EPGuru

Yo man you figured out a way to eliminate the 3minute delay bug? let us know, peace...

[f0xn3t]

ok folks, it's DONE, phew
I spend whole Friday and couple of hours today working on my script. I must say it was harder than I thought, but now I know what was wrong ...and this update should fix all problems, I hope


so, the first problem was in searching of tv shows at epguides.com. The fact is that epguides.com has deleted its own searching script and now uses google.com searching.
It wouldn't be so hard to fix it if google.com had better terms and condition from my point of view -> it means that eggdrop is unable to get search results because he gets Error 302 (Moved; google.com and altavista.com check URL and all "ampersands" change to "&") at first and even if it's succesfully redirected he gets Error 403 (Forbidden). So, there's no way to get rid of this Error 403 in case that I don't want to use any extra files

so, EPGuru uses Yahoo.com search engine and it's possible to use Logical operators like +,-,"", etc..
e.g. "!ep joey +2004" and you get Joey
"!ep joey" and you get Friends ..lol
... but it really depends on yahoo.com, it's not EPGuru bug


..and, the second problem was in 3min epguides.com response delay. This problem took me most of my time.

The situation is that epguides.com probably increased security level of server(s), because it's unable to "PING epguides.com" -> you get Timed-out...
My theory is that $socket in TCL is unable to overcome this problem because $socket probably pings the server before establishing connection... but $socket gets no response approx. for 150sec. ( -> this time is set by default in system, dunno where <- this was written on one developer forum

I thought this project is lost till upcoming v4.0 but suddenly I found one page with this command: "::http::geturl". This command reguires package http 2.0 (at least) but this package should be included in eggdrop, but I'm not 100% sure.

...but the important thing is it works great with epguides.com with NO DELAY. I get response from bot in 3sec


... so, this is my Friday and Saturday story
I hope you'll enjoy it and don't forget to tell your friends about this update

..and please, let me know if there are any problems with ::http:: or "package http 2.0", time-outs or so...

----------------------
Changelog:
- !!! IMPORTANT - Fixed Tvshowsondvd.com news parsing
- !!! IMPORTANT - Fixed Epguides.com searching (epguides.com search engine was deleted, so this script uses yahoo.com search engine (Google.com or Altavista.com can't be used due to terms and conditions))
- !!! IMPORTANT - Fixed approx. 3min delay in epguides.com response (caused by higher security of epguides.com server(s))
----------------------
DOWNLOAD

!ep smallvile
[EPGuru] Wrong page has been retrieved for "smallville" (http://epguides.com/calendars/)

!ep jag
[EPGuru] Wrong page has been retrieved for "jag" (http://epguides.com/menuj/)

*shrug*

enterprise works
joey works
joey +2004 works

can you make it auto +2004?? it seems redundant that we have to type +2004 to get the correct result

maybe i can mess with the script myself and add it...

awesome script though!

EDIT: !ep smallville +2004 WORKS! please make it auto +2004

[f0xn3t]

sorry, I can't make it auto to search series with +2004, because the script must be able to search old series as well

but here is a quick help for you...
this line:

if {[catch {set token [::http::geturl http://search.yahoo.com/search?_adv_prop=web&x=op&ei=UTF-8&prev_vm=p&fr=fp-top&va=$name&va_vt=any&vp=&vp_vt=any&vo=&vo_vt=any&ve=&ve_vt=any&vd=all&vst=on&vs=epguides.com&vf=all&vm=p&vc=&fl=0&n=10 -timeout 30000]}]} {

change to this:

if {[catch {set token [::http::geturl http://search.yahoo.com/search?_adv_prop=web&x=op&ei=UTF-8&prev_vm=p&fr=fp-top&va=$name+%2B2004&va_vt=any&vp=&vp_vt=any&vo=&vo_vt=any&ve=&ve_vt=any&vd=all&vst=on&vs=epguides.com&vf=all&vm=p&vc=&fl=0&n=10 -timeout 30000]}]} {

[wicked]

at first i'd like to thank f0xn3t for this excellent script!
i got one question tho...
does anyone know how to get the info for the show "24"?
it just doesnt work for me, i always get Epguides.com URL: http://epguides.com/SouthBankShow/ if i request "!ep 24"

using "!ep 24 +2004" also doesnt work, i just found out that works for "LOST" tho, which was another thing i wanted to ask here :P

the strange thing is, that 24 is listed on top position in the yahoo result the script queries for ( http://search.yahoo.com/search?_adv_...vc=&fl=0&n=10+ ) but its somehow not parsed the right way, it grabs southbankshow which is next in the list.
so would be nice if anyone could tell me how to fix this
thx

[f0xn3t]

Originally posted by wicked
at first i'd like to thank f0xn3t for this excellent script!
i got one question tho...
does anyone know how to get the info for the show "24"?
it just doesnt work for me, i always get Epguides.com URL: http://epguides.com/SouthBankShow/ if i request "!ep 24"

using "!ep 24 +2004" also doesnt work, i just found out that works for "LOST" tho, which was another thing i wanted to ask here :P

the strange thing is, that 24 is listed on top position in the yahoo result the script queries for ( http://search.yahoo.com/search?_adv_...vc=&fl=0&n=10+ ) but its somehow not parsed the right way, it grabs southbankshow which is next in the list.
so would be nice if anyone could tell me how to fix this
thx

here is a quick fix for everybody...
When you write "!ep 24 +2004" or "!ep lost +2004" you'll get the right info

This part of the script:

			if {[regexp {<div class=yschabstr>.*</div><em class=yschurl>.*</em>} $line] } {
				regsub {<div class=yschabstr>.*</div><em class=yschurl>(.*)</em>} $line {\1} page
				regsub -all "<b>" $page "" page
				regsub -all "</b>" $page "" page

Change to this:

			if {[regexp {(<div class=yschabstr>.*</div><em class=yschurl>.*</em>|<em class=yschurl>.*</em>)} $line] } {
				regsub {(<div class=yschabstr>.*</div><em class=yschurl>(.*)</em>|<em class=yschurl>(.*)</em>)} $line {\1} page
				regsub -all "<b>" $page "" page
				regsub -all "</b>" $page "" page
				regsub -all "<em class=yschurl>" $page "" page
				regsub -all "</em>" $page "" page
				regsub "<div class=yschabstr>.*</div>" $page "" page

[wicked]

hi, thx for the quick response.
kinda doesnt work for me tho...

"!ep 24 +2004" -> [EPGuru] Wrong page has been retrieved for "24 +2004" (http://epguides.com/grid/fall.shtml/)

"!ep 24" -> Title: The South Bank Show

lost works fine for me tho, but that also worked with +2004 before the code replace....

[f0xn3t]

Originally posted by wicked
hi, thx for the quick response.
kinda doesnt work for me tho...

"!ep 24 +2004" -> [EPGuru] Wrong page has been retrieved for "24 +2004" (http://epguides.com/grid/fall.shtml/)

"!ep 24" -> Title: The South Bank Show

lost works fine for me tho, but that also worked with +2004 before the code replace....

I'm sry, I used a bit different version of script

Keep the change I've written above and find the line which contains this code:

if {[catch {set token [::http::geturl http://search.yahoo.com/search

and whole line change to this:

		if {[catch {set token [::http::geturl http://search.yahoo.com/search?_adv_prop=web&x=op&ei=UTF-8&prev_vm=p&fr=fp-top&va=$name&va_vt=any&vp=&vp_vt=any&vo=&vo_vt=any&ve=&ve_vt=any&vd=all&vst=on&vs=epguides.com&vf=all&vm=p&vc=&fl=0&n=10 -timeout 30000]}]} {

[wicked]

hmm the exact same line is in the script i use already
its not working correct

[happy123]

hi f0xn3t, thanks a lot for this awesome script, works very well.
I had the same problem as wicked with a few shows like 24 and CSI, but perhaps I found a solution:
in the regexp

foreach line [split [::http::data $token] "\n"] {
if {[regexp {(<div class=yschabstr>.*</div><em class=yschurl>.*</em>|<em class=yschurl>.*</em> )} $line] } {

I removed the last blank space after |<em class=yschurl>.*</em> (also in the next line).

and it now lists correct results for CSI, CSI Miami and 24. Tried out the search on yahoo and the result page source text does not have any blank after the </em>.

ThanKs again for the script, great job :-)

[f0xn3t]

thx a lot happy123, you're right
I put blank space in regexp and regsub (below) by mistake here, so it should be:

			if {[regexp {(<div class=yschabstr>.*</div><em class=yschurl>.*</em>|<em class=yschurl>.*</em>)} $line] } {
				regsub {(<div class=yschabstr>.*</div><em class=yschurl>(.*)</em>|<em class=yschurl>(.*)</em>)} $line {\1} page

UPDATE: seems, it was caused by this forum because I can't get rid of the spaces... though in Edit form it's ok funny

[wicked]

/me so dumb

thx for the trick, i was playing around with that code as well, but didnt get it right

all working fine now

[teqno]

Can't get it to work with !ep 24

If someone got a working version then please post it

Thx in advance

[f0xn3t]

Originally posted by teqno
Can't get it to work with !ep 24

If someone got a working version then please post it

Thx in advance

read posts above

[f0xn3t]

here it comes... v4.0

It fully supports TVTome.com Guide (Show, Episode & Person)
... and moreover I get rid of some problems around EPGuides.com Guide, thx to a piece of code from wiki.tcl.tk.. so epguides.com guide uses google.com searching

... AND, I fixed a time limitation (1970-2036?) thx to time.tcl script (by Roger E. Critchlow Jr.) which I found.

d'oh.. it has many new features, so check Changelog

-----------------
CHANGELOG:
- !!! NEW TVTome.com Show Guide (just use command e.g. !ep ... supports releasename parser; more info in Help)
- !!! NEW TVTome.com Episode Guide (just use command e.g. !ep -ep 8-1 -id StargateSG1; more info in Help)
- !!! NEW Releasename parser for TVTome.com Episode Guide (just use command e.g. !ep -ep )
- Tvshowsondvd.com fix (a source code of this web page has changed a bit)
- Epguides.com Guide now uses google.com as it should (thx to a support page at wiki.tcl.tk)
- Improved Epguides.com searching (Logical operators work now to find a show you really need
- Improved Releasename parser (the code has been mostly changed; it should support almost all TV show releases
- Improved error handling (including ERROR 404 & wrong parameters checking in !ep cmd and ERR# + edited error description)
- Improved HTML codes mapping and replacing the codes with a proper char
- Improved anti-flood protection of Person Guide (last word in a note will be shown properly, so no missing letters)
- Unlimited number of records allowed in anti-flood protections of Person Guide
- Unlimited search results allowed in Person Guide
- Epguides.com now supports dates from 1930 to 2029 (thx to the script time.tcl by Roger E. Critchlow Jr. with "double" type of variable; shows older than 1970 will show its info correctly)
- Fixed epguides.com show title bug (!ep trace evidence)
- Fixed NFO link ("&" is "&" now)
- Fixed a bug in epguides.com episode names (!ep dr.vegas ... !ep wonderfalls)
- Fixed IMDB URL in Person Guide, because tvtome.com has changed source code a bit
- Changed name of all procedures (now they start "epg_*"; I've been noticed that there are some problems in dzsbot)
- Script uses HTTP package 2.0 instead of poor $socket
- Added customizable header, body and footer of TVTome.com Show/Episode Guide
- Added Exact match in search results of Person Guide
- Added start-line text for TVTome.com Show/Episode/Person Guide and Epguides.com Guide
- Person Guide is now fully customizable (header, body, footer)
- Better configuration part
-----------------
DOWNLOAD