###### tags: ``Team Admin`` # Language Design Calls: Rolling Agenda Please add topics that you want to discuss in our weekly Wednesday language design call with a link to an issue or forum post and your initials in []. PLEASE DO NOT DELETE PRIOR AGENDAS BUT ADD NEW ITEMS TO THE NEXT AGENDA OR COPY THEM IF YOU WANT TO RE-DISCUSS THEM. ## Next design call: Jannuary 2023 - inference of data locations together with mixed locations and tying locations to types - try...catch and decoding & extcodesize failures ### Backlog of topics to discuss - Schedule dedicated design calls for the upcoming larger tasks that are still under-defined: - Generics - owned by Daniel - Compile-time-constant expression evaluation - owned by Daniel - Better memory optimizations - owned by Chris, still in research phase, not yet in an explanatory stage, documented in hackMD here -> https://notes.ethereum.org/WnsIlf8RSASJFr8S7m5yvQ - Code data location / value type immutables - owned by Daniel - ...? - Decide whether it is worth adding type definitions to ABI and compiler I/O standard in solc-js (https://github.com/ethereum/solc-js/pull/630). And what would be the best way to maintain that (e.g., using a JSON schema). - Should we allow omitting location for function parameters in interfaces just like we decided to do for function pointers? - Do we want tangential features like [Yul formatter #13550](https://github.com/ethereum/solidity/issues/13550) or [Exposing evmasm::disassemble a "poor man's disassembler" #12753](https://github.com/ethereum/solidity/issues/12753) in the compiler? ### Agenda for 2022-11-23 - Picking up new tasks for people not assigned to any roadmap track: - Pick small tasks, that can you won't get stuck in for a long time. Ideally they should be finished the same week. Bug fixes are usually good candidates for that. - Remember to spend time on reviews (40-50% of your time). - Team discussion topic: what to do about holes in the documentation. - Decision: We'll be closing stale issues, including documentation issues. Documentation issues are very low priority. We'll address some of them but if we close an issue and it does not come up again, it wasn't important. ### Agenda for 2022-11-09 - Tooling promises: Import json AST via Standard JSON - [AST/asm import via Standard JSON #13690](https://github.com/ethereum/solidity/issues/13690) - Tooling promises: Export stack layout generator graphs. [need to check: how much libevmasm optimizer do we still need?] - We should create an issue. - Truffle requesting debugging data wrt internal function pointer values https://github.com/ethereum/solidity/issues/13425 - Named mappings, specifically nested named (or unamed) mappings PR #13384, Issue #11407 - The PR should have more tests (as already commented in it) - We should allow naming keys and values in nested mappings as well. Also it should be possible to name the value even when it's a mapping or an array. ### Discussion on 2022-11-07 - @GNSPS: Syntactic sugar for namespacing storage variables. > we were talking about the diamond pattern internally at one of our Show and Tells, today, at Dili, and we're really tired of having to deal with extremely bloated upgradeable storage patterns > and it occurred to me that some compiler-level namespacing for storage variables would solve 99% of the use cases > And this comment is exactly what Nick ended up using in the Diamond standard > https://github.com/ethereum/solidity/issues/8353#issuecomment-932474378 > that for some reason everybody is using now > but it is suuuuuper far from being readable > and the complexity to mental model code (as an auditor) that uses this extensively just adds up - Discussed with @frangio from OpenZeppelin. - We prefer a generic solution rather than special syntax promoting a specific use case. There were many solutions proposed in the issues already. We'd like someone to come forward with a concrete syntax proposal so that we can discuss it. @frangio will create a topic on the forum with such a proposal. ### Agenda for 2022-11-02 - [EIP 1153 (Transient Storage)](https://ethereum-magicians.org/t/eip-1153-transient-storage-opcodes/553) - opinions and should we react to it? - EOF (Ethereum Object Format): [info on EOF](https://twitter.com/teamipsilon/status/1565292837079597056?s=20&t=gEIFdOqNEsR5290tnhXmNA) - Daniel gives update from call with Ipsilon team - EOF concepts that influence Solidity the most - immediate static jumps - first-class citizen functions in the EVM - Can we build a prototype version in the next weeks? Or collect examples of function patterns that would be affected by those changes ### Agenda for 2022-10-26 ### Agenda for 2022-10-19 ### Agenda for 2022-10-12 _likely to be skipped due to Devcon_ ### Agenda for 2022-10-05 - gitcoin grants from 1inch - Stuff tagged `good first issue` may be too trivial. - In the long term we should have more issues labeled `outsourceable` ready at hand, but for now we'll need to hand-pick some. - These issues must have good descriptions and some pointers on how we expect them to be implemented. - We should coordinate on the Matrix channel. We'll propose some initial issues to work on. - Full "via IR" support in Hardhat and Truffle is reportedly blocked by "[IR-based compilation w/o optimization uses way too many stack slots, easily hits stack too deep errors #12533](https://github.com/ethereum/solidity/issues/12533)". - It's now possible to have a completely empty optimizer sequence. We should suggest that in the issue. - What's Hardhat's problem with optimized code? Heuristics. They can't be dropped without better debug info. - Issue with prioritizing this work: it's not more important than other roadmap issues but it's still important and the topic should be moving forward in the background as we work on other stuff. - For now we'll suggest an empty optimizer sequence and try to talk with them at Devcon. - OpenZeppelin is proposing the removal of the SPDX license warnings: "[Remove warning about missing license header #13598](https://github.com/ethereum/solidity/issues/13598)". - We'll ask people for more feedback on Twitter. - Anton's take: the warning itself does not seem to be a huge issue, but it's also not necessarily the job of the compiler to enforce this. It might not really be helpful to user's in practice. Getting this warning on its own won't make them understand licensing problems when they reuse others' code. - Do we want to accept [Enable the OpenSSF Scorecard Github Action #661](https://github.com/ethereum/solc-js/issues/661) in solc-js? Let's review the PR and give it a try. - No objections from the team. It's fully open-source so why not. - [feat: introduce block.prevrandao as alias for block.difficulty #13512](https://github.com/ethereum/solidity/issues/13512) ### Agenda for 2022-09-28 ### Agenda for 2022-09-21 ### Agenda for 2022-09-14 - Daniel: paris and prevrandao (https://github.com/ethereum/solidity/issues/13517 and https://github.com/ethereum/solidity/issues/13512) - just break it or keep things explicitly non-breaking even with the new evm-version setting? - Decision: we'll keep things compatible and support both `difficulty` and `prevrandao`. We'll output one or the other depending on the EVM version. ### Agenda for 2022-09-07 - Daniel (in absentio): https://github.com/ethereum/solidity/issues/13180 - some opinions on whether or not to take this more seriously than my suggestion would be nice. - would be nice to get some eyes on it, does not need to be discussed in the call - @aarlt: Literal suffixes vs postfix operators. - https://github.com/ethereum/solidity/pull/12362#discussion_r953938583 - https://github.com/ethereum/solidity/pull/12656/files - using { _kilometers as km(suffix), _meters as m(suffix), } for Meters; - [Functions taking `calldata` arguments are not assignable to function pointers of the same type #12778](https://github.com/ethereum/solidity/issues/12778): should we disallow such pointers or warn about them? Maybe adjust syntax so that they do not have locations? - We'll make specifying location in external function pointers optional. - In a breaking release we will disallow function pointers with calldata arguments. - We'll discuss doing the same for interfaces separately. ### Agenda for 2022-08-31 - Clang format (leo) - Daniel (in absentio): https://github.com/ethereum/solidity/pull/13378 - maybe it's not the worst thing in the world to reconsider our position of ever doing stuff like this. - Chris suggested to only enable this if optimizer is turned on. - Reentrancy guards by default: https://github.com/ethereum/solidity/issues/12996 ### Agenda for 2022-08-17 - Highlighting style for the docs - We agreed that we want the following general characteristics: higher contrast, more diverse colors, not having different syntactic elements colored the same way, prominent use of bold font. - The red color used by Github's theme is not great. - Our current theme is closer than Github's theme to what we want so it's a better starting point. - [feat(parsing): allow trailing commas in some places #13246](https://github.com/ethereum/solidity/pull/13246) - No one on the team is really against it, though many are indifferent. - We don't have an answer to the tuple ambiguity yet. - Proper branch protection for solc-bin/gh-pages - @franzihei will ask devops team to do it in solc-bin (and solc-js) ### Agenda for 2022-08-10 - Daniel: https://github.com/ethereum/solidity/issues/13348 [preliminary decision, but also clarify how best to triage such cases] - We'd actually consider changing the memory layout (the current one does not allow slicing, is wasteful, dynamic and static arrays are different). - This particular issue is not well specified and not good for discussing these changes. Let's close it. - [Option to disable metadata #13233](https://github.com/ethereum/solidity/issues/13233): naming of JSON and CBOR metadata. - `--no-cbor-metadata` and `settings.metadata.cborMetadata` is ok and we do not have better ideas. - We should still run this by @axic and make sure he has no objections. - Daniel: https://github.com/ethereum/solidity/issues/13308 [it's known, but maybe we should really do something more for it in the short term; and if not that at least improve docs] - [Functions taking `calldata` arguments are not assignable to function pointers of the same type #12778](https://github.com/ethereum/solidity/issues/12778): should we disallow such pointers or warn about them? Maybe adjust syntax so that they do not have locations? - No decision. We'll discuss it again next time. - https://github.com/ethereum/solidity/issues/11407 - The author reached out asking if he could start building it. - Decision: proceed. Asked the person to start implementing it. - add all team members to the solidity team on github ### Agenda for 2022-08-03 - `abi.encode` for custom errors: https://github.com/ethereum/solidity/issues/13339 - Decision: Let's implement `abi.encodeError()`. - Quick decision on merging https://github.com/ethereum/solidity/pull/13309 - We should do it but just in case reach out to Etherscan and auditing tools first. - Quick decision on closing https://github.com/ethereum/solidity/issues/13214 and https://github.com/ethereum/solidity/pull/13334 (or merging the latter). - Decision: close it, we don't want it to warn. - Fate of https://github.com/ethereum/solidity/pull/13168 - [Solidity 2.0](https://hackmd.io/@axic/r1KtcV0dq) discussion ### Agenda for 2022-07-13 - Decide whether to remove nobody or the entire team or the release author from the "thank all the contributors" list on release pages. - example: https://github.com/FuelLabs/sway/releases/tag/v0.17.0 - Should we deprecate the gas estimator? (discussion in [#8920](https://github.com/ethereum/solidity/issues/8920)) - Decision: remove it in 0.9.0, without doing deprecation warnings first. - [Turn on the optimizer by default #13223](https://github.com/ethereum/solidity/issues/13223) - Decision: We should do it in 0.9.0, with same default for both pipelines for consistency. - [Explicit exports](https://github.com/ethereum/solidity/issues/6482#issuecomment-659044989) - [Hackmd for conversation](https://notes.ethereum.org/f4E9L4uPT9SQE48zV7rhzQ) - [Allow constants in interfaces](https://github.com/ethereum/solidity/issues/8775) ### Agenda for 2022-06-15 - Project board for solc-js. ### Agenda for 2022-06-08 - Discuss bug reports: https://notes.ethereum.org/MkJN5aILRHWs6-lWWhLoWw ### Agenda for 2022-06-01 - Reverting the part of ["Adding event and error selector fields #12921"](https://github.com/ethereum/solidity/pull/12921) that allows accessing events from other contracts. Truffle reports that it will be problematic for them to release this without ["Export all events #10996"](https://github.com/ethereum/solidity/pull/10996). ### Agenda for 2022-05-11 - ABIv3: https://github.com/vyperlang/vyper/issues/2542 If we want to weigh in on this, we should probably do so soon - at devconnect they said, they might just go ahead and implement somthing like this... - Fe's plan to move away from Yul in the medium- to long-term and use a different intermediate language called Sonatina (https://github.com/fe-lang/sonatina) - Print: https://github.com/ethereum/solidity/issues/12995 ### Agenda for 2022-05-04 ### Agenda for 2022-04-13 - `bool` to `uint` conversions: https://github.com/ethereum/solidity/issues/12930 ### Agenda for 2022-03-30 - Do not avoid overflow in overflow checks [issue](https://github.com/ethereum/solidity/issues/12814) ### Agenda for 2022-03-30 - [HM] What should be pure: https://github.com/ethereum/solidity/pull/12861 ### Agenda for 2022-03-23 - [CR] Try/catch: [issue](https://github.com/ethereum/solidity/issues/12725) - instead of reverting, enter catch clause for failed extcodesize check (breaking) - [CR] Do not avoid overflow in overflow checks [issue](https://github.com/ethereum/solidity/issues/12814)