simplfy compiler patterns #5425
Merged
+387
−516
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This commit simplifies the compiler so the code paths are easier to understand and the API is easier to use. compiler.Job goes away as does the variations implementation of the runtime.Compiler. runtime.Compiler now just has one query method for both local and lake cases. We also moved compiler/data.Source to runtime/exec.Environment as that much more accurately potrays the role of this struct as representing the execution environment (lake or local) of the query. Down the road, we may want to store config options here (e.g., parallelism, strict compilation, etc). We also added the flag -files to "super compile" to ask it to compile the query assuming there are command-line input files.
nwt
reviewed
Nov 3, 2024
cmd/super/compile/shared.go
Outdated
| @@ -62,11 +61,16 @@ func (s *Shared) Run(ctx context.Context, args []string, lakeFlags *lakeflags.Fl | |||
| if len(args) == 1 { | |||
| query = args[0] | |||
| } | |||
| ast, err := parser.ParseQuery(query, s.includes...) | |||
| rctx := runtime.DefaultContext() | |||
| env := exec.NewEnvironment(nil, lk) | |||
There was a problem hiding this comment.
Nit: Declare this closer to its use.
nwt
reviewed
Nov 3, 2024
compiler/package.go
Outdated
Comment on lines
28
to
29
| func Analyze(rctx *runtime.Context, ast *parser.AST, env *exec.Environment, extInput bool) (dag.Seq, error) { | ||
| return semantic.Analyze(rctx.Context, ast, env, extInput) |
There was a problem hiding this comment.
Nit:
Suggested change
| func Analyze(rctx *runtime.Context, ast *parser.AST, env *exec.Environment, extInput bool) (dag.Seq, error) { | |
| return semantic.Analyze(rctx.Context, ast, env, extInput) | |
| func Analyze(ctx context.Context, ast *parser.AST, env *exec.Environment, extInput bool) (dag.Seq, error) { | |
| return semantic.Analyze(ctx, ast, env, extInput) |
compiler/package.go
Outdated
Comment on lines
32
to
35
| func Optimize(rctx *runtime.Context, seq dag.Seq, env *exec.Environment, parallel int) (dag.Seq, error) { | ||
| // Call optimize to possible push down a filter predicate into the | ||
| // kernel.Reader so that the zng scanner can do boyer-moore. | ||
| o := optimizer.New(rctx.Context, env) |
There was a problem hiding this comment.
Nit:
Suggested change
| func Optimize(rctx *runtime.Context, seq dag.Seq, env *exec.Environment, parallel int) (dag.Seq, error) { | |
| // Call optimize to possible push down a filter predicate into the | |
| // kernel.Reader so that the zng scanner can do boyer-moore. | |
| o := optimizer.New(rctx.Context, env) | |
| func Optimize(ctx context.Context, seq dag.Seq, env *exec.Environment, parallel int) (dag.Seq, error) { | |
| // Call optimize to possible push down a filter predicate into the | |
| // kernel.Reader so that the zng scanner can do boyer-moore. | |
| o := optimizer.New(ctx, env) |
nwt
approved these changes
Nov 3, 2024
compiler/package.go
Outdated
Comment on lines
36
to
37
| var err error | ||
| seq, err = o.Optimize(seq) |
There was a problem hiding this comment.
Nit:
Suggested change
| var err error | |
| seq, err = o.Optimize(seq) | |
| seq, err := o.Optimize(seq) |
compiler/package.go
Outdated
| if err != nil { | ||
| return nil, err | ||
| } | ||
| seq, err = o.Vectorize(seq) // XXX why is this here? |
There was a problem hiding this comment.
This is how we engage the vector runtime on lake queries when CSUP is available.
compiler/semantic/analyzer.go
Outdated
| ) | ||
|
|
||
| // Analyze performs a semantic analysis of the AST, translating it from AST | ||
| // to DAG form, resolving syntax ambiguities, and performing constant propagation. | ||
| // After semantic analysis, the DAG is ready for either optimization or compilation. | ||
| func Analyze(ctx context.Context, ast *parser.AST, source *data.Source, extInput bool) (dag.Seq, error) { | ||
| func Analyze(ctx context.Context, ast *parser.AST, source *exec.Environment, extInput bool) (dag.Seq, error) { |
There was a problem hiding this comment.
Nit:
Suggested change
| func Analyze(ctx context.Context, ast *parser.AST, source *exec.Environment, extInput bool) (dag.Seq, error) { | |
| func Analyze(ctx context.Context, ast *parser.AST, env *exec.Environment, extInput bool) (dag.Seq, error) { |
runtime/exec/environment.go
Outdated
| engine: engine, | ||
| lake: lake, | ||
| } | ||
| } | ||
|
|
||
| func (s *Source) IsLake() bool { | ||
| func (s *Environment) IsLake() bool { |
There was a problem hiding this comment.
Nit: Update receiver name on all these methods.
Suggested change
| func (s *Environment) IsLake() bool { | |
| func (e *Environment) IsLake() bool { |
zed_test.go
Outdated
| @@ -143,7 +144,7 @@ func runOneBoomerang(t *testing.T, format, data string) { | |||
| ast, err := parser.ParseQuery("fuse") | |||
| require.NoError(t, err) | |||
| rctx := runtime.NewContext(context.Background(), zctx) | |||
| q, err := compiler.NewCompiler().NewQuery(rctx, ast, []zio.Reader{dataReadCloser}) | |||
| q, err := compiler.NewCompiler(storage.NewLocalEngine()).NewQuery(rctx, ast, []zio.Reader{dataReadCloser}, 0) | |||
There was a problem hiding this comment.
Nit: This path doesn't use the engine.
Suggested change
| q, err := compiler.NewCompiler(storage.NewLocalEngine()).NewQuery(rctx, ast, []zio.Reader{dataReadCloser}, 0) | |
| q, err := compiler.NewCompiler(nil).NewQuery(rctx, ast, []zio.Reader{dataReadCloser}, 0) |
ztest/ztest.go
Outdated
| @@ -523,7 +524,7 @@ func runzq(path, zedProgram, input string, outputFlags []string, inputFlags []st | |||
| if err != nil { | |||
| return "", err | |||
| } | |||
| q, err := runtime.CompileQuery(context.Background(), zctx, compiler.NewCompiler(), ast, []zio.Reader{zrc}) | |||
| q, err := runtime.CompileQuery(context.Background(), zctx, compiler.NewCompiler(storage.NewLocalEngine()), ast, []zio.Reader{zrc}) | |||
There was a problem hiding this comment.
Nit: This path doesn't use the engine.
Suggested change
| q, err := runtime.CompileQuery(context.Background(), zctx, compiler.NewCompiler(storage.NewLocalEngine()), ast, []zio.Reader{zrc}) | |
| q, err := runtime.CompileQuery(context.Background(), zctx, compiler.NewCompiler(nil), ast, []zio.Reader{zrc}) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This commit simplifies the compiler so the code paths are easier to understand and the API is easier to use. compiler.Job goes away as does the variations implementation of the runtime.Compiler. runtime.Compiler now just has one query method for both local and lake cases. We also moved compiler/data.Source to runtime/exec.Environment as that much more accurately potrays the role of this struct as representing the execution environment (lake or local) of the query. Down the road, we may want to store config options here (e.g., parallelism, strict compilation, etc). We also added the flag -files to "super compile" to ask it to compile the query assuming there are command-line input files.