-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Locator flag with transaction list command #86
Comments
This command does not store locators, it only uses them, i.e. if they are already stored with It is important to know that you normally can't store completely arbitrary block intervals, this is intentional, to prevent people recording incomplete intervals:
I'll have to check if pacli rejects all other combinations correctly; will do so in the next days. Will also try to reproduce the error you got. |
I think this is a good news for my testing, because I was trying to create conditions to compare the results with and without the locator feature, which is not necessary because I can switch between the two modes using the I'd like to ask you whether it would make sense to setup the I'll also add a suggestion here to enrich the description of the |
FYI I've run
The address However the If I run
While the
The
while the
I'd assume that the |
I'd totally forgot this issue wasn't completely solved, but now I looked into it.
This is an unsupported combination. If you use locators, you have to provide an address to follow. Of course the error should be more meaningful, so I created a new message.
Correct observation, thanks. It would show actually all transactions in blocks which were already stored as locators. Reproduced and fixed it. I saw that I gave you wrong information above about the -l option (sorry, the transaction list command is so complex that I sometimes forget details):
It actually does store the blocks of the range which are behind the last checked block. As far as I remember the idea was to not "lose" the scanning work which is done in this case. I have decided to keep it this way, but to store the locator blockheights only if the start block is not behind the last cached block. Otherwise the user could be surprised that the process took so long because it would try to store all blocks up to the chosen endblock (example: you have already cached until 100000, startblock is 300000 and endblock 350000 - it would then cache between 100000 and 350000). The rest of that post is however correct: you cannot cache locators of arbitrary block ranges. About making Commit for the fixes is 32405b0. |
I've tried to store blocks using
Which is expected.
And of course the blocklocator.json file is not being updated. So I've say that the |
I fixed the UnboundLocalError, which appeared due to a mistake I made in the last related commit. Now the blockheights should be stored correctly in the cases I mentioned (at least here it works with the To check the blocklocator.json comfortably I added a command: Commit: 39ddb90 |
I've tested At the same time if I run |
While I didn't get that many transactions without the address in the case of mkwJijAwqXNFcEBxuC93j3Kni43wzXVuik, there were 4 txes in my output of |
Fixed the bug in commit 036949f. Transactions not related to addresses were indeed shown due to a bug in the branching logic. This has been fixed now and simplified too. EDIT: I recommend testing first the other commands until I post here again, I discovered some new issues with the block locator function I'd like to fix first. |
The issues I encountered are (hopefully) fixed and pre-tested as of commit 7c31f67. Block locator features ( It is however possible that due to the earlier bugs the blocklocator.json became corrupted. If anything looks strange then I highly recommend to simply delete or rename the blocklocator.json and start new. The exception is until now the integrity test. There I have still to test more. |
I've tested
I've thought there is some interference with the blocklocator.json file and have removed it - the error hasn't gone.
I've run
Then I've run
|
First two errors were fixed in commit 08247a9 (blocklocators.json was not the culprit here). The third problem is because this mode is unsupported, and this was also already mentioned in the help (Note 2). I have done however a workaround, in this case the locators are simply switched off. Alternatively I can raise a more meaningful error but here we run again into the problem: "how much have we to hold the hand of the user if he doesn't read the help properly?". This is in commit dc153e4. By the way: it's easier for me to detect the problems which cause "red errors" (like the well-known "General Error ...") if you post the Can be closed if everything works. |
It seems like there is still an issue with the |
Confirmed and fixed in commit 59163ee. Please close if everything works. |
Seems to work, closing. |
I was testing
transaction list
with the-l
flag.But probably I don't have a clear idea about the usage of this flag.
In fact while I was trying to cache the blocks from 156492 to 156902 by running
transaction list -x -l -f 156492 -e 156902
and alsotransaction list -x -f 156492 -e 156902 -l
I've got the following error:Should I first store the blocks using
deck cache
ordeck init
command as described in issue #50 or there is a way to store a specific interval of the blocks? If so what is the-l
usage?The text was updated successfully, but these errors were encountered: