first commit

This commit is contained in:
Tomas Dvorak
2026-02-22 10:42:17 +01:00
commit 55885a0e8f
239 changed files with 103690 additions and 0 deletions
+369
View File
@@ -0,0 +1,369 @@
package javadocs
import (
"net/url"
"strings"
"time"
"github.com/PuerkitoBio/goquery"
)
type Parser struct {
baseURL string
}
func NewParser() *Parser {
return &Parser{
baseURL: "https://docs.oracle.com",
}
}
func (p *Parser) ParsePackagePage(html string, docURL string) (*Package, error) {
doc, err := goquery.NewDocumentFromReader(strings.NewReader(html))
if err != nil {
return nil, err
}
pkg := &Package{
DocURL: docURL,
FetchedAt: time.Now(),
}
pkg.Name = p.extractPackageName(doc)
pkg.Doc = p.extractPackageDoc(doc)
pkg.Classes = p.extractClasses(doc, pkg.Name, docURL)
pkg.Interfaces = p.extractInterfaces(doc, pkg.Name, docURL)
pkg.Enums = p.extractEnums(doc, pkg.Name, docURL)
pkg.Exceptions = p.extractExceptions(doc, pkg.Name, docURL)
return pkg, nil
}
func (p *Parser) ParseSearchResults(html string) ([]*SearchResult, error) {
doc, err := goquery.NewDocumentFromReader(strings.NewReader(html))
if err != nil {
return nil, err
}
var results []*SearchResult
doc.Find(".result").Each(func(i int, s *goquery.Selection) {
result := &SearchResult{}
link := s.Find("a").First()
result.Name = strings.TrimSpace(link.Text())
if href, exists := link.Attr("href"); exists {
result.DocURL = resolveURL(p.baseURL, href)
}
result.Kind = s.Find(".result-kind").Text()
result.QualName = s.Find(".qualified-name").Text()
result.Package = s.Find(".package").Text()
result.Doc = strings.TrimSpace(s.Find(".description").Text())
results = append(results, result)
})
return results, nil
}
func (p *Parser) extractPackageName(doc *goquery.Document) string {
title := doc.Find("h1, .title").First().Text()
title = strings.TrimSpace(title)
if strings.Contains(title, "Package") {
parts := strings.Fields(title)
for i, part := range parts {
if part == "Package" && i+1 < len(parts) {
return parts[i+1]
}
}
}
if title != "" {
return title
}
return ""
}
func (p *Parser) extractPackageDoc(doc *goquery.Document) string {
docblock := doc.Find(".block, .description, #package-description").First()
return strings.TrimSpace(docblock.Text())
}
func (p *Parser) extractClasses(doc *goquery.Document, pkgName string, docURL string) []*Class {
var classes []*Class
doc.Find("table.type-summary tr, .class-summary .member, section.class tbody tr").Each(func(_ int, s *goquery.Selection) {
class := &Class{
Package: pkgName,
Kind: ClassKindClass,
}
link := s.Find("a").First()
class.Name = strings.TrimSpace(link.Text())
if class.Name == "" {
class.Name = strings.TrimSpace(s.Find(".member-name, td:first-child").Text())
}
if href, exists := link.Attr("href"); exists {
class.DocURL = resolveURL(docURL, href)
class.QualifiedName = pkgName + "." + class.Name
}
class.Doc = strings.TrimSpace(s.Find(".member-summary, td:last-child").Text())
if class.Name != "" && !strings.Contains(class.Name, "interface") {
classes = append(classes, class)
}
})
return classes
}
func (p *Parser) extractInterfaces(doc *goquery.Document, pkgName string, docURL string) []*Class {
var interfaces []*Class
doc.Find("table.interface-summary tr, .interface-summary .member").Each(func(_ int, s *goquery.Selection) {
iface := &Class{
Package: pkgName,
Kind: ClassKindInterface,
}
link := s.Find("a").First()
iface.Name = strings.TrimSpace(link.Text())
if iface.Name == "" {
iface.Name = strings.TrimSpace(s.Find(".member-name").Text())
}
if href, exists := link.Attr("href"); exists {
iface.DocURL = resolveURL(docURL, href)
iface.QualifiedName = pkgName + "." + iface.Name
}
iface.Doc = strings.TrimSpace(s.Find(".member-summary, td:last-child").Text())
if iface.Name != "" {
interfaces = append(interfaces, iface)
}
})
return interfaces
}
func (p *Parser) extractEnums(doc *goquery.Document, pkgName string, docURL string) []*Enum {
var enums []*Enum
doc.Find("table.enum-summary tr, .enum-summary .member").Each(func(_ int, s *goquery.Selection) {
enum := &Enum{
Package: pkgName,
}
link := s.Find("a").First()
enum.Name = strings.TrimSpace(link.Text())
if enum.Name == "" {
enum.Name = strings.TrimSpace(s.Find(".member-name").Text())
}
if href, exists := link.Attr("href"); exists {
enum.DocURL = resolveURL(docURL, href)
enum.QualifiedName = pkgName + "." + enum.Name
}
enum.Doc = strings.TrimSpace(s.Find(".member-summary, td:last-child").Text())
if enum.Name != "" {
enums = append(enums, enum)
}
})
return enums
}
func (p *Parser) extractExceptions(doc *goquery.Document, pkgName string, docURL string) []*Class {
var exceptions []*Class
doc.Find("table.exception-summary tr, .exception-summary .member").Each(func(_ int, s *goquery.Selection) {
exc := &Class{
Package: pkgName,
Kind: ClassKindClass,
}
link := s.Find("a").First()
exc.Name = strings.TrimSpace(link.Text())
if exc.Name == "" {
exc.Name = strings.TrimSpace(s.Find(".member-name").Text())
}
if href, exists := link.Attr("href"); exists {
exc.DocURL = resolveURL(docURL, href)
exc.QualifiedName = pkgName + "." + exc.Name
}
exc.Doc = strings.TrimSpace(s.Find(".member-summary, td:last-child").Text())
if exc.Name != "" {
exceptions = append(exceptions, exc)
}
})
return exceptions
}
func (p *Parser) ParseClassPage(html string, docURL string) (*Class, error) {
doc, err := goquery.NewDocumentFromReader(strings.NewReader(html))
if err != nil {
return nil, err
}
class := &Class{
DocURL: docURL,
}
header := doc.Find(".header, h1, .class-name").First()
class.Name = strings.TrimSpace(header.Text())
class.QualifiedName = class.Name
if idx := strings.LastIndex(class.Name, "."); idx > 0 {
class.Package = class.Name[:idx]
class.Name = class.Name[idx+1:]
}
class.Doc = strings.TrimSpace(doc.Find(".block, .description, .class-description").First().Text())
class.Methods = p.extractMethods(doc, class.Name, docURL)
class.Fields = p.extractFields(doc, class.Name, docURL)
class.Constructors = p.extractConstructors(doc, class.Name, docURL)
return class, nil
}
func (p *Parser) extractMethods(doc *goquery.Document, className string, docURL string) []*Method {
var methods []*Method
doc.Find("table.method-summary tr, .method-summary .member, section.method-detail > ul > li").Each(func(_ int, s *goquery.Selection) {
method := &Method{
IsConstructor: false,
}
link := s.Find("a").First()
method.Name = strings.TrimSpace(link.Text())
if method.Name == "" {
sig := s.Find(".member-signature, code").Text()
method.Name = extractMethodName(sig)
}
sigEl := s.Find(".member-signature, code, .sig")
method.Signature = strings.TrimSpace(sigEl.Text())
if id, exists := s.Attr("id"); exists {
method.DocURL = docURL + "#" + id
method.QualifiedName = className + "." + method.Name
} else if href, exists := link.Attr("href"); exists {
method.DocURL = resolveURL(docURL, href)
method.QualifiedName = className + "." + method.Name
}
method.Doc = strings.TrimSpace(s.Find(".block, .member-summary, dd").First().Text())
if method.Name != "" {
methods = append(methods, method)
}
})
return methods
}
func (p *Parser) extractFields(doc *goquery.Document, className string, docURL string) []*Field {
var fields []*Field
doc.Find("table.field-summary tr, .field-summary .member").Each(func(_ int, s *goquery.Selection) {
field := &Field{}
link := s.Find("a").First()
field.Name = strings.TrimSpace(link.Text())
if field.Name == "" {
field.Name = strings.TrimSpace(s.Find(".member-name, td:first-child").Text())
}
field.Type = strings.TrimSpace(s.Find(".member-type, td:nth-child(2)").Text())
field.Doc = strings.TrimSpace(s.Find(".member-summary, td:last-child").Text())
if id, exists := s.Attr("id"); exists {
field.DocURL = docURL + "#" + id
}
if field.Name != "" {
fields = append(fields, field)
}
})
return fields
}
func (p *Parser) extractConstructors(doc *goquery.Document, className string, docURL string) []*Method {
var constructors []*Method
doc.Find("table.constructor-summary tr, .constructor-summary .member").Each(func(_ int, s *goquery.Selection) {
ctor := &Method{
IsConstructor: true,
Name: className,
}
link := s.Find("a").First()
if name := strings.TrimSpace(link.Text()); name != "" {
ctor.Name = name
}
sigEl := s.Find(".member-signature, code")
ctor.Signature = strings.TrimSpace(sigEl.Text())
ctor.Doc = strings.TrimSpace(s.Find(".block, .member-summary, td:last-child").Text())
if id, exists := s.Attr("id"); exists {
ctor.DocURL = docURL + "#" + id
}
constructors = append(constructors, ctor)
})
return constructors
}
func extractMethodName(sig string) string {
sig = strings.TrimSpace(sig)
if idx := strings.Index(sig, "("); idx > 0 {
prefix := sig[:idx]
parts := strings.Fields(prefix)
if len(parts) > 0 {
return parts[len(parts)-1]
}
}
return ""
}
func resolveURL(base string, href string) string {
if strings.HasPrefix(href, "http") {
return href
}
baseURL, err := url.Parse(base)
if err != nil {
return href
}
hrefURL, err := url.Parse(href)
if err != nil {
return href
}
return baseURL.ResolveReference(hrefURL).String()
}
+115
View File
@@ -0,0 +1,115 @@
package javadocs
import (
"strings"
"testing"
"github.com/PuerkitoBio/goquery"
)
const testPackagePageHTML = `
<!DOCTYPE html>
<html>
<body>
<h1>Package java.util</h1>
<div class="block">Contains the collections framework, legacy collection classes, event model, date and time facilities.</div>
<table class="type-summary">
<tr>
<td><a href="ArrayList.html">ArrayList</a></td>
<td>Resizable-array implementation of the List interface.</td>
</tr>
<tr>
<td><a href="HashMap.html">HashMap</a></td>
<td>Hash table based implementation of the Map interface.</td>
</tr>
</table>
<table class="interface-summary">
<tr>
<td><a href="List.html">List</a></td>
<td>An ordered collection (also known as a sequence).</td>
</tr>
</table>
<table class="exception-summary">
<tr>
<td><a href="ConcurrentModificationException.html">ConcurrentModificationException</a></td>
<td>This exception may be thrown by methods that detect concurrent modification.</td>
</tr>
</table>
</body>
</html>
`
func TestParsePackagePage(t *testing.T) {
parser := NewParser()
pkg, err := parser.ParsePackagePage(testPackagePageHTML, "https://docs.oracle.com/en/java/javase/17/docs/api/java.base/java/util/package-summary.html")
if err != nil {
t.Fatalf("ParsePackagePage failed: %v", err)
}
if pkg.Name == "" {
t.Error("Expected non-empty package name")
}
if pkg.Doc == "" {
t.Error("Expected non-empty doc")
}
if len(pkg.Classes) == 0 {
t.Error("Expected at least one class")
}
if len(pkg.Interfaces) == 0 {
t.Error("Expected at least one interface")
}
if len(pkg.Exceptions) == 0 {
t.Error("Expected at least one exception")
}
}
func TestExtractClasses(t *testing.T) {
parser := NewParser()
doc, err := goquery.NewDocumentFromReader(strings.NewReader(testPackagePageHTML))
if err != nil {
t.Fatalf("Failed to parse HTML: %v", err)
}
classes := parser.extractClasses(doc, "java.util", "https://docs.oracle.com/test")
if len(classes) == 0 {
t.Fatal("Expected at least one class")
}
first := classes[0]
if first.Name == "" {
t.Error("Expected non-empty class name")
}
if first.Package != "java.util" {
t.Errorf("Expected package 'java.util', got %q", first.Package)
}
}
func TestResolveURL(t *testing.T) {
tests := []struct {
base string
href string
expected string
}{
{"https://docs.oracle.com", "/api/ArrayList.html", "https://docs.oracle.com/api/ArrayList.html"},
{"https://docs.oracle.com", "https://example.com/page", "https://example.com/page"},
}
for _, tt := range tests {
t.Run(tt.href, func(t *testing.T) {
got := resolveURL(tt.base, tt.href)
if got != tt.expected {
t.Errorf("resolveURL(%q, %q) = %q, want %q", tt.base, tt.href, got, tt.expected)
}
})
}
}
+116
View File
@@ -0,0 +1,116 @@
// Package javadocs provides parsing and extraction for Java documentation
// from docs.oracle.com and javadoc-generated sites.
package javadocs
import "time"
// Package represents a Java package's documentation.
type Package struct {
Name string `json:"name"`
Doc string `json:"doc,omitempty"`
Version string `json:"version,omitempty"`
DocURL string `json:"doc_url"`
Classes []*Class `json:"classes,omitempty"`
Interfaces []*Class `json:"interfaces,omitempty"`
Enums []*Enum `json:"enums,omitempty"`
Exceptions []*Class `json:"exceptions,omitempty"`
Annotations []*Class `json:"annotations,omitempty"`
FetchedAt time.Time `json:"fetched_at"`
}
// Class represents a Java class or interface.
type Class struct {
QualifiedName string `json:"qualified_name"`
Name string `json:"name"`
Package string `json:"package"`
Kind ClassKind `json:"kind"`
Doc string `json:"doc,omitempty"`
Modifiers []string `json:"modifiers,omitempty"`
SuperClass string `json:"super_class,omitempty"`
Interfaces []string `json:"interfaces,omitempty"`
Fields []*Field `json:"fields,omitempty"`
Methods []*Method `json:"methods,omitempty"`
Constructors []*Method `json:"constructors,omitempty"`
NestedClasses []*Class `json:"nested_classes,omitempty"`
DocURL string `json:"doc_url"`
Since string `json:"since,omitempty"`
Deprecated string `json:"deprecated,omitempty"`
}
// ClassKind represents the kind of class.
type ClassKind string
const (
ClassKindClass ClassKind = "class"
ClassKindInterface ClassKind = "interface"
ClassKindAnnotation ClassKind = "annotation"
ClassKindRecord ClassKind = "record"
ClassKindSealed ClassKind = "sealed"
)
// Enum represents a Java enum.
type Enum struct {
QualifiedName string `json:"qualified_name"`
Name string `json:"name"`
Package string `json:"package"`
Doc string `json:"doc,omitempty"`
Modifiers []string `json:"modifiers,omitempty"`
Constants []*EnumConst `json:"constants,omitempty"`
Methods []*Method `json:"methods,omitempty"`
DocURL string `json:"doc_url"`
Since string `json:"since,omitempty"`
}
// EnumConst represents an enum constant.
type EnumConst struct {
Name string `json:"name"`
Doc string `json:"doc,omitempty"`
DocURL string `json:"doc_url"`
}
// Field represents a class field.
type Field struct {
Name string `json:"name"`
Type string `json:"type"`
Doc string `json:"doc,omitempty"`
Modifiers []string `json:"modifiers,omitempty"`
Value string `json:"value,omitempty"`
DocURL string `json:"doc_url"`
Deprecated string `json:"deprecated,omitempty"`
}
// Method represents a method or constructor.
type Method struct {
Name string `json:"name"`
Doc string `json:"doc,omitempty"`
Modifiers []string `json:"modifiers,omitempty"`
ReturnType string `json:"return_type,omitempty"`
Parameters []*Parameter `json:"parameters,omitempty"`
Exceptions []string `json:"exceptions,omitempty"`
Signature string `json:"signature"`
QualifiedName string `json:"qualified_name"`
DocURL string `json:"doc_url"`
Since string `json:"since,omitempty"`
Deprecated string `json:"deprecated,omitempty"`
IsConstructor bool `json:"is_constructor"`
IsStatic bool `json:"is_static"`
IsDefault bool `json:"is_default"`
}
// Parameter represents a method parameter.
type Parameter struct {
Name string `json:"name"`
Type string `json:"type"`
Doc string `json:"doc,omitempty"`
}
// SearchResult represents a search result.
type SearchResult struct {
Name string `json:"name"`
QualName string `json:"qualified_name"`
Kind string `json:"kind"` // class, interface, enum, method, field
Package string `json:"package"`
Doc string `json:"doc,omitempty"`
DocURL string `json:"doc_url"`
Score int `json:"score"`
}