[TYPO3-english] Some cooluri questions

[Oliver Rowlands]

Hi Jörg,

Please see my comments below:

> 
>   1. There seems to be no .html at the end. Is this normal?

That is a mater of personal taste, by default CoolURI will not append 
the .html suffix at the end of it URIs, If you wish to change this add 
the following settings just after the opening <cooluri> node in your 
CoolUriConf.xml file:

<urlsuffix>.html</urlsuffix>
<removetrailingslash>1</removetrailingslash>

> 
>   2. I'd like to get rid of /blog/ in the url since it doesn't add
> anything descriptive. How would I go about that?

You can normally use the "Exclude this page from middle of a page path" 
option available in the page properties of every page. However this will 
only work if the page in question is in "the middle of a page page", I 
do no believe this will work in your situation. Without extending 
CoolURI I do not think that you will be able to solve the problem (apart 
from maybe moving your blog plugins to the home page).

> 
>   3. My wife's blog is in Japanese. If there is any roman character in
> the title, cooluri will just use that part for the url. However, if
> all the letters are Kanji-symbols, a URL like 
> http://www.example.com/kimagurenikki/2010/07/29/?tx_ttnews[tt_news]=1805
> is generated, because cooluri is "at a loss of words", so to speak. 
> 
> The part where it says kimagurenikki is the equivalent of "blog" in
> the above example. I just set the page title to the roman-character
> spoken equivalent of the symbols, something which every Japanese
> understands. Incidentally, that is also how asians generate their
> speaking urls, but which I guess would be something of its own
> doctorate thesis to implement. Any other suggestions how to beautify
> 3., though?

The CoolURI manual suggests using the following (though I have not 
tested it...):

<part>
	<parameter>tx_ttnews[tt_news]</parameter>
	<lookindb>
		<to>SELECT title FROM tt_news WHERE (uid=$1 or l18n_parent=$1) AND 
sys_language_uid={L=0}</to>
		<t3conv>1</t3conv>
	</lookindb>
</part>

Alternatively it would be possible to extend CoolURI (using the 
<userfunc> node) and use one of the romanisation libraries (e.g. KAKASI 
or MeCab) to convert the Japanese URL parts in rōmaji. Though as you can 
probably guess this would obviously require far work...

HTH,

Oliver

-- 
Oliver Rowlands
:: Liquid Light ::

E - oliver at liquidlight.co.uk
W - http://www.liquidlight.co.uk

T - 00 44 (0)845 6 58 88 35
F - 00 44 (0)845 6 58 44 35