We Keep Coding

Extracting only portion of JSON document with REST API search call in MarkLogic

I am looking out methods of Extracting only portion of JSON document with REST API search call in MarkLogic using JavaScript or XQuery.
I have tried using query options of re extract-document-data but was not successful. Tried checking my extract path using CTS.validextract path but that function was not recognised in Marklogic 9.0-1
Do I have to using specific search options like constraints or structured query.
Could you please help out? TIA.
I have below such sample document
{
"GenreType": {
"Name": "GenreType",
"LongName": "Genre Complex",
"AttributeDataType": "String",
"GenreType Instance Record": [
{
"Name": "GenreType Instance Record",
"Action": "NoChange",
"TitleGenre": [
"Test1"
],
"GenreL": [
"Test1"
],
"GenreSource": [
"ABC"
],
"GenreT": [
"Test1"
]
},
{
"Name": "GenreType Instance Record",
"Action": "NoChange",
"TitleGenre": [
"Test2"
],
"GenreL": [
"Test2"
],
"GenreSource": [
"PQR"
],
"GenreT": [
"Test2"
]
}
]
}
}
in which i need to search a document with attribute "TitleGenre" WHERE GenreSource = “ABC” in the GenreType complex attribute. It's an array in json document.
I was using the search option as below, (writing search option in XML, but searching the in json documents)
<extract-path>/GenreType/"GenreType Instance Record"[#GenreSource="ABC"]</extract-path>
I am still facing the issues. If possible could you please let me know how json documents can be searched for such specific requirement? #Wagner Michael

You can extract document data by using the extract-document-data option.
xquery version "1.0-ml";
let $doc := object-node {
"GenreType": object-node {
"Name": "GenreType",
"LongName": "Genre Complex",
"AttributeDataType": "String",
"GenreType-Instance-Record": array-node {
object-node {
"TitleGenre": array-node {
"Test1"
},
"GenreSource": array-node {
"ABC"
}
},
object-node {
"TitleGenre": array-node {
"Test2"
},
"GenreSource": array-node {
"PQR"
}
}}
}
}
return xdmp:document-insert("test.xml", $doc);
import module namespace search = "http://marklogic.com/appservices/search"
at "/MarkLogic/appservices/search/search.xqy";
search:search(
"Genre Complex",
<options xmlns="http://marklogic.com/appservices/search">
<extract-document-data>
<extract-path>/GenreType/GenreType-Instance-Record[GenreSource = "ABC"]</extract-path>
</extract-document-data>
</options>
)
In this case /GenreType/GenreType-Instance-Record is the xpath to the extracted element.
Relating to your comment, i also added a predicate [GenreSource = "ABC"]. This way only GenreType-Instance-Record which have a GenreSource of "ABC" are being extracted!
Result:
....
<search:extracted kind="array">[{"GenreType-Instance-Record":{"TitleGenre":["Test1"], "GenreSource":["ABC"]}}]
</search:extracted>
....
Note:
You can add multiple <search:extract-path> elements!
I had to change the name of GenreType Instance Record to GenreType-Instance-Record. I am not sure if you can have property names with whitespaces and access them with xpath. I couldn't get it working this way.
Please post your search options, if this does not work for you.
Edit: Added a predicate to the extract-path.

Thank you so much Wagner, for your prompt trials. Helped me look out for accurate solution to my problem as of now. I have used below extract path, as i could not modify the names in documents. /GenreType/array-node("GenreType Instance Record")/object-node()/TitleGenre[following-sibling::GenreSource="ABC"]

Facebook Graph API - Ads Insights with different Action Values from Post

I'm using facebook Graph API to retrieve action values from campaigns. But the JSON returned is always different from any values from facebook screen.
My Request from JavaScript
var d = new FormData();
d.append("access_token", "MY_ADS_INSIGHTS_TOKEN");
d.append("fields", "actions");
d.append("date_preset", "lifetime"); // I want lifetime data
return await (await fetch("https://graph.facebook.com/v3.1/" + campaignid + "/insights", {
method: "post",
body: d
})).json();
and after I access the report insights using the URL:
https://graph.facebook.com/v3.1/REPORT_RUN_ID/insights?access_token=MY_ADS_INSIGHTS_TOKEN
JSON returned after access report task
{
"data": [
{
"actions": [
{
"action_type": "comment",
"value": "2"
},
{
"action_type": "like",
"value": "4"
},
{
"action_type": "photo_view",
"value": "30"
},
{
"action_type": "post",
"value": "1"
},
{
"action_type": "link_click",
"value": "7"
},
{
"action_type": "page_engagement",
"value": "249"
},
{
"action_type": "post_engagement",
"value": "245"
},
{
"action_type": "post_reaction",
"value": "205"
}
],
"date_start": "2018-07-09",
"date_stop": "2018-07-15",
"ad_id": null // removed
}
],
"paging": {
"cursors": {
"before": "MAZDZD",
"after": "MAZDZD"
}
},
"__fb_trace_id__": null // removed
}
Facebook Post Results
I want to know:
Why Facebook Graph API return the post_reaction as 205 since from facebook view it is 160 or 150? the value doesn't match anything, happens to action like too
Notes:
I'm not using any SDK, but this isn't the problem.
The Ad has only ONE ads group and the group has only ONE campaign
I make the requisition at the same time as I see the post. There are no major interactions in this post, it is old enough that the values do not change.
I known that Facebook cache anything, but this Ad is from 10, July.
Ad Campaign Insights reference: https://developers.facebook.com/docs/marketing-api/reference/ad-campaign-group/insights/
I accept answers using SDKs or different programming languages like C#, php or Java as example, I want only know HOW make the right request or what is exactly happening.

How to get internal Database/Table content in TYPO3 and display via JavaScript, JQuery or AJAX

I am a beginner in TYPO3 and have to say its very complex but I am struggling my way through.
I just want a simple way to get content from an internal TYPO3 Database Table which I created and forward it to JavaScript/JQuery/AJAX.
My BootstrapTable is working but just displays static content from var data.
I know there are ways from pageType, eID and $GLOBALS['TYPO3_DB']->exec_SELECTgetRows( /* ... */ ); to get the content of the Database Table, but how to forward it to JavaScript and display it?
There should not be another database call with credentials, just a way to the internal Database.
Searching for this answer since 2 days but can't find any easy and plausible answer.
Or do I have to take the long road via eID or pageType?
Edit due to comment:
Here is my JavaScript with bootstrapTable:
$(function () {
$('#table').bootstrapTable({
idField: 'name',
pagination: true,
search: true,
data: data,
columns: [{
field: 'name',
title: 'Name'
}, {
field: 'stargazers_count',
title: 'Stars'
}, {
field: 'forks_count',
title: 'Forks'
}, {
field: 'description',
title: 'Description'
}],
onPostBody: function () {
$('#table').editableTableWidget({editor: $('<textarea>')});
}
});
});
var data = [{
"name": "bootstrap-table",
"stargazers_count": "526",
"forks_count": "122",
"description": "An extended Bootstrap table with radio, checkbox, sort, pagination, and other added features. (supports twitter bootstrap v2 and v3) "
}, {
"name": "multiple-select",
"stargazers_count": "288",
"forks_count": "150",
"description": "A jQuery plugin to select multiple elements with checkboxes :)"
}, {
"name": "bootstrap-show-password",
"stargazers_count": "32",
"forks_count": "11",
"description": "Show/hide password plugin for twitter bootstrap."
}, {
"name": "blog",
"stargazers_count": "13",
"forks_count": "4",
"description": "my blog"
}, {
"name": "scutech-redmine",
"stargazers_count": "6",
"forks_count": "3",
"description": "Redmine notification tools for chrome extension."
}
];
And calling it in my html with : <table id="table"></table>
My own created table in mysql database is: tx_table
I want the content of that table to be displayed in my bootstraptable Table

First solution. You can output your results directly into html using <script> tag and v:format.json.encode ViewHelper
action code:
$yourDataArrayFromPHP = array(
'var1' => 'val1',
'var2' => 'val2'
);
$this->view->assign('yourDataArrayFromPHP', $yourDataArrayFromPHP);
html:
<script>
var data = {v:format.json.encode(value: '{yourDataArrayFromPHP}')}
</script>
Now you can access this data in javascript data.var1
Also keep in mind that your data should be before bootstrapTable initialization.

Backbone? Can.js? Ghetto DIY? How should I work with this data?

I'm working on an application that lets our security dispatchers update a page that contains current road and campus conditions. The backend is a nodejs/express stack with and the data is a simple JSON structure that looks something like this:
{
"campus": {"condition": "open", "status": "normal"},
"roads": {"condition": "wet", "status": "alert"},
"adjacentroads": {"condition": "not applicable", "status": "warning"},
"transit": {"condition": "on schedule", "status": "normal"},
"classes": {"condition": "on schedule", "status": "normal"},
"exams": {"condition": "on schedule", "status": "normal"},
"announcements" : "The campus is currently under attack by a herd of wild velociraptors. It is recommended that you do not come to campus at this time. Busses are delayed.",
"sidebar": [
"<p>Constant traffic updates can be heard on radio station AM1234. Traffic updates also run every 10 minutes on AM5678 and AM901.</p>",
"<p>This report is also available at <strong>555-555-1234</strong> and will be updated whenever conditions change.</p>"
],
"links": [
{
"category": "Transportation Links",
"links": [
{
"url": "http://www.localtransit.whatever",
"text" : "Local Transit Agency"
},
{
"url": "http://m.localtransit.whatever",
"text" : "Local Transit Agency Mobile Site"
}
]
},
{
"category": "Weather Forecasts",
"links": [
{
"url": "http://weatheroffice.ec.gc.ca/canada_e.",
"text" : "Environment Canada"
},
{
"url": "http://www.theweathernetwork.com",
"text" : "The Weather Network"
}
]
},
{
"category": "Campus Notices & Conditions",
"links": [
{
"url": "http://www.foo.bar/security",
"text" : "Security Alerts & Traffic Notices"
},
{
"url": "http://foo.bar/athletics/whatever",
"text" : "Recreation & Athletics Conditions"
}
]
},
{
"category": "Wildlife Links",
"links": [
{
"url": "http://velociraptors.info",
"text" : "Velociraptor Encounters"
}
]
}
],
"lastupdated": 1333151930179
}
I'm wondering what the best way of working with this data on the client side would be (e.g. on the page that the dispatchers use to update the data). The page is a mix of selects (the campus, roads, etc conditions), TinyMCE textareas (announcements and sidebar) and text inputs (links). I'm open to changing this data structure if necessary but it seems to me to work well. I've been looking at Backbone, and also Can.JS but I'm not sure if either of those are suitable for this.
Some additional information:
there's no need to update an individual item in the data structure separatly; I plan on POSTing the entire structure when it's saved. That said...
there's actually two different views, one for the dispatchers and another for their supervisors. The dispatchers only have the ability to change the campus, roads, etc conditions through drop-downs and furthermore can only change the "condition" key; each possible condition has a default status assigned to it. Supervisors can override the default status, and have access to the announcements, sidebar and links keys. Maybe I do need to rethink the previous point about POSTing the whole thing at once?
the supervisors need to be able to add and remove links, as well as add and remove entire link categories. This means that DOM elements need to be added and removed, which is why I'm thinking of using something like Backbone or Can.js instead of just writing some ghetto solution that looks at all the form elements and builds the appropriate JSON to POST to the server.
Suggestions welcomed!

CanJS works great with nested data. can.Model is inheriting can.Observe which allows you to listen to any changes in the object structure.
If you include can.Observe.Delegate you have even more powerful event mechanism (example from the docs):
// create an observable
var observe = new can.Observe({
name : {
first : "Justin Meyer"
}
})
var handler;
//listen to changes on a property
observe.delegate("name.first","set",
handler = function(ev, newVal, oldVal, prop){
this //-> "Justin"
ev.currentTarget //-> observe
newVal //-> "Justin Meyer"
oldVal //-> "Justin"
prop //-> "name.first"
});
// change the property
observe.attr('name.first',"Justin")

jQuery loading data into jqGrid

I'm a first time user of jqGrid, so far I went trough official examples, I'm interested in loading data into grid either using json.
I'm currently looking at, Loading data(JSON Data):
http://trirand.com/blog/jqgrid/jqgrid.html
Here is a bit of javascript that creates grid :
jQuery("#list2").jqGrid(
{
url : '<c:url value="${webappRoot}/getdetails" />',
datatype : "json",
colNames : [ 'id', 'Location', 'Country Code', 'Type', 'Interval',
'Version', 'Last Active', 'Last Login', 'NOTE' ],
colModel : [
{ name : 'id', width : 10 },
{ name : 'location', width : 75 },
{ name : 'countryCode', width : 50 },
{ name : 'type', width : 40 },
{ name : 'interval', width : 30 },
{ name : 'version', width : 45 },
{ name : 'lastactive', width : 50, align : "right" },
{ name : 'lastlogin', width : 50, sortable : false },
{ name : 'note', width : 50, sortable : false}
],
rowNum : 10,
rowList : [ 10, 20, 30 ],
pager : '#pager2',
width: gridWidth,
sortname : 'id',
viewrecords : true,
sortorder : "desc",
caption : "JSON Example"
});
jQuery("#list2").jqGrid('navGrid', '#pager2',
{ edit : false, add : false, del : false});
${webappRoot}/getdetails transforms path to my project like http://localhost/myProject/getdetails, I'm using spring MVC(it might be irrelevant).
When I look in firebug this generates this http request :
GET http://localhost/newProject/getdetails?_search=false&nd=1304638787511&rows=10&page=1&sidx=id&sord=desc
200 OK
135ms
Here is the response :
{
"id": 1,
"location": "office_2782",
"countryCode": "UK",
"quarter": "500",
"version": "v3.05",
"lastactive": "yesterday",
"lastlogin": "today",
"note": "no note",
"type": "read-only"
}
When I navigate to JSON tab it all seems same as this, any idea what I'm doing wrong?
I'm trying to load only one record for start, and I can't get it working, any help is appriciated.

First of all you are not the first person who has problems understanding how the JSON data should be constructed, what the parameters sent from jqGrid to the server mean and so on. The official jqGrid documentation doesn't contain enough introduction, so the first steps of the jqGrid usage can be a little more difficult than one expect.
The problem which exists in your JSON response from the server is that it contains only one item of data instead of an array (or list) of items representing the grid rows. The data should be at least
[
{
"id": 1,
"location": "office_2782",
"countryCode": "UK",
"quarter": "500",
"version": "v3.05",
"lastactive": "yesterday",
"lastlogin": "today",
"note": "no note",
"type": "read-only"
}
]
or better as
{
"total": 1,
"page": 1,
"records": 1,
"rows": [
{
"id": 1,
"location": "office_2782",
"countryCode": "UK",
"quarter": 500,
"version": "v3.05",
"lastactive": "yesterday",
"lastlogin": "today",
"note": "no note",
"type": "read-only"
}
]
}
or even as
{
"total": 1,
"page": 1,
"records": 1,
"rows": [
{
"id": 1,
"row": [ "1", "office_2782", "UK", "500", "v3.05",
"yesterday", "today", "no note", "read-only" ]
}
]
}
or
{
"total": 1,
"page": 1,
"records": 1,
"rows": [
[ "1", "office_2782", "UK", "500", "v3.05", "yesterday", "today",
"no note", "read-only" ]
]
}
The reason of such strange at the first glance JSON data is that jqGrid is designed to support paging, sorting and filtering/searching of data implemented on the server. So the parameters rows=10&page=1&sidx=id&sord=desc from the url sent to the server mean that jqGrid asks the server to get the first page (page=1) of the data with the page having 10 rows per page (rows=10). The data should be previously sorted by id (sidx=id) in the descending order (sord=desc). If you has small number of rows (under some hundert for example) you can use client based sorting, paging and filtering if you add loadonce:true parameter of the jqGrid, but the server based implementation allows you to work with really large dataset having many hundred thousands rows of data with very good performace.
I recommend you to read my this answer where I tried to explain how the additional elements of the server response "total", "page" and "records" will be used. The values of the parameters can be encoded in JSON either as numbers or as strings (on your taste).
If the user clicks on the column header of the 'location' column for example jqGrid will send new request to the server having sidx=location&sord=asc in the url. So it is important to understand, that the server can be asked to provide the data for the grid not once per grid, but many times and the request will contain some parameters chosen by the user who works with the jqGrid.
Defining of jsonReader (and sometimes additional jsonmap parameters for every column) you describe the structure of the server response. Using the information jqGrid read the response and fill the grid.
The demo shows that with the corresponding jsonReader you can read even your original JSON data.
The last advice for you from me would be to consider at the beginning to use loadError event handle which helps to inform the user about the errors reported by the server. In the answer I have shown how it can be implemented in the case of ASP.NET MVC. I don't use spring MVC myself so I can't give you direct examples of how to better implement the error reporting in spring MVC, but the main idea is the same in any server technology: in case of errors the server should respond with the response having an error HTTP status code. Inside of your implementation of the loadError event handle you decode the response and display the information about the error.