On Wed, Dec 16, 2009 at 2:36 PM, Christopher Wright <[email protected]>wrote:
> It works as expected for me (booted in 10.5.8), without having to re-enter >> anything, or anything buggy. Haven't tested in 10.6. >> > > > Just as a quick question, what exactly is "expected"? (I understand this > is mostly philosophical -- I'm just curious, and haven't had time to > experiment) > I assumed that you were shooting for matching the string replace function of the grep patch. It does that fine, with no flakiness. > > I ask because "Grep" is a command line tool for matching strings/regular > expressions (http://en.wikipedia.org/wiki/Grep). It doesn't do actual > replacement, which the grep patch does -- that's more like Sed ( > http://en.wikipedia.org/wiki/Sed), which will replace stuff. > That's a good point. I tend to think of grep as a search tool. Grep commands can be used for reorganizing strings, for instance. So, I would expect a grep patch to search for *something* and then do* something. * * * I think you're right on though; it would seem more intuitive for a grep patch to be a tool for matching strings that returned boolean, or perhaps returned a structure where the member matched the amount of words in the string, and the members would have either 0 or 1 for true/false. > > Shouldn't this patch be called "string replacer" or something (to parallel > string printer, string case, string components, string truncate, et al.)? > Grep should return a boolean (whether or not the string matched the search > string/expression). > Agreed. Also, we already have the "string compare" patch, which should likely be called "grep". > > The description explains what happens when Replace String isn't empty, but > what happens when it is? is it supposed to return unadulterated input (i.e. > no replace took place)? no output (no match)? replace matches with empty > string? > > It should probably search for empty spots and be able to return a structure, or simple boolean result, and the patch should likely have an on/off for the search. I guess an alternative is that a string length patch could be used to shut the render patch off if the string was empty, and by default it would just search for the "empty spaces". -George Toledo
_______________________________________________ Do not post admin requests to the list. They will be ignored. Quartzcomposer-dev mailing list ([email protected]) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/quartzcomposer-dev/archive%40mail-archive.com This email sent to [email protected]
