Differential D14382 Diff 49727 source/blender/blenlib/BLI_length_parameterize.hh

Changeset View

Standalone View

source/blender/blenlib/BLI_length_parameterize.hh

This file was added.

				/* SPDX-License-Identifier: GPL-2.0-or-later */

				#pragma once

				/** \file
				* \ingroup bli
				*/

				#include "BLI_math_base.hh"
				#include "BLI_math_color.hh"
				#include "BLI_math_vector.hh"
				#include "BLI_vector.hh"

				namespace blender::length_parameterize {

				/**
				* Return the size of the necessary lengths array for a group of points, taking into account the
				* possible last cyclic segment.
				*
				* \note This is the same as #bke::curves::curve_segment_size.
				*/
				inline int lengths_num(const int points_num, const bool cyclic)
				{
				return (cyclic && points_num > 2) ? points_num : points_num - 1;
				}
				JacquesLuckeUnsubmitted Done Inline Actions I think it is not obvious that a curve cannot be cyclic when it only has two points. I don't see a fundamental reason for why this should be the case. If there is a reason for this behavior, it should be mentioned here. JacquesLucke: I think it is not obvious that a curve cannot be cyclic when it only has two points. I don't…
				HooglyBooglyAuthorUnsubmitted Done Inline Actions I think I was copying the logic from existing curve code, but I also can't think of a fundamental reason, at this level. At a higher level, for a two point poly curve, cyclic might not make sense since since it can't be reflected visually. HooglyBoogly: I think I was copying the logic from existing curve code, but I also can't think of a…
				JacquesLuckeUnsubmitted Done Inline Actions Well, there are always ways to visualize it. We could just use the arrows that you see when adjusting the tilt. Then better mention that reason in a comment. JacquesLucke: Well, there are always ways to visualize it. We could just use the arrows that you see when…
				HooglyBooglyAuthorUnsubmitted Done Inline Actions Honestly I don't remember the reasoning well enough to document it. I'll just remove that 2 point check for now I think. If it becomes a problem we can add it back. HooglyBoogly: Honestly I don't remember the reasoning well enough to document it. I'll just remove that 2…

				/**
				* Accumulate the length of the next segment into each point.
				*/
				template<typename T>
				void accumulate_lengths(const Span<T> values, const bool cyclic, MutableSpan<float> lengths)
				{
				BLI_assert(lengths.size() == lengths_num(values.size(), cyclic));
				float length = 0.0f;
				for (const int i : IndexRange(values.size() - 1)) {
				length += math::distance(values[i], values[i + 1]);
				lengths[i] = length;
				}
				if (cyclic) {
				lengths.last() = length + math::distance(values.last(), values.first());
				}
				}

				struct ResultSamples {
				/**
				* The index of the sample's previous point. Note that this and #factors don't include
				* the first result point, which always copies the value of the first source point.
				*/
				Vector<int> indices;
				/** The sample's factor from the previous point to the next point. */
				Vector<float> factors;
				/** Sample factors for the last cyclic segment connecting the first and last point. */
				Vector<float> cyclic_samples;

				bool cyclic;

				#ifdef DEBUG
				int src_points_size;
				#endif

				/* Do linear interpolation of the values from the samples' neighboring points. */
				template<typename T> void interpolate_linear(const Span<T> src, MutableSpan<T> dst) const
				{
				#ifdef DEBUG
				BLI_assert(src.size() == this->src_points_size);
				#endif

				dst.first() = src.first();
				dst = dst.drop_front(1);

				if (cyclic) {
				for (const int i : this->indices.index_range()) {
				const int index = this->indices[i];
				const float factor = this->factors[i];
				dst[i] = math::interpolate(src[index], src[index + 1], factor);
				}
				MutableSpan<T> cyclic_results = dst.take_back(cyclic_samples.size());
				for (const int i : cyclic_samples.index_range()) {
				cyclic_results[i] = math::interpolate(src.last(), src.first(), cyclic_samples[i]);
				}
				}
				else {
				for (const int i : this->indices.index_range()) {
				const int index = this->indices[i];
				const float factor = this->factors[i];
				dst[i] = math::interpolate(src[index], src[index + 1], factor);
				}
				dst.last() = src.last();
				}
				}
				};

				/**
				* Find the given number of points, evenly spaced along the provided length. For non-cyclic
				* sequences, the first and last point will always be included, for cyclic sequences, the first
				* point will always be included.
				JacquesLuckeUnsubmitted Done Inline Actions Even when `count` is zero or one? JacquesLucke: Even when `count` is zero or one?
				HooglyBooglyAuthorUnsubmitted Done Inline Actions Good point, I adjusted the comment. And the function expects the count to be greater than zero so I added that too. HooglyBoogly: Good point, I adjusted the comment. And the function expects the count to be greater than zero…
				*/
				void create_uniform_samples(Span<float> lengths,
				const bool cyclic,
				const int count,
				ResultSamples &result);

				} // namespace blender::length_parameterize