Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/wragge/trovebot


https://github.com/wragge/trovebot

code4lib troveaustralia twitter-bot

Last synced: about 2 months ago
JSON representation

Awesome Lists containing this project

README 

        
trovebot

============



Tweeting content from the following Trove zones:



* Journals, articles and data sets

* Books

* Pictures, photos, and objects

* Music, sound, and video

* Maps

* Diaries, letters, and archives



See me in action at [@TroveBot](http://twitter.com/trovebot).



For digitised newspapers see [@TroveNewsBot](http://twitter.com/trovenewsbot).



Built using the [Trove API](http://trove.nla.gov.au/general/api), the Twitter API, and the [AlchemyAPI](http://www.alchemyapi.com/).



### Making a bot query



Simply tweet some keywords or a url to TroveBot. If it's a url, TroveBot will use AlchemyAPI to extract keywords from the page.



TroveBot will look across Trove zones to see where there are matches. It'll then choose a zone at random and return the most relevant result.



### Modifying your bot query



To limit your query to a particular zone or format, simply add one of these facets to your tweet:



* #artwork ('Art work' facet in the picture zone) 

* #article (anything in the article zone)

* #chapter ('Article/Book chapter' facet in the article zone)

* #paper ('Article/Conference paper' facet in article zone)

* #report ('Article/Report' facet in article zone)

* #review ('Article/Review' facet in article zone)

* #book (anything in the book zone)

* #proceedings ('Conference Proceedings' facet in book zone)

* #data ('Data set' facet in article zone)

* #map (anything in the map zone)

* #object ('Object' facet in picture zone)

* #periodical ('Periodical', 'Periodical/Journal, magazine, other', 'Periodical/Newspaper' facets in article zone

* #photo ('Photograph' facet in picture zone)

* #picture (anything in picture zone)

* #poster ('Poster, chart, other' facet in picture zone)

* #archives (anything in the archives zone)

* #score ('Printed music' facet in the music zone)

* #sound (anything in the music zone)

* #interview ('Sound/Interview, lecture, talk' facet in the music zone)

* #music ('Sound/Recorded music' facet in the music zone)

* #thesis ('Thesis' facet in the book)

* #video ('Video' facet in music zone)

* #abcrn (limit to ABC Radio National content in music zone)



You can also add the following filters:



* #aus (limit to 'Australian' content)

* #online (limit to content freely available online)



Both of these are very metadata quality dependent, so they mightn't always be accurate.



By default all keywords are required for a match. To change this you can add the #any tag. This will match records that contain *any* of your keywords.



In theory, your 'keywords' could be anything that works in Trove's simple search box. This includes things like fielded searches -- eg creator:("Wragge, Clement"). 



### Random goodness



If you supply a query TroveBot will normally return the most relevant record it can find. If you want to dig deeper, you can include the #luckydip tag to make the bot deliver a random record from the matching results.



If you just want to play around without specifying a query, you can:



* include the tag #luckydip by itself to get a random record from somewhere in Trove

* include one of the facets above by itself to get a random record from that zone/format.



### Examples



* Search for a book matching 'wragge': *@TroveBot wragge #book*

* Search for an Australian thesis related to a wikipedia page: *@TroveBot http://en.wikipedia.org/wiki/Clement_Lindley_Wragge #thesis #aus*

* Search for a random photo available online: *@TroveBot #photo #online*



### Automatic botness



* Several times a day TroveBot will tweet a random item.



Released under CC0 licence.