Hi Stephen,
Last I checked a couple months ago, a robot can't remove any
participants (including itself). The reason being that Google are
still considering how this will work in practice. The only 'robot'
that can remove other robots being Bouncy which is actually more than
a standard robot
Yeah, I think I found that too, antimatter15 - since Bouncy worked for
my human account at the time I tried it on the same wave I was testing
my robot on.
The blip content regex solution won't help robots that don't remove
the triggering text content: e.g. text management robots, notifying,
etc.
Thanks Pamela! =)
On Dec 5, 7:53 am, pamela (Google Employee) pamela...@gmail.com
wrote:
Hi cmdskp-
Apologies; I also am not able to set it manually (even after realizing
that it's actually conv/contentcreationtime). I based its existence
on what Tweety does, and examining the annotations
Yeah, I found my robot tonight went over quota - I expected this to
happen weeks ago, but hoped it'd take longer. It appears to be due to
many attachments in a wave rather than gadgets as expected - though
I'm not sure...the wave that seemed to exhaust my quota was reporting:
This wave can no
Try listening in on DOCUMENT_CHANGED - robots seem to get those from
other robots last I checked a week ago or so.
On Dec 10, 3:24 am, Gabriel Militello gabriel.milite...@gmail.com
wrote:
Tryed to make two robots interact with each other, but found it
impossible.
I don´t know if i´m making the
If you're using Java, you could try loading a synchronized static
global object from an AppEngine database in a global static block.
This will trigger only when the servlet is first loaded by any server
and so you get both the speed of simple variables for requests without
having to call the
I agree, a gadget open in all viewers' browsers can submit a state
change on a short Javascript timer event leading to a horrendous
amount of potential events.
Even worse if you consider malicious possibilities, like having a
robot insert one of these small gadgets in every blip in a wave. Upon
ATM, you can get all blips their content (I've tested in the Java
API) on WAVELET_SELF_ADDED by setting a context=children
content=true in the capabilities.xml (remember to change its version
number there to ensure Wave picks up the change in event tracking)
You could right now just use
I just found the same problem - although not ideal, you can use
DOCUMENT_CHANGED, which fires as soon as someone creates a blip (even
before they type a character).
However, it will fire a lot since it triggers on any changes in the
blip or its gadgets/buttons.
On Nov 27, 5:37 am, Elanchezhiyan
I'm bearing in mind the draft status on that specification document
and the Roadmap section about:
Wave is being actively developed and some features of the
conversation model have not been documented, or may change in the
short term.
Has anyone had any luck getting cursor annotations recently?
9
So, it appears the actual user's cursor position is unavailable
currently - this makes sense now with other answers from Google I've
come across.
On Nov 25, 6:14 pm, cmdskp cmd...@gmail.com wrote:
I'm bearing in mind the draft status on that specification document
and the Roadmap section
lots of code processing tempText or aBlip for
delete()
}
On Nov 26, 4:46 am, Stephen George sfgeo...@gmail.com wrote:
Thanks cmdskp,
I must have missed your response to the other thread, sorry. I'd be
happy to swap notes. Here's my traversal approach. Apologies for
this long message
capabilities.xml as you suggested. Unfortunately that
doesn't seem to have changed the number of blips I can read through
(for my robot at least).
S
On Nov 23, 10:04 pm, cmdskp cmd...@gmail.com wrote:
Good day, Stephen!
Have a look at the reply I just left on that thread How to delete
On Nov 23, 10:04 pm, cmdskp cmd...@gmail.com wrote:
Good day, Stephen!
Have a look at the reply I just left on that thread How to delete all
the blips in a wave except the root blip??
I'd given a shorter version of how the event design for robots is
specified before on the issue tracker
My robot is: clean...@appspot.com
Please test it yourself - just create a new wave and add lots of
nested blips with a lowercase word in each, then add my robot. If any
have missing capitals it'll correct them.
The only types of blips I've not tested are internal comment blips.
So, do not
for the modifications.
On Nov 25, 1:52 am, cmdskp cmd...@gmail.com wrote:
My robot is: clean...@appspot.com
Please test it yourself - just create a new wave and add lots of
nested blips with a lowercase word in each, then add my robot. If any
have missing capitals it'll correct them.
The only types
am, Olreich olre...@gmail.com wrote:
The only remaining question is:
Can one specify both parent and child context, and thus get all
blip_data with a single event?
On Nov 23, 10:53 pm, cmdskp cmd...@gmail.com wrote:
@qMax: As of today, it appears to delete only the blip and its inline
@qMax: As of today, it appears to delete only the blip and its inline
replies (the collapseable blips) - not standard reply children.
@shitu: Last night I was testing deleting of all blips on
WAVELET_SELF_ADDED. It worked fine then, I recursively went through
the children and all were deleted -
Good day, Stephen!
Have a look at the reply I just left on that thread How to delete all
the blips in a wave except the root blip??
I'd given a shorter version of how the event design for robots is
specified before on the issue tracker to help Olreich a few days ago.
In essence you can
purpose is natural.
By design is deliberate, thought out. It's also flippant and means
nothing, and I was grasping for answers, so I took the most meaningful
route. It happens I suppose, though I am sorry to mis-quote you.
On Nov 20, 12:30 am, cmdskp comdudes...@hotmail.com wrote:
I didn't
Still, no worries, text is notorious for getting taken out of
context. More emoticons, I say. :)
On Nov 21, 4:00 am, cmdskp comdudes...@hotmail.com wrote:
They did design it that way - but it doesn't mean they designed it to
not allow it - you're inferring too much from two words
I really should read the rest of a thread before jumping to reply to
someone asking for an answer in a link above...
Anyway, qMax gave the same answer I just did on your issue Olreich. I
didn't star the issue, so didn't get any notification.
As you can imagine, this has been a feature lots of
Yes, the gadget inserting robot works - since I use that. =)
The trick I used to determine when my robot should add a gadget to a
wave it's added to and when not to - is to use a proxy robot address
(e.g. instead of ro...@appspot.com, I use something like robot
+addgad...@appspot.com).
I then
I use state.submitDelta(...) or state.submitValue(...) to update the
state to the server. For clearing values, I've set them to null (as
recommended in the guides) with those functions. Haven't tried
state.reset()
I'd expect updateState to trigger more than once as more state is
sent. It's
. Which event should one
use: BLIP_SUBMITTED, DOCUMENT_CHANGED, ..?
On Nov 17, 5:15 am, cmdskp comdudes...@hotmail.com wrote:
Ah, found where I read it - the draft version of the next Robot API
here:
http://code.google.com/apis/wave/extensions/robots/protocol.html
On Nov 17, 2:45 am
There are many interesting ideas raised by considering robots as tools
to act on a users behalf or in concert with them.
With respect to folder management, I do not see any problem with a
robot requesting movement via the API of waves into or between a
user's folders (deleting can be handled by
by the user later.
On Nov 18, 4:26 am, cmdskp comdudes...@hotmail.com wrote:
There are many interesting ideas raised by considering robots as tools
to act on a users behalf or in concert with them.
With respect to folder management, I do not see any problem with a
robot requesting movement via
GADGET_STATE_CHANGED does not exist currently - but has been mentioned
by Google as coming in a future version of the API, AFAIK, this hasn't
been released yet.
Possibly I read about it on the Wave API Issues.
On Nov 16, 9:58 pm, qMax qwigly...@gmail.com wrote:
And GADGET_STATE_CHANGED is not
Ideally, I would like an event that only gets triggered on a text
change.
The reason being that my robot currently gets a lot of resubmitted
blips because of a gadget in them being clicked on. They can contain
a lot of text which is still the same and has been seen by the robot.
This eats up
*and BLIP_SUBMITTED can still be used for both.
On Nov 12, 3:36 pm, cmdskp comdudes...@hotmail.com wrote:
Ideally, I would like an event that only gets triggered on a text
change.
The reason being that my robot currently gets a lot of resubmitted
blips because of a gadget in them being
If you end up using a Java robot and a static object variable -
remember to ensure you synchronize on it or use a synchronized
object. Otherwise, simultaneous Servlet requests can end up changing
the shared static object's value before each request is done with it.
On Nov 11, 10:05 pm, Niels H.
Hallo,
You can retrieve and store the dice value from the gadget's wave state
like so:
var tempState = wave.getState();
var tempDiceValue = tempState.get(dice); // gets the dice
value
if (! tempDiceValue) { // check we don't already have a value
Check out the Extension Installers guide (http://code.google.com/apis/
wave/extensions/installers/index.html) for adding menu actions under
'New Wave' that will allow you to automatically add robots/
participants when creating a new wave.
Please note, that guide is out-of-date though, you need
Another solution is to alert the user before entering a wave that has
gadgets that 'requireViewerId' to the privacy issue and allow them to
view the wave with those gadgets disabled until they choose to
activate them or edit the wave.
Or disabling all gadgets until activated or wave edit. This
Check your App Engine versions page - ensure the default is your
latest version...and change your capabilities.xml version if you've
changed the events the bot listens to. =)
My bot responds fine to 'BLIP_SUBMITTED' even when other bots are also
altering things (e.g. emoticon replacement).
35 matches
Mail list logo
