Indexes, videos, and tasks
The GET methods of the /indexes, /indexes/videos, and /tasks endpoints support filtering.
To filter the results of an API query, you must specify one or more conditions after the endpoint in the URL of the call. To separate the endpoint and the conditions, you must use a ? symbol. If you add multiple conditions, you must separate them using the & symbol.
A condition consists of an expression that specifies:
- The field you want to filter on
- An operator
- The value to compare to
To indicate the relationship between a field and its value, you can use the exact match or comparison operators.
Exact match operator
Use the exact match operator (=) to match only the results that equal the value you specify. The syntax is as follows: <field>=<value>.
Comparison operators
Use the comparison operators ([lte]= and [gte]=) to match based on the arithmetic comparison. The syntax is as follows <field>[lte]=<value> and <field>[gte]=<value>.
Filter composition
You can add multiple filters to a single query. For example, if you want to filter all the video indexing tasks created on a specific date and whose status is pending, you must specify both the created_at and status fields, as shown in this example: created_at=2022-05-17&status=pending.
The examples in this section are specific to using the /tasks endpoint. However, the principles demonstrated are similar when using the /indexes and /indexes/videos endpoints.
Note
Depending on the endpoint, you can filter on various fields. For details, see the API Reference section.
You can interact with the platform using one of the available SDKs or an HTTP client like requests or axios. Follow the steps in one of the sections below, depending on your use case.
Prerequisites
- You're familiar with the concepts that are described on the Platform overview page.
- You've already created an index.
Use an SDK
Refer to this section if you use one of the available SDKs.
Filter on a specific date
The following example code uses the created_at parameter to return all the tasks created on a specific date:
from twelvelabs import TwelveLabs
client = TwelveLabs(api_key="<YOUR_API_KEY>")
indexes = client.task.list_pagination(
created_at="2022-05-17"
)
# Utility function to print a specific page
def print_page(page):
for task in page:
print(f"id={task.id} status={task.status} created_at={task.created_at}")
print_page(indexes.data)
while True:
try:
print_page(next(indexes))
except StopIteration:
break
import { TwelveLabs, Task } from 'twelvelabs-js';
const client = new TwelveLabs({ apiKey: '<YOUR_API_KEY>'});
const tasks = await client.task.listPagination({ createdAt: '2022-05-17' });
tasks.data.forEach((task: Task) => {
console.log(`id=${task.id} status=${task.status}`);
});
while (true) {
const nextPageData = await tasks.next();
if (!nextPageData) {
break;
}
nextPageData.forEach((task: Task) => {
console.log(`id=${task.id} status=${task.status} created_at=${task.createdAt}`);
});
}
The following example output was truncated for brevity:
id=6283ab5ef18e21febd160bc3 status=ready created_at=2022-05-17T14:04:14.001Z
id=6283a0a8f18e21febd160bbf status=ready created_at=2022-05-17T13:18:32.388Z
id=6283a00df18e21febd160bbc status=ready created_at=2022-05-17T13:15:57.053Z
Filter based on a file name
To filter based on a file name, use the filename parameter as shown in the example below:
from twelvelabs import TwelveLabs
client = TwelveLabs(api_key="<YOUR_API_KEY>")
tasks = client.task.list_pagination(
filename="<YOUR_FILENAME>" # Example: filename="01.mp4"
)
# Utility function to print a specific page
def print_page(page):
for task in page:
print(f"id={task.id} status={task.status} created_at={task.created_at} metadata={task.metadata}")
print_page(tasks.data)
while True:
try:
print_page(next(tasks))
except StopIteration:
break
import { TwelveLabs, Task } from 'twelvelabs-js';
const client = new TwelveLabs({ apiKey: '<YOUR_API_KEY>'});
const tasks = await client.task.listPagination({ filename: '<YOUR_FILE_NAME>' }); // Example: filename="01.mp4"
tasks.data.forEach((task: Task) => {
console.log(`id=${task.id} status=${task.status} metadata=${task.metadata}`);
});
while (true) {
const nextPageData = await tasks.next();
if (!nextPageData) {
break;
}
nextPageData.forEach((task: Task) => {
console.log(`id=${task.id} status=${task.status} created_at=${task.createdAt} metadata=${task.metadata}`);
});
}
The following output was truncated for brevity:
id=65e5ae52bb29f13bdd6f37c8 status=ready created_at=2024-03-04T11:19:46.901Z metadata={'filename': '01.mp4', 'duration': 579.12, 'width': 640, 'height': 360}
id=65d46ef879384c4778ab30fa status=ready created_at=2024-02-20T09:20:56.321Z metadata={'filename': '01.mp4', 'duration': 579.12, 'width': 640, 'height': 360}
Filter based on duration
To filter based on the duration of a video, use the duration parameter and either the exact match or the comparison operators.
Use the exact match operator
The example below uses the exact match operator to display all the videos of a specific duration (810 seconds):
from twelvelabs import TwelveLabs
client = TwelveLabs(api_key="<YOUR_API_KEY>")
indexes = client.task.list_pagination(
duration=810
)
# Utility function to print a specific page
def print_page(page):
for task in page:
print(f"id={task.id} status={task.status} created_at={task.created_at} metadata={task.metadata}")
print_page(indexes.data)
while True:
try:
print_page(next(indexes))
except StopIteration:
break
import { TwelveLabs, Task } from 'twelvelabs-js';
const client = new TwelveLabs({ apiKey: '<YOUR_API_KEY>'});
const tasks = await client.task.listPagination({ duration: 810 });
tasks.data.forEach((task: Task) => {
console.log(`id=${task.id} status=${task.status}`);
});
while (true) {
const nextPageData = await tasks.next();
if (!nextPageData) {
break;
}
nextPageData.forEach((task: Task) => {
console.log(`id=${task.id} status=${task.status} created_at=${task.createdAt} metadata=${task.metadata}`);
});
}
The following example output was truncated for brevity:
id=65e5ae52bb29f13bdd6f37c8 status=ready created_at=2024-03-04T11:19:46.901Z metadata={'filename': '02.mp4', 'duration': 810, 'width': 640, 'height': 360}
id=65d46ef879384c4778ab30fa status=ready created_at=2024-02-20T09:20:56.321Z metadata={'filename': '02.mp4', 'duration': 810, 'width': 640, 'height': 360}
Use an HTTP client
Refer to this section if you use clients such as requests or axios.
Filter on a specific date
The following example code uses the created_at query parameter and the exact match operator to return all the items created on a specific date:
TASKS_URL = f"{API_URL}/tasks?created_at=2022-05-17"
response = requests.get(TASKS_URL, headers=headers)
print (f”Status code: {response.status_code}”)
pprint (response.json())
const TASKS_URL = `${API_URL}/tasks?created_at=2022-05-17`
const config = {
method: 'get',
url: TASKS_URL,
headers: headers,
}
const resp = await axios(config)
const response = await resp.data
console.log(`Status code: ${resp.status}`)
console.log(response)
The following example output was truncated for brevity:
Status code: 200
{
"data": [
{
"_id": "6283ab5ef18e21febd160bc3",
"created_at": "2022-05-17T14:04:14.001Z",
"estimated_time": "2022-05-17T14:15:14.667Z",
"index_id": "6283aa51f18e21febd160bc0",
"metadata": {
"duration": 966.598967,
"filename": "animal-encounters-03.mp4",
"height": 480,
"width": 854
},
"status": "ready",
"updated_at": "2022-05-17T14:16:44.857Z"
},
{
"_id": "6283a0a8f18e21febd160bbf",
"created_at": "2022-05-17T13:18:32.388Z",
"estimated_time": "2022-05-17T13:31:26.808Z",
"index_id": "6283990ff18e21febd160bbb",
"metadata": {
"duration": 966.598967,
"filename": "animal-encounters-03.mp4",
"height": 480,
"width": 854
},
"status": "ready",
"updated_at": "2022-05-17T13:31:28.316Z"
},
{
"_id": "6283a00df18e21febd160bbc",
"created_at": "2022-05-17T13:15:57.053Z",
"estimated_time": "2022-05-17T13:28:11.817Z",
"index_id": "6283990ff18e21febd160bbb",
"metadata": {
"duration": 531.998133,
"filename": "animal-encounters-01.mp4",
"height": 480,
"width": 854
},
"status": "ready",
"updated_at": "2022-05-17T13:29:56.485Z"
}
],
"page_info": {
"limit_per_page": 10,
"page": 1,
"total_page": 1,
"total_results": 7
}
}
Filter on a date range
To display all the items created in a date range, you must combine the lte and gte operators. The following example code returns all the video indexing tasks created between 2022-05-15 and 2022-05-16:
TASKS_URL = f"{API_URL}/tasks?created_at[gte]=2022-05-15&created_at[lte]=2022-05-16"
response = requests.get(TASKS_URL, headers=headers)
print (f”Status code: {response.status_code}”)
pprint (response.json())
const TASKS_URL = `${API_URL}/tasks?created_at[gte]=2022-05-15&created_at[lte]=2022-05-16`
const config = {
method: 'get',
url: TASKS_URL,
headers: headers,
}
const resp = await axios(config)
const response = await resp.data
console.log(`Status code: ${resp.status}`)
console.log(response)
The output should look similar to the following one:
Status code: 200
{
"data": [
{
"_id": "62823d4df18e21febd160bb5",
"created_at": "2022-05-16T12:02:20.977Z",
"estimated_time": "2022-05-16T12:14:32.26Z",
"index_id": "627de802f18e21febd160ba9",
"metadata": {
"duration": 810.84,
"filename": "best-racing-moments.mp4",
"height": 480,
"width": 854
},
"status": "ready",
"updated_at": "2022-05-16T12:14:34.824Z"
}
],
"page_info": {
"limit_per_page": 10,
"page": 1,
"total_page": 1,
"total_results": 1
}
}
Filter based on a file name
To filter based on a file name, use the filename query parameter and the exact match operator as shown in the example below:
TASKS_URL = f"{API_URL}/tasks?filename=best-racing-moments.mp4"
response = requests.get(TASKS_URL, headers=headers)
print (f”Status code: {response.status_code}”)
pprint (response.json())
const TASKS_URL = `${API_URL}/tasks?filename=best-racing-moments.mp4`
const config = {
method: 'get',
url: TASKS_URL,
headers: headers,
}
const resp = await axios(config)
const response = await resp.data
console.log(`Status code: ${resp.status}`)
console.log(response)
Filter based on duration
To filter based on the duration of a video, use the duration query parameter and either the exact match or the comparison operators.
Use the exact match operator
The example below uses the exact match operator to display all the videos of a specific duration (810 seconds):
TASKS_URL = f"{API_URL}/tasks?duration=810"
response = requests.get(TASKS_URL, headers=headers)
print (f”Status code: {response.status_code}”)
pprint (response.json())
const TASKS_URL = `${API_URL}/tasks?duration=810`
const config = {
method: 'get',
url: TASKS_URL,
headers: headers,
}
const resp = await axios(config)
const response = await resp.data
console.log(`Status code: ${resp.status}`)
console.log(response)
The output should look similar to the following one:
Status code: 200
{
"data": [
{
"_id": "62823d4df18e21febd160bb5",
"created_at": "2022-05-16T12:02:20.977Z",
"estimated_time": "2022-05-16T12:14:32.26Z",
"index_id": "627de802f18e21febd160ba9",
"metadata": {
"duration": 810.84,
"filename": "best-racing-moments.mp4",
"height": 480,
"width": 854
},
"status": "ready",
"updated_at": "2022-05-16T12:14:34.824Z"
}
],
"page_info": {
"limit_per_page": 10,
"page": 1,
"total_page": 1,
"total_results": 1
}
}
Note
The API rounds down the duration of the videos when using the exact match operator to filter based on duration.
Use the comparison operators
The example below uses the comparison operators to filter videos longer than 600 seconds and shorter than 820 seconds:
TASKS_URL = f"{API_URL}/tasks?duration[gte]=600&duration[lte]=820"
response = requests.get(TASKS_URL, headers=headers)
print (f”Status code: {response.status_code}”)
pprint (response.json())
const TASKS_URL = `${API_URL}/tasks?duration[gte]=600&duration[lte]=820`
const config = {
method: 'get',
url: TASKS_URL,
headers: headers,
}
const resp = await axios(config)
const response = await resp.data
console.log(`Status code: ${resp.status}`)
console.log(response)
The output should look similar to the following one:
Status code: 200
{
"data": [
{
"_id": "6283a07ef18e21febd160bbe",
"created_at": "2022-05-17T13:17:50.564Z",
"estimated_time": "2022-05-17T13:28:55.578Z",
"index_id": "6283990ff18e21febd160bbb",
"metadata": {
"duration": 700.24,
"filename": "animal-encounters-02.mp4",
"height": 480,
"width": 854
},
"status": "ready",
"updated_at": "2022-05-17T13:30:54.078Z"
},
{
"_id": "6283a059f18e21febd160bbd",
"created_at": "2022-05-17T13:17:13.524Z",
"estimated_time": "2022-05-17T13:28:55.569Z",
"index_id": "6283990ff18e21febd160bbb",
"metadata": {
"duration": 700.24,
"filename": "animal-encounters-01.mp4",
"height": 480,
"width": 854
},
"status": "ready",
"updated_at": "2022-05-17T13:30:41.95Z"
},
{
"_id": "62823d4df18e21febd160bb5",
"created_at": "2022-05-16T12:02:20.977Z",
"estimated_time": "2022-05-16T12:14:32.26Z",
"index_id": "627de802f18e21febd160ba9",
"metadata": {
"duration": 810.84,
"filename": "best-racing-moments.mp4",
"height": 480,
"width": 854
},
"status": "ready",
"updated_at": "2022-05-16T12:14:34.824Z"
}
],
"page_info": {
"limit_per_page": 10,
"page": 1,
"total_page": 1,
"total_results": 3
}
}
Filter on custom metadata
The example code below filters your list of videos on a custom field named views of type integer. For details about specifying custom metadata, see the Provide custom metadata page. The platform will return only the videos for which the value of the views field equals 120000.
INDEXES_VIDEOS_URL = f"{API_URL}/indexes/{INDEX_ID}/videos"
queryString = {
"views" : 120000
}
response = requests.get(INDEXES_VIDEOS_URL, headers=headers, params=queryString)
print (f'Status code: {response.status_code}')
pprint(response.json())
const INDEXES_VIDEOS_URL = `${API_URL}/indexes/${INDEX_ID}/videos`
const resp = await axios.get(
INDEXES_VIDEOS_URL,
{
'params': {
'views': 120000
},
'headers': {
'x-api-key': API_KEY
}
}
)
const { data: response } = resp;
console.log(`Status code: ${resp.status}`)
console.log(JSON.stringify(response,null,4))
Updated 28 days ago