platform/upstream/libHarfBuzzSharp.git
5 years agoRevert "Add harfbuzzjs build configuration (#1636)" (#1675)
Ebrahim Byagowi [Fri, 19 Apr 2019 19:14:09 +0000 (12:14 -0700)]
Revert "Add harfbuzzjs build configuration (#1636)" (#1675)

This reverts commit 694cb1beeefe1c54b2e613d2d566a21e248a2c9c.

5 years agoAdd harfbuzzjs build configuration (#1636)
Ebrahim Byagowi [Fri, 19 Apr 2019 14:51:04 +0000 (07:51 -0700)]
Add harfbuzzjs build configuration (#1636)

5 years agoMinor warning fix again
Behdad Esfahbod [Thu, 18 Apr 2019 23:04:59 +0000 (19:04 -0400)]
Minor warning fix again

5 years agoadd spaces
Michiharu Ariza [Thu, 18 Apr 2019 22:17:10 +0000 (15:17 -0700)]
add spaces

5 years agofix oss-fuzz issue 14345
Michiharu Ariza [Thu, 18 Apr 2019 21:53:35 +0000 (14:53 -0700)]
fix oss-fuzz issue 14345

5 years agosilence MVC warnings 3rd attempt
Michiharu Ariza [Thu, 18 Apr 2019 20:54:58 +0000 (13:54 -0700)]
silence MVC warnings 3rd attempt

5 years agosilence MVC warnings 2nd attempt
Michiharu Ariza [Thu, 18 Apr 2019 18:32:10 +0000 (11:32 -0700)]
silence MVC warnings 2nd attempt

5 years agosilence MVC warnings
Michiharu Ariza [Thu, 18 Apr 2019 17:38:57 +0000 (10:38 -0700)]
silence MVC warnings

5 years agoMinor
Behdad Esfahbod [Thu, 18 Apr 2019 16:21:25 +0000 (12:21 -0400)]
Minor

5 years ago[array] Simplify copy assignment/constructor
Behdad Esfahbod [Thu, 18 Apr 2019 14:04:10 +0000 (10:04 -0400)]
[array] Simplify copy assignment/constructor

To fix bogus MSVC warnings:

  c:\projects\harfbuzz\src\hb-array.hh(189): warning C4521: 'hb_array_t<Type>': multiple copy constructors specified [C:\projects\harfbuzz\build\harfbuzz.vcxproj]
  c:\projects\harfbuzz\src\hb-array.hh(189): warning C4522: 'hb_array_t<Type>': multiple assignment operators specified [C:\projects\harfbuzz\build\harfbuzz.vcxproj]

5 years ago[serialize] Fix offset calculation
Behdad Esfahbod [Wed, 17 Apr 2019 21:59:39 +0000 (17:59 -0400)]
[serialize] Fix offset calculation

5 years ago[subset] Assert offsets are zero during relocation
Behdad Esfahbod [Wed, 17 Apr 2019 21:58:13 +0000 (17:58 -0400)]
[subset] Assert offsets are zero during relocation

If they're not, it's a bug in our subsetting logic somewhere.  So check.

5 years ago[serializer] Add copy()
Behdad Esfahbod [Wed, 17 Apr 2019 15:00:08 +0000 (11:00 -0400)]
[serializer] Add copy()

Calls obj.copy() or obj.operator=() in that order.

5 years agoAdd HB_RETURN
Behdad Esfahbod [Wed, 17 Apr 2019 14:20:02 +0000 (10:20 -0400)]
Add HB_RETURN

5 years agoComment out ot_languages where fallback suffices
David Corbett [Tue, 16 Apr 2019 21:29:34 +0000 (17:29 -0400)]
Comment out ot_languages where fallback suffices

5 years agoFix "hb_script_" doc typo
David Corbett [Wed, 17 Apr 2019 13:11:44 +0000 (09:11 -0400)]
Fix "hb_script_" doc typo

5 years agoOne more auto return type
Behdad Esfahbod [Tue, 16 Apr 2019 22:33:51 +0000 (18:33 -0400)]
One more auto return type

5 years agoRename HB_AUTO_RETURN_EXPR to HB_AUTO_RETURN
Behdad Esfahbod [Tue, 16 Apr 2019 22:28:17 +0000 (18:28 -0400)]
Rename HB_AUTO_RETURN_EXPR to HB_AUTO_RETURN

5 years agoUse HB_AUTO_RETURN_EXPR in hb_min/max
Behdad Esfahbod [Tue, 16 Apr 2019 22:27:25 +0000 (18:27 -0400)]
Use HB_AUTO_RETURN_EXPR in hb_min/max

5 years agoUse auto return type for hb_first/hb_second
Behdad Esfahbod [Tue, 16 Apr 2019 22:26:30 +0000 (18:26 -0400)]
Use auto return type for hb_first/hb_second

5 years agoUse hb_deref_pointer() to reduce number of overloads
Behdad Esfahbod [Tue, 16 Apr 2019 21:34:06 +0000 (17:34 -0400)]
Use hb_deref_pointer() to reduce number of overloads

5 years agoProperly prioritize hb_hash()
Behdad Esfahbod [Tue, 16 Apr 2019 21:28:18 +0000 (17:28 -0400)]
Properly prioritize hb_hash()

5 years agoMove around
Behdad Esfahbod [Tue, 16 Apr 2019 21:22:29 +0000 (17:22 -0400)]
Move around

5 years agoFix priorities
Behdad Esfahbod [Tue, 16 Apr 2019 20:50:07 +0000 (16:50 -0400)]
Fix priorities

5 years agoUse type aliasing for meta-functions, ie. those returning a type
Behdad Esfahbod [Tue, 16 Apr 2019 20:45:53 +0000 (16:45 -0400)]
Use type aliasing for meta-functions, ie. those returning a type

5 years agoReduce LangTag from 3 language system tags to 1
David Corbett [Tue, 16 Apr 2019 14:04:45 +0000 (10:04 -0400)]
Reduce LangTag from 3 language system tags to 1

5 years agoReduce NullPool size
Behdad Esfahbod [Tue, 16 Apr 2019 15:35:09 +0000 (11:35 -0400)]
Reduce NullPool size

5 years ago[meta] Flesh out hb_invoke()
Behdad Esfahbod [Tue, 16 Apr 2019 15:24:42 +0000 (11:24 -0400)]
[meta] Flesh out hb_invoke()

5 years agoComment
Behdad Esfahbod [Tue, 16 Apr 2019 15:20:16 +0000 (11:20 -0400)]
Comment

5 years ago[meta] Add hb_invoke()
Behdad Esfahbod [Tue, 16 Apr 2019 14:50:22 +0000 (10:50 -0400)]
[meta] Add hb_invoke()

5 years ago[meta] Add HB_AUTO_RETURN_EXPR, HB_VOID_RETURN_EXPR, hb_priority, hb_has(), hb_get()
Behdad Esfahbod [Tue, 16 Apr 2019 14:45:20 +0000 (10:45 -0400)]
[meta] Add HB_AUTO_RETURN_EXPR, HB_VOID_RETURN_EXPR, hb_priority, hb_has(), hb_get()

The first three based on range-v3.

5 years agoAdd hb_void_tt<> ala std::void_t
Behdad Esfahbod [Tue, 16 Apr 2019 13:59:08 +0000 (09:59 -0400)]
Add hb_void_tt<> ala std::void_t

5 years agoFix copyright
Behdad Esfahbod [Mon, 15 Apr 2019 21:36:09 +0000 (17:36 -0400)]
Fix copyright

5 years ago[fuzzing] Run valgrind with --leak-check=full
Behdad Esfahbod [Mon, 15 Apr 2019 20:53:10 +0000 (16:53 -0400)]
[fuzzing] Run valgrind with --leak-check=full

5 years ago[fuzzing] Fail if valgrind is requested but not found
Behdad Esfahbod [Mon, 15 Apr 2019 20:52:21 +0000 (16:52 -0400)]
[fuzzing] Fail if valgrind is requested but not found

5 years agoMore tweaks to previous commit
Behdad Esfahbod [Mon, 15 Apr 2019 20:43:34 +0000 (16:43 -0400)]
More tweaks to previous commit

Delete assignment operator of OffsetTo<> instead of Offset<>.

In simple ArrayOf<>::sanitize() assert that Type has assignment operator.
Ideally we should SFINAE this and fallback to calling Type::sanitize()
if assignment operator is not available.  But we don't have a case of
that in the codebase.

5 years agoDelete default assignment operator Offset<>
Behdad Esfahbod [Mon, 15 Apr 2019 20:00:20 +0000 (16:00 -0400)]
Delete default assignment operator Offset<>

5 years agoAdd HB_FUNCOBJ()
Behdad Esfahbod [Mon, 15 Apr 2019 19:39:03 +0000 (15:39 -0400)]
Add HB_FUNCOBJ()

Fixes https://github.com/harfbuzz/harfbuzz/issues/1651

5 years ago[Usermanual]: fix Tamil error in Why-do-I-need-a-shaping-engine section.
Nathan Willis [Mon, 15 Apr 2019 17:05:14 +0000 (18:05 +0100)]
[Usermanual]: fix Tamil error in Why-do-I-need-a-shaping-engine section.

5 years agoMove static const to post-struct for a function object
Behdad Esfahbod [Mon, 15 Apr 2019 16:44:31 +0000 (12:44 -0400)]
Move static const to post-struct for a function object

Just sending this to bots to see if all happy, then turn it into macro and
apply everywhere.

Part of https://github.com/harfbuzz/harfbuzz/issues/1651

5 years agoUgh. Another try, to unbreak gcc this time!
Behdad Esfahbod [Mon, 15 Apr 2019 16:07:00 +0000 (12:07 -0400)]
Ugh.  Another try, to unbreak gcc this time!

Jenga.

5 years agoMove location of HB_UNUSED to make MSVC happy
Behdad Esfahbod [Mon, 15 Apr 2019 15:59:57 +0000 (11:59 -0400)]
Move location of HB_UNUSED to make MSVC happy

5 years agoFix gcc warning
Behdad Esfahbod [Mon, 15 Apr 2019 15:31:40 +0000 (11:31 -0400)]
Fix gcc warning

5 years ago[ci] Fix macOS glib issue
Ebrahim Byagowi [Sun, 14 Apr 2019 20:04:04 +0000 (00:34 +0430)]
[ci] Fix macOS glib issue

5 years agoAdd HB_NO_SUBSET_LAYOUT
Behdad Esfahbod [Sat, 13 Apr 2019 02:50:22 +0000 (22:50 -0400)]
Add HB_NO_SUBSET_LAYOUT

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years agoRevert "Hide symbols in hb-iter"
Behdad Esfahbod [Sat, 13 Apr 2019 02:42:44 +0000 (22:42 -0400)]
Revert "Hide symbols in hb-iter"

This reverts commit 98f14c4cdb837a962083a6702f401d41b4c1ec5c.

Same as previous commit.

5 years agoRevert "Hide more symbols"
Behdad Esfahbod [Sat, 13 Apr 2019 02:39:38 +0000 (22:39 -0400)]
Revert "Hide more symbols"

This reverts commit 2e86d50915cf1a791da9acb95245aa820a3d70f4.

I think the setup that caused me to do this is faulty and not hiding inlines.

5 years agoHide symbols in hb-iter
Behdad Esfahbod [Fri, 12 Apr 2019 22:11:18 +0000 (18:11 -0400)]
Hide symbols in hb-iter

Painful.  All template methods need to be explicitly hidden :(.

Maybe we should switch to -fvisibility=hidden pragma.

A LOT more to go.

5 years agoHide more symbols
Behdad Esfahbod [Fri, 12 Apr 2019 22:07:42 +0000 (18:07 -0400)]
Hide more symbols

Exposed by:

$ make -j5 CPPFLAGS="-O0" CXXFLAGS=-flto=thin LDFLAGS=-lc++ && ./check-symbols.sh

5 years agoHide a few more symbols
Behdad Esfahbod [Fri, 12 Apr 2019 21:59:18 +0000 (17:59 -0400)]
Hide a few more symbols

Exposed by:

$ make CPPFLAGS=-O0

5 years agoHide a few static methods
Behdad Esfahbod [Fri, 12 Apr 2019 21:50:03 +0000 (17:50 -0400)]
Hide a few static methods

Looks like static methods that do not get inlined end up exported.
We have a lot more.  Need to protect all at some point.  Wish there
was an easier way, like the visibility flag we pass that automatically
hides all inline methods.

Was exposed by check-symbols.sh when compiling on OS X 10.14 with:

$ make CPPFLAGS=-Oz CXXFLAGS=-flto=thin LDFLAGS=-lc++

5 years agoAdd HB_NO_OPTIONS
Behdad Esfahbod [Fri, 12 Apr 2019 20:21:58 +0000 (16:21 -0400)]
Add HB_NO_OPTIONS

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years agoAdd HB_NO_NAME_TABLE_AAT
Behdad Esfahbod [Fri, 12 Apr 2019 19:00:37 +0000 (15:00 -0400)]
Add HB_NO_NAME_TABLE_AAT

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years agoAdd HB_NO_OT_FONT_BITMAP
Behdad Esfahbod [Fri, 12 Apr 2019 18:58:53 +0000 (14:58 -0400)]
Add HB_NO_OT_FONT_BITMAP

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years agoAdd HB_NO_OT_FONT_CFF
Behdad Esfahbod [Fri, 12 Apr 2019 18:57:49 +0000 (14:57 -0400)]
Add HB_NO_OT_FONT_CFF

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years agoAdd HB_NO_OT_SHAPE_COMPLEX_HEBREW_FALLBACK
Behdad Esfahbod [Fri, 12 Apr 2019 14:16:12 +0000 (10:16 -0400)]
Add HB_NO_OT_SHAPE_COMPLEX_HEBREW_FALLBACK

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years agoAdd HB_NO_OT_SHAPE_FALLBACK
Behdad Esfahbod [Fri, 12 Apr 2019 14:12:11 +0000 (10:12 -0400)]
Add HB_NO_OT_SHAPE_FALLBACK

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years agoAdd HB_NO_OT_SHAPE_COMPLEX_THAI_FALLBACK
Behdad Esfahbod [Fri, 12 Apr 2019 13:35:29 +0000 (09:35 -0400)]
Add HB_NO_OT_SHAPE_COMPLEX_THAI_FALLBACK

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years agoAdd HB_NO_OT_LAYOUT_BLACKLIST
Behdad Esfahbod [Fri, 12 Apr 2019 13:33:25 +0000 (09:33 -0400)]
Add HB_NO_OT_LAYOUT_BLACKLIST

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years agoAdd HB_NO_OT_SHAPE_COMPLEX_VOWEL_CONSTRAINTS
Behdad Esfahbod [Thu, 11 Apr 2019 21:54:38 +0000 (17:54 -0400)]
Add HB_NO_OT_SHAPE_COMPLEX_VOWEL_CONSTRAINTS

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years agoUpdate gtk-doc annotations for inout counts on various getter functions.
Nathan Willis [Tue, 2 Apr 2019 17:49:40 +0000 (18:49 +0100)]
Update gtk-doc annotations for inout counts on various getter functions.

5 years ago[Docs] Minor edits to gtk-doc inline comment review.
Nathan Willis [Sun, 24 Mar 2019 15:07:07 +0000 (15:07 +0000)]
[Docs] Minor edits to gtk-doc inline comment review.

5 years ago[Docs] Minor; fix formatting for gtk-doc multiple-annotations.
Nathan Willis [Mon, 18 Mar 2019 14:03:16 +0000 (14:03 +0000)]
[Docs] Minor; fix formatting for gtk-doc multiple-annotations.

5 years ago[Docs] Annotate gtk-doc formatting with some un-annotated (out)s.
Nathan Willis [Sun, 17 Mar 2019 14:50:47 +0000 (14:50 +0000)]
[Docs] Annotate gtk-doc formatting with some un-annotated (out)s.

5 years ago[Docs] Fix gtk-doc formatting for (out) and (inout).
Nathan Willis [Sun, 17 Mar 2019 14:43:06 +0000 (14:43 +0000)]
[Docs] Fix gtk-doc formatting for (out) and (inout).

5 years ago[Docs] Fix gtk-doc references to 'kern' table functions, clarifying that GPOS is...
Nathan Willis [Sun, 17 Mar 2019 14:27:27 +0000 (14:27 +0000)]
[Docs] Fix gtk-doc references to 'kern' table functions, clarifying that GPOS is not examined.

5 years ago[Docs] Add inline gtk-doc documentation of GDEF glyph classes.
Nathan Willis [Sat, 16 Mar 2019 15:38:08 +0000 (15:38 +0000)]
[Docs] Add inline gtk-doc documentation of GDEF glyph classes.

5 years ago[Docs] Add inlind gtk-doc comments for hb-ot-layout functions.
Nathan Willis [Sat, 16 Mar 2019 15:10:21 +0000 (15:10 +0000)]
[Docs] Add inlind gtk-doc comments for hb-ot-layout functions.

5 years agoUse injected class name to simplify macros
Behdad Esfahbod [Thu, 11 Apr 2019 15:20:10 +0000 (11:20 -0400)]
Use injected class name to simplify macros

5 years ago[C++11] Use deleted methods
Behdad Esfahbod [Thu, 11 Apr 2019 15:18:04 +0000 (11:18 -0400)]
[C++11] Use deleted methods

5 years agoRename a few macros
Behdad Esfahbod [Thu, 11 Apr 2019 15:16:01 +0000 (11:16 -0400)]
Rename a few macros

5 years ago[arabic] Disable fallback shaping if HB_NO_OT_SHAPE_COMPLEX_ARABIC_FALLBACK defined
Behdad Esfahbod [Wed, 10 Apr 2019 19:53:48 +0000 (15:53 -0400)]
[arabic] Disable fallback shaping if HB_NO_OT_SHAPE_COMPLEX_ARABIC_FALLBACK defined

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years agoDon't install ot-font funcs on new fonts if HB_NO_OT_FONT defined
Behdad Esfahbod [Wed, 10 Apr 2019 19:40:03 +0000 (15:40 -0400)]
Don't install ot-font funcs on new fonts if HB_NO_OT_FONT defined

Currently linker cannot GC hb-ot-font completely because we install
it on fonts by default.  Don't do that if HB_NO_OT_FONT defined.

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years agoDon't use any default unicode funcs if HB_NO_UNICODE_FUNCS is defined
Behdad Esfahbod [Wed, 10 Apr 2019 19:38:15 +0000 (15:38 -0400)]
Don't use any default unicode funcs if HB_NO_UNICODE_FUNCS is defined

Part of https://github.com/harfbuzz/harfbuzz/issues/1652

5 years ago[coretext] Fix unused-variable error
Behdad Esfahbod [Mon, 8 Apr 2019 18:50:58 +0000 (14:50 -0400)]
[coretext] Fix unused-variable error

Fixes https://github.com/harfbuzz/harfbuzz/issues/1659

5 years agoFix MSVC C4068 warning (#1656)
Maks Naumov [Fri, 5 Apr 2019 18:46:27 +0000 (21:46 +0300)]
Fix MSVC C4068 warning (#1656)

5 years agoreplace test font SourceHanSans with its subet
Michiharu Ariza [Fri, 5 Apr 2019 17:15:08 +0000 (10:15 -0700)]
replace test font SourceHanSans with its subet

5 years agoFix MSVC C4138 warning (#1657)
Maks Naumov [Fri, 5 Apr 2019 09:29:56 +0000 (12:29 +0300)]
Fix MSVC C4138 warning (#1657)

5 years agoMinor
Behdad Esfahbod [Wed, 3 Apr 2019 23:24:12 +0000 (16:24 -0700)]
Minor

5 years ago[GDEF] Don't assume glyphlist is sorted
Behdad Esfahbod [Wed, 3 Apr 2019 23:06:55 +0000 (16:06 -0700)]
[GDEF] Don't assume glyphlist is sorted

As was hit by the fuzzer.

Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=14032

5 years ago[algs] Add hb_min() and hb_max()
Behdad Esfahbod [Wed, 3 Apr 2019 23:06:34 +0000 (16:06 -0700)]
[algs] Add hb_min() and hb_max()

5 years ago[serialize] Make putting breakpoint on out-of-memory easier
Behdad Esfahbod [Wed, 3 Apr 2019 22:48:27 +0000 (15:48 -0700)]
[serialize] Make putting breakpoint on out-of-memory easier

5 years ago[iter] Tweak SFINAE again
Behdad Esfahbod [Wed, 3 Apr 2019 22:31:53 +0000 (15:31 -0700)]
[iter] Tweak SFINAE again

Don't think we need hb_is_same().

5 years ago[array] Add compy assignment operator since copy constructor is explicit
Behdad Esfahbod [Wed, 3 Apr 2019 22:23:06 +0000 (15:23 -0700)]
[array] Add compy assignment operator since copy constructor is explicit

5 years ago[array] Add default copy constructor
Behdad Esfahbod [Wed, 3 Apr 2019 21:32:15 +0000 (14:32 -0700)]
[array] Add default copy constructor

MSVC seems to need it.

5 years ago[iter] Use different SFINAE scheme to make MSVC happy
Behdad Esfahbod [Wed, 3 Apr 2019 21:18:19 +0000 (14:18 -0700)]
[iter] Use different SFINAE scheme to make MSVC happy

From Orvid King: TLDR; MSVC has some issues using sizeof(declval<T>()) for
SFINAE of templated types, so I just used SFINAE in a different context where
MSVC doesn't have the issue.

5 years ago[meta] Add hb_is_same()
Behdad Esfahbod [Wed, 3 Apr 2019 21:15:01 +0000 (14:15 -0700)]
[meta] Add hb_is_same()

5 years ago[serialize] Actually reclaim storage from duplicate objects
Behdad Esfahbod [Wed, 3 Apr 2019 06:10:03 +0000 (23:10 -0700)]
[serialize] Actually reclaim storage from duplicate objects

5 years ago[serialize] Port to use object pool
Behdad Esfahbod [Wed, 3 Apr 2019 05:42:22 +0000 (22:42 -0700)]
[serialize] Port to use object pool

Tested, but feels fragile :(.

5 years ago[pool] Uses memset() instead of assigning Null()
Behdad Esfahbod [Wed, 3 Apr 2019 05:41:06 +0000 (22:41 -0700)]
[pool] Uses memset() instead of assigning Null()

Assignment is invalid on invalid object.

5 years agoAdd hb_pool_t<> for pooled memory allocation
Behdad Esfahbod [Wed, 3 Apr 2019 04:46:40 +0000 (21:46 -0700)]
Add hb_pool_t<> for pooled memory allocation

5 years ago[serialize] Disable packed_map again
Behdad Esfahbod [Wed, 3 Apr 2019 03:20:53 +0000 (20:20 -0700)]
[serialize] Disable packed_map again

Ugh.  Need to think of something else.

5 years ago[map] Protect more against pointer deref
Behdad Esfahbod [Wed, 3 Apr 2019 03:17:27 +0000 (20:17 -0700)]
[map] Protect more against pointer deref

5 years agoMore
Behdad Esfahbod [Wed, 3 Apr 2019 03:13:16 +0000 (20:13 -0700)]
More

5 years ago[serialize] Fix hb_hashmap_t<> for pointers and use in packed_map
Behdad Esfahbod [Wed, 3 Apr 2019 02:27:02 +0000 (19:27 -0700)]
[serialize] Fix hb_hashmap_t<> for pointers and use in packed_map

5 years ago[iter] Remove passing pointer to hb_iter()
Behdad Esfahbod [Wed, 3 Apr 2019 01:41:33 +0000 (18:41 -0700)]
[iter] Remove passing pointer to hb_iter()

While doable with hb_deref_pointer() as well, we also would then
need to do it in a ton of places.  Not worth it / messy.

5 years ago[array] Use dagger for hashing array
Behdad Esfahbod [Wed, 3 Apr 2019 01:22:39 +0000 (18:22 -0700)]
[array] Use dagger for hashing array

Also switch to better mixing.

5 years ago[iter] Accept pointers in hb_iter()
Behdad Esfahbod [Wed, 3 Apr 2019 01:12:01 +0000 (18:12 -0700)]
[iter] Accept pointers in hb_iter()

No idea how to avoid dupicating code.  Was hoping hb_deref_pointer()
would do it, but looks like a pointer can't bind to a universal
reference T&&.  Humm.

5 years ago[serialize] Towards maintaining packed_map
Behdad Esfahbod [Wed, 3 Apr 2019 00:49:52 +0000 (17:49 -0700)]
[serialize] Towards maintaining packed_map

5 years ago[serialize] Minor
Behdad Esfahbod [Wed, 3 Apr 2019 00:42:10 +0000 (17:42 -0700)]
[serialize] Minor