Fix tex2lyx parsing of consecutive hyphens

tex2lyx was already behaving as if bug #3647 was fixed, but currently -- and
--- have the same special meaning in LyX as in LaTeX.

src/tex2lyx/test/test-insets.lyx.lyx

@@ -6768,6 +6768,70 @@ and a spif\SpecialChar \textcompwordmark{}
 fy ligature break.
 \end_layout
 
+\begin_layout Standard
+There are dashes: endash in short form -- and long form –, emdash is alike: --- and —. If we really want several hyphens in a row, we need to separate them: -
+\begin_inset ERT
+status collapsed
+
+\begin_layout Plain Layout
+{}
+\end_layout
+
+\end_inset
+
+-, -
+\begin_inset ERT
+status collapsed
+
+\begin_layout Plain Layout
+{}
+\end_layout
+
+\end_inset
+
+-
+\begin_inset ERT
+status collapsed
+
+\begin_layout Plain Layout
+{}
+\end_layout
+
+\end_inset
+
+-, -
+\begin_inset ERT
+status collapsed
+
+\begin_layout Plain Layout
+{}
+\end_layout
+
+\end_inset
+
+-
+\begin_inset ERT
+status collapsed
+
+\begin_layout Plain Layout
+{}
+\end_layout
+
+\end_inset
+
+-
+\begin_inset ERT
+status collapsed
+
+\begin_layout Plain Layout
+{}
+\end_layout
+
+\end_inset
+
+- etc.
+\end_layout
+
 \begin_layout Standard
 
 \begin_inset ERT

src/tex2lyx/test/test-insets.tex

@@ -656,6 +656,10 @@ And what about special characters like hyphe\-nation mark,
 ellipsis\ldots, and end-of-sentence\@. LyX also supports a menu
 separator\lyxarrow{}and a spif\textcompwordmark{}fy ligature break.
 
+There are dashes: endash in short form -- and long form \textendash,
+emdash is alike: --- and \textemdash. If we really want several hyphens
+in a row, we need to separate them: -{}-, -{}-{}-, -{}-{}-{}- etc.
+
 LyX translates the phrases LyX, TeX, LaTeX2e and LaTeX
 to the commands \LyX{}, \TeX{}, \LaTeXe{} and \LaTeX{}.
 If these phrases occur as part of other words (like 1LyX or aTeX or LaTeX3)

src/tex2lyx/text.cpp

@@ -2420,10 +2420,8 @@ void parse_text(Parser & p, ostream & os, unsigned flags, bool outer,
 				// {}
 				Token const prev = p.prev_token();
 				p.get_token();
-				if (p.next_token().character() == '`' ||
-				    (prev.character() == '-' &&
-				     p.next_token().character() == '-'))
-					; // ignore it in {}`` or -{}-
+				if (p.next_token().character() == '`')
+					; // ignore it in {}``
 				else
 					output_ert_inset(os, "{}", context);
 			} else if (next.cat() == catEscape &&