workflowy
by mholzen
markdown.go•10 kB
package formatter
import (
"strings"
"github.com/mholzen/workflowy/pkg/workflowy"
)
type MarkdownFormatter struct {
config *MarkdownConfig
}
type MarkdownConfig struct {
ExcludeTag string
H1Tag string
H2Tag string
H3Tag string
PTag string
ListTag string
}
func DefaultMarkdownConfig() *MarkdownConfig {
return &MarkdownConfig{
ExcludeTag: "#exclude",
H1Tag: "#h1",
H2Tag: "#h2",
H3Tag: "#h3",
PTag: "#p",
ListTag: "#list",
}
}
func NewMarkdownFormatter() *MarkdownFormatter {
return &MarkdownFormatter{
config: DefaultMarkdownConfig(),
}
}
func NewMarkdownFormatterWithConfig(config *MarkdownConfig) *MarkdownFormatter {
return &MarkdownFormatter{
config: config,
}
}
func (f *MarkdownFormatter) FormatTree(items []*workflowy.Item) (string, error) {
var result strings.Builder
for _, item := range items {
output := f.formatNode(item, 1)
if output != "" {
result.WriteString(output)
}
}
return strings.TrimRight(result.String(), "\n") + "\n", nil
}
func (f *MarkdownFormatter) formatNode(item *workflowy.Item, headerLevel int) string {
if f.shouldExclude(item) {
return ""
}
layoutMode := f.getLayoutMode(item)
name := f.stripAllTags(item.Name)
switch layoutMode {
case "h1":
return f.formatAsHeader(item, name, 1)
case "h2":
return f.formatAsHeader(item, name, 2)
case "h3":
return f.formatAsHeader(item, name, 3)
case "p":
return f.formatAsParagraphWithChildren(item, name)
case "ol":
return f.formatAsOrderedList(item, name)
case "quote":
return f.formatAsQuote(item, name)
case "code":
return f.formatAsCode(item, name)
case "divider":
return "---\n\n"
default:
return f.formatBulletsNode(item, name, headerLevel)
}
}
func (f *MarkdownFormatter) formatBulletsNode(item *workflowy.Item, name string, headerLevel int) string {
if len(item.Children) == 0 {
return ""
}
if IsListPattern(item) {
return f.formatWithListChildren(item, name, headerLevel)
}
if f.childrenAreAllSubheaders(item.Children) {
return f.formatAsHeaderWithSubheaders(item, name, headerLevel)
}
return f.formatAsHeaderWithParagraph(item, name, headerLevel)
}
func (f *MarkdownFormatter) childrenAreAllSubheaders(children []*workflowy.Item) bool {
hasAnyWithGrandchildren := false
for _, child := range children {
if f.shouldExclude(child) {
continue
}
if IsEmpty(child.Name) {
continue
}
if len(child.Children) > 0 {
if IsListPattern(child) {
return false
}
hasAnyWithGrandchildren = true
} else {
if hasAnyWithGrandchildren {
return false
}
}
}
return hasAnyWithGrandchildren
}
func (f *MarkdownFormatter) formatAsHeader(item *workflowy.Item, name string, level int) string {
var result strings.Builder
result.WriteString(HeaderPrefix(level))
result.WriteString(Capitalize(name))
result.WriteString("\n\n")
for _, child := range item.Children {
childOutput := f.formatNode(child, level+1)
result.WriteString(childOutput)
}
return result.String()
}
func (f *MarkdownFormatter) formatAsHeaderWithParagraph(item *workflowy.Item, name string, headerLevel int) string {
var result strings.Builder
result.WriteString(HeaderPrefix(headerLevel))
result.WriteString(Capitalize(name))
result.WriteString("\n")
paragraphs := f.collectParagraphs(item.Children)
for _, para := range paragraphs {
if para == "" {
result.WriteString("\n")
continue
}
result.WriteString(para)
result.WriteString("\n")
}
result.WriteString("\n")
return result.String()
}
func (f *MarkdownFormatter) collectParagraphs(children []*workflowy.Item) []string {
var paragraphs []string
var currentSentences []string
needsBlankBefore := false
for _, child := range children {
if f.shouldExclude(child) {
continue
}
childName := f.stripAllTags(child.Name)
if IsEmptyBullet(child) || IsEmpty(childName) {
if len(currentSentences) > 0 {
paragraphs = append(paragraphs, strings.Join(currentSentences, " "))
currentSentences = nil
}
needsBlankBefore = true
continue
}
if needsBlankBefore && len(paragraphs) > 0 {
paragraphs = append(paragraphs, "")
needsBlankBefore = false
}
if IsListPattern(child) {
currentSentences = append(currentSentences, FormatAsSentence(childName))
intro := strings.Join(currentSentences, " ")
currentSentences = nil
listOutput := f.formatInlineListWithIntro(child, intro)
paragraphs = append(paragraphs, listOutput)
needsBlankBefore = true
continue
}
currentSentences = append(currentSentences, FormatAsSentence(childName))
}
if len(currentSentences) > 0 {
if needsBlankBefore && len(paragraphs) > 0 {
paragraphs = append(paragraphs, "")
}
paragraphs = append(paragraphs, strings.Join(currentSentences, " "))
}
return paragraphs
}
func (f *MarkdownFormatter) formatInlineList(item *workflowy.Item, name string) string {
return f.formatInlineListWithIntro(item, FormatAsSentence(name))
}
func (f *MarkdownFormatter) formatInlineListWithIntro(item *workflowy.Item, intro string) string {
var result strings.Builder
result.WriteString(intro)
result.WriteString("\n")
for _, child := range item.Children {
if f.shouldExclude(child) {
continue
}
childName := f.stripAllTags(child.Name)
if !IsEmpty(childName) {
result.WriteString("- ")
result.WriteString(childName)
result.WriteString("\n")
}
}
return strings.TrimRight(result.String(), "\n")
}
func (f *MarkdownFormatter) formatAsHeaderWithSubheaders(item *workflowy.Item, name string, headerLevel int) string {
var result strings.Builder
result.WriteString(HeaderPrefix(headerLevel))
result.WriteString(Capitalize(name))
result.WriteString("\n\n")
for _, child := range item.Children {
if f.shouldExclude(child) {
continue
}
childName := f.stripAllTags(child.Name)
if len(child.Children) > 0 {
childOutput := f.formatBulletsNode(child, childName, headerLevel+1)
result.WriteString(childOutput)
} else if !IsEmpty(childName) {
result.WriteString(FormatAsSentence(childName))
result.WriteString("\n\n")
}
}
return result.String()
}
func (f *MarkdownFormatter) formatWithListChildren(item *workflowy.Item, name string, headerLevel int) string {
var result strings.Builder
result.WriteString(HeaderPrefix(headerLevel))
result.WriteString(Capitalize(name))
result.WriteString("\n")
for _, child := range item.Children {
if f.shouldExclude(child) {
continue
}
childName := f.stripAllTags(child.Name)
if !IsEmpty(childName) {
result.WriteString("- ")
result.WriteString(childName)
result.WriteString("\n")
}
}
result.WriteString("\n")
return result.String()
}
func (f *MarkdownFormatter) formatAsParagraph(text string) string {
return FormatAsSentence(text)
}
func (f *MarkdownFormatter) formatAsParagraphWithChildren(item *workflowy.Item, name string) string {
var result strings.Builder
result.WriteString(f.formatAsParagraph(name))
result.WriteString("\n\n")
if len(item.Children) > 0 {
for _, child := range item.Children {
if f.shouldExclude(child) {
continue
}
childName := f.stripAllTags(child.Name)
if !IsEmpty(childName) {
result.WriteString("- ")
result.WriteString(childName)
result.WriteString("\n")
}
}
result.WriteString("\n")
}
return result.String()
}
func (f *MarkdownFormatter) formatAsOrderedList(item *workflowy.Item, name string) string {
var result strings.Builder
if name != "" {
result.WriteString(name)
result.WriteString("\n")
}
for i, child := range item.Children {
if f.shouldExclude(child) {
continue
}
childName := f.stripAllTags(child.Name)
if !IsEmpty(childName) {
result.WriteString(strings.Repeat(" ", 0))
result.WriteString(string(rune('1'+i)) + ". ")
result.WriteString(childName)
result.WriteString("\n")
}
}
result.WriteString("\n")
return result.String()
}
func (f *MarkdownFormatter) formatAsQuote(item *workflowy.Item, name string) string {
var result strings.Builder
if name != "" {
result.WriteString("> ")
result.WriteString(name)
result.WriteString("\n")
}
for _, child := range item.Children {
if f.shouldExclude(child) {
continue
}
childName := f.stripAllTags(child.Name)
if !IsEmpty(childName) {
result.WriteString("> ")
result.WriteString(childName)
result.WriteString("\n")
}
}
result.WriteString("\n")
return result.String()
}
func (f *MarkdownFormatter) formatAsCode(item *workflowy.Item, name string) string {
var result strings.Builder
result.WriteString("```\n")
if name != "" {
result.WriteString(name)
result.WriteString("\n")
}
for _, child := range item.Children {
if f.shouldExclude(child) {
continue
}
childName := f.stripAllTags(child.Name)
result.WriteString(childName)
result.WriteString("\n")
}
result.WriteString("```\n\n")
return result.String()
}
func (f *MarkdownFormatter) shouldExclude(item *workflowy.Item) bool {
return HasTag(item.Name, f.config.ExcludeTag)
}
func (f *MarkdownFormatter) getLayoutMode(item *workflowy.Item) string {
if HasTag(item.Name, f.config.H1Tag) {
return "h1"
}
if HasTag(item.Name, f.config.H2Tag) {
return "h2"
}
if HasTag(item.Name, f.config.H3Tag) {
return "h3"
}
if HasTag(item.Name, f.config.PTag) {
return "p"
}
if HasTag(item.Name, f.config.ListTag) {
return "list"
}
if item.Data != nil {
if mode, ok := item.Data["layoutMode"].(string); ok && mode != "" {
return mode
}
}
return "bullets"
}
func (f *MarkdownFormatter) stripAllTags(text string) string {
text = StripTag(text, f.config.ExcludeTag)
text = StripTag(text, f.config.H1Tag)
text = StripTag(text, f.config.H2Tag)
text = StripTag(text, f.config.H3Tag)
text = StripTag(text, f.config.PTag)
text = StripTag(text, f.config.ListTag)
return strings.TrimSpace(text)
}
func FormatItemsAsMarkdown(items []*workflowy.Item) (string, error) {
formatter := NewMarkdownFormatter()
return formatter.FormatTree(items)
}
Latest Blog Posts
- MCP Moves to the Linux Foundation: Neutral Stewardship for Agentic InfrastructureBy Om-Shree-0709 on .mcpanthropicLinux Foundation
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/mholzen/workflowy'
If you have feedback or need assistance with the MCP directory API, please join our Discord server