egghelp.org community

[random2]

Did you change it everywhere?

Find every place that has "GET" (in capitals) and that's every place that needs to have a similar change made. I count three places. (It only *really* needs to be in one place for this to work, but I did it in all three for consistency [and the one place isn't the line number in the diff])

[redwolf]

changed it in all 3 places, are you on irc at all?
i was runningh 4.2.4
is the port still 80?

[redwolf]

Tcl error [pub_wz]: can't read "outtitle": no such variable

[redwolf]

if {[catch {puts $wzsock "GET $pn_url HTTP/1.0"} sockerr]} { puthelp "NOTICE $nick :$sockerr"; puthelp "NOTICE $nick :Try again later, look out the window till then!"; close $wzsock; return 0 }

[random2]

[random2]

[cache]

I wonder if Dragon is still around? I tried to go to his chat room and forum as mentioned in the .tcl but its gone no longer there

[Yog]

@random2: Yeah I figured it'd be something like that from watching the HTTP GETs using live headers and ngrep. Was going to form those full HTTP headers and try again, but ran out of time.

As for the legal issues, who knows. I've been using this script for years and years. I think it was derived from another script. I don't think it was Murf who originally wrote it (I could be wrong about this though). I sort of remember when he took it over. Was a long time ago though, so memory may be a bit foggy.

I also fixed it a few times over the years and sent diffs to Murf before he stopped supporting it and Dragnlord took it over (I'm the "Jim B" that's thanked in the change logs ).

By the way, incith also has a wunderground script that is actively supported, and works well. It doesn't have as many features, and works a bit differently, and the output is a bit more terse, but it works. Check it out here: http://incith.com/eggdrop/

[Moonster]

[cache]

[hangtough]

I prefer Dragnlord's script, but I got this Weather v2112 and it was simple enough and works.

Weather v2112
http://www.egghelp.org/cgi-bin/tcl_archive.tcl?mode=download&id=1365

I'd much rather have the original. I will patiently wait as well.

[stoneone]

Dragonlords script will work with the fixes mentioned before. My 4.2.4 had the mtemp problem and the outtitle problems also. For clarification the patches I made are shown below:

Line 878:
puts $wzsock "GET $pn_url HTTP/1.0"
puts $wzsock "Host: www.wunderground.com"
puts $wzsock "Cookie: Units=both"
puts $wzsock ""
flush $wzsock

where before it was only:
puts $wzsock "GET $pn_url"
flush $wzsock

also line 745:
if {[catch {puts $wzsock "GET $webpage HTTP/1.0"; puts $wzsock "Host: www.wunderground.com"; puts $wzsock "Cookie: Units=both"; puts $wzsock ""} sockerr]} { puthelp "NOTICE $nick :$sockerr"; puthelp "NOTICE $nick :Try again later, look out the window till then!"; close $wzsock; return 0 }
flush $wzsock


where before it was:
if {[catch {puts $wzsock "GET $webpage"} sockerr]} { puthelp "NOTICE $nick :$sockerr"; puthelp "NOTICE $nick :Try again later, look out the window till then!"; close $wzsock; return 0 }
flush $wzsock

Thanks.

Edit: Changed line 745 to execute as catch block.

[Suntop]

I hope he will come along and fix it

I looked at the code, but dont know how to code with websites
_________________
--==> Wizard Suntop <==--

[stoneone]

Just make the changes to the .tcl file that I mentioned, .rehash, and it will work just fine.

[random2]

stoneone, the 745 fix needs to be inside the catch block (yep, in that big now-one-longer-line -- side note, the original coding style and back in weather4 even was better). Otherwise if there's a problem, you'll be writing to the closed socket. It needs to go right after the "GET ...".

The inconsistencies in the coding style here (look at the difference in all three of the places "GET" is in the code makes this rather unmaintainable code.

The script hangtough mentioned is, like this one, also a successor to murf's. If you want the "original" go back to murf's. I've looked at that one as well. I'm not fond of the output format of that one either, but note that the parser on that one is better (it looks at the properties of the html, sort of the way you would in XML [simplified explanation]) rather than the structure of the page. It also didn't need to be updated for this page breakage.