refactor: split README logic into separate file

cmd/readmevalidation/contributors.go

@@ -1,7 +1,6 @@
 package main
 
 import (
-	"bufio"
 	"errors"
 	"fmt"
 	"net/url"
@@ -13,17 +12,7 @@ import (
 	"gopkg.in/yaml.v3"
 )
 
-const rootRegistryPath = "./registry"
-
-var (
-	validContributorStatuses   = []string{"official", "partner", "community"}
-	supportedAvatarFileFormats = []string{".png", ".jpeg", ".jpg", ".gif", ".svg"}
-)
-
-type readme struct {
-	filePath string
-	rawText  string
-}
+var validContributorStatuses = []string{"official", "partner", "community"}
 
 type contributorProfileFrontmatter struct {
 	DisplayName    string `yaml:"display_name"`
@@ -65,40 +54,6 @@ func (vpe validationPhaseError) Error() string {
 	return msg
 }
 
-func extractFrontmatter(readmeText string) (string, error) {
-	if readmeText == "" {
-		return "", errors.New("README is empty")
-	}
-
-	const fence = "---"
-	fm := ""
-	fenceCount := 0
-	lineScanner := bufio.NewScanner(
-		strings.NewReader(strings.TrimSpace(readmeText)),
-	)
-	for lineScanner.Scan() {
-		nextLine := lineScanner.Text()
-		if fenceCount == 0 && nextLine != fence {
-			return "", errors.New("README does not start with frontmatter fence")
-		}
-
-		if nextLine != fence {
-			fm += nextLine + "\n"
-			continue
-		}
-
-		fenceCount++
-		if fenceCount >= 2 {
-			break
-		}
-	}
-
-	if fenceCount == 1 {
-		return "", errors.New("README does not have two sets of frontmatter fences")
-	}
-	return fm, nil
-}
-
 func validateContributorGithubUsername(githubUsername string) error {
 	if githubUsername == "" {
 		return errors.New("missing GitHub username")
@@ -297,7 +252,7 @@ func validateContributorYaml(yml contributorProfile) []error {
 }
 
 func parseContributorProfile(rm readme) (contributorProfile, error) {
-	fm, err := extractFrontmatter(rm.rawText)
+	fm, _, err := separateFrontmatter(rm.rawText)
 	if err != nil {
 		return contributorProfile{}, fmt.Errorf("%q: failed to parse frontmatter: %v", rm.filePath, err)
 	}

cmd/readmevalidation/readmes.go

@@ -0,0 +1,127 @@
+package main
+
+import (
+	"bufio"
+	"errors"
+	"fmt"
+	"strings"
+)
+
+const rootRegistryPath = "./registry"
+
+var supportedAvatarFileFormats = []string{".png", ".jpeg", ".jpg", ".gif", ".svg"}
+
+// Readme represents a single README file within the repo (usually within the
+// "/registry" directory).
+type readme struct {
+	filePath string
+	rawText  string
+}
+
+// separateFrontmatter attempts to separate a README file's frontmatter content
+// from the main README body, returning both values in that order. It does not
+// validate whether the structure of the frontmatter is valid (i.e., that it's
+// structured as YAML).
+func separateFrontmatter(readmeText string) (string, string, error) {
+	if readmeText == "" {
+		return "", "", errors.New("README is empty")
+	}
+
+	const fence = "---"
+	fm := ""
+	body := ""
+	fenceCount := 0
+	lineScanner := bufio.NewScanner(
+		strings.NewReader(strings.TrimSpace(readmeText)),
+	)
+	for lineScanner.Scan() {
+		nextLine := lineScanner.Text()
+		if fenceCount < 2 && nextLine == fence {
+			fenceCount++
+			continue
+		}
+		// Break early if the very first line wasn't a fence, because then we
+		// know for certain that the README has problems
+		if fenceCount == 0 {
+			break
+		}
+
+		// It should be safe to trim each line of the frontmatter on a per-line
+		// basis, because there shouldn't be any extra meaning attached to the
+		// indentation. The same does NOT apply to the README; best we can do is
+		// gather all the lines, and then trim around it
+		if inReadmeBody := fenceCount >= 2; inReadmeBody {
+			body += nextLine + "\n"
+		} else {
+			fm += strings.TrimSpace(nextLine) + "\n"
+		}
+	}
+	if fenceCount < 2 {
+		return "", "", errors.New("README does not have two sets of frontmatter fences")
+	}
+	if fm == "" {
+		return "", "", errors.New("readme has frontmatter fences but no frontmatter content")
+	}
+
+	return fm, strings.TrimSpace(body), nil
+}
+
+// validationPhase represents a specific phase during README validation. It is
+// expected that each phase is discrete, and errors during one will prevent a
+// future phase from starting.
+type validationPhase int
+
+const (
+	// validationPhaseFilesystemRead indicates when a README file is being read
+	// from the file system
+	validationPhaseFilesystemRead validationPhase = iota
+
+	// validationPhaseReadmeParsing indicates when a README's frontmatter is being
+	// parsed as YAML. This phase does not include YAML validation.
+	validationPhaseReadmeParsing
+
+	// validationPhaseReadmeValidation indicates when a README's frontmatter is
+	// being validated as proper YAML with expected keys.
+	validationPhaseReadmeValidation
+
+	// validationPhaseAssetCrossReference indicates when a README's frontmatter
+	// is having all its relative URLs be validated for whether they point to
+	// valid resources.
+	validationPhaseAssetCrossReference
+)
+
+func (p validationPhase) String() string {
+	switch p {
+	case validationPhaseFilesystemRead:
+		return "Filesystem reading"
+	case validationPhaseReadmeParsing:
+		return "README parsing"
+	case validationPhaseReadmeValidation:
+		return "README validation"
+	case validationPhaseAssetCrossReference:
+		return "Cross-referencing asset references"
+	default:
+		return "Unknown validation phase"
+	}
+}
+
+var _ error = ValidationPhaseError{}
+
+// ValidationPhaseError represents an error that occurred during a specific
+// phase of README validation. It should be used to collect ALL validation
+// errors that happened during a specific phase, rather than the first one
+// encountered.
+type ValidationPhaseError struct {
+	phase  validationPhase
+	errors []error
+}
+
+func (vpe ValidationPhaseError) Error() string {
+	msg := fmt.Sprintf("Error during %q phase of README validation:", vpe.phase.String())
+	for _, e := range vpe.errors {
+		msg += fmt.Sprintf("\n- %v", e)
+	}
+	msg += "\n"
+
+	return msg
+}