Can we have better testing by those making commits?
Over christmas we had a rather glaring bug resulting from improper inclusion of the red background on the wrong div, leading to giant blue stripes across the bottom of the page when the forum wasnt tall enough to fill the window.
Right now someone has added a css3 drop shadow, which fails spectacularly in the main menu ( the drop shadow starts far too low leaving a double rounded corner, and when you resize the window below the maximum width, the drop shadow persists despite media queries putting the bar full width ).
Over christmas we had a rather glaring bug resulting from improper inclusion of the red background on the wrong div, leading to giant blue stripes across the bottom of the page when the forum wasnt tall enough to fill the window.
Right now someone has added a css3 drop shadow, which fails spectacularly in the main menu ( the drop shadow starts far too low leaving a double rounded corner, and when you resize the window below the maximum width, the drop shadow persists despite media queries putting the bar full width ).
Was gonna mention these things but figured somebody else would eventually
Clearly it didn't, and in some cases the issues are obvious. What's likely happened is that changes were made to the markup and committed without them being applied to a local test environment to make sure they added/fixed the issue without introducing other issues.
There are similar issues on the footer.
The community news RSS feed addition on the frontpage also has issues, it's pulling in RSS1 not RSS2, and it's only caching for 5 minutes, which is pointless, caching for say 6 or more hours would be better
Over christmas we had a rather glaring bug resulting from improper inclusion of the red background on the wrong div, leading to giant blue stripes across the bottom of the page when the forum wasnt tall enough to fill the window.
You should be happy that someone spends time to make a gift for everyone and created a christmas theme for the website ...
Also you have a wrong idea of the workflow. In contrast to wiki formatting (thx to css extension) only a few ppl have access to update the website (and it's good like this). So you make changes in the css, see that the html itself needs updated - the website code isn't 100% clean, so the sizes of many objects aren't really static or defined with evil hacks instead of simple css-height/width, esp. the header is full of such stuff - and so you make changes to them and commit and wait 3h upto 2days until the website gets updated. Fact is I don't want ssh access neither do I want to disturb others the whole time to update the website's repo.
Last edited by jK on 10 Jan 2012, 13:35, edited 1 time in total.
I managed to update the site, and I don't have access.
If anything it's a good thing that most people don't have access, because working on the live copy is a terrible way of doing things, especially when we have versioned source code.
What I do, and how it should be tested:
Fork the github repo
Set up a local install using X/W/L/AMP
Install the site using the instructions
Make your changes
Test them in your browsers locally
Commit, push
Issue a pull request
Nobody, not a single person, should be making any changes to the live site other than git commands. To do otherwise is dangerous and negligent. What's worse is to continually request changes be put live because you didn't test things and had to go back and make yet more changes.
It is to be noted that in firefox box shadows don't behave quite the same in as in chrome. In chrome they are purely visual, in some versions of firefox it has layout implications.
Users browsing this forum: No registered users and 0 guests
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot post attachments in this forum