Converting to title case

[James Hurley]

>
> Message: 3
> Date: Wed, 25 Aug 2010 15:44:01 -0300
> From: Andre Garzia <andre at andregarzia.com>
> Subject: Re: Converting to title case
> To: How to use Revolution <use-revolution at lists.runrev.com>
> Message-ID:
> 	<AANLkTikm63Ci3PKLtP9Qu57gfbb4O99aaHM1TmzDSXOt at mail.gmail.com>
> Content-Type: text/plain; charset=ISO-8859-1
>
> jim,
>
> why not this:
>
> put fld "list" into tContent
>
> repeat for each word tName in tContent
>  replace tName with (toUpper(char 1 of tName) & toLower(char 2 to -1  
> of
> tName))
> end repeat
>
> This will capitalize all your names right?
>
> -- 
> http://www.andregarzia.com All We Do Is Code.
>

Andrea,

Thanks for your suggestion. An interesting alternative.

But don't forget the "container" at the end of the "replace" command.  
If the container is tContent there will be considerable overlap.

For example, if tName is say "Auburn," a city in our county, and there  
were 1,000 instances of that word (since this is an election DB there  
will be thousands of instances of each city in the county) you will be  
replacing 1000 instance of "auburn" with "Auburn" and doing so 1000  
times. That is 1,000,000 replacements--most of them of course will be  
substituting "Auburn" for "Auburn."

What one could do is build a dictionary of the DB and then replace  
each incidence of each word from the dictionary. The advantage of this  
would be that there would be no danger of overlooking a white space  
that i hadn't recognized.  But the down side is that it is VERY slow.  
I waited an hour and it hadn't finished.

On the other hand, running through each character, even through there  
are 26,631,931 of them, takes only about 40 seconds on my very slow  
Mac Mini. But I only have to do it once for each new release of the  
election DB, so I have no incentive to optimize the script. My only  
worry, as I said, was that I hadn't overlooked some white spaces  
besides tab, space, enter and return.

Jim Hurley