geo/apps/desktop-client/src/main/adapters/dongchedi-content.ts

import { renderTablesAsParagraphRows } from '../../../../../packages/publisher-platforms/src/baijiahao'

export function prepareDongchediArticleHtml(html: string, title = ''): string {
  let next = renderTablesAsParagraphRows(html).trim()
  next = next.replace(/<hr\b[^>]*\/?>/gi, '')
  next = normalizeDongchediImageTags(next)
  next = next.replace(/<h[1-6]\b[^>]*>([\s\S]*?)<\/h[1-6]>/gi, (_match, content: string) => {
    const text = normalizeInlineContent(content)
    return text ? `<p><strong>${text}</strong></p>` : ''
  })
  next = next.replace(/<li\b[^>]*>([\s\S]*?)<\/li>/gi, (_match, content: string) => {
    const text = normalizeInlineContent(content)
    return text ? `<p>${text}</p>` : ''
  })
  next = next.replace(/<\/?(?:ol|ul)\b[^>]*>/gi, '')
  next = next.replace(/<p\b[^>]*>/gi, '<p>')
  next = next.replace(/<(strong|b)\b[^>]*>/gi, '<strong>').replace(/<\/b>/gi, '</strong>')
  next = next.replace(/<br\s*\/?>/gi, ' ')
  next = next.replace(/<(?!\/?(?:p|strong|img)\b)[^>]+>/gi, '')
  next = next.replace(/<p>\s*<\/p>/gi, '')
  next = next.replace(/\s+/g, ' ')
  next = next.replace(/>\s+</g, '><')
  next = stripLeadingDuplicateTitleParagraph(next, title)
  return next.trim()
}

export function countDongchediContentWordCount(html: string): number {
  const text = decodeDongchediHtmlEntities(html)
    .replace(/<[^>]*>/g, '')
    .replace(/\s+/g, '')
    .trim()
  return [...text].length
}

function decodeDongchediHtmlEntities(value: string): string {
  return value
    .replace(/&nbsp;/gi, ' ')
    .replace(/&amp;/gi, '&')
    .replace(/&lt;/gi, '<')
    .replace(/&gt;/gi, '>')
    .replace(/&quot;/gi, '"')
    .replace(/&#39;|&apos;/gi, "'")
    .replace(/&#(\d+);/g, (_match, code: string) => {
      const value = Number.parseInt(code, 10)
      return Number.isFinite(value) && value >= 0 && value <= 0x10ffff
        ? String.fromCodePoint(value)
        : ''
    })
    .replace(/&#x([0-9a-f]+);/gi, (_match, code: string) => {
      const value = Number.parseInt(code, 16)
      return Number.isFinite(value) && value >= 0 && value <= 0x10ffff
        ? String.fromCodePoint(value)
        : ''
    })
}

function normalizeDongchediImageTags(html: string): string {
  return html.replace(/<img\b[^>]*>/gi, (imageTag: string) => {
    const src = extractAttribute(imageTag, 'src')
    if (!src) {
      return ''
    }
    const alt = extractAttribute(imageTag, 'alt') ?? ''
    return `<img src="${escapeAttribute(src)}" alt="${escapeAttribute(alt)}" />`
  })
}

function normalizeInlineContent(value: string): string {
  return value
    .replace(/<br\s*\/?>/gi, ' ')
    .replace(
      /<\/?(?:p|div|section|article|header|footer|ol|ul|li|blockquote|h[1-6])\b[^>]*>/gi,
      ' ',
    )
    .replace(/\s+/g, ' ')
    .trim()
}

function stripLeadingDuplicateTitleParagraph(html: string, title: string): string {
  const normalizedTitle = normalizeComparableText(title)
  if (!normalizedTitle) {
    return html
  }

  return html.replace(/^\s*<p>([\s\S]*?)<\/p>/i, (match, paragraphContent: string) => {
    return normalizeComparableText(paragraphContent) === normalizedTitle ? '' : match
  })
}

function normalizeComparableText(value: string): string {
  return decodeDongchediHtmlEntities(value)
    .replace(/<[^>]*>/g, '')
    .replace(/\s+/g, '')
    .trim()
}

function extractAttribute(sourceTag: string, attributeName: string): string | null {
  const matcher = new RegExp(
    `\\s${attributeName}\\s*=\\s*(?:"([^"]*)"|'([^']*)'|([^\\s"'>]+))`,
    'i',
  )
  const match = matcher.exec(sourceTag)
  const value = match?.[1] ?? match?.[2] ?? match?.[3] ?? ''
  const trimmed = value.trim()
  return trimmed ? trimmed : null
}

function escapeAttribute(value: string): string {
  return value
    .replace(/&/g, '&amp;')
    .replace(/"/g, '&quot;')
    .replace(/</g, '&lt;')
    .replace(/>/g, '&gt;')
}