keywords duplicates and unwanted keywords
 

I am using version 1.8.7.

My keywords table has entries that end with characters such as:
,.-:;# and 's

I would like to ensure that these values are stripped from words when indexing.

I checked the source against the instructions in this thread and found that the new version has this code implemented.

My kewords table also includes entries that end with:
.doc .mid .shtml .html .gif .jpg .pnp

I would like to ensure that these entries are no longer indexed.

I would appreciate any help.

That, fortunately or unfortunately, is how PhpDig 1.8.7 works to deal with words followed by punctuation marks. PhpDig 1.8.8 RC1 avoids the issue.

Hi Charter,

Due to the new requirements, I cannot use PhpDig 1.8.8 RC1. Is it possible that I can use some of the code from 1.8.8 RC1 in 1.8.7 to resolve the problem?

Sorry, I don't have a list of code changes from 1.8.7 to 1.8.8 RC1, and 1.8.8 RC1 uses mb_* for multi-byte processing. :(

Charter,

What is muti-byte processing and what will it do?

PhpDig v.1.8.8 RC1 converts text to UTF-8 and stores it as UTF-8 in MySQL. Basically, some languages use more than one byte per letter/character, and different sites use different encodings, so the mb_* functions try to deal with these things. Check out http://www.php.net/manual/en/ref.mbstring.php for an intro to multibyte functions.