[CMSP-27] Implement sites.yml parsing (#1)

.gitignore

@@ -4,3 +4,5 @@
 .DS_Store
 
 dist/
+
+*.out

README.MD

@@ -1 +1,28 @@
 # Sites.yml | Pantheon.yml Validator
+
+A utility for validating a sites.yml file on a pantheon site during WordPress multisites' search-replace tasks. Asprirationally to include pantheon.yml validation in the future.
+
+# Usage
+
+## Sites.yml
+```
+$ pyml-validator sites -f path/to/sites.yml
+```
+
+See [this annotated fixture](./fixtures/sites/valid.yml) for an example of a valid sites.yml file.
+
+## Pantheon.yml
+Note, validation of pantheon.yml is unimplemented, so any file reads as valid.
+```
+$ pyml-validator pantheon -f path/to/pantheon.yml
+```
+
+# Testing
+
+[![Coverage Status](https://coveralls.io/repos/github/pantheon-systems/pyml-validator/badge.svg?t=PGhafd)](https://coveralls.io/github/pantheon-systems/pyml-validator)
+
+`make test` runs linting and testing.
+
+# Releases
+
+Automatically releases on merge to main via autotag + goreleaser. See [Autotag Readme](https://github.com/pantheon-systems/autotag) for details on how the SemVer is determined. Note, with goreleaser, each commit merged will become a line item in the release's Changelog. Take note to use squashing and/or rebase to ensure helpful and informative commit messages.

cmd/root.go

@@ -0,0 +1,30 @@
+package cmd
+
+import (
+	"log"
+
+	"github.com/spf13/cobra"
+)
+
+var FilePath string
+
+var rootCmd = &cobra.Command{
+	Use:   "pyml-validator",
+	Short: "Pyml-validator validates pantheon.yml, sites.yml, etc.",
+	Long: `Pyml-validator is a validator for pantheon.yml or sites.yml.
+Ensures that the given config file can be used by the platform.`,
+}
+
+func Execute() {
+	if err := rootCmd.Execute(); err != nil {
+		log.Fatal(err)
+	}
+}
+
+func init() {
+	rootCmd.PersistentFlags().StringVarP(&FilePath, "file", "f", "", "path/to/file.yml")
+	err := rootCmd.MarkPersistentFlagRequired("file")
+	if err != nil {
+		log.Fatal(err)
+	}
+}

cmd/validators.go

@@ -0,0 +1,49 @@
+package cmd
+
+import (
+	"fmt"
+	"pyml-validator/pkg/validator"
+
+	"github.com/spf13/cobra"
+)
+
+func validatorCommand(cmd *cobra.Command) error {
+	// Is there a better way to do this? Without this we print usage on error exits.
+	// If we override at the root level, we don't get usage when we _do_ want it.
+	cmd.SilenceUsage = true
+
+	v, err := validator.ValidatorFactory(cmd.Use)
+	if err != nil {
+		return err
+	}
+
+	err = v.ValidateFromFilePath(FilePath)
+	if err != nil {
+		return err
+	}
+	fmt.Printf("✨ %s.yml is valid\n", cmd.Use)
+	return nil
+}
+
+var sitesCommand = &cobra.Command{
+	Use:   "sites",
+	Short: "validate sites.yml",
+	Long:  `Validate sites.yml`,
+	RunE: func(cmd *cobra.Command, args []string) error {
+		return validatorCommand(cmd)
+	},
+}
+
+var pantheonCommand = &cobra.Command{
+	Use:   "pantheon",
+	Short: "validate pantheon.yml",
+	Long:  `Validate pantheon.yml. For more information, see https://pantheon.io/docs/pantheon-yml`,
+	RunE: func(cmd *cobra.Command, args []string) error {
+		return validatorCommand(cmd)
+	},
+}
+
+func init() {
+	rootCmd.AddCommand(pantheonCommand)
+	rootCmd.AddCommand(sitesCommand)
+}

fixtures/sites/invalid_api_version_only.yml

@@ -0,0 +1,2 @@
+---
+api_version: 2

fixtures/sites/valid.yml

@@ -0,0 +1,31 @@
+---
+api_version: 1 # Currently only one api version.
+
+# "domain_maps" is a collection of blog URLs for each environment used to
+# facilitate search-replace of a WordPress Multisite (WPMS) across pantheon
+# environments. Each key of "domain_maps" must be a valid environment name.
+domain_maps:
+  # environment: <collection of domains to be used on this environment>
+  # i.e. dev, test, live, feat-branch, &c.
+  dev:
+    # each environment collection maps the blog ID to its URL. A url must be
+    # set in both the target and source environments for search-replace to be
+    # run.
+    # i.e. 1: blog1-mysite.com
+    1: about.dev-mysite.pantheonsite.io  
+    2: employee-resources.dev-mysite.pantheonsite.io  
+    3: staff-portal.dev-mysite.pantheonsite.io  
+  test:
+    1: about.test-mysite.pantheonsite.io
+    2: employee-resources.test-mysite.pantheonsite.io
+    3: staff-portal.test-mysite.pantheonsite.io
+  live:
+    1: about.mysite.com
+    2: employee-resources.mysite.com
+    3: staff-portal.mysite.com
+  autopilot:
+    1: about.autopilot-mysite.pantheonsite.io
+    2: employee-resources.autopilot-mysite.pantheonsite.io
+    3: staff-portal.autopilot-mysite.pantheonsite.io
+
+# Anything else in the file will be ignored, but not rejected.

fixtures/sites/valid_api_version_only.yml

@@ -0,0 +1,2 @@
+---
+api_version: 1

go.mod

@@ -1,3 +1,18 @@
 module pyml-validator
 
 go 1.19
+
+require (
+	github.com/spf13/cobra v1.6.1
+	github.com/stretchr/testify v1.8.1
+	gopkg.in/yaml.v3 v3.0.1
+)
+
+require (
+	github.com/davecgh/go-spew v1.1.1 // indirect
+	github.com/inconshreveable/mousetrap v1.1.0 // indirect
+	github.com/pmezard/go-difflib v1.0.0 // indirect
+	github.com/spf13/pflag v1.0.5 // indirect
+)
+
+replace gopkg.in/yaml.v3 => gopkg.in/yaml.v3 v3.0.1

go.sum

@@ -0,0 +1,25 @@
+github.com/cpuguy83/go-md2man/v2 v2.0.2/go.mod h1:tgQtvFlXSQOSOSIRvRPT7W67SCa46tRHOmNcaadrF8o=
+github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
+github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/inconshreveable/mousetrap v1.0.1/go.mod h1:vpF70FUmC8bwa3OWnCshd2FqLfsEA9PFc4w1p2J65bw=
+github.com/inconshreveable/mousetrap v1.1.0 h1:wN+x4NVGpMsO7ErUn/mUI3vEoE6Jt13X2s0bqwp9tc8=
+github.com/inconshreveable/mousetrap v1.1.0/go.mod h1:vpF70FUmC8bwa3OWnCshd2FqLfsEA9PFc4w1p2J65bw=
+github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
+github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/russross/blackfriday/v2 v2.1.0/go.mod h1:+Rmxgy9KzJVeS9/2gXHxylqXiyQDYRxCVz55jmeOWTM=
+github.com/spf13/cobra v1.6.1 h1:o94oiPyS4KD1mPy2fmcYYHHfCxLqYjJOhGsCHFZtEzA=
+github.com/spf13/cobra v1.6.1/go.mod h1:IOw/AERYS7UzyrGinqmz6HLUo219MORXGxhbaJUqzrY=
+github.com/spf13/pflag v1.0.5 h1:iy+VFUOCP1a+8yFto/drg2CJ5u0yRoB7fZw3DKv/JXA=
+github.com/spf13/pflag v1.0.5/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
+github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
+github.com/stretchr/objx v0.4.0/go.mod h1:YvHI0jy2hoMjB+UWwv71VJQ9isScKT/TqJzVSSt89Yw=
+github.com/stretchr/objx v0.5.0/go.mod h1:Yh+to48EsGEfYuaHDzXPcE3xhTkx73EhmCGUpEOglKo=
+github.com/stretchr/testify v1.7.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
+github.com/stretchr/testify v1.8.0/go.mod h1:yNjHg4UonilssWZ8iaSj1OCr/vHnekPRkoO+kdMU+MU=
+github.com/stretchr/testify v1.8.1 h1:w7B6lhMri9wdJUVmEZPGGhZzrYTPvgJArz7wNPgYKsk=
+github.com/stretchr/testify v1.8.1/go.mod h1:w2LPCIKwWwSfY2zedu0+kehJoqGctiVI29o6fzry7u4=
+gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405 h1:yhCVgyC4o1eVCa2tZl7eS0r+SDo693bJlVdllGtEeKM=
+gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
+gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
+gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=

main.go

@@ -1,3 +1,7 @@
 package main
 
-func main() {}
+import "pyml-validator/cmd"
+
+func main() {
+	cmd.Execute()
+}

pkg/model/sites.go

@@ -0,0 +1,12 @@
+package model
+
+// SitesYml is used to map domains across environments for search and replace with WPMS sites.
+type SitesYml struct {
+	APIVersion int        `yaml:"api_version"`
+	DomainMaps DomainMaps `yaml:"domain_maps"`
+}
+
+type DomainMaps map[string]DomainMapByEnvironment
+
+// DomainMapByEnvironment is a map of site (blog) domains keyed by blog ID.
+type DomainMapByEnvironment map[int]string

pkg/validator/pantheon.go

@@ -0,0 +1,22 @@
+package validator
+
+import (
+	"fmt"
+	"os"
+)
+
+type PantheonValidator struct{}
+
+// ValidateFromYaml asserts a given pantheon.yaml file is valid.
+// As this has not been implemented, nothing is invalid.
+func (v *PantheonValidator) ValidateFromYaml(y []byte) error {
+	return nil
+}
+
+func (v *PantheonValidator) ValidateFromFilePath(filePath string) error {
+	yFile, err := os.ReadFile(filePath)
+	if err != nil {
+		return fmt.Errorf("error reading YAML file: %w", err)
+	}
+	return v.ValidateFromYaml(yFile)
+}

pkg/validator/sites.go

@@ -0,0 +1,81 @@
+package validator
+
+import (
+	"fmt"
+	"os"
+	"pyml-validator/pkg/model"
+	"regexp"
+
+	"gopkg.in/yaml.v3"
+)
+
+const (
+	MaxDomainMaps = 25 // This could be raised
+)
+
+var (
+	// See https://github.com/pantheon-systems/titan-mt/blob/master/yggdrasil/lib/pantheon_yml/pantheon_yml_v1_schema.py
+	ValidHostnameRegex     = regexp.MustCompile(`^([a-zA-Z0-9]|[a-zA-Z0-9][a-zA-Z0-9\-]{0,61}[a-zA-Z0-9])(\.([a-zA-Z0-9]|[a-zA-Z0-9][a-zA-Z0-9\-]{0,61}[a-zA-Z0-9]))*$`)
+	ValidMultidevNameRegex = regexp.MustCompile(`^[a-z0-9\-]{1,11}$`)
+)
+
+type SitesValidator struct{}
+
+// ValidateFromYaml asserts a given sites.yaml file is valid.
+func (v *SitesValidator) ValidateFromYaml(y []byte) error {
+	var s model.SitesYml
+
+	err := yaml.Unmarshal(y, &s)
+	if err != nil {
+		return err
+	}
+	return v.validate(s)
+}
+
+func (v *SitesValidator) ValidateFromFilePath(filePath string) error {
+	yFile, err := os.ReadFile(filePath)
+	if err != nil {
+		return fmt.Errorf("error reading YAML file: %w", err)
+	}
+	return v.ValidateFromYaml(yFile)
+}
+
+// validate asserts all aspects of sites.yml are valid.
+func (v *SitesValidator) validate(sites model.SitesYml) error {
+	err := validateAPIVersion(sites.APIVersion)
+	if err != nil {
+		return err
+	}
+	return validateDomainMaps(sites.DomainMaps)
+}
+
+// validateDomainMaps ensures the domain maps provided in sites.yml are valid
+// by asserting cloud development environments names are valid, there are not
+// too many domain maps listed for any environment, and that the hostnames
+// provided are valid Pantheon hostnames.
+func validateDomainMaps(domainMaps map[string]model.DomainMapByEnvironment) error {
+	for env, domainMap := range domainMaps {
+		if !ValidMultidevNameRegex.MatchString(env) {
+			return fmt.Errorf("%q is not a valid environment name", env)
+		}
+		domainMapCount := len(domainMap)
+		if domainMapCount > MaxDomainMaps {
+			return fmt.Errorf("%q has too many domains listed (%d). Maximum is %d", env, domainMapCount, MaxDomainMaps)
+		}
+		for _, domain := range domainMap {
+			if !ValidHostnameRegex.MatchString(domain) {
+				return fmt.Errorf("%q is not a valid hostname", domain)
+			}
+		}
+	}
+	return nil
+}
+
+// validateAPIVersion asserts if sites.yml has a valid api version set. Once
+// more than one version is valid, this will need to be more more robust.
+func validateAPIVersion(apiVersion int) error {
+	if apiVersion != 1 {
+		return ErrInvalidAPIVersion
+	}
+	return nil
+}

pkg/validator/sites_errors.go

@@ -0,0 +1,10 @@
+package validator
+
+import "errors"
+
+var (
+	ErrInvalidAPIVersion = errors.New("Invalid API Version. Must be '1'")
+)
+
+// TODO: More dynamic errors could be refactored here, but likely only worth
+// pursuing once we are passing errors back to customers