kovidgoyal/kitty · error · ValueError

Linear easing curve must have strictly increasing points: {p

Error message

Linear easing curve must have strictly increasing points: {params} does not

What it means

When kitty fills in missing x-axis values for a linear easing curve it computes a delta; if the implied/computed x points are not strictly increasing (delta <= 0), the curve is invalid and this error is raised.

Source

Thrown at kitty/options/utils.py:1676

            raise ValueError('cubic-bezier easing function must have four points')
        return cls(type='cubic-bezier', cubic_bezier_points=(unit_float(parts[0]), float(parts[1]), unit_float(parts[2]), float(parts[3])))

    @classmethod
    def linear(cls, params: str) -> 'EasingFunction':
        parts = params.split(',')
        if len(parts) < 2:
            raise ValueError('Must specify at least two points for the linear easing function')
        xaxis: list[float] = []
        yaxis: list[float] = []

        def balance(end: float) -> None:
            extra = len(yaxis) - len(xaxis)
            if extra <= 0:
                return
            start = xaxis[-1] if xaxis else 0.0
            delta = (end - start) / max(1, extra - 1)
            if delta <= 0.0:
                raise ValueError(f'Linear easing curve must have strictly increasing points: {params} does not')
            if xaxis:
                for i in range(extra):
                    xaxis.append(start + (i + 1) * delta)
            else:
                for i in range(extra):
                    xaxis.append(i * delta)

        def add_point(y: float, x: float | None = None) -> None:
            if x is None:
                yaxis.append(y)
            else:
                x = unit_float(x)
                balance(x)
                xaxis.append(x)
                yaxis.append(y)

        for r in parts:
            points = r.strip().split()

View on GitHub (pinned to 6d5d0c4406)

Solutions

1. Ensure all x coordinates strictly increase from 0 toward 1
2. Omit x values entirely and let kitty balance them, giving only y values

Example fix

# before
cursor_blink_interval 0.5:linear(0 0, 0 1)
# after
cursor_blink_interval 0.5:linear(0 0, 1 1)

def valid_linear_xs(parts: list[str]) -> bool:
    xs = [float(p.split()[0]) for p in parts if len(p.split()) == 2]
    return all(b > a for a, b in zip(xs, xs[1:]))

When it happens

Trigger: Supplying x-axis values in a linear() easing that repeat or decrease relative to the auto-balanced points, e.g. linear(0, 0, 1) style inputs causing delta <= 0.

Common situations: Hand-authoring linear easing with duplicate or descending x coordinates, or mixing y-only and x,y points inconsistently.

Related errors

• cubic-bezier easing function must have four points
• Must specify at least two points for the linear easing funct
• {r} has too many points for a linear easing curve parameter
• {jt} is not a valid jump type for a linear easing function
• {spec} specified more than two easing functions

AI-assisted analysis of kovidgoyal/kitty@6d5d0c4406 (2026-08-27). Data as JSON: /api/errors/5276664f9a8b3cf9. Report an issue: GitHub.