Even with the pervasiveness of XHTML, some web developers forget that the XML specification only defines five built-in character entities and most of the converted named entities that TextMate produces actually break some apps that rely on this strict interpretation.
It would be convenient if there was another conversion command that would use the numeric ISO character in place of the named entity. For example, instead of of converting a ' character to ' the proposed command would output ’ ...and – would instead be –
I would assume that TextMate would need to store an updated HTML -> XML lookup table to do this, similar to the old command.
On 19/9/2006, at 22:18, Glenn Batuyong wrote:
[...] It would be convenient if there was another conversion command that would use the numeric ISO character in place of the named entity. For example, instead of of converting a ' character to ' the proposed command would output ’ ...and – would instead be –
We can put a version in the XML bundle which only does numeric entities.
But seeing how XML has no encoding problems, would you need entities for anything but the special characters?
I would assume that TextMate would need to store an updated HTML -> XML lookup table to do this, similar to the old command.
Come again? Why would the command need to know named entities? You want it to first decode HTML entities, and then recode them as numeric entities?
-
Allan Odgaard -
Glenn Batuyong