[Note, after proofreading this message a bit, I see that I use the terms 'add-on' and 'extension' interchangeably, which is probably not unusual for FF-related groups.]
John J Barton wrote: > First let's consider that Firebug is used by tens or hundreds of > thousands of people every day. Then let's read your post. It seems to > me that a better title for the post would be "Mavi Gozler has a > Nightmare" ;-) Granted. I know that Firebug is not thrown into the universe to work with Firefox by itself, and that its functions and features and those of FF can be altered by other, often badly written executable code. I am only saying that when Firebug was enabled, I was observing this behavior. I certainly consider the possibility that FB in combination with other badly written extensions---and even problems with FF upon which it sits---was causing this behavior. Also, surely if THOUSANDS have installed and used FB, then they can and have probably reported which setups and/or add-ons are reported or confirmed to interfere with the operation of FB. Such information would be useful to include in group documentation or its FAQ. I could find nothing in the group literature. A search of posts regarding interferences would be more complicated...essentially it would require reading often every post, many of which have improperly specified or even unspecified subject headings (my own header would be much more specific if there were a specific problem I had). > > The three most common causes of these "Firebug Nightmares" are > 1) Extension conflicts: some other extension which works fine > interferes with Firebug's operation, > 2) Some data in the Firefox profile that breaks Firefox in a way > that is only seen when Firebug operates, > 3) An unusual set of options or pages that Firebug developers have > never seen. Okay, with much gnashing of teeth and other kind of effort, I have eliminated the one and only profile running on FF, and started with a clean slate. FB is the first add-on, and I have also installed an HTML validator tool. Over the course of a few weeks, I will see how FB runs on top of FF in this very minimal environment. If it is advisable to create a separate profile in FF strictly for development, I would like to know. If it requires re-start of FF to move from one profile to another, does anyone find this an annoyance? I multi-task a great deal, so I use FF for surfing often and then switch tabs to continue web development. I would like to know the habits of other developers, such as hobbyists/amateurs like myself, and perhaps from you who actually may be earning a living in whole or in part from this. > We can eliminate the first two as a cause if you install Firebug in a > new Firefox extension and repeat your testcase. > http://kb.mozillazine.org/Profile_Manager#Creating_a_new_profile Yes, it could be that my profile, filled with extensions in which FB was one of the last rather than the first and only, were getting in the way and explaining all the behavior in FF and FB that I was seeing. Time---several days to a couple of weeks---will now inform me of this. > On Nov 20, 11:51 pm, Mavi Gozler <[EMAIL PROTECTED]> wrote: >> Firefox 3.0.4 with Firebug 1.2.1 >> >> 1. When a reload of a page is done, the loading hangs. My guess is >> that this is related to Firebug being in the middle of a debug >> session. It does not seem to occur when Firebuy is NOT in the middle >> of a debug session--that is, when the Continue funciton is used to get >> Firebug out of a script. > > Firebug does not have a Continue function. Really? When the Script tab is active, and in the middle of debugging a script, and Firebug becomes a panel within the FF window, there is a blue right arrowhead with a key equivalent F8 which "continues" execution of the script without stepping. When you put the mouse cursor over the blue right arrowhead, the title "Continue (F8)" even appears. To be "in the middle of a debug session," I mean that Firebug is controlling the execution of Javascript and is stepping through it, either as the result of (1) use of a "debugger" keyword invocation, (2) setting of a breakpoint using FB's script control (breakpoint-setting) feature, or (3) halting on error or exception. > >> 2. If a page is reloaded with a script, its script will stop working, >> even though the script is bug-free. This is probably related to the >> related to Firebug being in the middle of a session on another page. > > This paragraph makes no sense. "its script" means which script? What > mechanism would cause Firebug to suddenly decide to sabotage a script > in a page? What does it mean for Firebug to be "in the middle of a > session"? I define "middle of a session" above: that is, FB takes over control of Javascript/ECMAscript execution. Let me show you the following HTML page fragment: ============== START HTML DOCUMENT =============== <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd";> <html> <head> <title>Title</title> <script type="text/javascript"> function initializePage() { for (var i = 0; i < document.forms.length; i++) document.forms[i].reset(); } function showLogin() { document.getElementById("login-form").style.display = "block"; } function showPassword(chkboxObj) { if (chkboxObj.checked == false) document.getElementById("systempwd").type = "password"; else document.getElementById("systempwd").type = "text"; } </script> </head> <body onload="initializePage();"> <p id="title"> Page Titl<span onclick="showLogin();">e</span> <form action="OpinionSessionHome.php" id="login-form"> <legend style="font:bold 110% serif;background-color:#aaa;color:maroon; border:2px solid blue;">System Login</legend> <table id="login-table" style="margin-top:2em;"> <tr><th>username</th><td><input name="uname" size="10"/></td></tr> <tr><th>password</th><td><input type="password" id="systempwd" name="pass" size="10"/></td></tr> <tr><td colspan="2" style="font:normal 80% 'Courier New',Courier,monospace;"> <input type="checkbox" onchange="showPassword(this);" />Show Password</td></tr> </table> <button type="submit"><b> Login </b></button> <button type="reset">Clear Form</button> </form> </body></html> ============== END HTML DOCUMENT =============== In this document, processed through PHP and its code eliminated, the administrative use clicks on the letter 'e' (which has an 'onclick' attribute set within a span element) of the page title, which I have changed to "Page Title", and form appears in the page for logging into the document control/management system. Clicking on the 'e' always worked without fail. Then when FF and FB get into this interacting 'funk' I describe, it does not respond. It requires a complete re-start of FF again, with no strict requirement to suspend FB's operation at any rate. > >> 3. The suspension and resumption of Firebug does not seem to work at >> all. A breakpoint that is set won't be caught. This is true when >> Firefox is even re-started. > > Suspension and resumption of Firebug has nothing to do with > breakpoints. True. I meant to express two different thoughts in this paragraph, rather than make them appear to be connected. What I am saying here is that when I want to resume FB control of a document---analysis of its structure (DOM/HTML) and supervision of its script---it doesn't seem to do it. I purposely put errors into script to see if FB stops on them after resuming, and it does not catch the error. In other moments, I set a breakpoint in script, then reload the page, hit a form control that I know will run a script with a breakpoint set inside it. The script runs, and it does not stop on the breakpoint. > >> 4. In clicking on the pages of these discussion groups, Firebug came >> up and I disabled it for use on this site. I then clicked the next >> page for discussion, and loading was hanging. When reloaded the >> intial page and then clicked next page, it still hung. When I clicked >> away the tab and re-opened a new tab, the loading of the next page >> still hung, despite Firebug having been disabled. > > Firebug probably sensed that you were going to post something > unpleasant to the newsgroup and tried to prevent it. Possibly. > >> 5. In closing down Firefox, I am unable to restart it. This is >> because Firefox continues as an invisible, high CPU-using (50%) >> process that must be ended from within the Task Manager. > > Every time I have looked into this kind of problem, the cause has been > an error message raised within the error handling code that triggers > on exit, causing errors, which trigger the error handling code, ... > > The looping behavior is a Firefox bug. The best we can do is try to > protect against error propagation within Firebug's error handler. Has someone written their own handler and offered it as a library code of some kind? > >> 6. All these problems seem to go away when Firebug is disabled. > > This however does not mean that Firebug is the cause. Agreed. FB may be a catalyst in the expression of another's bad code or interaction with the entire application environment that is FF. There is no doubt that a soup can be made in which one starts with water (FF) and then one adds one ingredient at a time, say onion and/or garlic, then salt, then may be another vegetable, and then spices one by one. We can't add everything to the soup obviously, or else it tastes awful. The question is what to add, and in what order to add, and then when to stop adding. So you tell me how you think this soup should be made in the end, given that I have already gone back to getting new water (FF with profile deletion), and am now adding ingredients (so far FB and the validator tool). I hope you can relate to this metaphor. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Firebug" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/firebug?hl=en -~----------~----~----~----~------~----~------~--~---
