# Html method on watir don't return the real html

**URL:** <https://www.ruby-forum.com/t/html-method-on-watir-dont-return-the-real-html/168250>\
**Category:** Ruby\
**Created:** [May 18, 2009, 7:17pm UTC](https://www.ruby-forum.com/t/html-method-on-watir-dont-return-the-real-html/168250 "2009-05-18T19:17:16Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![tcblues](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/db5fbb/32.png) [@tcblues](https://www.ruby-forum.com/u/tcblues)\
**Post date:** [May 18, 2009, 7:17pm UTC](https://www.ruby-forum.com/t/html-method-on-watir-dont-return-the-real-html/168250/1 "2009-05-18T19:17:16Z")

</div>

Hi,  
I’m using watir to get the html of a page in order to verify every  
single page on [w3c.org](http://w3c.org) but what I get with watir is something like:

Certus Games ....

And the real content is:

\<?xml version="1.0" encoding="UTF-8" ?\> ....

Any idea how to get the real html content?

thank you.

---

<div class="post-metadata">

**Author:** ![tcblues](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/db5fbb/32.png) [@tcblues](https://www.ruby-forum.com/u/tcblues)\
**Post date:** [May 20, 2009, 3:55pm UTC](https://www.ruby-forum.com/t/html-method-on-watir-dont-return-the-real-html/168250/2 "2009-05-20T15:55:00Z")

</div>

Since I think it’s impossible to get the real html I’m trying to do it  
using net/http library.

In my case I need first to login the user onto the webpage and then go  
to the pages I want to get the html.  
This is what I’m doing… but it seems to me the user can’t login:

```
require 'net/http'
require 'activesupport'
server="myserver.com"
port=80

http = Net::HTTP.new(server, port)
http.start()

data={":loginName=>"myLoginName",
  :password=>"myPassord"
}

resp,data=http.post("/login.do",data.to_query)
if resp.code=="200" then
  headers={
    "Cookie"=>resp["set-cookie"].to_s()
  }
end
resp,data=http.get('/MyAccount.do',headers)
html=data
http.finish()

```

Any idea???

Thanks.

Mario R. wrote:

> Hi,  
> I’m using watir to get the html of a page in order to verify every  
> single page on [w3c.org](http://w3c.org) but what I get with watir is something like:
> 
> Certus Games

---

<div class="post-metadata">

**Author:** ![tcblues](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/db5fbb/32.png) [@tcblues](https://www.ruby-forum.com/u/tcblues)\
**Post date:** [May 21, 2009, 9:36am UTC](https://www.ruby-forum.com/t/html-method-on-watir-dont-return-the-real-html/168250/3 "2009-05-21T09:36:25Z")

</div>

Hi,

2009/5/19 Mario R. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid):

> rel=stylesheet\>\<LINK media=print href=“/CF/css/print.css” type=text/css
> 
> Any idea how to get the real html content?

If you are using Ruby 1.8.6 on Windows, try this code:

require ‘win32/api’  
require ‘watir’

IID\_IPersistStreamInit =  
[0x7FD52380,0x4E07,0x101B,0xAE,0x2D,0x08,0x00,0x2B,0x2E,0xC7,0x13].pack(‘LSSC8’)

CreateStreamOnHGlobal = Win32::API.new(‘CreateStreamOnHGlobal’, ‘LLP’,  
‘L’,‘ole32’)  
OleSaveToStream = Win32::API.new(‘OleSaveToStream’, ‘LL’, ‘L’,‘ole32’)  
GetHGlobalFromStream = Win32::API.new(‘GetHGlobalFromStream’, ‘LP’,  
‘L’,‘ole32’)  
GlobalSize = Win32::API.new(‘GlobalSize’, ‘L’, ‘L’)  
GlobalLock = Win32::API.new(‘GlobalLock’, ‘L’, ‘L’)  
CopyMemory = Win32::API.new(‘RtlMoveMemory’, ‘PLL’, ‘V’)  
GlobalUnlock = Win32::API.new(‘GlobalUnlock’, ‘L’, ‘L’)

browser = Watir::Browser.new  
browser.goto(“[http://www.google.com](http://www.google.com)”)

ptr = browser.document.inspect.scan(/:(0x[\da-f]+)?\>/).to\_s.hex  
p = 0.chr \* 4  
CopyMemory.call(p,ptr+16,4)  
data = p.unpack(‘L’).first  
p = 0.chr \* 4  
CopyMemory.call(p,data,4)  
dispatch = p.unpack(‘L’).first

lpVtbl = 0.chr \* 4  
table = 0.chr \* 28  
CopyMemory.call(lpVtbl,dispatch,4)  
CopyMemory.call(table,lpVtbl.unpack(‘L’).first,28)  
table = table.unpack(‘L\*’)  
queryInterface = Win32::API::Function.new(table[0],‘PPP’,‘L’)  
p = 0.chr \* 4  
hr = queryInterface.call(dispatch,IID\_IPersistStreamInit,p)  
persiststream = p.unpack(‘L’).first

p = 0.chr \* 4  
CreateStreamOnHGlobal.call(0, 1, p)  
stream = p.unpack(‘L’).first  
if OleSaveToStream.call(persiststream,stream) == 0  
p = 0.chr \* 4  
GetHGlobalFromStream.call(stream,p)  
h = p.unpack(‘L’).first  
size = GlobalSize.call(h)  
buf = 0.chr \* size  
ptr = GlobalLock.call(h)  
if ptr != 0  
CopyMemory.call(buf,ptr,size)  
GlobalUnlock.call(ptr)  
html = buf[16…-1]  
else  
html = nil  
end  
end

puts html

Regards,  
Park H.

---

<div class="post-metadata">

**Author:** ![tcblues](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/db5fbb/32.png) [@tcblues](https://www.ruby-forum.com/u/tcblues)\
**Post date:** [May 25, 2009, 1:05pm UTC](https://www.ruby-forum.com/t/html-method-on-watir-dont-return-the-real-html/168250/4 "2009-05-25T13:05:32Z")

</div>

It’s working!!! Thanks a lot.

Since I’m not using the last Watir version I changed Watir::Browser.new  
for Watir::IE.new and it’s working fine.

Again… thanks a lot.

Heesob P. wrote:

> Hi,
> 
> 2009/5/19 Mario R. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid):
> 
> > rel=stylesheet\>\<LINK media=print href=“/CF/css/print.css” type=text/css
> > 
> > Any idea how to get the real html content?