That is correct. The attribute really has nothing to do with the problem. Also, there is no problem if default formating, textDevice is used. I think Michael is probably following the right path in attempting to locate whatever is causing empty <brl> elements. It will take a lot of tracing through the transcribe_math.c module. John On Wed, Mar 05, 2014 at 01:10:18PM +0000, Keith Creasy wrote: > Hi John. > > It isn't the attribute itself that is probably causing the problem it is the > fact that the value changes which initiates a new function call. The > attribute itself is just showing Michael something about what is happening. A > good thing I believe. > > Keith > > > -----Original Message----- > From: liblouis-liblouisxml-bounce@xxxxxxxxxxxxx > [mailto:liblouis-liblouisxml-bounce@xxxxxxxxxxxxx] On Behalf Of Michael > Whapples > Sent: Wednesday, March 05, 2014 6:07 AM > To: liblouis-liblouisxml@xxxxxxxxxxxxx > Subject: [liblouis-liblouisxml] Re: UTDML semantics attribute > > Hello, > To explain what I was really asking about: APH have a problem with multiple > brl elements being placed after some math elements. > > From my inspection of the code it appears what might be causing this is when > an element has a style (the particular example I am thinking of is mtable > which has style matrix). > > From what I find in the code when a style applies to a MathML element (eg. > mtable) then mathTrans is called when starting the style and when ending the > style. This will lead to two brl elements. In addition the top level math > element also makes a call to mathTrans which creates another brl element. > > Are all these calls to mathTrans required and why? Having multiple brl > elements next to each other do cause issues. > > I am trying to establish whether the code is doing what was intended or > whether this is an unintended consequence. > > Michael Whapples > On 05/03/2014 10:23, John J. Boyer wrote: > > The semantics attribute was added at the request of APH. It is used in > > BrailleBlaster. The part before the comma is either action or style. The > > part after the comma is the name of the action or style. > > These names are derived from the semantic-action and configuration files > > that are used. > > > > John > > > > On Wed, Mar 05, 2014 at 09:34:31AM +0000, Michael Whapples wrote: > >> Hello, > >> I don't think semantics attribute was part of the original > >> specification for UTDML and I don't think its purpose has been documented > >> anywhere. > >> > >> Could someone explain the semantics attribute which gets added to > >> elements in UTDML documents? > >> > >> All of the examples I have seen the semantics attribute contains two > >> values separated by a comma. It appears may be the second is the > >> action name from the semantic action file which got applied to the > >> element. The first value though I am not sure about. Mostly it seems > >> to be set to action, but occasionally it is style. What does style > >> for the first value in semantics attribute mean? > >> > >> 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 > For a description of the software, to download it and links to > project pages go to http://www.abilitiessoft.com -- John J. Boyer; President, Chief Software Developer Abilitiessoft, Inc. http://www.abilitiessoft.com Madison, Wisconsin USA Developing software for people with disabilities For a description of the software, to download it and links to project pages go to http://www.abilitiessoft.com