Error loading DICOM with Japanese Patient Name (ISO 2022 IR 87): JSON parse error

[mitsuki60]

Summary

VolView fails with a JSON parse error when loading a DICOM file that contains Japanese characters in Patient Name (PN).

Error:
SyntaxError: Unterminated string in JSON at position XXXX

Environment

• VolView version: HEAD (latest)
• Deployment: Web version
• OS: Windows
• Browser: Chrome (latest)
• Reproducibility: Always

Steps to Reproduce

1. Prepare a DICOM file with:
   • Specific Character Set: ISO 2022 IR 87
   • Patient Name (0010,0010) containing Japanese characters
2. Open VolView
3. Load the DICOM file
4. Observe the error in the browser console

Expected Behavior

The DICOM file should load successfully, and Japanese characters in Patient Name should be handled correctly.

Actual Behavior

VolView throws the following error and fails to load the data:
SyntaxError: Unterminated string in JSON at position XXXX

Additional Notes

• If the same DICOM file is modified so that Patient Name contains only ASCII characters, the error does NOT occur.
• Removing Japanese characters from Patient Name avoids the issue.
• If the Specific Character Set tag (0008,0005) is removed and Patient Name contains only English characters, the error does NOT occur.
• The issue appears to depend solely on the DICOM content and is reproducible without any additional configuration.

[PaulHax]

Thanks for the clear report. I wonder if you have tried to load that DICOM file in current main VolView or on https://volview-dev.kitware.app/ which tracks main?

We had the same error with a date tag:
#799

[mitsuki60]

We tested this using the current main branch (built from the latest commit as of Jan 12) and confirmed that the patch from PR #799 is applied in our build.

We verified that getPipelineModuleOutputJSON() in
node_modules/itk-wasm/dist/pipeline/internal/run-pipeline-emscripten.js
includes the control-character sanitization (dataJSON = dataJSON.replace(/[\x00-\x08\x0B\x0C\x0E-\x1F\x7F]/g, ...)), so the mitigation introduced in #799 is in effect.

At the same time, we still observe the issue when Japanese characters are present in the Patient Name (PN) on main. The error reported is:

SyntaxError: Unterminated string in JSON at position 2148 (line 137 column 18)

When the PN contains only ASCII characters, the file loads as expected.