This repository has been archived on 2023-10-09. You can view files and clone it, but cannot push or open issues or pull requests.
Files
blender-archive/source/blender/blenkernel/intern/lib_remap.c

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

711 lines
25 KiB
C
Raw Normal View History

ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
/*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software Foundation,
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*/
/** \file
* \ingroup bke
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
*
* Contains management of ID's for remapping.
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
*/
#include "CLG_log.h"
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
#include "BLI_utildefines.h"
#include "DNA_collection_types.h"
#include "DNA_object_types.h"
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
#include "BKE_armature.h"
Render Layers and Collections (merge from render-layers) Design Documents ---------------- * https://wiki.blender.org/index.php/Dev:2.8/Source/Layers * https://wiki.blender.org/index.php/Dev:2.8/Source/DataDesignRevised User Commit Log --------------- * New Layer and Collection system to replace render layers and viewport layers. * A layer is a set of collections of objects (and their drawing options) required for specific tasks. * A collection is a set of objects, equivalent of the old layers in Blender. A collection can be shared across multiple layers. * All Scenes have a master collection that all other collections are children of. * New collection "context" tab (in Properties Editor) * New temporary viewport "collections" panel to control per-collection visibility Missing User Features --------------------- * Collection "Filter" Option to add objects based on their names * Collection Manager operators The existing buttons are placeholders * Collection Manager drawing The editor main region is empty * Collection Override * Per-Collection engine settings This will come as a separate commit, as part of the clay-engine branch Dev Commit Log -------------- * New DNA file (DNA_layer_types.h) with the new structs We are replacing Base by a new extended Base while keeping it backward compatible with some legacy settings (i.e., lay, flag_legacy). Renamed all Base to BaseLegacy to make it clear the areas of code that still need to be converted Note: manual changes were required on - deg_builder_nodes.h, rna_object.c, KX_Light.cpp * Unittesting for main syncronization requirements - read, write, add/copy/remove objects, copy scene, collection link/unlinking, context) * New Editor: Collection Manager Based on patch by Julian Eisel This is extracted from the layer-manager branch. With the following changes: - Renamed references of layer manager to collections manager - I doesn't include the editors/space_collections/ draw and util files - The drawing code itself will be implemented separately by Julian * Base / Object: A little note about them. Original Blender code would try to keep them in sync through the code, juggling flags back and forth. This will now be handled by Depsgraph, keeping Object and Bases more separated throughout the non-rendering code. Scene.base is being cleared in doversion, and the old viewport drawing code was poorly converted to use the new bases while the new viewport code doesn't get merged and replace the old one. Python API Changes ------------------ ``` - scene.layers + # no longer exists - scene.objects + scene.scene_layers.active.objects - scene.objects.active + scene.render_layers.active.objects.active - bpy.context.scene.objects.link() + bpy.context.scene_collection.objects.link() - bpy_extras.object_utils.object_data_add(context, obdata, operator=None, use_active_layer=True, name=None) + bpy_extras.object_utils.object_data_add(context, obdata, operator=None, name=None) - bpy.context.object.select + bpy.context.object.select = True + bpy.context.object.select = False + bpy.context.object.select_get() + bpy.context.object.select_set(action='SELECT') + bpy.context.object.select_set(action='DESELECT') -AddObjectHelper.layers + # no longer exists ```
2017-02-07 10:18:38 +01:00
#include "BKE_collection.h"
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
#include "BKE_curve.h"
#include "BKE_layer.h"
#include "BKE_lib_id.h"
#include "BKE_lib_query.h"
#include "BKE_lib_remap.h"
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
#include "BKE_main.h"
#include "BKE_material.h"
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
#include "BKE_mball.h"
#include "BKE_modifier.h"
#include "BKE_multires.h"
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
#include "BKE_node.h"
#include "BKE_object.h"
#include "DEG_depsgraph.h"
#include "DEG_depsgraph_build.h"
2020-02-11 11:51:12 +11:00
#include "lib_intern.h" /* own include */
static CLG_LogRef LOG = {.identifier = "bke.lib_remap"};
BKE_library_free_notifier_reference_cb free_notifier_reference_cb = NULL;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
void BKE_library_callback_free_notifier_reference_set(BKE_library_free_notifier_reference_cb func)
{
free_notifier_reference_cb = func;
}
BKE_library_remap_editor_id_reference_cb remap_editor_id_reference_cb = NULL;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
void BKE_library_callback_remap_editor_id_reference_set(
BKE_library_remap_editor_id_reference_cb func)
{
remap_editor_id_reference_cb = func;
}
typedef struct IDRemap {
Main *bmain; /* Only used to trigger depsgraph updates in the right bmain. */
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
ID *old_id;
ID *new_id;
/** The ID in which we are replacing old_id by new_id usages. */
ID *id_owner;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
short flag;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
/* 'Output' data. */
short status;
2020-09-24 14:18:04 +10:00
/** Number of direct use cases that could not be remapped (e.g.: obdata when in edit mode). */
int skipped_direct;
2020-09-24 14:18:04 +10:00
/** Number of indirect use cases that could not be remapped. */
int skipped_indirect;
2020-09-24 14:18:04 +10:00
/** Number of skipped use cases that refcount the data-block. */
int skipped_refcounted;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
} IDRemap;
/* IDRemap->flag enums defined in BKE_lib.h */
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
/* IDRemap->status */
enum {
/* *** Set by callback. *** */
ID_REMAP_IS_LINKED_DIRECT = 1 << 0, /* new_id is directly linked in current .blend. */
ID_REMAP_IS_USER_ONE_SKIPPED = 1 << 1, /* There was some skipped 'user_one' usages of old_id. */
};
static int foreach_libblock_remap_callback(LibraryIDLinkCallbackData *cb_data)
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
{
const int cb_flag = cb_data->cb_flag;
if (cb_flag & IDWALK_CB_EMBEDDED) {
return IDWALK_RET_NOP;
}
ID *id_owner = cb_data->id_owner;
ID *id_self = cb_data->id_self;
ID **id_p = cb_data->id_pointer;
IDRemap *id_remap_data = cb_data->user_data;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
ID *old_id = id_remap_data->old_id;
ID *new_id = id_remap_data->new_id;
2020-03-11 17:29:20 +01:00
/* Those asserts ensure the general sanity of ID tags regarding 'embedded' ID data (root
* nodetrees and co). */
BLI_assert(id_owner == id_remap_data->id_owner);
BLI_assert(id_self == id_owner || (id_self->flag & LIB_EMBEDDED_DATA) != 0);
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
if (!old_id) { /* Used to cleanup all IDs used by a specific one. */
BLI_assert(!new_id);
old_id = *id_p;
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
if (*id_p && (*id_p == old_id)) {
/* Better remap to NULL than not remapping at all,
* then we can handle it as a regular remap-to-NULL case. */
if ((cb_flag & IDWALK_CB_NEVER_SELF) && (new_id == id_self)) {
new_id = NULL;
}
const bool is_reference = (cb_flag & IDWALK_CB_OVERRIDE_LIBRARY_REFERENCE) != 0;
const bool is_indirect = (cb_flag & IDWALK_CB_INDIRECT_USAGE) != 0;
const bool skip_indirect = (id_remap_data->flag & ID_REMAP_SKIP_INDIRECT_USAGE) != 0;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
/* Note: proxy usage implies LIB_TAG_EXTERN, so on this aspect it is direct,
* on the other hand since they get reset to lib data on file open/reload it is indirect too.
* Edit Mode is also a 'skip direct' case. */
const bool is_obj = (GS(id_owner->name) == ID_OB);
const bool is_obj_proxy = (is_obj &&
(((Object *)id_owner)->proxy || ((Object *)id_owner)->proxy_group));
const bool is_obj_editmode = (is_obj && BKE_object_is_in_editmode((Object *)id_owner));
const bool is_never_null = ((cb_flag & IDWALK_CB_NEVER_NULL) && (new_id == NULL) &&
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
(id_remap_data->flag & ID_REMAP_FORCE_NEVER_NULL_USAGE) == 0);
const bool skip_reference = (id_remap_data->flag & ID_REMAP_SKIP_OVERRIDE_LIBRARY) != 0;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
const bool skip_never_null = (id_remap_data->flag & ID_REMAP_SKIP_NEVER_NULL_USAGE) != 0;
const bool force_user_refcount = (id_remap_data->flag & ID_REMAP_FORCE_USER_REFCOUNT) != 0;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
"Fix" crash when deleting linked object which has indirect usages. This is in fact very hairy situation here... Objects are only refcounted by scenes, any other usage is 'free', which means once all object instanciations are gone Blender considers it can delete it. There is a trap here though: indirect usages. Typically, we should never modify linked data (because it is essencially useless, changes would be ignored and ost on next reload or even undo/redo). This means indirect usages are not affected by default 'safe' remapping/unlinking. For unlinking preceeding deletion however, this is not acceptable - we are likely to end with a zero-user ID (aka deletable one) which is still actually used by other linked data. Solution choosen here is double: I) From 'user-space' (i.e. outliner, operators...), we check for cases where deleting datablocks should not be allowed (indirect data or indirectly used data), and abort (with report) if needed. II) From 'lower' level (BKE_library_remap and RNA), we also unlink from linked data, which makes actual deletion possible and safe. Note that with previous behavior (2.77 one), linked object would be deleted, including from linked data - but then, once file is saved and reloaded, indirect usage would link back the deleted object, without any instanciation in scene, which made it somehow virtual and unreachable... With new behavior, this is no more possible, but on the other hand it means that in situations of dependency cycles (two linked objects using each other), linked objects become impossible to delete (from user space). Not sure what's best here, behavior with those corner cases of library linking is very poorly defined... :(
2016-07-01 17:51:08 +02:00
#ifdef DEBUG_PRINT
printf(
"In %s (lib %p): Remapping %s (%p) to %s (%p) "
"(is_indirect: %d, skip_indirect: %d, is_reference: %d, skip_reference: %d)\n",
id->name,
id->lib,
old_id->name,
old_id,
new_id ? new_id->name : "<NONE>",
new_id,
is_indirect,
skip_indirect,
is_reference,
skip_reference);
"Fix" crash when deleting linked object which has indirect usages. This is in fact very hairy situation here... Objects are only refcounted by scenes, any other usage is 'free', which means once all object instanciations are gone Blender considers it can delete it. There is a trap here though: indirect usages. Typically, we should never modify linked data (because it is essencially useless, changes would be ignored and ost on next reload or even undo/redo). This means indirect usages are not affected by default 'safe' remapping/unlinking. For unlinking preceeding deletion however, this is not acceptable - we are likely to end with a zero-user ID (aka deletable one) which is still actually used by other linked data. Solution choosen here is double: I) From 'user-space' (i.e. outliner, operators...), we check for cases where deleting datablocks should not be allowed (indirect data or indirectly used data), and abort (with report) if needed. II) From 'lower' level (BKE_library_remap and RNA), we also unlink from linked data, which makes actual deletion possible and safe. Note that with previous behavior (2.77 one), linked object would be deleted, including from linked data - but then, once file is saved and reloaded, indirect usage would link back the deleted object, without any instanciation in scene, which made it somehow virtual and unreachable... With new behavior, this is no more possible, but on the other hand it means that in situations of dependency cycles (two linked objects using each other), linked objects become impossible to delete (from user space). Not sure what's best here, behavior with those corner cases of library linking is very poorly defined... :(
2016-07-01 17:51:08 +02:00
#endif
if ((id_remap_data->flag & ID_REMAP_FLAG_NEVER_NULL_USAGE) &&
(cb_flag & IDWALK_CB_NEVER_NULL)) {
id_owner->tag |= LIB_TAG_DOIT;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
/* Special hack in case it's Object->data and we are in edit mode, and new_id is not NULL
* (otherwise, we follow common NEVER_NULL flags).
* (skipped_indirect too). */
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
if ((is_never_null && skip_never_null) ||
(is_obj_editmode && (((Object *)id_owner)->data == *id_p) && new_id != NULL) ||
(skip_indirect && is_indirect) || (is_reference && skip_reference)) {
"Fix" crash when deleting linked object which has indirect usages. This is in fact very hairy situation here... Objects are only refcounted by scenes, any other usage is 'free', which means once all object instanciations are gone Blender considers it can delete it. There is a trap here though: indirect usages. Typically, we should never modify linked data (because it is essencially useless, changes would be ignored and ost on next reload or even undo/redo). This means indirect usages are not affected by default 'safe' remapping/unlinking. For unlinking preceeding deletion however, this is not acceptable - we are likely to end with a zero-user ID (aka deletable one) which is still actually used by other linked data. Solution choosen here is double: I) From 'user-space' (i.e. outliner, operators...), we check for cases where deleting datablocks should not be allowed (indirect data or indirectly used data), and abort (with report) if needed. II) From 'lower' level (BKE_library_remap and RNA), we also unlink from linked data, which makes actual deletion possible and safe. Note that with previous behavior (2.77 one), linked object would be deleted, including from linked data - but then, once file is saved and reloaded, indirect usage would link back the deleted object, without any instanciation in scene, which made it somehow virtual and unreachable... With new behavior, this is no more possible, but on the other hand it means that in situations of dependency cycles (two linked objects using each other), linked objects become impossible to delete (from user space). Not sure what's best here, behavior with those corner cases of library linking is very poorly defined... :(
2016-07-01 17:51:08 +02:00
if (is_indirect) {
id_remap_data->skipped_indirect++;
if (is_obj) {
Object *ob = (Object *)id_owner;
if (ob->data == *id_p && ob->proxy != NULL) {
/* And another 'Proudly brought to you by Proxy Hell' hack!
* This will allow us to avoid clearing 'LIB_EXTERN' flag of obdata of proxies... */
id_remap_data->skipped_direct++;
}
}
}
else if (is_never_null || is_obj_editmode || is_reference) {
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
id_remap_data->skipped_direct++;
}
else {
"Fix" crash when deleting linked object which has indirect usages. This is in fact very hairy situation here... Objects are only refcounted by scenes, any other usage is 'free', which means once all object instanciations are gone Blender considers it can delete it. There is a trap here though: indirect usages. Typically, we should never modify linked data (because it is essencially useless, changes would be ignored and ost on next reload or even undo/redo). This means indirect usages are not affected by default 'safe' remapping/unlinking. For unlinking preceeding deletion however, this is not acceptable - we are likely to end with a zero-user ID (aka deletable one) which is still actually used by other linked data. Solution choosen here is double: I) From 'user-space' (i.e. outliner, operators...), we check for cases where deleting datablocks should not be allowed (indirect data or indirectly used data), and abort (with report) if needed. II) From 'lower' level (BKE_library_remap and RNA), we also unlink from linked data, which makes actual deletion possible and safe. Note that with previous behavior (2.77 one), linked object would be deleted, including from linked data - but then, once file is saved and reloaded, indirect usage would link back the deleted object, without any instanciation in scene, which made it somehow virtual and unreachable... With new behavior, this is no more possible, but on the other hand it means that in situations of dependency cycles (two linked objects using each other), linked objects become impossible to delete (from user space). Not sure what's best here, behavior with those corner cases of library linking is very poorly defined... :(
2016-07-01 17:51:08 +02:00
BLI_assert(0);
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
if (cb_flag & IDWALK_CB_USER) {
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
id_remap_data->skipped_refcounted++;
}
else if (cb_flag & IDWALK_CB_USER_ONE) {
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
/* No need to count number of times this happens, just a flag is enough. */
id_remap_data->status |= ID_REMAP_IS_USER_ONE_SKIPPED;
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
else {
if (!is_never_null) {
*id_p = new_id;
DEG_id_tag_update_ex(id_remap_data->bmain,
id_self,
ID_RECALC_COPY_ON_WRITE | ID_RECALC_TRANSFORM | ID_RECALC_GEOMETRY);
if (id_self != id_owner) {
DEG_id_tag_update_ex(id_remap_data->bmain,
id_owner,
ID_RECALC_COPY_ON_WRITE | ID_RECALC_TRANSFORM | ID_RECALC_GEOMETRY);
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
if (cb_flag & IDWALK_CB_USER) {
/* NOTE: by default we don't user-count IDs which are not in the main database.
2019-06-12 09:04:10 +10:00
* This is because in certain conditions we can have data-blocks in
* the main which are referencing data-blocks outside of it.
* For example, BKE_mesh_new_from_object() called on an evaluated
* object will cause such situation.
*/
if (force_user_refcount || (old_id->tag & LIB_TAG_NO_MAIN) == 0) {
id_us_min(old_id);
}
if (new_id != NULL && (force_user_refcount || (new_id->tag & LIB_TAG_NO_MAIN) == 0)) {
/* We do not want to handle LIB_TAG_INDIRECT/LIB_TAG_EXTERN here. */
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
new_id->us++;
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
else if (cb_flag & IDWALK_CB_USER_ONE) {
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
id_us_ensure_real(new_id);
/* We cannot affect old_id->us directly, LIB_TAG_EXTRAUSER(_SET)
* are assumed to be set as needed, that extra user is processed in final handling. */
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
if (!is_indirect || is_obj_proxy) {
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
id_remap_data->status |= ID_REMAP_IS_LINKED_DIRECT;
}
/* We need to remap proxy_from pointer of remapped proxy... sigh. */
if (is_obj_proxy && new_id != NULL) {
Object *ob = (Object *)id_owner;
if (ob->proxy == (Object *)new_id) {
ob->proxy->proxy_from = ob;
}
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
return IDWALK_RET_NOP;
}
static void libblock_remap_data_preprocess(IDRemap *r_id_remap_data)
{
switch (GS(r_id_remap_data->id_owner->name)) {
case ID_OB: {
ID *old_id = r_id_remap_data->old_id;
if (!old_id || GS(old_id->name) == ID_AR) {
Object *ob = (Object *)r_id_remap_data->id_owner;
/* Object's pose holds reference to armature bones. sic */
/* Note that in theory, we should have to bother about linked/non-linked/never-null/etc.
* flags/states.
* Fortunately, this is just a tag, so we can accept to 'over-tag' a bit for pose recalc,
* and avoid another complex and risky condition nightmare like the one we have in
* foreach_libblock_remap_callback(). */
if (ob->pose && (!old_id || ob->data == old_id)) {
BLI_assert(ob->type == OB_ARMATURE);
ob->pose->flag |= POSE_RECALC;
/* We need to clear pose bone pointers immediately, some code may access those before
* pose is actually recomputed, which can lead to segfault. */
BKE_pose_clear_pointers(ob->pose);
}
}
break;
}
default:
break;
}
}
/**
* Can be called with both old_ob and new_ob being NULL,
* this means we have to check whole Main database then.
*/
static void libblock_remap_data_postprocess_object_update(Main *bmain,
Object *old_ob,
Object *new_ob)
{
Collections and groups unification OVERVIEW * In 2.7 terminology, all layers and groups are now collection datablocks. * These collections are nestable, linkable, instanceable, overrideable, .. which opens up new ways to set up scenes and link + override data. * Viewport/render visibility and selectability are now a part of the collection and shared across all view layers and linkable. * View layers define which subset of the scene collection hierarchy is excluded for each. For many workflows one view layer can be used, these are more of an advanced feature now. OUTLINER * The outliner now has a "View Layer" display mode instead of "Collections", which can display the collections and/or objects in the view layer. * In this display mode, collections can be excluded with the right click menu. These will then be greyed out and their objects will be excluded. * To view collections not linked to any scene, the "Blender File" display mode can be used, with the new filtering option to just see Colleciton datablocks. * The outliner right click menus for collections and objects were reorganized. * Drag and drop still needs to be improved. Like before, dragging the icon or text gives different results, we'll unify this later. LINKING AND OVERRIDES * Collections can now be linked into the scene without creating an instance, with the link/append operator or from the collections view in the outliner. * Collections can get static overrides with the right click menu in the outliner, but this is rather unreliable and not clearly communicated at the moment. * We still need to improve the make override operator to turn collection instances into collections with overrides directly in the scene. PERFORMANCE * We tried to make performance not worse than before and improve it in some cases. The main thing that's still a bit slower is multiple scenes, we have to change the layer syncing to only updated affected scenes. * Collections keep a list of their parent collections for faster incremental updates in syncing and caching. * View layer bases are now in a object -> base hash to avoid quadratic time lookups internally and in API functions like visible_get(). VERSIONING * Compatibility with 2.7 files should be improved due to the new visibility controls. Of course users may not want to set up their scenes differently now to avoid having separate layers and groups. * Compatibility with 2.8 is mostly there, and was tested on Eevee demo and Hero files. There's a few things which are know to be not quite compatible, like nested layer collections inside groups. * The versioning code for 2.8 files is quite complicated, and isolated behind #ifdef so it can be removed at the end of the release cycle. KNOWN ISSUES * The G-key group operators in the 3D viewport were left mostly as is, they need to be modified still to fit better. * Same for the groups panel in the object properties. This needs to be updated still, or perhaps replaced by something better. * Collections must all have a unique name. Less restrictive namespacing is to be done later, we'll have to see how important this is as all objects within the collections must also have a unique name anyway. * Full scene copy and delete scene are exactly doing the right thing yet. Differential Revision: https://developer.blender.org/D3383 https://code.blender.org/2018/05/collections-and-groups/
2018-04-30 15:57:22 +02:00
if (new_ob == NULL) {
/* In case we unlinked old_ob (new_ob is NULL), the object has already
* been removed from the scenes and their collections. We still have
* to remove the NULL children from collections not used in any scene. */
Collections and groups unification OVERVIEW * In 2.7 terminology, all layers and groups are now collection datablocks. * These collections are nestable, linkable, instanceable, overrideable, .. which opens up new ways to set up scenes and link + override data. * Viewport/render visibility and selectability are now a part of the collection and shared across all view layers and linkable. * View layers define which subset of the scene collection hierarchy is excluded for each. For many workflows one view layer can be used, these are more of an advanced feature now. OUTLINER * The outliner now has a "View Layer" display mode instead of "Collections", which can display the collections and/or objects in the view layer. * In this display mode, collections can be excluded with the right click menu. These will then be greyed out and their objects will be excluded. * To view collections not linked to any scene, the "Blender File" display mode can be used, with the new filtering option to just see Colleciton datablocks. * The outliner right click menus for collections and objects were reorganized. * Drag and drop still needs to be improved. Like before, dragging the icon or text gives different results, we'll unify this later. LINKING AND OVERRIDES * Collections can now be linked into the scene without creating an instance, with the link/append operator or from the collections view in the outliner. * Collections can get static overrides with the right click menu in the outliner, but this is rather unreliable and not clearly communicated at the moment. * We still need to improve the make override operator to turn collection instances into collections with overrides directly in the scene. PERFORMANCE * We tried to make performance not worse than before and improve it in some cases. The main thing that's still a bit slower is multiple scenes, we have to change the layer syncing to only updated affected scenes. * Collections keep a list of their parent collections for faster incremental updates in syncing and caching. * View layer bases are now in a object -> base hash to avoid quadratic time lookups internally and in API functions like visible_get(). VERSIONING * Compatibility with 2.7 files should be improved due to the new visibility controls. Of course users may not want to set up their scenes differently now to avoid having separate layers and groups. * Compatibility with 2.8 is mostly there, and was tested on Eevee demo and Hero files. There's a few things which are know to be not quite compatible, like nested layer collections inside groups. * The versioning code for 2.8 files is quite complicated, and isolated behind #ifdef so it can be removed at the end of the release cycle. KNOWN ISSUES * The G-key group operators in the 3D viewport were left mostly as is, they need to be modified still to fit better. * Same for the groups panel in the object properties. This needs to be updated still, or perhaps replaced by something better. * Collections must all have a unique name. Less restrictive namespacing is to be done later, we'll have to see how important this is as all objects within the collections must also have a unique name anyway. * Full scene copy and delete scene are exactly doing the right thing yet. Differential Revision: https://developer.blender.org/D3383 https://code.blender.org/2018/05/collections-and-groups/
2018-04-30 15:57:22 +02:00
BKE_collections_object_remove_nulls(bmain);
}
BKE_main_collection_sync_remap(bmain);
if (old_ob == NULL) {
for (Object *ob = bmain->objects.first; ob != NULL; ob = ob->id.next) {
if (ob->type == OB_MBALL && BKE_mball_is_basis(ob)) {
DEG_id_tag_update(&ob->id, ID_RECALC_GEOMETRY);
}
}
}
else {
for (Object *ob = bmain->objects.first; ob != NULL; ob = ob->id.next) {
if (ob->type == OB_MBALL && BKE_mball_is_basis_for(ob, old_ob)) {
DEG_id_tag_update(&ob->id, ID_RECALC_GEOMETRY);
break; /* There is only one basis... */
}
}
}
}
/* Can be called with both old_collection and new_collection being NULL,
* this means we have to check whole Main database then. */
static void libblock_remap_data_postprocess_collection_update(Main *bmain,
Collection *owner_collection,
Collection *UNUSED(old_collection),
Collection *new_collection)
{
Collections and groups unification OVERVIEW * In 2.7 terminology, all layers and groups are now collection datablocks. * These collections are nestable, linkable, instanceable, overrideable, .. which opens up new ways to set up scenes and link + override data. * Viewport/render visibility and selectability are now a part of the collection and shared across all view layers and linkable. * View layers define which subset of the scene collection hierarchy is excluded for each. For many workflows one view layer can be used, these are more of an advanced feature now. OUTLINER * The outliner now has a "View Layer" display mode instead of "Collections", which can display the collections and/or objects in the view layer. * In this display mode, collections can be excluded with the right click menu. These will then be greyed out and their objects will be excluded. * To view collections not linked to any scene, the "Blender File" display mode can be used, with the new filtering option to just see Colleciton datablocks. * The outliner right click menus for collections and objects were reorganized. * Drag and drop still needs to be improved. Like before, dragging the icon or text gives different results, we'll unify this later. LINKING AND OVERRIDES * Collections can now be linked into the scene without creating an instance, with the link/append operator or from the collections view in the outliner. * Collections can get static overrides with the right click menu in the outliner, but this is rather unreliable and not clearly communicated at the moment. * We still need to improve the make override operator to turn collection instances into collections with overrides directly in the scene. PERFORMANCE * We tried to make performance not worse than before and improve it in some cases. The main thing that's still a bit slower is multiple scenes, we have to change the layer syncing to only updated affected scenes. * Collections keep a list of their parent collections for faster incremental updates in syncing and caching. * View layer bases are now in a object -> base hash to avoid quadratic time lookups internally and in API functions like visible_get(). VERSIONING * Compatibility with 2.7 files should be improved due to the new visibility controls. Of course users may not want to set up their scenes differently now to avoid having separate layers and groups. * Compatibility with 2.8 is mostly there, and was tested on Eevee demo and Hero files. There's a few things which are know to be not quite compatible, like nested layer collections inside groups. * The versioning code for 2.8 files is quite complicated, and isolated behind #ifdef so it can be removed at the end of the release cycle. KNOWN ISSUES * The G-key group operators in the 3D viewport were left mostly as is, they need to be modified still to fit better. * Same for the groups panel in the object properties. This needs to be updated still, or perhaps replaced by something better. * Collections must all have a unique name. Less restrictive namespacing is to be done later, we'll have to see how important this is as all objects within the collections must also have a unique name anyway. * Full scene copy and delete scene are exactly doing the right thing yet. Differential Revision: https://developer.blender.org/D3383 https://code.blender.org/2018/05/collections-and-groups/
2018-04-30 15:57:22 +02:00
if (new_collection == NULL) {
/* XXX Complex cases can lead to NULL pointers in other collections than old_collection,
* and BKE_main_collection_sync_remap() does not tolerate any of those, so for now always check
* whole existing collections for NULL pointers.
* I'd consider optimizing that whole collection remapping process a TODO for later. */
BKE_collections_child_remove_nulls(bmain, owner_collection, NULL /*old_collection*/);
Collections and groups unification OVERVIEW * In 2.7 terminology, all layers and groups are now collection datablocks. * These collections are nestable, linkable, instanceable, overrideable, .. which opens up new ways to set up scenes and link + override data. * Viewport/render visibility and selectability are now a part of the collection and shared across all view layers and linkable. * View layers define which subset of the scene collection hierarchy is excluded for each. For many workflows one view layer can be used, these are more of an advanced feature now. OUTLINER * The outliner now has a "View Layer" display mode instead of "Collections", which can display the collections and/or objects in the view layer. * In this display mode, collections can be excluded with the right click menu. These will then be greyed out and their objects will be excluded. * To view collections not linked to any scene, the "Blender File" display mode can be used, with the new filtering option to just see Colleciton datablocks. * The outliner right click menus for collections and objects were reorganized. * Drag and drop still needs to be improved. Like before, dragging the icon or text gives different results, we'll unify this later. LINKING AND OVERRIDES * Collections can now be linked into the scene without creating an instance, with the link/append operator or from the collections view in the outliner. * Collections can get static overrides with the right click menu in the outliner, but this is rather unreliable and not clearly communicated at the moment. * We still need to improve the make override operator to turn collection instances into collections with overrides directly in the scene. PERFORMANCE * We tried to make performance not worse than before and improve it in some cases. The main thing that's still a bit slower is multiple scenes, we have to change the layer syncing to only updated affected scenes. * Collections keep a list of their parent collections for faster incremental updates in syncing and caching. * View layer bases are now in a object -> base hash to avoid quadratic time lookups internally and in API functions like visible_get(). VERSIONING * Compatibility with 2.7 files should be improved due to the new visibility controls. Of course users may not want to set up their scenes differently now to avoid having separate layers and groups. * Compatibility with 2.8 is mostly there, and was tested on Eevee demo and Hero files. There's a few things which are know to be not quite compatible, like nested layer collections inside groups. * The versioning code for 2.8 files is quite complicated, and isolated behind #ifdef so it can be removed at the end of the release cycle. KNOWN ISSUES * The G-key group operators in the 3D viewport were left mostly as is, they need to be modified still to fit better. * Same for the groups panel in the object properties. This needs to be updated still, or perhaps replaced by something better. * Collections must all have a unique name. Less restrictive namespacing is to be done later, we'll have to see how important this is as all objects within the collections must also have a unique name anyway. * Full scene copy and delete scene are exactly doing the right thing yet. Differential Revision: https://developer.blender.org/D3383 https://code.blender.org/2018/05/collections-and-groups/
2018-04-30 15:57:22 +02:00
}
else {
/* Temp safe fix, but a "tad" brute force... We should probably be able to use parents from
* old_collection instead? */
BKE_main_collections_parent_relations_rebuild(bmain);
}
BKE_main_collection_sync_remap(bmain);
}
static void libblock_remap_data_postprocess_obdata_relink(Main *bmain, Object *ob, ID *new_id)
{
if (ob->data == new_id) {
switch (GS(new_id->name)) {
case ID_ME:
multires_force_sculpt_rebuild(ob);
break;
case ID_CU:
BKE_curve_type_test(ob);
break;
default:
break;
}
BKE_modifiers_test_object(ob);
BKE_object_materials_test(bmain, ob, new_id);
}
}
static void libblock_remap_data_postprocess_nodetree_update(Main *bmain, ID *new_id)
{
/* Update all group nodes using a node group. */
ntreeUpdateAllUsers(bmain, new_id);
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
/**
2019-06-12 09:04:10 +10:00
* Execute the 'data' part of the remapping (that is, all ID pointers from other ID data-blocks).
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
*
* Behavior differs depending on whether given \a id is NULL or not:
* - \a id NULL: \a old_id must be non-NULL, \a new_id may be NULL (unlinking \a old_id) or not
* (remapping \a old_id to \a new_id).
* The whole \a bmain database is checked, and all pointers to \a old_id
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
* are remapped to \a new_id.
* - \a id is non-NULL:
* + If \a old_id is NULL, \a new_id must also be NULL,
* and all ID pointers from \a id are cleared
2019-06-12 09:04:10 +10:00
* (i.e. \a id does not references any other data-block anymore).
* + If \a old_id is non-NULL, behavior is as with a NULL \a id, but only within given \a id.
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
*
* \param bmain: the Main data storage to operate on (must never be NULL).
2019-06-12 09:04:10 +10:00
* \param id: the data-block to operate on
* (can be NULL, in which case we operate over all IDs from given bmain).
2019-06-12 09:04:10 +10:00
* \param old_id: the data-block to dereference (may be NULL if \a id is non-NULL).
* \param new_id: the new data-block to replace \a old_id references with (may be NULL).
* \param r_id_remap_data: if non-NULL, the IDRemap struct to use
* (uselful to retrieve info about remapping process).
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
*/
ATTR_NONNULL(1)
static void libblock_remap_data(
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
Main *bmain, ID *id, ID *old_id, ID *new_id, const short remap_flags, IDRemap *r_id_remap_data)
{
IDRemap id_remap_data;
const int foreach_id_flags = ((remap_flags & ID_REMAP_NO_INDIRECT_PROXY_DATA_USAGE) != 0 ?
IDWALK_NO_INDIRECT_PROXY_DATA_USAGE :
IDWALK_NOP) |
((remap_flags & ID_REMAP_FORCE_INTERNAL_RUNTIME_POINTERS) != 0 ?
IDWALK_DO_INTERNAL_RUNTIME_POINTERS :
IDWALK_NOP);
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
if (r_id_remap_data == NULL) {
r_id_remap_data = &id_remap_data;
}
r_id_remap_data->bmain = bmain;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
r_id_remap_data->old_id = old_id;
r_id_remap_data->new_id = new_id;
r_id_remap_data->id_owner = NULL;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
r_id_remap_data->flag = remap_flags;
r_id_remap_data->status = 0;
r_id_remap_data->skipped_direct = 0;
r_id_remap_data->skipped_indirect = 0;
r_id_remap_data->skipped_refcounted = 0;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
if (id) {
#ifdef DEBUG_PRINT
printf("\tchecking id %s (%p, %p)\n", id->name, id, id->lib);
#endif
r_id_remap_data->id_owner = id;
libblock_remap_data_preprocess(r_id_remap_data);
BKE_library_foreach_ID_link(
NULL, id, foreach_libblock_remap_callback, (void *)r_id_remap_data, foreach_id_flags);
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
else {
/* Note that this is a very 'brute force' approach,
* maybe we could use some depsgraph to only process objects actually using given old_id...
* sounds rather unlikely currently, though, so this will do for now. */
ID *id_curr;
FOREACH_MAIN_ID_BEGIN (bmain, id_curr) {
if (BKE_library_id_can_use_idtype(id_curr, GS(old_id->name))) {
/* Note that we cannot skip indirect usages of old_id here (if requested),
* we still need to check it for the user count handling...
* XXX No more true (except for debug usage of those skipping counters). */
r_id_remap_data->id_owner = id_curr;
libblock_remap_data_preprocess(r_id_remap_data);
BKE_library_foreach_ID_link(NULL,
id_curr,
foreach_libblock_remap_callback,
(void *)r_id_remap_data,
foreach_id_flags);
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
}
FOREACH_MAIN_ID_END;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
if ((remap_flags & ID_REMAP_SKIP_USER_CLEAR) == 0) {
/* XXX We may not want to always 'transfer' fake-user from old to new id...
* Think for now it's desired behavior though,
* we can always add an option (flag) to control this later if needed. */
if (old_id && (old_id->flag & LIB_FAKEUSER)) {
id_fake_user_clear(old_id);
id_fake_user_set(new_id);
}
id_us_clear_real(old_id);
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
if (new_id && (new_id->tag & LIB_TAG_INDIRECT) &&
(r_id_remap_data->status & ID_REMAP_IS_LINKED_DIRECT)) {
new_id->tag &= ~LIB_TAG_INDIRECT;
new_id->flag &= ~LIB_INDIRECT_WEAK_LINK;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
new_id->tag |= LIB_TAG_EXTERN;
}
#ifdef DEBUG_PRINT
printf("%s: %d occurrences skipped (%d direct and %d indirect ones)\n",
__func__,
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
r_id_remap_data->skipped_direct + r_id_remap_data->skipped_indirect,
r_id_remap_data->skipped_direct,
r_id_remap_data->skipped_indirect);
#endif
}
/**
* Replace all references in given Main to \a old_id by \a new_id
* (if \a new_id is NULL, it unlinks \a old_id).
*/
void BKE_libblock_remap_locked(Main *bmain, void *old_idv, void *new_idv, const short remap_flags)
{
IDRemap id_remap_data;
ID *old_id = old_idv;
ID *new_id = new_idv;
int skipped_direct, skipped_refcounted;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
BLI_assert(old_id != NULL);
BLI_assert((new_id == NULL) || GS(old_id->name) == GS(new_id->name));
BLI_assert(old_id != new_id);
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
libblock_remap_data(bmain, NULL, old_id, new_id, remap_flags, &id_remap_data);
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
if (free_notifier_reference_cb) {
free_notifier_reference_cb(old_id);
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
/* We assume editors do not hold references to their IDs... This is false in some cases
* (Image is especially tricky here),
* editors' code is to handle refcount (id->us) itself then. */
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
if (remap_editor_id_reference_cb) {
remap_editor_id_reference_cb(old_id, new_id);
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
skipped_direct = id_remap_data.skipped_direct;
skipped_refcounted = id_remap_data.skipped_refcounted;
if ((remap_flags & ID_REMAP_SKIP_USER_CLEAR) == 0) {
/* If old_id was used by some ugly 'user_one' stuff (like Image or Clip editors...), and user
* count has actually been incremented for that, we have to decrease once more its user
* count... unless we had to skip some 'user_one' cases. */
if ((old_id->tag & LIB_TAG_EXTRAUSER_SET) &&
!(id_remap_data.status & ID_REMAP_IS_USER_ONE_SKIPPED)) {
id_us_clear_real(old_id);
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
if (old_id->us - skipped_refcounted < 0) {
CLOG_ERROR(&LOG,
"Error in remapping process from '%s' (%p) to '%s' (%p): "
"wrong user count in old ID after process (summing up to %d)",
old_id->name,
old_id,
new_id ? new_id->name : "<NULL>",
new_id,
old_id->us - skipped_refcounted);
BLI_assert(0);
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
if (skipped_direct == 0) {
/* old_id is assumed to not be used directly anymore... */
if (old_id->lib && (old_id->tag & LIB_TAG_EXTERN)) {
old_id->tag &= ~LIB_TAG_EXTERN;
old_id->tag |= LIB_TAG_INDIRECT;
}
}
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
/* Some after-process updates.
* This is a bit ugly, but cannot see a way to avoid it.
* Maybe we should do a per-ID callback for this instead? */
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
switch (GS(old_id->name)) {
case ID_OB:
libblock_remap_data_postprocess_object_update(bmain, (Object *)old_id, (Object *)new_id);
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
break;
case ID_GR:
Collections and groups unification OVERVIEW * In 2.7 terminology, all layers and groups are now collection datablocks. * These collections are nestable, linkable, instanceable, overrideable, .. which opens up new ways to set up scenes and link + override data. * Viewport/render visibility and selectability are now a part of the collection and shared across all view layers and linkable. * View layers define which subset of the scene collection hierarchy is excluded for each. For many workflows one view layer can be used, these are more of an advanced feature now. OUTLINER * The outliner now has a "View Layer" display mode instead of "Collections", which can display the collections and/or objects in the view layer. * In this display mode, collections can be excluded with the right click menu. These will then be greyed out and their objects will be excluded. * To view collections not linked to any scene, the "Blender File" display mode can be used, with the new filtering option to just see Colleciton datablocks. * The outliner right click menus for collections and objects were reorganized. * Drag and drop still needs to be improved. Like before, dragging the icon or text gives different results, we'll unify this later. LINKING AND OVERRIDES * Collections can now be linked into the scene without creating an instance, with the link/append operator or from the collections view in the outliner. * Collections can get static overrides with the right click menu in the outliner, but this is rather unreliable and not clearly communicated at the moment. * We still need to improve the make override operator to turn collection instances into collections with overrides directly in the scene. PERFORMANCE * We tried to make performance not worse than before and improve it in some cases. The main thing that's still a bit slower is multiple scenes, we have to change the layer syncing to only updated affected scenes. * Collections keep a list of their parent collections for faster incremental updates in syncing and caching. * View layer bases are now in a object -> base hash to avoid quadratic time lookups internally and in API functions like visible_get(). VERSIONING * Compatibility with 2.7 files should be improved due to the new visibility controls. Of course users may not want to set up their scenes differently now to avoid having separate layers and groups. * Compatibility with 2.8 is mostly there, and was tested on Eevee demo and Hero files. There's a few things which are know to be not quite compatible, like nested layer collections inside groups. * The versioning code for 2.8 files is quite complicated, and isolated behind #ifdef so it can be removed at the end of the release cycle. KNOWN ISSUES * The G-key group operators in the 3D viewport were left mostly as is, they need to be modified still to fit better. * Same for the groups panel in the object properties. This needs to be updated still, or perhaps replaced by something better. * Collections must all have a unique name. Less restrictive namespacing is to be done later, we'll have to see how important this is as all objects within the collections must also have a unique name anyway. * Full scene copy and delete scene are exactly doing the right thing yet. Differential Revision: https://developer.blender.org/D3383 https://code.blender.org/2018/05/collections-and-groups/
2018-04-30 15:57:22 +02:00
libblock_remap_data_postprocess_collection_update(
bmain, NULL, (Collection *)old_id, (Collection *)new_id);
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
break;
case ID_ME:
case ID_CU:
case ID_MB:
case ID_HA:
case ID_PT:
case ID_VO:
if (new_id) { /* Only affects us in case obdata was relinked (changed). */
for (Object *ob = bmain->objects.first; ob; ob = ob->id.next) {
libblock_remap_data_postprocess_obdata_relink(bmain, ob, new_id);
}
}
break;
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
default:
break;
}
/* Node trees may virtually use any kind of data-block... */
/* XXX Yuck!!!! nodetree update can do pretty much any thing when talking about py nodes,
* including creating new data-blocks (see T50385), so we need to unlock main here. :(
* Why can't we have re-entrent locks? */
BKE_main_unlock(bmain);
libblock_remap_data_postprocess_nodetree_update(bmain, new_id);
BKE_main_lock(bmain);
2017-08-10 16:04:01 +02:00
/* Full rebuild of DEG! */
DEG_relations_tag_update(bmain);
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
void BKE_libblock_remap(Main *bmain, void *old_idv, void *new_idv, const short remap_flags)
{
BKE_main_lock(bmain);
BKE_libblock_remap_locked(bmain, old_idv, new_idv, remap_flags);
BKE_main_unlock(bmain);
}
/**
* Unlink given \a id from given \a bmain
* (does not touch to indirect, i.e. library, usages of the ID).
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
*
* \param do_flag_never_null: If true, all IDs using \a idv in a 'non-NULL' way are flagged by
* #LIB_TAG_DOIT flag (quite obviously, 'non-NULL' usages can never be unlinked by this function).
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
*/
"Fix" crash when deleting linked object which has indirect usages. This is in fact very hairy situation here... Objects are only refcounted by scenes, any other usage is 'free', which means once all object instanciations are gone Blender considers it can delete it. There is a trap here though: indirect usages. Typically, we should never modify linked data (because it is essencially useless, changes would be ignored and ost on next reload or even undo/redo). This means indirect usages are not affected by default 'safe' remapping/unlinking. For unlinking preceeding deletion however, this is not acceptable - we are likely to end with a zero-user ID (aka deletable one) which is still actually used by other linked data. Solution choosen here is double: I) From 'user-space' (i.e. outliner, operators...), we check for cases where deleting datablocks should not be allowed (indirect data or indirectly used data), and abort (with report) if needed. II) From 'lower' level (BKE_library_remap and RNA), we also unlink from linked data, which makes actual deletion possible and safe. Note that with previous behavior (2.77 one), linked object would be deleted, including from linked data - but then, once file is saved and reloaded, indirect usage would link back the deleted object, without any instanciation in scene, which made it somehow virtual and unreachable... With new behavior, this is no more possible, but on the other hand it means that in situations of dependency cycles (two linked objects using each other), linked objects become impossible to delete (from user space). Not sure what's best here, behavior with those corner cases of library linking is very poorly defined... :(
2016-07-01 17:51:08 +02:00
void BKE_libblock_unlink(Main *bmain,
void *idv,
const bool do_flag_never_null,
const bool do_skip_indirect)
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
{
"Fix" crash when deleting linked object which has indirect usages. This is in fact very hairy situation here... Objects are only refcounted by scenes, any other usage is 'free', which means once all object instanciations are gone Blender considers it can delete it. There is a trap here though: indirect usages. Typically, we should never modify linked data (because it is essencially useless, changes would be ignored and ost on next reload or even undo/redo). This means indirect usages are not affected by default 'safe' remapping/unlinking. For unlinking preceeding deletion however, this is not acceptable - we are likely to end with a zero-user ID (aka deletable one) which is still actually used by other linked data. Solution choosen here is double: I) From 'user-space' (i.e. outliner, operators...), we check for cases where deleting datablocks should not be allowed (indirect data or indirectly used data), and abort (with report) if needed. II) From 'lower' level (BKE_library_remap and RNA), we also unlink from linked data, which makes actual deletion possible and safe. Note that with previous behavior (2.77 one), linked object would be deleted, including from linked data - but then, once file is saved and reloaded, indirect usage would link back the deleted object, without any instanciation in scene, which made it somehow virtual and unreachable... With new behavior, this is no more possible, but on the other hand it means that in situations of dependency cycles (two linked objects using each other), linked objects become impossible to delete (from user space). Not sure what's best here, behavior with those corner cases of library linking is very poorly defined... :(
2016-07-01 17:51:08 +02:00
const short remap_flags = (do_skip_indirect ? ID_REMAP_SKIP_INDIRECT_USAGE : 0) |
(do_flag_never_null ? ID_REMAP_FLAG_NEVER_NULL_USAGE : 0);
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
BKE_main_lock(bmain);
BKE_libblock_remap_locked(bmain, idv, NULL, remap_flags);
BKE_main_unlock(bmain);
}
/**
* Similar to libblock_remap, but only affects IDs used by given \a idv ID.
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
*
* \param old_idv: Unlike BKE_libblock_remap, can be NULL,
* in which case all ID usages by given \a idv will be cleared.
* \param us_min_never_null: If \a true and new_id is NULL,
* 'NEVER_NULL' ID usages keep their old id, but this one still gets its user count decremented
* (needed when given \a idv is going to be deleted right after being unlinked).
*/
/* Should be able to replace all _relink() funcs (constraints, rigidbody, etc.) ? */
/* XXX Arg! Naming... :(
* _relink? avoids confusion with _remap, but is confusing with _unlink
* _remap_used_ids?
* _remap_datablocks?
* BKE_id_remap maybe?
* ... sigh
*/
void BKE_libblock_relink_ex(
Main *bmain, void *idv, void *old_idv, void *new_idv, const short remap_flags)
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
{
ID *id = idv;
ID *old_id = old_idv;
ID *new_id = new_idv;
/* No need to lock here, we are only affecting given ID, not bmain database. */
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
BLI_assert(id);
if (old_id) {
BLI_assert((new_id == NULL) || GS(old_id->name) == GS(new_id->name));
BLI_assert(old_id != new_id);
}
else {
BLI_assert(new_id == NULL);
}
libblock_remap_data(bmain, id, old_id, new_id, remap_flags, NULL);
/* Some after-process updates.
* This is a bit ugly, but cannot see a way to avoid it.
* Maybe we should do a per-ID callback for this instead?
*/
switch (GS(id->name)) {
case ID_SCE:
case ID_GR: {
/* Note: here we know which collection we have affected, so at lest for NULL children
* detection we can only process that one.
* This is also a required fix in case `id` would not be in Main anymore, which can happen
* e.g. when called from `id_delete`. */
Collection *owner_collection = (GS(id->name) == ID_GR) ? (Collection *)id :
((Scene *)id)->master_collection;
if (old_id) {
switch (GS(old_id->name)) {
case ID_OB:
libblock_remap_data_postprocess_object_update(
bmain, (Object *)old_id, (Object *)new_id);
break;
case ID_GR:
Collections and groups unification OVERVIEW * In 2.7 terminology, all layers and groups are now collection datablocks. * These collections are nestable, linkable, instanceable, overrideable, .. which opens up new ways to set up scenes and link + override data. * Viewport/render visibility and selectability are now a part of the collection and shared across all view layers and linkable. * View layers define which subset of the scene collection hierarchy is excluded for each. For many workflows one view layer can be used, these are more of an advanced feature now. OUTLINER * The outliner now has a "View Layer" display mode instead of "Collections", which can display the collections and/or objects in the view layer. * In this display mode, collections can be excluded with the right click menu. These will then be greyed out and their objects will be excluded. * To view collections not linked to any scene, the "Blender File" display mode can be used, with the new filtering option to just see Colleciton datablocks. * The outliner right click menus for collections and objects were reorganized. * Drag and drop still needs to be improved. Like before, dragging the icon or text gives different results, we'll unify this later. LINKING AND OVERRIDES * Collections can now be linked into the scene without creating an instance, with the link/append operator or from the collections view in the outliner. * Collections can get static overrides with the right click menu in the outliner, but this is rather unreliable and not clearly communicated at the moment. * We still need to improve the make override operator to turn collection instances into collections with overrides directly in the scene. PERFORMANCE * We tried to make performance not worse than before and improve it in some cases. The main thing that's still a bit slower is multiple scenes, we have to change the layer syncing to only updated affected scenes. * Collections keep a list of their parent collections for faster incremental updates in syncing and caching. * View layer bases are now in a object -> base hash to avoid quadratic time lookups internally and in API functions like visible_get(). VERSIONING * Compatibility with 2.7 files should be improved due to the new visibility controls. Of course users may not want to set up their scenes differently now to avoid having separate layers and groups. * Compatibility with 2.8 is mostly there, and was tested on Eevee demo and Hero files. There's a few things which are know to be not quite compatible, like nested layer collections inside groups. * The versioning code for 2.8 files is quite complicated, and isolated behind #ifdef so it can be removed at the end of the release cycle. KNOWN ISSUES * The G-key group operators in the 3D viewport were left mostly as is, they need to be modified still to fit better. * Same for the groups panel in the object properties. This needs to be updated still, or perhaps replaced by something better. * Collections must all have a unique name. Less restrictive namespacing is to be done later, we'll have to see how important this is as all objects within the collections must also have a unique name anyway. * Full scene copy and delete scene are exactly doing the right thing yet. Differential Revision: https://developer.blender.org/D3383 https://code.blender.org/2018/05/collections-and-groups/
2018-04-30 15:57:22 +02:00
libblock_remap_data_postprocess_collection_update(
bmain, owner_collection, (Collection *)old_id, (Collection *)new_id);
break;
default:
break;
}
}
else {
Collections and groups unification OVERVIEW * In 2.7 terminology, all layers and groups are now collection datablocks. * These collections are nestable, linkable, instanceable, overrideable, .. which opens up new ways to set up scenes and link + override data. * Viewport/render visibility and selectability are now a part of the collection and shared across all view layers and linkable. * View layers define which subset of the scene collection hierarchy is excluded for each. For many workflows one view layer can be used, these are more of an advanced feature now. OUTLINER * The outliner now has a "View Layer" display mode instead of "Collections", which can display the collections and/or objects in the view layer. * In this display mode, collections can be excluded with the right click menu. These will then be greyed out and their objects will be excluded. * To view collections not linked to any scene, the "Blender File" display mode can be used, with the new filtering option to just see Colleciton datablocks. * The outliner right click menus for collections and objects were reorganized. * Drag and drop still needs to be improved. Like before, dragging the icon or text gives different results, we'll unify this later. LINKING AND OVERRIDES * Collections can now be linked into the scene without creating an instance, with the link/append operator or from the collections view in the outliner. * Collections can get static overrides with the right click menu in the outliner, but this is rather unreliable and not clearly communicated at the moment. * We still need to improve the make override operator to turn collection instances into collections with overrides directly in the scene. PERFORMANCE * We tried to make performance not worse than before and improve it in some cases. The main thing that's still a bit slower is multiple scenes, we have to change the layer syncing to only updated affected scenes. * Collections keep a list of their parent collections for faster incremental updates in syncing and caching. * View layer bases are now in a object -> base hash to avoid quadratic time lookups internally and in API functions like visible_get(). VERSIONING * Compatibility with 2.7 files should be improved due to the new visibility controls. Of course users may not want to set up their scenes differently now to avoid having separate layers and groups. * Compatibility with 2.8 is mostly there, and was tested on Eevee demo and Hero files. There's a few things which are know to be not quite compatible, like nested layer collections inside groups. * The versioning code for 2.8 files is quite complicated, and isolated behind #ifdef so it can be removed at the end of the release cycle. KNOWN ISSUES * The G-key group operators in the 3D viewport were left mostly as is, they need to be modified still to fit better. * Same for the groups panel in the object properties. This needs to be updated still, or perhaps replaced by something better. * Collections must all have a unique name. Less restrictive namespacing is to be done later, we'll have to see how important this is as all objects within the collections must also have a unique name anyway. * Full scene copy and delete scene are exactly doing the right thing yet. Differential Revision: https://developer.blender.org/D3383 https://code.blender.org/2018/05/collections-and-groups/
2018-04-30 15:57:22 +02:00
/* No choice but to check whole objects/collections. */
libblock_remap_data_postprocess_collection_update(bmain, owner_collection, NULL, NULL);
libblock_remap_data_postprocess_object_update(bmain, NULL, NULL);
}
2016-07-19 10:23:51 +10:00
break;
}
case ID_OB:
if (new_id) { /* Only affects us in case obdata was relinked (changed). */
libblock_remap_data_postprocess_obdata_relink(bmain, (Object *)id, new_id);
}
break;
default:
break;
}
DEG_relations_tag_update(bmain);
ID-Remap - Step one: core work (cleanup and rework of generic ID datablock handling). This commit changes a lot of how IDs are handled internally, especially the unlinking/freeing processes. So far, this was very fuzy, to summarize cleanly deleting or replacing a datablock was pretty much impossible, except for a few special cases. Also, unlinking was handled by each datatype, in a rather messy and prone-to-errors way (quite a few ID usages were missed or wrongly handled that way). One of the main goal of id-remap branch was to cleanup this, and fatorize ID links handling by using library_query utils to allow generic handling of those, which is now the case (now, generic ID links handling is only "knwon" from readfile.c and library_query.c). This commit also adds backends to allow live replacement and deletion of datablocks in Blender (so-called 'remapping' process, where we replace all usages of a given ID pointer by a new one, or NULL one in case of unlinking). This will allow nice new features, like ability to easily reload or relocate libraries, real immediate deletion of datablocks in blender, replacement of one datablock by another, etc. Some of those are for next commits. A word of warning: this commit is highly risky, because it affects potentially a lot in Blender core. Though it was tested rather deeply, being totally impossible to check all possible ID usage cases, it's likely there are some remaining issues and bugs in new code... Please report them! ;) Review task: D2027 (https://developer.blender.org/D2027). Reviewed by campbellbarton, thanks a bunch.
2016-06-22 17:29:38 +02:00
}
static int id_relink_to_newid_looper(LibraryIDLinkCallbackData *cb_data)
{
const int cb_flag = cb_data->cb_flag;
if (cb_flag & IDWALK_CB_EMBEDDED) {
return IDWALK_RET_NOP;
}
ID **id_pointer = cb_data->id_pointer;
ID *id = *id_pointer;
if (id) {
/* See: NEW_ID macro */
if (id->newid) {
BKE_library_update_ID_link_user(id->newid, id, cb_flag);
id = id->newid;
*id_pointer = id;
}
if (id->tag & LIB_TAG_NEW) {
id->tag &= ~LIB_TAG_NEW;
BKE_libblock_relink_to_newid(id);
}
}
return IDWALK_RET_NOP;
}
/**
* Similar to #libblock_relink_ex,
* but is remapping IDs to their newid value if non-NULL, in given \a id.
*
* Very specific usage, not sure we'll keep it on the long run,
* currently only used in Object/Collection duplication code...
*/
void BKE_libblock_relink_to_newid(ID *id)
{
if (ID_IS_LINKED(id)) {
return;
}
BKE_library_foreach_ID_link(NULL, id, id_relink_to_newid_looper, NULL, 0);
}