Page MenuHome
Differential D12722 Diff 43326 source/blender/blenkernel/intern/lib_id.c

Changeset View

Standalone View

View Options

source/blender/blenkernel/intern/lib_id.c

Show First 20 Lines • Show All 1,115 Lines • ▼ Show 20 Lines if (id) {
} }
id->icon_id = 0; id->icon_id = 0;
*((short *)id->name) = type; *((short *)id->name) = type;
if ((flag & LIB_ID_CREATE_NO_USER_REFCOUNT) == 0) { if ((flag & LIB_ID_CREATE_NO_USER_REFCOUNT) == 0) {
id->us = 1; id->us = 1;
} }
if ((flag & LIB_ID_CREATE_NO_MAIN) == 0) { if ((flag & LIB_ID_CREATE_NO_MAIN) == 0) {
/* Note that 2.8x versioning has tested not to cause conflicts. */ /* Note that 2.8x versioning has tested not to cause conflicts. Node trees are
BLI_assert(bmain->is_locked_for_linking == false || ELEM(type, ID_WS, ID_GR)); * skipped in this check to allow adding a geometry node tree for versioning. */
BLI_assert(bmain->is_locked_for_linking == false || ELEM(type, ID_WS, ID_GR, ID_NT));
JacquesLuckeUnsubmitted
Not Done
Inline Actions

I wonder if is_locked_for_linking should already be false when calling do_versions_after_linking_300. The after linking suggests that linking is already done at this point. Not sure what kinds of errors one can run into when creating data blocks in versioning code. (I can see that removing them can be an issue though)

JacquesLucke: I wonder if `is_locked_for_linking` should already be false when calling…
HooglyBooglyAuthorUnsubmitted
Done
Inline Actions

The reasoning makes sense, but the file reading code is not that logical-- it's still set to true in do_versions_after_linking.

HooglyBoogly: The reasoning makes sense, but the file reading code is not that logical-- it's still set to…
mont29Unsubmitted
Not Done
Inline Actions

I think this is wrongly handled currently, this exception makes no sense to me. Created a design task for it (T92333: BMain locking during the 'after liblink' doversion).

But for the time being think you will need that hack yes, fixing this is outside the scope of that patch.

mont29: I think this is wrongly handled currently, this exception makes no sense to me. Created a…
ListBase *lb = which_libbase(bmain, type); ListBase *lb = which_libbase(bmain, type);
BKE_main_lock(bmain); BKE_main_lock(bmain);
BLI_addtail(lb, id); BLI_addtail(lb, id);
BKE_id_new_name_validate(lb, id, name, false); BKE_id_new_name_validate(lb, id, name, false);
bmain->is_memfile_undo_written = false; bmain->is_memfile_undo_written = false;
/* alphabetic insertion: is in new_id */ /* alphabetic insertion: is in new_id */
BKE_main_unlock(bmain); BKE_main_unlock(bmain);
▲ Show 20 Lines • Show All 1,343 Lines • Show Last 20 Lines