Well done John. It looks like you are making amazing progress with the parser, 
which I would say was the single most important feature to ease the future 
development of FlexWiki. With a good parser in place it should be easier to add 
new functionality and push FlexWiki to the forefront of wiki products again.

Thanks,

Derek.


From: John Davidson 
Sent: Saturday, May 10, 2008 2:48 PM
To: FlexWiki Users Mailing List 
Subject: [Flexwiki-users] New Features in New Parser


Progress on the new parser can be followed at [1]. The rule sets are nearing 
completion with the current implementation of tables done and most links now 
being managed in addition to handling of emoticons, lists, headers, etc. 


One of the frustrations from the original parser was that some links that ended 
with a period worked, while others did not. This condition was documented, but 
still caused problems. The new parser will now allow punctuation at the end of 
a link and ignore it.


More importantly the new parser will make it possible to add new features 
easily, where this would have been extremely difficult in the original parser. 
Feedback and discussion on any of these proposals is welcomed.


First is the addition of the 'valign' style property to table rows or table 
cells. The choices will be 'VT' - top, 'VM' - middle, 'VB' - bottom or 'VBL' - 
baseline. The token must be enclosed in the style sequence '{.....}'. If the 
token is in the first cell in the row it will be applied to the html <tr> 
element. If it is in any other cell in a row it will be applied to the 
corresponding cell html <td> element.


Similar treatments can be added for table padding and spacing to override the 
default values in CSS for the entire table. These would again be applied at the 
corresponding row or cell. The token would be 'Pdd' or 'Sdd' where 'dd' are 
digit values of pixels.


Definition lists - html element <dl> with sub-elements <dt> and <dd> - will be 
added with the indentation level controlled by each tab or 8 spaces equalling 
one level of indent up to 7 levels. The actual for of the input will start with 
a semicolon, then the term being defined, followed by a colon and then the 
actual definition of the term (see below):


       ;term:definition


Acronym - html element <acronym> - will be added by having the acronym directly 
followed by an opening bracket, then the full text representation of the 
acronym followed by the closing bracket. This can occur anywhere in text. This 
element is useful for style sheets and search engines. An example follows:


     ABC(American Bridge Club)


Finally the big one. Tables with multiple lines in any cell, where those lines 
can include any wiki text except another embedded table. For example, this will 
allow the easy placement of lists within a table cell. In order to achieve this 
it is necessary to define a new token to identify table row start and end. 
These will be '[||' and '||]' respectively, where the row start is found at the 
start of a line of wiki text and the row end is at the end of a line of wiki 
text, with any number of lines of wiki text intervening between the two tokens. 
The new cell token will remain as '||'. The existing single-line table row 
definitions will still be supported and current topics that use wikitalk to 
simulate multiline table will continue to function correctly. An example of the 
new format is shown below (note that the position of the cell break token '||' 
is unimportant):


[||{T^}        * first item of unordered list
        * second unordered list item
        * third unordered list item (all in first cell of first row ||
The second cell of the first row ||]
[|| first cell of second row
|| second cell of second row
contain LF/CR in text ||]


I have not yet really looked at including any microformat definitions, but 
certainly these could be included if I receive input about which ones should be 
supported. It should also be noted that the implementation of these rule sets 
will allow local wiki administrators to update and include their own rule 
definitions, as the rule sets are in xml configuration files that are read by 
the application. 




[1] http://ods.dyndns.org/NewParser/default.aspx


John Davidson


--------------------------------------------------------------------------------


-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
Don't miss this year's exciting event. There's still time to save $100. 
Use priority code J8TL2D2. 
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone 


--------------------------------------------------------------------------------


_______________________________________________
Flexwiki-users mailing list
Flexwiki-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flexwiki-users
-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
Don't miss this year's exciting event. There's still time to save $100. 
Use priority code J8TL2D2. 
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
Flexwiki-users mailing list
Flexwiki-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flexwiki-users

Reply via email to