This is the mail archive of the
mailing list for the Cygwin project.
- From: Achim Gratz <Stromeko at nexgo dot de>
- To: cygwin-apps at cygwin dot com
- Date: Fri, 12 Jun 2015 19:05:33 +0200
- Subject: Re: setup
- Authentication-results: sourceware.org; auth=none
- References: <87oakqnkfi dot fsf at Rainer dot invalid> <20150608193154 dot GU3416 at calimero dot vinschen dot de> <878ubtor9g dot fsf at Rainer dot invalid> <20150609095604 dot GA4993 at calimero dot vinschen dot de> <874mmghlf8 dot fsf at Rainer dot invalid> <20150610080526 dot GC31537 at calimero dot vinschen dot de> <871thjtq0m dot fsf at Rainer dot invalid> <20150610185417 dot GL31537 at calimero dot vinschen dot de> <87wpzbs2yj dot fsf at Rainer dot invalid> <87mw060xg3 dot fsf at Rainer dot invalid> <20150612102945 dot GS31537 at calimero dot vinschen dot de>
Corinna Vinschen writes:
> "known good" == -m?
> Thank you for getting rid of IniParseFindVisitor :)
I wasn't all that sure about this, but it simply made no sense to me for
the task at hand.
> However, IniParseFindVisitor searched the tree for ini files, so, does
> the new code still work for the standard local install with ini files
> copied from various mirrors, w/o -m option? It seems the answer is yes
> (you're iterating over the site list), but it would be helpful if this
> had been tested.
What's the expected directory structure here? The FindVisitor searched
the tree (two levels deep), but it rejected almost all files anyway. I
might have to go back to the original code to see what exactly it would
accept, but an almost exact copy of it is still in fromcwd (we might
want to get rid of that too at one point). From just reading that
pointer orgy it would appear that if dir doesn't end in a slash or
backslash and doesn't end in SETUP_INI_DIR (x86 or x86_64), then it's
not recognized anyway. That's the long winded way of saying that it
should look for a setup file either in the current directory or one
level down, I guess.
I think as it stands at the moment the search one level down from the
current directory is actually cut, since it was done again in the
IniParseFindVistor that I've removed. I think this should be added into
fromcwd to build a list that then gets iterated over in do_local_ini
just like the sitelist in the remote case.
> The patch is a bit big and mixes several things into one. It would
> be helpful for potential later blaming to split it into independent
> chunks to commit separately. E.g., in at least one instance you just
> reordered a few lines without functional change. This may better go
> in as a separate style patch.
I'll see how to split it once I have thought about the search again. It
ended up that way since I didn't find an easy way to keep the separate
parts both compiling and working, but since it's now in a more stable
form I can revisit that question again.
> Otherwise it looks ok, but I'd be more comfortable to see the splitted
> patches before ok'ing it.
No need for hurrying. I'm on the opposite side of the planet for a
week, so I'll take another look when I'm back.
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+
Waldorf MIDI Implementation & additional documentation: