CMS System WTF



  • We use a very expensive and very crappy CMS system here called SiteRefresh.

    There isn't a single user who is not proficient at HTML/CSS, yet we are all forced to enter content changes into this system.

    It does allow you to turn off its horrible WYSWIG editor, and do plain HTML, however when you save your change the system reparses your entry.

     Some of the most annoying "Features":

    •  Attributes that are crossbrowser compatible (for example: align="center") automaticly get replaced with align="middle" and other IE only attributes.
    • Table cells get randomly reordered, I believe this is just a bug in the parser.
    • If you have a line break in your HTML, it randomly inserts <br/> tags.
    • Ampersands are always escaped, even if you put them inside of quotes for an URL, this is a real PITA.
    • There is no support for javascript, we have hacked in a token replace system to allow us to put some javascript into the system, but it will escape characters such as < and > making any sort of javascript difficult.
    • External CSS styles are not allowed, the Style tag automaticly gets removed, and you are forced to use inline CSS.
    • Inline CSS gets parsed and reformatted, and often broken by the parser.
    • The parser decides when it wants to add additional attributes to a tag, for example, if you don't explicity set a width attribute on a div, or a image, or a table then it sticks width="100%" in.
    • Images are stored internally, and the only way to look one up is to search for its asset name, and then the editor will return both the image path (which is a huge long string of guids), and add a special key attribute to the tag. This breaks HTML validation, and without the key attribute on the tag, the image won't show up

     I've been fighting this thing all morning, and I'm ready to kill someone.



  • You have a Content Management System System?



  • @MasterPlanSoftware said:

    You have a Content Management System System?

     

    Actually it sounds like the hacked-together JS support is a Content Mangement System Management System



  • @MasterPlanSoftware said:

    You have a Content Management System System?
     

     MasterPlan, do you ever have anything useful to contribute?



  • @morbiuswilters said:

     MasterPlan, do you ever have anything useful to contribute?

     

    This is coming from someone with 3 posts. How about you redirect the question at yourself?



  • @MasterPlanSoftware said:

    You have a Content Management System System?

     

     The Department of Redundancy Department called.

     They want their joke back.

     



  • @MasterPlanSoftware said:

    @morbiuswilters said:

     MasterPlan, do you ever have anything useful to contribute?

     

    This is coming from someone with 3 posts. How about you redirect the question at yourself?

     

    Disregarding my replies to you, I've had something to say twice, which is evidently two more times than you.  Then again, I'm not a poorly-diguised spambot gaming for an inflated post count.  Seriously, at least Lysis and SpectateSwamp are amusing trolls.



  • @morbiuswilters said:

    Disregarding my replies to you, I've had something to say twice, which is evidently two more times than you.  Then again, I'm not a poorly-diguised spambot gaming for an inflated post count.  Seriously, at least Lysis and SpectateSwamp are amusing trolls.
     

    Riiight. Because no one on this forum has ever made a joke before...

    Every reply is insightful and critical to the post...

    You are a retard.



  •  @Jonathan Holland said:

    It does allow you to turn off its horrible WYSWIG editor, and do plain HTML, however when you save your change the system reparses your entry.

     Some of the most annoying "Features":

    • [stuff]

    Much source parsing fuckedness can be attributed to the fuckedness of IE's source parser. Have a look at the source when saving a webpage.



  • @Jonathan Holland said:

    Ampersands are always escaped, even if you put them inside of quotes for an URL, this is a real PITA.

    Actually, that particular behavior is absolutely correct - even inside link tags, you MUST escape ampersands in URLs used to separate parameters (i.e. href="index.php?page=1&amp;section=2".

    Now, if it was re-encoding ampersands attached to existing attributes (e.g. turning &amp; into &amp;amp;), THEN there's a problem.



  •  What kind of an asspod would authorize buying a Content Management CMS System from a company with a site that has rendering errors in both IE and Firefox?

    http://refreshsoftware.com/ 



  • Try comparing that site to Opera's site: http://www.opera.com
    "Inspiration" or more than that?



  • @Quietust said:

    @Jonathan Holland said:
    Ampersands are always escaped, even if you put them inside of quotes for an URL, this is a real PITA.

    Actually, that particular behavior is absolutely correct - even inside link tags, you MUST escape ampersands in URLs used to separate parameters (i.e. href="index.php?page=1&amp;section=2".

    Now, if it was re-encoding ampersands attached to existing attributes (e.g. turning &amp; into &amp;amp;), THEN there's a problem.

     

    However, pass an url with escaped ampersands to ASP.NET, and you won't be able to pull the querystring.

    Try testing this in PHP, or another web framework and see if it does.



  • @Jonathan Holland said:

    However, pass an url with escaped ampersands to ASP.NET, and you won't be able to pull the querystring.

    Try testing this in PHP, or another web framework and see if it does.

     

    Why would you post an escaped URL back to the server? After all, it's the browsers job to unescape it and process it in its correct form. 



  • @PSWorx said:

    @Jonathan Holland said:

    However, pass an url with escaped ampersands to ASP.NET, and you won't be able to pull the querystring.

    Try testing this in PHP, or another web framework and see if it does.

     

    Why would you post an escaped URL back to the server? After all, it's the browsers job to unescape it and process it in its correct form. 

     

     Then the browser is fail :)

    For example, if the CMS system takes: "http://www.fictionalSite.com/page.aspx?Stuff=1&Stuff2=2"

    and escapes it to:

     http://www.fictionalSite.com/page.aspx?Stuff=1&amp;Stuff2=2

    The browser does not convert it back, and the framework parsing the querystring does not return Stuff2. 



  • Drupal is my CMS of choice… it's free, of course. We've been making a lot of sites with it at work recently, and once you learn how to write for it, it really is a joy to use. Seriously, who would pay for a half-assed CMS when you could use the entirely non-assed Drupal for free?



  • @Albright said:

    Drupal is my CMS of choice… it's free, of course. We've been making a lot of sites with it at work recently, and once you learn how to write for it, it really is a joy to use. Seriously, who would pay for a half-assed CMS when you could use the entirely non-assed Drupal for free?

     

    Errm... Surely that should be "fully-assed" at least according to here, the only other etymology I could find was that "half-assed" is a humorous mispronounciation of "haphazard".



  •  @Jonathan Holland said:

     Then the browser is fail :)

    For example, if the CMS system takes: "http://www.fictionalSite.com/page.aspx?Stuff=1&Stuff2=2"

    and escapes it to:

     http://www.fictionalSite.com/page.aspx?Stuff=1&amp;Stuff2=2

    The browser does not convert it back, and the framework parsing the querystring does not return Stuff2. 

    How is your escaped URL embedded into the page HTML?

    At least firefox does unescape it correctly, and I'd be very surprised if many other browsers were any different. There may be many quirks, but entity resolving is in this form there since the very beginnings of HTML. There shouldn't be any major implementation differences in such a basic part, should there?

    Firefox example (in lack of a webspace):

    [code]data:text/html,%3ca%20href=%22http://www.google.com/search?someparam=0%26amp;someotherparam=1%26amp;q=The+REAL+WTF%22%3eThe%20REAL%20WTF%3c/a%3e[/code] 



  •  @superjer said:

     What kind of an asspod would authorize buying a Content Management CMS System from a company with a site that has rendering errors in both IE and Firefox?

    I believe they're generally known as 'managers'.

    I used to have to work with a system called Evolve CMS - written in classic ASP - which was a pretty big WTF. It wouldn't work in anything except IE (and then only sometimes), you couldn't upload any images / other files to the server (and given I didn't have any other file transfer access this was a pretty big problem),  lots of the pages wouldn't actually work (i.e. lots of ASP errors...)

    It was made by a fairly small company a few years back, but still... it looks like it doesn't quite compare with Refresh's CMS. 



  • @PhillS said:

     @superjer said:

     What kind of an asspod would authorize buying a Content Management CMS System from a company with a site that has rendering errors in both IE and Firefox?

    I believe they're generally known as 'managers'.

    I really don't understand how managers get sold on this BS so easily. When I worked at a gov't office, the management's first question when faced with a problem was "what can we buy to fix this?"

    Hello? You already have 4 developers on contract here... so you're going to pay for some bloated enterprise software, then pay for us to spend time setting it up and customizing it (which is occasionally longer than it would take to solve the problem on our own), and they invariable pay for support contracts, and end up bringing in consultants from the enterprise Bloatware company...? Is that right? Yeah? OK...



  • @morbiuswilters said:

    Lysis and SpectateSwamp are amusing trolls

    True on the latter, false on the former.  They both spout nonsense, but Lysis is not amusing 



  • @Jonathan Holland said:

    @PSWorx said:

    @Jonathan Holland said:

    However, pass an url with escaped ampersands to ASP.NET, and you won't be able to pull the querystring.

    Try testing this in PHP, or another web framework and see if it does.

     

    Why would you post an escaped URL back to the server? After all, it's the browsers job to unescape it and process it in its correct form. 

     

     Then the browser is fail :)

    For example, if the CMS system takes: "http://www.fictionalSite.com/page.aspx?Stuff=1&Stuff2=2"

    and escapes it to:

     http://www.fictionalSite.com/page.aspx?Stuff=1&amp;Stuff2=2

    The browser does not convert it back, and the framework parsing the querystring does not return Stuff2. 

    It must be getting escaped twice or something, then. This very thread has &amp; in several links.



  •  @Cap'n Steve said:

    @Jonathan Holland said:

    For example, if the CMS system takes: "http://www.fictionalSite.com/page.aspx?Stuff=1&Stuff2=2"

    and escapes it to:

     http://www.fictionalSite.com/page.aspx?Stuff=1&amp;Stuff2=2

    The browser does not convert it back, and the framework parsing the querystring does not return Stuff2. 

    It must be getting escaped twice or something, then. This very thread has &amp; in several links.

    A browser will definitely turn a source-"&amp;" into a regular &.

    So I'm assuming the browser gets &amp;amp; 


Log in to reply