hi everyone this is Sarah McKenna from
sequentum thank you for joining our
webinar today on how to embed data
collection
capabilities into your operation via API
um we have our Engineers XI Jiang here
today to do a walk through of our API
and to talk through
the process flow and how it would work
to embed our system in yours with that
I'm going to pass it over to him
hi guys welcome to a webinar on API
access for Content Grabber Enterprise
Edition I think the most important thing
that we should go over first is setting
up a
uthentication token in order to start
using our API
so then under your organization first
head over to organization and click on
users
for one of our users over here jevin we
can set up API access under his account
I've already set one up beforehand that
expires uh
tomorrow but up from here if you want to
add a new access token you can just
press add new access token set up an
expiration date and maybe a description
if you want to but we have already set
up a
pair token that we would need to use for
every single one of our requests
so going back to our API uh the second
most important thing to do from here is
to get Asian details in order to get a
basically an agent ID for all of your
agents so in our repository over here if
we head back to repository you can see
that we have a couple agents here set up
uh Amazon I think I'm going to be using
uh either the Amazon us demo or the
Victoria's Secret one so that in order
to get the agent ID and start using our
API I'm just going to head over to my
editor over here and uh
basically for the API call I think it
was yep API agent agent path so all we
need to do is specify the agent path in
our request in order to get a
the agent ID back in our example
response over here so then we can also
do this uh like via python or a postman
uh I'm first just going to do it quickly
using like a python request and then for
the rest of them I'll probably just be
using Postman because it's actually
simpler
okay so then from here we can just uh
import our requests
um set up a URL which is
one that we're currently using is
ECC demo that's so quantum.com API
agent and then we specify our agent path
the agent path is just the root
directory which is retail over here
slash whichever agent that you want uh
to get the ID for
so then from here we'll specify retail
slash probably Victoria's Secrets
is that paste that in
uh set up a body that could be blank for
now set up our headers which is uh
our authorization token
authorization
come back over here I actually did not
save the token I mean copy it
oops
and copy this token
and then we can
set up the response
Quest Dot
quests
get
set up the URL the headers
then uh the body
and then we can just print the response
from here
oops that's spelled oops will responds
wrong up here
response yep
run this
and then you can see that uh the agent
ID the name the directory ID and some of
the description for the agent actually
gets returned all of these can actually
be found in
the web portal over here like scrape
agent ID but we can also do this in
Postman as well if we want to get ID for
a specific agent like uh
retail slash Victorious
Victoria Secrets
this also works so then basically the
agent ID is our first endpoint into
every other request that we have here so
in after we have
the agent ID we can actually start up
one of the agents so then from over here
uh if I go back to the API in order to
start up an agent all we need is the
agent ID in order to start it up
um in the body a lot of these options
are actually optional so then you won't
you won't actually need them
four like some simple simple requests
like log logging level production runs
input parameters we have any
but for uh for now which is gonna enter
in agent ID
uh have the content type equals
application.json authorization with the
bearer token and then I guess one level
logging and then we can press send
you can see that uh the status here
is a 2-2 which means it has started so
from here we can check
our agent Victoria's Secrets you can see
that has actually started running
history it takes a couple seconds to run
so it's looks like it already completed
from here so from here you can actually
like download the data view the I'm not
exactly sure what this agent does again
but here I can just open up the data to
see uh what it's actually doing
oops close this
looks like we're just capturing some ID
in one of the particular fields
okay so from here uh there's a lot of
use cases for our API and one of the
biggest ones and most frequent one is
running an agent via ad hoc where
an end user will pass in a couple of
input parameters for the agent to run
and four different runs there might be
different input parameters associated
with that run so then here I've set up
an agent that actually uses an input
parameter
and the input files here it's called
Amazon us demo
from here uh if you go click on input
data you can see that there are actually
no input data currently for the agent so
then we will actually need to upload an
uh upload a input file for the agent run
so first uh
under my input.text I will set up a
specific URL to scrape from Amazon
I had one page open yeah this one I
guess we'll scrape some chocolate so I'm
just going to copy this URL
come back to my text file
paste it in here so for this agent I
guess right now it's just only gonna
script from one particular
page it's chocolate page press save now
on the postman if I want to
actually come back to API first
uploading so this is all online on
Postman and under our documentations so
then you guys can all follow along if
you want to it's under uh
this Postman link is also you can also
come here directly from uh send desk
help manual as well
but right now what we want to do is uh
want to upload agent input files in
order for uh the agent to run with these
input files
so if I come back to a postman over here
if I want to upload it
I want to upload agent file all I need
is uh the agent ID
again so from here
um
this is the right agent ID
let me get the agent ID for the Amazon
one instead of the Victoria's Secrets
One
Amazon
Amazon us demo
copy this
copy the agent ID
then from here set up the authorization
and then in the body here
we can select our particular files that
we want to use I think my files
currently saved under on my desktop the
one that I just saved open that up
I can press send
so status here is a 202. so if I come
back over here to my Asian directory if
I go to Amazon device demo and then I
navigate to input data you can see that
an input file has actually been uploaded
to the agent from here I can edit it or
download it for right now I'm just going
to check it out
and it seems like yeah it has uploaded
the chocolate URL
so I can actually start up this agent to
run with that input
so from here if I start the agent again
this one here I need to change the agent
ID
the agent actually ran with that
particular input file so if I come back
to us agent
see that's currently running and if I go
to agent run history
instead just completed successfully
right now and from here I can download
the data
uh no one wants to buy that
oops
you can see that uh the agent scrape
this particular link got some additional
Fields such as like price and so forth
accounts but now there are some
instances where you will want to run the
agent via API like a different set of
inputs so in this case all you need to
do is just uh edit your input for like a
different page so say from here if we
want to
get some maybe laptops or something
like uh let's get the first one
copy this link and then save it directly
into our input
and then in Postman upload the new one
this is still the same file all we did
was uh change the content in the file
upload the new file
and then start up the agent again
you see that agent is running currently
going to ship it relatively quick
complete it now if I download this data
you can see that now it's scraped from
the new URL this laptop Bluetooth
personal
and yep that's how we would go about
uploading different input files for you
guys to run via ad hoc
and then there's some cases where uh you
will actually also want to download the
data itself directly as well instead of
downloading directly from the portal you
can download via API
so from this we have
a download data specific
get data specific agent run yep from
here all we need is the agent run ID
this agent run ID is generated whenever
we run an agent
which is under when I started the agent
this run here had this ID so I'm going
to copy this ID
copy that under download data
paste that in there the authorization
token is good
that is how we will go about uh
using our internal API to set some jobs
runs
so we went over a starting an agent I
mean getting a setting up the
authorization token getting agent
details starting an agent uploading
input files for the agent to run
updating those input files and then also
downloading
the data afterwards
we still have a lot of other
API calls in here as well
I don't think any of them are that
important to go over on this webinar
but I mean if anyone sees anyone that
they think is important that they want
me to go over I could go over them but
are there any particular questions right
now regarding
this API
I don't see any questions I do see a
number of people are still on the call
um if any of you have any particular
questions about how to embed
this data collection engine into your
system please do raise them now
otherwise you can feel free to email us
at sales sequentum.com
with any questions following this
webinar in your uh you can always access
our manual as well
um thank you for joining our webinar
today and we look forward to
um to hearing from you
bye