Docx reader: Only use LTR when it is overriding BiDi setting

The left-to-right direction setting in docx is used in the spec only
for overriding an explicit right-to-left setting. We only process it
when it happens in a paragraph set with BiDi.

This is especially important for docs exported from Google Docs, which
explicitly (and unnecessarily) set "rtl=0" for every paragraph.

Closes: #5723

3 files changed, 14 insertions, 2 deletions

diff --git a/src/Text/Pandoc/Readers/Docx.hs b/src/Text/Pandoc/Readers/Docx.hs
index 8b0b295ab..387c3c7e2 100644
--- a/src/Text/Pandoc/Readers/Docx.hs
+++ b/src/Text/Pandoc/Readers/Docx.hs
@@ -124,10 +124,11 @@ instance Default DState where
 
 data DEnv = DEnv { docxOptions       :: ReaderOptions
                  , docxInHeaderBlock :: Bool
+                 , docxInBidi        :: Bool
                  }
 
 instance Default DEnv where
-  def = DEnv def False
+  def = DEnv def False False
 
 type DocxContext m = ReaderT DEnv (StateT DState m)
 
@@ -290,7 +291,10 @@ runStyleToTransform rPr
       return $ spanWith ("",[],[("dir","rtl")]) . transform
   | Just False <- isRTL rPr = do
       transform <- runStyleToTransform rPr{isRTL = Nothing}
-      return $ spanWith ("",[],[("dir","ltr")]) . transform
+      inBidi <- asks docxInBidi
+      return $ if inBidi
+               then spanWith ("",[],[("dir","ltr")]) . transform
+               else transform
   | Just SupScrpt <- rVertAlign rPr = do
       transform <- runStyleToTransform rPr{rVertAlign = Nothing}
       return $ superscript . transform
@@ -561,6 +565,10 @@ normalizeToClassName = map go . fromStyleName
 
 bodyPartToBlocks :: PandocMonad m => BodyPart -> DocxContext m Blocks
 bodyPartToBlocks (Paragraph pPr parparts)
+  | Just True <- pBidi pPr = do
+      let pPr' = pPr { pBidi = Nothing }
+      local (\s -> s{ docxInBidi = True })
+        (bodyPartToBlocks (Paragraph pPr' parparts))
   | isCodeDiv pPr = do
       transform <- parStyleToTransform pPr
       return $
diff --git a/src/Text/Pandoc/Readers/Docx/Parse.hs b/src/Text/Pandoc/Readers/Docx/Parse.hs
index 8c5c94bb9..889bd80fc 100644
--- a/src/Text/Pandoc/Readers/Docx/Parse.hs
+++ b/src/Text/Pandoc/Readers/Docx/Parse.hs
@@ -224,6 +224,7 @@ data ParagraphStyle = ParagraphStyle { pStyle      :: [ParStyle]
                                      , indentation :: Maybe ParIndentation
                                      , dropCap     :: Bool
                                      , pChange     :: Maybe TrackedChange
+                                     , pBidi       :: Maybe Bool
                                      }
                       deriving Show
 
@@ -232,6 +233,7 @@ defaultParagraphStyle = ParagraphStyle { pStyle = []
                                        , indentation = Nothing
                                        , dropCap     = False
                                        , pChange     = Nothing
+                                       , pBidi       = Just False
                                        }
 
 
@@ -998,6 +1000,7 @@ elemToParagraphStyle ns element sty
                                          isElem ns "w" "moveFrom" e
                                   ) >>=
                       getTrackedChange ns
+      , pBidi = checkOnOff ns pPr (elemName ns "w" "bidi")
       }
 elemToParagraphStyle _ _ _ =  defaultParagraphStyle
 
diff --git a/src/Text/Pandoc/Readers/Docx/Parse/Styles.hs b/src/Text/Pandoc/Readers/Docx/Parse/Styles.hs
index c2d27e7fb..ac2d6fa07 100644
--- a/src/Text/Pandoc/Readers/Docx/Parse/Styles.hs
+++ b/src/Text/Pandoc/Readers/Docx/Parse/Styles.hs
@@ -40,6 +40,7 @@ module Text.Pandoc.Readers.Docx.Parse.Styles (
   , getNumInfo
   , elemToRunStyle
   , defaultRunStyle
+  , checkOnOff
   ) where
 import Prelude
 import Codec.Archive.Zip