[Haskell-cafe] Converting wiki pages into pdf

[mukesh tiwari]

Hello all
I am trying to write a Haskell program which download html pages from
wikipedia including images and convert them into pdf . I wrote a
small script

import Network.HTTP
import Data.Maybe
import Data.List

main = do
x <- getLine
htmlpage <- getResponseBody =<< simpleHTTP ( getRequest x ) --
open url
--print.words $ htmlpage
let ind_1 = fromJust . ( \n -> findIndex ( n `isPrefixOf`) .
tails $ htmlpage ) $ "<!-- content -->"
ind_2 = fromJust . ( \n -> findIndex ( n `isPrefixOf`) .
tails $ htmlpage ) $ "<!-- /content -->"
tmphtml = drop ind_1 $ take ind_2 htmlpage
writeFile "down.html" tmphtml

and its working fine except some symbols are not rendering as it
should be. Could some one please suggest me how to accomplish this
task.

Thank you
Mukesh Tiwari

_______________________________________________
Haskell-Cafe mailing list
Haskel...@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe

[Max Rabkin]

This doesn't answer your Haskell question, but Wikpedia has
PDF-generation facilities ("Books"). Take a look at
http://en.wikipedia.org/wiki/Help:Book (for single articles, just use
the "download PDF" option in the sidebar).

--Max

[mukesh tiwari]

Is it possible to automate this process rather than manually clicking and downloading  using Haskell ?

Thank You
Mukesh Tiwari

[mukesh tiwari]

I tried to use the PDF-generation facilities . I wrote a script which
generates the rendering url . When i am pasting rendering url in
browser its generating the download file but when i am trying to get
the tags , its empty. Could some one please tell me what is wrong with
code.
Thank You
Mukesh Tiwari

import Network.HTTP
import Text.HTML.TagSoup
import Data.Maybe

parseHelp :: Tag String -> Maybe String
parseHelp ( TagOpen _ y ) = if ( filter ( \( a , b ) -> b == "Download
a PDF version of this wiki page" ) y ) /= []
then Just $ "http://en.wikipedia.org" ++ ( snd $
y !! 0 )
else Nothing

parse :: [ Tag String ] -> Maybe String
parse [] = Nothing
parse ( x : xs )
| isTagOpen x = case parseHelp x of
Just s -> Just s
Nothing -> parse xs
| otherwise = parse xs

main = do
x <- getLine

tags_1 <- fmap parseTags $ getResponseBody =<< simpleHTTP

( getRequest x ) --open url

let lst = head . sections ( ~== "<div class=portal id=p-coll-
print_export>" ) $ tags_1
url = fromJust . parse $ lst --rendering url
putStrLn url
tags_2 <- fmap parseTags $ getResponseBody =<< simpleHTTP
( getRequest url )
print tags_2

[Daniel Patterson]

It looks to me that the link is generated by javascript, so unless you can script an actual browser into the loop, it may not be a viable approach.

[mukesh tiwari]

Thank your for reply Daniel. Considering my limited knowledge of web programming and javascript , first i need to simulated the some sort of browser in my program which will run the javascript and will generate the pdf. After that i can download the pdf . Is this you mean ?  Is Network.Browser any helpful for this purpose ? Is there  way to solve this problem ? 
Sorry for  many questions but this  is my first web application program and i am trying hard to finish it. 

Thank you

Mukesh Tiwari

[Conrad Parker]

On Sep 9, 2011 7:33 AM, "mukesh tiwari" <mukeshtiw...@gmail.com> wrote:
>
> Thank your for reply Daniel. Considering my limited knowledge of web programming and javascript , first i need to simulated the some sort of browser in my program which will run the javascript and will generate the pdf. After that i can download the pdf . Is this you mean ?  Is Network.Browser any helpful for this purpose ? Is there  way to solve this problem ? 
> Sorry for  many questions but this  is my first web application program and i am trying hard to finish it. 
>

Have you tried finding out if simple URLs exist for this, that don't require Javascript? Does Wikipedia have a policy on this?

Conrad.

[Kyle Murphy]

It's worth pointing out at this point (as alluded to by Conrad) that what you're attempting might be considered somewhat rude, and possibly slightly illegal (depending on the insanity of the legal system in question). Automated site scraping (what you're essentially doing) is generally frowned upon by most hosts unless it follows some very specific guidelines, usually at a minimum respecting the restrictions specified in the robots.txt file contained in the domains root. Furthermore, depending on the type of data in question, and if a EULA was agreed to if the site requires an account, doing any kind of automated processing might be disallowed. Now, I think wikipedia has a fairly lenient policy, or at least I hope it does considering it's community driven, but depending on how much of wikipedia you're planning on crawling you might at the very least consider severly throttling the process to keep from sucking up too much bandwidth.

On the topic of how to actually perform that crawl, you should probably check out the format of the link provided in the download PDF element. After looking at an article (note, I'm basing this off a quick glance at a single page) it looks like you should be able to modify the URL provided in the "Permanent link" element to generate the PDF link by changing the title argument to arttitle, adding a new title argument with the value "Special:Book", and adding the new arguments "bookcmd=render_article" and "writer=rl". For example if the permanent link to the article is:

http://en.wikipedia.org/w/index.php?title=Shapinsay&oldid=449266269

Then the PDF URL is:

http://en.wikipedia.org/w/index.php?arttitle=Shapinsay&oldid=449266269&title=Special:Book&bookcmd=render_article&write=rl

This is all rather hacky as well, and none of it has been tested so it might not actually work, although I see no reason why it shouldn't. It's also fragile, as if wikipedia changes just about anything it could all brake, but that's the risk you run anytime you resort of site scraping.

-R. Kyle Murphy
--
Curiosity was framed, Ignorance killed the cat.

[Matti Oinas]

The whole wikipedia database can also be downloaded if that is any help.

http://en.wikipedia.org/wiki/Wikipedia:Database_download

There is also text in that site saying "Please do not use a web
crawler to download large numbers of articles. Aggressive crawling of
the server can cause a dramatic slow-down of Wikipedia."

Matti

2011/9/9 Kyle Murphy <orc...@gmail.com>:

--
/*******************************************************************/

try {
   log.trace("Id=" + request.getUser().getId() + " accesses " +
manager.getPage().getUrl().toString())
} catch(NullPointerException e) {}

/*******************************************************************/

This is a real code, but please make the world a bit better place and
don’t do it, ever.

* http://www.javacodegeeks.com/2011/01/10-tips-proper-application-logging.html *

[mukesh tiwari]

Thank you all for replying. I managed to write a python script. It depends on PyQt4 . I am curious if we have any thing like PyQt4  in Haskell.

import sys
from PyQt4.QtCore import *
from PyQt4.QtGui import *
from PyQt4.QtWebKit import *

#http://www.rkblog.rk.edu.pl/w/p/webkit-pyqt-rendering-web-pages/
#http://pastebin.com/xunfQ959
#http://bharatikunal.wordpress.com/2010/01/31/converting-html-to-pdf-with-python-and-qt/
#http://www.riverbankcomputing.com/pipermail/pyqt/2009-January/021592.html

def convertFile( ):
                web.print_( printer )
                print "done"
                QApplication.exit()


if __name__=="__main__":
        url = raw_input("enter url:")
        filename = raw_input("enter file name:")
        app = QApplication( sys.argv )
        web = QWebView()
        web.load(QUrl( url ))
        #web.show()
        printer = QPrinter( QPrinter.HighResolution )
        printer.setPageSize( QPrinter.A4 )
        printer.setOutputFormat( QPrinter.PdfFormat )
        printer.setOutputFileName(  filename + ".pdf" )
        QObject.connect( web ,  SIGNAL("loadFinished(bool)"), convertFile  )
        sys.exit(app.exec_())

[Michael Snoyman]

On Fri, Sep 9, 2011 at 3:16 PM, mukesh tiwari

I've actually used wkhtmltopdf[1] for this kind of stuff in the past.

[1] http://code.google.com/p/wkhtmltopdf/