I've noticed that there has been a flurry of site improvements to the site lately. That's great! I do have a question, though. Whenever I go to create a node (whether it be a reply or a completely new node), I get the following:
Warning: When this page was created your message appeared to be entirely wrapped in pmsig tags. You should ensure that only your signature is within these tags otherwise some users may see an empty node. See Signatures and Node Templates for details.
I get this before I've done anything to post the message. That is to say I hit an "Offer your reply" link and that warning comes up right away. I know that this warning functionality has existed for a while, but what gives?

thor

Replies are listed 'Best First'.
Re: New Behavior for Signatures?
by Chady (Priest) on Jun 03, 2004 at 13:52 UTC

    As the text states, some people may turn off signatures, if you post inside the signature divs, some people will not be able to see your node.

    As a solution, there is now Node Template in User Settings where you can specify a default template for your node (info)

    I took care of the warning by placing a single <p> tag in my node template, so everytime I try to compose a node, I have a starting tag waiting for me, and no warnings.


    He who asks will be a fool for five minutes, but he who doesn't ask will remain a fool for life.

    Chady | http://chady.net/
      That doesn't answer his question. The question is, Why does PM give that warning BEFORE the post has been submitted? Iow, why does PM (seem to) check whether the post appears to be entirely inside pmsig tags before the poster has had a chance to write a single word? (Not that I've observed this phenomenon myself...)

        I see,

        I had a quick glance at add_signature; it issues its warning without looking for a op=preview param.

        If the param is set, then it should return the warning, else, it should assume that the display is not a preview and thus not show the warning. But I guess it is implemented that way to warn people who do not preview their nodes.


        He who asks will be a fool for five minutes, but he who doesn't ask will remain a fool for life.

        Chady | http://chady.net/

      While jdporter didn't appreciate Chady's first response, I found that it provided the info needed to get the message off of my page. In other words, I appreciated Chady's node and info provided. Thanks tye for the further info.

      Paulster2


      You're so sly, but so am I. - Quote from the movie Manhunter.
Re: New Behavior for Signatures?
by halley (Prior) on Jun 03, 2004 at 16:46 UTC
    I don't understand why the .sig needs to be part of the node contents at all. Sigs are separate data like subject lines. If a user wants to edit a node, fill the empty edit box with JUST their preferred template (if any), and let them edit JUST their new node's contents. When the server renders a node to HTML, render the subject, then contents, then sig. Keep them separate.

    Another web forum I frequent has a lot of Internet-newbies on it, and this forum code does the same thing: it inserts the .sig into an empty edit box, expecting the user to edit above that space. These users always put their new contents UNDER the -- line, or even under their name, because they're not familiar with the convention.

    I'm sure this is just an artifact of slow growth on top of the Everything codebase, which never considered signature lines at all, or has much use for them.

    --
    [ e d @ h a l l e y . c c ]

      Appending the sig at rendering time appeals to me. Previous proposals to append the sig but only after "create" is pushed did not appeal to me (sorry, no time to explain ATM).

      You'd lose the ability to see how my sig has changed over the years, but I find that an acceptable trade-off.

      Best of all, it would prevent the frustration of super searching for information about the "Flort:Bangle" module and finding every single node by Tye because his stupid sig says "Flort::Bangle *rocks*!!" just because he wrote it.

      The problem of backward compat is not trivial and may be enough to prevent this change happening. But this idea is really starting to grow on me.

      - tye