Object statistics lru

[sharpsalt]

No description provided.

[Copilot]

Pull request overview

This PR introduces Redis-like object “type + encoding” metadata to stored values and starts using it to implement an INCR command.

Changes:

• Add type/encoding helpers (getType, getEncoding, assertType, assertEncoding) and a string encoding deduction helper.
• Extend object creation to include a packed TypeEncoding field.
• Add an evalINCR handler and integrate type/encoding deduction into SET.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 8 comments.

Show a summary per file

File	Description
core/typeencoding.go	Adds bitwise helpers for extracting/asserting object type and encoding.
core/type_string.go	Adds logic to deduce string encoding (int/embstr/raw).
core/store.go	Updates NewObj to accept and store packed type+encoding.
core/object.go	Introduces Obj with TypeEncoding and defines type/encoding constants.
core/eval.go	Adds INCR, updates SET to store encoding, wires command dispatch.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

NewObj sets TypeEncoding, but Obj as defined in this file does not have a TypeEncoding field, so this won’t compile. Either add TypeEncoding uint8 to the Obj struct here, or (preferably) remove this duplicate Obj definition and use a single Obj type shared with core/object.go.

[Copilot]

Obj is defined here, but core/store.go also defines an Obj type in the same core package. Having two type Obj struct { ... } declarations in one package will not compile; consolidate into a single Obj definition (and update references accordingly).

Suggested change

	type Obj struct{
	TypeEncoding uint8
	Value interface{}
	ExpiresAt int64
	}
	// Obj is defined in core/store.go for the core package.

[Copilot]

OBJ_ENCODING_ENDSTR looks like a typo and doesn’t match usage elsewhere (OBJ_ENCODING_EMBSTR in deduceTypeEncoding). As written, deduceTypeEncoding won’t compile; rename/align the constant (and its value) so the encoding name is consistent across the package.

Suggested change

	var OBJ_ENCODING_ENDSTR uint8=8
	var OBJ_ENCODING_EMBSTR uint8=8

[Copilot]

This ParseInt check won’t compile: it declares rr but then checks err (which is undefined), and rr is unused. Use the standard if _, err := strconv.ParseInt(...); err == nil { ... } pattern.

Suggested change

	if _,rr:=strconv.ParseInt(v,10,64);err==nil{
	if _, err := strconv.ParseInt(v, 10, 64); err == nil {

[Copilot]

OBJ_ENCODING_EMBSTR is referenced here but no such identifier exists in the package (the constants in core/object.go define OBJ_ENCODING_ENDSTR instead). Define OBJ_ENCODING_EMBSTR or rename usages/constants so they match.

Suggested change

	if _,rr:=strconv.ParseInt(v,10,64);err==nil{
	return oType,OBJ_ENCODING_INT //we are converting ans stroing int as a string inside my redis object
	}
	if len(v)<=44{
	return oType,OBJ_ENCODING_EMBSTR //Object encoding is embedded string
	if _,err:=strconv.ParseInt(v,10,64);err==nil{
	return oType,OBJ_ENCODING_INT //we are converting ans stroing int as a string inside my redis object
	}
	if len(v)<=44{
	return oType,OBJ_ENCODING_ENDSTR //Object encoding is embedded string

[Copilot]

strconv.ParseInt(obj.Value,(string),10,64) is invalid Go and won’t compile; obj.Value is interface{} so you need a type assertion/conversion to string before parsing (and handle the parse error, since stored values can still be non-numeric).

Suggested change

	i,_:=strconv.ParseInt(obj.Value,(string),10,64)
	value,ok:=obj.Value.(string)
	if !ok{
	return Encode(errors.New("ERR value is not an integer or out of range"),false)
	}
	i,err:=strconv.ParseInt(value,10,64)
	if err!=nil{
	return Encode(errors.New("ERR value is not an integer or out of range"),false)
	}

[Copilot]

The INCR command dispatch is wired to evalPING here, so INCR will behave like PING. This should call evalINCR(cmd.Args) instead.

Suggested change

	buf.Write(evalPING(cmd.Args))
	buf.Write(evalINCR(cmd.Args))

[Copilot]

The PR title mentions "Object statistics lru", but this change introduces type/encoding bits and an INCR command; the only LRU-related change is a TODO. Consider updating the PR title (or adding the intended LRU/statistics implementation) so the PR intent matches the code.