Well, yes, that's what a bootstrap.dat file *is*. You just write all the blocks in the main chain into a file one after the other.
I added a command 'dumpbootstrap' to clamd that creates the bootstrap.dat file rather than writing a separate Python script to do it.
You can probably just merge the commit into your coin.
Hmm.. I thought that python script stripped out all the orphan blocks.. and "to create a no-fork best chain version" as it states.
If I write out all the blocks into one file, I see the client trying to work through all the orphan blocks when importing.. (tail -f debug.log to see the messages).
I'll look at dumpbootstrap, thank you.
Note that I said "in the main chain".
Orphan blocks are by definition not in the main chain.
See
https://github.com/nochowderforyou/clams/commit/3bb9c0ed for the initial commit.
At its core it just does this - finds each block in order, reads it from disk, writes it to the bootstrap file:
+ for (int nHeight = 0; nHeight <= nBlocks; nHeight++)
+ {
+ CBlockIndex* pblockindex = FindBlockByHeight(nHeight);
+ block.ReadFromDisk(pblockindex, true);
+ fileout << FLATDATA(Params().MessageStart()) << fileout.GetSerializeSize(block) << block;
+ }
