pseudo block c and enum, creates wrong line counting

WIth version 28.1 on Master branch following code will create wrong line counting:
```
#c
enum myEnum {
_A = 0,
_B = 1,
_C = 2
};
#endc
{#enum myEnum {underscoreToCase} deriving( Show, Eq) #}
```

Will create following code snippet
```
...
toEnum 2 = C
toEnum unmatched = error ("myEnum.toEnum: Cannot match " ++ show unmatched)

{-# LINE 8 "test.chs" #-}
...
```
This `LINE 8` should be `LINE 9`. This can be simply verified by repeating this the first code block. Each time, c2hs looses one line, eg.:
```
#c
enum myEnum {
_A = 0,
_B = 1,
_C = 2
};
#endc
{#enum myEnum {underscoreToCase} deriving( Show, Eq) #}

#c
enum myEnum2 {
_A2 = 0,
_B2 = 1,
_C2 = 2
};
#endc
{#enum myEnum2 {underscoreToCase} deriving( Show, Eq) #}
```
This leads to wrong line numbering in error/warnings.

Dumping the c2hs process gives following code for the pseudo-block of c:
```
#ce
#endc
#cn
#endc
#cu
#endc
#cm
#endc
#c
#endc
#cm
#endc
#cy
#endc
#cE
#endc
#cn
#endc
#cu
#endc
#cm
#endc
#c
#endc
#c{
#endc
#c

#endc
#c
#endc
#c
#endc
#c_
#endc
#cA
#endc
#c
#endc
#c=
#endc
#c
#endc
#c0
#endc
#c,
#endc
#c

...

#endc
#c}
#endc
#c;
#endc
#c

#endc
#c

#endc{#enum myEnum {underscoreToCase} deriving (Show, Eq) #}
```

I'm not sure, if this is how it should look like.
Any suggestions?