Sparklines 1.0.0, in major update available

[deeplook]

Maybe of interest for this project being a large user of sparklines. I've just released 1.0.0 today with major new features and bug fixes. Here are some of them:

• Mixed positive/negative data is now auto-detected and rendered as a
  proportional split sparkline automatically — no flags required. Upward bars
  appear on top, downward bars below.
• New -n / --num-lines accepts three forms: a positive integer (total rows,
  allocated proportionally), 'auto' (smallest proportional row count), or
  'up:down' (explicit per-side layout). Integer and auto modes use a shared
  max scale; up:down uses independent per-side scaling.
• New --zero flag: 'up' (default) places zeros on the positive baseline;
  'none' omits zeros from both sides.
• Proportional row allocation: up_rows/total == pos_max/range, so the zero
  line stays meaningful as the visual boundary between halves.
• New index-slice form for -e/--emphasize: "color:[start:stop:step]" selects
  bars by position rather than value, using standard Python slice syntax
  (e.g. "red:[0:3]", "blue:[::2]", "yellow:[-1:]"). Value and slice
  expressions can be mixed freely.