UNOFFICIAL incith-google 2.1x (Nov30,2o12)

[speechles]

It shouldn't matter which country you chose to designate as the default. What might be happening tho, is you have two copies of this script in your eggdrop.conf, maybe the old one 1.9.5 and the new 1.9.6, whichever is sourced last will become dominant. If this isn't the problem, and you don't mind.. Can you please post your entire config section as you have it presently, hopefully with that I or someone else can deduce what is wrong for you.

[Elfriede]

MY FAULT - I AM SOOOOO SORRRRY ! I had loaded an old version ((

Sorry sorry.. but thanx for helping !

The Script works perfect for me - 10 of 10 Points !

greets

[transacid]

hey there.
I'm using v1.9.6 but get this error:

Code: Select all

Socket Error accessing 'http://www.google.com/search?q=foo&safe=off&btnG=Search&lr=lang_all&num=10' .. Does it exist?

did i do something wrong?

[speechles]

hey there.
I'm using v1.9.6 but get this error:

Code: Select all

Socket Error accessing 'http://www.google.com/search?q=foo&safe=off&btnG=Search&lr=lang_all&num=10' .. Does it exist?

did i do something wrong?

Nah, that just happens when your bot cannot create a socket to the internet (bot is lagged, low bandwidth, etc). It's one of the few error messages that will always be in english because it's script created (all script created errors will only display in english, as I am an american) not dynamically read from the page like the others.

But on the lighter side, I noticed some problems with recent website changes: !news and !book specifically. These have been fixed and will now return results as usual. I've also fixed the long broken !groups parser, it will now work too. Have a FUN!

Download here or at any v1.9.6 link above, all are updated the same.

[mabus]

Tcl error [incith::google::public_message]: invalid command name "stripcodes"

In the status window of the bot, i get this message whenever i try to use the wikipedia. It's not a problem with what i'm imputting because i've tried to copy paste the exact lines displayed in here on this thread {apparently working}, but all i get is the error message. Anyone else?

[speechles]

Tcl error [incith::google::public_message]: invalid command name "stripcodes"

In the status window of the bot, i get this message whenever i try to use the wikipedia. It's not a problem with what i'm imputting because i've tried to copy paste the exact lines displayed in here on this thread {apparently working}, but all i get is the error message. Anyone else?

It's an easy fix, guess you didnt read the top comments.

Code: Select all

# tested on:                                                    #
#   eggdrop v1.6.17 GNU/LINUX with Tcl 8.4                      #
#   windrop v1.6.17 CYGWIN_NT with Tcl 8.4 (http.tcl v2.4)      #

You need to update your eggdrop to either 1.6.17 or 1.6.18
Eggdrop 1.6.17 Notes: Added a "stripcodes" Tcl command for efficiantly removing control codes from strings (see doc/tcl-commands.doc).

[mabus]

ICK, I hate making Newbie mistakes. My apologies and thank you

[mabus]

Well that didn't work. Brand new Eggdrop V1.6.18 installation, same exact error message. Even installed the latest TCL V8.4.5 and Http TCL V2.5.001.

Any ideas?

[danzigrules]

whats the trick for translate to work?

[danzigrules] !translate en @ it testing

[danzigrules] !translate en@it testing


2] Tcl error [incith::google::public_message]: Illegal characters in URL path

Thanks

[Alchera]

Well that didn't work. Brand new Eggdrop V1.6.18 installation, same exact error message. Even installed the latest TCL V8.4.5 and Http TCL V2.5.001.

Any ideas?

Installed tcl8.4-dev?

[speechles]

whats the trick for translate to work?

[danzigrules] !translate en @ it testing

[danzigrules] !translate en@it testing


2] Tcl error [incith::google::public_message]: Illegal characters in URL path

Thanks

Oh no!
But in all honesty, that translate function is sorta limited at the moment until I figure out encodings. Pretty much the same problem every proc in this script has, it doesn't support russian and other funky charsets properly. It's hard to work a fulltime job and maintain this script at the level of sophistication it needs. It will get done soon mind you, just I can't say when. Since juggling a career and this project, the career always wins. ($$ Money > *)

set url "http://www.google.com/translate_t[b][i]?[/i][/b]langpair=${link}|${desc}"

Now the reason your getting illegal characters in url path, is the silly question mark and pipe (see them in bold above?). Not sure if I can \escape them and get a working url or not, but I'm pretty sure simply enclosing the url in quotes during the ::http::geturl solves it. Find the 'Proc Trans' section and within it do like I have below.

set url "http://www.google.com/translate_t?langp ... nk}|${desc}
set query [::http::formatQuery text $titem langpair "${link}\|${desc}" ]
set http [::http::geturl $url -query $query -timeout [expr 1000 * 10]]

set url "http://www.google.com/translate_t?"
set query [::http::formatQuery text $titem langpair "${link}\|${desc}" ]
set http [::http::geturl "$url" -query $query -timeout [expr 1000 * 10]]

I see my rookie mistake now as clear as rain on a Tuesday morning. I was redundant in my querying. I have ::http::formatQuery build the query correctly, but I left the 'langpair=' hardcoded into the url as well. This creates redundancy and perhaps url errors...ugh. Hopefully this fix solves the problem for the time being.

<speechles> !tr en@it hello everyone at egghelp
<sp33chy> Google says: (en->it) ciao tutto a egghelp
<speechles> !tr it@en ciao tutto a egghelp
<sp33chy> Google says: (it->en) hello all to egghelp

Google translations leave alot to be desired, but the bot can only be as smart as the source site it parses, unfortunately.

[mabus]

Yes, but still getting the error message

[danzigrules]

fixed me all up.


Thanks much

[speechles]

Latest unofficial build: v1.9.6 August 23rd, 2oo7
This only fixes a few broken triggers, but includes all the quick-fix updates listed above. Still working on the promised encoding issues, will have it solved one day. Grab a copy Here or at any of the v1.96 links above. Have a fun

[rosc2112]

Unfortunately, encoding issues are inherent in tcl, until it supports unicode properly :<

Even using string map to put odd chars into the right codes for html, you're going to eventually come across the problem of tcl not handling the higher numbered unicode chars.