Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • Jan 31 2019 21:12
    codecov-io commented #425
  • Jan 31 2019 20:52
    NicolasDP synchronize #425
  • Jan 31 2019 20:52

    NicolasDP on update-ledger-mock

    make the Genesis leader selecti… change name of the am_leader_at… (compare)

  • Jan 31 2019 20:23
    codecov-io commented #425
  • Jan 31 2019 20:10
    NicolasDP assigned #425
  • Jan 31 2019 20:10
    NicolasDP review_requested #425
  • Jan 31 2019 20:10
    NicolasDP labeled #425
  • Jan 31 2019 20:10
    NicolasDP opened #425
  • Jan 31 2019 20:08

    NicolasDP on update-ledger-mock

    network-grpc: Hide implementati… Merge pull request #420 from in… CircleCI: remove junit result c… and 7 more (compare)

  • Jan 31 2019 19:26

    qnikst on ntt-cli

    WIP (compare)

  • Jan 31 2019 16:28

    vincenthz on master

    network: Rename StreamBlocksTo*… Merge pull request #422 from in… (compare)

  • Jan 31 2019 16:28
    vincenthz closed #422
  • Jan 31 2019 16:27

    vincenthz on master

    use unified update model Merge pull request #424 from in… (compare)

  • Jan 31 2019 16:27
    vincenthz closed #424
  • Jan 31 2019 15:42
    NicolasDP labeled #424
  • Jan 31 2019 15:42
    NicolasDP assigned #424
  • Jan 31 2019 15:42
    NicolasDP review_requested #424
  • Jan 31 2019 15:41
    NicolasDP opened #424
  • Jan 31 2019 15:41
    vincenthz labeled #423
  • Jan 31 2019 15:41
    vincenthz opened #423
Rinor Hoxha
@rinor
yes, exactly
Rinor Hoxha
@rinor
Question 1: What is the long-term usage/vision of JCLI? (is it here to stay or just a temporary tool that will be deprecated)
Just asking because have a feeling the the community got a bit "tired" with the actual usage of shell scripts related to jcli, among other things, don't like the fact that those scripts can't be used easily and in the same way in all OS platforms without the need of installing specific shells. There is also this comment https://github.com/input-output-hk/jormungandr/pull/644#issuecomment-516555241 from Vincent.
Lately I've been thinking a bit about this, and beside what is mentioned on that comment, it would be nice to have some jcli (the executable binary) wrappers in some statically typed language that is small and easy enough to be understood that could be used for/like scripting and in the same time be able to produce executables if needed from the same codebase/scripts.
Before checking with the community, I would like to have an opinion if is worth pursuing it because would like it to be helpful/worth more then in short-term.
I understand that short-term/long-term are subject of interpretation, but I hope they make sense in this context.
Question 2: Is there any plan/way to use https://github.com/input-output-hk/chain-libs with a standalone JIT WebAssembly runtime? (smth like https://github.com/wasmerio/wasmer would be epic)?
https://github.com/input-output-hk/js-chain-libs is awesome, and it would be nice to have it also for other languages.
Nicolas Di Prima
@NicolasDP

Response to question 1: JCLI won't be deprecated. But it will never be much than what it is now:
a command line tool that emulate the raw functions of addresses, keys, etc... The vision for jcli
is that is ought to be a tool for the devs to play with the different objects, do some testing.
IOHK's haskell wallet backend utilises the jcli to validate what they do for example.

The scripts are interesting for some dev/qa manual testings. Like you said they
are not great for the general users. There are lot of issues with them and they don't support every
platforms.

A community project to provide a more user friendly application for users to monitor/control the node
would be great.

Response to question 2: We are building libraries and APIs for developers to use. There is the
Rust library in chain-libs and the JS bindings in js-chain-libs. We will provide C bindings too
but we need to put a bit more thoughts on the APIs we plan to provide. From there we will have access to
every languages with C FFI capabilities (Haskell, Go, Java, Objective-C...).

Follow up on these questions: We have focused mainly on the JS bindings as it seemed to us this is the most common language used for cross platform applications. An electron app to monitor/control the node would be great for example. And not that hard to write. Just need some time.

Rinor Hoxha
@rinor
Understood. Thank you for your time and detailed responses.
Rinor Hoxha
@rinor

A community project to provide a more user friendly application for users to monitor/control the node
would be great.

here we go https://github.com/rinor/jorcli

just started it and will see how far it goes
Nicolas Di Prima
@NicolasDP
Great. A project in Go! \o/
Markus
@gufmar
Awesome!
Rinor Hoxha
@rinor
thank you for the new SCENARIO TEST SUITE
Nicolas Di Prima
@NicolasDP
you tried it? still lot of things to work on on this. Will get more interesting by the end of the week
Rinor Hoxha
@rinor
Yes I did and liked it. Also it gave me some ideas I may implement in go apps.
Rinor Hoxha
@rinor
Quick question:
Beside digging the source code and/or graphiql server, is there any available standardized source for the followings:
1) Graphql full schemas for explorer
2) Protobuf schemas for grpc
Rinor Hoxha
@rinor
Thank you. For 2) I was looking at jormungandr source and missed chain-libs 🙈
Rinor Hoxha
@rinor
congrats, awesome job with the latest changes on jörmungandr node. Kudos to all of you
Nicolas Di Prima
@NicolasDP
thanks for your help too :) your contributions are not unnoticed
KoT_B_KocMoce
@hodlonaut
what's up folks
what does 'reason: bootstrap pull stream failed' mean
KoT_B_KocMoce
@hodlonaut
sorry about the spam
Rinor Hoxha
@rinor
yes it is a known issue
KoT_B_KocMoce
@hodlonaut
thanks
Nicolas Di Prima
@NicolasDP
feel free to ask this kind of questions or look for answers in shelley-testnet repository: https://github.com/input-output-hk/shelley-testnet the team have accumulated a great deal of common issues and help there.
hugolev
@hugolev
woot
testnet channel
finaly
are we live?
KoT_B_KocMoce
@hodlonaut
no
KoT_B_KocMoce
@hodlonaut
@NicolasDP & co: in current Shelley networking testnet, in the ideal scenario is there meant to be one leadership event assigned to some leader node per each slot in the epoch? or is it fine for there to be slot gaps and thus inconsistency in the block generation frequency? had trouble ascertaining that from the documentation
Rinor Hoxha
@rinor
Gaps are expexted and desired in genesis_praos
KoT_B_KocMoce
@hodlonaut
why are they desired/expected?
i.e. why wouldn't we want to have a leader / block per slot
i mean, yes i found a page that says "In Genesis-Praos slot time duration is constant, however the frequency of creating blocks is not stable, since the creation of blocks is a probability that is linked to the stake and consensus_genesis_praos_active_slot_coeff." but that still doesn't answer my question i.e. why wouldn't we strive to have consistent block creation throughput so to speak
KoT_B_KocMoce
@hodlonaut
or is it this way because it needs to be this randomness-involved lottery without cross-sharing of data between nodes
Rinor Hoxha
@rinor
If I remember correctly that it a PRAOS feature initially and has to do with the semi-synchronous notion and those "empty slots" or as called also "silence period" is used to help synchronization between partecipants.
KoT_B_KocMoce
@hodlonaut
after leader finishes with 'block from leader event successfully stored' as a part of their leader event processing which other nodes/peers/etc are meant to get notified of this new block? every peer/passive node/leader/etc on the current network?
KoT_B_KocMoce
@hodlonaut
also what's the quickest way for me to add logging messages to, say, try_send in comm.rs - do i have to pass in Logger parameter or is there some way to get a static logger initializer? (if there is such a thing in Rust)
Nicolas Di Prima
@NicolasDP
Hi, thanks a lot for your interests. and thanks @rinor for the help here too. The protocol is not really working like this. There are some concepts explained here: https://input-output-hk.github.io/jormungandr/concepts/blockchain.html

There is 2 things to take into consideration:

  • it is a lottery, multiple stake pool can win at the same time;
  • it is possible no one has won the current draw.

Because of these notions, it is possible to have slots where more than one block is valid for the same slot and it is possible to have empty slots.

it is normal to have empty slots and it is normal to have competitive blocks.
taipa-ibl
@taipa-ibl
@NicolasDP
Do rust-cardano have functions to generate the account address like "Daedalus - Balance check" on the Testnet ? Address is like: addr1sdslyh6nh99zj02ckyra8rluva8ks6z6fy6f9qgxfzuh0znvhe355jw034y
Nicolas Di Prima
@NicolasDP
@taipa-ibl : yes.
with prefix: --prefix=addr on jcli
are you looking for Javascript library or Rust Library function ?
taipa-ibl
@taipa-ibl
@NicolasDP
Thank you. Yes, i'm looking for Javascript library.
Nicolas Di Prima
@NicolasDP
and there is a nodejs version too there: https://www.npmjs.com/package/js-chain-libs-node
both contains typescript annotations too
taipa-ibl
@taipa-ibl
Great, thank you @NicolasDP
Omotoshogabriel
@Omotoshogabriel_twitter
Hi,we're recruiting technical evangelists, researchers, developers. It is strategically invested by Softbank. which aims to achieve public chain TPS sharing and intelligent scheduling through bRPC protocol with multiple public chains. Enjoying unlimited expansion of TPS and optimal GAS fee among the entire blockchain network.
You can join our channel to learn about project. https://gitter.im/marcopoloprotocols/community?utm_source=share-link&utm_medium=link&utm_campaign=share-link