sm-sitemap-plugin/models/settings.go

package models

import (
	"net/url"
	"strconv"
	"strings"
	"time"
	"unicode"
	"unicode/utf8"

	"git.golem15.com/golem15/summercms/modules/cabana"
	"git.golem15.com/golem15/summercms/modules/lagoon"
)

// errFrontendURL is the phrase key of the frontend_url refusal.
const errFrontendURL = "golem15.sitemap::lang.errors.frontend_url"

// The settings keys of the five priority tiers, in the order PHP checked
// them. A tier rule names one of them.
const (
	PriorityHomepage   = "priority_homepage"
	PriorityServices   = "priority_services"
	PriorityBlog       = "priority_blog"
	PriorityCategories = "priority_categories"
	PriorityLegal      = "priority_legal"
)

// PriorityDefaults are the PHP defaults of the priority settings.
var PriorityDefaults = map[string]string{
	PriorityHomepage:   "1.0",
	PriorityServices:   "0.8",
	PriorityBlog:       "0.6",
	PriorityCategories: "0.4",
	PriorityLegal:      "0.2",
}

// Settings is the sitemap settings singleton (row id 1) on
// golem15_sitemap_settings, the PHP Golem15\Sitemap\Models\Settings that
// WinterCMS kept in system_settings under golem15_sitemap_settings.
// FrontendURL replaces the backend URL in the generated XML;
// FrontendDefaultLocale is the locale whose URLs carry no prefix;
// UpdatesManifestPath names a JSON file of URL path to last-modified date;
// EnablePriorityTiers turns the tier priorities on, and the priority fields
// hold each tier's value as text. A null priority reads as its PHP default.
// The PHP settings form validated nothing; the port checks FrontendURL in
// BeforeValidate.
type Settings struct {
	ID                    uint    `gorm:"column:id;primaryKey"`
	FrontendURL           *string `gorm:"column:frontend_url"`
	FrontendDefaultLocale *string `gorm:"column:frontend_default_locale"`
	UpdatesManifestPath   *string `gorm:"column:updates_manifest_path"`
	EnablePriorityTiers   bool    `gorm:"column:enable_priority_tiers"`
	PriorityHomepage      *string `gorm:"column:priority_homepage"`
	PriorityServices      *string `gorm:"column:priority_services"`
	PriorityBlog          *string `gorm:"column:priority_blog"`
	PriorityCategories    *string `gorm:"column:priority_categories"`
	PriorityLegal         *string `gorm:"column:priority_legal"`

	SiteURL             *string                     `gorm:"column:site_url"`
	FileSize            int                         `gorm:"column:file_size"`
	DefaultLocale       *string                     `gorm:"column:default_locale"`
	ExtraURLs           lagoon.Jsonable[[]ExtraURL] `gorm:"column:extra_urls"`
	RebuildPendingUntil *time.Time                  `gorm:"column:rebuild_pending_until"`
	NumberedSize        *int                        `gorm:"column:numbered_size"`
	BuiltAt             *time.Time                  `gorm:"column:built_at"`
	RebuildRequestedAt  *time.Time                  `gorm:"column:rebuild_requested_at"`
}

// DefaultFileSize is the number of addresses per sitemap file when the
// setting is not stored (D-30); MaxFileSize is the protocol's limit.
const (
	DefaultFileSize = 1000
	MaxFileSize     = 50000
)

// ExtraURL is one additional address of the settings: URL is a path (or an
// absolute http(s) URL on the site), LastMod its optional last change as
// the admin's date picker sends it.
type ExtraURL struct {
	URL     string `json:"url"`
	LastMod string `json:"lastmod"`
}

// EffectiveFileSize is the stored file size, or DefaultFileSize when it is
// outside 1..MaxFileSize.
func (s Settings) EffectiveFileSize() int {
	if s.FileSize < 1 || s.FileSize > MaxFileSize {
		return DefaultFileSize
	}
	return s.FileSize
}

// Extra returns the additional addresses.
func (s Settings) Extra() []ExtraURL { return s.ExtraURLs.Get() }

func (Settings) TableName() string { return "golem15_sitemap_settings" }

// Fillable lists the columns the settings form edits.
func (Settings) Fillable() []string {
	return []string{"site_url", "file_size", "extra_urls"}
}

// Rules is empty: the PHP settings form validated nothing. The one check of
// the port, on frontend_url, is in BeforeValidate.
func (Settings) Rules() map[string]string { return map[string]string{} }

// The phrase keys of the other settings refusals.
const (
	errFileSize  = "golem15.sitemap::lang.errors.file_size"
	errURLScheme = "golem15.sitemap::lang.errors.url_scheme"
	errLastmod   = "golem15.sitemap::lang.errors.lastmod"
)

// Validate answers every problem Problems finds as a validation error on
// its field; the settings controller calls it before a save, so the save
// answers 422 and the row is left as it was.
func (s Settings) Validate() error {
	problems := s.Problems()
	if len(problems) == 0 {
		return nil
	}
	details := make(map[string]any, len(problems))
	for field, keys := range problems {
		details[field] = keys
	}
	return &cabana.ValidationError{Details: details}
}

// Problems lists the settings' refusals by field (UI-SPEC C26): a site
// address that AllowedFrontendURL refuses (site_url and the Phase 17
// frontend_url), a file size outside 1..MaxFileSize and an additional
// address that is not a path or an http(s) URL (extra_urls.<i>.url) or
// whose lastmod is not a date (extra_urls.<i>.lastmod).
func (s Settings) Problems() map[string][]string {
	out := map[string][]string{}
	if !AllowedFrontendURL(Text(s.SiteURL)) {
		out["site_url"] = []string{errFrontendURL}
	}
	if !AllowedFrontendURL(Text(s.FrontendURL)) {
		out["frontend_url"] = []string{errFrontendURL}
	}
	if s.FileSize < 1 || s.FileSize > MaxFileSize {
		out["file_size"] = []string{errFileSize}
	}
	for i, e := range s.Extra() {
		if !AllowedAddress(strings.TrimSpace(e.URL)) {
			out["extra_urls."+strconv.Itoa(i)+".url"] = []string{errURLScheme}
		}
		if mod := strings.TrimSpace(e.LastMod); mod != "" && !ValidDate(mod) {
			out["extra_urls."+strconv.Itoa(i)+".lastmod"] = []string{errLastmod}
		}
	}
	return out
}

// AllowedFrontendURL reports whether u may be the frontend_url setting: the
// empty string, or an absolute http or https URL with a host and without
// whitespace, control characters, a double quote, <, > or &. The PHP plugin
// stored any text. The value is written into every loc, every hreflang link
// and the stylesheet instruction of the sitemap, and sitemap.xsl renders
// each loc as a link on the public origin, so a value outside this form
// (a javascript: URL, or text that closes the attribute it is written into)
// would reach the readers of /sitemap.xml (T-17-15).
func AllowedFrontendURL(u string) bool {
	if u == "" {
		return true
	}
	if !utf8.ValidString(u) {
		return false
	}
	if strings.ContainsFunc(u, func(r rune) bool {
		return unicode.IsControl(r) || unicode.IsSpace(r) || r == '"' || r == '<' || r == '>' || r == '&'
	}) {
		return false
	}
	parsed, err := url.Parse(u)
	if err != nil {
		return false
	}
	if parsed.Scheme != "http" && parsed.Scheme != "https" {
		return false
	}
	return parsed.Opaque == "" && parsed.Hostname() != ""
}

// DefaultSettings is the settings PHP read when nothing was saved: tiers on
// with the default priorities.
func DefaultSettings() Settings {
	return Settings{ID: 1, EnablePriorityTiers: true, FileSize: DefaultFileSize}
}

// Priority returns the value of a priority setting: the stored text, or the
// PHP default when it is null.
func (s Settings) Priority(key string) string {
	var p *string
	switch key {
	case PriorityHomepage:
		p = s.PriorityHomepage
	case PriorityServices:
		p = s.PriorityServices
	case PriorityBlog:
		p = s.PriorityBlog
	case PriorityCategories:
		p = s.PriorityCategories
	case PriorityLegal:
		p = s.PriorityLegal
	}
	if p == nil {
		return PriorityDefaults[key]
	}
	return *p
}

func init() {
	Register(Settings{})
}

// dateLayouts are the lastmod forms an additional address may carry.
var dateLayouts = []string{time.RFC3339Nano, "2006-01-02 15:04:05", "2006-01-02T15:04:05", "2006-01-02"}

// ParseDate reads a lastmod as the settings form or an import gives it
// (YYYY-MM-DD, "YYYY-MM-DD hh:mm:ss" or RFC 3339; UTC unless it names a
// zone); nil when empty or unreadable.
func ParseDate(s string) *time.Time {
	s = strings.TrimSpace(s)
	if s == "" {
		return nil
	}
	for _, layout := range dateLayouts {
		if t, err := time.Parse(layout, s); err == nil {
			t = t.UTC()
			return &t
		}
	}
	return nil
}

// ValidDate reports whether s is a lastmod ParseDate reads.
func ValidDate(s string) bool { return ParseDate(s) != nil }