shithub: pdffs

Info  •  Files  •  Log  •  Branches

improve pdf2txt heuristics

--- a/op.c

+++ b/op.c

@@ -342,6 +342,13 @@

 static int

 tpmove(Op *op, Page *p)

 {

+	Object *x, *y;

+	x = arrayget(p->stack, 0);

+	y = arrayget(p->stack, 1);

+	if(y->num.d != 0)

+		fprint(1, "\n");

+	else if(x->num.d < 50)

+		fprint(1, " ");

 	USED(op, p);

 	return 0;

 }

@@ -378,7 +385,7 @@

 		o = arrayget(arr, i);

 		if(o->type == Ostr)

 			fprint(1, "%s", o->str);

-		else if(o->num.d < -14)

+		else if(o->num.d < -150)

 			fprint(1, " ");

 	}

 	USED(op);