[PATCH 10/34] HP-PA: use is_end_of_stmt()

[Jan Beulich]

On 21.02.2025 06:04, Hans-Peter Nilsson wrote:
> On Thu, 20 Feb 2025, Jan Beulich wrote:
>> is
>> this yet another of the many assumptions towards the scrubber eating
>> white space, breaking when scrubbing is skipped (and potentially
>> breaking when scrubber behavior changes)?
> 
> Surely you don't intend to change the scrubber output format?

I certainly continue to have plans to make another attempt to do so, yes.
It is simply too aggressive in the whitespace removal it does.

> (Say, "helpfully" allowing a space after comma because that's 
> low-effort to allow #NO_APP.)
> 
> The scrubber is intended to *allow* these "assumptions"!

Some, but not all. It is intended to "canonicalize" the input (e.g. by
collapsing successive whitespace and comments into a single blank), but
it surely isn't supposed to e.g. convert "< <" into a shift operator
(going from memory here; things along these lines certainly happen).

Plus parsing code, when not allowing for any whitespace on the assumption
that the scrubber would have removed it, would break when -f is passed to
gas.

Jan