Add filetype (ex. .jpg, .png, .tiff) parameter to Image Search API2 votes
Support for getting search results with actual links (not https://www.bing.com/cr?... prefixed ones)
For my use case, I want to get the actual URL's of the given news items (or websearch items). This was the default in v2, but the documentation is unclear in v5 on how to go about it.3 votes
I am using the current Bing API v2 to search for feed results only. This works fantastic by limiting the results with the WebFileType parameter like this:
As far as I can tell, there is nothing that replaces this functionality in the new v5 API - this is very sad, as I will have to move away from Bing now, despite being really satisfied with the results I got.
So please, implement this (and other missing features the old API supported) - it's not very customer friendly to take away the old version when you can't offer it's features in the new one (yet).
So please, implement this (and other missing features the old API supported) - it's not very customer friendly to take away the old version when you can't offer it's features…3 votes
This functionality will not supported on the new version of the API.
But the Bing Custom Search API might help.
When site visitors perform a web search on our website using the Web Search API it would be nice if the API could track and maintain a database of searched terms done through our site. This way we could see what the most popular terms searched by our site's visitors are each day and how many visitors searched for that same keyword.3 votes
News Photos that are 16:9 will scale up better in the News Answer of the Web Search Results on the mobile screen. The current aspect ratio of the photos is taller than it is wide and it doesn't scale well when trying to make the photos take up more of the screen's width.1 vote
We desperately need a Places or Phonebook Results in the Web Search so that mobile users that search for local businesses can find business information (web address, list of locations, map of locations, call button, hours of operation, directions) through the Web API.5 votes
Try the Bing Local Business Search service.
For example a URL such as https://www.microsoft.com/bingapis/searchapi in the Search Results would be shortened to microsoft.com > bing apis > search2 votes
This is so that users can update their location manually if their location is identified incorrectly based on their I.P. address and the search results are displaying irrelevant local search results based on that incorrect location2 votes
For example when there are multiple Yahoo Answers related to the keyword searched they should display as expanded deep links within the same search results entry, as well as other popular forums like Reddit3 votes
For example when doing a search for Match.com the app for Match at the respective app store as well as common alternative apps should come up in the search results2 votes
Hi, we have deep links to Wikipedia. Do you have example queries where these are not returning results.
Thank you for the sharing this idea. There are limitations that prevent sharing encumbrance data due to legal and licensing requirements.
The new Bing Search API 5.0 does not support SortyBy Date for search results, though this functionality is supported in the previous Bing Search API 2.0.
This is a very useful capability, without it one can not query for all news search results from the past year (or any other date window longer than one month)9 votes
I'm seeing weird behaviour when paginating results. It is impossible to figure out when to stop asking for more results.
I cannot use the number of duplicate results returned as a way of figuring out when to stop either. (The first page returned 68 unique results, the 2nd 7, the 3rd 2 and the 4th zero results; but page 7 gave me 1 result)
The really annoying part was at page 119 where it suddenly returned 33 new results. (see attached for details)
Why are we seeing this weird behaviour (see 3), and what is the recommended method of paginating the results.
I cannot use the number of duplicate results returned as a way of figuring out when to stop either. (The first page returned 68 unique results, the 2nd 7, the 3rd 2…
The &responseFilter= value is missing from the docs into the new api.1 vote
1) Search using "100,000.51 euros to dollars" in Bing
2) Note that the from value is correctly shown as 100000.51
3) Do the same search using the Web Search API
4) Examine the JSON object returned from this query and look at the "currency" member which contains the values for the currency card
Expect: "fromValue": 100000.51
Actual: "fromValue": 100000.51 vote
Using intitle: before a keyword appears to do something, but doesn't actually restrict results to those with that keyword in the title. This appears to be a problem with bing search itself, not just the api. For instance, see
Most results include 4l60e in the title, but not all do.1 vote
As described. Would be nice to know when pages were created, not just when Bing last looked at them. A separate field for the last time the page changed would be great too, although not as important.
Also, the ability to sort results by creation (and ideally last-updated) date is really critical.9 votes
We would like to know what type of scenarios you have that would help us understand to see if there is something we can provide.
It looks like the web results api currently returns 50 results per page max. It would be nice to increase that to at least 100, if not more (or at least to count multiple page requests with the same query as a single transaction).
Speaking of which, when will pricing be announced?2 votes
As I understand it, your URL has a limit of 2048 characters. Obviously when searching a small handful of sites or the entire web that is more than sufficient. However, if a user wants to curate a large but selective set of sources, the space will quickly be filled up with site:one.example.com OR site:two.example.com OR etc..
One way to get around this is what Google does with Custom Search, by allowing the user to specify a url to a file, which is expected to contain some XML including a set of URL patterns. The search results are then limited to those URLs. I'm sure there are other ways. Alternatively, although less elegantly, you could just allow much longer input urls.
One way to get around this is what Google does with Custom Search, by allowing the user to specify a url to a file, which is expected to contain some XML including a set of URL patterns. The search results are then limited to…3 votes
- Don't see your idea?