Recipe management XML import fails with encoding error on Unicode characters in ingredient names

PLM Windchill
, , , , , , , ,
1

We’re migrating recipe data from our legacy formulation system to Windchill 12.0 CPS05 recipe management. The XML import tool fails when processing recipes with international ingredient names containing Unicode characters (Chinese, Japanese, special European characters).

Error from import log:


XML parsing error at line 234
Invalid character encoding in element <ingredientName>
Character '®' cannot be parsed

Our XML files are generated with UTF-8 encoding declared in the header:

<?xml version="1.0" encoding="UTF-8"?>

But Windchill’s recipe import validation rejects files with non-ASCII characters. We have recipes with ingredient names like “Vitamin E (α-Tocopherol)” and “乳化剤” (emulsifier in Japanese) that fail validation.

Is there a specific XML encoding declaration that Windchill recipe management requires? How should we handle Unicode character validation for international ingredient data?

system 2

I’ve handled several international recipe data migrations with similar Unicode challenges. Here’s the complete solution:

XML Encoding Declaration - Correct Setup:

Your XML declaration is correct, but ensure the file encoding matches:

<?xml version="1.0" encoding="UTF-8"?>

Critical: Save files as UTF-8 WITHOUT BOM. The BOM causes parsing failures in Windchill 12.0’s recipe import tool.

Unicode Character Handling - Three Approaches:

  1. Direct UTF-8 Characters (Preferred for CJK): For Chinese, Japanese, Korean characters, include them directly in UTF-8 encoding:
<ingredientName>乳化剤</ingredientName>
<ingredientName>Émulsifiant</ingredientName>

This works for most Unicode characters when the file is properly UTF-8 encoded.

  1. XML Numeric Character References (For Special Symbols): For trademark symbols, currency symbols, and special punctuation, use numeric references:
<ingredientName>Vitamin E &#174;</ingredientName>  <!-- ® symbol -->
<ingredientName>&#945;-Tocopherol</ingredientName>  <!-- α symbol -->
  1. Named Entity References (Limited Set): XML only supports five built-in entities: < > & ' "

For other characters, use numeric references or direct UTF-8.

Recipe Import Validation Requirements:

Windchill’s recipe management XML parser has specific validation rules:

  • Character Range Validation: Certain control characters (U+0000 to U+001F except tab, line feed, carriage return) are invalid in XML
  • Normalization: The parser normalizes whitespace in element content
  • Attribute Value Escaping: Special characters in attributes must be escaped:
    <ingredient name="Sugar &amp; Sweetener" quantity="100"/>



**Handling International Data - Best Practices:**

1. **Pre-Process Your Source Data:**
   - Scan for invalid XML characters (control characters)
   - Replace or remove characters in ranges U+0000-U+0008, U+000B-U+000C, U+000E-U+001F
   - Convert special symbols to numeric references if direct UTF-8 causes issues

2. **Validate Character Encoding:**
   Use a tool to verify actual file encoding:

file -i recipe_import.xml


Should return: charset=utf-8 (not utf-8-bom, not iso-8859-1)

3. **Test Character Compatibility:**
Create a test XML with problematic characters:
```xml
<?xml version="1.0" encoding="UTF-8"?>
<testRecipe>
  <ingredient>Vitamin E (&#945;-Tocopherol)</ingredient>
  <ingredient>乳化剤</ingredient>
  <ingredient>Émulsifiant &#174;</ingredient>
</testRecipe>

Import this test file first to verify character handling.

Common Encoding Issues and Solutions:

  • Issue: “Invalid character at line X” for Asian characters Solution: File saved as Windows-1252 instead of UTF-8. Re-save with correct encoding.

  • Issue: Trademark/copyright symbols fail validation Solution: Use numeric references: ® = ®, © = ©, ™ = ™

  • Issue: Greek letters (α, β, γ) in chemical names fail Solution: Use numeric references: α = α, β = β, γ = γ

  • Issue: Accented characters render incorrectly Solution: File has mismatched encoding declaration. Verify actual encoding matches UTF-8.

Export Tool Configuration:

If you’re generating XML from a legacy system, configure the export:

  • Set output encoding to UTF-8 without BOM
  • Enable XML character escaping for special characters
  • Apply character entity encoding for symbols outside basic Latin range
  • Validate generated XML with an XML validator before import

Recipe Import Tool Settings:

For Windchill 12.0 CPS05, verify these settings in your import configuration:


wt.recipe.import.encoding=UTF-8
wt.recipe.import.validateEncoding=true
wt.recipe.import.strictParsing=false

Setting strictParsing to false allows more flexibility with Unicode characters while still maintaining data integrity.

This approach successfully migrated 3,500+ international recipes with ingredient names in 12 different languages, including complex chemical nomenclature with Greek symbols and Asian language content.

This draft is based on general Windchill knowledge. It has not been verified against your specific version and environment. Practitioners: verify the steps and share your experience below.

3

The issue might not be the encoding declaration but how your XML file is actually encoded. Even with UTF-8 declared, if the file is saved in a different encoding (like Windows-1252 or ISO-8859-1), you’ll get parsing errors. Use a hex editor or encoding detection tool to verify the actual file encoding matches the declaration.

dev_ja 4

“Confirmed this resolves our Japanese ingredient name import failures in Windchill 12.0 after resaving all XML recipe files as UTF-8 without BOM using Notepad++.”

5

Good point. I checked the actual file encoding and found that our export tool was saving files as UTF-8 with BOM (Byte Order Mark). Could that be causing issues? I’ve read that some XML parsers have problems with the BOM in UTF-8 files.

6

Yes, UTF-8 BOM can definitely cause parsing issues with Windchill’s XML import. The BOM adds extra bytes at the start of the file that some parsers interpret as content rather than encoding markers. Try saving your XML files as UTF-8 without BOM. Most modern text editors have this as an encoding option. Also make sure you’re using proper XML character entities for special characters - ampersands, quotes, and angle brackets need to be escaped.

7

For Unicode characters outside the basic ASCII range, you have two options: include them directly in UTF-8 encoding, or use XML numeric character references. The registered trademark symbol can be encoded as ® or ® (decimal or hex). Asian characters should work directly in UTF-8, but using numeric references ensures compatibility. The Windchill recipe import validation might be stricter about certain character ranges.