Re: [PATCH 1 of 2] follow: Fix args for $.post() call in toggleFollowingRepo()

2020-02-09 Thread Mads Kiilerich

On 2/8/20 9:21 PM, Ross Thomas wrote:

On 2/8/20 6:05 AM, Mads Kiilerich wrote:

Thanks. I do agree this is better. But is it fixing an actual problem?
The old manual encoding seems to work fine too?

Yep, it worked but it was mostly fortunate 'luck'.  It ended up using
matrix mode instead of true query params.



Interesting. I had never heard about "matrix mode". 
https://www.w3.org/DesignIssues/MatrixURIs.html . That idea was 
stillborn. Quite unfortunate that urllib.parse implement it without any 
documentation or code comments. I guess that will bite some users, and 
perhaps in some cases cause security issues in some applications.


Thanks for the patch.

/Mads


___
kallithea-general mailing list
kallithea-general@sfconservancy.org
https://lists.sfconservancy.org/mailman/listinfo/kallithea-general


Re: [PATCH 1 of 2] follow: Fix args for $.post() call in toggleFollowingRepo()

2020-02-08 Thread Ross Thomas
On 2/8/20 6:05 AM, Mads Kiilerich wrote:
> Thanks. I do agree this is better. But is it fixing an actual problem? 
> The old manual encoding seems to work fine too?

Yep, it worked but it was mostly fortunate 'luck'.  It ended up using
matrix mode instead of true query params.

The resulting query string was _literally_:
   ?follows_repository_id=xxx_session_csrf_secret_token=yyy

which was then processed as:
   param:  follows_repository_id = xxx
   param:  amp =
   matrix: _session_csrf_secret_token = yyy

If you look at the network call in the browser's dev tools you'll see
that the request payload consists of two items:
   param: follows_repository_id = xxx
   param: amp;_session_csrf_secret_token = yyy

So, yes, it will continue to work.  This just seemed cleaner.

Ross.
___
kallithea-general mailing list
kallithea-general@sfconservancy.org
https://lists.sfconservancy.org/mailman/listinfo/kallithea-general


Re: [PATCH 1 of 2] follow: Fix args for $.post() call in toggleFollowingRepo()

2020-02-08 Thread Mads Kiilerich

On 2/8/20 2:06 AM, Ross Thomas wrote:

# HG changeset patch
# User Ross Thomas 
# Date 1581112527 28800
#  Fri Feb 07 13:55:27 2020 -0800
# Branch stable
# Node ID f273a7b53fe035c12a55b6a3e34f849af245d797
# Parent  2571674c4a2fd6247a6dee7d9b4e27df13df1429
follow: Fix args for $.post() call in toggleFollowingRepo()



Thanks. I do agree this is better. But is it fixing an actual problem? 
The old manual encoding seems to work fine too?


/Mads



diff -r 2571674c4a2f -r f273a7b53fe0 kallithea/public/js/base.js
--- a/kallithea/public/js/base.js   Mon Feb 03 20:50:07 2020 +0100
+++ b/kallithea/public/js/base.js   Fri Feb 07 13:55:27 2020 -0800
@@ -457,8 +457,10 @@
  }
  
  var toggleFollowingRepo = function(target, follows_repository_id){

-var args = 'follows_repository_id=' + follows_repository_id;
-args += '_session_csrf_secret_token=' + _session_csrf_secret_token;
+var args = {
+'follows_repository_id': follows_repository_id,
+'_session_csrf_secret_token': _session_csrf_secret_token
+}
  $.post(TOGGLE_FOLLOW_URL, args, function(data){
  _onSuccessFollow(target);
  });
___
kallithea-general mailing list
kallithea-general@sfconservancy.org
https://lists.sfconservancy.org/mailman/listinfo/kallithea-general



___
kallithea-general mailing list
kallithea-general@sfconservancy.org
https://lists.sfconservancy.org/mailman/listinfo/kallithea-general