"bytes"
"fmt"
"strings"
"bytes"
"strconv"
"strings"
"unicode"
"bytes"
"fmt"
"strings"
"unicode"
"unicode/utf8"
"bytes"
"fmt"
"strings"
"unicode/utf8"
"golang.org/x/text/cases"
"bytes"
"fmt"
"strings"
"bytes"
"bytes"
"strings"
"strings"
"bytes"
"bytes"
"fmt"
"strings"
"bytes"
"strings"
"bytes"
"fmt"
"reflect"
"slices"
"strings"
var blockType = *ast.CallExprvar blocksType = *ast.CallExpremoji maps known emoji names to their UTF-8 emoji forms.
var emoji = map[string]string{...}htmlEntity maps known HTML entity sequences to their meanings.
var htmlEntity = map[string]string{...}var htmlEscaper = *ast.CallExprvar htmlLinkEscaper = *ast.CallExprvar htmlQuoteEscaper = *ast.CallExprhtmlTags lists the known HTML tags for HTML blocks.
var htmlTags = []string{...}var inlinesType = *ast.CallExprInline code span markers must fit on punched cards, to match cmark-gfm.
const maxBackticks = 80const maxEmojiLen = 36var mdUnescaper = *ast.CallExprvar news = []func(*parseState, line) (line, bool){...}type tableTrimmed stringBlock is implemented by: CodeBLock Document Empty HTMLBlock Heading Item List Paragraph Quote Text ThematicBreak
type Block interface {
Pos() Position
PrintHTML(buf *bytes.Buffer)
printMarkdown(buf *bytes.Buffer, s mdState)
}type Inline interface {
PrintHTML(*bytes.Buffer)
PrintText(*bytes.Buffer)
printMarkdown(*bytes.Buffer)
}type blockBuilder interface {
extend(p *parseState, s line) (line, bool)
build(buildState) Block
}type buildState interface {
blocks() []Block
pos() Position
last() Block
deleteLast()
link(label string) *Link
defineLink(label string, link *Link)
newText(pos Position, text string) *Text
}type AutoLink struct {
Text string
URL string
}type Code struct {
Text string
}type CodeBlock struct {
Position
Fence string
Info string
Text []string
}type Del struct {
Marker string
Inner []Inline
}type Document struct {
Position
Blocks []Block
Links map[string]*Link
}type Emoji struct {
Name string
Text string
}type Emph struct {
Marker string
Inner []Inline
}type Empty struct {
Position
}type Escaped struct {
Plain
}type HTMLBlock struct {
Position
Text []string
}type HTMLTag struct {
Text string
}type HardBreak struct {
}type Heading struct {
Position
Level int
Text *Text
ID string
}type Image struct {
Inner []Inline
URL string
Title string
TitleChar byte
corner bool
}type Item struct {
Position
Blocks []Block
width int
}type Link struct {
Inner []Inline
URL string
Title string
TitleChar byte
corner bool
}type List struct {
Position
Bullet rune
Start int
Loose bool
Items []Block
}type Paragraph struct {
Position
Text *Text
}A Parser is a Markdown parser. The exported fields in the struct can be filled in before calling [Parser.Parse] in order to customize the details of the parsing process. A Parser is safe for concurrent use by multiple goroutines.
type Parser struct {
HeadingIDs bool
Strikethrough bool
TaskListItems bool
AutoLinkText bool
AutoLinkAssumeHTTP bool
Table bool
Emoji bool
SmartDot bool
SmartDash bool
SmartQuote bool
}type Plain struct {
Text string
}type Position struct {
StartLine int
EndLine int
}type Quote struct {
Position
Blocks []Block
}type SoftBreak struct {
}type Strong struct {
Marker string
Inner []Inline
}type Table struct {
Position
Header []*Text
Align []string
Rows [][]*Text
}type Task struct {
Checked bool
}type Text struct {
Position
Inline []Inline
raw string
}type ThematicBreak struct {
Position
raw string
}type backtickParser struct {
last [maxBackticks]int
scanned bool
}type emphPlain struct {
Plain
canOpen bool
canClose bool
i int
n int
}type fenceBuilder struct {
fence string
info string
n int
text []string
}type htmlBuilder struct {
endBlank bool
text []string
endFunc func(string) bool
}type itemBuilder struct {
list *listBuilder
width int
haveContent bool
}type line struct {
spaces int
i int
tab int
text string
nl byte
}type listBuilder struct {
bullet rune
num int
loose bool
item *itemBuilder
todo func() line
}type mdState struct {
prefix string
prefix1 string
bullet rune
num int
}type openBlock struct {
builder blockBuilder
inner []Block
pos Position
}type openPlain struct {
Plain
i int
}type paraBuilder struct {
text []string
table *tableBuilder
}type parseState struct {
*Parser
root *Document
links map[string]*Link
lineno int
stack []openBlock
lineDepth int
corner bool
s string
emitted int
list []Inline
lists []*List
texts []*Text
backticks backtickParser
}For indented code blocks.
type preBuilder struct {
indent string
text []string
}type quoteBuilder struct {
}type rootBuilder struct {
}type tableBuilder struct {
hdr tableTrimmed
delim tableTrimmed
rows []tableTrimmed
}type validDomainChecker struct {
s string
cut int
}func (x *Strong) Inline()func (*Emoji) Inline()func (x *Task) Inline()func (*Plain) Inline()func (x *Del) Inline()func (*HardBreak) Inline()func (*Emph) Inline()func (*AutoLink) Inline()func (*Code) Inline()func (*SoftBreak) Inline()func (*Image) Inline()func (*Link) Inline()func (*HTMLTag) Inline()func (p *Parser) Parse(text string) *Documentfunc (p Position) Pos() Positionfunc (x *Del) PrintHTML(buf *bytes.Buffer)func (x *Strong) PrintHTML(buf *bytes.Buffer)func (b *Heading) PrintHTML(buf *bytes.Buffer)func (x *AutoLink) PrintHTML(buf *bytes.Buffer)func (b *CodeBlock) PrintHTML(buf *bytes.Buffer)func (b *Quote) PrintHTML(buf *bytes.Buffer)func (x *Link) PrintHTML(buf *bytes.Buffer)func (x *SoftBreak) PrintHTML(buf *bytes.Buffer)func (x *Emph) PrintHTML(buf *bytes.Buffer)func (t *Table) PrintHTML(buf *bytes.Buffer)func (b *Text) PrintHTML(buf *bytes.Buffer)func (b *Empty) PrintHTML(buf *bytes.Buffer)func (b *List) PrintHTML(buf *bytes.Buffer)func (x *HTMLTag) PrintHTML(buf *bytes.Buffer)func (x *Image) PrintHTML(buf *bytes.Buffer)func (x *HardBreak) PrintHTML(buf *bytes.Buffer)func (b *Document) PrintHTML(buf *bytes.Buffer)func (b *ThematicBreak) PrintHTML(buf *bytes.Buffer)func (x *Plain) PrintHTML(buf *bytes.Buffer)func (b *Item) PrintHTML(buf *bytes.Buffer)func (b *Paragraph) PrintHTML(buf *bytes.Buffer)func (b *HTMLBlock) PrintHTML(buf *bytes.Buffer)func (x *Emoji) PrintHTML(buf *bytes.Buffer)func (x *Code) PrintHTML(buf *bytes.Buffer)func (x *Task) PrintHTML(buf *bytes.Buffer)func (x *AutoLink) PrintText(buf *bytes.Buffer)func (x *Link) PrintText(buf *bytes.Buffer)func (x *Task) PrintText(buf *bytes.Buffer)func (x *Emoji) PrintText(buf *bytes.Buffer)func (x *Plain) PrintText(buf *bytes.Buffer)func (x *Strong) PrintText(buf *bytes.Buffer)func (x *HTMLTag) PrintText(buf *bytes.Buffer)func (x *SoftBreak) PrintText(buf *bytes.Buffer)func (x *HardBreak) PrintText(buf *bytes.Buffer)func (x *Del) PrintText(buf *bytes.Buffer)func (x *Code) PrintText(buf *bytes.Buffer)func (x *Image) PrintText(buf *bytes.Buffer)func (x *Emph) PrintText(buf *bytes.Buffer)func ToHTML(b Block) stringfunc ToMarkdown(b Block) stringfunc (p *parseState) addBlock(c blockBuilder)func (p *parseState) addLine(s line)func (b *tableBuilder) addRow(row string)func (p *parseState) autoLinkPlain(s string) []InlineautoLinkMore rewrites any extended autolinks in the body and returns the result. body is a list of Plain, Emph, Strong, and Del nodes. Two Plains only appear consecutively when one is a potential emphasis marker that ended up being plain after all, like "_" or "**". There are no Link nodes. The GitHub “spec” declares that “autolinks can only come at the beginning of a line, after whitespace, or any of the delimiting characters *, _, ~, and (”. However, the GitHub web site does not enforce this rule: text like "$abc@def.ghi is my email" links the text following the $ as an email address. It appears the actual rule is that autolinks cannot come after ASCII letters, although they can come after numbers or Unicode letters. Since the only point of implementing GitHub Flavored Markdown is to match GitHub's behavior, we do what they do, not what they say, at least for now.
func (p *parseState) autoLinkText(list []Inline) []Inlinefunc (p *parseState) blocks() []Blockfunc (b *rootBuilder) build(p buildState) Blockfunc (c *fenceBuilder) build(p buildState) Blockfunc (b *preBuilder) build(p buildState) Blockfunc (b *listBuilder) build(p buildState) Blockfunc (b *paraBuilder) build(p buildState) Blockfunc (b *tableBuilder) build(p buildState) Blockfunc (b *quoteBuilder) build(p buildState) Blockfunc (b *itemBuilder) build(p buildState) Blockfunc (c *htmlBuilder) build(p buildState) Blockfunc (p *parseState) closeBlock() Blockfunc (p *parseState) curB() blockBuilderfunc (p *parseState) defineLink(label string, link *Link)func (p *parseState) deleteLast()func (p *parseState) doneBlock(b Block)func dump(b Block) stringfunc (p *parseState) emit(i int)func (ps *parseState) emph(dst []Inline, src []Inline) []Inlinefunc (s *line) eof() boolfunc (c *preBuilder) extend(p *parseState, s line) (line, bool)func (c *rootBuilder) extend(p *parseState, s line) (line, bool)func (c *fenceBuilder) extend(p *parseState, s line) (line, bool)func (b *quoteBuilder) extend(p *parseState, s line) (line, bool)func (c *htmlBuilder) extend(p *parseState, s line) (line, bool)func (c *listBuilder) extend(p *parseState, s line) (line, bool)func (c *itemBuilder) extend(p *parseState, s line) (line, bool)func (b *paraBuilder) extend(p *parseState, s line) (line, bool)extractID removes an ID attribute from s if one is present. It returns the attribute value and the resulting string. The attribute has the form "{#...}", where the "..." can contain any character other than '}'. The attribute must be followed only by whitespace.
func extractID(p *parseState, s string) (id string, s2 string)func hasEndPre(s string) boolfunc (p *parseState) inline(s string) []Inlinefunc ins(first Inline, x []Inline) []Inlinefunc (s *line) isBlank() boolfunc isDigit(c byte) boolfunc isHR(s line) boolfunc isHexDigit(c byte) boolfunc isLDH(c byte) boolfunc isLetter(c byte) boolfunc isLetterDigit(c byte) boolfunc isPunct(c byte) boolfunc isScheme(c byte) boolfunc isTableSpace(c byte) boolfunc isTableStart(hdr1 string, delim1 string) boolfunc isURL(c byte) boolfunc isUnicodePunct(r rune) boolfunc isUnicodeSpace(r rune) boolfunc isUser(c byte) boolfunc (p *parseState) last() Blockfunc (p *parseState) link(label string) *Linkfunc linkCorner(url string) boolfunc listCorner(list *List) boollongestSequence returns the length of the longest sequence of consecutive bytes b in s.
func longestSequence(s string, b byte) intfunc mdUnescape(s string) stringfunc (p *parseState) mergePlain(list []Inline) []Inlinefunc mergePlain1(list []Inline) *Plainfunc newATXHeading(p *parseState, s line) (line, bool)func newFence(p *parseState, s line) (line, bool)func newHR(p *parseState, s line) (line, bool)func newHTML(p *parseState, s line) (line, bool)func newListItem(p *parseState, s line) (line, bool)func newPara(p *parseState, s line) (line, bool)func newPre(p *parseState, s line) (line, bool)func newQuote(p *parseState, s line) (line, bool)func newSetextHeading(p *parseState, s line) (line, bool)func (p *parseState) newText(pos Position, text string) *Textfunc (p *parseState) nextB() blockBuilderfunc nop(text string) stringfunc normalizeLabel(s string) stringfunc (p *parseState) para() *paraBuilderfunc (p *Parser) parse(text string) (d *Document, corner bool)func (b *tableBuilder) parseAlign(delim tableTrimmed, n int) []stringfunc parseAttr(p *parseState, s string, i int) (string, int, bool)func parseAttrName(s string, i int) (string, int, bool)func parseAttrValue(s string, i int) (string, int, bool)func parseAttrValueSpec(p *parseState, s string, i int) (string, int, bool)func (p *parseState) parseAutoEmail(s string, i int) (before string, link *Link, after string, ok bool)parseAutoWWW parses an extended www autolink. https://github.github.com/gfm/#extended-www-autolink
func (p *parseState) parseAutoHTTP(scheme string, s string, textstart int, start int, min int, vd *validDomainChecker) (link *Link, after string, found bool)func parseAutoLinkEmail(s string, i int) (Inline, int, bool)func parseAutoLinkOrHTML(p *parseState, s string, i int) (Inline, int, int, bool)func parseAutoLinkURI(s string, i int) (Inline, int, bool)func (p *parseState) parseAutoMailto(s string, i int) (link *Link, after string, ok bool)func (p *parseState) parseAutoProto(s string, i int, vd *validDomainChecker) (link *Link, after string, found bool)func (p *parseState) parseAutoXmpp(s string, i int) (link *Link, after string, ok bool)func parseBreak(_ *parseState, s string, i int) (Inline, int, int, bool)func (b *backtickParser) parseCodeSpan(p *parseState, s string, i int) (Inline, int, int, bool)func parseDash(p *parseState, s string, i int) (Inline, int, int, bool)func parseDot(p *parseState, s string, i int) (Inline, int, int, bool)func parseEmoji(p *parseState, s string, i int) (Inline, int, int, bool)func parseEmph(p *parseState, s string, i int) (Inline, int, int, bool)func parseEscape(p *parseState, s string, i int) (Inline, int, int, bool)func parseHTMLCDATA(s string, i int) (Inline, int, bool)func parseHTMLClosingTag(p *parseState, s string, i int) (Inline, int, bool)func parseHTMLComment(s string, i int) (Inline, int, bool)func parseHTMLDecl(p *parseState, s string, i int) (Inline, int, bool)func parseHTMLEntity(_ *parseState, s string, i int) (Inline, int, int, bool)func parseHTMLMarker(s string, i int, prefix string, suffix string) (Inline, int, bool)func parseHTMLOpenTag(p *parseState, s string, i int) (Inline, int, bool)func parseHTMLProcInst(s string, i int) (Inline, int, bool)func parseHTMLTag(p *parseState, s string, i int) (Inline, int, bool)func parseImageOpen(_ *parseState, s string, i int) (Inline, int, int, bool)func (p *parseState) parseLinkClose(s string, i int, open *openPlain) (*Link, int, bool)func parseLinkDest(s string, i int) (string, int, bool)func parseLinkLabel(p *parseState, s string, i int) (string, int, bool)func parseLinkOpen(_ *parseState, s string, i int) (Inline, int, int, bool)func parseLinkRefDef(p buildState, s string) (int, bool)func parseLinkTitle(s string, i int) (title string, char byte, next int, found bool)func (b *tableBuilder) parseRow(p buildState, row tableTrimmed, line int, width int) []*Textfunc parseTagName(s string, i int) (string, int, bool)parseValidDomain parses a valid domain. https://github.github.com/gfm/#valid-domain If s starts with a valid domain, parseValidDomain returns the length of that domain and true. If s does not start with a valid domain, parseValidDomain returns n, false, where n is the length of a prefix guaranteed not to be acceptable to any future call to parseValidDomain. “A valid domain consists of segments of alphanumeric characters, underscores (_) and hyphens (-) separated by periods (.). There must be at least one period, and no underscores may be present in the last two segments of the domain.” The spec does not spell out whether segments can be empty. Empirically, in GitHub's implementation they can.
func (v *validDomainChecker) parseValidDomain(start int) (n int, found bool)func (s *line) peek() bytefunc (p *parseState) pos() Positionfunc printLinkTitleMarkdown(buf *bytes.Buffer, title string, titleChar byte)func (b *Quote) printMarkdown(buf *bytes.Buffer, s mdState)func (x *Image) printMarkdown(buf *bytes.Buffer)func (b *CodeBlock) printMarkdown(buf *bytes.Buffer, s mdState)func (t *Table) printMarkdown(buf *bytes.Buffer, s mdState)func (x *AutoLink) printMarkdown(buf *bytes.Buffer)func (b *Heading) printMarkdown(buf *bytes.Buffer, s mdState)func (x *Link) printMarkdown(buf *bytes.Buffer)func (x *Task) printMarkdown(buf *bytes.Buffer)func (b *HTMLBlock) printMarkdown(buf *bytes.Buffer, s mdState)func (x *Emph) printMarkdown(buf *bytes.Buffer)func (x *Del) printMarkdown(buf *bytes.Buffer)func (b *Empty) printMarkdown(*bytes.Buffer, mdState)func (x *Strong) printMarkdown(buf *bytes.Buffer)func (b *Paragraph) printMarkdown(buf *bytes.Buffer, s mdState)func (x *Code) printMarkdown(buf *bytes.Buffer)func (x *Escaped) printMarkdown(buf *bytes.Buffer)func (x *Plain) printMarkdown(buf *bytes.Buffer)func (x *HTMLTag) printMarkdown(buf *bytes.Buffer)func (b *List) printMarkdown(buf *bytes.Buffer, s mdState)func (b *Document) printMarkdown(buf *bytes.Buffer, s mdState)func (x *SoftBreak) printMarkdown(buf *bytes.Buffer)func (b *Text) printMarkdown(buf *bytes.Buffer, s mdState)func (x *HardBreak) printMarkdown(buf *bytes.Buffer)func (b *Item) printMarkdown(buf *bytes.Buffer, s mdState)func (x *Emoji) printMarkdown(buf *bytes.Buffer)func (b *ThematicBreak) printMarkdown(buf *bytes.Buffer, s mdState)func printMarkdownBlocks(bs []Block, buf *bytes.Buffer, s mdState)func (x *Link) printRemainingMarkdown(buf *bytes.Buffer)func printb(buf *bytes.Buffer, b Block, prefix string)func printi(buf *bytes.Buffer, in Inline)func printis(buf *bytes.Buffer, ins []Inline)func printslice(buf *bytes.Buffer, v reflect.Value, prefix string)func (b *backtickParser) reset()func (v *validDomainChecker) skip(i int)func (p *parseState) skip(i int)func skipDomainElem(s string) (int, bool)func skipSpace(s string, i int) intfunc (s *line) skipSpace()func (b *tableBuilder) start(hdr string, delim string)func (p *parseState) startHTML(s *line) boolfunc (p *parseState) startListItem(s *line) boolfunc (s *line) string() stringfunc tableAlign(cell string) stringfunc tableCount(row tableTrimmed) intfunc tableTrimOuter(row string) tableTrimmedfunc tableTrimSpace(s string) stringfunc tableUnescape(text string) stringfunc (p *parseState) taskList(list *List)func toPlain(x Inline) *Plainfunc (s *line) trim(c byte) boolfunc (s *line) trimFence(fence *string, info *string, n *int) boolfunc (s *line) trimHeading(width *int) boolfunc trimLeftSpaceTab(s string) stringfunc trimQuote(s line) (line, bool)func trimRightSpaceTab(s string) stringfunc (s *line) trimSetext(n *int) boolfunc (s *line) trimSpace(min int, max int, eolOK bool) boolfunc trimSpace(s string) stringfunc (s *line) trimSpaceString() stringfunc trimSpaceTab(s string) stringfunc trimSpaceTabNewline(s string) stringfunc (p *parseState) trimStack(depth int)func (s *line) trimString() stringGenerated with Arrow