fix, Linting problems

cmd/eval-dev-quality/cmd/evaluate.go

@@ -135,7 +135,9 @@ func (command *Evaluate) Initialize(args []string) (evaluationContext *evaluate.
 		if err != nil {
 			command.logger.Panicf("ERROR: %s", err)
 		}
-		configurationFile.Close()
+		if err := configurationFile.Close(); err != nil {
+			panic(err)
+		}
 
 		command.Models = configuration.Models.Selected
 		command.Repositories = configuration.Repositories.Selected
@@ -360,9 +362,7 @@ func (command *Evaluate) Initialize(args []string) (evaluationContext *evaluate.
 			sort.Strings(command.Repositories)
 		}
 		evaluationContext.RepositoryPaths = command.Repositories
-		for _, r := range command.Repositories {
-			evaluationConfiguration.Repositories.Selected = append(evaluationConfiguration.Repositories.Selected, r)
-		}
+		evaluationConfiguration.Repositories.Selected = append(evaluationConfiguration.Repositories.Selected, command.Repositories...)
 	}
 
 	// Make the resolved selected languages available in the command.
@@ -494,7 +494,11 @@ func (command *Evaluate) Execute(args []string) (err error) {
 	if err != nil {
 		command.logger.Panicf("ERROR: cannot create configuration file: %s", err)
 	}
-	defer configurationFile.Close()
+	defer func() {
+		if err := configurationFile.Close(); err != nil {
+			panic(err)
+		}
+	}()
 	if err := evaluationConfiguration.Write(configurationFile); err != nil {
 		command.logger.Panicf("ERROR: %s", err)
 	}
@@ -773,7 +777,9 @@ func (command *Evaluate) evaluateKubernetes(ctx *evaluate.Context) (err error) {
 
 		parallel.Execute(func() {
 			var tmplData bytes.Buffer
-			jobTmpl.Execute(&tmplData, data)
+			if err := jobTmpl.Execute(&tmplData, data); err != nil {
+				panic(err)
+			}
 
 			commandOutput, err := util.CommandWithResult(context.Background(), command.logger, &util.Command{
 				Command: kubeCommand,
@@ -836,7 +842,9 @@ func (command *Evaluate) evaluateKubernetes(ctx *evaluate.Context) (err error) {
 		}
 
 		var tmplData bytes.Buffer
-		storageTmpl.Execute(&tmplData, data)
+		if err := storageTmpl.Execute(&tmplData, data); err != nil {
+			panic(err)
+		}
 
 		// Create the storage access pod.
 		output, err := util.CommandWithResult(context.Background(), command.logger, &util.Command{

cmd/eval-dev-quality/cmd/report.go

@@ -47,7 +47,11 @@ func (command *Report) Execute(args []string) (err error) {
 	if evaluationCSVFile, err = os.OpenFile(filepath.Join(command.ResultPath, "evaluation.csv"), os.O_CREATE|os.O_EXCL|os.O_WRONLY, 0755); err != nil {
 		command.logger.Panicf("ERROR: %s", err)
 	}
-	defer evaluationCSVFile.Close()
+	defer func() {
+		if err := evaluationCSVFile.Close(); err != nil {
+			panic(err)
+		}
+	}()
 
 	// Collect all evaluation CSV file paths.
 	allEvaluationPaths := map[string]bool{}
@@ -86,7 +90,11 @@ func (command *Report) Execute(args []string) (err error) {
 	if err != nil {
 		command.logger.Panicf("ERROR: %s", err)
 	}
-	defer modelsMetaInformationCSVFile.Close()
+	defer func() {
+		if err := modelsMetaInformationCSVFile.Close(); err != nil {
+			panic(err)
+		}
+	}()
 
 	// Fetch all openrouter models since it is the only provider that currently supports querying meta information.
 	provider := openrouter.NewProvider().(*openrouter.Provider)
@@ -139,7 +147,6 @@ func collectAllEvaluationLogFiles(evaluationCSVFilePaths []string) (evaluationLo
 		if err != nil {
 			continue
 		}
-		filepath.Base(evaluationDirectory)
 		evaluationLogFilePaths = append(evaluationLogFilePaths, filepath.Join(filepath.Base(evaluationDirectory), "evaluation.log"))
 	}
 

cmd/eval-dev-quality/cmd/report_test.go

@@ -313,7 +313,7 @@ func TestPathsFromGlobPattern(t *testing.T) {
 		Before: func(workingDirectory string) {
 			file, err := os.Create(filepath.Join(workingDirectory, "not-an-evaluation.csv"))
 			require.NoError(t, err)
-			file.Close()
+			require.NoError(t, file.Close())
 		},
 
 		EvaluationGlobPattern: "not-an-evaluation.csv",
@@ -393,7 +393,7 @@ func TestCollectAllEvaluationLogFiles(t *testing.T) {
 
 			file, err := os.Create(filepath.Join(workingDirectory, "someModel", "evaluation.csv"))
 			require.NoError(t, err)
-			file.Close()
+			require.NoError(t, file.Close())
 		},
 
 		EvaluationCSVFilePaths: []string{
@@ -442,9 +442,9 @@ func createEvaluationDirectoryWithLogFiles(t *testing.T, workingDirectory string
 
 	file, err := os.Create(filepath.Join(workingDirectory, "evaluation.csv"))
 	require.NoError(t, err)
-	file.Close()
+	require.NoError(t, file.Close())
 
 	file, err = os.Create(filepath.Join(workingDirectory, "evaluation.log"))
 	require.NoError(t, err)
-	file.Close()
+	require.NoError(t, file.Close())
 }

evaluate/evaluate.go

@@ -7,7 +7,6 @@ import (
 
 	"github.com/symflower/eval-dev-quality/evaluate/report"
 	evaluatetask "github.com/symflower/eval-dev-quality/evaluate/task"
-	"github.com/symflower/eval-dev-quality/language"
 	evallanguage "github.com/symflower/eval-dev-quality/language"
 	"github.com/symflower/eval-dev-quality/log"
 	evalmodel "github.com/symflower/eval-dev-quality/model"
@@ -79,7 +78,11 @@ func Evaluate(ctx *Context) (assessments *report.AssessmentStore, totalScore uin
 	if err != nil {
 		ctx.Log.Panicf("ERROR: unable to create evaluation CSV file: %+v", err)
 	}
-	defer evaluationCSVFile.Close()
+	defer func() {
+		if err := evaluationCSVFile.Close(); err != nil {
+			panic(err)
+		}
+	}()
 	evaluationFile, err := report.NewEvaluationFile(evaluationCSVFile)
 	if err != nil {
 		ctx.Log.Panicf("ERROR: %+v", err)
@@ -131,7 +134,7 @@ func Evaluate(ctx *Context) (assessments *report.AssessmentStore, totalScore uin
 					}
 
 					for _, taskIdentifier := range temporaryRepository.SupportedTasks() {
-						task, err := evaluatetask.TaskForIdentifier(taskIdentifier)
+						task, err := evaluatetask.ForIdentifier(taskIdentifier)
 						if err != nil {
 							logger.Fatal(err)
 						}
@@ -172,7 +175,9 @@ func Evaluate(ctx *Context) (assessments *report.AssessmentStore, totalScore uin
 								}
 								assessments.AddAssessmentPerTask(model, language, repositoryPath, assessment)
 								// Write the task assessment to the evaluation CSV file.
-								evaluationFile.WriteEvaluationRecord(model, language, temporaryRepository.Name(), runCount, assessment)
+								if err := evaluationFile.WriteEvaluationRecord(model, language, temporaryRepository.Name(), runCount, assessment); err != nil {
+									panic(err)
+								}
 							}
 						})
 					}
@@ -191,7 +196,7 @@ func Evaluate(ctx *Context) (assessments *report.AssessmentStore, totalScore uin
 	// Create temporary repositories for each language so the repository is copied only once per language.
 	temporaryRepositories := map[string]*evaluatetask.Repository{}
 	for _, l := range ctx.Languages {
-		relativeRepositoryPaths, err := language.RepositoriesForLanguage(l, ctx.TestdataPath)
+		relativeRepositoryPaths, err := evallanguage.RepositoriesForLanguage(l, ctx.TestdataPath)
 		if err != nil {
 			ctx.Log.Panicf("ERROR: %s", err)
 		}
@@ -256,7 +261,7 @@ func Evaluate(ctx *Context) (assessments *report.AssessmentStore, totalScore uin
 						continue
 					}
 					for _, taskIdentifier := range temporaryRepository.Tasks {
-						task, err := evaluatetask.TaskForIdentifier(taskIdentifier)
+						task, err := evaluatetask.ForIdentifier(taskIdentifier)
 						if err != nil {
 							logger.Fatal(err)
 						}
@@ -291,7 +296,9 @@ func Evaluate(ctx *Context) (assessments *report.AssessmentStore, totalScore uin
 								}
 								assessments.AddAssessmentPerTask(model, language, repositoryPath, assessment)
 								// Write the task assessment to the evaluation CSV file.
-								evaluationFile.WriteEvaluationRecord(model, language, temporaryRepository.Name(), runCount, assessment)
+								if err := evaluationFile.WriteEvaluationRecord(model, language, temporaryRepository.Name(), runCount, assessment); err != nil {
+									panic(err)
+								}
 							}
 						})
 					}

evaluate/evaluate_test.go

@@ -22,7 +22,6 @@ import (
 	"github.com/symflower/eval-dev-quality/language"
 	"github.com/symflower/eval-dev-quality/language/golang"
 	"github.com/symflower/eval-dev-quality/log"
-	"github.com/symflower/eval-dev-quality/model"
 	evalmodel "github.com/symflower/eval-dev-quality/model"
 	"github.com/symflower/eval-dev-quality/model/llm"
 	modeltesting "github.com/symflower/eval-dev-quality/model/testing"
@@ -863,7 +862,7 @@ func TestEvaluate(t *testing.T) {
 			mockedModelID := "testing-provider/testing-model"
 			mockedModel := modeltesting.NewMockCapabilityWriteTestsNamed(t, mockedModelID)
 			mockedProviderID := "testing-provider"
-			mockedProvider := providertesting.NewMockProviderNamedWithModels(t, mockedProviderID, []model.Model{mockedModel})
+			mockedProvider := providertesting.NewMockProviderNamedWithModels(t, mockedProviderID, []evalmodel.Model{mockedModel})
 			mockedLoader := providertesting.NewMockLoader(t)
 			embeddedProvider := &struct {
 				provider.Provider
@@ -947,7 +946,7 @@ func TestEvaluate(t *testing.T) {
 			mockedModelID := "testing-provider/testing-model"
 			mockedModel := modeltesting.NewMockCapabilityWriteTestsNamed(t, mockedModelID)
 			mockedProviderID := "testing-provider"
-			mockedProvider := providertesting.NewMockProviderNamedWithModels(t, mockedProviderID, []model.Model{mockedModel})
+			mockedProvider := providertesting.NewMockProviderNamedWithModels(t, mockedProviderID, []evalmodel.Model{mockedModel})
 			mockedLoader := providertesting.NewMockLoader(t)
 			embeddedProvider := &struct {
 				provider.Provider

evaluate/metrics/testing/assessments.go

@@ -11,40 +11,40 @@ import (
 
 // Clean deletes all empty and nondeterministic keys from the assessment.
 func Clean(assessment metrics.Assessments) metrics.Assessments {
-	copy := metrics.Assessments{}
-	maps.Copy(copy, assessment)
+	c := metrics.Assessments{}
+	maps.Copy(c, assessment)
 
-	delete(copy, metrics.AssessmentKeyProcessingTime)
+	delete(c, metrics.AssessmentKeyProcessingTime)
 
 	for _, key := range metrics.AllAssessmentKeysStrings {
-		if copy[metrics.AssessmentKey(key)] == 0 {
-			delete(copy, metrics.AssessmentKey(key))
+		if c[metrics.AssessmentKey(key)] == 0 {
+			delete(c, metrics.AssessmentKey(key))
 		}
 	}
 
-	return copy
+	return c
 }
 
 // CleanSlice deletes all empty and nondeterministic keys from the assessments.
 func CleanSlice(assessments []metrics.Assessments) []metrics.Assessments {
-	copy := make([]metrics.Assessments, len(assessments))
+	c := make([]metrics.Assessments, len(assessments))
 
 	for i, assessment := range assessments {
-		copy[i] = Clean(assessment)
+		c[i] = Clean(assessment)
 	}
 
-	return copy
+	return c
 }
 
 // CleanMap deletes all empty and nondeterministic keys from the assessments.
 func CleanMap[E comparable](assessments map[E]metrics.Assessments) map[E]metrics.Assessments {
-	copy := map[E]metrics.Assessments{}
+	c := map[E]metrics.Assessments{}
 
 	for key, assessment := range assessments {
-		copy[key] = Clean(assessment)
+		c[key] = Clean(assessment)
 	}
 
-	return copy
+	return c
 }
 
 // AssessmentsWithProcessingTime is an empty assessment collection with positive processing time.
@@ -61,8 +61,10 @@ type AssessmentTuple struct {
 	Assessment     metrics.Assessments
 }
 
+// AssessmentTuples holds a list of all parameters uniquely defining to which run an assessment belongs to.
 type AssessmentTuples []*AssessmentTuple
 
+// ToMap converts a list of assessment tuples to a mapping.
 func (at AssessmentTuples) ToMap() (lookup map[model.Model]map[language.Language]map[string]map[task.Identifier]metrics.Assessments) {
 	lookup = map[model.Model]map[language.Language]map[string]map[task.Identifier]metrics.Assessments{}
 	for _, t := range at {

evaluate/report/csv.go

@@ -72,7 +72,7 @@ func (e *EvaluationFile) WriteLines(records [][]string) (err error) {
 	return nil
 }
 
-// evaluationHeader returns the CSV header for the evaluation CSV.
+// EvaluationHeader returns the CSV header for the evaluation CSV.
 func EvaluationHeader() (header []string) {
 	return append([]string{"model-id", "language", "repository", "task", "run", "score"}, metrics.AllAssessmentKeysStrings...)
 }
@@ -84,12 +84,18 @@ func RecordsFromEvaluationCSVFiles(evaluationCSVFilePaths []string) (records [][
 		if err != nil {
 			return nil, pkgerrors.WithStack(err)
 		}
-		defer file.Close()
+		defer func() {
+			if err := file.Close(); err != nil {
+				panic(err)
+			}
+		}()
 
 		csv := csv.NewReader(file)
 
 		// Ignore the CSV header.
-		csv.Read()
+		if _, err := csv.Read(); err != nil {
+			panic(err)
+		}
 
 		evaluationRecords, err := csv.ReadAll()
 		if err != nil {

evaluate/task/task-code-repair.go → evaluate/task/code-repair.go

@@ -14,26 +14,26 @@ import (
 	evaltask "github.com/symflower/eval-dev-quality/task"
 )
 
-// TaskCodeRepair holds the code repair task.
-type TaskCodeRepair struct {
+// CodeRepair holds the code repair task.
+type CodeRepair struct {
 }
 
-// TaskArgumentsCodeRepair holds extra arguments to be used in a query prompt.
-type TaskArgumentsCodeRepair struct {
+// ArgumentsCodeRepair holds extra arguments to be used in a query prompt.
+type ArgumentsCodeRepair struct {
 	// Mistakes holds the list of compilation errors for a package.
 	Mistakes []string
 }
 
-var _ evaltask.Task = (*TaskCodeRepair)(nil)
+var _ evaltask.Task = (*CodeRepair)(nil)
 
 // Identifier returns the code repair task identifier.
-func (t *TaskCodeRepair) Identifier() evaltask.Identifier {
+func (t *CodeRepair) Identifier() evaltask.Identifier {
 	return IdentifierCodeRepair
 }
 
 // Run performs source code repairing in a repository with compilation errors.
 // This task requires the repository to consist of multiple packages, with each containing one faulty implementation file and a corresponding test file.
-func (t *TaskCodeRepair) Run(ctx evaltask.Context) (repositoryAssessment map[evaltask.Identifier]metrics.Assessments, problems []error, err error) {
+func (t *CodeRepair) Run(ctx evaltask.Context) (repositoryAssessment map[evaltask.Identifier]metrics.Assessments, problems []error, err error) {
 	modelCapability, ok := ctx.Model.(model.CapabilityRepairCode)
 	if !ok {
 		return nil, nil, pkgerrors.Wrap(evaltask.ErrTaskUnsupportedByModel, fmt.Sprintf("%q does not support %q", ctx.Model.ID(), string(t.Identifier())))
@@ -76,7 +76,7 @@ func (t *TaskCodeRepair) Run(ctx evaltask.Context) (repositoryAssessment map[eva
 			RepositoryPath: packagePath,
 			FilePath:       sourceFile,
 
-			Arguments: &TaskArgumentsCodeRepair{
+			Arguments: &ArgumentsCodeRepair{
 				Mistakes: mistakes,
 			},
 
@@ -115,7 +115,7 @@ func (t *TaskCodeRepair) Run(ctx evaltask.Context) (repositoryAssessment map[eva
 }
 
 // unpackCodeRepairPackage validates a package under test and returns the source file path and the list of compilation errors found.
-func (t *TaskCodeRepair) unpackCodeRepairPackage(ctx evaltask.Context, fileLogger *log.Logger, packagePath string) (sourceFilePath string, mistakes []string, err error) {
+func (t *CodeRepair) unpackCodeRepairPackage(ctx evaltask.Context, fileLogger *log.Logger, packagePath string) (sourceFilePath string, mistakes []string, err error) {
 	mistakes, err = ctx.Language.Mistakes(ctx.Logger, packagePath)
 	if err != nil {
 		return "", nil, pkgerrors.WithStack(err)