From 88b914c8f65f78e31ae591bc00f09cf3192f502d Mon Sep 17 00:00:00 2001
From: Jabier Arraiza Cenoz <jabier.arraiza@marker.es>
Date: Fri, 18 Nov 2016 18:59:18 +0100
Subject: Fix #1627817. Bug in knot LPE

Fixed bugs:
  - https://launchpad.net/bugs/1627817

(bzr r15259)
---
 src/live_effects/lpe-knot.cpp | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

(limited to 'src')

diff --git a/src/live_effects/lpe-knot.cpp b/src/live_effects/lpe-knot.cpp
index fbc32cf3e..474523aa3 100644
--- a/src/live_effects/lpe-knot.cpp
+++ b/src/live_effects/lpe-knot.cpp
@@ -26,6 +26,7 @@
 #include <2geom/sbasis-to-bezier.h>
 #include <2geom/bezier-to-sbasis.h>
 #include <2geom/basic-intersection.h>
+#include "helper/geom.h"
 
 // for change crossing undo
 #include "verbs.h"
@@ -392,14 +393,14 @@ LPEKnot::doEffect_path (Geom::PathVector const &path_in)
     if (gpaths.size()==0){
         return path_in;
     }
-
-    for (unsigned comp=0; comp<path_in.size(); comp++){
+    Geom::PathVector const original_pathv = pathv_to_linear_and_cubic_beziers(path_in);
+    for (unsigned comp=0; comp<original_pathv.size(); comp++){
 
         //find the relevant path component in gpaths (required to allow groups!)
         //Q: do we always receive the group members in the same order? can we rest on that?
         unsigned i0 = 0;
         for (i0=0; i0<gpaths.size(); i0++){
-            if (path_in[comp]==gpaths[i0]) break;
+            if (original_pathv[comp]==gpaths[i0]) break;
         }
         if (i0 == gpaths.size() ) {THROW_EXCEPTION("lpe-knot error: group member not recognized");}// this should not happen...
 
@@ -514,7 +515,7 @@ collectPathsAndWidths (SPLPEItem const *lpeitem, Geom::PathVector &paths, std::v
             c = SP_SHAPE(lpeitem)->getCurve();
         }
         if (c) {
-            Geom::PathVector subpaths = c->get_pathvector();
+            Geom::PathVector subpaths = pathv_to_linear_and_cubic_beziers(c->get_pathvector());
             for (unsigned i=0; i<subpaths.size(); i++){
                 paths.push_back(subpaths[i]);
                 //FIXME: do we have to be more carefull when trying to access stroke width?
-- 
cgit v1.2.3