各位老师, 我用的是ORD中国版, 我需要从三维模型中提取廊道线, 效率特别慢, 有什么好的方法吗

Question

     public static void GetAllAlignmentFromActiveDgnModel( out List<Element> listLeft, out List<Element> listRight)
        {
            BIM.Application app = BMI.Utilities.ComApp;
            //获取所有廊道路线
            app.ActiveDesignFile.Models["Default-3D"].Activate();

            ModelElementsCollection cols = Session.Instance.GetActiveDgnModel().GetGraphicElements();
            listLeft = new List<Element>();
            listRight = new List<Element>();

            List<string> StrLeft = new List<string>() {
                "PKX-", "PJX-",
                "BG-L-2","BG-L-3","BG-L-4","BG-L-5",
                "JSG-L-2","JSG-L-3","JSG-L-4","JSG-L-5"
            , "PSG-L-1", "PSG-L-8", "PSG-L-7", "PSG-L-6"
            ,"YDJ-","-WBP","-TBT","ZFD-","-TBP"
            ,"CL" ,"XCD-","LYD-","YLJ-","TLJ-","LJTT-T-L-0"};

            List<string> StrRight = new List<string>() {
                "PKX-", "PJX-",
                "BG-R-2","BG-R-3","BG-R-4","BG-R-5",
                "JSG-R-2","JSG-R-3","JSG-R-4","JSG-R-5"
            , "PSG-R-1", "PSG-R-8", "PSG-R-7", "PSG-R-6"
            ,"YDJ-","-WBP","-TBT","ZFD-","-TBP"
            ,"CL" ,"XCD-","LYD-","YLJ-","TLJ-","LJTT-T-R-0"};

            //List<Element> list = new List<Element>();
            //获取当前模型下所有廊道线(out list);
            foreach (var element in cols)
            {
                string featureName = element.GetFeaturizedEntityName();

                //左侧
                foreach (string str in StrLeft)
                {
                    if (featureName.Contains(str) && featureName.Contains("-L"))
                    {
                        listLeft.Add(element);

                        break;
                    }
                }

                //右侧
                foreach (string str in StrRight)
                {
                    if (featureName.Contains(str) && featureName.Contains("-R"))
                    {
                        listRight.Add(element);

                        break;
                    }
                }
            }

        }

测试文件太大, 只能放进网盘了
链接：pan.baidu.com/.../1H-OUWOFKIC31epXIPDgWTA
提取码：i47x
--来自百度网盘超级会员V3的分享

Yongan.Fu · Answer

两个建议： 
 1、廊道线都可能是什么元素类型的？可以考虑用ScanCriteria根据元素类型先筛查一下，然后再进入你这个详细判断中。这样过滤后要考察的元素个数肯定会比你当前这个cols中的元素要少非常多。 
 2、将您的-L和-R直接写入列表StrLeft和StrRight中构成一个完整的FeatureName，这样就可以用字符串相等的比较运算而不用Contains运算了。设想一下，Contains函数是包含的意思，那底层算法一定是很复杂的。这样，您的 
 foreach (string str in StrLeft)
 {
 if (featureName.Contains(str) && featureName.Contains("-L"))
 {
 listLeft.Add(element);

 break;
 }
 } 
 就可以简写成： 
 for (int i=0; i<StrLeft.Count; i++)
{
 if (featureName == StrLeft[i])
 {
 listLeft.Add (element);
 break;
 }
} 
 
 【注意】：我还改用for取代了foreach。据说for能比foreach又快很多 
 3、如果还想进一步优化程序的话，其实listLeft和listRight中可以保存元素的ID，待需要用到时再读取出来。这样占用内存会小许多。一个element会占用较多的内存的。当然，这一步不是为了提高速度的。可以暂时忽略。