• Josh (16/16) Feb 10 2019 I have a website (say https://website.com) that I need to log in
• Anonymouse (16/19) Feb 10 2019 Assuming I'm understanding your question (you want to get the

Josh <moonburntm gmail.com> writes:

I have a website (say https://website.com) that I need to log in 
to, it will do a few 302 redirects and then I will end up with a 
unique session ID in the URL (such as 
https://website.com/welcome.html?s=636853677441448706). Is there 
some way of extracting this ID (I'll need it later for other 
things)?

I'm using std.net.curl's HTTP, and when I turn verbose on and do 
the login POST I can see that it is in fact redirecting to the 
URL with the session ID. I can see that etc.c.curl has 
CurlInfo.effective_url and CurlInfo.redirect_url that I can use 
with curl_easy_getinfo, but that needs a CURL*. 
HTTP.handle.handle is private, so I can't get a CURL* out of the 
Curl struct.

Is there something I'm missing? It it possible to access the 
private handle? Is there a better way to do what I'm trying to 
achieve?

Anonymouse <asdf asdf.net> writes:

On Sunday, 10 February 2019 at 14:12:02 UTC, Josh wrote:
 Is there something I'm missing? It it possible to access the 
 private handle? Is there a better way to do what I'm trying to 
 achieve?

Assuming I'm understanding your question (you want to get the 
path of a URL after redirects), if you're not adverse to 
dependencies you could use requests; 
https://code.dlang.org/packages/requests.

Create a Request, query an URL for a Response and then access its 
finalURI member. Then parse the path string normally to get the 
session ID, such as with std.algorithm.findSplit and friends or 
with regex.

import requests;

Request req;
Response res = req.get("https://website.com/");

string id = res.finalURI.path
     .findSplit("?s=")[2]
     .findSplitBefore("&")[0];

if (!id.length) { /* no ID in URL */ }