At present, I'm pretty sure Apply will just track every line which comes through it, including those which are the same as prior lines only one column over. Line tracking should be reduced, both to save space and to make the results more useful to users. Reduction should be according to the following rules:
First Line will be the first line and column in the parse, excluding leading whitespace
Last Line will be the last line and column in the parse, unless the lineNo is the same as the lineNo of the first line, in which case it should be discarded
All lines with distinct lineNo values must be included at least once
Lines which have the same lineNo and a different colNo value from a line which is already included by some other rule should be discarded.
Lines should be ordered by lineNo
Thus, any given lineNo will appear at most once, with the first and last Line instances bounding the start/end columns for the fragment.
At present, I'm pretty sure
Applywill just track every line which comes through it, including those which are the same as prior lines only one column over. Line tracking should be reduced, both to save space and to make the results more useful to users. Reduction should be according to the following rules:Linewill be the first line and column in the parse, excluding leading whitespaceLinewill be the last line and column in the parse, unless thelineNois the same as thelineNoof the first line, in which case it should be discardedlineNovalues must be included at least oncelineNoand a differentcolNovalue from a line which is already included by some other rule should be discarded.lineNoThus, any given
lineNowill appear at most once, with the first and lastLineinstances bounding the start/end columns for the fragment.