diff --git a/src/Text/Pandoc/Readers/HTML.hs b/src/Text/Pandoc/Readers/HTML.hs index e2be1c5bd..2b667c63c 100644 --- a/src/Text/Pandoc/Readers/HTML.hs +++ b/src/Text/Pandoc/Readers/HTML.hs @@ -1133,6 +1133,7 @@ htmlTag :: (HasReaderOptions st, Monad m) -> ParserT [Char] st m (Tag String, String) htmlTag f = try $ do lookAhead (char '<') + startpos <- getPosition inp <- getInput let ts = canonicalizeTags $ parseTagsOptions parseOptions{ optTagWarning = False @@ -1153,11 +1154,17 @@ htmlTag f = try $ do [] -> False (c:cs) -> isLetter c && all isNameChar cs - let endAngle = try $ do char '>' - pos <- getPosition - guard $ (sourceLine pos == ln && - sourceColumn pos >= col) || - sourceLine pos > ln + let endpos = if ln == 1 + then setSourceColumn startpos + (sourceColumn startpos + (col - 1)) + else setSourceColumn (setSourceLine startpos + (sourceLine startpos + (ln - 1))) + col + let endAngle = try $ + do char '>' + pos <- getPosition + guard $ pos >= endpos + let handleTag tagname = do -- basic sanity check, since the parser is very forgiving -- and finds tags in stuff like x<y) diff --git a/test/command/3989.md b/test/command/3989.md index 32b5e0574..bf078b2e4 100644 --- a/test/command/3989.md +++ b/test/command/3989.md @@ -1,7 +1,7 @@ ``` pandoc -f markdown -t native <span title="1st line of text <br> 2nd line of text">foo</span> -<span title="1st line of text <br> 2nd line of text">foo</span> + <span title="1st line of text <br> 2nd line of text">foo</span> ^D [Para [Span ("",[],[("title","1st line of text <br> 2nd line of text")]) [Str "foo"],SoftBreak,Span ("",[],[("title","1st line of text <br> 2nd line of text")]) [Str "foo"]]] ```