[Pmwiki-users] comment/vote on proposed changes

[Patrick R. Michaud]

On Tue, May 06, 2003 at 12:14:17AM -0600, Patrick R. Michaud wrote:
> 
> 5.  Backticks in character strings.  John Rankin has suggested a scheme
> where a backtick (` == accent grave, not a single quote) in the middle of
> a word is inline replaced by an empty string.  

Here are a couple of discussion items on the backticks proposal:

1.  In April, John Rankin proposed the following substitution pattern
for backticks:
	([A-Za-z0-9.\\/])`([A-Z])
Basically this says that any backtick after a letter, digit, dot, or
slash and before an uppercase letter is removed.  Thus CI`Os would
remove the backtick but Gr`oupWise would not.  However, 
Text`FormattingRules would still have FormattingRules as a WikiWord
and would have to be in the markup as Text`Formatting`Rules to avoid this.

1a.  Is there an advantage to be gained (and what disadvantages exist)
in using the pattern 
        ([A-Za-z0-9.\\/])`([A-Za-z])
instead?  This would remove backticks in front of lowercase letters
in addition to uppercase ones, thus T`extFormattingRules would not
have any WikiWords in it because WikiWords cannot begin with a lowercase
letter.

1b.  Since backticks are used rarely, how about simply saying that
any CamelCase string preceded with a backtick is not a WikiWord and has the
backtick removed.  Thus `WikiWord, `TextFormattingRules, `CIOs, etc.
are all non-wikiwords.  If someone really needs a backtick in front
of a WikiWord link, they can always write [=`=]WikiWord or `{{WikiWord}}
or any of a number of other combinations to achieve it.  (Note that this
proposal can be combined with the previous ones if compatibility is
an issue.)

2.  One person privately commented that we may be eating up a valuable
future markup character (the backtick) when there's already a mechanism 
for preventing WikiWord stuff (the [=...=] sequence).  Anyone have the
same concern?

Comments greatly desired.

Pm