Earlier I wrote:
> I can update the article to use a non-numeric name, but before I do
> it's worth exploring that relative to whichever version you have on
> Github, since there was at least one later version which appeared to
> be faster but turned out to cover fewer edge cases.
I've done some initial testing, and with the common test data in the
article both functions return the same result. The one from your GitHub
site is about 30% slower, but apparently for good reason: when testing
on other CSV files I use locally, I found at least one edge case where
the new algo handles some escaped escapes better.
I've updated my article with the new code, along with updating the URL
that points to LiveCode and the URL to the source, now pointing to your
Github repository:
<http://www.fourthworld.com/embassy/articles/csv-must-die.html>
The only suggestion I would make for the Github repo is in the Read Me:
While I believe I may have initiative the first of the several threads
suggesting we should pull our minds together and solve the CSV issue for
LiveCode, we've also had so many contributions to the source I no longer
recall who suggested the first posted version, and many people
contributed ideas along the way.
So where your Read Me says:
Originally a project by Alex Tweedley to improve Richard's code,
we eventually put it in the community...
...it may be more accurate to say:
After a community process with many contributors, Alex Tweedly
came up with what is currently the most robust general-purpose
CSV parser.
--
Richard Gaskin
Fourth World Systems
Software Design and Development for the Desktop, Mobile, and the Web
____________________________________________________________________
ambassa...@fourthworld.com http://www.FourthWorld.com
_______________________________________________
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
