OpenEuphoria

2. Re: Testing

• Posted by ChrisB (moderator) Oct 01, 2008
• 1338 views

No, everyones left euphoria, and gone to program in python

Chris

4. Re: Testing

• Posted by jacquesd Oct 01, 2008
• 1303 views
• Last edited Oct 02, 2008

Strange! opera is the only browser that let me post to euforum!
ie7 and firefox 3.0.2 just send me back to the list of messages after the 5 seconds delay
No error messages.

jacques

6. Re: Testing

• Posted by DerekParnell (admin) Oct 01, 2008
• 1331 views
• Last edited Oct 02, 2008

Replying using Safari (WinXP)

8. Re: Testing

• Posted by DerekParnell (admin) Oct 01, 2008
• 1374 views
• Last edited Oct 02, 2008

Replying using IE7 v7.0.5730.11

10. Re: Testing

• Posted by jacquesd Oct 01, 2008
• 1304 views
• Last edited Oct 02, 2008

I tried on 3 different computers 2 of them in windows xp pro sp3
ie 7.0.5730.11 same problem on both
firefox 3.0.2. same problem on both
opera 9.50 OK
on 3th laptop with zenwalk linux 5.0 and iceweisel (a renamed firefox) it fail too.

Conclusion Opera have cursed all others browsers on my computers. :)

12. Re: Testing

• Posted by jemima Oct 02, 2008
• 1283 views

It may be that you /MUST/ check the "remember me forever" box when logging on?

14. Re: Testing

• Posted by bernie Oct 02, 2008
• 1270 views

The problem is that a user has to click remember me forever. 
Please put the login back the way it was I don't like storing 
any password on my system or on any host at any time. 

16. Re: Testing

• Posted by DanM Oct 03, 2008
• 1283 views

Finally I can login! ! ! But only if I check the "Remember Me Forever" box EACH TIME I log in!

So apparently now we HAVE to check, and re-check EVERY TIME, the "Remember Me Forever" box in order to login????

Sorry, but this seems stupid! !

In the past, I did check "Remember Me Forever", once, and my password auto-filled in just fine as soon as I hit Enter on keyboard, every time thereafter, without having to re-check that box.

Except now.

What's the point of saying "Forever", if you got to check that box EVERY TIME? ? ? That just doesn't seem to make sense. It seems to end up meaning "remember me forever as long as I keep reminding you to remember me", which obviously doesn't make any sense.

Please excuse me if I have misunderstood something, but I've been trying to login for a couple of days, then left a "comment" on Jeremy's website about the problem, then tried to email him also (but the message was undeliverable), and it was only by reading Bernie's post that it dawned on me to re-check that "Remember Me Forever" box.

At the very least, if the method of logging in was changed, it would have made sense to tell us so, right?

Anyway, thank you Bernie for your post!

And Jeremy, could you please delete my "Comments" on your website?

Dan

18. Re: Testing

• Posted by DerekParnell (admin) Oct 03, 2008
• 1310 views

Its working fine for me. I do not have to check the "Forever" box. I logged in weeks ago and haven't had to log in again. And that is using two different computers and five different browsers.

It just works here.

20. Re: Testing

• Posted by DanM Oct 03, 2008
• 1261 views

Ok, but I just tried it again, and if I did NOT re-check the "Forever" box, I DIDN'T get logged in, but after I then did check it, I did get logged in.

And previously, logging in worked for me essentially as you describe (except I think you meant you didn't have to type in your password when you logged in, not that you didn't have to log in in order to be..logged in?)

So maybe it is a combination problem: some (most?) browsers still work fine, presumably even after some change in the forum software, but others (one?), now require re-checking the "Forever" box every time user wants to log in? ?

If anyone else is still having trouble logging in, you might try re-checking the "Remember Me Forever" box, & if that logs you in, please indicate that here in this thread (or start another one about log in problem?), & note what browser you're using .

I'm using IE 6.0 on WinXP sp2.

Dan

22. Re: Testing

• Posted by DanM Oct 03, 2008
• 1261 views

Ok, applogies to Derek, he probably meant the same thing, exactly as he said, but that doesn't change the fact that I used to be able to log in by just typing my account name & then hitting "log in", without entering my password, after once previously checking off "Remember Me Forever", & now I have to re-check the "Forever" button every time. I'm not sure it's a cookies issue, though I can't say for sure, but nonetheless, something has changed to occasion this difficulty.

So, has there been a change in the forum software that could be making this problem?

Dan

24. Re: Testing

• Posted by DanM Oct 03, 2008
• 1254 views

Um, you wouldn't by chance be not logging out would you? I tried that & it, as one might expect, does allow browsing to the forum & finding oneself "automagically" logged in, not a big surprise, but maybe not a good thing to do? ? Dunno.

Dan

26. Re: Testing

• Posted by euphoric (admin) Oct 03, 2008
• 1254 views

Of course I'm not logging out. I'm at my PC and nobody else uses it. Why would I log out anyway?

Sometimes I get logged out for whatever reason, but then I just log back in and stay logged in for forever (or until another browser glitch logs me out).

28. Re: Testing

• Posted by SDPringle Oct 03, 2008
• 1255 views
• Last edited Oct 04, 2008

I was able to get in by adding openeuphoria.org to "Trusted Sites." I still cannot login through Opera. Perhaps the problem is the kind of cookie asked for? I had setup the machines to ask me to confirm if I wanted each cookie. In that case I never got a request from the euphoria domain.

I think the routine add_cookie should be called with the domain set to the domain name: openeuphoria.org instead of '/'. The prototype is as follows:

export procedure add_cookie(sequence name, sequence value, object path=0, object expires=0, object domain=0)

The procedure is called with '/' supplied as the domain in users.e line 68.

Shawn Pringle

30. Re: Testing

• Posted by jacquesd Oct 04, 2008
• 1259 views

I will look at it this evening

jacques

32. Re: Testing

• Posted by bernie Oct 04, 2008
• 1227 views

Hi Jim and others:

I looked at the cookie on that eu deposited in my browser.
Even though I clicked remember me forever the cookie information says that it expires at end of session.

My browser is set to accept cookies and destroy them when I close my browser.

So the way things work with different browsers is dependent on your cookie control settings.
Bernie

34. Re: Testing

• Posted by jimcbrown (admin) Oct 04, 2008
• 1207 views
• Last edited Oct 05, 2008

 
 expire = dt:now_gmt() 
  if web:post_has("login_forever") then 
    expire = dt:add(expire,1,YEARS) 
  else 
    expire = dt:add(expire,1,HOURS) -- this return date prior du current date!! 
  end if 
 

With this simple test program, I couldn't reproduce the bug. I am at -0400, and at the time of writing here are my times:

$ date Sat Oct 4 20:11:51 EDT 2008 $ date utc Sun Oct 5 00:11:53 UTC 2008 $

include std/datetime.e 
 
object expire = now_gmt() 
? expire 
expire = add(expire, 1, HOURS) 
? expire 

$ exu dt.e {2008,10,5,0,9,26} {2008,10,5,1,9,26}

It is not clear to me why expire would be considered as being before the current date.

36. Re: Testing

• Posted by jimcbrown (admin) Oct 04, 2008
• 1227 views
• Last edited Oct 05, 2008

One possibility that occurs to me is that the server that the code is running on might be running slow... though it'd have to be off by more than an hour for the cookie to be rejected immediately by the browser.

38. Re: Testing

• Posted by DerekParnell (admin) Oct 04, 2008
• 1196 views
• Last edited Oct 05, 2008

I was also thinking that the changes to daylight saving might be an influence. This event is happening around the globe about now but at slightly different times in different places.

Maybe the logic should be to add 121 minutes to the "not forever" situation rather than 1 hour.

40. Re: Testing

• Posted by jacquesd Oct 04, 2008
• 1222 views
• Last edited Oct 05, 2008

 
 expire = dt:now_gmt() 
  if web:post_has("login_forever") then 
    expire = dt:add(expire,1,YEARS) 
  else 
    expire = dt:add(expire,1,HOURS) -- this return date prior du current date!! 
  end if 
 

include std/datetime.e 
 
object expire = now_gmt() 
? expire 
expire = add(expire, 1, HOURS) 
? expire 

This time I got it right. The problem in really in datetime.e. here's what I found in datetime.e

public sequence month_abbrs = { "Jan", "Feb", "Mar", "Apr", "May", "Jun", "Jul", "Aug", "Sep","Aug", "Sep", "Oct", "Nov", "Dec" } 

Now we know why the bug manifested just at the beginning of october.

jacques