PrusaSlicer-NonPlainar/src/libslic3r/Print.hpp

538 lines
24 KiB
C++

#ifndef slic3r_Print_hpp_
#define slic3r_Print_hpp_
#include "libslic3r.h"
#include <atomic>
#include <set>
#include <vector>
#include <string>
#include <functional>
#include "BoundingBox.hpp"
#include "Flow.hpp"
#include "PrintConfig.hpp"
#include "Point.hpp"
#include "Layer.hpp"
#include "Model.hpp"
#include "PlaceholderParser.hpp"
#include "Slicing.hpp"
#include "GCode/ToolOrdering.hpp"
#include "GCode/WipeTower.hpp"
#include "tbb/atomic.h"
// tbb/mutex.h includes Windows, which in turn defines min/max macros. Convince Windows.h to not define these min/max macros.
#ifndef NOMINMAX
#define NOMINMAX
#endif
#include "tbb/mutex.h"
namespace Slic3r {
class Print;
class PrintObject;
class ModelObject;
class GCode;
class GCodePreviewData;
// Print step IDs for keeping track of the print state.
enum PrintStep {
psSkirt, psBrim, psWipeTower, psGCodeExport, psCount,
};
enum PrintObjectStep {
posSlice, posPerimeters, posPrepareInfill,
posInfill, posSupportMaterial, posCount,
};
class CanceledException : public std::exception {
public:
const char* what() const throw() { return "Background processing has been canceled"; }
};
// To be instantiated over PrintStep or PrintObjectStep enums.
template <class StepType, size_t COUNT>
class PrintState
{
public:
PrintState() { for (size_t i = 0; i < COUNT; ++ i) m_state[i].store(INVALID, std::memory_order_relaxed); }
enum State {
INVALID,
STARTED,
DONE,
};
// With full memory barrier.
bool is_done(StepType step) const { return m_state[step] == DONE; }
// Set the step as started. Block on mutex while the Print / PrintObject / PrintRegion objects are being
// modified by the UI thread.
// This is necessary to block until the Print::apply_config() updates its state, which may
// influence the processing step being entered.
void set_started(StepType step, tbb::mutex &mtx) {
mtx.lock();
m_state[step].store(STARTED, std::memory_order_relaxed);
mtx.unlock();
}
// Set the step as done. Block on mutex while the Print / PrintObject / PrintRegion objects are being
// modified by the UI thread.
void set_done(StepType step, tbb::mutex &mtx) {
mtx.lock();
m_state[step].store(DONE, std::memory_order_relaxed);
mtx.unlock();
}
// Make the step invalid.
// The provided mutex should be locked at this point, guarding access to m_state.
// In case the step has already been entered or finished, cancel the background
// processing by calling the cancel callback.
template<typename CancelationCallback>
bool invalidate(StepType step, tbb::mutex &mtx, CancelationCallback &cancel) {
bool invalidated = m_state[step].load(std::memory_order_relaxed) != INVALID;
if (invalidated) {
#if 0
if (mtx.state != mtx.HELD) {
printf("Not held!\n");
}
#endif
mtx.unlock();
cancel();
mtx.lock();
}
return invalidated;
}
// Make all steps invalid.
// The provided mutex should be locked at this point, guarding access to m_state.
// In case any step has already been entered or finished, cancel the background
// processing by calling the cancel callback.
template<typename CancelationCallback>
bool invalidate_all(tbb::mutex &mtx, CancelationCallback &cancel) {
bool invalidated = false;
for (size_t i = 0; i < COUNT; ++ i)
if (m_state[i].load(std::memory_order_relaxed) != INVALID) {
if (! invalidated) {
mtx.unlock();
cancel();
mtx.lock();
invalidated = true;
}
m_state[i].store(INVALID, std::memory_order_relaxed);
}
return invalidated;
}
private:
std::atomic<State> m_state[COUNT];
};
// A PrintRegion object represents a group of volumes to print
// sharing the same config (including the same assigned extruder(s))
class PrintRegion
{
friend class Print;
// Methods NOT modifying the PrintRegion's state:
public:
const Print* print() const { return m_print; }
const PrintRegionConfig& config() const { return m_config; }
Flow flow(FlowRole role, double layer_height, bool bridge, bool first_layer, double width, const PrintObject &object) const;
// Average diameter of nozzles participating on extruding this region.
coordf_t nozzle_dmr_avg(const PrintConfig &print_config) const;
// Average diameter of nozzles participating on extruding this region.
coordf_t bridging_height_avg(const PrintConfig &print_config) const;
// Methods modifying the PrintRegion's state:
public:
Print* print() { return m_print; }
void config_apply_only(const ConfigBase &other, const t_config_option_keys &keys, bool ignore_nonexistent = false) { this->m_config.apply_only(other, keys, ignore_nonexistent); }
private:
Print *m_print;
PrintRegionConfig m_config;
PrintRegion(Print* print) : m_print(print) {}
PrintRegion(Print* print, const PrintRegionConfig &config) : m_print(print), m_config(config) {}
~PrintRegion() {}
};
typedef std::vector<Layer*> LayerPtrs;
typedef std::vector<SupportLayer*> SupportLayerPtrs;
class BoundingBoxf3; // TODO: for temporary constructor parameter
class PrintObject
{
friend class Print;
public:
// vector of (vectors of volume ids), indexed by region_id
std::vector<std::vector<int>> region_volumes;
t_layer_height_ranges layer_height_ranges;
// Profile of increasing z to a layer height, to be linearly interpolated when calculating the layers.
// The pairs of <z, layer_height> are packed into a 1D array to simplify handling by the Perl XS.
// layer_height_profile must not be set by the background thread.
std::vector<coordf_t> layer_height_profile;
// There is a layer_height_profile at both PrintObject and ModelObject. The layer_height_profile at the ModelObject
// is used for interactive editing and for loading / storing into a project file (AMF file as of today).
// This flag indicates that the layer_height_profile at the UI has been updated, therefore the backend needs to get it.
// This flag is necessary as we cannot safely clear the layer_height_profile if the background calculation is running.
bool layer_height_profile_valid;
// this is set to true when LayerRegion->slices is split in top/internal/bottom
// so that next call to make_perimeters() performs a union() before computing loops
bool typed_slices;
Vec3crd size; // XYZ in scaled coordinates
Print* print() { return m_print; }
const Print* print() const { return m_print; }
ModelObject* model_object() { return m_model_object; }
const ModelObject* model_object() const { return m_model_object; }
const PrintObjectConfig& config() const { return m_config; }
void config_apply(const ConfigBase &other, bool ignore_nonexistent = false) { this->m_config.apply(other, ignore_nonexistent); }
void config_apply_only(const ConfigBase &other, const t_config_option_keys &keys, bool ignore_nonexistent = false) { this->m_config.apply_only(other, keys, ignore_nonexistent); }
const LayerPtrs& layers() const { return m_layers; }
const SupportLayerPtrs& support_layers() const { return m_support_layers; }
const Points& copies() const { return m_copies; }
bool add_copy(const Vec2d &point);
bool delete_last_copy();
bool delete_all_copies() { return this->set_copies(Points()); }
bool set_copies(const Points &points);
bool reload_model_instances();
// since the object is aligned to origin, bounding box coincides with size
BoundingBox bounding_box() const { return BoundingBox(Point(0,0), to_2d(this->size)); }
// adds region_id, too, if necessary
void add_region_volume(unsigned int region_id, int volume_id) {
if (region_id >= region_volumes.size())
region_volumes.resize(region_id + 1);
region_volumes[region_id].push_back(volume_id);
}
// This is the *total* layer count (including support layers)
// this value is not supposed to be compared with Layer::id
// since they have different semantics.
size_t total_layer_count() const { return this->layer_count() + this->support_layer_count(); }
size_t layer_count() const { return m_layers.size(); }
void clear_layers();
Layer* get_layer(int idx) { return m_layers[idx]; }
const Layer* get_layer(int idx) const { return m_layers[idx]; }
// print_z: top of the layer; slice_z: center of the layer.
Layer* add_layer(int id, coordf_t height, coordf_t print_z, coordf_t slice_z);
size_t support_layer_count() const { return m_support_layers.size(); }
void clear_support_layers();
SupportLayer* get_support_layer(int idx) { return m_support_layers[idx]; }
SupportLayer* add_support_layer(int id, coordf_t height, coordf_t print_z);
SupportLayerPtrs::const_iterator insert_support_layer(SupportLayerPtrs::const_iterator pos, int id, coordf_t height, coordf_t print_z, coordf_t slice_z);
void delete_support_layer(int idx);
// methods for handling state
bool invalidate_state_by_config_options(const std::vector<t_config_option_key> &opt_keys);
bool invalidate_step(PrintObjectStep step);
bool invalidate_all_steps();
bool is_step_done(PrintObjectStep step) const { return m_state.is_done(step); }
// To be used over the layer_height_profile of both the PrintObject and ModelObject
// to initialize the height profile with the height ranges.
bool update_layer_height_profile(std::vector<coordf_t> &layer_height_profile) const;
// Process layer_height_ranges, the raft layers and first layer thickness into layer_height_profile.
// The layer_height_profile may be later modified interactively by the user to refine layers at sloping surfaces.
bool update_layer_height_profile();
void reset_layer_height_profile();
void adjust_layer_height_profile(coordf_t z, coordf_t layer_thickness_delta, coordf_t band_width, int action);
// Collect the slicing parameters, to be used by variable layer thickness algorithm,
// by the interactive layer height editor and by the printing process itself.
// The slicing parameters are dependent on various configuration values
// (layer height, first layer height, raft settings, print nozzle diameter etc).
SlicingParameters slicing_parameters() const;
// Called when slicing to SVG (see Print.pm sub export_svg), and used by perimeters.t
void slice();
// Helpers to slice support enforcer / blocker meshes by the support generator.
std::vector<ExPolygons> slice_support_enforcers() const;
std::vector<ExPolygons> slice_support_blockers() const;
private:
void make_perimeters();
void prepare_infill();
void infill();
void generate_support_material();
void _slice();
std::string _fix_slicing_errors();
void _simplify_slices(double distance);
void _make_perimeters();
bool has_support_material() const;
void detect_surfaces_type();
void process_external_surfaces();
void discover_vertical_shells();
void bridge_over_infill();
void clip_fill_surfaces();
void discover_horizontal_shells();
void combine_infill();
void _generate_support_material();
bool is_printable() const { return ! m_copies.empty(); }
Print *m_print;
ModelObject *m_model_object;
PrintObjectConfig m_config;
// Slic3r::Point objects in scaled G-code coordinates
Points m_copies;
// scaled coordinates to add to copies (to compensate for the alignment
// operated when creating the object but still preserving a coherent API
// for external callers)
Point m_copies_shift;
LayerPtrs m_layers;
SupportLayerPtrs m_support_layers;
PrintState<PrintObjectStep, posCount> m_state;
// TODO: call model_object->get_bounding_box() instead of accepting
// parameter
PrintObject(Print* print, ModelObject* model_object, const BoundingBoxf3 &modobj_bbox);
~PrintObject() {}
void set_started(PrintObjectStep step);
void set_done(PrintObjectStep step);
std::vector<ExPolygons> _slice_region(size_t region_id, const std::vector<float> &z, bool modifier);
std::vector<ExPolygons> _slice_volumes(const std::vector<float> &z, const std::vector<const ModelVolume*> &volumes) const;
};
struct WipeTowerData
{
// Following section will be consumed by the GCodeGenerator.
// Tool ordering of a non-sequential print has to be known to calculate the wipe tower.
// Cache it here, so it does not need to be recalculated during the G-code generation.
ToolOrdering tool_ordering;
// Cache of tool changes per print layer.
std::unique_ptr<WipeTower::ToolChangeResult> priming;
std::vector<std::vector<WipeTower::ToolChangeResult>> tool_changes;
std::unique_ptr<WipeTower::ToolChangeResult> final_purge;
std::vector<float> used_filament;
int number_of_toolchanges;
// Depth of the wipe tower to pass to GLCanvas3D for exact bounding box:
float depth;
void clear() {
tool_ordering.clear();
priming.reset(nullptr);
tool_changes.clear();
final_purge.reset(nullptr);
used_filament.clear();
number_of_toolchanges = -1;
depth = 0.f;
}
};
struct PrintStatistics
{
PrintStatistics() { clear(); }
std::string estimated_normal_print_time;
std::string estimated_silent_print_time;
double total_used_filament;
double total_extruded_volume;
double total_cost;
double total_weight;
double total_wipe_tower_cost;
double total_wipe_tower_filament;
std::map<size_t, float> filament_stats;
void clear() {
estimated_normal_print_time.clear();
estimated_silent_print_time.clear();
total_used_filament = 0.;
total_extruded_volume = 0.;
total_cost = 0.;
total_weight = 0.;
total_wipe_tower_cost = 0.;
total_wipe_tower_filament = 0.;
filament_stats.clear();
}
};
typedef std::vector<PrintObject*> PrintObjectPtrs;
typedef std::vector<PrintRegion*> PrintRegionPtrs;
// The complete print tray with possibly multiple objects.
class Print
{
public:
Print() { restart(); }
~Print() { clear_objects(); }
// Methods, which change the state of Print / PrintObject / PrintRegion.
// The following methods are synchronized with process() and export_gcode(),
// so that process() and export_gcode() may be called from a background thread.
// In case the following methods need to modify data processed by process() or export_gcode(),
// a cancellation callback is executed to stop the background processing before the operation.
void clear_objects();
void delete_object(size_t idx);
void reload_object(size_t idx);
bool reload_model_instances();
void add_model_object(ModelObject* model_object, int idx = -1);
bool apply_config(DynamicPrintConfig config);
void process();
void export_gcode(const std::string &path_template, GCodePreviewData *preview_data);
// SLA export, temporary.
void export_png(const std::string &dirpath);
// methods for handling state
bool is_step_done(PrintStep step) const { return m_state.is_done(step); }
bool is_step_done(PrintObjectStep step) const;
bool has_infinite_skirt() const;
bool has_skirt() const;
PrintObjectPtrs get_printable_objects() const;
float get_wipe_tower_depth() const { return m_wipe_tower_data.depth; }
// Returns an empty string if valid, otherwise returns an error message.
std::string validate() const;
BoundingBox bounding_box() const;
BoundingBox total_bounding_box() const;
double skirt_first_layer_height() const;
Flow brim_flow() const;
Flow skirt_flow() const;
std::vector<unsigned int> object_extruders() const;
std::vector<unsigned int> support_material_extruders() const;
std::vector<unsigned int> extruders() const;
double max_allowed_layer_height() const;
bool has_support_material() const;
// Make sure the background processing has no access to this model_object during this call!
void auto_assign_extruders(ModelObject* model_object) const;
const PrintConfig& config() const { return m_config; }
const PrintObjectConfig& default_object_config() const { return m_default_object_config; }
const PrintRegionConfig& default_region_config() const { return m_default_region_config; }
const PrintObjectPtrs& objects() const { return m_objects; }
PrintObject* get_object(size_t idx) { return m_objects[idx]; }
const PrintObject* get_object(size_t idx) const { return m_objects[idx]; }
const PrintRegionPtrs& regions() const { return m_regions; }
const PlaceholderParser& placeholder_parser() const { return m_placeholder_parser; }
PlaceholderParser& placeholder_parser() { return m_placeholder_parser; }
// Returns extruder this eec should be printed with, according to PrintRegion config:
static int get_extruder(const ExtrusionEntityCollection& fill, const PrintRegion &region);
const ExtrusionEntityCollection& skirt() const { return m_skirt; }
const ExtrusionEntityCollection& brim() const { return m_brim; }
const PrintStatistics& print_statistics() const { return m_print_statistics; }
// Wipe tower support.
bool has_wipe_tower() const;
const WipeTowerData& wipe_tower_data() const { return m_wipe_tower_data; }
std::string output_filename() const;
std::string output_filepath(const std::string &path) const;
typedef std::function<void(int, const std::string&)> status_callback_type;
// Default status console print out in the form of percent => message.
void set_status_default() { m_status_callback = nullptr; }
// No status output or callback whatsoever, useful mostly for automatic tests.
void set_status_silent() { m_status_callback = [](int, const std::string&){}; }
// Register a custom status callback.
void set_status_callback(status_callback_type cb) { m_status_callback = cb; }
// Calls a registered callback to update the status, or print out the default message.
void set_status(int percent, const std::string &message) {
if (m_status_callback) m_status_callback(percent, message);
else printf("%d => %s\n", percent, message.c_str());
}
typedef std::function<void()> cancel_callback_type;
// Various methods will call this callback to stop the background processing (the Print::process() call)
// in case a successive change of the Print / PrintObject / PrintRegion instances changed
// the state of the finished or running calculations.
void set_cancel_callback(cancel_callback_type cancel_callback) { m_cancel_callback = cancel_callback; }
// Has the calculation been canceled?
bool canceled() const { return m_canceled; }
// Cancel the running computation. Stop execution of all the background threads.
void cancel() { m_canceled = true; }
// Cancel the running computation. Stop execution of all the background threads.
void restart() { m_canceled = false; }
// Accessed by SupportMaterial
const PrintRegion* get_region(size_t idx) const { return m_regions[idx]; }
protected:
void set_started(PrintStep step) { m_state.set_started(step, m_mutex); throw_if_canceled(); }
void set_done(PrintStep step) { m_state.set_done(step, m_mutex); throw_if_canceled(); }
bool invalidate_step(PrintStep step);
bool invalidate_all_steps() { return m_state.invalidate_all(m_mutex, m_cancel_callback); }
// methods for handling regions
PrintRegion* get_region(size_t idx) { return m_regions[idx]; }
PrintRegion* add_region();
PrintRegion* add_region(const PrintRegionConfig &config);
private:
bool invalidate_state_by_config_options(const std::vector<t_config_option_key> &opt_keys);
PrintRegionConfig _region_config_from_model_volume(const ModelVolume &volume);
// If the background processing stop was requested, throw CanceledException.
// To be called by the worker thread and its sub-threads (mostly launched on the TBB thread pool) regularly.
void throw_if_canceled() const { if (m_canceled) throw CanceledException(); }
void _make_skirt();
void _make_brim();
void _make_wipe_tower();
void _simplify_slices(double distance);
PrintState<PrintStep, psCount> m_state;
// Mutex used for synchronization of the worker thread with the UI thread:
// The mutex will be used to guard the worker thread against entering a stage
// while the data influencing the stage is modified.
mutable tbb::mutex m_mutex;
// Has the calculation been canceled?
tbb::atomic<bool> m_canceled;
// Callback to be evoked regularly to update state of the UI thread.
status_callback_type m_status_callback;
// Callback to be evoked to stop the background processing before a state is updated.
cancel_callback_type m_cancel_callback = [](){};
PrintConfig m_config;
PrintObjectConfig m_default_object_config;
PrintRegionConfig m_default_region_config;
PrintObjectPtrs m_objects;
PrintRegionPtrs m_regions;
PlaceholderParser m_placeholder_parser;
// Ordered collections of extrusion paths to build skirt loops and brim.
ExtrusionEntityCollection m_skirt;
ExtrusionEntityCollection m_brim;
// Following section will be consumed by the GCodeGenerator.
WipeTowerData m_wipe_tower_data;
// Estimated print time, filament consumed.
PrintStatistics m_print_statistics;
// To allow GCode to set the Print's GCodeExport step status.
friend class GCode;
// Allow PrintObject to access m_mutex and m_cancel_callback.
friend class PrintObject;
};
#define FOREACH_BASE(type, container, iterator) for (type::const_iterator iterator = (container).begin(); iterator != (container).end(); ++iterator)
#define FOREACH_OBJECT(print, object) FOREACH_BASE(PrintObjectPtrs, (print)->m_objects, object)
#define FOREACH_LAYER(object, layer) FOREACH_BASE(LayerPtrs, (object)->m_layers, layer)
#define FOREACH_LAYERREGION(layer, layerm) FOREACH_BASE(LayerRegionPtrs, (layer)->m_regions, layerm)
}
#endif