Make subxt-core ready for publishing #1508

jsdw · 2024-03-28T14:51:17Z

Tadeo did some great work getting a bunch of no-std compatible bits moved out of subxt into this new subxt-core crate. The aim of this PR is to tidy up and polish this crate so that it's ready to be released and made use of directly.

Some things I did here:

Move Extrinsic decoding stuff to subxt_core
Move Transaction creation/signing stuff to subxt_core
Expose methods in storage and runtime APIs to encode/decode things and use in Subxt
Provide a top level example for each of the areas (tx, runtime_apis, events, constants, custom_values, storage, blocks) to show how to build/encode/decode the relevant bits, and overall docs to link to these main areas.
Various other tweaks to consistify the structure and such.

… for storage and txs

…untime API fns

jsdw · 2024-04-09T10:20:54Z

cli/src/commands/explore/pallets/calls.rs

+    let runtime_version = subxt::client::RuntimeVersion {
+        spec_version: 9370,
+        transaction_version: 20,
+    };


I reverted to just a struct so that the args (of the same type) were named

This is more specific, I like it

jsdw · 2024-04-09T10:22:24Z

codegen/src/lib.rs

+        if absolute_path(crate_path.clone()).is_err() {
+            // Throw an error here, because otherwise we end up with a harder to comprehend error when
+            // substitute types don't begin with an absolute path.
+            panic!("The provided crate path must be an absolute path, ie prefixed with '::' or 'crate'");
+        }


We shouldn't need to clone crate_path here but need to change scale-typegen to avoid.

jsdw · 2024-04-09T10:24:41Z

core/src/constants/address.rs

@@ -10,7 +12,7 @@ use derive_where::derive_where;

 /// This represents a constant address. Anything implementing this trait
 /// can be used to fetch constants.
-pub trait ConstantAddress {
+pub trait AddressT {


I standardised on all addresses/payloads being either AddressT (just looking something up; constants & storage) or PayloadT (submitting/calling something; TX's & runtime APIs)

perhaps modify the docs then above 👆

What do you think needs modifying? I think they are still relevant? :)

yepp, I misunderstood your comment

jsdw · 2024-04-09T10:25:21Z

core/src/constants/mod.rs

+///
+/// When the provided `address` is dynamic (and thus does not come with any expectation of the
+/// shape of the constant value), this just returns `Ok(())`
+pub fn validate<Address: AddressT>(metadata: &Metadata, address: &Address) -> Result<(), Error> {


instead of validate_constant etc, aim for constants::validate etc.

jsdw · 2024-04-09T10:38:28Z

core/src/metadata/metadata_type.rs

-}
-
-/// Some extension methods on [`subxt_metadata::Metadata`] that return Errors instead of Options.
-pub trait MetadataExt {


I didn't see a need for this offhand and just added the one missing function above on the concrete type instead.

jsdw · 2024-04-09T10:39:35Z

core/src/runtime_api/payload.rs

This was just moved from mod.rs; I should have done a git mv because after then changing mod.rs it didn't pick up the move.

jsdw · 2024-04-09T10:40:37Z

core/src/storage/utils.rs

-/// Outputs the [`storage_address_root_bytes`] as well as any additional bytes that represent
-/// a lookup in a storage map at that location.
-pub fn storage_address_bytes<Address: StorageAddress>(
-    addr: &Address,
-    metadata: &Metadata,
-) -> Result<Vec<u8>, Error> {
-    let mut bytes = Vec::new();
-    write_storage_address_root_bytes(addr, &mut bytes);
-    addr.append_entry_bytes(metadata, &mut bytes)?;
-    Ok(bytes)
-}
-
-/// Outputs a vector containing the bytes written by [`write_storage_address_root_bytes`].
-pub fn storage_address_root_bytes<Address: StorageAddress>(addr: &Address) -> Vec<u8> {
-    let mut bytes = Vec::new();
-    write_storage_address_root_bytes(addr, &mut bytes);
-    bytes
-}


moved to mod.rs as get_address_bytes and get_address_root_bytes

artifacts/demo_chain_specs/polkadot.json

artifacts/polkadot_metadata_small.scale

core/src/blocks/extrinsics.rs

niklasad1 · 2024-04-15T11:51:36Z

core/src/client.rs

-    pub fn genesis_hash(&self) -> C::Hash {
-        self.genesis_hash
-    }
+    /// Genesis hash.


why pub now?

I think is more in-line with RuntimeVersion, but either way is good for me

It's probably because we use it outside subxt-core but ideally we wouldn't make it pub. It's hard to add stuff without breaking the API...

I would probably add doc(hidden) to it to be on the safe-side

Yeah, I made runtime version things pub just because it felt better constructing it with named params (and it's basically just a named tuple). I think I followed the same logic here and just made it pub since it's just a tuple of 3 things you can fetch individually otherwise. I could go either way really, but at the time preferred the simplicity I guess of it just being a simple data struct :)

I don't think it should be doc(hidden) personally because it's a used as a param in a couple of places so people should be able to name it.

lexnv · 2024-04-15T11:58:43Z

core/src/tx/signer.rs

@@ -1,4 +1,4 @@
-// Copyright 2019-2023 Parity Technologies (UK) Ltd.
+// Copyright 2019-2024 Parity Technologies (UK) Ltd.


nit: Should we rename Singer -> SignerT for consistency with AddrT / PayloadT?
I've also seen it imported as SignerT for testing

Good question! There are probably a bunch of other traits here that we could, if we wanted, add T on the end of! In this PR I just standardised the "address/payload" traits to have a T but maybe we should go further! What do you reckon? (miight be better to do a full standarisation as a separate PR but I'd have to check hoiw many traits we have left to add the T too :D)

I had a look at the other traits we have and now I'm not sure which way is best! I liked the standardising here but also think that adding T onto everything feels a bit naff (and it's not what Rust does on any of the basic traits).

Lemme have a play; maybe I should undo the T thing and do something else..

(I'll sort this out in a small separate PR anyways :))

lexnv

LGTM! Nice one 🚀

Polkadot-Forum · 2024-05-01T07:32:05Z

This pull request has been mentioned on Polkadot Forum. There might be relevant details there:

https://forum.polkadot.network/t/paritytech-update-for-april/7646/1

jsdw added 2 commits March 28, 2024 12:15

Move Extrinsic decoding things to subxt_core and various tidy-ups

adf1f0d

A couple more fixes and fmt

5076651

jsdw changed the title ~~Jsdw subxt core polish~~ Polish subxt-core Mar 28, 2024

jsdw added 12 commits March 28, 2024 16:42

first pass moving tx logic to subxt_core

49268fd

cargo fmt

b571bb1

fix wasm example

6f54031

clippy

b9b63b4

more clippy

1ba85f5

WIP Adding examples and such

65345a7

Move storage functionality more fully to subxt_core and nice examples…

d22338a

… for storage and txs

Add example for events

0a33d38

consistify how addresses/payloads are exposed in subxt-core and add r…

b3c782b

…untime API fns

Add runtime API core example

4ef4585

fmt

a164951

remove scale-info patch

191237d

jsdw commented Apr 9, 2024

View reviewed changes

jsdw added 2 commits April 9, 2024 11:29

Add a little to the top level docs

e9e59d3

Merge branch 'master' into jsdw-subxt-core-polish

42fc42a

jsdw commented Apr 9, 2024

View reviewed changes

jsdw added 6 commits April 9, 2024 12:03

swap args around

e8a96ac

clippy

44b91e6

Merge branch 'master' into jsdw-subxt-core-polish

53b7d8e

cargo fmt and fix wasm-example

6e42e91

doc fixes

2d98704

no-std-ise new subxt-core additions

c2987f2

jsdw added 3 commits April 9, 2024 13:38

alloc, not core

8b8bfec

more no-std fixes

b518286

A couple more fixes

50db863

jsdw marked this pull request as ready for review April 9, 2024 16:16

jsdw requested a review from a team as a code owner April 9, 2024 16:16

jsdw changed the title ~~Polish subxt-core~~ Make subxt-core ready for publishing Apr 9, 2024

Merge branch 'master' into jsdw-subxt-core-polish

75eb502

niklasad1 reviewed Apr 15, 2024

View reviewed changes

artifacts/demo_chain_specs/polkadot.json Outdated Show resolved Hide resolved

niklasad1 reviewed Apr 15, 2024

View reviewed changes

artifacts/polkadot_metadata_small.scale Outdated Show resolved Hide resolved

niklasad1 reviewed Apr 15, 2024

View reviewed changes

core/src/blocks/extrinsics.rs Outdated Show resolved Hide resolved

niklasad1 reviewed Apr 15, 2024

View reviewed changes

lexnv reviewed Apr 15, 2024

View reviewed changes

lexnv approved these changes Apr 15, 2024

View reviewed changes

Add back extrinsic decode test

1c6b55f

niklasad1 approved these changes Apr 15, 2024

View reviewed changes

jsdw merged commit 1e111ea into master Apr 15, 2024
13 checks passed

jsdw deleted the jsdw-subxt-core-polish branch April 15, 2024 14:20

jsdw mentioned this pull request Apr 15, 2024

Create subxt-core crate; a no_std subset of subxt functionality? #1394

Closed

jsdw mentioned this pull request May 16, 2024

Prep for 0.36.0 release #1587

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Make subxt-core ready for publishing #1508

Make subxt-core ready for publishing #1508

jsdw commented Mar 28, 2024 •

edited

Loading

jsdw Apr 9, 2024

lexnv Apr 15, 2024

jsdw Apr 9, 2024

jsdw Apr 9, 2024

niklasad1 Apr 15, 2024

jsdw Apr 15, 2024

niklasad1 Apr 15, 2024

jsdw Apr 9, 2024

jsdw Apr 9, 2024

jsdw Apr 9, 2024 •

edited

Loading

jsdw Apr 9, 2024 •

edited

Loading

niklasad1 Apr 15, 2024

lexnv Apr 15, 2024

niklasad1 Apr 15, 2024

niklasad1 Apr 15, 2024

jsdw Apr 15, 2024

lexnv Apr 15, 2024

jsdw Apr 15, 2024

jsdw Apr 15, 2024

jsdw Apr 15, 2024

lexnv left a comment

Polkadot-Forum commented May 1, 2024

		@@ -1,4 +1,4 @@
		// Copyright 2019-2023 Parity Technologies (UK) Ltd.
		// Copyright 2019-2024 Parity Technologies (UK) Ltd.

Make subxt-core ready for publishing #1508

Make subxt-core ready for publishing #1508

Conversation

jsdw commented Mar 28, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jsdw Apr 9, 2024 • edited Loading

Choose a reason for hiding this comment

jsdw Apr 9, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

lexnv left a comment

Choose a reason for hiding this comment

Polkadot-Forum commented May 1, 2024

jsdw commented Mar 28, 2024 •

edited

Loading

jsdw Apr 9, 2024 •

edited

Loading

jsdw Apr 9, 2024 •

edited

Loading