I think it has helped me understand what needs to be done. I might do things a little differently to how you describe.As an example, to indicate locked content I would probably have a separate locked attribute which could take value true or false. It just feels more normal to do it that way. Also may be by doing that it leads to simpler processing when multiple things apply to a brl node (eg. what if modifiers were to have another value as well, such as notext because its an equation as well as it being locked, one needs to parse a delimited string. Its simpler to check separate attributes.
One thing XML is normal not, is compact.Dealing with locked Braille when the original text changes is probably the biggest challenge, that is very unclear to me what should happen.
Do transcriber notes have a "original" text? Michael Whapples On 12/05/2014 14:20, John J. Boyer wrote:
at present liblouisutdml does not handle locked braille. This is because no decision was ever made as to how it should be implemented. The span tag inside <brl> nodes has a class attribute. In the case of locked braille this attribute could be given the value locked . However, it would be better to use the modifiers attribute on the <brl> node itself and give it the value locked. Some modifications on how liblouisutdml handles <brl> nodes during a retranslation will also be necessary. If the translation of something in the original text is modified and locked that portion of the text will have to be marked up in some way. There is a notranslate semantic action. This might be applied to brl,modifiers,locked . In this case the <brl> node would contain braille directly, with no index. Hope this helps, John On Mon, May 12, 2014 at 01:17:08PM +0100, Michael Whapples wrote:Hello, Can liblouisutdml yet handle locked Braille for things like transcriber's notes? We are needing to implement this in BrailleBlaster and so LibLouisUTDML will need to support it if BrailleBlaster is to work properly. If LibLouisUTDML does support locked Braille how does it currently support it (IE. what XML markup should be used)? If nothing has been defined then we were thinking of doing something similar to how page numbers are done in UTDML. Thinking about it though, we would have to mark locked Braille in another way because page numbers would not be expected to be maintained through a translation as potentially the pages may have changed if content has changed, whereas locked Braille should be maintained through the translation process. I seem to have a memory that it had been suggested that locked Braille would be noted by a locked attribute, possibly on the span or the brl node. Also were there cases where locked Braille may be associated to print/ink content? I think the reason for that would be where a manual translation correction had to be made. Michael Whapples For a description of the software, to download it and links to project pages go to http://www.abilitiessoft.com
For a description of the software, to download it and links to project pages go to http://www.abilitiessoft.com