kopt: add fallbacks for cases where kctx is not in cache

There were a handful of cases where if there was no cached kctx there
was no fallback and several KoptInterface methods would return nil,
causing issues in various parts of KOReader (this happened with the
migration to selected_text everywhere but it's unclear how that change
caused this regression).

In any case, from a correctness perspective it makes sense to have the
corresponding fallback paths to create a new kctx if we couldn't find a
cached one.

Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>

frontend/document/koptinterface.lua

@@ -538,19 +538,22 @@ function KoptInterface:getReflowedTextBoxes(doc, pageno)
     local hash = table.concat(hash_list, "|")
     local cached = DocCache:check(hash)
     if not cached then
+        local kc
         local kctx_hash = hash:gsub("^rfpgboxes|", "kctx|")
         cached = DocCache:check(kctx_hash)
-        if cached then
+        if not cached then
-            local kc = self:waitForContext(cached.kctx)
+            kc = self:getCachedContext(doc, pageno)
-            --kc:setDebug()
+        else
-            local fullwidth, fullheight = kc:getPageDim()
+            kc = self:waitForContext(cached.kctx)
-            local boxes, nr_word = kc:getReflowedWordBoxes("dst", 0, 0, fullwidth, fullheight)
+        end
-            if not boxes then
+        --kc:setDebug()
-                return nil
+        local fullwidth, fullheight = kc:getPageDim()
-            end
+        local boxes, nr_word = kc:getReflowedWordBoxes("dst", 0, 0, fullwidth, fullheight)
-            DocCache:insert(hash, CacheItem:new{ rfpgboxes = boxes, size = 192 * nr_word }) -- estimation
+        if not boxes then
-            return boxes
+            return
         end
+        DocCache:insert(hash, CacheItem:new{ rfpgboxes = boxes, size = 192 * nr_word }) -- estimation
+        return boxes
     else
         return cached.rfpgboxes
     end
@@ -566,19 +569,27 @@ function KoptInterface:getNativeTextBoxes(doc, pageno)
     local hash = table.concat(hash_list, "|")
     local cached = DocCache:check(hash)
     if not cached then
+        local kc
         local kctx_hash = hash:gsub("^nativepgboxes|", "kctx|")
         cached = DocCache:check(kctx_hash)
-        if cached then
+        if not cached then
-            local kc = self:waitForContext(cached.kctx)
+            kc = self:createContext(doc, pageno)
-            --kc:setDebug()
+            DocCache:insert(kctx_hash, ContextCacheItem:new{
-            local fullwidth, fullheight = kc:getPageDim()
+                persistent = true,
-            local boxes, nr_word = kc:getNativeWordBoxes("dst", 0, 0, fullwidth, fullheight)
+                size = self.last_context_size or self.default_context_size,
-            if not boxes then
+                kctx = kc,
-                return nil
+            })
-            end
+        else
-            DocCache:insert(hash, CacheItem:new{ nativepgboxes = boxes, size = 192 * nr_word }) -- estimation
+            kc = self:waitForContext(cached.kctx)
-            return boxes
+        end
+        --kc:setDebug()
+        local fullwidth, fullheight = kc:getPageDim()
+        local boxes, nr_word = kc:getNativeWordBoxes("dst", 0, 0, fullwidth, fullheight)
+        if not boxes then
+            return
         end
+        DocCache:insert(hash, CacheItem:new{ nativepgboxes = boxes, size = 192 * nr_word }) -- estimation
+        return boxes
     else
         return cached.nativepgboxes
     end
@@ -596,21 +607,24 @@ function KoptInterface:getReflowedTextBoxesFromScratch(doc, pageno)
     local hash = table.concat(hash_list, "|")
     local cached = DocCache:check(hash)
     if not cached then
+        local reflowed_kc
         local kctx_hash = hash:gsub("^scratchrfpgboxes|", "kctx|")
         cached = DocCache:check(kctx_hash)
-        if cached then
+        if not cached then
-            local reflowed_kc = self:waitForContext(cached.kctx)
+            reflowed_kc = self:getCachedContext(doc, pageno)
-            local fullwidth, fullheight = reflowed_kc:getPageDim()
+        else
-            local kc = self:createContext(doc, pageno)
+            reflowed_kc = self:waitForContext(cached.kctx)
-            kc:copyDestBMP(reflowed_kc)
+        end
-            local boxes, nr_word = kc:getNativeWordBoxes("dst", 0, 0, fullwidth, fullheight)
+        local fullwidth, fullheight = reflowed_kc:getPageDim()
-            kc:free()
+        local kc = self:createContext(doc, pageno)
-            if not boxes then
+        kc:copyDestBMP(reflowed_kc)
-                return nil
+        local boxes, nr_word = kc:getNativeWordBoxes("dst", 0, 0, fullwidth, fullheight)
-            end
+        kc:free()
-            DocCache:insert(hash, CacheItem:new{ scratchrfpgboxes = boxes, size = 192 * nr_word }) -- estimation
+        if not boxes then
-            return boxes
+            return
         end
+        DocCache:insert(hash, CacheItem:new{ scratchrfpgboxes = boxes, size = 192 * nr_word }) -- estimation
+        return boxes
     else
         return cached.scratchrfpgboxes
     end
@@ -723,17 +737,20 @@ function KoptInterface:getReflewOCRWord(doc, pageno, rect)
     local hash = table.concat(hash_list, "|")
     local cached = DocCache:check(hash)
     if not cached then
+        local kc
         local kctx_hash = hash:gsub("^rfocrword|", "kctx|")
         cached = DocCache:check(kctx_hash)
-        if cached then
+        if not cached then
-            local kc = self:waitForContext(cached.kctx)
+            kc = self:getCachedContext(doc, pageno)
-            local _, word = pcall(
+        else
-                kc.getTOCRWord, kc, "dst",
+            kc = self:waitForContext(cached.kctx)
-                rect.x, rect.y, rect.w, rect.h,
-                self.tessocr_data, self.ocr_lang, self.ocr_type, 0, 1)
-            DocCache:insert(hash, CacheItem:new{ rfocrword = word, size = #word + 64 }) -- estimation
-            return word
         end
+        local _, word = pcall(
+            kc.getTOCRWord, kc, "dst",
+            rect.x, rect.y, rect.w, rect.h,
+            self.tessocr_data, self.ocr_lang, self.ocr_type, 0, 1)
+        DocCache:insert(hash, CacheItem:new{ rfocrword = word, size = #word + 64 }) -- estimation
+        return word
     else
         return cached.rfocrword
     end