input.c was capable of generating a paragraph structure with no text

in it, if the input paragraph contained (say) an unrecognised
control command and nothing else. Such paragraphs can confuse back
ends later on, so input.c should refrain from generating them. Added
a check and a polite error message (just in case the user manages to
generate an empty paragraph using otherwise legal syntax).

[originally from svn r5629]

3 files changed, 19 insertions, 1 deletions

diff --git a/error.c b/error.c
index 3635f0f..b1e74a1 100644
--- a/error.c
+++ b/error.c
@@ -278,6 +278,11 @@ static void do_error(int code, va_list ap) {
 	flags = FILEPOS;
 	sfree(sp);
 	break;
+      case err_emptypara:
+	fpos = *va_arg(ap, filepos *);
+	sprintf(error, "found no text in paragraph");
+	flags = FILEPOS;
+	break;
       case err_whatever:
 	sp = va_arg(ap, char *);
         vsprintf(error, sp, ap);
diff --git a/halibut.h b/halibut.h
index ed2a0da..99b9a0e 100644
--- a/halibut.h
+++ b/halibut.h
@@ -244,6 +244,7 @@ enum {
     err_text_codeline,		       /* \c line too long in text backend */
     err_htmlver,		       /* unrecognised HTML version keyword */
     err_charset,		       /* unrecognised character set name */
+    err_emptypara,		       /* paragraph contains no actual text */
     err_whatever                       /* random error of another type */
 };
 
diff --git a/input.c b/input.c
index 493449c..d8d44f9 100644
--- a/input.c
+++ b/input.c
@@ -1529,7 +1529,19 @@ static void read_file(paragraph ***ret, input *in, indexdata *idx,
 	}
 	stk_free(parsestk);
 	prev_para_type = par.type;
-	addpara(par, ret);
+	/*
+	 * Before we add the paragraph to the output list, we
+	 * should check that there was any text in it at all; there
+	 * might not be if (for example) the paragraph contained
+	 * nothing but an unrecognised command sequence, and if we
+	 * put an empty paragraph on the list it may confuse the
+	 * back ends later on.
+	 */
+	if (par.words) {
+	    addpara(par, ret);
+	} else {
+	    error(err_emptypara, &par.fpos);
+	}
 	if (t.type == tok_eof)
 	    already = TRUE;
     }