*Developers may request up to 1000 results in a single API call
I have to say though that the behavior of providing results 990 - 1000
repetitively after that is quite odd.
Hello Mike,
The current BOSS design purposely has the following behavior:
* The universal argument of "start=___" is limited to a max value of
1000.
* If a "start=___" value is greater than 1000, it is always replaced
with a value of "990".
----------
The above design decisions explain why you are seeing the results you
are seeing, with your experiments.
There are currently no plans to modify this above behavior. But if at
some point we do, we will let you know.
Thanks for the responses. I have one more question with the following
background. Is there a limit of 1000 images returned? Seems so when you
read the following.
Image query = "sci fi spaceship"
deep hits = 9418
total hits = 9418
I have changed the code to loop through and gather images in groups of
50 based on the max count = 50 per request by incrementing the :start
by 50 each time. I have noticed that I am getting duplicate images
(image.url, image.referererurl...) starting at :start = 1000.
Regardless of asking for 1000 --> 1050 I am actually getting 5 sets
of 990 --> 1000. This pattern repeats - request for 1100 --> 1150
returns 5 sets of 990 --> 1000, and so on. I confirmed the above by
storing all image.url and image.refererurl in a hash and comparing all
new entries for its existence in the hash. As soon as I hit image 1000
the following images and corrresponding image.refererurl already exist
in the hash.
I am using the ruby gem eshopworks-rboss (0.1.9) to
access the
api
and if I set the count = anything greater than 50 it still only returns
50. The result set totalhits = 9986. If I set count=5 I get 5, count =
20 I get 20. If count = is set to anything greater than 50 I always
only get 50.
Is anybody else having this problem. Their is nothing in the
documentation to state that there is a limit on the number returned.
Mike
--
*************************
Mike Johnson
Dir Professional Services
Image Fortress Corp.
978-846-0139 (work)
--
*************************
Mike Johnson
Dir Professional Services
Image Fortress Corp.
978-846-0139 (work)
--
*************************
Mike Johnson
Dir Professional Services
Image Fortress Corp.
978-846-0139 (work)
Hi,
As far as I know there is no option to restrict BOSS searches to a single
country like "pages from Australia". When will this option be available?
This feature is really important for people coming from countries like UK,
Australia, Canada, Switzerland and Austria.
Best,
ChrischiK
Thanks for the response. I had actually been playing around with very similar
ideas myself and managed to display calcualated/aggragated results fairly
satisfactorily. In the case of my app, this seems like needless work, though:
clearly, Yahoo does all the calculating on its own - hence the way the 'related
concepts' are displayed when using regular search - and I wish there were a way
to access those 'related concepts' directly through BOSS. My app is going to be
run on Google App Engine: hence, the need for being super conservative with
using CPU resources.
- Shailen
--- In ysearchboss@yahoogroups.com, "Ted Drake" <tdrake@...> wrote:
>
> I do something like this. It's actually fairly easy with all of the
> information you can get from BOSS.
>
> Here's an example page: http://insiderarts.com/mona+lisa.html
>
>
>
> I'm building the related terms for each result by analyzing:
>
> 1. The keyterms associated by boss
>
> 2. The delicious tags used by people that have saved a link in
> Delicious
>
> 3. Microformatted tags used by the web site via the search monkey
> option.
>
> I create an array of all the associated keyterms and then look to see if
> any have been used more than once. Those are then moved to the front and
> I list the top four keyterms for each result.
>
>
>
> I do a similar logic to create the related search terms in the right
> column, only I use the combined values for all of the results for the
> page. I also use that uber array to resort the results. Notice the
> first result is for the National Gallery of Art and not the louvre. That
> is because the #1 keyterm for the national gallery's page is "Mona Lisa"
> and that matches the number 1 keyterm for the page. The number one
> keyterm for the louvre is closer look.
>
>
>
> Here's an example where the top 5 matched yves klein as their number 1
> keyterm: http://insiderarts.com/yves+klein.html
>
>
>
> Here's a blog post I wrote about this concept:
> http://developer.yahoo.net/blog/archives/2009/04/boss_updates.html
>
>
>
> Ted
>
>
>
> From: ysearchboss@yahoogroups.com [mailto:ysearchboss@yahoogroups.com]
> On Behalf Of tulishailen
> Sent: Monday, December 07, 2009 11:57 AM
> To: ysearchboss@yahoogroups.com
> Subject: [ysearchboss] "Explore related concepts'
>
>
>
>
>
> Yahoo search has a very useful 'Explore related concepts' area. Is there
> any way to replicate the results using BOSS? I've been playing around
> with view=keyterms, but that results in suggested searches/keyterms for
> every retrieved result. I'm looking for the sort of functionality
> provided by the 'Explore related concepts' service: a handful of
> keywords related to the query in general. Anyone know of a way to do
> this with BOSS? Many thanks in advance.
>
> - Shailen Tuli
>
I’m building the related terms for each result by
analyzing:
1.The keyterms associated by boss
2.The delicious tags used by people that have saved a link in
Delicious
3.Microformatted tags used by the web site via the search monkey
option.
I create an array of all the associated keyterms and then look
to see if any have been used more than once. Those are then moved to the front
and I list the top four keyterms for each result.
I do a similar logic to create the related search terms in the right
column, only I use the combined values for all of the results for the
page. I also use that uber array to resort the results. Notice the first
result is for the National Gallery of Art and not the louvre. That is because the
#1 keyterm for the national gallery’s page is “Mona Lisa” and
that matches the number 1 keyterm for the page. The number one keyterm for the
louvre is closer look.
From:
ysearchboss@yahoogroups.com [mailto:ysearchboss@yahoogroups.com] On Behalf
Of tulishailen Sent: Monday, December 07, 2009 11:57 AM To: ysearchboss@yahoogroups.com Subject: [ysearchboss] "Explore related concepts'
Yahoo search has a very useful 'Explore related
concepts' area. Is there any way to replicate the results using BOSS? I've been
playing around with view=keyterms, but that results in suggested
searches/keyterms for every retrieved result. I'm looking for the sort of
functionality provided by the 'Explore related concepts' service: a handful of
keywords related to the query in general. Anyone know of a way to do this with BOSS?
Many thanks in advance.
Yahoo search has a very useful 'Explore related concepts' area. Is there any way
to replicate the results using BOSS? I've been playing around with
view=keyterms, but that results in suggested searches/keyterms for every
retrieved result. I'm looking for the sort of functionality provided by the
'Explore related concepts' service: a handful of keywords related to the query
in general. Anyone know of a way to do this with BOSS? Many thanks in advance.
- Shailen Tuli
i am able to extract all the results through the folowing code. But the results may get repeated. So could u plz tell me how can i avoid this. Moreover could you plz tell me
To accomplish finding search results that contain the text "FAQ" within
the URL's themselves, you can use the "inurl:xxxxx" query filter, such
as this BOSS API example (but with the APP ID taken out, on purpose):
...and then do a Find in your browser for the web page text of:
"Related Content for a Blog"
...to then see this text:
>>
>> BOSS allows you to use most of the common query filters
>> available in an advanced search. This list includes:
>>
>> * inurl (search for the term in a web site's url)
----------
I'll try and pursue getting this added to our official public BOSS API
docs, but cannot guarantee.
I would not be able to help you specifically on your PHP code (your
follow-up email, after this one). But hopefully the above gets you on
the correct track, or if you still need help on your PHP code, somebody
else in this user group could assist you with that.
----------
Hope this helps,
Jay
kailash sati wrote:
Hi Team
I want to extract all the urls which contain word "FAQ "or
"queries" in them.
Hi everyone,
I am new in yahooboss. I am developing a search engine with some moderation for
one of my course project. Can anyone help me regarding how I can use the dates
of the search result and manipulate them like show them in a graph or something.
I am doing it in php.
Thanks for any kind of help.
Renesa
*Developers may request up to 1000
results in a single API call
I have to say though that the behavior of providing results 990 - 1000 repetitively after that is quite odd.
Mike
On Thu, Dec 3, 2009 at 5:19 PM, Mike Johnson <mike.johnson@...> wrote:
Hello Jay,
Thanks for the responses. I have one more question with the following background. Is there a limit of 1000 images returned? Seems so when you read the following.
Image query = "sci fi spaceship"
deep hits = 9418 total hits = 9418
I have changed the code to loop through and gather images in groups of
50 based on the max count = 50 per request by incrementing the :start
by 50 each time. I have noticed that I am getting duplicate images (image.url, image.referererurl...) starting at :start = 1000. Regardless of asking for 1000 --> 1050 I am actually getting 5 sets of 990 --> 1000. This pattern repeats - request for 1100 --> 1150 returns 5 sets of 990 --> 1000, and so on. I confirmed the above by storing all image.url and image.refererurl in a hash and comparing all new entries for its existence in the hash. As soon as I hit image 1000 the following images and corrresponding image.refererurl already exist in the hash.
I am using the ruby gem eshopworks-rboss (0.1.9) to access the
api
and if I set the count = anything greater than 50 it still only returns
50. The result set totalhits = 9986. If I set count=5 I get 5, count =
20 I get 20. If count = is set to anything greater than 50 I always
only get 50.
Is anybody else having this problem. Their is nothing in the
documentation to state that there is a limit on the number returned.
Mike
--
*************************
Mike Johnson
Dir Professional Services
Image Fortress Corp.
978-846-0139 (work)
-- *************************
Mike Johnson Dir Professional Services Image Fortress Corp. 978-846-0139 (work)
-- *************************
Mike Johnson Dir Professional Services Image Fortress Corp. 978-846-0139 (work)
Thanks for the responses. I have one more question with the following background. Is there a limit of 1000 images returned? Seems so when you read the following.
Image query = "sci fi spaceship"
deep hits = 9418 total hits = 9418
I have changed the code to loop through and gather images in groups of
50 based on the max count = 50 per request by incrementing the :start
by 50 each time. I have noticed that I am getting duplicate images (image.url, image.referererurl...) starting at :start = 1000. Regardless of asking for 1000 --> 1050 I am actually getting 5 sets of 990 --> 1000. This pattern repeats - request for 1100 --> 1150 returns 5 sets of 990 --> 1000, and so on. I confirmed the above by storing all image.url and image.refererurl in a hash and comparing all new entries for its existence in the hash. As soon as I hit image 1000 the following images and corrresponding image.refererurl already exist in the hash.
I am using the ruby gem eshopworks-rboss (0.1.9) to access the
api
and if I set the count = anything greater than 50 it still only returns
50. The result set totalhits = 9986. If I set count=5 I get 5, count =
20 I get 20. If count = is set to anything greater than 50 I always
only get 50.
Is anybody else having this problem. Their is nothing in the
documentation to state that there is a limit on the number returned.
Mike
--
*************************
Mike Johnson
Dir Professional Services
Image Fortress Corp.
978-846-0139 (work)
-- *************************
Mike Johnson Dir Professional Services Image Fortress Corp. 978-846-0139 (work)
I am using the ruby gem eshopworks-rboss (0.1.9) to access the
api
and if I set the count = anything greater than 50 it still only returns
50. The result set totalhits = 9986. If I set count=5 I get 5, count =
20 I get 20. If count = is set to anything greater than 50 I always
only get 50.
Is anybody else having this problem. Their is nothing in the
documentation to state that there is a limit on the number returned.
Mike
--
*************************
Mike Johnson
Dir Professional Services
Image Fortress Corp.
978-846-0139 (work)
The following is my code. It returns only max 10 results. If I search on yahoo.com it gives me lots of reults. Iam trying to display all the urls which has "faq" word in their url.
I am using the ruby gem eshopworks-rboss (0.1.9) to access the api
and if I set the count = anything greater than 50 it still only returns
50. The result set totalhits = 9986. If I set count=5 I get 5, count =
20 I get 20. If count = is set to anything greater than 50 I always
only get 50.
Is anybody else having this problem. Their is nothing in the
documentation to state that there is a limit on the number returned.
Mike
-- *************************
Mike Johnson Dir Professional Services Image Fortress Corp. 978-846-0139 (work)
I am using the ruby gem eshopworks-rboss (0.1.9) to access the api
and if I set the count = anything greater than 50 it still only returns
50. The result set totalhits = 9986. If I set count=5 I get 5, count =
20 I get 20. If count = is set to anything greater than 50 I always
only get 50.
Is anybody else having this problem. Their is nothing in the
documentation to state that there is a limit on the number returned.
I am using the ruby gem eshopworks-rboss (0.1.9) to access the api and if I set
the count = anything greater than 50 it still only returns 50. The result set
totalhits = 9986. If I set count=5 I get 5, count = 20 I get 20. If count = is
set to anything greater than 50 I always only get 50.
Is anybody else having this problem. Their is nothing in the documentation to
state that there is a limit on the number returned.
Mike
We want to use boss to search an index of 900+ websites. We get
errors about the length of the HTTP request when we attempt this. Is
there any way for BOSS to search a large number of sites in a single
request?
Regarding #1 below: I wouldn't be able to help you with this, but
hopefully somebody else here can.
Regarding #2 below: Yes, you would need to take care of this yourself,
in your php code. BOSS does not have an API parameter / ability to
accomplish this.
I have just set the boss search on a site I am developing, using
the WebAssist extension.
1. I would like to change the appearance of the returned keyword form
<b> to an alternative css tag.
2. I noticed that if the search word is matched exactly, it is returned
in the result set, but part of the word does not.
is there a setting to add wildcards to the search parameter, or do i
need to do this in php.
Thanks. It was a mistake on my part that caused my error. I did
not encode the web sites used in the search properly and therefore the query
did not work.
Thanks,
Izzy
From:
ysearchboss@yahoogroups.com [mailto:ysearchboss@yahoogroups.com] On Behalf
Of Ted Drake Sent: Tuesday, December 01, 2009 11:40 PM To: ysearchboss@yahoogroups.com Subject: RE: [ysearchboss] New to BOSS - and I have some problems
What host are you using?
Is it Dreamhost?
They may not allow you to make a direct
call in your php code. You may need to use a curl request instead of just
loading it.
It would help if you include the request
you are making, but delete your app id from the example.
I use this function to get the request
and turn it into simplexml
function My_simplexml_load_file($URL)
{
$ch = curl_init($URL);
curl_setopt($ch,
CURLOPT_RETURNTRANSFER, 1);
curl_setopt($ch, CURLOPT_HEADER,
0);
$xml = simplexml_load_string(curl_exec($ch));
curl_close($ch);
return $xml;
}
Good luck
Ted DRAKE
From:
ysearchboss@yahoogroups.com [mailto:ysearchboss@yahoogroups.com] On Behalf
Of Izzy Sent: Tuesday, December 01, 2009 1:36 PM To: ysearchboss@yahoogroups.com Subject: [ysearchboss] New to BOSS - and I have some problems
Hello All,
I just starting getting into BOSS - and made some nice progress. However, I am
now stuck.
I build a search and issue it through BOSS via PHP and get back no results,
HOWEVER if I copy the same BOSS search to my browser window I get results.
I recommended DMX as an ads platform last year. This was
when we were trying to figure out how to place ads on our sites while still
complying with the original terms of BOSS. BOSS has since removed the
restriction of only using Yahoo ads.
Right Media, the ad exchange
that was acquired
by Yahoo in 2007 for a reported $680 million in cash and stock, has
recently seen its founder Michael
Walrath walk away from
his job as Yahoo’s SVP of Advertising Strategy.
When the announcement of Walrath’s
departure was made, Yahoo
at the same time declared that it would be shifting
strategy for Right Media
by placing more emphasis on serving as a high-quality, premium marketplace
rather than contuining to be an exchange platform for high volumes of ad
networks, including many small ones (zing).
They’re now making good on that promise by confirming that Right
Media’s DMX (Direct Media
Exchange) platform will be shut down on January
31, 2010.
They may not allow you to make a direct call in your php code.
You may need to use a curl request instead of just loading it.
It would help if you include the request you are making, but
delete your app id from the example.
I use this function to get the request and turn it into
simplexml
function My_simplexml_load_file($URL)
{
$ch = curl_init($URL);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
curl_setopt($ch, CURLOPT_HEADER, 0);
$xml = simplexml_load_string(curl_exec($ch));
curl_close($ch);
return $xml;
}
Good luck
Ted DRAKE
From:
ysearchboss@yahoogroups.com [mailto:ysearchboss@yahoogroups.com] On Behalf
Of Izzy Sent: Tuesday, December 01, 2009 1:36 PM To: ysearchboss@yahoogroups.com Subject: [ysearchboss] New to BOSS - and I have some problems
Hello All,
I just starting getting into BOSS - and made some nice progress. However, I am
now stuck.
I build a search and issue it through BOSS via PHP and get back no results,
HOWEVER if I copy the same BOSS search to my browser window I get results.
Hello All,
I just starting getting into BOSS - and made some nice progress. However, I am
now stuck.
I build a search and issue it through BOSS via PHP and get back no results,
HOWEVER if I copy the same BOSS search to my browser window I get results.
What gives?
I am using PHP and JSON.
Thanks,
Izzy
I wish you all the best in your new future. I thank you for what you have done. It has made a difference.
Best regards and Cheers,
Dev
On Tue, Dec 1, 2009 at 6:10 PM, Jacob B <jacobbbc@...> wrote:
Hello Vik Singh,
Thanks a lot for all your support. Never thought you would leave Yahoo!. Thought you are there always. Anyway best of luck for your future. I hope your depart will not adversely affect Yahoo BOSS and its developers.
Thanks and Regards,
Jacob Baby
--- In ysearchboss@yahoogroups.com, Matthew Terenzio <mterenzio@...> wrote:
>
> Thanks for all your work and Good Luck.
>
> On Mon, Nov 30, 2009 at 5:14 PM, Vik Singh <viksi@...> wrote:
>
> >
> >
> > Hey everyone
> >
> > I�m leaving Yahoo! today to explore the entrepreneurial world but just
> > wanted to say thank you for all the BOSS support. It�s because of you guys
> > we were able to push the boundaries of the platform. Let�s continue to root
> > for its future because honestly the full potential of BOSS has yet to be
> > reached: open sourcing key search code, multiple revenue making
> > opportunities, indexing on the server, bulkload site restricts, web crawl
> > access, etc. to name just a few innovations left to be done. I�m hoping to
> > see the day when full open search exists because I�m a firm believer that
> > this will produce insane innovation and real competition in the search and
> > general web spaces.
> >
> > I�m going to definitely miss working with the many great people that make
> > BOSS happen each day. Thank you again for this awesome opportunity.
> >
> > -- Vik Singh
> >
> >
>
Hello Vik Singh,
Thanks a lot for all your support. Never thought you would leave Yahoo!. Thought
you are there always. Anyway best of luck for your future. I hope your depart
will not adversely affect Yahoo BOSS and its developers.
Thanks and Regards,
Jacob Baby
--- In ysearchboss@yahoogroups.com, Matthew Terenzio <mterenzio@...> wrote:
>
> Thanks for all your work and Good Luck.
>
> On Mon, Nov 30, 2009 at 5:14 PM, Vik Singh <viksi@...> wrote:
>
> >
> >
> > Hey everyone
> >
> > I�m leaving Yahoo! today to explore the entrepreneurial world but just
> > wanted to say thank you for all the BOSS support. It�s because of you guys
> > we were able to push the boundaries of the platform. Let�s continue to
root
> > for its future because honestly the full potential of BOSS has yet to be
> > reached: open sourcing key search code, multiple revenue making
> > opportunities, indexing on the server, bulkload site restricts, web crawl
> > access, etc. to name just a few innovations left to be done. I�m hoping to
> > see the day when full open search exists because I�m a firm believer that
> > this will produce insane innovation and real competition in the search and
> > general web spaces.
> >
> > I�m going to definitely miss working with the many great people that make
> > BOSS happen each day. Thank you again for this awesome opportunity.
> >
> > -- Vik Singh
> >
> >
>
Thank you Vik for everything you have done at Yahoo! for YBOSS. I hope you're
not leaving Yahoo! because somethings regarding YBOSS are happening behind the
scenes after the Microsof-Yahoo deal.
I wish you all the best in your future projects.
Regards,
Reza Sardeha
--- In ysearchboss@yahoogroups.com, Vik Singh <viksi@...> wrote:
>
> Hey everyone
>
> I¹m leaving Yahoo! today to explore the entrepreneurial world but just
> wanted to say thank you for all the BOSS support. It¹s because of you guys
> we were able to push the boundaries of the platform. Let¹s continue to root
> for its future because honestly the full potential of BOSS has yet to be
> reached: open sourcing key search code, multiple revenue making
> opportunities, indexing on the server, bulkload site restricts, web crawl
> access, etc. to name just a few innovations left to be done. I¹m hoping to
> see the day when full open search exists because I¹m a firm believer that
> this will produce insane innovation and real competition in the search and
> general web spaces.
>
> I¹m going to definitely miss working with the many great people that make
> BOSS happen each day. Thank you again for this awesome opportunity.
>
> -- Vik Singh
>