PsAccelerator: tweet2word

This time I will create a word document (open xml) in F# without using the Open XML SDK and using XML libraries. Things have been intentionally kept simple.

I start by modelling a simple word document:

There are Two types of paragraphs: normal and bold ones:

 //reference to the package 
#r "WindowsBase"
 
open System
open System.IO
open System.IO.Packaging
 
type Paragraph =
    | Normal of string
    | Bold of string
 

And a document is a list of paragraphs:

 type Document =
    | Paragraphs of Paragraph list
 

This is the test:

 //test
let list = [Normal("test1"); Normal("test2"); Bold("test3"); Normal("test4")]
let doc = Paragraphs(list)
 

Next we have the to create an xml version of document:

//helpers

// http://weblogs.sqlteam.com/mladenp/archive/2008/10/21/Different-ways-how-to-escape-an-XML-string-in-C.aspx

let encodedXml (text: string) =

    text.Replace("&", "&amp;").Replace("<", "&lt;").Replace(">", "&gt;").Replace("\"", "&quot;").Replace("'", "&apos;")

let xml1 = "<node>it's my \"node\" & i like it<node>"

let encode = encodedXml xml1

let addString acc item = acc + "\r\t" + item

let createParagraphNormal text =

    "<w:p>

      <w:r>

        <w:t>" + encodedXml text + 

        "</w:t>

      </w:r>

    </w:p>"

let createParagraphBold text =

    "<w:p>

      <w:r>

        <w:rPr>

         <w:b />

        </w:rPr>

        <w:t>" + encodedXml text + 

        "</w:t>

      </w:r>

    </w:p>"

let createParagraph = function

    | Normal(text) -> createParagraphNormal text

    | Bold (text) -> createParagraphBold text

let createParagraphs = function

    | Paragraphs(list) -> list|> List.map createParagraph |> List.reduce addString

//test

let p = createParagraphs doc

let createDocument doc =

    let startDocument = @"<?xml version=""1.0"" encoding=""utf-8""?>

                    <w:document xmlns:w=""http://schemas.openxmlformats.org/wordprocessingml/2006/main"">

                      <w:body>"

    let endDocument = "  </w:body>

                        </w:document>"

    let content = createParagraphs doc

    startDocument + content + endDocument

Then we have to create a package, add the document and save it to disk:

//add the document to package and save

let createFile doc (fileName:string) =

    using (Package.Open(fileName, FileMode.Create, FileAccess.ReadWrite))(fun package ->

        let uri = new Uri("/word/document.xml", UriKind.Relative )

        let partDocumentXML = 

            package.CreatePart(

                                uri,

                                "application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml" )

        using(new StreamWriter(partDocumentXML.GetStream(FileMode.Create, FileAccess.Write)))(fun stream ->

            doc |> createDocument |> stream.Write

        )

        package.CreateRelationship(

                                    uri,

                                    TargetMode.Internal,

                                    "http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument",

                                    "rId1") |> ignore

    )

//test

let fileName = @"D:\Tmp\test.docx"

createFile doc fileName;;

This is the result:

I have created a map from twitter to my document model

type UserStatus = 

    { UserName : string; 

      ProfileImage : string; 

      Status : string; 

      StatusDate : DateTime }

let mapUserStatus (us:UserStatus) =

    let date = "Date: " + us.StatusDate.ToShortDateString()  + " time: " + us.StatusDate.ToLongTimeString()

    [Normal(date); Bold(us.UserName); Normal(us.Status); Normal("")]

let addUserStatus acc item = item @ acc

let result = xml|> parseTweetXml |> List.map mapUserStatus|> List.reduce addUserStatus