Jump to content

Quoting in IPB 3.4, not satisifed


TSP

Recommended Posts

  • Management

Yes I know. I'm beating a dead horse like the rest of this community. It's satire. You're supposed to laugh. I don't expect a bug free product out of you (I was actually surprised at how long gallery 5.0.3 went without a report) but I do expect obvious ones that come out and slap you in the face to be resolved on a dev site before you push it to here. As we've said before though, that notion is wrong of me. Obviously.

At least I gave you an excuse to ignore the rest of the topic. :tongue:

EDIT: Actually, may I ask you one thing? When you fix a bug on your svn thing (sorry I don't know the details) do you actually go and subsequently check if the bug is actually fixed and check if any related functionality was unaffected? Or do you just say "Well, that looks right and seems ok. What's next?" I'm just wondering. You guys obviously do what's been working for you.

I didn't see it as humour. One has to be careful with such things because other people read that kind of thing and are led to believe that IP.Board is being released 'buggy'. When in reality, at the time of your post there were only 6 open bug reports. Which isn't a lot.

Dear Matt, how can you say this when you have bugs in STD mode that cuts posts? (Yes, I do realize you haven't actually relased 3.4, but still...)

...


And there we have another bug that must've materialized somehow when I switched mode. Gets better and better. Will have to fix my post now...


edit: post fixed, you can view how it ended up originally here:

http://community.invisionpower.com/topic/373132-post-code-from-editor-bug/#entry2335355

As per the bug report you've been posting in, this has hopefully been fixed.

Link to comment
Share on other sites

  • Replies 112
  • Created
  • Last Reply

EDIT: Actually, may I ask you one thing? When you fix a bug on your svn thing (sorry I don't know the details) do you actually go and subsequently check if the bug is actually fixed and check if any related functionality was unaffected? Or do you just say "Well, that looks right and seems ok. What's next?" I'm just wondering. You guys obviously do what's been working for you.

This is a VERY good question that I think should be answered (I hope). Does IPB regression test changes to the code base after fixes are made? A standard set of functional tests of key portions of the board to ensure there are no undesired consequences. This should be standard practice before you commit a change back via svn (or what ever version control is being used).

Link to comment
Share on other sites

  • Management

Of course we test the fix. Then we release betas to further test the fix and then we update our forum here to get even further testing.

For code as complex and entwined as IP.Board, it's almost impossible to design cookie-cutter test models and regression test models.

Link to comment
Share on other sites



I didn't see it as humour. One has to be careful with such things because other people read that kind of thing and are led to believe that IP.Board is being released 'buggy'. When in reality, at the time of your post there were only 6 open bug reports. Which isn't a lot.


As per the bug report you've been posting in, this has hopefully been fixed.

Ok well I apologize for that. Of course I don't think you would ever intentionally release a buggy product, I just can't help but feel you can improve a little on pretesting. Call it stubborn if you want.
Link to comment
Share on other sites

  • Management

We're very open to suggestion.

As mentioned above, we currently:

- Test fixes / new features as they are written

- Build for an internal testing board for wider staff testing

- Build copies for an internal QA team

- Build copies for customer beta testing

- Upgrade our company forums (often with 1,000 online users, over 1,000,000 posts and lots of activity)

We go through this process perhaps a dozen times with each major release.

If you have any ideas on how to improve this, please let me know. I've worked in software development for about 12 years now and I think we do pretty well.

Link to comment
Share on other sites

We're very open to suggestion.

As mentioned above, we currently:

- Test fixes / new features as they are written
- Build for an internal testing board for wider staff testing
- Build copies for an internal QA team
- Build copies for customer beta testing
- Upgrade our company forums (often with 1,000 online users, over 1,000,000 posts and lots of activity)

We go through this process perhaps a dozen times with each major release.

If you have any ideas on how to improve this, please let me know. I've worked in software development for about 12 years now and I think we do pretty well.

I already gave you my idea. It was pretty simple.

If you make a change in the editor to fix a bug then make sure that the bug is fixed also in PMs, comments, and that it didn't have any adverse affects on submitting content via the mobile skin. I believe that's a very simple step you can add to the testing of bugs as you fix them. Additionally you should have an internal list of special characters that have historically given you trouble. For example I have seen it happen where you will fix a bug with a specific character and then two versions later the bug is back. Instead of relying on your current testing phase which requires initiative on behalf of the tester you could just use this list every time before a release and not have to worry about people freaking out over that one Greek letter not working and ruining almost all their posts.

Basically what I'm saying is to have a list of the most common reoccurring bugs that you test at each step of the build cycle. Things like special characters, certain bbcodes, and getting stuck in the mobile skin are bugs that seem to pop up too often. I don't think waiting for someone else to find those bugs is a very acceptable solution. It reflects better on you if those constantly reoccurring bugs just stop...reoccurring.

I think there's integrity in making sure you don't have reoccurring bugs. I understand that it's not possible to map out the entire suite, but having a testing protocol for the most common bug reports isn't impossible.
Link to comment
Share on other sites

If you have any ideas on how to improve this, please let me know. I've worked in software development for about 12 years now and I think we do pretty well.


In addition to what Rimi mentions, you should probarbly create a default test post with various bbcodes. The post would contain bbcode that would be tested in STD, RTE, fast reply and normal reply. The bbcode would be mixed in various ways in the post, and then you would see if you saw any problems or things out of the ordinary with the post.
Link to comment
Share on other sites

In addition to what Rimi mentions, you should probarbly create a default test post with various bbcodes. The post would contain bbcode that would be tested in STD, RTE, fast reply and normal reply. The bbcode would be mixed in various ways in the post, and then you would see if you saw any problems or things out of the ordinary with the post.

I think that's a good idea. Basically like this (http://www.ipbforumskins.com/skindemo/index.php?/topic/6-main-test-topic/) except you want to test it within the editor in STD and RTE at time of posting as well.

Link to comment
Share on other sites

  • Management

I do have a unit test for the editor actually.

" alt="http__localhost_invisionboard3__unittest">

It has a lot of test cases I've identified over the past six months or so.

The issues you (TSP) experienced only occur when manually typing in BBCode.

In a lot of cases, certain character issues don't show locally as we use UTF-8 whereas we use ISO-8559-1 on the company forums. And even then, character issues may not show unless you're using a German, Russian or Turkish character set.

The sheer number of different possible combinations makes it impossible to test completely locally which is why we have beta releases and encourage our customers to install them using the test license and test them out as fully as they can.

Link to comment
Share on other sites

That was just an example, but I do appreciate that something like actually exists in your debugging and am impressed enough to shut up now and maybe never bring this up again. That wasn't in your list though.

Make one for the mobile skin. </3

Link to comment
Share on other sites

Do you have problems with the styling of the bbcode and code?

I did so i just decided to re-design it and i have to say that i fairly like it:

One on top is quote, one on bottom is code.

tFxHm.png' alt='' class='ipsImage' width" alt="tFxHm.png">

TBH i like the new one :/. Really good to work with. Made it so i can tell the difference :/


what i cant stand is the (c ) (r ) and (tm ) conversions so i removed it :/

Link to comment
Share on other sites

Do you have problems with the styling of the bbcode and code?

No. The problem is the limited functionality of RTE quotes. For example I just trimmed your quote. But what if I wanted to reply to two parts of it separately? This is now incredibly more tedious in the RTE.

Link to comment
Share on other sites

No. The problem is the limited functionality of RTE quotes. For example I just trimmed your quote. But what if I wanted to reply to two parts of it separately? This is now incredibly more tedious in the RTE.

Oh yea that does pose a problem, which is why i created a mod several weeks ago to fix that issue for me :|

Though it uses the old format for IPB which shows the quote code instead of the Styling borders. Have to fix that.

Link to comment
Share on other sites


Oh yea that does pose a problem, which is why i created a mod several weeks ago to fix that issue for me :|

Though it uses the old format for IPB which shows the quote code instead of the Styling borders. Have to fix that.

How did you quote my post without it sending me a notification?
Link to comment
Share on other sites

  • 2 weeks later...

Here's what I want to know. When quoting multiple people (multiquote), how do we know who we are typing a response to when in RTE? Don't see the person or the date/time of the post quoted. Have to switch to source to see that information. That's a huge complaint I have about RTE quoting.

Link to comment
Share on other sites

If you want to type out the tags because you're old skool, what's wrong with the plain textbox mode that the lightswitch provides?


You guys realize you COULD solve all these complaints about the RTE by enabling the buttons to insert the bbcode in the STD editor... and adding 1 setting to the ACP to control which mode to automatically turn on.

All you have to do is make it splat in the tags, then you can do the RTE conversion on post.

It's far easier to put in then developing the CDN, the Backup service and all these extras you are making... and you'd get far more positive feedback for making this change.

What is your priority? Is it customer satisfaction? If it was, you'd be doing what the community has asked for many times.
Link to comment
Share on other sites

Editor keeps moving MY reply to within the quote tags. My response is in bold. (Bug report here: http://community.invisionpower.com/resources/bugs.html/_/ip-board/unable-to-reply-outside-of-quote-tags-r40262 oof..)

How did you quote my post without it sending me a notification?

I selected your post [Highlighted] and pressed quote :/

You do realize that highlighting text and then pressing the Quote button won't quote only the highlighted text, right? It quotes the entire post.


What the heck?! Come on editor, stop putting my reply within the quote tags!!!!

Link to comment
Share on other sites

Is there any way to roll back to 3.3.4. because this version is full of bugs.

No. The only way is to restore a back up of your database and forums.

Don't you people ever learn. Never ever upgrade when the .0.0 version comes out. You're essentially public beta testing. Pretty much all software developers do this so it's not just IPS. The release version that the masses get brings up bugs that didn't get picked up with Beta testing. Likely because it's not being testing on live sites with lots of posts and members.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.

×
×
  • Create New...