`Uncaught SyntaxError` when importing certain PDFs via text loader

[inad9300]

What version of Bun is running?

1.1.21+70ca2b76c

What platform is your computer?

Darwin 21.6.0 x86_64 i386

What steps can reproduce the bug?

Bundling the following code:

import doc from './sample.pdf' with { type: 'text' }
console.log(doc.length)

And running it from a browser, it produces:

main.js:59 Uncaught SyntaxError: Invalid or unexpected token (at main.js:59:22)

Example of problematic PDFs:

• https://dl.acm.org/doi/pdf/10.1145/1283920.1283927
• https://arvindguptatoys.com/arvindgupta/whatislife-schrodinger.pdf

Example of working PDFs:

• https://wiki.creativecommons.org/images/3/35/Creativecommons-what-is-creative-commons_eng.pdf
• https://upload.wikimedia.org/wikipedia/meta/8/8e/Wikipedia-logo-big.pdf

What is the expected behavior?

No response

What do you see instead?

No response

Additional information

No response

[paperdave]

the bundled file contains a newline in a string. i looked at the first file and it seems that the \r that is 355069 characters in gets emitted as \n by bun build, but bundling the byte sequence nearby that does not cause the issue.