Agreed, Larry. A real debugger is the best. I prefer to use Firebug or Safari's error console to dig into my errors.
So, when I test in IE and get an error notification, I just treat that as a signal, and switch back over to FF or Safari for the real diagnosis. My point to the OP was mainly to employ some means of surfacing errors. I don't like the meme "works in [browser 1], not in [browser 2]" because that's usually a matter of HOW browsers cope with coding errors, not WHETHER there are coding errors. Leaving aside IE's non-standard DOM and related javascript methods/ properties, I've actually found it quite rare that a javascript problem is due to code that is INCORRECT only in IE. It's more that IE isn't as tolerant of coding errors as some other browsers. When I started teaching HTML in 1996, the most difficult challenge was when students would say things like "It looks fine in Netscape, so who cares about Internet Explorer?" or "I'm developing this page in Windows, so I need it to work in Internet Explorer." The legacy of those non-standard, overly fault-tolerant days is still with us, and shows up on this list, as when someone says "We're developing our map in Firefox." On May 8, 9:40 am, "[email protected]" <[email protected]> wrote: > On May 8, 9:22 am, Bruce Van Allen <[email protected]> wrote: > > > I just just keep IE's preference enabled for showing script errors. > > > In IE7, it's under Tools -> Internet Options -> Advanced -> Browsing - > > > > Display a Notification About Every Script Error. > > I always do that, but I find IE's errors pretty terse, it tells you > the line number on which the error occured, but not what file that > line is in. Experience helps to interpret them but when I have the > script debugger available, it is much more useful for tracking down > the errors, particularly IE specific ones... > > -- Larry > > > > > You might search around for more verbose utilities, but the above does > > give useful clues. > > > As a related matter, I think good practice would call for checking > > your points before attempting to create markers with them. > > > E.g., as you loop through the location data, check whether you get a > > valid LatLng, and skip creating a marker for that point if not. > > > var debug = true; // global variable usually set to false; set to > > true for testing. > > > // later, inside your loop through location data: > > > var newPoint = new GLatLng(lat,lon); > > if (newPoint = null) { > > if (debug) GLog.write('Point error with lat ' + lat + ', lon ' > > + lon + '. Location skipped.'); > > continue; > > } > > > Even if you quietly skip a bad point (that is, without reporting via > > GLog), your script will be able to carry on. > > > Invalid coordinates show up in lots of ways, not just hand-coded XML, > > so it pays to take precautions and report out failures. Storing map > > coordinates that haven't been checked risks problems later. In my > > previous life as a woodworker, we had a name for what you get when a > > mistake made early on isn't noticed: "time bomb". You can count on > > them exploding right when you deliver the cabinet for installation, > > most likely with the client standing right there! > > > On May 8, 8:00 am, Matthew <[email protected]> wrote: > > > > Larry, > > > Thanks for catching that. I had to manually generate the xml file > > > from a excel sheet of addresses. I then added in the lat lon for each > > > one. That's the problem with doing it all by hand you can miss one > > > small thing and mess everything up. Ok last question how did you > > > figure out that the function Pi was generating the error? I have not > > > found a good way to view the exact source that IE is using to build > > > the page. In Firefox I just use Firebug. Is there something similar > > > for Internet Explorer? Anyway thanks again. > > > > Matthew Sturdevant > > > > On May 7, 10:07 pm, "[email protected]" <[email protected]> > > > wrote: > > > > > On May 7, 8:06 pm, Matthew <[email protected]> wrote: > > > > > > Yup your right I don't call it. Must of been left over from something > > > > > I borrowed. Anyway removed the showAddress function and got the same > > > > > error. But the error doesn't seem to mean anything. Thanks for > > > > > catching that. I should of caught it. > > > > > The error is here: > > > > function Pi(a,b){a.style.left=I(b)} > > > > a is a div > > > > b is NaN > > > > > That comes from the API processing this call: > > > > map.addOverlay(marker); > > > > > where: > > > > var marker = createMarker(markers[i], i); > > > > > i = 84 > > > > > This entry in your xml is invalid: > > > > <marker lat="38.648945" lng=",-121.382452," label="California Check > > > > Cashing" address="4630 Watt Ave" city="N Highlands" state="CA" > > > > zipcode="95660" weekdayHours="8:00AM-9:00PM" satHours="9:00AM-6:00PM" > > > > sunHours="10:00AM-6:00PM" /> > > > > > lng=",-121.382452," is not a number (NaN)... > > > > > -- Larry > > > > > > On May 7, 5:17 pm, Ralph Ames <[email protected]> wrote: > > > > > > > Interesting if you delete that function > > > > > > you get another error message. > > > > > > > But the map works in the same way. > > > > > > > Where is this function called? > > > > > > > Ralph- Hide quoted text - > > > - Show quoted text - --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Google Maps API" 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/Google-Maps-API?hl=en -~----------~----~----~----~------~----~------~--~---
