Ah k, so it would be situational. After reading back in the thread some, I suspect the tests would be 2 different strings, so testing for both would be necessary. I was just not grokking the double. Probably just easier to set up a single string and check for it twice.
Either way, it runs through 100k records amazingly fast. (using 6.6.2 right now) Good to have the reminder that -or- checks only progress till the first hit, or failure to hit. On Thu, Apr 23, 2015 at 4:48 PM, Jerry Jensen <j...@jhj.com> wrote: > Actually it may be a bit faster with the two tests if a lot of them match > at the beginning. Testing at the beginning (begins with) is a lot faster > than spinning through the entire thing (contains), and if the first clause > of the OR is satisfied, it won’t execute the second clause. So, maybe. > > I probably wouldn’t write it with the two tests unless I was really in a > hurry and proved it to be an advantage with typical test data. > > Jerry > > > On Apr 23, 2015, at 3:38 PM, Mike Bonner <bonnm...@gmail.com> wrote: > > > > Did some testing out of curiosity, and WOW this is fast. > > > > > > repeat for each key tKey in tArray > > if tArray[tKey] begins with tSearchString or \ > > tArray[tKey] contains tSearchString then > > put tKey & return after tResults > > end if > > end repeat > > > > The only question I have is.. why search for the same thing twice? If it > > "contains" it also "begins with" so its a redundant check. (making it > > faster) Am I missing something here? > > > > On Thu, Apr 23, 2015 at 4:01 PM, Geoff Canyon <gcan...@gmail.com> wrote: > > > >> On Wed, Apr 22, 2015 at 6:22 PM, Peter Haworth <p...@lcsql.com> wrote: > >> > >>> Out of interest, I added a test which used combine and filter. It took > >>> around 3 times longer than the other two tests. > >>> > >> > >> Yeah, I didn't expect this to be competitive except under specific > >> circumstances -- if the filter command happens to be suited to the > >> (relatively complex) search performed, for example. I didn't see 3x, > more > >> like 1.5x to 2x, but still, iterating is generally the way to go. > >> _______________________________________________ > >> use-livecode mailing list > >> use-livecode@lists.runrev.com > >> Please visit this url to subscribe, unsubscribe and manage your > >> subscription preferences: > >> http://lists.runrev.com/mailman/listinfo/use-livecode > >> > > _______________________________________________ > > use-livecode mailing list > > use-livecode@lists.runrev.com > > Please visit this url to subscribe, unsubscribe and manage your > subscription preferences: > > http://lists.runrev.com/mailman/listinfo/use-livecode > > > _______________________________________________ > use-livecode mailing list > use-livecode@lists.runrev.com > Please visit this url to subscribe, unsubscribe and manage your > subscription preferences: > http://lists.runrev.com/mailman/listinfo/use-livecode > _______________________________________________ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
