231 lines
7.7 KiB
JavaScript
231 lines
7.7 KiB
JavaScript
#!/usr/bin/env node
|
||
|
||
const fs = require('fs');
|
||
const path = require('path');
|
||
|
||
const PROCESSED_DIR = path.join(__dirname, '..', 'data', 'processed');
|
||
const ASSET_MAP_PATH = path.join(PROCESSED_DIR, 'asset-map.json');
|
||
|
||
// Load asset map
|
||
const assetMap = JSON.parse(fs.readFileSync(ASSET_MAP_PATH, 'utf8'));
|
||
|
||
// Create ID to path mapping
|
||
const idToPath = {};
|
||
for (const [wpUrl, localPath] of Object.entries(assetMap)) {
|
||
const patterns = [/\/(\d+)-/, /\/(\d+)\./, /id=(\d+)/];
|
||
for (const pattern of patterns) {
|
||
const match = wpUrl.match(pattern);
|
||
if (match) {
|
||
idToPath[match[1]] = localPath;
|
||
break;
|
||
}
|
||
}
|
||
}
|
||
|
||
// Add manual mappings
|
||
idToPath['45569'] = '/media/45569-Still-2025-02-10-104337_1.1.1.webp';
|
||
idToPath['10648'] = '/media/10648-low-voltage-scaled.webp';
|
||
idToPath['6486'] = '/media/6486-Low-Voltage.svg';
|
||
idToPath['10649'] = '/media/10649-medium-voltage-scaled.webp';
|
||
idToPath['6487'] = '/media/6487-Medium-Voltage.svg';
|
||
idToPath['46786'] = '/media/46786-na2xsfl2y-rendered.webp';
|
||
idToPath['6485'] = '/media/6485-High-Voltage.svg';
|
||
idToPath['46359'] = '/media/46359-3.webp';
|
||
idToPath['6484'] = '/media/6484-Solar.svg';
|
||
idToPath['6527'] = '/media/6527-high-voltage-category.webp';
|
||
idToPath['6519'] = '/media/6519-solar-category.webp';
|
||
idToPath['6521'] = '/media/6521-low-voltage-category.webp';
|
||
idToPath['6517'] = '/media/6517-medium-voltage-category.webp';
|
||
|
||
console.log('Found', Object.keys(idToPath).length, 'media ID mappings');
|
||
|
||
// HTML entity decoding - handles decimal, hex, and named entities
|
||
function decodeHTMLEntities(text) {
|
||
if (!text) return '';
|
||
|
||
let result = text;
|
||
|
||
// First, handle numeric entities (decimal and hex)
|
||
result = result
|
||
.replace(/&#(\d+);/g, (match, dec) => {
|
||
const char = String.fromCharCode(parseInt(dec, 10));
|
||
return char;
|
||
})
|
||
.replace(/&#x([0-9a-fA-F]+);/g, (match, hex) => {
|
||
const char = String.fromCharCode(parseInt(hex, 16));
|
||
return char;
|
||
});
|
||
|
||
// Handle common named entities and Unicode characters
|
||
const entityMap = {
|
||
' ': ' ',
|
||
'‘': "'",
|
||
'’': "'",
|
||
'“': '"',
|
||
'”': '"',
|
||
'″': '"', // Double prime (8243)
|
||
'–': '-',
|
||
'—': '—',
|
||
'…': '…',
|
||
'•': '•',
|
||
'€': '€',
|
||
'©': '©',
|
||
'®': '®',
|
||
'™': '™',
|
||
'°': '°',
|
||
'±': '±',
|
||
'×': '×',
|
||
'÷': '÷',
|
||
'−': '−',
|
||
'¢': '¢',
|
||
'£': '£',
|
||
'¥': '¥',
|
||
'§': '§',
|
||
'¶': '¶',
|
||
'µ': 'µ',
|
||
'«': '«',
|
||
'»': '»',
|
||
'·': '·'
|
||
};
|
||
|
||
// Replace all named entities
|
||
for (const [entity, char] of Object.entries(entityMap)) {
|
||
result = result.replace(new RegExp(entity, 'g'), char);
|
||
}
|
||
|
||
// Clean up any remaining ampersand patterns
|
||
result = result.replace(/&([a-zA-Z]+);/g, (match, name) => {
|
||
return entityMap[`&${name};`] || match;
|
||
});
|
||
|
||
return result;
|
||
}
|
||
|
||
// Process files
|
||
const files = ['pages.json', 'posts.json', 'products.json'];
|
||
|
||
files.forEach(file => {
|
||
const filePath = path.join(PROCESSED_DIR, file);
|
||
if (!fs.existsSync(filePath)) return;
|
||
|
||
const items = JSON.parse(fs.readFileSync(filePath, 'utf8'));
|
||
let updated = false;
|
||
let updateCount = 0;
|
||
let decodeCount = 0;
|
||
|
||
items.forEach(item => {
|
||
let contentChanged = false;
|
||
let wasDecoded = false;
|
||
|
||
if (item.contentHtml) {
|
||
// Decode entities first
|
||
const original = item.contentHtml;
|
||
item.contentHtml = decodeHTMLEntities(item.contentHtml);
|
||
if (item.contentHtml !== original) {
|
||
wasDecoded = true;
|
||
decodeCount++;
|
||
}
|
||
|
||
// Now replace IDs with local paths
|
||
for (const [id, localPath] of Object.entries(idToPath)) {
|
||
// Pattern 1: bg_image="45569" (standard quotes)
|
||
const patterns = [
|
||
{ search: 'bg_image="' + id + '"', replace: 'bg_image="' + localPath + '"' },
|
||
{ search: 'background_image="' + id + '"', replace: 'background_image="' + localPath + '"' },
|
||
{ search: 'image_url="' + id + '"', replace: 'image_url="' + localPath + '"' },
|
||
{ search: 'custom_icon_image="' + id + '"', replace: 'custom_icon_image="' + localPath + '"' },
|
||
{ search: 'poster="' + id + '"', replace: 'poster="' + localPath + '"' },
|
||
{ search: 'column_background_image="' + id + '"', replace: 'column_background_image="' + localPath + '"' },
|
||
];
|
||
|
||
patterns.forEach(({ search, replace }) => {
|
||
if (item.contentHtml.includes(search)) {
|
||
item.contentHtml = item.contentHtml.split(search).join(replace);
|
||
contentChanged = true;
|
||
}
|
||
});
|
||
|
||
// Also check for HTML-encoded attribute values (after decodeHTMLEntities, these become regular quotes)
|
||
// But we need to handle the case where the HTML entities haven't been decoded yet
|
||
const encodedPatterns = [
|
||
{ search: 'bg_image=”' + id + '″', replace: 'bg_image="' + localPath + '"' },
|
||
{ search: 'bg_image=”' + id + '”', replace: 'bg_image="' + localPath + '"' },
|
||
{ search: 'bg_image="' + id + '"', replace: 'bg_image="' + localPath + '"' },
|
||
];
|
||
|
||
encodedPatterns.forEach(({ search, replace }) => {
|
||
if (item.contentHtml.includes(search)) {
|
||
item.contentHtml = item.contentHtml.split(search).join(replace);
|
||
contentChanged = true;
|
||
}
|
||
});
|
||
}
|
||
}
|
||
|
||
if (item.excerptHtml) {
|
||
const original = item.excerptHtml;
|
||
item.excerptHtml = decodeHTMLEntities(item.excerptHtml);
|
||
|
||
for (const [id, localPath] of Object.entries(idToPath)) {
|
||
// Standard pattern
|
||
const search = 'bg_image="' + id + '"';
|
||
const replace = 'bg_image="' + localPath + '"';
|
||
if (item.excerptHtml.includes(search)) {
|
||
item.excerptHtml = item.excerptHtml.split(search).join(replace);
|
||
contentChanged = true;
|
||
}
|
||
|
||
// Also check for HTML-encoded patterns that might remain (after decode)
|
||
// Handle various quote combinations
|
||
const encodedPatterns = [
|
||
'bg_image="' + id + '"', // Already decoded
|
||
'bg_image="' + id + '″', // Opening regular, closing double prime
|
||
'bg_image="' + id + '"', // Both regular
|
||
];
|
||
|
||
encodedPatterns.forEach(search => {
|
||
if (item.excerptHtml.includes(search)) {
|
||
item.excerptHtml = item.excerptHtml.split(search).join(replace);
|
||
contentChanged = true;
|
||
}
|
||
});
|
||
}
|
||
|
||
if (item.excerptHtml !== original && !contentChanged) contentChanged = true;
|
||
}
|
||
|
||
if (contentChanged || wasDecoded) {
|
||
updated = true;
|
||
if (contentChanged) updateCount++;
|
||
}
|
||
});
|
||
|
||
if (updated) {
|
||
fs.writeFileSync(filePath, JSON.stringify(items, null, 2));
|
||
console.log('✅ Updated ' + file + ' (' + updateCount + ' replacements, ' + decodeCount + ' decoded)');
|
||
} else {
|
||
console.log('ℹ️ No changes for ' + file);
|
||
}
|
||
});
|
||
|
||
// Verify
|
||
const pages = JSON.parse(fs.readFileSync(path.join(PROCESSED_DIR, 'pages.json'), 'utf8'));
|
||
const homeEn = pages.find(p => p.slug === 'corporate-3-landing-2' && p.locale === 'en');
|
||
const homeDe = pages.find(p => p.slug === 'start' && p.locale === 'de');
|
||
|
||
console.log('\n✅ Verification:');
|
||
console.log('EN home images:', (homeEn?.contentHtml?.match(/\/media\//g) || []).length);
|
||
console.log('DE home images:', (homeDe?.contentHtml?.match(/\/media\//g) || []).length);
|
||
|
||
// Check for remaining IDs
|
||
const remainingIds = homeEn?.contentHtml?.match(/bg_image="\d+"/g) || [];
|
||
console.log('Remaining IDs in EN:', remainingIds.length > 0 ? remainingIds : 'None');
|
||
|
||
// Show examples
|
||
if (homeEn?.contentHtml) {
|
||
const matches = homeEn.contentHtml.match(/bg_image="[^"]+"/g);
|
||
if (matches) {
|
||
console.log('\nEN bg_image examples:', matches.slice(0, 3));
|
||
}
|
||
}
|