Suggestion Backlog Slog, 8/13/2018

[RyanCavanaugh]

• Enum/number assignability review
• any vs unknown on DT
• Proposal: Overload Function.bind for when no argArray is provided. #22669 Add a new overload to Function#bind
• tsconfig cannot be specified along with an explicit list of files to compile - TS5042 #22649 Combine tsconfig with file list
• Proposal: better type narrowing in overloaded functions through overload set pruning #22609 Do some type guards based on overload when a good discriminant is present
• Provide a back-compat mechanism for bundled .d.ts files in package.json #22605 Provide a back-compat mechanism for bundled .d.ts files in package.json
• Mapped types enumerating keys in string behave poorly #22509 Something something mapped types
• Allow bitwise test on undefined #22230 Allow implicit undefined->0 coercion under strictNullChecks for bitwise ops
• foo.ts is resolved before foo.d.ts even if the latter is in files[] #22208 Prefer .d.ts over .ts if it's in files array
• Add a strict flag forbidding impossible generic declarations #26129 Forbid illogical generic declarations
• typeof union.membername errors. #22598 Allow speculative union type property access in typeof (expression) operands

[RyanCavanaugh]

• Enum/number assignability review
  • Keep getting duplicates on this
  • Constant source of surprise when people notice
  • Back compat constraints limit our options
  • A commandline flag really is not what you want
  • Open a longer issue explaining the state of things
• any vs unknown on DT
  • Impending disaster as people argue over what return types on DT should be
  • It's not up to randos on DT to decide your strictness policy
  • Ideally unknown would have acted as any under non-strict
    • Oops
  • Spitball: mixed type that is unknown in strict and any otherwise
    • Ugh
  • Anyone who would want strictUnknown probably actually wants strictAny
  • Revive strictAny and try to work through its problems again
  • Do not accept DT PRs for replacing any in output positions
  • unknown in input positions is OK
  • Bivariant positions: 🤷‍♂️
• Proposal: Overload Function.bind for when no argArray is provided. #22669 Add a new overload to Function#bind
  • Sure let's try it
• tsconfig cannot be specified along with an explicit list of files to compile - TS5042 #22649 Combine tsconfig with file list
  • -1
• Proposal: better type narrowing in overloaded functions through overload set pruning #22609 Do some type guards based on overload when a good discriminant is present
  • Recommend using conditional types in lieu of overloads whenever possible
• Provide a back-compat mechanism for bundled .d.ts files in package.json #22605 Provide a back-compat mechanism for bundled .d.ts files in package.json
  • Ryan to write proposal
• Mapped types enumerating keys in string behave poorly #22509 Something something mapped types
  • ?
• foo.ts is resolved before foo.d.ts even if the latter is in files[] #22208 Prefer .d.ts over .ts if it's in files array
  • +1
• Add a strict flag forbidding impossible generic declarations #26129 Forbid illogical generic declarations
  • -1 missed the boat and can't break the DT world

[simonbuchan]

It's not up to randos on DT to decide your strictness policy

As one of those randos, this is a continuous concern of mine! It would be good if DT had a fairly complete style guide about things along these lines, but that's a different issue (specifically DefinitelyTyped/DefinitelyTyped#2001 which lead to the now quite out of date and incomplete http://definitelytyped.org/guides/best-practices.html)