Open dunnl opened 1 year ago
Ah, indeed — I haven't thought about tabs at all.
The reST spec says:
Spaces are recommended for indentation, but tabs may also be used. Tabs will be converted to spaces. Tab stops are at every 8th column (processing systems may make this value configurable).
How bad would it be to convert all tabs to spaces as part of the conversion process? Not exactly in line with the whole "reversible conversion" aspect, but very simple to implement ^^
I imagine the alternative would be to adjust the way the code measures indentation, but it will require in-depth testing to make sure that all cases are properly handled.
Ah, but this isn't sufficient, because dedenting a line may require replacing a tab with spaces before removing leading spaces. I will look into improving things, but given that tabs aren't used very commonly with reST, I'm curious to hear your thoughts on
How bad would it be to convert all tabs to spaces as part of the conversion process?
I think that is reasonable. I encountered this issue when some editing I was doing silently inserted tab characters I didn't want.
Excellent. I will try to implement this.
In
alectryon.el
, going from reST mode to Coq mode (or saving a file while opened in reST mode) does not handle tab characters that appear incoq
blocks. Instead Alectryon seems to begin a new prose block.Sample input (in reST mode). Note that
auto
here is indented with a hard tab character.Sample output from saving or switching to Coq mode:
Output obtained if the tab is replaced with spaces: