reporting: Start span start line mis-calculation

[feds01]

In the following error report:

The highlighter mistakingly highlights the 9 line number. The span of the expression does not actually begin on the 9th line. This might be a mistake within the lexing implementation or a miscalculation within the report offset to (col, row) calculation. It's clear that the line of the span is for some reason being recorded as the end of line 9 where there is a \n character at the end.

Further investigation yielded that the actual span of expression (as printed from the raw source is):

loc=134:195, actual="IoError = struct(
    error: IoErrorType,
    message: str,
);"

Ideally, the report should not highlight the line number 9, like so:

This could be fixed by changing offset_col_row by not counting the \n in the case of the initial span offset, an implementation could be:

pub(crate) fn offset_col_row(offset: usize, source: &str, non_inclusive: bool) -> (/* col: */ usize, /* row: */ usize) {
    let source_lines = source.split('\n');

    let mut bytes_skipped = 0;
    let mut total_lines: usize = 0;
    let mut last_line_len = 0;

    let mut line_index = None;
    for (line_idx, line) in source_lines.enumerate() {
        // One byte for the newline
        let skip_width = line.len() + 1;


        // Here, we don't want an inclusive range because we don't want to get the last byte because
        // that will always point 
        let range = if non_inclusive {
            bytes_skipped..bytes_skipped + skip_width
        } else {
            bytes_skipped..bytes_skipped + skip_width + 1
        };

        if range.contains(&offset) {
            line_index = Some((offset - bytes_skipped, line_idx));
            break;
        }

        bytes_skipped += skip_width;
        total_lines += 1;
        last_line_len = line.len();
    }

    line_index.unwrap_or((
        last_line_len.saturating_sub(1),
        total_lines.saturating_sub(1),
    ))
}

[kontheocharis]

What determines if a report span should be non_inclusive?

[feds01]

When the start_row, start_col is being computed

[kontheocharis]

It would be interesting to see the col:row start and end points for this example.