1 2016-03-22 00:33:29 0|Chris_Stewart_5|I have a question about BIP66 and this test case inside of script_valid.json
2 2016-03-22 00:33:51 0|Chris_Stewart_5|[
3 2016-03-22 00:33:53 0|Chris_Stewart_5|"1",
4 2016-03-22 00:33:55 0|Chris_Stewart_5|"0x21 0x038282263212c609d9ea2a6e3e172de238d8c39cabd5ac1ca10646e23fd5f51508 CHECKSIG NOT",
5 2016-03-22 00:33:57 0|Chris_Stewart_5|"",
6 2016-03-22 00:33:59 0|Chris_Stewart_5|"BIP66 example 6, without DERSIG"
7 2016-03-22 00:34:01 0|Chris_Stewart_5|]
8 2016-03-22 00:34:21 0|Chris_Stewart_5|BIP66 says that "if the signature does not pass the IsValidSignatureEncoding check below, the entire script evaluates to false immediately."
9 2016-03-22 00:35:31 0|Chris_Stewart_5|why would this script evaluate to true? Wouldn't it fail immediately after seeing "1" since it is not DER compliant
10 2016-03-22 00:35:50 0|Chris_Stewart_5|shouldn't it just fail immediately
11 2016-03-22 00:35:52 0|Chris_Stewart_5|?
12 2016-03-22 00:37:35 0|gmaxwell|Chris_Stewart_5: I assume because it is not run with the DERSIG flag. (as is the case for most of the blocks in the chain)
13 2016-03-22 00:40:35 0|Chris_Stewart_5|gmaxwell: So after BIP66 this would fail, pre BIP66 this would pass, correct?
14 2016-03-22 00:40:45 0|Chris_Stewart_5|and this is indicated with the flags
15 2016-03-22 04:08:46 0|phantomcircuit|do i need a lock for Params() ?
16 2016-03-22 04:18:35 0|sipa|i'd say no
17 2016-03-22 05:46:41 0|phantomcircuit|sipa: what about CBlockIndex objects? I would assume yes but that they remain valid between lock acquisitions
18 2016-03-22 05:47:28 0|phantomcircuit|(i mean pointers)
19 2016-03-22 05:58:59 0|sipa|phantomcircuit: some fields are immutable
20 2016-03-22 06:00:18 0|sipa|if you take a CBlockIndex object from mapBlockIndex, its CBlockHeader fields, chainWork, pprev, ... are immutable
21 2016-03-22 06:00:27 0|sipa|so you can use GetAncestor without lock
22 2016-03-22 06:00:59 0|sipa|nStatus, nFile, nBlockPos, nUndoPos, ... are mutable and need cs_main
23 2016-03-22 06:05:24 0|phantomcircuit|sipa: just interested in whether a CBlockIndex pointed from mapBlockIndex will continue to work when passed to chainActive.Next
24 2016-03-22 06:11:23 0|sipa|phantomcircuit: yes, it will
25 2016-03-22 06:11:39 0|sipa|pointers to CBlockIndex remain vlaid forever
26 2016-03-22 06:12:04 0|sipa|chainActive.Next will return NULL if the passed CBlockIndex is no longer in the main chaim
27 2016-03-22 06:44:24 0|sipa|phantomcircuit: i'm wrong
28 2016-03-22 06:44:35 0|sipa|chainActive itself needs cs_main
29 2016-03-22 06:44:52 0|sipa|though the CBlockIndex pointer do remain valid across different lockings
30 2016-03-22 06:45:11 0|sipa|which may be enough
31 2016-03-22 07:30:06 0|jonasschnelli|Is there a chance to escape IBD in regtest (getheaders command reject with reason "node is in IBD)"?
32 2016-03-22 07:34:26 0|sipa|first mine a block?
33 2016-03-22 07:37:19 0|jonasschnelli|:} ... let me try that... I didn't tried that because the IBD warning icon was also there in Qt after I mined couple of blocks...
34 2016-03-22 07:37:29 0|sipa|hmm
35 2016-03-22 07:37:37 0|jonasschnelli|(1min)
36 2016-03-22 07:39:03 0|jonasschnelli|sipa: thanks! Was my stupidity (of not mining a block)...
37 2016-03-22 07:46:57 0|GitHub185|[13bitcoin] 15jonasschnelli opened pull request #7732: [Qt] Debug window: replace "Build date" with "Datadir" (06master...062016/03/qt_datadir) 02https://github.com/bitcoin/bitcoin/pull/7732