diff --git a/Shake.hs b/Shake.hs
index 8069dcd90..89962772b 100755
--- a/Shake.hs
+++ b/Shake.hs
@@ -113,7 +113,11 @@ sed      = "sed -E"
 fromsrcmd = "-f markdown-smart-tex_math_dollars"
 
 -- The kind of markdown we like to generate for the website.
-towebmd = "-t markdown-smart-fenced_divs --atx-headers"
+-- This will be consumed by sphinx extensions:
+--  recommonmark (Commonmark syntax, https://spec.commonmark.org)
+--  sphinx-markdown-tables (PHP Markdown Extra table syntax, https://michelf.ca/projects/php-markdown/extra/#table)
+-- XXX trying to force the use of pipe_tables here, but sometimes it uses html instead
+towebmd = "-t markdown-smart-fenced_divs-simple_tables-multiline_tables-grid_tables --atx-headers"
 
 
 main = do
diff --git a/hledger/hledger_options.m4.md b/hledger/hledger_options.m4.md
index e1c5801b5..f2fb1b77c 100644
--- a/hledger/hledger_options.m4.md
+++ b/hledger/hledger_options.m4.md
@@ -168,12 +168,12 @@ but it can also be one of several other formats, listed below.
 hledger detects the format automatically based on the file extension,
 or if that is not recognised, by trying each built-in "reader" in turn:
 
-| Reader:         | Reads:                                                | Used for file extensions:                          
-|-----------------|-------------------------------------------------------|-------------------------------------------
-| `journal`       | hledger's journal format, also some Ledger journals   | `.journal` `.j` `.hledger` `.ledger`   
-| `timeclock`     | timeclock files (precise time logging)                | `.timeclock`                              
-| `timedot`       | timedot files (approximate time logging)              | `.timedot`                                
-| `csv`           | comma-separated values (data interchange)             | `.csv`                                    
+| Reader:     | Reads:                                              | Used for file extensions:                           |
+|-------------|-----------------------------------------------------|-----------------------------------------------------|
+| `journal`   | hledger's journal format, also some Ledger journals | `.journal` `.j` `.hledger` `.ledger`                |
+| `timeclock` | timeclock files (precise time logging)              | `.timeclock`                                        |
+| `timedot`   | timedot files (approximate time logging)            | `.timedot`                                          |
+| `csv`       | comma-separated values (data interchange)           | `.csv`                                              |
 
 If needed (eg to ensure correct error messages when a file has the "wrong" extension), 
 you can force a specific reader/format by prepending it to the file path with a colon.