Dear All,
somewhere between bonito-open-4.24.6 and bonito-open-5.55.5 the XML export behavior is changed. Currently the code looks like this:
```
left = ' '.join([n(scoll(x)) for x in l['Left']])
kwic = ' '.join([n(s(x)) for x in l['Kwic']])
right = ' '.join([n(scoll(x)) for x in l['Right']])
if params.get('viewmode') == 'kwic':
outf.write(' <left>%s</left>' % left + nl)
outf.write(' <kwic>%s</kwic>' % kwic + nl)
outf.write(' <right>%s</right>' % right + nl)
else:
outf.write(' %s %s %s' % (left, kwic, right) + nl)
```
In the old behavior the else branch did not exist and even when selecting "sentence" display mode instead of "kwic" the left, kwic and right tags were written to the output, currently their content is joined by spaces making them distinguishable (In the browser the kwic is still displayed in red, clearly distinguishable from the left and right context. One would expect the same in the XML output.)
I want to use the old behavior (i.e. one full sentence in each concordance with left, kwic, right tags separated/distinguishable automatically without external education). I cannot see any use case where one could not easily merge the content of left, kwic, right tags when needed, while the inverse operation is impossible.
1. Is there a rationale behind this new behavior?
2. Can I request reverting it back or adding an option to achieve it?
Thank you,
Balázs INDIG