[reportlab-users] paragraph mark up in unicode reportlab release

[Tim Roberts]

On Sun, 26 Mar 2006 09:59:14 +0000, Robin Becker <robin at reportlab.com>

>  
>
>yes eg in the utf8 branch we have
>
>greeks = {
>	'alefsym': '\xe2\x84\xb5',
>	'Alpha': '\xce\x91',
>	'alpha': '\xce\xb1',
>	'and': '\xe2\x88\xa7',
>	'ang': '\xe2\x88\xa0',
>	'asymp': '\xe2\x89\x88',
>	'Beta': '\xce\x92',
>	'beta': '\xce\xb2',
>	'bull': '\xe2\x80\xa2',
>	'cap': '\xe2\x88\xa9',
>	'Chi': '\xce\xa7',
>	'chi': '\xcf\x87',
>	'clubs': '\xe2\x99\xa3',
>	'cong': '\xe2\x89\x85',
>	'cup': '\xe2\x88\xaa',
>........................
>
>the main problem is how to support the idea of escaping to another font 
>when we see a problem in the rendering phase.
>

Granted that I'm chiming in at the end of a lengthy process, but
wouldn't it have been better to have a table like this entirely Unicode,
and use the Python codecs module to convert them to utf8 as needed? 
Each of those maps to a single Unicode code point, which is identical
regardless of the final encoding.  It seems to me what you have is an
opportunity for hard-to-find typos.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.