Re: [9front] sed: fix moving '^' match

[ieliedonge@wilsonb.com]

> > we must be disagreeing about some fundamental understanding instead.
> > it's hard for me to see still where *exactly* we differ.
>
> ...
>
> In some sense, this interpretation of 'g' is considering that 'g' makes an
> infinite looping of the regex, applied to the resulting string and that should
> only end if no modification is made. Seems incorrect to me.

Oh, nice! I believe you lasered in on the crux here. IIUC, hiro's mental model
seems to be that s///g induces a simple loop over the string where '^' can
match the beginning of string multiple times, while mine (and others'?) is that
g identifies all matches *before replacement*.

I'm not sure what *should* be in this case, but the latter interpretation is
definitely consistent with the 5 or so different regex implementations I
regularly interact with.

> The current behavior may be bug compatible with V8. It remains a bug
> IMHO.

Just gave V8 a spin:

    # echo aabbccd | sed 's/^..//g'
    bbccd