Last active
November 10, 2016 18:34
-
-
Save SecretDeveloper/e0ec251820d41f8f06a462a3c109e756 to your computer and use it in GitHub Desktop.
unsong.fs
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| // Create epub file from unsongbook.com and write to console. | |
| open FSharp.Data | |
| open System.Collections.Generic | |
| type Chapter(title:string, body:string, url:string)= | |
| member x.Title = title | |
| member x.Body = body | |
| member x.URL = url | |
| let chapters = new List<Chapter>() | |
| let slug = fun(value:string)-> | |
| value.Replace(" ", "_").Replace("\n","").Replace(":","").Replace("?","").Replace("'","").Replace("#","").Replace("=","").Replace(",","") | |
| let rec gatherPages = fun (url : string) -> | |
| let page = HtmlDocument.Load(url) | |
| let content = page.CssSelect("div#pjgm-content") | |
| let title = content.CssSelect("h1.pjgm-posttitle").[0].InnerText() | |
| let body = | |
| content.CssSelect("div.pjgm-postcontent").[0].Elements() | |
| |> Seq.filter(fun (t:HtmlNode) -> t.HasClass("sharedaddy")=false) | |
| |> Seq.map (fun s -> s.ToString()) | |
| |> Seq.reduce (fun acc item -> acc + " " + item) | |
| chapters.Add(new Chapter(title=title, body=body, url=url)) | |
| let nextUrl = content.CssSelect("a[rel|=next]").[0].AttributeValue("href").ToString() | |
| gatherPages(nextUrl) | |
| [<EntryPoint>] | |
| let main argv = | |
| gatherPages "http://unsongbook.com/prologue-2/" | |
| printfn "<html><head><title>Unsong book</title></head><body>" | |
| //TOC | |
| printfn "<h1>Table of Contents</h1><ol>" | |
| for chap in chapters do | |
| let s = slug chap.Title | |
| printfn "<li><h2><a href='#%s'>%s</a></h2></li>\n" s chap.Title | |
| printfn "</ol>" | |
| //CONTENT | |
| for chap in chapters do | |
| let s = slug chap.Title | |
| printfn "<div><h2 id='%s'>%s</h2>\n<div>%s</div>\n<div>\n" s chap.Title chap.Body | |
| printfn "</body></html>" | |
| 0 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment