Re: [cowiki-dev] Extension to coWiki text formatting

Archie Campbell wrote:

> I think that dtg & paulh are at cross-purposes. I'll go with dtg's 
> fore-and-back scheme. No need for a bug.

It's OK to accept both as long as the reconstructed document is the same 
as the original.  The reason I don't like the <plugin ..> is that it 
implies the need for a </plugin>, which is invalid.  Also anything 
between the tags is lost and not available to the plugin.

> Re bug #195, WikiParser accepting <u>, <i>, <b>... I could fix a 
> scheme for user-defined attributes in the output HTML, such that <u 
> origin="wiki"> back-parses to "__" and <u origin="html"> back-parses 
> to "<u>". Any takers? Comments on expanding this? Put downs?

I can't think of a probem with it...  I haven't looked at the code 
handling this, my off-the-top-of-my-head fear was that a new tag was 
needed to represent every formatting source.  I see now that is not 
necessary.  The light bounces back and forth inside my management-type 
head!  :-)

>
> BTW, I wasn't quoting the source, I was attaching the files for you 
> guys to test. I need some feedback on the changes to the four files 
> class.WikiParser.php, class.WikiReverseParser.php and in a minor way, 
> class.FrontHTMLTransformer.php.

I understood that.  Realistically I'm not going to get to it till 
Thursday, maybe Friday at the earliest.

>
> What is the consensus on <p>-aragraphs? The parsers, together, seem 
> [sic] to like putting everything on one line unless broken by <br/>. 
> This is true-to-html behaviour. Thing is, I could be slotting 
> <p>...</p> around these lines without much effort. I'm thinking of 
> somebody's future javascript, where the <p></p> is the only containing 
> element we have, except for a <div> within which justification takes 
> place. Comment?

Are you talking about internal representation, what we display back to 
the user when they edit, or the html output?

In Wiki-text, the end of line, in a non-blank line, is a paragraph 
break.  Multiple empty lines produce no output and silently disappear 
from the page text.

>
> If you like, I'll commit to cvs, but I don't want to break the HEAD again.
>
> Talking of cvs-HEAD, my "D_ALL" tree stops dead after loading 
> StorageMySQL.

My cvs HEAD seems to work OK...  Have you tried a new sandbox?

>
> My working tree is 0.3.4 with patches to the three files I've been 
> working on. Works fine.
>
> Regards,
>
> Archie
>
> Daniel T. Gorski wrote:
>
>>On 18 Apr 08:46, PAUL HANCHETT wrote:
>>  
>>
>>>I think the syntax for plugin needs to change-- It should be <plugin ... 
>>>/>.  I see no problem with accepting <plugin ... > and changing it to 
>>>the proper version.
>>>    
>>>
>>>Should I bug this?
>>>    
>>>
>>
>>It would be OK, if the "reversed" (from XML to coWiki) version is <plugin
>>...> again.
>>
>>coWiki document: <plugin ...> or <plugin .../> will be accepted
>>Redisplay: <plugin ...>
>>
>>regards dtg
>>
>>P.S. Please, please do not quote the source code in your replies.
>>
>>---------------------------------------------------------------------
>>To unsubscribe, e-mail: [email protected]
>>For additional commands, e-mail: [email protected]
>>  
>>