[liblouis-liblouisxml] Re: Update output positions during multi pass (forward translation only)

  • From: "Arend Arends" <mada73bg@xxxxxxxxxx>
  • To: <liblouis-liblouisxml@xxxxxxxxxxxxx>
  • Date: Mon, 2 Nov 2015 17:33:39 +0100

Hi Susan,

Thank you for sharing your thoughts. In my case I try to provide this for any (LibLouis) table. It doesn't have to be perfect, but the replacements by the multipasses break the output positions array at the moment. It would be easier to make a special test like you describe if it was just for one table, but I have to deal with the preferences of all the people that use my program.

Arend

-----Oorspronkelijk bericht----- From: Susan Jolly
Sent: Monday, November 2, 2015 4:17 PM
To: liblouis-liblouisxml@xxxxxxxxxxxxx
Subject: [liblouis-liblouisxml] Re: Update output positions during multi pass (forward translation only)

Hi Arend,

Thanks for the explanation. I see now what I was missing.

There is somewhat of a conceptual prblem although it may not affect the
liblouis implementation of outpos. In UEB at least sometimes what we think
of as an indicator is actually part of a symbol. For example, a standalone
number sign is not an indicator. The UEB Numeric Indicator is the two-cell
symbol consisting of the number sign and the following digit or special
character. This symbol thus has a dual purpose; it represents a print
character and it also causes a mode switch. (This interpretation is
important for supporting automated backtranslation.)

Also according to the UEB rules a single capital letter is a two-cell
symbol. However, the indicators used for sequences of capital letters are
just indicators with similar logic to typeform indicators which I believe
are always considered as indicators since they are like tags in HTML.

This type of complexity is one thing I hoped to avoid with extended braille.

Here is a thought experiment of how you could use an extended braille
approach to identify indicators in liblouis at least for six-dot braille.
You would need to change the translation table to use 8-dot cells with at
least one dot in the bottom row for any braille cells that you want to
identify as indicators for your purpose. If I understand how liblouis works
the braille output would retain these 8-dot cells. You could then
postprocess the braille output to identify the indicators and also convert
it to standard six-dot braille. I realize this is probably not something
you'd want to do right now but it might turn out to be a useful idea for
future work.

Thanks again,
Susan

For a description of the software, to download it and links to
project pages go to http://liblouis.org

For a description of the software, to download it and links to
project pages go to http://liblouis.org

Other related posts: