* addition becomes concat with a string operand
* improve handling of edgier cases
* arithmetic in float like JS (more places need cast to int?)
* increase test coverage
* extend single indexing with improved RE (should probably use/have used _separate_at_paren())
* fix some cases that should have given undefined, not throwing
* standardise RE group names
* support length of objects, like {1: 2, 3: 4, length: 42}
* reimplement loose equality according to MDN (eg, 1 == "1")
* improve strict equality (eg, "abc" === "abc" but 'abc' is not 'abc')
* add tests for above
* support subset `... else if ...`
* support `while`
* add `RegExp` class
* generalise `new` support
* limited more debug strings
* matching test changes
* add NaN
* allow any white-space character for `after_op`
* align with yt-dlp f26af78a8ac11d9d617ed31ea5282cfaa5bcbcfa (charcodeAt and bitwise overflow)
* allow escaping in regex, fixing player c57c113c
* back-port from yt-dlp 8f53dc44a0cc1c2d98c35740b9293462c080f5d0, thanks pukkandan
* also support void, improve <</>> precedence, improve expressions in comma-list
* add more tests
In pycodestyle 2.1.0, E305 was introduced, which requires two blank
lines after top level declarations, too.
See https://github.com/PyCQA/pycodestyle/issues/400
See also #10689; thanks @stepshal for first mentioning this issue and
initial patches
We need a proper lexer to be able to understand YouTube's code, which contains /* inside of strings.
For now it's sufficient to just disable comment support altogether.
Fixes#4976, fixes#4979, fixes#4980, fixes#4981, fixes#4982.
Closes#4977.
