How to read https webpage..

BlitzMax Forums/BlitzMax Programming/How to read https webpage..

Ifq200

(Posted 2013) [#1]

Hi, i have this example,


' opens a read stream to the blitzbasic.com website and
' dumps the homepage to the console using readline and print

in=ReadStream("http::blitzbasic.com")

If Not in RuntimeError "Failed to open a ReadStream to file http::www.blitzbasic.com"

While Not Eof(in)
	Print ReadLine(in)
Wend
CloseStream in

Is it possible to do the same thing with https ??

Brucey

(Posted 2013) [#2]

HTTPS works via the SSL/TLS protocol, and requires you to handle all of the security features that those provide - in order to get access to the pages.

A library such as libcurl built with OpenSSL support can do this for you.

Otherwise, you may want to look into interfacing with some SSL libraries and handling all the SSL certification functions yourself.

With libcurl you might do it like this:

Framework BaH.libcurlssl
Import BRL.StandardIO
Import BRL.FileSystem

Local curl:TCurlEasy = TCurlEasy.Create()

curl.setWriteString()
curl.setOptInt(CURLOPT_FOLLOWLOCATION, 1)

curl.setOptString(CURLOPT_URL, "https://some.web.page.com")

Local res:Int = curl.perform()

If res Then
	Print CurlError(res)
	End
End If

curl.cleanup()

Print curl.toString()

Ifq200

(Posted 2013) [#3]

Thanks for reply, i'll try

BlitzSupport

(Posted 2013) [#4]

Hi Brucey, that looks nice and simple, but do you know how the certificate stuff works? For any site I've tried (eg. "https://www.google.co.uk/"), the above code returns:

"Peer certificate cannot be authenticated with given CA certificates".

I tried replacing libcurlssl.mod/certificates/cacert.pem with this one, and even rebuilt the module just in case, but it still gives the same result.

(Interestingly, the linked cacert.pem is smaller, despite being automatically kept up to date. Not sure if that's to do with the Diginotar stuff being removed or what... )