{"id":2777,"date":"2016-03-18T13:41:10","date_gmt":"2016-03-18T13:41:10","guid":{"rendered":"http:\/\/www.ericwhite.com\/home2\/bm8qcmjy\/public_html\/blog\/?page_id=2777"},"modified":"2016-03-20T11:48:14","modified_gmt":"2016-03-20T11:48:14","slug":"transforming-a-docx-with-hierarchical-numbering-into-different-xml","status":"publish","type":"page","link":"https:\/\/www.ericwhite.com\/blog\/transforming-a-docx-with-hierarchical-numbering-into-different-xml\/","title":{"rendered":"Transforming a Docx with Hierarchical Numbering into Different XML"},"content":{"rendered":"

The new ListItemRetriever.cs module<\/a> in PowerTools for Open XML<\/a> enables us to find out lots of information about the numbered and bulleted lists in our documents.  We can write a small amount of code and transform a DOCX that contains hierarchical numbered lists into an alternative form of XML.  For instance, we may want to transform a document that looks like this:<\/p>\n

\"2656.DocxWithOutline\"<\/p>\n

Into XML that looks like this:<\/p>\n

<?xml version=\"1.0\" encoding=\"utf-8\"?>\r\n<Root>\r\n  <Paragraph>Video provides a powerful way to help you prove your point.<\/Paragraph>\r\n  <Indent Level=\"1\">\r\n    <Heading>Overview<\/Heading>\r\n    <Indent Level=\"1.1\">\r\n      <Heading>Dr. Johnson’s Vitae<\/Heading>\r\n      <Paragraph>You can also type a keyword.<\/Paragraph>\r\n    <\/Indent>\r\n    <Indent Level=\"1.2\">\r\n      <Heading>Team List<\/Heading>\r\n      <Paragraph>John<\/Paragraph>\r\n      <Paragraph>Bob<\/Paragraph>\r\n      <Paragraph>Bill<\/Paragraph>\r\n      <Paragraph>Frank<\/Paragraph>\r\n    <\/Indent>\r\n    <Indent Level=\"1.3\">\r\n      <Heading>Motivations<\/Heading>\r\n      <Paragraph>Lorem ipsum dolor sit amet.<\/Paragraph>\r\n    <\/Indent>\r\n    <Indent Level=\"1.4\">\r\n      <Heading>Benefits<\/Heading>\r\n      <Paragraph>Nunc viverra imperdiet enim.<\/Paragraph>\r\n      <Indent Level=\"1.4.1\">\r\n        <Heading>Aaa<\/Heading>\r\n      <\/Indent>\r\n      <Indent Level=\"1.4.2\">\r\n        <Heading>Bbb<\/Heading>\r\n      <\/Indent>\r\n      <Indent Level=\"1.4.3\">\r\n        <Heading>CCC<\/Heading>\r\n      <\/Indent>\r\n    <\/Indent>\r\n  <\/Indent>\r\n  <Indent Level=\"2\">\r\n    <Heading>Synopsis<\/Heading>\r\n    <Indent Level=\"2.1\">\r\n      <Heading>Day 1 – Compression<\/Heading>\r\n      <Paragraph>Aenean nec lorem.<\/Paragraph>\r\n    <\/Indent>\r\n    <Indent Level=\"2.2\">\r\n      <Heading>Day 2 – Distribution<\/Heading>\r\n      <Paragraph>In porttitor.<\/Paragraph>\r\n    <\/Indent>\r\n    <Indent Level=\"2.3\">\r\n      <Heading>Day 3 – Time Line<\/Heading>\r\n      <Paragraph>Donec laoreet nonummy augue.<\/Paragraph>\r\n    <\/Indent>\r\n  <\/Indent>\r\n  <Indent Level=\"3\">\r\n    <Heading>Technical Details<\/Heading>\r\n    <Indent Level=\"3.1\">\r\n      <Heading>Engine<\/Heading>\r\n    <\/Indent>\r\n    <Indent Level=\"3.2\">\r\n      <Heading>Passenger Compartment<\/Heading>\r\n    <\/Indent>\r\n    <Indent Level=\"3.3\">\r\n      <Heading>Trunk<\/Heading>\r\n    <\/Indent>\r\n    <Indent Level=\"3.4\">\r\n      <Heading>Tires<\/Heading>\r\n    <\/Indent>\r\n  <\/Indent>\r\n  <Indent Level=\"4\">\r\n    <Heading>Summary<\/Heading>\r\n    <Indent Level=\"4.1\">\r\n      <Heading>Contraindications<\/Heading>\r\n    <\/Indent>\r\n    <Indent Level=\"4.2\">\r\n      <Heading>Index<\/Heading>\r\n    <\/Indent>\r\n  <\/Indent>\r\n<\/Root><\/pre>\n
The ListItemRetreiver assembles all kinds of information about each item in a numbered list, and it is easy to retrieve this information – it is stored as annotations on the paragraph XML elements after calling ListItemRetriever.RetrieveListItem.  The following code accomplishes the above transform:<\/p>\n
\/***************************************************************************\r\n\r\nCopyright (c) Microsoft Corporation 2014.\r\n\r\nThis code is licensed using the Microsoft Public License (Ms-PL).  The text of the license\r\ncan be found here:\r\n\r\nhttp:\/\/www.microsoft.com\/resources\/sharedsource\/licensingbasics\/publiclicense.mspx\r\n\r\n***************************************************************************\/\r\n\r\nusing System;\r\nusing System.Collections.Generic;\r\nusing System.IO;\r\nusing System.Linq;\r\nusing System.Text;\r\nusing System.Threading.Tasks;\r\nusing System.Xml.Linq;\r\nusing DocumentFormat.OpenXml.Packaging;\r\nusing OpenXmlPowerTools;\r\n\r\nclass ListItemRetriever01\r\n{\r\n    private class XmlStackItem\r\n    {\r\n        public XElement Element;\r\n        public int[] LevelNumbers;\r\n    }\r\n\r\n    \/*\r\n     * This example loads each document into a byte array, then into a memory stream,\r\n     * so that the document can be opened for writing without modifying the source document.\r\n     *\/\r\n\r\n    static void Main(string[] args)\r\n    {\r\n        var thisDir = new DirectoryInfo(\".\");\r\n        foreach (var xmlFile in thisDir.GetFiles(\"*.xml\"))\r\n            xmlFile.Delete();\r\n        using (WordprocessingDocument wDoc =\r\n            WordprocessingDocument.Open(\"..\/..\/NumberedListTest.docx\", false))\r\n        {\r\n            int abstractNumId = 0;\r\n            XElement xml = ConvertDocToXml(wDoc, abstractNumId);\r\n            Console.WriteLine(xml);\r\n            xml.Save(\"Out.xml\");\r\n        }\r\n        Console.WriteLine(\"Press Enter\");\r\n        Console.ReadKey();\r\n    }\r\n\r\n    private static XElement ConvertDocToXml(WordprocessingDocument wDoc, int abstractNumId)\r\n    {\r\n        XDocument xd = wDoc.MainDocumentPart.GetXDocument();\r\n\r\n        \/\/ First, call RetrieveListItem so that all paragraphs are initialized with ListItemInfo\r\n        var firstParagraph = xd.Descendants(W.p).FirstOrDefault();\r\n        var listItem = ListItemRetriever.RetrieveListItem(wDoc, firstParagraph);\r\n\r\n        XElement xml = new XElement(\"Root\");\r\n        var current = new Stack<XmlStackItem>();\r\n        current.Push(\r\n            new XmlStackItem()\r\n            {\r\n                Element = xml,\r\n                LevelNumbers = new int[] { },\r\n            });\r\n        foreach (var paragraph in xd.Descendants(W.p))\r\n        {\r\n            \/\/ The following does not take into account documents that have tracked revisions.\r\n            \/\/ As necessary, call RevisionAccepter.AcceptRevisions before converting to XML.\r\n            var text = paragraph.Descendants(W.t).Select(t => (string)t).StringConcatenate();\r\n            ListItemRetriever.ListItemInfo lii = \r\n                paragraph.Annotation<ListItemRetriever.ListItemInfo>();\r\n            if (lii.IsListItem && lii.AbstractNumId == abstractNumId)\r\n            {\r\n                ListItemRetriever.LevelNumbers levelNums = \r\n                    paragraph.Annotation<ListItemRetriever.LevelNumbers>();\r\n                if (levelNums.LevelNumbersArray.Length == current.Peek().LevelNumbers.Length)\r\n                {\r\n                    current.Pop();\r\n                    var levelNumsForThisIndent = levelNums.LevelNumbersArray;\r\n                    string levelText = levelNums\r\n                        .LevelNumbersArray\r\n                        .Select(l => l.ToString() + \".\")\r\n                        .StringConcatenate()\r\n                        .TrimEnd('.');\r\n                    var newCurrentElement = new XElement(\"Indent\",\r\n                        new XAttribute(\"Level\", levelText));\r\n                    current.Peek().Element.Add(newCurrentElement);\r\n                    current.Push(\r\n                        new XmlStackItem()\r\n                        {\r\n                            Element = newCurrentElement,\r\n                            LevelNumbers = levelNumsForThisIndent,\r\n                        });\r\n                    current.Peek().Element.Add(new XElement(\"Heading\", text));\r\n                }\r\n                else if (levelNums.LevelNumbersArray.Length > current.Peek().LevelNumbers.Length)\r\n                {\r\n                    for (int i = current.Peek().LevelNumbers.Length; \r\n                        i < levelNums.LevelNumbersArray.Length; \r\n                        i++)\r\n                    {\r\n                        var levelNumsForThisIndent = levelNums\r\n                            .LevelNumbersArray\r\n                            .Take(i + 1)\r\n                            .ToArray();\r\n                        string levelText = levelNums\r\n                            .LevelNumbersArray\r\n                            .Select(l => l.ToString() + \".\")\r\n                            .StringConcatenate()\r\n                            .TrimEnd('.');\r\n                        var newCurrentElement = new XElement(\"Indent\",\r\n                            new XAttribute(\"Level\", levelText));\r\n                        current.Peek().Element.Add(newCurrentElement);\r\n                        current.Push(\r\n                            new XmlStackItem()\r\n                            {\r\n                                Element = newCurrentElement,\r\n                                LevelNumbers = levelNumsForThisIndent,\r\n                            });\r\n                        current.Peek().Element.Add(new XElement(\"Heading\", text));\r\n                    }\r\n                }\r\n                else if (levelNums.LevelNumbersArray.Length < current.Peek().LevelNumbers.Length)\r\n                {\r\n                    for (int i = current.Peek().LevelNumbers.Length;\r\n                        i > levelNums.LevelNumbersArray.Length; \r\n                        i--)\r\n                        current.Pop();\r\n                    current.Pop();\r\n                    var levelNumsForThisIndent = levelNums.LevelNumbersArray;\r\n                    string levelText = levelNums\r\n                        .LevelNumbersArray\r\n                        .Select(l => l.ToString() + \".\")\r\n                        .StringConcatenate()\r\n                        .TrimEnd('.');\r\n                    var newCurrentElement = new XElement(\"Indent\",\r\n                        new XAttribute(\"Level\", levelText));\r\n                    current.Peek().Element.Add(newCurrentElement);\r\n                    current.Push(\r\n                        new XmlStackItem()\r\n                        {\r\n                            Element = newCurrentElement,\r\n                            LevelNumbers = levelNumsForThisIndent,\r\n                        });\r\n                    current.Peek().Element.Add(new XElement(\"Heading\", text));\r\n                }\r\n            }\r\n            else\r\n            {\r\n                current.Peek().Element.Add(new XElement(\"Paragraph\", text));\r\n            }\r\n        }\r\n        return xml;\r\n    }\r\n}<\/pre>\n
This example of the use of the ListItemRetriever module is important enough that I’ve incorporated it as one of the examples that I deliver as part of the Open XML PowerTools core examples.<\/p>\n
One note about the above code – I didn’t write it in the pure functional style.  I write it as procedural code that uses a stack.  Previously, I had posted some LINQ code that took a recursive, functional approach<\/a>.  I am not sure which is more approachable for less experienced developers, but I suspect that the procedural approach that uses a stack might be easier.<\/p>\n
This code isn’t data driven – it produces a specific XML structure that is hard coded into the example.  It could be possible to convert this code to another more flexible form that could produce a variety of XML shapes.  The code is only about 150 lines long – not too complicated, so it should be easy for developers to morph the code into a form that fits their scenario.<\/p>\n
Cheers, Eric<\/p>\n","protected":false},"excerpt":{"rendered":"
The new ListItemRetriever.cs module in PowerTools for Open XML enables us to find out lots of information about the numbered and bulleted lists in our documents.  We can write a small amount of code and transform a DOCX that contains hierarchical numbered lists into an alternative form of XML.  For instance, we may want to […]<\/p>\n","protected":false},"author":10567,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_bbp_topic_count":0,"_bbp_reply_count":0,"_bbp_total_topic_count":0,"_bbp_total_reply_count":0,"_bbp_voice_count":0,"_bbp_anonymous_reply_count":0,"_bbp_topic_count_hidden":0,"_bbp_reply_count_hidden":0,"_bbp_forum_subforum_count":0,"_s2mail":"","footnotes":""},"class_list":["post-2777","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.ericwhite.com\/blog\/wp-json\/wp\/v2\/pages\/2777","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.ericwhite.com\/blog\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.ericwhite.com\/blog\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.ericwhite.com\/blog\/wp-json\/wp\/v2\/users\/10567"}],"replies":[{"embeddable":true,"href":"https:\/\/www.ericwhite.com\/blog\/wp-json\/wp\/v2\/comments?post=2777"}],"version-history":[{"count":5,"href":"https:\/\/www.ericwhite.com\/blog\/wp-json\/wp\/v2\/pages\/2777\/revisions"}],"predecessor-version":[{"id":3102,"href":"https:\/\/www.ericwhite.com\/blog\/wp-json\/wp\/v2\/pages\/2777\/revisions\/3102"}],"wp:attachment":[{"href":"https:\/\/www.ericwhite.com\/blog\/wp-json\/wp\/v2\/media?parent=2777"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}