

Hello,

Thank you so much for developing this patch.
I am trying to install it right now on my bitcoin server.
I followed your instructions that are in the manual and all went fine.

I tried:

Code:

bitcoin getblockcount

Wich obviously worked flawlessly.

Only when I restarted my Linux server and try for instance:

Code:

bitcoin getblock 71995

I get the error:

Code:

error: {"code":-32601,"message":"Method not found"}

I builded the patch with wxWidgets.

Thanks again.

xf2_org

member

Activity: 98

Merit: 13

getblockbycount / getblockbyhash RPCs updated to latest git.

davout

legendary

Activity: 1372

Merit: 1007

1davout

Could it be possible that getblockbyhash returns an "orphan_block" flag ?

See this example that I generated on testnet :

Code:

david@bankbox:~/rails/bitcoin-pool$ bitcoin getblockbyhash 00000000003d7e088aba9d12e34cc3f75cd10ffa48a026fa535c0771e5c1b0c8
{
    "hash" : "00000000003d7e088aba9d12e34cc3f75cd10ffa48a026fa535c0771e5c1b0c8",
    "version" : 1,
    "prev_block" : "0000000000176fa50f3c759b2025acdbb0b4f311372e2b94acb0e69f73ae7c6b",
    "mrkl_root" : "46689b6020b67534e63fb81ddf379bb666a96ff55de27f8eebe8dffc924c09e7",
    "time" : 1307350469,
    "bits" : 470193746,
    "nonce" : 3293063839,
    "n_tx" : 1,
    "size" : 215,
    "tx" : [
        {
            "hash" : "46689b6020b67534e63fb81ddf379bb666a96ff55de27f8eebe8dffc924c09e7",
            "version" : 1,
            "lock_time" : 0,
            "size" : 134,
            "in" : [
                {
                    "prev_out" : {
                        "hash" : "0000000000000000000000000000000000000000000000000000000000000000",
                        "n" : 4294967295
                    },
                    "coinbase" : "045296061c0112"
                }
            ],
            "out" : [
                {
                    "value" : 50.00000000,
                    "scriptPubKey" : "04d98927faa89cd92df4861f08933ff68f50182850775fc590965049e34de80f6277cbcc9d5446ed43abf9794d70df9813b9fb7c589805a70676ddaee8e803d806 OP_CHECKSIG"
                }
            ]
        }
    ],
    "mrkl_tree" : [
        "46689b6020b67534e63fb81ddf379bb666a96ff55de27f8eebe8dffc924c09e7"
    ]
}

Nothing wrong here, but when I want to get details about the generation tx this is what I get :

Code:

david@bankbox:~/rails/bitcoin-pool$ bitcoin gettransaction 46689b6020b67534e63fb81ddf379bb666a96ff55de27f8eebe8dffc924c09e7
{
    "amount" : 0.00000000,
    "confirmations" : 0,
    "txid" : "46689b6020b67534e63fb81ddf379bb666a96ff55de27f8eebe8dffc924c09e7",
    "time" : 1307350469,
    "details" : [
        {
            "account" : "",
            "category" : "orphan",
            "amount" : 50.00000000
        }
    ]
}

Also this should get pulled in, it's incredibly useful when building decent bitcoin handling apps.

error

hero member

Activity: 588

Merit: 500

Quote from: Luke-Jr on January 29, 2011, 11:07:49 AM

Any reason this shouldn't be merged with the main tree?

Seconded.

xf2_org

member

Activity: 98

Merit: 13

getblockbycount RPC updated to latest git. As the top post notes, this work is now maintained at github, jgarzik/bitcoin's "getblockbycount" branch.

Notable addition: new RPC "getblockbyhash" added, to retrieve block by block header hash.

fahhem

newbie

Activity: 11

Merit: 0

Quote from: jgarzik on February 23, 2011, 08:44:18 PM

Patch updated to latest git. See URL at top of thread.

I updated it to the latest git. http://pastebin.com/FdrYDVF5

I made it as a git format-patch so it can be applied using git-apply. To make it a normal patch, just delete the first 9 lines.

jgarzik

legendary

Activity: 1596

Merit: 1091

Patch updated to latest git. See URL at top of thread.

Luke-Jr

legendary

Activity: 2576

Merit: 1186

Any reason this shouldn't be merged with the main tree?

jgarzik

legendary

Activity: 1596

Merit: 1091

Patch updated for latest SVN.

jgarzik

legendary

Activity: 1596

Merit: 1091

Patch updated to latest SVN, with additional block/TX size output.

theymos

administrator

Activity: 5166

Merit: 12850

To get block/tx size, you can add this to block processing:

Code:

obj.push_back(Pair("size", (int)::GetSerializeSize(block, SER_NETWORK)));

And add this to tx processing:

Code:

txobj.push_back(Pair("size", (int)::GetSerializeSize(block.vtx[i], SER_NETWORK)));

I don't know that this is the best way of doing this (I'm not really familiar with C++). That cast to integer is probably not necessary, but it works and I don't want to test another version (overflow is impossible for the foreseeable future).

jgarzik

legendary

Activity: 1596

Merit: 1091

Updated to SVN r188 (accounts).

Topic: [PATCH] getblockbycount, getblockbyhash RPCs (Read 30500 times)