From 463ae07e9982e337b9340cb24598db57d008e33c Mon Sep 17 00:00:00 2001
From: Garret Rieger <grieger@google.com>
Date: Wed, 23 Nov 2022 18:41:23 +0000
Subject: [PATCH] [subset] In the preprocess subset call always use long loca.

Long loca is needed so that we can store the trimmed glyph bytes to allow us to safely skip trimming in the later subset.
---
 src/OT/glyf/glyf.hh    | 9 ++++++---
 src/hb-subset-input.cc | 5 +++++
 src/hb-subset-input.hh | 4 ++++
 src/hb-subset-plan.cc  | 1 +
 src/hb-subset-plan.hh  | 1 +
 5 files changed, 17 insertions(+), 3 deletions(-)

diff --git a/src/OT/glyf/glyf.hh b/src/OT/glyf/glyf.hh
index fe2f1d7ad..f714d793d 100644
--- a/src/OT/glyf/glyf.hh
+++ b/src/OT/glyf/glyf.hh
@@ -90,9 +90,12 @@ struct glyf
     | hb_map (&glyf_impl::SubsetGlyph::padded_size)
     ;
 
-    unsigned max_offset = + padded_offsets | hb_reduce (hb_add, 0);
-    bool use_short_loca = max_offset < 0x1FFFF;
-
+    bool use_short_loca = false;
+    if (likely (!c->plan->force_long_loca))
+    {
+      unsigned max_offset = + padded_offsets | hb_reduce (hb_add, 0);
+      use_short_loca = max_offset < 0x1FFFF;
+    }
 
     glyf_prime->serialize (c->serializer, glyphs.writer (), use_short_loca, c->plan, font);
     if (!use_short_loca) {
diff --git a/src/hb-subset-input.cc b/src/hb-subset-input.cc
index 880c5eea2..18a07fead 100644
--- a/src/hb-subset-input.cc
+++ b/src/hb-subset-input.cc
@@ -489,6 +489,11 @@ hb_subset_preprocess (hb_face_t *source)
                             HB_SUBSET_FLAGS_RETAIN_GIDS);
   input->attach_accelerator_data = true;
 
+  // Always use long loca in the preprocessed version. This allows
+  // us to store the glyph bytes unpadded which allows the future subset
+  // operation to run faster by skipping the trim padding step.
+  input->force_long_loca = true;
+
   hb_face_t* new_source = hb_subset_or_fail (source, input);
   hb_subset_input_destroy (input);
 
diff --git a/src/hb-subset-input.hh b/src/hb-subset-input.hh
index cb9652d6c..7c4645efd 100644
--- a/src/hb-subset-input.hh
+++ b/src/hb-subset-input.hh
@@ -60,6 +60,10 @@ struct hb_subset_input_t
 
   unsigned flags;
   bool attach_accelerator_data = false;
+
+  // If set loca format will always be the long version.
+  bool force_long_loca = false;
+
   hb_hashmap_t<hb_tag_t, float> *axes_location;
 #ifdef HB_EXPERIMENTAL_API
   hb_hashmap_t<unsigned, hb_bytes_t> *name_table_overrides;
diff --git a/src/hb-subset-plan.cc b/src/hb-subset-plan.cc
index 2d989e9c6..1d4f2c4a3 100644
--- a/src/hb-subset-plan.cc
+++ b/src/hb-subset-plan.cc
@@ -883,6 +883,7 @@ hb_subset_plan_create_or_fail (hb_face_t	 *face,
   void* accel = hb_face_get_user_data(face, hb_subset_accelerator_t::user_data_key());
 
   plan->attach_accelerator_data = input->attach_accelerator_data;
+  plan->force_long_loca = input->force_long_loca;
   if (accel)
     plan->accelerator = (hb_subset_accelerator_t*) accel;
 
diff --git a/src/hb-subset-plan.hh b/src/hb-subset-plan.hh
index 74b603bb5..656db706b 100644
--- a/src/hb-subset-plan.hh
+++ b/src/hb-subset-plan.hh
@@ -108,6 +108,7 @@ struct hb_subset_plan_t
   bool successful;
   unsigned flags;
   bool attach_accelerator_data = false;
+  bool force_long_loca = false;
 
   // For each cp that we'd like to retain maps to the corresponding gid.
   hb_set_t *unicodes;