*! Drop any headwords which begin with a slash. They all seem gibberish and

always duplicate the normal ones.
This commit is contained in:
Konstantin Isakov 2009-04-27 12:23:10 +00:00
parent 589e56dda2
commit f616a040d9

View file

@ -38,7 +38,7 @@ namespace
enum enum
{ {
Signature = 0x584c4742, // BGLX on little-endian, XLGB on big-endian Signature = 0x584c4742, // BGLX on little-endian, XLGB on big-endian
CurrentFormatVersion = 12 + BtreeIndexing::FormatVersion CurrentFormatVersion = 13 + BtreeIndexing::FormatVersion
}; };
struct IdxHeader struct IdxHeader
@ -148,6 +148,12 @@ namespace
// Strip any leading or trailing whitespaces // Strip any leading or trailing whitespaces
trimWs( word ); trimWs( word );
// If the word starts with a slash, we drop it. There are quite a lot
// of them, and they all seem to be redudant duplicates.
if ( word.size() && word[ 0 ] == '/' )
return;
// Check the input word for a superscript postfix ($1$, $2$ etc), which // Check the input word for a superscript postfix ($1$, $2$ etc), which
// signifies different meaning in Bgl files. We emit different meaning // signifies different meaning in Bgl files. We emit different meaning
// as different articles, but they appear in the index as the same word. // as different articles, but they appear in the index as the same word.