dev: comment parsing tests/refactoring, fix indented timedot comments

hledger-lib/Hledger/Read/Common.hs

@@ -96,7 +96,7 @@ module Hledger.Read.Common (
   isSameLineCommentStart,
   multilinecommentp,
   emptyorcommentlinep,
-  emptyorcommentlinep',
+  emptyorcommentlinep2,
   followingcommentp,
   transactioncommentp,
   commenttagsp,
@@ -407,6 +407,10 @@ setYear y = modify' (\j -> j{jparsedefaultyear=Just y})
 getYear :: JournalParser m (Maybe Year)
 getYear = fmap jparsedefaultyear get
 
+dp :: String -> TextParser m ()
+dp = const $ return ()  -- no-op
+-- dp = dbgparse 1  -- trace parse state at this --debug level
+
 -- | Get the decimal mark that has been specified for parsing, if any
 -- (eg by the CSV decimal-mark rule, or possibly a future journal directive).
 -- Return it as an AmountStyle that amount parsers can use.
@@ -1260,9 +1264,10 @@ multilinecommentp = startComment *> anyLine `skipManyTill` endComment
 
 -- | A blank or comment line in journal format: a line that's empty or
 -- containing only whitespace or whose first non-whitespace character
--- is semicolon, hash, or star.
+-- is semicolon, hash, or star. See also emptyorcommentlinep2.
 emptyorcommentlinep :: TextParser m ()
 emptyorcommentlinep = do
+  dp "emptyorcommentlinep"
   skipNonNewlineSpaces
   skiplinecommentp <|> void newline
   where
@@ -1275,25 +1280,18 @@ emptyorcommentlinep = do
 
 {-# INLINABLE emptyorcommentlinep #-}
 
-dp :: String -> TextParser m ()
-dp = const $ return ()  -- no-op
--- dp = dbgparse 1  -- trace parse state at this --debug level
-
--- | A new comment line parser (from TimedotReader).
--- Parse empty lines, all-blank lines, and lines beginning with any of
--- the provided comment-beginning characters.
-emptyorcommentlinep' :: [Char] -> TextParser m ()
-emptyorcommentlinep' cs = do
-  dp "emptyorcommentlinep'"
-  label ("empty line or comment line beginning with "++cs) $
-    void commentp <|> void (try $ skipNonNewlineSpaces >> newline)
-  where
-    commentp = do
-      choice (map (some.char) cs)
-      void $ takeWhileP Nothing (/='\n')
-      void $ optional newline
-
-{-# INLINABLE emptyorcommentlinep' #-}
+-- | A newer comment line parser.
+-- Parses a line which is empty, all blanks, or whose first non-blank character is one of those provided.
+emptyorcommentlinep2 :: [Char] -> TextParser m ()
+emptyorcommentlinep2 cs =
+  label ("empty line or comment line beginning with "++cs) $ do
+    dp "emptyorcommentlinep2"
+    skipNonNewlineSpaces
+    void newline <|> void commentp
+    where
+      commentp = do
+        choice (map (some.char) cs)
+        takeWhileP Nothing (/='\n') <* newline
 
 -- | Is this a character that, as the first non-whitespace on a line,
 -- starts a comment line ?

hledger-lib/Hledger/Read/TimedotReader.hs

@@ -50,7 +50,7 @@ import Text.Megaparsec hiding (parse)
 import Text.Megaparsec.Char
 
 import Hledger.Data
-import Hledger.Read.Common hiding (emptyorcommentlinep)
+import Hledger.Read.Common
 import Hledger.Utils
 import Data.Decimal (roundTo)
 import Data.Functor ((<&>))
@@ -112,7 +112,7 @@ timedotp = preamblep >> many dayp >> eof >> get
 preamblep :: JournalParser m ()
 preamblep = do
   dp "preamblep"
-  void $ many $ notFollowedBy datelinep >> (lift $ emptyorcommentlinep "#;*")
+  void $ many $ notFollowedBy datelinep >> (lift $ emptyorcommentlinep2 "#;*")
 
 -- | Parse timedot day entries to multi-posting time transactions for that day.
 -- @
@@ -156,7 +156,7 @@ datelinep = do
 commentlinesp :: JournalParser m ()
 commentlinesp = do
   dp "commentlinesp"
-  void $ many $ try $ lift $ emptyorcommentlinep "#;"
+  void $ many $ try $ lift $ emptyorcommentlinep2 "#;"
 
 -- orgnondatelinep :: JournalParser m ()
 -- orgnondatelinep = do
@@ -274,18 +274,3 @@ letterquantitiesp =
           | t@(c:_) <- group $ sort $ letter1:letters
           ]
     return groups
-
--- | XXX new comment line parser, move to Hledger.Read.Common.emptyorcommentlinep
--- Parse empty lines, all-blank lines, and lines beginning with any of the provided
--- comment-beginning characters.
-emptyorcommentlinep :: [Char] -> TextParser m ()
-emptyorcommentlinep cs =
-  label ("empty line or comment line beginning with "++cs) $ do
-    dbgparse tracelevel "emptyorcommentlinep"
-    skipNonNewlineSpaces
-    void newline <|> void commentp
-    where
-      commentp = do
-        choice (map (some.char) cs)
-        takeWhileP Nothing (/='\n') <* newline
-

hledger/Hledger/Cli/Conf.hs

@@ -98,8 +98,8 @@ dp = const $ return ()  -- no-op
 
 whitespacep, commentlinesp, restoflinep :: TextParser Identity ()
 whitespacep   = void $ {- dp "whitespacep"   >> -} many spacenonewline
-commentlinesp = void $ {- dp "commentlinesp" >> -} many (emptyorcommentlinep' "#")
-restoflinep   = void $ {- dp "restoflinep"   >> -} whitespacep >> emptyorcommentlinep' "#"
+commentlinesp = void $ {- dp "commentlinesp" >> -} many (emptyorcommentlinep2 "#")
+restoflinep   = void $ {- dp "restoflinep"   >> -} whitespacep >> emptyorcommentlinep2 "#"
 
 confp :: TextParser Identity [ConfSection]  -- a monadic TextParser to allow reusing other hledger parsers
 confp = do

hledger/test/journal/comments.test

@@ -31,7 +31,7 @@ $ hledger -f - print
 
 >= 0
 
-# ** 2. transaction comments must use ;
+# ** 2. Same-line transaction comments start only with ;. The description can contain # or *.
 <
 2017/1/1 this # and * are not ; the comment
 $ hledger -f - print
@@ -39,16 +39,33 @@ $ hledger -f - print
 
 >= 0
 
-# ** 3. posting comments must use ;
+# ** 3. Same-line posting comments start only with ;.
 <
 2017/1/1
-  a  0  # hash & star not allowed for posting comments
+  a  0  # posting comments can't start with # or *
+
 $ hledger -f - print
 > //
 >2 //
 >= 1
 
-# ** 4. register does not show comments
+# ** 4. Postings can be commented by a ; or a non-indented # (indented # or * does not start a comment).
+# Account names can contain # or (not at start) *.
+<
+2024-01-01
+  #a   1  ; posting to #a account
+  b*  -1  ; posting to b* account
+  *c   0  ; posting to c account, with * status mark
+  ;d   0  ; a comment line attached to the c posting above, part of the transcction
+; e    0  ; top level comment line, not part of the transaction
+# f    0  ; top level comment line, not part of the transaction
+
+$ hledger -f - accounts
+#a
+b*
+c
+
+# ** 5. register does not show comments.
 <
 2010/1/1 x
     a  1 ; comment

hledger/test/timedot.test

@@ -79,3 +79,14 @@ $ hledger -ftimedot:- print
     (g)            0.25
 
 >=
+# ** 5. Lines can be commented by a leading # or ;, indented or not.
+<
+2024-01-01
+;a
+ ;b
+#c
+ #d
+e
+
+$ hledger -ftimedot:- accounts
+e