naterush / goblocks Goto Github PK

Simplifying the calling of blaze to eliminate default options.

Assume the chain is at block b.

If we run goblocks with --startBlock b-100 --nBlocks 1000 the processing will run off the end of the chain. This is probably okay because the code that processes blocks will simply error out (although we could make that more graceful).

The trouble is that as the processor gets closer and closer to the front of the chain there is more and more of a chance that the block will re-organize. The way I've handled this elsewhere is to write the results files to a different folder if the block is within a certain time to the front of the chain. For example, if blockTimestamp >= five minutes from front of chain, write to ./raw folder. If blockTimeStamp < five minutes from front of chain, write to ./unripe folder.

In my code (TrueBlocks), I can handle this by only calling in with block numbers older than five minutes, but we need to figure this out because my current scrape is getting close to head of chain.

As a temp solution in my code, I'm going to send in the path into which blaze should write its data. If I want the files written to ./raw, I will send that path in. If I want the files written to ./unripe, I will send that in.

https://github.com/Great-Hill-Corporation/trueblocks-core/issues/677
https://github.com/Great-Hill-Corporation/trueblocks-core/issues/1044

If I set startBlock to zero, I get an index out of range error. Change it to '1' and it works.

The files that are created during the scrape are given these permissions:

-rwxr-xr-x  1 jrush  staff   59 Jul 10 07:56 000001073.txt

There's no reason to give the file executable permissions. Write by owner, read by all is enough.