Re: [Nano-devel] Update

[David Lawrence Ramsey]

Jordi Mallach:
>Yes, David came back with some replacement for the dollar patch that is
>in the 1.2 branch right now. I can't find the url for it in the archive
>though.

Never mind about that one, it turned out to be broken too.  For example,
it would sometimes replace on the starting line twice.  What I said
before about having the search/replace skip lines it'd already been to
in the cases of zero-length regexes was incorrect before since the code
actually skipped positions (I don't know what I was thinking when I
wrote that) but turned out to be the right solution, at least according
to all the test cases I checked in an attempt to finally get it right:

forward replace "^" with a blank
forward replace "$" with a blank
forward replace "^$" with a blank
forward replace "^" with a non-blank
forward replace "$" with a non-blank
forward replace "^$" with a non-blank
backward replace "^" with a blank
backward replace "$" with a blank
backward replace "^$" with a blank
backward replace "^" with a non-blank
backward replace "$" with a non-blank
backward replace "^$" with a non-blank

So it's in for now.  It still is a bit weird in that it defines two
local match_len's, but I'll clean that up soon.

Along with that, I've committed what I said needed to go in before
1.3.1, since it all seems to be holding (although for some strange
reason the name on the commits ended up being "uid67109" instead of
"dolorous").  That means:

* nanodbcut.patch, so the cutting code doesn't reference editbot anymore
and the related fixes/extensions to the file-writing code (including the
selection-writing code) are in

* nanoinput2.patch, so most escape sequences are supported again and  
some input problems/design oddities are fixed

* an additional bit of code to highlight zero-length matches as single
highlighted spaces, as Mike Frysinger suggested awhile back, since it's
relatively simple

The FAQ entry for verbatim input is still needed; I'll get to it next. 
In the meantime, there seems to be enough in the way of code changes to
warrant starting stability testing again so that 1.3.1 can be released
in the near future.  So would everyone start testing the code to make 
absolutely sure that it works as it should?  Thanks.


_____________________________________________________________
Sluggy.Net: The Sluggy Freelance Community!