Fix minor crash for certain oddly formed *.js files

[PierceLBrooks]

Description

• Source branch in your fork has meaningful name (not main)

Hello there! I recently discovered an interesting piece of apparent javascript code that I wanted to inspect further at https://www.scribd.com/document/512791356/Player - which has been attached here - and it was found that neither the https://beautifier.io/ web interface nor this python module could beautify the content without either hanging indefinitely or throwing an exception like this one below:

pbrooks@MacBook-Pro Downloads % python3 ./jsbeautify.py file://$PWD/Player.js ./Vimeo.js
Traceback (most recent call last):
  File "/Users/pbrooks/Downloads/./jsbeautify.py", line 40, in <module>
    main()
  File "/Users/pbrooks/Downloads/./jsbeautify.py", line 33, in main
    js = beauty(r["content"])
  File "/Users/pbrooks/Downloads/./jsbeautify.py", line 12, in beauty
    return jsbeautifier.beautify(content.decode())
  File "/opt/homebrew/lib/python3.10/site-packages/jsbeautifier/__init__.py", line 82, in beautify
    return b.beautify(string, opts)
  File "/opt/homebrew/lib/python3.10/site-packages/jsbeautifier/javascript/beautifier.py", line 187, in beautify
    self._tokens = Tokenizer(source_text, self._options).tokenize()
  File "/opt/homebrew/lib/python3.10/site-packages/jsbeautifier/core/tokenizer.py", line 71, in tokenize
    current = self.__get_next_token_with_comments(previous, open_token)
  File "/opt/homebrew/lib/python3.10/site-packages/jsbeautifier/core/tokenizer.py", line 93, in __get_next_token_with_comments
    current = self._get_next_token(previous, open_token)
  File "/opt/homebrew/lib/python3.10/site-packages/jsbeautifier/javascript/tokenizer.py", line 226, in _get_next_token
    token = token or self._read_regexp(c, previous_token)
  File "/opt/homebrew/lib/python3.10/site-packages/jsbeautifier/javascript/tokenizer.py", line 331, in _read_regexp
    if c == "/" and self.allowRegExOrXML(previous_token):
  File "/opt/homebrew/lib/python3.10/site-packages/jsbeautifier/javascript/tokenizer.py", line 514, in allowRegExOrXML
    and previous_token.opened.previous.type == TOKEN.RESERVED
AttributeError: 'NoneType' object has no attribute 'previous'

The beautification script I used can be located on this page... https://gist.github.com/PierceLBrooks/7a29d53f4915f847b82e41e2954ca000

Before Merge Checklist

These items can be completed after PR is created.

(Check any items that are not applicable (NA) for this PR)

• JavaScript implementation
• Python implementation (NA if HTML beautifier)
• Added Tests to data file(s)
• Added command-line option(s) (NA if
• README.md documents new feature/option(s)

Player.txt

[bitwiseman]

Could you add a test with the input that crashes without this fix?

[UlyssesWu]

AttributeError: 'NoneType' object has no attribute 'previous'

Same problem here. This fix still works.

[bitwiseman]

@UlyssesWu
What is an input that causes this?

[PierceLBrooks]

@bitwiseman:

@UlyssesWu What is an input that causes this?

https://github.com/beautify-web/js-beautify/files/10437339/Player.txt

[bitwiseman]

@PierceLBrooks
Please add the following the regression tests in test/data/javascript/tests.js

        {
          comment: 'Issue #2128 - NPE in python implementation',
          fragment: true,
          unchanged: [
            ') / a / g'
          ]
        },

This input fails before this PR and succeeds with this change in place.

[PierceLBrooks]

@bitwiseman I forgot all about your original unit test request! My bad.

PierceLBrooks@de2b0c6

^ Does this suffice? Your unit test framework is unfamiliar to me unfortunately.

[PierceLBrooks]

The failure encountered by the https://github.com/beautifier/js-beautify/actions/runs/13642772279 CI run appears to have been on account of the issue manifesting not within the fixed Python code, but by the at-that-time-yet-to-be-fixed Javascript code's version of the same bug. This should now be resolved by https://github.com/beautifier/js-beautify/actions/runs/13643064944