Merge pull request #46 from ping9719/main

 高级替换 <t>abs{</t><t>{</t> 测试文档

Author: shps951023

samples/docx/TestDemo04.docx

@@ -0,0 +1,142 @@
+namespace MiniSoftware.Extensions
+{
+    using DocumentFormat.OpenXml.Wordprocessing;
+    using System;
+    using System.Collections;
+    using System.Collections.Generic;
+    using System.ComponentModel;
+    using System.Dynamic;
+    using System.Linq;
+    using System.Text;
+
+    internal static class OpenXmlExtension
+    {
+        /// <summary>
+        /// 高级搜索：得到段落里面的连续字符串
+        /// </summary>
+        /// <param name="paragraph">段落</param>
+        /// <returns>Item1：连续文本；Item2：块；Item3：块文本</returns>
+        internal static List<Tuple<string, List<Run>, List<Text>>> GetContinuousString(this Paragraph paragraph)
+        {
+            List<Tuple<string, List<Run>, List<Text>>> tuples = new List<Tuple<string, List<Run>, List<Text>>>();
+            if (paragraph == null)
+                return tuples;
+
+            var sb = new StringBuilder();
+            var runs = new List<Run>();
+            var texts = new List<Text>();
+
+            //段落：所有子级
+            foreach (var pChildElement in paragraph.ChildElements)
+            {
+                //块
+                if (pChildElement is Run run)
+                {
+                    //文本块
+                    if (run.IsText())
+                    {
+                        var text = run.GetFirstChild<Text>();
+                        runs.Add(run);
+                        texts.Add(text);
+                        sb.Append(text.Text);
+                    }
+                    else
+                    {
+                        if (runs.Any())
+                            tuples.Add(new Tuple<string, List<Run>, List<Text>>(sb.ToString(), runs, texts));
+
+                        sb = new StringBuilder();
+                        runs = new List<Run>();
+                        texts = new List<Text>();
+                    }
+                }
+                //公式，书签...
+                else
+                {
+                    //跳过的类型
+                    if (pChildElement is BookmarkStart || pChildElement is BookmarkEnd)
+                    {
+
+                    }
+                    else
+                    {
+                        if (runs.Any())
+                            tuples.Add(new Tuple<string, List<Run>, List<Text>>(sb.ToString(), runs, texts));
+
+                        sb = new StringBuilder();
+                        runs = new List<Run>();
+                        texts = new List<Text>();
+                    }
+                }
+            }
+
+            if (runs.Any())
+                tuples.Add(new Tuple<string, List<Run>, List<Text>>(sb.ToString(), runs, texts));
+
+            sb = null;
+            runs = null;
+            texts = null;
+
+            return tuples;
+        }
+
+        /// <summary>
+        /// 整理字符串到连续字符串块中
+        /// </summary>
+        /// <param name="texts">连续字符串块</param>
+        /// <param name="text">待整理字符串</param>
+        internal static void TrimStringToInContinuousString(this IEnumerable<Text> texts, string text)
+        {
+            /*
+            //假如块为：[A][BC][DE][FG][H]
+            //假如替换：[AB][E][GH]
+            //优化块为：[AB][C][DE][FGH][]
+             */
+
+            var allTxtx = string.Concat(texts.SelectMany(o => o.Text));
+            var indexState = allTxtx.IndexOf(text);
+            if (indexState == -1)
+                return;
+
+            int indexEnd = indexState + text.Length - 1;
+            List<Tuple<int, char>> yl = new List<Tuple<int, char>>(allTxtx.Length);
+            int iRun = 0;
+            int iIndex = 0;
+            int iRunOf = -1;
+            foreach (var item in texts)
+            {
+                foreach (var item2 in item.Text)
+                {
+                    if (indexState <= iIndex && iIndex <= indexEnd)
+                    {
+                        if (iRunOf == -1)
+                            iRunOf = iRun;
+
+                        yl.Add(new Tuple<int, char>(iRunOf, item2));
+                    }
+                    else
+                    {
+                        yl.Add(new Tuple<int, char>(iRun, item2));
+                    }
+
+                    iIndex++;
+                }
+                iRun++;
+            }
+
+            int i = 0;
+            foreach (var item in texts)
+            {
+                item.Text = string.Concat(yl.Where(o => o.Item1 == i).Select(o => o.Item2));
+                i++;
+            }
+
+        }
+
+
+        internal static bool IsText(this Run run)
+        {
+            return run.Elements().All(o => o is Text || o is RunProperties);
+        }
+    }
+}

src/MiniWord/Extensions/OpenXmlExtension.cs

@@ -3,6 +3,7 @@
     using DocumentFormat.OpenXml;
     using DocumentFormat.OpenXml.Packaging;
     using DocumentFormat.OpenXml.Wordprocessing;
+    using MiniSoftware.Extensions;
     using MiniSoftware.Utility;
     using System;
     using System.Collections;
@@ -42,7 +43,9 @@ private static void SaveAsByTemplateImpl(Stream stream, byte[] template, Diction
         private static void Generate(this OpenXmlElement xmlElement, WordprocessingDocument docx, Dictionary<string, object> tags)
         {
             // avoid {{tag}} like <t>{</t><t>{</t> 
-            AvoidSplitTagText(xmlElement);
+            //AvoidSplitTagText(xmlElement);
+            // avoid {{tag}} like <t>aa{</t><t>{</t>  test in...
+            AvoidSplitTagText(xmlElement, tags.Select(o => "{{" + o.Key + "}}"));
 
             //Tables
             var tables = xmlElement.Descendants<Table>().ToArray();
@@ -144,6 +147,20 @@ private static void AvoidSplitTagText(OpenXmlElement xmlElement)
             }
         }
 
+        private static void AvoidSplitTagText(OpenXmlElement xmlElement, IEnumerable<string> txt)
+        {
+            foreach (var paragraph in xmlElement.Elements<Paragraph>())
+            {
+                foreach (var continuousString in paragraph.GetContinuousString())
+                {
+                    foreach (var text in txt.Where(o => continuousString.Item1.Contains(o)))
+                    {
+                        continuousString.Item3.TrimStringToInContinuousString(text);
+                    }
+                }
+            }
+        }
+
         private static void ReplaceText(OpenXmlElement xmlElement, WordprocessingDocument docx, Dictionary<string, object> tags)
         {
             var paragraphs = xmlElement.Descendants<Paragraph>().ToArray();