We could lose AccuWeather folks!!!!!

[sgtevmckay]

These are excellent ideas, and creating an XML of the appropriate page is easy enough in FireFox

All you have to do is go to the page with the appropriate XML

Example:
http://rainmeter.accu-weather.com/widget/rainmeter/weather-data.asp?Location=87505&metric=0
The above is for US only!!!

Then go to File>Save As
And save the page as an XML to your Hard Drive.

Now the first issue I can see is that I do not believe that JSMorley's Regexp tool will parse a local XML, so I would have to defer to him in regards to this.

Update note Here:
The Regexp tool will open the XML, but it is outlayed Interestingly in the tool

Now you will have to keep in mind that you will have to make the appropriate adjustments for the URL when you are finished, otherwise the skin would continue to look on the local drive for an XML that will not exist
You would also need to start your Parse at "Units" as listed on the page. The XML saved is not exactly what you see from the direct XML page, but no major changes.

Also when building skins, remember do set up the location and Unit Variables prior to release.

allow me to also kick around the idea of placing US and International XML's in both Fahrenheit (SAE) and Celsius (Metric) from an additional page set from the forum server.
This would allow for some self beta testing, but the information would always remain the same until the skin goes live with the Accuweather XML.

I like this idea.
Let's kick the concept around for a bit and discuss feasibility

As for advertising. I think this is addressed as it needs to be here, and I do not believe I can be any clearer.
Please refer to the following thread:
http://forum.rainmeter.net/viewtopic.php?f=53&t=1348&start=0#p9673

[jsmorley]

You guys are making this way too hard. It's easy. Go into RegExp and load the actual URL for the feed one time. Then click "save" on the "HTML" panel and save it with any name.

Now whenever you want to test a RegExp to modify a weather skin, just load up RegExp and click "Browse" Load the file you saved and there is the HTML from AccuWeather without hitting their site. You can also us it for real time testing by pointing the URL in your skin to the file instead of the actual website so you can test meter layout and all that by hitting refresh while not actually hitting the web site.

[Alex2539]

I typically just set "Debug=2" then point WebParser to file://C:/WebParserDump.txt. If you do it that way, you can keep the real link commented out, then just switch it, refresh, and switch it back to update it if needed. I never actually need see the file.

[sgtevmckay]

all this is true, but what is being requested is a static page away from the AccuWeather XML that can be tested and refreshed numerous times without placing a load on the AccuWeather server.

As you all know: that as we develop and tweak our skins, we must refresh several times, when we refresh a skin, it access the URL that is placed in the skin.

So what we are looking for is a non-volatile way to develop a skin, and not hit the AccuWeather server until the skin is ready to go live.

JSMorley also recommended using his Regexp tool to save the page and then to parse the saved file, creating a url to that file in the skin, then just adding the appropriate URL before the skin goes live.

Does DeBug=2 save the entire XML from the requested site?
If so then Alex's concept is also an excellent idea.

[jsmorley]

all this is true, but what is being requested is a static page away from the AccuWeather XML that can be tested and refreshed numerous times without placing a load on the AccuWeather server.

As you all know: that as we develop and tweak our skins, we must refresh several times, when we refresh a skin, it access the URL that is placed in the skin.

So what we are looking for is a non-volatile way to develop a skin, and not hit the AccuWeather server until the skin is ready to go live.

JSMorley also recommended using his Regexp tool to save the page and then to parse the saved file, creating a url to that file in the skin, then just adding the appropriate URL before the skin goes live.

Does DeBug=2 save the entire XML from the requested site?
If so then Alex's concept is also an excellent idea.

Debug=2 does work, but don't forget it saves it in the root directory, which I believe is a dangerous place to save things and do work. You can use debug=2 then move the file somewhere safer, or use RainRegExp and save the html someplace. Either works just as well.

[sgtevmckay]

I am thinking the Regexp is probably the safer way, and allows more freedom, so this I will base a tutorial on the Regexp tool.

especially as when I work on skins, almost everything is on the desktop, as the desktop was meant for an information and staging area.

Either way, this is great, as it will allow skin builders the opportunity to build skins, repeatedly refresh, refresh faster, and not hit the AccuWeather server until the skin builder is about to go live with their skin.

I think this is a worth wild concept, and should be listed as a possibility.

Any additional thoughts here?

[Yelleke]

Isn't it possible to make a weather plugin with the update vaulues (update and updaterate) already inside.
Perhaps with a special code inside to let accuweather known, it comes from a validated skin/plugin, and blocking all other skins, forcing the Rainees to use this plugin.

the plugin could write the xml page (every 15 minutes) to disk (plugin directory).
The plugin could also prevent hitting accuweather nomatter how manny times I use the Refresh button, it will allways read the stored xml file.

Maybe, this plugin could retrieve all data from the xml file and use a uniform StringIndex, so every skinbuilder have the same StringIndex numbers.
Wich makes it easyer to build skins, no hasle anymore with RegExp, just look to the help page to see wich StringIndex number to use for a weathericon, etc.

I also noted that the skin from our sarge, AccuWeather_Vision_Plus, takes more data and time to load than the skin I use, wich only reads 5 strings (who needs all this data anyway)
the weather skin I use reloads in 2 seconds

what I need is temperature, weathericon, rainammount, cloudcover, and txtshort; all the rest is not interesting at all for me.
rainamount is not shown on my skin, unless the value is more than 0.0 wich will then hide the cloudcover. (when it rains, there are clouds I think)

UpdateRate=900
Url=http://rainmeter.accu-weather.com/widget/rainmeter/weather-data.asp?Location=#Location#&metric=1
RegExp="(?siU)<temperature>(.*)</temperature>.*<weathericon>(.*)</weathericon>.*<cloudcover>(.*)</cloudcover>.*<txtshort>(.*)</txtshort>.*<rainamount>(.*)</rainamount>"


Just my 2 cents
Yelleke

[sgtevmckay]

Greetings Yelleke, to the forums

I think that your proposal is an interesting one. In part something similar to this has been discussed, but your concept of having the Addon (DLL) load the XML to a Rainmeter Temp file is intriguing.
I think (and do not hold me to it) is that creating a Temp file for internet drawn down has never been resolved.
Some of this concept would assist Rainmeter over-all, the controll
concept for the flow from AccuWeather is great, but then we need someone to to build it, a working proof, and then testing.
I would not detour this, I would actually embrace this concept. It would have to be developed as a plugin, and can be instituted into the core package.

I admit your are correct: My skin takes forever to initially load, and it is not for everyone. I just happen to be an information monger.
Interestingly enough, a similar amount of information being loaded from weather.com takes significantly less time to load than from AccuWeather.com.
I have been unable to resolve this.
I like your layout, and I think you have hit on something many folks forget: Rainmeter is completely customizable to our needs, and WANTS.
The skin you are using, if properly developed with the appropriate links, will draw more traffic to AccuWeather, than my monster, as a lot of information is already given, so travel to AccuWeather.com site will be less. With the smaller skin, and less information immediately available, more people would be drawn to the web site for more information.

All in all we develop the skins for ourselves, and post them. If folks like them, they download them, and this is one of the beauties of Rainmeter.

Once again, I welcome to you

[sgtevmckay]

For the rest of us that give a damn, and for all those that should.
There has been a development, and I am sorry that most of you are missing it!

I received an e-mail from AccuWeather today
It is very short an simple.
Details here:
http://forum.rainmeter.net/viewtopic.php?f=53&t=1461#p10432

A very simple solution has been rendered, and I am grateful that this will work for many of the Rainmeter end users, but I am unsatisfied.
Many of us should be.
I know that many of you are satisfied with the simple this is this and that is that, but I want more.
I made an attempt to acquire this for us, as a community, but apparently there was no need.

I will continue negotiations with AccuWeather, but there is no overnight solution to their demands. Therefore I suggest you folks take the next 15 days and some hours we have left and find your own alternate solutions.

Rainmeter was built on this principle, but I am disappointed to find that only a handful have worked to render solutions, and these few of us are all tired and PO'ed. Working our fingers to teh bone, and getting slapped at both ends for it!

My depression at this time is beyond parallel, If someone, anyone, can present an international solution, please come forward and do so quickly.
I will even open negotiations with designated concept proposed sites.

As to the other 450, 000+ down loaders of the Default skin and AccuWeather users/skinners: My heart goes out to the entire community as the weathers skins will most likely go dark sometime October 15th 2009.

Finding weather hear in the US is easy, but international concerns are vague and far between.
If you can direct me to a site that will allow free XMl feeds or a builders API, let us know, and fast.

The current rendered solution is not for us all, and I do not want to believe in the "I got what I want, the rest can fend for themselves!"
this is a piss poor approach.

I have at least 2 more site contacts that have yet to return current contact, but I am not holding by breath!

Yes I'm Pissed!!!
I can not express this boldly enough!

[maminscris]

that is bad,

i do not understand why they are so unsatisfied