Improve external types and JSON codec

[turbolent]

Closes #2167

Description

• Remove unnecessary link decoding from JSON-CDC codec. Only encoding is necessary
• Add support for account links to cadence (add cadence.AccountLink and encoding in JSON-CDC)

Also:

• Remove unnecessary/old type ID from external function type and restricted type
  (See https://dapperlabs.slack.com/archives/C034CHNNRFD/p1679610220252009. Noticed this in the new CCF codec and realized it is there for historic reasons: We used to only have the ID, but added parameters and return type later)
• Improve the way type IDs are generated in sema (for e.g. reference types, capability types, function types) and use those functions from cadence, instead of duplicating the ID generation code (and introducing a possibility for diverging logic, as e.g. attempted in 7a1b3f5 (#2364) and 54c6987#diff-b990161986c40e0867e71c60779bc98a8730a9eb0ca5866ae8189e126863c059R1736-R1751)
• Include missing type parameters in function type

---

• Targeted PR against master branch
• Linked to Github issue with discussion and accepted design OR link to spec that describes this work
• Code follows the standards mentioned here
• Updated relevant documentation
• Re-reviewed Files changed in the Github PR explorer
• Added appropriate labels

[github-actions]

Cadence Benchstat comparison

This branch with compared with the base branch onflow:master commit 2a3eac6
The command for i in {1..N}; do go test ./... -run=XXX -bench=. -benchmem -shuffle=on; done was used.
Bench tests were run a total of 7 times on each branch.

Collapsed results for better readability

	old.txt	new.txt
	time/op		delta
CheckContractInterfaceFungibleTokenConformance-2	113µs ± 0%	111µs ± 0%	~	(p=1.000 n=1+1)
ContractInterfaceFungibleToken-2	38.8µs ± 0%	38.5µs ± 0%	~	(p=1.000 n=1+1)
ExportType/composite_type-2	323ns ± 0%	472ns ± 0%	~	(p=1.000 n=1+1)
ExportType/simple_type-2	52.8ns ± 0%	55.9ns ± 0%	~	(p=1.000 n=1+1)
InterpretRecursionFib-2	2.66ms ± 0%	2.50ms ± 0%	~	(p=1.000 n=1+1)
NewInterpreter/new_interpreter-2	1.09µs ± 0%	1.09µs ± 0%	~	(p=1.000 n=1+1)
NewInterpreter/new_sub-interpreter-2	576ns ± 0%	571ns ± 0%	~	(p=1.000 n=1+1)
ParseArray-2	7.74ms ± 0%	8.24ms ± 0%	~	(p=1.000 n=1+1)
ParseDeploy/byte_array-2	12.5ms ± 0%	11.2ms ± 0%	~	(p=1.000 n=1+1)
ParseDeploy/decode_hex-2	1.20ms ± 0%	1.19ms ± 0%	~	(p=1.000 n=1+1)
ParseFungibleToken/With_memory_metering-2	180µs ± 0%	183µs ± 0%	~	(p=1.000 n=1+1)
ParseFungibleToken/Without_memory_metering-2	141µs ± 0%	142µs ± 0%	~	(p=1.000 n=1+1)
ParseInfix-2	7.01µs ± 0%	7.00µs ± 0%	~	(p=1.000 n=1+1)
QualifiedIdentifierCreation/One_level-2	2.48ns ± 0%	2.41ns ± 0%	~	(p=1.000 n=1+1)
QualifiedIdentifierCreation/Three_levels-2	116ns ± 0%	115ns ± 0%	~	(p=1.000 n=1+1)
RuntimeResourceDictionaryValues-2	5.14ms ± 0%	5.12ms ± 0%	~	(p=1.000 n=1+1)
RuntimeScriptNoop-2	4.28µs ± 0%	7.43µs ± 0%	~	(p=1.000 n=1+1)
SuperTypeInference/arrays-2	297ns ± 0%	296ns ± 0%	~	(p=1.000 n=1+1)
SuperTypeInference/composites-2	132ns ± 0%	133ns ± 0%	~	(p=1.000 n=1+1)
SuperTypeInference/integers-2	97.0ns ± 0%	95.6ns ± 0%	~	(p=1.000 n=1+1)
ValueIsSubtypeOfSemaType-2	88.6ns ± 0%	93.9ns ± 0%	~	(p=1.000 n=1+1)
	alloc/op		delta
CheckContractInterfaceFungibleTokenConformance-2	48.9kB ± 0%	48.9kB ± 0%	~	(all equal)
ContractInterfaceFungibleToken-2	23.2kB ± 0%	23.2kB ± 0%	~	(all equal)
ExportType/composite_type-2	136B ± 0%	136B ± 0%	~	(all equal)
ExportType/simple_type-2	0.00B	0.00B	~	(all equal)
InterpretRecursionFib-2	1.00MB ± 0%	1.00MB ± 0%	~	(all equal)
NewInterpreter/new_interpreter-2	768B ± 0%	768B ± 0%	~	(all equal)
NewInterpreter/new_sub-interpreter-2	200B ± 0%	200B ± 0%	~	(all equal)
ParseArray-2	2.72MB ± 0%	2.65MB ± 0%	~	(p=1.000 n=1+1)
ParseDeploy/byte_array-2	4.09MB ± 0%	4.09MB ± 0%	~	(p=1.000 n=1+1)
ParseDeploy/decode_hex-2	214kB ± 0%	214kB ± 0%	~	(p=1.000 n=1+1)
ParseFungibleToken/With_memory_metering-2	28.9kB ± 0%	28.9kB ± 0%	~	(all equal)
ParseFungibleToken/Without_memory_metering-2	28.9kB ± 0%	28.9kB ± 0%	~	(all equal)
ParseInfix-2	1.92kB ± 0%	1.92kB ± 0%	~	(all equal)
QualifiedIdentifierCreation/One_level-2	0.00B	0.00B	~	(all equal)
QualifiedIdentifierCreation/Three_levels-2	64.0B ± 0%	64.0B ± 0%	~	(all equal)
RuntimeResourceDictionaryValues-2	2.28MB ± 0%	2.28MB ± 0%	~	(p=1.000 n=1+1)
RuntimeScriptNoop-2	2.70kB ± 0%	2.70kB ± 0%	~	(all equal)
SuperTypeInference/arrays-2	96.0B ± 0%	96.0B ± 0%	~	(all equal)
SuperTypeInference/composites-2	0.00B	0.00B	~	(all equal)
SuperTypeInference/integers-2	0.00B	0.00B	~	(all equal)
ValueIsSubtypeOfSemaType-2	48.0B ± 0%	48.0B ± 0%	~	(all equal)
	allocs/op		delta
CheckContractInterfaceFungibleTokenConformance-2	806 ± 0%	806 ± 0%	~	(all equal)
ContractInterfaceFungibleToken-2	370 ± 0%	370 ± 0%	~	(all equal)
ExportType/composite_type-2	3.00 ± 0%	3.00 ± 0%	~	(all equal)
ExportType/simple_type-2	0.00	0.00	~	(all equal)
InterpretRecursionFib-2	18.9k ± 0%	18.9k ± 0%	~	(all equal)
NewInterpreter/new_interpreter-2	13.0 ± 0%	13.0 ± 0%	~	(all equal)
NewInterpreter/new_sub-interpreter-2	4.00 ± 0%	4.00 ± 0%	~	(all equal)
ParseArray-2	59.6k ± 0%	59.6k ± 0%	~	(p=1.000 n=1+1)
ParseDeploy/byte_array-2	89.4k ± 0%	89.4k ± 0%	~	(p=1.000 n=1+1)
ParseDeploy/decode_hex-2	63.0 ± 0%	63.0 ± 0%	~	(all equal)
ParseFungibleToken/With_memory_metering-2	768 ± 0%	768 ± 0%	~	(all equal)
ParseFungibleToken/Without_memory_metering-2	768 ± 0%	768 ± 0%	~	(all equal)
ParseInfix-2	48.0 ± 0%	48.0 ± 0%	~	(all equal)
QualifiedIdentifierCreation/One_level-2	0.00	0.00	~	(all equal)
QualifiedIdentifierCreation/Three_levels-2	2.00 ± 0%	2.00 ± 0%	~	(all equal)
RuntimeResourceDictionaryValues-2	36.9k ± 0%	36.9k ± 0%	~	(p=1.000 n=1+1)
RuntimeScriptNoop-2	43.0 ± 0%	43.0 ± 0%	~	(all equal)
SuperTypeInference/arrays-2	3.00 ± 0%	3.00 ± 0%	~	(all equal)
SuperTypeInference/composites-2	0.00	0.00	~	(all equal)
SuperTypeInference/integers-2	0.00	0.00	~	(all equal)
ValueIsSubtypeOfSemaType-2	1.00 ± 0%	1.00 ± 0%	~	(all equal)

[codecov]

Codecov Report

Merging #2401 (3f90c88) into master (27e7705) will increase coverage by 0.05%.
The diff coverage is 91.52%.

❗ Current head 3f90c88 differs from pull request most recent head 09eea5d. Consider uploading reports for the commit 09eea5d to get more accurate results

@@            Coverage Diff             @@
##           master    #2401      +/-   ##
==========================================
+ Coverage   78.55%   78.60%   +0.05%     
==========================================
  Files         316      316              
  Lines       68596    68848     +252     
==========================================
+ Hits        53885    54118     +233     
- Misses      12911    12928      +17     
- Partials     1800     1802       +2     

Flag	Coverage Δ
unittests	78.60% <91.52%> (+0.05%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
runtime/common/memorykind_string.go	40.00% <ø> (ø)
runtime/common/metering.go	92.32% <ø> (ø)
runtime/sema/resources.go	88.44% <0.00%> (ø)
runtime/ast/expression.go	92.46% <50.00%> (ø)
runtime/convertTypes.go	87.45% <69.23%> (-2.18%)	⬇️
values.go	75.31% <71.42%> (-0.08%)	⬇️
runtime/convertValues.go	82.94% <87.50%> (+0.19%)	⬆️
runtime/sema/type.go	89.80% <91.66%> (-0.02%)	⬇️
encoding/json/decode.go	88.64% <100.00%> (+0.21%)	⬆️
encoding/json/encode.go	94.72% <100.00%> (+0.20%)	⬆️
... and 1 more

... and 9 files with indirect coverage changes

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[fxamacker]

Looks good! I mostly focused on external types, values, and JSON-CDC encoding.

Only have one question and a few very minor nits.

[fxamacker]

@turbolent To confirm my understanding of this change's impact on CCF specs and CCF codec #2364:

• CCF doesn't need to add support for AccountLink and PathLink values.
• CCF needs to add encoding for type parameters in function type.

Is this correct?

[turbolent]

@fxamacker

To confirm my understanding of this change's impact on CCF specs and CCF codec #2364:

* CCF doesn't need to add support for `AccountLink` and `PathLink` values.

* CCF needs to add encoding for type parameters in function type.

Is this correct?

• Yes, CCF does not need support for links for now. Encoding of links to JSON is only used by the Playground for now, but will eventually be refactored, so I don' think we'll need support for them in CCF. We can add it later if needed.

• Yes, we should maybe encode type parameters for function types in CCF from the get-go