wpimport

type AuditFailure struct {
	URL      string `json:"url"`
	Status   int    `json:"status"`
	Location string `json:"location,omitempty"`
	Reason   string `json:"reason"`
}

type AuditResult struct {
	Checked    int            `json:"checked"`
	OK         int            `json:"ok"`         // answered 200 directly
	Redirected int            `json:"redirected"` // answered one 301 to a 200
	Failures   []AuditFailure `json:"failures,omitempty"`
	// Unresolved are old URLs that answer, but with a page that is not the
	// imported content (see Result.URLConflicts).
	Unresolved []AuditFailure `json:"unresolved,omitempty"`
	// Hidden are old author archives that answer 404 on purpose: the person
	// imported for that author is not public yet. Making the person public
	// gives them a page; a re-run then redirects the old URL to it.
	Hidden []AuditFailure `json:"hidden,omitempty"`
}

type Author struct {
	Key       string // login or slug: what items refer to
	Name      string
	Slug      string
	Bio       string
	Link      string // old author archive URL
	AvatarURL string // only set for avatars hosted on the site itself
}

type Client struct {
	Base *url.URL
	// contains filtered or unexported fields
}

type ClientOptions struct {
	// URL is the site's home URL.
	URL string
	// User and AppPassword are WordPress application-password credentials;
	// without them only public content is read (no drafts, no menus).
	User, AppPassword string
	// AllowPrivate lets the importer reach a site on a private or loopback
	// address (an intranet WordPress, the test fixture). Off by default: the
	// guard that protects plugin http.fetch applies here too.
	AllowPrivate bool
	// RatePerSec limits requests to the old site; 0 means 10.
	RatePerSec float64
	// MaxRetries is the number of retries after the first attempt; 0 means 4.
	MaxRetries int
	// Timeout is per request; 0 means 30s.
	Timeout time.Duration
	// MaxTotalBytes bounds everything read through Do (REST pages, sitemaps)
	// over the life of the client; 0 means DefaultMaxTotalBytes.
	MaxTotalBytes int64
	// MaxPages is the hard ceiling of pages of one collection; 0 means DefaultMaxPages.
	MaxPages int
	// Lookup overrides name resolution (tests).
	Lookup safefetch.Lookup
	// Sleep overrides the backoff wait (tests).
	Sleep func(ctx context.Context, d time.Duration) error
}

type Converted struct {
	HTML       string
	Shortcodes []string // names of shortcodes found (and removed)
	Forms      []string // form plugins whose markup or shortcode was removed
	Builders   []string // page builders whose markers were found
	Embeds     int      // YouTube and Vimeo embeds kept
	Dropped    []string // other things removed: unsupported iframes
}

type CustomType struct {
	Slug         string
	Name         string
	RESTBase     string
	Hierarchical bool
	HasArchive   bool
}

type Deps struct {
	DB         *database.DB
	Entities   *entity.Service
	SchemaRepo entity.SchemaRepository
	Media      *media.Service
	Taxonomy   *taxonomy.Service
	TermRepo   taxonomy.Repository
	Menus      *menu.Service
	People     *person.Store
	Redirects  *redirect.Service
	Site       *siteconfig.Store
	Logger     *slog.Logger
}

type DepsConfig struct {
	MediaDir     string
	MediaBaseURL string
	MediaMaxMB   int
}

type Detection struct {
	SEOPlugin     string
	PageBuilders  []string
	Forms         []string
	Multilingual  []string
	DefaultLang   string
	OtherPlugins  []string // REST namespaces or plugin markers nothing maps
	Shortcodes    map[string]int
	Namespaces    []string
	CustomTypes   []CustomType
	PermalinkNote string
}

type Dump struct {
	Site    Site
	Authors []Author
	Terms   []Term
	Media   []Media
	Items   []Item
	Menus   []Menu
	// OldURLs are the public URLs the old site lists in its sitemap (REST), or
	// the links of its published items (WXR); the audit checks each.
	OldURLs []string
	// Detect holds what the scan found out about plugins and builders.
	Detect Detection
	// RawBodies is true when item bodies are as typed (WXR): block comments,
	// unexpanded shortcodes, paragraphs not yet wrapped.
	RawBodies bool
	// Warnings are non-fatal problems met while reading.
	Warnings []string
}

type EntityRef struct {
	ID   uuid.UUID
	Path string // the new public path
}

type Fidelity struct {
	High    int `json:"high"`
	Medium  int `json:"medium"`
	Low     int `json:"low"`
	Percent int `json:"percent"` // share of items imported without loss
}

type HTTPError struct {
	Status int
	URL    string
	Body   string
}

type HTTPProber struct {
	Base   string
	Client *http.Client
}

type HandlerProber struct {
	Handler http.Handler
	// Host is the Host header; empty means example.test.
	Host string
}

type HostedProber interface {
	Prober
	Hosts() []string
}

type Item struct {
	ID       string
	Type     string // post, page or a custom post type slug
	Status   string // WordPress status
	Slug     string
	Title    string
	Content  string // HTML as the reader sees it (REST) or raw (WXR)
	Raw      string // raw content with block comments and shortcodes, when known
	Excerpt  string
	Link     string // old public URL
	Date     time.Time
	Modified time.Time
	ParentID string
	// AuthorKey refers to Author.Key.
	AuthorKey string
	// FeaturedMedia is a Media.ID.
	FeaturedMedia string
	Terms         []TermRef
	Protected     bool   // password protected
	Lang          string // language tag as the source gave it (WPML or Polylang); empty when it gave none
	SEO           SEO
	Meta          map[string]string
}

type ItemError struct {
	Kind  string `json:"kind"`
	ID    string `json:"id,omitempty"`
	Title string `json:"title,omitempty"`
	Err   string `json:"error"`
}

type ItemNote struct {
	ID     string `json:"id"`
	Type   string `json:"type"`
	Title  string `json:"title"`
	Link   string `json:"link"`
	Reason string `json:"reason"`
}

type Media struct {
	ID      string
	URL     string
	Sizes   []string // URLs of generated sizes, mapped to the same media
	Mime    string
	Title   string
	Alt     string
	Caption string
	Date    time.Time
}

type Menu struct {
	ID       string
	Name     string
	Slug     string
	Location string
	Items    []MenuItem
}

type MenuItem struct {
	ID       string
	ParentID string
	Title    string
	URL      string
	Type     string // post_type, taxonomy or custom
	Object   string // page, post, category, ...
	ObjectID string // item id, or "taxonomy:slug" for terms
	NewTab   bool
	Order    int
}

type Options struct {
	// SourceRef identifies the source site; it scopes the id map. Set from the Dump when empty.
	SourceRef string
	// Actor is recorded as the creator of imported rows (nil: none).
	Actor *uuid.UUID
	// AllowPrivate lets media be fetched from a private-network site. It turns
	// the SSRF guard off for everything the source names: set it only from an
	// operator's own command line. Never bind it from request data and never
	// restore it from a stored run's options.
	AllowPrivate bool
	// Client downloads media; built from the site URL when nil (WXR).
	Client *Client
	// MediaConcurrency bounds parallel downloads; 0 means 4.
	MediaConcurrency int
	// KeepPermalinks sets the post and page URL patterns to the WordPress
	// structure. Default true (set NoPermalinks to disable).
	NoPermalinks bool
	// WorkDir holds temporary downloads; empty uses the system temp dir.
	WorkDir string
	// Progress receives one line per step.
	Progress func(phase, msg string)
	// Purge clears rendered-page and sitemap caches after the import. The CLI is a
	// separate process and cannot reach the server's caches: it leaves this nil.
	Purge func(ctx context.Context)
	// Audit checks every old URL against the new site; nil skips it.
	Audit Prober
	// KeepEdited leaves entities alone that were edited in Hatamoto after the
	// last import. By default a re-run overwrites them and lists them in the
	// result.
	KeepEdited bool
	// SkipMedia imports no files: the library is skipped and images and
	// attachments inside bodies are removed or unlinked, as for a file that
	// cannot be fetched (no wp-content URL is left behind).
	SkipMedia bool
	// Options are written to the run (no secrets) so --resume can reuse them.
	Stored map[string]any
}

type Prober interface {
	Probe(ctx context.Context, path string) (status int, location string, err error)
}

type RESTOptions struct {
	ClientOptions
	// MaxItems caps posts, pages and custom post types together (0: 50000).
	MaxItems int
	// MaxCollection caps each of terms, users, media and menu items (0: DefaultMaxCollection).
	MaxCollection int
	// MaxDuration bounds the whole read (0: 4 hours).
	MaxDuration time.Duration
	// Concurrency bounds the per-item requests (SEO data); 0 means 4.
	Concurrency int
	// Progress receives short status lines.
	Progress func(string)
}

type RESTSource struct {
	// contains filtered or unexported fields
}

type Report struct {
	Source string `json:"source"`
	Site   Site   `json:"site"`

	Posts       int            `json:"posts"`
	Pages       int            `json:"pages"`
	CustomTypes map[string]int `json:"custom_types,omitempty"`
	Drafts      int            `json:"drafts"`
	Private     int            `json:"private"`
	Media       int            `json:"media"`
	Terms       int            `json:"terms"`
	Authors     int            `json:"authors"`
	Menus       int            `json:"menus"`
	OldURLs     int            `json:"old_urls"`

	// SiteMode is "empty" or "existing" (see InspectSite); set by ApplySiteState.
	SiteMode        SiteMode `json:"site_mode,omitempty"`
	ExistingContent int      `json:"existing_content,omitempty"`
	ModeNote        string   `json:"mode_note,omitempty"`

	Permalink        string `json:"permalink"`
	PermalinkPattern string `json:"permalink_pattern,omitempty"`
	// PermalinkKept is true when the new URL pattern reproduces the old URLs;
	// otherwise every changed URL gets a redirect.
	PermalinkKept bool `json:"permalink_kept"`

	SEOPlugin          string         `json:"seo_plugin,omitempty"`
	PageBuilders       []string       `json:"page_builders,omitempty"`
	Forms              []string       `json:"forms,omitempty"`
	Multilingual       []string       `json:"multilingual,omitempty"`
	DefaultLang        string         `json:"default_language,omitempty"`
	Languages          map[string]int `json:"languages,omitempty"`
	Shortcodes         map[string]int `json:"shortcodes,omitempty"`
	UnsupportedPlugins []string       `json:"unsupported_plugins,omitempty"`

	LowFidelity []ItemNote `json:"low_fidelity,omitempty"`
	MembersOnly []ItemNote `json:"members_only,omitempty"`
	Embeds      int        `json:"embeds"`
	Fidelity    Fidelity   `json:"fidelity"`
	Warnings    []string   `json:"warnings,omitempty"`
}

type Result struct {
	RunID uuid.UUID `json:"run_id"`

	Created int `json:"created"`
	Updated int `json:"updated"`
	Failed  int `json:"failed"`

	Terms         int `json:"terms"`
	Authors       int `json:"authors"`
	MediaImported int `json:"media_imported"`
	MediaReused   int `json:"media_reused"`
	MediaFailed   int `json:"media_failed"`
	Menus         int `json:"menus"`
	Redirects     int `json:"redirects"`

	MediaRewritten int      `json:"media_rewritten"`
	LinksRewritten int      `json:"links_rewritten"`
	Unresolved     []string `json:"unresolved,omitempty"`
	// WPContentLeft lists wp-content URLs still present in imported content;
	// the importer's promise is that it is empty.
	WPContentLeft []string `json:"wp_content_left,omitempty"`

	// SiteMode is how the target was treated: "empty" or "existing".
	SiteMode SiteMode `json:"site_mode"`
	// URLConflicts are old URLs that could not be redirected because a live
	// page or a redirect of the owner already answers there.
	URLConflicts []URLConflict `json:"url_conflicts,omitempty"`
	// RedirectsRemoved counts redirects of an earlier run that are no longer needed.
	RedirectsRemoved int `json:"redirects_removed,omitempty"`
	// OverwrittenEdits and KeptEdits list entities edited in Hatamoto since the
	// last import, which a re-run overwrote or (with KeepEdited) left alone.
	OverwrittenEdits []string `json:"overwritten_edits,omitempty"`
	KeptEdits        []string `json:"kept_edits,omitempty"`

	Audit    *AuditResult `json:"audit,omitempty"`
	Errors   []ItemError  `json:"errors,omitempty"`
	Warnings []string     `json:"warnings,omitempty"`
	Report   *Report      `json:"report,omitempty"`
}

type RewriteStats struct {
	MediaRewritten int
	LinksRewritten int
	// Unresolved are wp-content URLs that could not be mapped; the element
	// carrying them was removed (an img) or unlinked (an anchor).
	Unresolved []string
}

type Rewriter struct {
	// Hosts are the hostnames of the old site (without port).
	Hosts map[string]bool
	// Media maps a URL of the old site's files to the new media URL. ok is
	// false when the file could not be imported.
	Media func(u *url.URL) (newURL string, ok bool)
	// Entity resolves a path (or ?p= / ?page_id= id) of the old site to an
	// imported entity.
	Entity func(path string, query url.Values) (EntityRef, bool)
	// Path maps any other old-site path to its new one (terms, authors); nil keeps it.
	Path func(path string) (string, bool)
}

type Run struct {
	ID        uuid.UUID
	Source    string
	SourceRef string
	Status    string
	Phase     string
	Options   json.RawMessage
	Report    json.RawMessage
	Error     string
}

type Runner struct {
	// contains filtered or unexported fields
}

type SEO struct {
	Title       string
	Description string
	Canonical   string
	OGImage     string
	NoIndex     bool
	Source      string // yoast, rankmath, seopress, aioseo, html
}

type Selection struct {
	// Types lists the post types to import ("post", "page", a custom type);
	// nil means all of them.
	Types []string
	// Drafts includes drafts, pending, private and password-protected items
	// (imported as drafts). Off, they are left out.
	Drafts bool
	// Menus includes the navigation menus.
	Menus bool
}

type Site struct {
	URL         string // home URL, no trailing slash
	Title       string
	Tagline     string
	Language    string // BCP 47-ish ("en-US"); empty when unknown
	Timezone    string
	ShowOnFront string // "posts" or "page"
	FrontPageID int
	PostsPageID int
	// Permalink is the WordPress structure, for example "/%year%/%monthnum%/%postname%/",
	// inferred from the links of posts ("" when unknown).
	Permalink string
}

type SiteMode string

type SiteState struct {
	Mode SiteMode `json:"mode"`
	// Existing counts published entities that are neither starter content nor
	// imported from this source.
	Existing int `json:"existing"`
	// Starter counts untouched starter and first-boot entities (ignored).
	Starter int `json:"starter"`
	// Examples are a few slugs of the existing content.
	Examples []string `json:"examples,omitempty"`
}

type Store struct {
	// contains filtered or unexported fields
}

type Term struct {
	Taxonomy    string // "category", "post_tag" or custom
	Slug        string
	Name        string
	Description string
	ParentSlug  string
	Link        string
}

type TermRef struct{ Taxonomy, Slug string }

type URLConflict struct {
	URL    string `json:"url"`
	Target string `json:"target,omitempty"` // where the imported item lives
	Reason string `json:"reason"`
}

type WXROptions struct {
	MaxBytes int64 // 0: DefaultWXRMaxBytes
	MaxItems int   // 0: DefaultMaxItems
	// MaxElements caps the elements of the whole file (0: DefaultMaxWXRElements).
	MaxElements int
}