sam/claude-plugin-runescape
Public Access
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Activity

Update parser

Browse Source
This commit is contained in:
Sam Myers
2026-03-05 12:51:31 -06:00
parent b20d68c9f0
commit c283444d68
7 changed files with 603 additions and 103 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
scripts/rsw/go.mod
View File

@@ -1,10 +1,11 @@
module github.com/runescape-wiki/rsw
go 1.22
go 1.25.0
require github.com/spf13/cobra v1.8.1
require (
github.com/inconshreveable/mousetrap v1.1.0 // indirect
github.com/spf13/pflag v1.0.5 // indirect
golang.org/x/net v0.51.0 // indirect
)

2
scripts/rsw/go.sum
View File

@@ -6,5 +6,7 @@ github.com/spf13/cobra v1.8.1 h1:e5/vxKd/rZsfSJMUX1agtjeTDf+qv1/JdBF8gg5k9ZM=
github.com/spf13/cobra v1.8.1/go.mod h1:wHxEcudfqmLYa8iTfL+OuZPbBZkmvliBWKIezN3kD9Y=
github.com/spf13/pflag v1.0.5 h1:iy+VFUOCP1a+8yFto/drg2CJ5u0yRoB7fZw3DKv/JXA=
github.com/spf13/pflag v1.0.5/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
golang.org/x/net v0.51.0 h1:94R/GTO7mt3/4wIKpcR5gkGmRLOuE/2hNGeWq/GBIFo=
golang.org/x/net v0.51.0/go.mod h1:aamm+2QF5ogm02fjy5Bb7CQ0WMt1/WVM7FtyaTLlA9Y=
gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=

82
scripts/rsw/internal/cmd/page.go
View File

@@ -4,7 +4,7 @@ import (
"fmt"
"strings"
"github.com/runescape-wiki/rsw/internal/extract"
"github.com/runescape-wiki/rsw/internal/htmlconv"
"github.com/runescape-wiki/rsw/internal/render"
"github.com/runescape-wiki/rsw/internal/wiki"
"github.com/spf13/cobra"
@@ -27,80 +27,54 @@ Examples:
title := args[0]
client := wiki.NewClient(GameBaseURL())
page, err := client.GetPage(title)
if Raw() {
page, err := client.GetPage(title)
if err != nil {
return fmt.Errorf("failed to fetch page: %w", err)
}
fmt.Println(page.Wikitext)
return nil
}
page, err := client.GetPageHTML(title)
if err != nil {
return fmt.Errorf("failed to fetch page: %w", err)
}
wikitext := page.Wikitext
if pageSection != "" {
idx := wiki.FindSectionIndex(page.Sections, pageSection)
if idx == -1 {
needle := strings.ToLower(pageSection)
// Case-insensitive exact match
for _, s := range page.Sections {
if strings.ToLower(s.Line) == needle {
fmt.Sscanf(s.Index, "%d", &idx)
break
}
}
// Case-insensitive prefix match (e.g. "Location" → "Locations")
if idx == -1 {
for _, s := range page.Sections {
if strings.HasPrefix(strings.ToLower(s.Line), needle) {
fmt.Sscanf(s.Index, "%d", &idx)
break
}
}
}
// Case-insensitive contains match
if idx == -1 {
for _, s := range page.Sections {
if strings.Contains(strings.ToLower(s.Line), needle) {
fmt.Sscanf(s.Index, "%d", &idx)
break
}
}
}
}
if idx == -1 {
body := htmlconv.ExtractSection(page.HTML, pageSection)
if body == "" {
sections := htmlconv.ListSections(page.HTML)
return fmt.Errorf("section %q not found. Available sections: %s",
pageSection, listSections(page.Sections))
pageSection, formatSectionNames(sections))
}
sectionPage, err := client.GetPageSection(title, idx)
if err != nil {
return fmt.Errorf("failed to fetch section: %w", err)
}
wikitext = sectionPage.Wikitext
}
if Raw() {
fmt.Println(wikitext)
md := render.New()
md.H1(page.Title)
md.Line(body)
fmt.Print(md.String())
return nil
}
md := render.New()
md.H1(page.Title)
if pageSection == "" && len(page.Sections) > 0 {
sections := htmlconv.ListSections(page.HTML)
if len(sections) > 0 {
md.H2("Sections")
for _, s := range page.Sections {
for _, s := range sections {
indent := ""
if s.Level == "3" {
if s.Level == 3 {
indent = " "
} else if s.Level == "4" {
} else if s.Level >= 4 {
indent = " "
}
md.Line(fmt.Sprintf("%s- %s", indent, s.Line))
md.Line(fmt.Sprintf("%s- %s", indent, s.Name))
}
md.Newline()
md.HR()
}
plain := extract.ExtractPlainText(wikitext)
md.P(plain)
md.Line(htmlconv.Convert(page.HTML))
fmt.Print(md.String())
return nil
},
@@ -110,10 +84,10 @@ Examples:
return cmd
}
func listSections(sections []wiki.Section) string {
func formatSectionNames(sections []htmlconv.SectionInfo) string {
names := make([]string, len(sections))
for i, s := range sections {
names[i] = s.Line
names[i] = s.Name
}
return strings.Join(names, ", ")
}

84
scripts/rsw/internal/cmd/skill.go
View File

@@ -4,7 +4,7 @@ import (
"fmt"
"strings"
"github.com/runescape-wiki/rsw/internal/extract"
"github.com/runescape-wiki/rsw/internal/htmlconv"
"github.com/runescape-wiki/rsw/internal/render"
"github.com/runescape-wiki/rsw/internal/wiki"
"github.com/spf13/cobra"
@@ -31,19 +31,27 @@ Examples:
wikiClient := wiki.NewClient(GameBaseURL())
trainingTitle := capitalizeFirst(strings.ToLower(skillName)) + " training"
page, err := wikiClient.GetPage(trainingTitle)
if Raw() {
page, err := wikiClient.GetPage(trainingTitle)
if err != nil {
page, err = wikiClient.GetPage(capitalizeFirst(strings.ToLower(skillName)))
if err != nil {
return fmt.Errorf("failed to fetch skill page: %w", err)
}
}
fmt.Println(page.Wikitext)
return nil
}
page, err := wikiClient.GetPageHTML(trainingTitle)
if err != nil {
page, err = wikiClient.GetPage(capitalizeFirst(strings.ToLower(skillName)))
page, err = wikiClient.GetPageHTML(capitalizeFirst(strings.ToLower(skillName)))
if err != nil {
return fmt.Errorf("failed to fetch skill page: %w", err)
}
}
if Raw() {
fmt.Println(page.Wikitext)
return nil
}
md := render.New()
md.H1(fmt.Sprintf("%s Training Guide", page.Title))
@@ -51,11 +59,13 @@ Examples:
md.P("*Showing methods suitable for ironman accounts (no GE access).*")
}
if len(page.Sections) > 0 {
sections := htmlconv.ListSections(page.HTML)
if len(sections) > 0 {
md.H2("Contents")
for _, s := range page.Sections {
if s.Level == "2" {
md.Bullet(s.Line)
for _, s := range sections {
if s.Level == 2 {
md.Bullet(s.Name)
}
}
md.Newline()
@@ -63,30 +73,23 @@ Examples:
if levelRange != "" {
found := false
for _, s := range page.Sections {
if strings.Contains(strings.ToLower(s.Line), strings.ToLower(levelRange)) ||
sectionMatchesLevelRange(s.Line, levelRange) {
idx := 0
fmt.Sscanf(s.Index, "%d", &idx)
if idx > 0 {
sectionPage, err := wikiClient.GetPageSection(page.Title, idx)
if err == nil {
plain := extract.ExtractPlainText(sectionPage.Wikitext)
if strings.TrimSpace(plain) != "" {
md.H2(s.Line)
md.P(plain)
found = true
}
}
for _, s := range sections {
if strings.Contains(strings.ToLower(s.Name), strings.ToLower(levelRange)) ||
sectionMatchesLevelRange(s.Name, levelRange) {
body := htmlconv.ExtractSection(page.HTML, s.Name)
if strings.TrimSpace(body) != "" {
md.H2(s.Name)
md.Line(body)
found = true
}
}
}
if !found {
md.P(fmt.Sprintf("*No section found matching level range %q. Showing full guide.*", levelRange))
renderFullGuide(md, page, wikiClient)
renderFullGuideHTML(md, page.HTML, sections)
}
} else {
renderFullGuide(md, page, wikiClient)
renderFullGuideHTML(md, page.HTML, sections)
}
fmt.Print(md.String())
@@ -98,24 +101,15 @@ Examples:
return cmd
}
func renderFullGuide(md *render.Builder, page *wiki.ParsedPage, client *wiki.Client) {
for _, s := range page.Sections {
if s.Level != "2" {
func renderFullGuideHTML(md *render.Builder, pageHTML string, sections []htmlconv.SectionInfo) {
for _, s := range sections {
if s.Level != 2 {
continue
}
idx := 0
fmt.Sscanf(s.Index, "%d", &idx)
if idx <= 0 {
continue
}
sectionPage, err := client.GetPageSection(page.Title, idx)
if err != nil {
continue
}
plain := extract.ExtractPlainText(sectionPage.Wikitext)
if strings.TrimSpace(plain) != "" {
md.H2(s.Line)
md.P(plain)
body := htmlconv.ExtractSection(pageHTML, s.Name)
if strings.TrimSpace(body) != "" {
md.H2(s.Name)
md.Line(body)
}
}
}

44
scripts/rsw/internal/extract/infobox.go
View File

@@ -202,6 +202,7 @@ func cleanWikiLinks(s string) string {
// CleanWikitext strips templates, wiki links, and HTML but preserves line structure
// and converts wiki list markers (* items) to readable bullet points.
func CleanWikitext(s string) string {
s = StripTransclusionTags(s)
s = expandKnownTemplates(s)
s = removeTemplates(s)
s = cleanWikiLinks(s)
@@ -239,6 +240,7 @@ func CleanWikitext(s string) string {
// ExtractPlainText strips all wikitext markup to produce plain text.
func ExtractPlainText(wikitext string) string {
s := wikitext
s = StripTransclusionTags(s)
s = expandKnownTemplates(s)
s = removeTemplates(s)
s = cleanWikiLinks(s)
@@ -371,6 +373,48 @@ func tryExpandTemplate(inner string) (string, bool) {
return "", false
}
// StripTransclusionTags handles MediaWiki transclusion directives in raw wikitext.
// When viewing a page directly (not transcluding):
// - <noinclude>content</noinclude> → keep content (strip tags only)
// - <includeonly>content</includeonly> → remove entirely (tags + content)
// - <onlyinclude>content</onlyinclude> → keep content (strip tags only)
func StripTransclusionTags(s string) string {
// Remove <includeonly>...</includeonly> blocks entirely
for {
lower := strings.ToLower(s)
start := strings.Index(lower, "<includeonly>")
if start == -1 {
break
}
end := strings.Index(lower[start:], "</includeonly>")
if end == -1 {
// Unclosed tag — remove to end of string
s = s[:start]
break
}
s = s[:start] + s[start+end+len("</includeonly>"):]
}
// Unwrap <noinclude> and <onlyinclude> (keep content, remove tags)
for _, tag := range []string{"noinclude", "onlyinclude"} {
s = removeTagKeepContent(s, tag)
}
return s
}
func removeTagKeepContent(s, tag string) string {
open := "<" + tag + ">"
close := "</" + tag + ">"
s = strings.ReplaceAll(s, open, "")
s = strings.ReplaceAll(s, close, "")
// Case-insensitive variants
upper := "<" + strings.ToUpper(tag) + ">"
upperClose := "</" + strings.ToUpper(tag) + ">"
s = strings.ReplaceAll(s, upper, "")
s = strings.ReplaceAll(s, upperClose, "")
return s
}
func removeTemplates(s string) string {
var b strings.Builder
depth := 0

483
scripts/rsw/internal/htmlconv/htmlconv.go Normal file
View File

@@ -0,0 +1,483 @@
package htmlconv
import (
"fmt"
"strings"
"golang.org/x/net/html"
"golang.org/x/net/html/atom"
)
// Convert takes MediaWiki-rendered HTML and returns clean markdown.
func Convert(rawHTML string) string {
doc, err := html.Parse(strings.NewReader(rawHTML))
if err != nil {
return rawHTML
}
var w walker
w.walk(doc)
return w.finish()
}
// ExtractSection extracts a single section by heading text and converts it to markdown.
// Matches case-insensitively. Returns empty string if not found.
func ExtractSection(rawHTML string, sectionName string) string {
doc, err := html.Parse(strings.NewReader(rawHTML))
if err != nil {
return ""
}
needle := strings.ToLower(strings.TrimSpace(sectionName))
// Find the mw-parser-output container.
container := findParserOutput(doc)
if container == nil {
return ""
}
// Walk top-level children to find the target heading div, then collect
// siblings until the next heading of equal or lesser depth.
var (
collecting bool
targetLvl int
collected []*html.Node
)
for c := container.FirstChild; c != nil; c = c.NextSibling {
if isHeadingDiv(c) {
lvl, text := headingInfo(c)
if collecting {
if lvl <= targetLvl {
break
}
}
if !collecting && strings.ToLower(strings.TrimSpace(text)) == needle {
collecting = true
targetLvl = lvl
continue
}
}
if collecting {
collected = append(collected, c)
}
}
if len(collected) == 0 {
return ""
}
var w walker
for _, n := range collected {
w.walk(n)
}
return w.finish()
}
// ListSections returns section names and levels from the HTML.
func ListSections(rawHTML string) []SectionInfo {
doc, err := html.Parse(strings.NewReader(rawHTML))
if err != nil {
return nil
}
container := findParserOutput(doc)
if container == nil {
return nil
}
var sections []SectionInfo
for c := container.FirstChild; c != nil; c = c.NextSibling {
if isHeadingDiv(c) {
lvl, text := headingInfo(c)
if text != "" {
sections = append(sections, SectionInfo{Name: text, Level: lvl})
}
}
}
return sections
}
// SectionInfo holds a section heading name and level.
type SectionInfo struct {
Name string
Level int
}
// --- DOM helpers ---
func findParserOutput(n *html.Node) *html.Node {
if n.Type == html.ElementNode && n.DataAtom == atom.Div && hasClass(n, "mw-parser-output") {
return n
}
for c := n.FirstChild; c != nil; c = c.NextSibling {
if found := findParserOutput(c); found != nil {
return found
}
}
return nil
}
func isHeadingDiv(n *html.Node) bool {
return n.Type == html.ElementNode && n.DataAtom == atom.Div && hasClassPrefix(n, "mw-heading")
}
func headingInfo(div *html.Node) (level int, text string) {
for c := div.FirstChild; c != nil; c = c.NextSibling {
if c.Type == html.ElementNode {
switch c.DataAtom {
case atom.H1:
return 1, textContent(c)
case atom.H2:
return 2, textContent(c)
case atom.H3:
return 3, textContent(c)
case atom.H4:
return 4, textContent(c)
case atom.H5:
return 5, textContent(c)
case atom.H6:
return 6, textContent(c)
}
}
}
return 0, ""
}
func textContent(n *html.Node) string {
if n.Type == html.TextNode {
return n.Data
}
// Skip edit section spans.
if n.Type == html.ElementNode && hasClass(n, "mw-editsection") {
return ""
}
var sb strings.Builder
for c := n.FirstChild; c != nil; c = c.NextSibling {
sb.WriteString(textContent(c))
}
return sb.String()
}
func hasClass(n *html.Node, cls string) bool {
for _, a := range n.Attr {
if a.Key == "class" {
for _, c := range strings.Fields(a.Val) {
if c == cls {
return true
}
}
}
}
return false
}
func hasClassPrefix(n *html.Node, prefix string) bool {
for _, a := range n.Attr {
if a.Key == "class" {
for _, c := range strings.Fields(a.Val) {
if strings.HasPrefix(c, prefix) {
return true
}
}
}
}
return false
}
func getAttr(n *html.Node, key string) string {
for _, a := range n.Attr {
if a.Key == key {
return a.Val
}
}
return ""
}
func isHidden(n *html.Node) bool {
style := getAttr(n, "style")
return strings.Contains(style, "display:none") || strings.Contains(style, "display: none")
}
// --- walker: recursive HTML-to-markdown converter ---
type walker struct {
sb strings.Builder
listCtx []listContext
}
type listContext struct {
ordered bool
index int
}
func (w *walker) finish() string {
out := w.sb.String()
// Collapse runs of 3+ newlines to 2.
for strings.Contains(out, "\n\n\n") {
out = strings.ReplaceAll(out, "\n\n\n", "\n\n")
}
return strings.TrimSpace(out) + "\n"
}
func (w *walker) walk(n *html.Node) {
switch n.Type {
case html.TextNode:
w.sb.WriteString(n.Data)
return
case html.ElementNode:
// skip
case html.DocumentNode:
w.walkChildren(n)
return
default:
return
}
if isHidden(n) {
return
}
if w.shouldSkip(n) {
return
}
switch n.DataAtom {
case atom.H1, atom.H2, atom.H3, atom.H4, atom.H5, atom.H6:
w.renderHeading(n)
case atom.P:
w.ensureNewline()
w.walkChildren(n)
w.sb.WriteString("\n\n")
case atom.Br:
w.sb.WriteString("\n")
case atom.B, atom.Strong:
w.sb.WriteString("**")
w.walkChildren(n)
w.sb.WriteString("**")
case atom.I, atom.Em:
w.sb.WriteString("*")
w.walkChildren(n)
w.sb.WriteString("*")
case atom.A:
w.walkChildren(n)
case atom.Ul:
w.renderList(n, false)
case atom.Ol:
w.renderList(n, true)
case atom.Li:
w.renderListItem(n)
case atom.Table:
w.renderTable(n)
case atom.Img:
alt := getAttr(n, "alt")
if alt != "" {
w.sb.WriteString(alt)
}
case atom.Div:
if isHeadingDiv(n) {
lvl, text := headingInfo(n)
if text != "" {
w.ensureNewline()
w.sb.WriteString(strings.Repeat("#", lvl))
w.sb.WriteString(" ")
w.sb.WriteString(text)
w.sb.WriteString("\n\n")
}
return
}
w.walkChildren(n)
default:
w.walkChildren(n)
}
}
func (w *walker) walkChildren(n *html.Node) {
for c := n.FirstChild; c != nil; c = c.NextSibling {
w.walk(c)
}
}
func (w *walker) shouldSkip(n *html.Node) bool {
if n.Type != html.ElementNode {
return false
}
if getAttr(n, "id") == "toc" || getAttr(n, "id") == "toctemplate" {
return true
}
if hasClass(n, "navigation-not-searchable") {
return true
}
if hasClass(n, "mw-editsection") {
return true
}
// Skip infobox tables — noisy in CLI. RS3 uses "rsw-infobox", OSRS uses "infobox".
if n.DataAtom == atom.Table && (hasClass(n, "infobox") || hasClassPrefix(n, "rsw-infobox")) {
return true
}
// Skip navbox (bottom-of-page navigation templates).
if n.DataAtom == atom.Table && hasClass(n, "navbox") {
return true
}
// Skip messagebox / disambig boxes.
if n.DataAtom == atom.Table && hasClass(n, "messagebox") {
return true
}
// Skip "clear" divs.
if n.DataAtom == atom.Div && hasClass(n, "clear-template") {
return true
}
return false
}
func (w *walker) ensureNewline() {
s := w.sb.String()
if len(s) > 0 && s[len(s)-1] != '\n' {
w.sb.WriteString("\n")
}
}
func (w *walker) renderHeading(n *html.Node) {
level := 0
switch n.DataAtom {
case atom.H1:
level = 1
case atom.H2:
level = 2
case atom.H3:
level = 3
case atom.H4:
level = 4
case atom.H5:
level = 5
case atom.H6:
level = 6
}
w.ensureNewline()
w.sb.WriteString(strings.Repeat("#", level))
w.sb.WriteString(" ")
w.sb.WriteString(textContent(n))
w.sb.WriteString("\n\n")
}
func (w *walker) renderList(n *html.Node, ordered bool) {
w.listCtx = append(w.listCtx, listContext{ordered: ordered})
w.ensureNewline()
for c := n.FirstChild; c != nil; c = c.NextSibling {
w.walk(c)
}
w.listCtx = w.listCtx[:len(w.listCtx)-1]
if len(w.listCtx) == 0 {
w.sb.WriteString("\n")
}
}
func (w *walker) renderListItem(n *html.Node) {
depth := len(w.listCtx)
if depth == 0 {
depth = 1
}
indent := strings.Repeat(" ", depth-1)
ctx := &w.listCtx[len(w.listCtx)-1]
if ctx.ordered {
ctx.index++
w.sb.WriteString(fmt.Sprintf("%s%d. ", indent, ctx.index))
} else {
w.sb.WriteString(indent + "- ")
}
// Walk children inline — but nested lists should go on their own lines.
for c := n.FirstChild; c != nil; c = c.NextSibling {
if c.Type == html.ElementNode && (c.DataAtom == atom.Ul || c.DataAtom == atom.Ol) {
w.sb.WriteString("\n")
w.walk(c)
} else {
w.walk(c)
}
}
w.ensureNewline()
}
func (w *walker) renderTable(n *html.Node) {
rows := collectTableRows(n)
if len(rows) == 0 {
return
}
w.ensureNewline()
w.sb.WriteString("\n")
var headers []string
var dataRows [][]string
for i, row := range rows {
var cells []string
allHeaders := true
for _, cell := range row {
text := strings.TrimSpace(textContent(cell))
text = strings.ReplaceAll(text, "|", "/")
text = strings.ReplaceAll(text, "\n", " ")
cells = append(cells, text)
if cell.DataAtom != atom.Th {
allHeaders = false
}
}
if i == 0 && allHeaders {
headers = cells
} else {
dataRows = append(dataRows, cells)
}
}
if headers == nil && len(dataRows) > 0 {
headers = dataRows[0]
dataRows = dataRows[1:]
}
if len(headers) == 0 {
return
}
// Header row.
w.sb.WriteString("| " + strings.Join(headers, " | ") + " |\n")
seps := make([]string, len(headers))
for i := range seps {
seps[i] = "---"
}
w.sb.WriteString("| " + strings.Join(seps, " | ") + " |\n")
for _, row := range dataRows {
for len(row) < len(headers) {
row = append(row, "")
}
w.sb.WriteString("| " + strings.Join(row[:len(headers)], " | ") + " |\n")
}
w.sb.WriteString("\n")
}
func collectTableRows(table *html.Node) [][]*html.Node {
var rows [][]*html.Node
var visit func(*html.Node)
visit = func(n *html.Node) {
if n.Type == html.ElementNode && n.DataAtom == atom.Tr {
if isHidden(n) {
return
}
var cells []*html.Node
for c := n.FirstChild; c != nil; c = c.NextSibling {
if c.Type == html.ElementNode && (c.DataAtom == atom.Th || c.DataAtom == atom.Td) {
cells = append(cells, c)
}
}
if len(cells) > 0 {
rows = append(rows, cells)
}
return
}
for c := n.FirstChild; c != nil; c = c.NextSibling {
visit(c)
}
}
visit(table)
return rows
}

8
scripts/rsw/internal/wiki/parse.go
View File

@@ -97,9 +97,10 @@ func (c *Client) GetPageSection(title string, sectionIndex int) (*ParsedPage, er
// GetPageHTML fetches the rendered HTML for a page.
func (c *Client) GetPageHTML(title string) (*ParsedPage, error) {
params := url.Values{
"action": {"parse"},
"page": {title},
"prop": {"text|sections"},
"action": {"parse"},
"page": {title},
"prop": {"text|sections"},
"redirects": {"1"},
}
var resp parseResponse
@@ -133,3 +134,4 @@ func FindSectionIndex(sections []Section, name string) int {
}
return -1
}