Add html builtin module

2023-06-25 13:07:38 -07:00 · 2023-06-25 13:07:38 -07:00 · f5c8ba2282
parent a30d805f9b
commit f5c8ba2282
6 changed files with 255 additions and 3 deletions
--- a/.gitignore
+++ b/.gitignore
@ -1,3 +1,4 @@
 /lure-updater.toml
 /lure-updater
-/dist/
+/dist/
+*.star
--- a/go.mod
+++ b/go.mod
@ -3,6 +3,7 @@ module go.elara.ws/lure-updater
 go 1.20

 require (
+	github.com/PuerkitoBio/goquery v1.8.1
 	github.com/caarlos0/env/v8 v8.0.0
 	github.com/go-git/go-git/v5 v5.7.0
 	github.com/pelletier/go-toml/v2 v2.0.8
@ -21,6 +22,7 @@ require (
 	github.com/Microsoft/go-winio v0.5.2 // indirect
 	github.com/ProtonMail/go-crypto v0.0.0-20230518184743-7afd39499903 // indirect
 	github.com/acomagu/bufpipe v1.0.4 // indirect
+	github.com/andybalholm/cascadia v1.3.1 // indirect
 	github.com/cloudflare/circl v1.3.3 // indirect
 	github.com/emirpasic/gods v1.18.1 // indirect
 	github.com/go-git/gcfg v1.5.1-0.20230307220236-3a3c6141e376 // indirect
--- a/go.sum
+++ b/go.sum
@ -4,8 +4,12 @@ github.com/Microsoft/go-winio v0.5.2 h1:a9IhgEQBCUEk6QCdml9CiJGhAws+YwffDHEMp1VM
 github.com/Microsoft/go-winio v0.5.2/go.mod h1:WpS1mjBmmwHBEWmogvA2mj8546UReBk4v8QkMxJ6pZY=
 github.com/ProtonMail/go-crypto v0.0.0-20230518184743-7afd39499903 h1:ZK3C5DtzV2nVAQTx5S5jQvMeDqWtD1By5mOoyY/xJek=
 github.com/ProtonMail/go-crypto v0.0.0-20230518184743-7afd39499903/go.mod h1:8TI4H3IbrackdNgv+92dI+rhpCaLqM0IfpgCgenFvRE=
+github.com/PuerkitoBio/goquery v1.8.1 h1:uQxhNlArOIdbrH1tr0UXwdVFgDcZDrZVdcpygAcwmWM=
+github.com/PuerkitoBio/goquery v1.8.1/go.mod h1:Q8ICL1kNUJ2sXGoAhPGUdYDJvgQgHzJsnnd3H7Ho5jQ=
 github.com/acomagu/bufpipe v1.0.4 h1:e3H4WUzM3npvo5uv95QuJM3cQspFNtFBzvJ2oNjKIDQ=
 github.com/acomagu/bufpipe v1.0.4/go.mod h1:mxdxdup/WdsKVreO5GpW4+M/1CE2sMG4jeGJ2sYmHc4=
+github.com/andybalholm/cascadia v1.3.1 h1:nhxRkql1kdYCc8Snf7D5/D3spOX+dBgjA6u8x004T2c=
+github.com/andybalholm/cascadia v1.3.1/go.mod h1:R4bJ1UQfqADjvDa4P6HZHLh/3OxWWEqc0Sk8XGwHqvA=
 github.com/anmitsu/go-shlex v0.0.0-20200514113438-38f4b401e2be h1:9AeTilPcZAjCFIImctFaOjnTIavg87rW78vTPkQqLI8=
 github.com/armon/go-socks5 v0.0.0-20160902184237-e75332964ef5 h1:0CwZNZbxp69SHPdPJAN/hZIm0C4OItdklCFmMRWYpio=
 github.com/bwesterb/go-ristretto v1.2.0/go.mod h1:fUIoIZaG73pV5biE2Blr2xEzDoMj7NFEuV9ekS419A0=
@ -124,8 +128,8 @@ go.elara.ws/logger v0.0.0-20230421022458-e80700db2090 h1:RVC8XvWo6Yw4HUshqx4TSzu
 go.elara.ws/logger v0.0.0-20230421022458-e80700db2090/go.mod h1:qng49owViqsW5Aey93lwBXONw20oGbJIoLVscB16mPM=
 go.elara.ws/pcre v0.0.0-20230421030233-daf2d2e6973f h1:ZwR0xvBeP5BHHv63fgfuwhZIj+Si5rp79WSDUE73ZVA=
 go.elara.ws/pcre v0.0.0-20230421030233-daf2d2e6973f/go.mod h1:EF48C6VnP4wBayzFGk6lXqbiLucH7EfiaYOgiiCe5k4=
-go.elara.ws/vercmp v0.0.0-20230622213404-1ef50f776de8 h1:Q4fiNpQH1qOlPUVt/cVoW3VtIvOe+jo7mFTX0Sor7O4=
-go.elara.ws/vercmp v0.0.0-20230622213404-1ef50f776de8/go.mod h1:/7PNW7nFnDR5W7UXZVc04gdVLR/wBNgkm33KgIz0OBk=
+go.elara.ws/vercmp v0.0.0-20230622214216-0b2b067575c4 h1:Ep54XceQlKhcCHl9awG+wWP4kz4kIP3c3Lzw/Gc/zwY=
+go.elara.ws/vercmp v0.0.0-20230622214216-0b2b067575c4/go.mod h1:/7PNW7nFnDR5W7UXZVc04gdVLR/wBNgkm33KgIz0OBk=
 go.etcd.io/bbolt v1.3.7 h1:j+zJOnnEjF/kyHlDDgGnVL/AIqIJPq8UoB2GSNfkUfQ=
 go.etcd.io/bbolt v1.3.7/go.mod h1:N9Mkw9X8x5fupy0IKsmuqVtoGDyxsaDlbk4Rd05IAQw=
 go.starlark.net v0.0.0-20230525235612-a134d8f9ddca h1:VdD38733bfYv5tUZwEIskMM93VanwNIi5bIKnDrJdEY=
@ -155,9 +159,11 @@ golang.org/x/net v0.0.0-20190404232315-eb5bcb51f2a3/go.mod h1:t9HGtf8HONx5eT2rtn
 golang.org/x/net v0.0.0-20190620200207-3b0461eec859/go.mod h1:z5CRVTTTmAJ677TzLLGU+0bjPO0LkuOLi4/5GtJWs/s=
 golang.org/x/net v0.0.0-20201021035429-f5854403a974/go.mod h1:sp8m0HH+o8qH0wwXwYZr8TS3Oi6o0r6Gce1SSxlDquU=
 golang.org/x/net v0.0.0-20210226172049-e18ecbb05110/go.mod h1:m0MpNAwzfU5UDzcl9v0D8zg8gWTRqZa9RBIspLL5mdg=
+golang.org/x/net v0.0.0-20210916014120-12bc252f5db8/go.mod h1:9nx3DQGgdP8bBQD5qxJ1jj9UTztislL4KSBs9R2vV5Y=
 golang.org/x/net v0.0.0-20211112202133-69e39bad7dc2/go.mod h1:9nx3DQGgdP8bBQD5qxJ1jj9UTztislL4KSBs9R2vV5Y=
 golang.org/x/net v0.0.0-20220722155237-a158d28d115b/go.mod h1:XRhObCWvk6IyKnWLug+ECip1KBveYUHfp+8e9klMJ9c=
 golang.org/x/net v0.6.0/go.mod h1:2Tu9+aMcznHK/AK1HMvgo6xiTLG5rD5rZLDS+rp2Bjs=
+golang.org/x/net v0.7.0/go.mod h1:2Tu9+aMcznHK/AK1HMvgo6xiTLG5rD5rZLDS+rp2Bjs=
 golang.org/x/net v0.8.0/go.mod h1:QVkue5JL9kW//ek3r6jTKnTFis1tRmNAW2P1shuFdJc=
 golang.org/x/net v0.10.0 h1:X2//UzNDwYmtCLn7To6G58Wr6f5ahEAQgKNzv9Y951M=
 golang.org/x/net v0.10.0/go.mod h1:0qNGK6F8kojg2nk9dLZ2mShWaEBan6FAoqfSigmmuDg=
--- a/internal/builtins/html.go
+++ b/internal/builtins/html.go
@ -0,0 +1,229 @@
+package builtins
+
+import (
+	"io"
+	"strings"
+
+	"github.com/PuerkitoBio/goquery"
+	"go.starlark.net/starlark"
+	"go.starlark.net/starlarkstruct"
+)
+
+var (
+	_ starlark.Iterable  = (*starlarkSelection)(nil)
+	_ starlark.Sliceable = (*starlarkSelection)(nil)
+	_ starlark.Sequence  = (*starlarkSelection)(nil)
+	_ starlark.Value     = (*starlarkSelection)(nil)
+)
+
+var htmlModule = &starlarkstruct.Module{
+	Name: "html",
+	Members: starlark.StringDict{
+		"parse": starlark.NewBuiltin("html.parse", htmlParse),
+	},
+}
+
+func htmlParse(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	var val starlark.Value
+	err := starlark.UnpackArgs("html.selection.find", args, kwargs, "from", &val)
+	if err != nil {
+		return nil, err
+	}
+
+	var r io.ReadCloser
+	switch val := val.(type) {
+	case starlark.String:
+		r = io.NopCloser(strings.NewReader(string(val)))
+	case starlark.Bytes:
+		r = io.NopCloser(strings.NewReader(string(val)))
+	case starlarkReader:
+		r = val
+	}
+	defer r.Close()
+
+	doc, err := goquery.NewDocumentFromReader(r)
+	if err != nil {
+		return nil, err
+	}
+
+	return newStarlarkSelection(doc.Selection), nil
+}
+
+type starlarkSelection struct {
+	sel *goquery.Selection
+	*starlarkstruct.Struct
+}
+
+func newStarlarkSelection(sel *goquery.Selection) starlarkSelection {
+	ss := starlarkSelection{sel: sel}
+	ss.Struct = starlarkstruct.FromStringDict(starlark.String("html.selection"), starlark.StringDict{
+		"text":           starlark.NewBuiltin("html.selection.text", ss.text),
+		"html":           starlark.NewBuiltin("html.selection.html", ss.html),
+		"children":       starlark.NewBuiltin("html.selection.children", ss.children),
+		"parent":         starlark.NewBuiltin("html.selection.parent", ss.parent),
+		"find":           starlark.NewBuiltin("html.selection.find", ss.find),
+		"add":            starlark.NewBuiltin("html.selection.add", ss.add),
+		"attr":           starlark.NewBuiltin("html.selection.attr", ss.attr),
+		"has_class":      starlark.NewBuiltin("html.selection.has_class", ss.hasClass),
+		"index_selector": starlark.NewBuiltin("html.selection.index_selector", ss.indexSelector),
+		"and_self":       starlark.NewBuiltin("html.selection.and_self", ss.andSelf),
+		"first":          starlark.NewBuiltin("html.selection.first", ss.first),
+		"last":           starlark.NewBuiltin("html.selection.last", ss.last),
+		"next":           starlark.NewBuiltin("html.selection.last", ss.next),
+		"next_all":       starlark.NewBuiltin("html.selection.next_all", ss.nextAll),
+		"next_until":     starlark.NewBuiltin("html.selection.next_until", ss.nextUntil),
+		"prev":           starlark.NewBuiltin("html.selection.prev", ss.prev),
+		"prev_all":       starlark.NewBuiltin("html.selection.prev_all", ss.prevAll),
+		"prev_until":     starlark.NewBuiltin("html.selection.prev_until", ss.prevUntil),
+	})
+	return ss
+}
+
+func (ss starlarkSelection) Truth() starlark.Bool {
+	return len(ss.sel.Nodes) > 0
+}
+
+func (ss starlarkSelection) Len() int {
+	return ss.sel.Length()
+}
+
+func (ss starlarkSelection) Index(i int) starlark.Value {
+	return newStarlarkSelection(ss.sel.Slice(i, i+1))
+}
+
+func (ss starlarkSelection) Slice(start, end, _ int) starlark.Value {
+	return newStarlarkSelection(ss.sel.Slice(start, end))
+}
+
+func (ss starlarkSelection) Iterate() starlark.Iterator {
+	return newSelectionIterator(ss.sel)
+}
+
+func (ss starlarkSelection) text(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	return starlark.String(ss.sel.Text()), nil
+}
+
+func (ss starlarkSelection) html(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	s, err := ss.sel.Html()
+	return starlark.String(s), err
+}
+
+func (ss starlarkSelection) children(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	return newStarlarkSelection(ss.sel.Children()), nil
+}
+
+func (ss starlarkSelection) parent(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	return newStarlarkSelection(ss.sel.Parent()), nil
+}
+
+func (ss starlarkSelection) find(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	var selector string
+	err := starlark.UnpackArgs("html.selection.find", args, kwargs, "selector", &selector)
+	if err != nil {
+		return nil, err
+	}
+
+	return newStarlarkSelection(ss.sel.Find(selector)), nil
+}
+
+func (ss starlarkSelection) add(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	var selector string
+	err := starlark.UnpackArgs("html.selection.add", args, kwargs, "selector", &selector)
+	if err != nil {
+		return nil, err
+	}
+	return newStarlarkSelection(ss.sel.Add(selector)), nil
+}
+
+func (ss starlarkSelection) indexSelector(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	var selector string
+	err := starlark.UnpackArgs("html.selection.index_selector", args, kwargs, "selector", &selector)
+	if err != nil {
+		return nil, err
+	}
+	return starlark.MakeInt(ss.sel.IndexSelector(selector)), nil
+}
+
+func (ss starlarkSelection) attr(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	var name, def string
+	err := starlark.UnpackArgs("html.selection.find", args, kwargs, "name", &name, "default??", &def)
+	if err != nil {
+		return nil, err
+	}
+	return starlark.String(ss.sel.AttrOr(name, def)), nil
+}
+
+func (ss starlarkSelection) hasClass(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	var name string
+	err := starlark.UnpackArgs("html.selection.has_class", args, kwargs, "name", &name)
+	if err != nil {
+		return nil, err
+	}
+	return starlark.Bool(ss.sel.HasClass(name)), nil
+}
+
+func (ss starlarkSelection) andSelf(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	return newStarlarkSelection(ss.sel.AndSelf()), nil
+}
+
+func (ss starlarkSelection) first(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	return newStarlarkSelection(ss.sel.First()), nil
+}
+
+func (ss starlarkSelection) last(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	return newStarlarkSelection(ss.sel.Last()), nil
+}
+
+func (ss starlarkSelection) next(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	return newStarlarkSelection(ss.sel.Next()), nil
+}
+
+func (ss starlarkSelection) nextAll(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	return newStarlarkSelection(ss.sel.NextAll()), nil
+}
+
+func (ss starlarkSelection) nextUntil(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	var selector string
+	err := starlark.UnpackArgs("html.selection.next_until", args, kwargs, "selector", &selector)
+	if err != nil {
+		return nil, err
+	}
+	return newStarlarkSelection(ss.sel.NextUntil(selector)), nil
+}
+
+func (ss starlarkSelection) prev(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	return newStarlarkSelection(ss.sel.Prev()), nil
+}
+
+func (ss starlarkSelection) prevAll(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	return newStarlarkSelection(ss.sel.PrevAll()), nil
+}
+
+func (ss starlarkSelection) prevUntil(thread *starlark.Thread, fn *starlark.Builtin, args starlark.Tuple, kwargs []starlark.Tuple) (starlark.Value, error) {
+	var selector string
+	err := starlark.UnpackArgs("html.selection.prev_until", args, kwargs, "selector", &selector)
+	if err != nil {
+		return nil, err
+	}
+	return newStarlarkSelection(ss.sel.PrevUntil(selector)), nil
+}
+
+type starlarkSelectionIterator struct {
+	sel   *goquery.Selection
+	index int
+}
+
+func newSelectionIterator(sel *goquery.Selection) *starlarkSelectionIterator {
+	return &starlarkSelectionIterator{sel: sel}
+}
+
+func (ssi *starlarkSelectionIterator) Next(v *starlark.Value) bool {
+	if ssi.index == ssi.sel.Length() {
+		return false
+	}
+	*v = newStarlarkSelection(ssi.sel.Slice(ssi.index, ssi.index+1))
+	ssi.index++
+	return true
+}
+
+func (ssi *starlarkSelectionIterator) Done() {}
--- a/internal/builtins/reader.go
+++ b/internal/builtins/reader.go
@ -200,3 +200,16 @@ func (sr starlarkReader) closeReader(thread *starlark.Thread, b *starlark.Builti
 	}
 	return starlark.None, nil
 }
+
+// Read implements the io.ReadCloser interface
+func (sr starlarkReader) Read(b []byte) (int, error) {
+	return sr.br.Read(b)
+}
+
+// Close implements the io.ReadCloser interface
+func (sr starlarkReader) Close() error {
+	if sr.closeFunc != nil {
+		return sr.closeFunc()
+	}
+	return nil
+}
--- a/internal/builtins/register.go
+++ b/internal/builtins/register.go
@ -44,5 +44,6 @@ func Register(sd starlark.StringDict, opts *Options) {
 	sd["log"] = logModule(opts.Name)
 	sd["json"] = starlarkjson.Module
 	sd["utils"] = utilsModule
+	sd["html"] = htmlModule
 	sd["register_webhook"] = registerWebhook(opts.Mux, opts.Config, opts.Name)
 }