Explore the workings of the Trove API (v3)
| parameter | value |
|---|---|
encoding | json |
reclevel | full |
{
"id" : "217926119",
"url" : "https://api.trove.nla.gov.au/v3/newspaper/217926119",
"identifier" : "https://nla.gov.au/nla.news-article217926119",
"heading" : "'Wragge' on Wragge.",
"category" : "Article",
"title" : {
"id" : "902",
"title" : "Glen Innes Examiner and General Advertiser (NSW : 1874 - 1908)"
},
"date" : "1902-12-12",
"page" : "6",
"pageSequence" : "6",
"troveUrl" : "https://nla.gov.au/nla.news-article217926119",
"trovePageUrl" : "https://nla.gov.au/nla.news-page23894193",
"illustrated" : "N",
"wordCount" : "136",
"correctionCount" : "0",
"tagCount" : "0",
"commentCount" : "0",
"listCount" : "1",
"pdf" : [ "https://trove.nla.gov.au/ndp/imageservice/nla.news-page23894193/print" ]
}
Version 3 of the API was made available in July 2023. For more information about the changes made in version 3 see the official release notes. I've also created a summary of breaking changes for users wanting to update their code from version 2 to version 3.
Click on any of the examples below to open them in this console, then modify them to suit. If you want to use them in your own code you'll need to get an API key and add the key to your request either using the key parameter, or setting X-API-KEY in the request headers.
See the Trove API v3 documentation for a full list of available parameters and detailed information on constructing queries.
For more examples, tools, and hacks using the Trove API see the GLAM Workbench.
See the Trove API v3 status page to check for problems.
https://api.trove.nla.gov.au/v3/result?q=wragge&category=all&encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
all
|
possible values are 'all', 'newspaper', 'magazine', 'image', 'research', 'book', 'diary', 'music', 'people', 'list' |
encoding
|
format of results |
json
|
possible values are 'xml' (default) or 'json' |
https://api.trove.nla.gov.au/v3/result?q=wragge&category=book,newspaper&encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
book,newspaper
|
separate multiple categorys with commas |
encoding
|
format of results |
json
|
possible values are 'xml' (default) or 'json' |
https://api.trove.nla.gov.au/v3/result?q=wragge&category=newspaper&encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
possible values are 'xml' (default) or 'json' |
https://api.trove.nla.gov.au/v3/result?q=wragge&category=newspaper&encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query |
wragge
|
|
category
|
category(ies) to search in |
newspaper
|
|
encoding
|
format of results |
json
|
possible values are 'xml' (default) or 'json' |
https://api.trove.nla.gov.au/v3/result?q=wragge&category=newspaper&encoding=json&n=100
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query |
wragge
|
|
category
|
category(ies) to search in |
newspaper
|
|
encoding
|
format of results |
json
|
possible values are 'xml' (default) or 'json' |
n
|
number of results |
100
|
possible values are 0 to 100 (default is 20) |
https://api.trove.nla.gov.au/v3/result?q=wragge&category=newspaper&encoding=json&sortby=dateasc
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query |
wragge
|
|
category
|
category(ies) to search in |
newspaper
|
|
encoding
|
format of results |
json
|
possible values are 'xml' (default) or 'json' |
sortby
|
order of results |
dateasc
|
possible values are 'relevance' (default), 'dateasc', 'datedesc' |
https://api.trove.nla.gov.au/v3/result?q=wragge&category=newspaper&encoding=json&reclevel=full
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
reclevel
|
request additional article metadata |
full
|
Adding reclevel=full adds a number of additional fields to the article metadata, the actual fields added depends on the category. In newspapers and gazettes it adds illustrated, wordCount, correctionCount, tagCount, commentCount, listCount, trovePageUrl, and pdf (which is a link to a PDF of the whole page on which the article is published). In works it adds tagCount, commentCount, and listCount.
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query |
wragge
|
|
category
|
category(ies) to search in |
newspaper
|
|
encoding
|
format of results |
json
|
possible values are 'xml' (default) or 'json' |
include
|
include extra information |
tags
|
possible values include 'tags', 'comments', and 'lists'; repeat parameter for multiple values |
include
|
include extra information |
comments
|
possible values include 'tags', 'comments', and 'lists'; repeat parameter for multiple values |
The accepted values for the include parameter differ across categorys, but some values, like 'tags' and 'comments', are consistent. Other examples are included below.
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query |
wragge
|
|
category
|
category(ies) to search in |
newspaper
|
you can only harvest from one category at a time, so this must be set to a single category such as 'newspaper' |
encoding
|
format of results |
json
|
possible values are 'xml' (default) or 'json' |
bulkHarvest
|
setting this to 'true' ensures that results will stay in the same order when you request subsequent pages; without it results might be duplicated or missing |
true
|
|
s
|
supply a resumption token to get the next page of results |
*
|
use '*' for the first request, then replace with the value of |
Three parameters are important when harvesting a complete result set: category must be set to a single category, bulkHarvest set to 'true', and s must be updated after each request with the value of nextStart. Note that the nextStart value will only appear if there is another page of results to download, so you can use it in your loop as a check to see if you've finished.
https://api.trove.nla.gov.au/v3/result?q=wragge&category=newspaper&encoding=json&l-category=Article
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
l-category
|
category to include |
Article
|
see Trove help documentation, or use the |
This parameter can be used multiple times, however, this will behave like an AND query and only those articles in all of the specified categories will be returned. So adding &l-category=Article&l-category=Advertising will return zero results, as no articles are in both the 'Article' and 'Advertising' categories.
https://api.trove.nla.gov.au/v3/result?q=wragge&category=newspaper&encoding=json&l-state=Victoria
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
l-state
|
place in which articles were published |
Victoria
|
possible values are 'ACT', 'International', 'National', 'New South Wales', 'Northern Territory', 'Queensland', 'South Australia', 'Tasmania', 'Victoria' |
This parameter can be used multiple times. Unlike the category facet, this will behave like an OR query, returning articles published in any of the specified places. So adding &l-state=Victoria&l-state=ACT will return articles published in either Victoria or the ACT.
https://api.trove.nla.gov.au/v3/result?q=wragge&category=newspaper&encoding=json&l-decade=192
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
l-decade
|
limit to articles from this decade |
192
|
192 includes the years 1920 to 1929 |
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
l-decade
|
limit to articles from this decade |
192
|
|
l-year
|
limit to articles from this year ( |
1924
|
limit the results to articles from 1924, the |
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
l-decade
|
limit to articles from this decade |
192
|
|
l-year
|
limit to articles from this year ( |
1924
|
limit the results to articles from 1924 |
l-month
|
limit to articles from this month ( |
3
|
limit the results to articles from March 1924, possible values are numbers from 1 to 12 |
https://api.trove.nla.gov.au/v3/result?q=firstpageseq:1&category=newspaper&encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
firstpageseq:1
|
use the |
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
date:[1942-10-31T00:00:00Z TO 1942-11-30T00:00:00Z]
|
use the |
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge date:[1942-11-01T00:00:00Z TO 1942-11-02T00:00:00Z]
|
use the |
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
https://api.trove.nla.gov.au/v3/result?q=wragge&category=newspaper&encoding=json&l-illustrated=true
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
l-illustrated
|
limit to articles with illustrations |
true
|
accepted values are 'true' or 'false' |
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
l-illustrated
|
limit to articles with illustrations |
true
|
accepted values are 'true' or 'false' |
l-illustrationType
|
limit to articles with this type of illustration |
Photo
|
the |
This parameter can be used multiple times, however, this will behave like an AND query and only those articles with all of the specified illustration types will be returned.
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
l-wordCount
|
filter by number of words in the article |
<100 Words
|
accepted values are '<100 Words', '100 - 1000 Words', or '1000+ Words' |
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge has:tags
|
add |
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
include
|
tags
|
you need to specifically |
https://api.trove.nla.gov.au/v3/result?q=wragge&category=newspaper&encoding=json&include=articletext
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
include
|
add additional information to the record |
articletext
|
this adds the full OCRd text to the record, other possible values are 'tags', 'comments' |
The OCRd text is return as HTML, so depending on you use you might need to strip the tags out.
https://api.trove.nla.gov.au/v3/result?q=%00&category=newspaper&encoding=json&facet=category&n=0
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
%00
|
'%00' is an empty search, so will return everything |
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
n
|
0
|
we don't need results, just the facets, so we can set |
|
facet
|
include facet data |
category
|
break down results by category; see the Trove API documentation for a full list of facets |
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
%00
|
'%00' is an empty search, so will return everything |
category
|
category(ies) to search in (required) |
newspaper
|
|
encoding
|
format of results |
json
|
|
n
|
number of results |
0
|
we don't need results, just the facets, so we can set |
l-decade
|
set the decade |
192
|
needs to be set to find the totals by year |
facet
|
include facet data |
year
|
break down results by year in the given decade; see the Trove API documentation for a full list of facets |
For more examples using the year facet see Visualise Trove newspaper searches over time in the GLAM Workbench.
https://api.trove.nla.gov.au/v3/result?q=weather&category=book&encoding=json&l-format=Thesis
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
weather
|
|
category
|
category(ies) to search in (required) |
book
|
|
encoding
|
format of results |
json
|
|
l-format
|
format or type of work |
Thesis
|
see the Trove help documentation for a list of formats |
This parameter can be used multiple times, however, this will behave like an AND query and only those articles in all of the specified formats will be returned.
https://api.trove.nla.gov.au/v3/result?q=weather NOT format:Book &category=image&encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
weather NOT format:Book
|
by adding |
category
|
category(ies) to search in (required) |
image
|
|
encoding
|
format of results |
json
|
You can use the format index in both the query and as a facet, that means you can say that you don't want works with particular formats. This is handy for filtering out noise.
https://api.trove.nla.gov.au/v3/result?q=weather date:[* TO 1900]&category=book&encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
weather date:[* TO 1900]
|
using the |
category
|
category(ies) to search in (required) |
book
|
|
encoding
|
format of results |
json
|
https://api.trove.nla.gov.au/v3/result?q=weather imageInd:thumbnail &category=image&encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
weather imageInd:thumbnail
|
set |
category
|
category(ies) to search in (required) |
image
|
|
encoding
|
format of results |
json
|
The actual thumbnail link is included in the identifier field with a linktype of 'thumbnail'. To ensure you get all the possible values for identifier you can also add &include=links to your query.
https://api.trove.nla.gov.au/v3/result?q=wragge&category=image&encoding=json&include=workversions
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
wragge
|
|
category
|
category(ies) to search in (required) |
image
|
|
encoding
|
format of results |
json
|
|
include
|
format of results |
workversions
|
include the metadata for each individual version of this work |
The way that versions are grouped as works in Trove can be quite confusing. The work level metadata is generally quite consistent, but it can leave out useful metadata that's attached to an individual version. The version metadata is richer, but is also less consistent, so you have to be prepared for some trial and error if you want to extract information from it.
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
weather fullTextInd:y
|
set |
category
|
category(ies) to search in (required) |
magazine
|
|
encoding
|
format of results |
json
|
|
include
|
workversions
|
the full text is buried down in the version-level metadata, so you need to add this parameter if you want to actually get to it |
The fullTextInd index seems quite misleading as it includes articles from commercial databases where only a truncated sample of the full text is actually available, and electronic legal deposit works where access to the text is restricted. It would seem better to combine this indicator with a search for "nla.obj" or nuc:ANL:DL to try and limit to resources digitised by the NLA and partners.
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
weather
|
|
category
|
category(ies) to search in (required) |
magazine
|
|
encoding
|
format of results |
json
|
|
l-title
|
limit to this journal title |
The bulletin
|
There's no controlled list of journal titles to use with this facet, so you probably need to get the values from the web interface.
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
weather "https://nla.gov.au/nla.obj-68375465"
|
|
category
|
category(ies) to search in (required) |
magazine
|
|
encoding
|
format of results |
json
|
|
l-format
|
limit to this format |
Article
|
limit the results to articles so we don't get the records for the periodicals themselves |
An alternative to using the title facet is to include the identifier of the parent journal in the q parameter. This is how the Trove Digitised Journals app works.
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
the search query (optional) |
rights:Free
|
see the Trove help documentation for a full list of possible values for |
category
|
category(ies) to search in (required) |
image
|
|
encoding
|
format of results |
json
|
|
l-format
|
format of work |
Photo
|
|
include
|
workversions
|
the rights statements are attached to versions, so to see them you need to ask for the full versions data |
This facet only works with records where the copyright status is specified in the record. Some out-of-copyright photographs might be missing from results because their status has not been identified. A search for photos created before 1955 using date[* TO 1954] might find additional out-of-copyright images.
Facets break the result set into sections and can provide a useful overview. Note that there are limits on the number of facets returned, so they might not include the whole result set.
https://api.trove.nla.gov.au/v3/result?category=all&facet=decade&n=0&encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
category
|
category(ies) to search in (required) |
all
|
|
encoding
|
format of results |
json
|
|
facet
|
facet to return |
decade
|
|
n
|
number of results |
0
|
We just want facets, not the list of results. |
https://api.trove.nla.gov.au/v3/result?category=all&facet=partnerNuc&n=0&encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
category
|
category(ies) to search in (required) |
all
|
|
encoding
|
format of results |
json
|
|
facet
|
facet to return |
partnerNuc
|
|
n
|
number of results |
0
|
We just want facets, not the list of results. |
Contributors are identified by NUC identifiers.
Article or work identifiers are supplied as part of the url rather than as a query parameter. For example, https://api.trove.nla.gov.au/v3/newspaper/41697877 returns the details of the article with the identifier 41697877.
| parameter | parameter note | value | value note |
|---|---|---|---|
encoding
|
format of results |
json
|
|
reclevel
|
amount of detail in record |
full
|
|
include
|
include extra information |
articletext,tags,comments
|
include the OCRd text, as well as any tags or comments, in the record |
| parameter | parameter note | value | value note |
|---|---|---|---|
encoding
|
format of results |
json
|
|
reclevel
|
amount of detail in record |
full
|
|
include
|
include extra information |
workversions,tags,comments
|
include metadata from all grouped versions, as well as any tags or comments |
https://api.trove.nla.gov.au/v3/list/43805?encoding=json&reclevel=full&include=listitems
| parameter | parameter note | value | value note |
|---|---|---|---|
encoding
|
format of results |
json
|
|
reclevel
|
amount of detail in record |
full
|
|
include
|
include extra information |
listitems
|
include details of all items in the list |
You only get the brief record version for each of the list items, so to get extra information, such as OCRd text for a newspaper article, you'll have to request each article individually. Also, the list items are not paginated, so if you have a long list this will return a lot of data.
There's a bug in the API so that a request for newspaper titles also returns the gazette titles. See the GLAM Workbench for a workaround.
https://api.trove.nla.gov.au/v3/newspaper/titles?encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
encoding
|
format of results |
json
|
https://api.trove.nla.gov.au/v3/gazette/titles?encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
encoding
|
format of results |
json
|
https://api.trove.nla.gov.au/v3/newspaper/titles?encoding=json&state=vic
| parameter | parameter note | value | value note |
|---|---|---|---|
encoding
|
format of results |
json
|
|
state
|
limit to this state |
vic
|
possible values are 'nsw', 'act', 'qld', 'tas', 'sa', 'nt', 'wa', 'vic', 'national' |
https://api.trove.nla.gov.au/v3/newspaper/title/35?encoding=json&include=years
| parameter | parameter note | value | value note |
|---|---|---|---|
encoding
|
format of results |
json
|
|
include
|
years
|
list the years available on Trove, with the number of issues per year |
The newspaper identifier is included in the url as indicated. This example will return details of the Sydney Morning Herald which has an identifier of '35'.
| parameter | parameter note | value | value note |
|---|---|---|---|
encoding
|
format of results |
json
|
|
include
|
years
|
this needs to be included to retrieve the full issue information |
|
range
|
date range of issues to return |
18420101-18421231
|
dates are in YYYYMMDD format; this example will return details of all issues of the Sydney Morning Herald published in 1842 |
The issue details include a publication date and a url which redirects to the first page of the issue.
Although this endpoint is included in the technical documention, as of 23 May 2023 it was returning no results. Hopefully it'll start working soon!
https://api.trove.nla.gov.au/v3/magazine/titles?encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
encoding
|
format of results |
json
|
Organisations can be nested under other organisations (as children), this makes the data structure a bit complex.
https://api.trove.nla.gov.au/v3/contributor?q=&encoding=json
| parameter | parameter note | value | value note |
|---|---|---|---|
q
|
search query (optional) |
|
leave blank for everything |
encoding
|
format of results |
json
|
https://api.trove.nla.gov.au/v3/contributor/ANL?encoding=json&reclevel=full
| parameter | parameter note | value | value note |
|---|---|---|---|
encoding
|
format of results |
json
|
|
reclevel
|
amount of detail to return |
full
|
The organisation's identifier is included in the url as indicated. This example retrieves details for the National Library of Australia, whose identifier is 'ANL'.