FQL Your statement is not indexable, but fine in console - facebook-graph-api

I'm using this query for FQL:
SELECT pid, object_id, caption, like_info, comment_info, src, src_small, src_big, images FROM photo WHERE album_object_id="10151088306597851" ORDER BY like_info DESC
which works fine in the console, check:
https://developers.facebook.com/tools/explorer?method=GET&path=fql%3Fq%3DSELECT%20pid%2C%20object_id%2C%20caption%2C%20like_info%2C%20comment_info%2C%20src%2C%20src_small%2C%20src_big%2C%20images%20FROM%20photo%20WHERE%20album_object_id%3D%2210151088306597851%22%20ORDER%20BY%20like_info%20DESC
But when I use the Javascript SDK with FB.api() method it fails with error 604 "Your statement is not indexable. The WHERE clause must contain an indexable column. Such columns are marked with * in the tables linked from http://developers.facebook.com/docs/reference/fql"
But according to the docs of table "photo" "album_object_id" IS indexable..
This is my Javascript code:
if (album_object_id) {
var query = 'SELECT pid, object_id, caption, like_info, comment_info, src, src_small, src_big, images FROM photo WHERE album_object_id="' + album_object_id + '" ORDER BY like_info DESC';
//console.log(query);return;
FB.api('/fql?q=' + encodeURI(query), callback);
}
When I log query and copy/paste it into the graph API console, it works fine..
Any clues??
cheers
Sjoerd

I think there is a query encoding issue introduced by encodeURI on this syntax. The following code works perfectly and returns 82 items:
var query = 'SELECT pid, object_id, caption, like_info, comment_info, src, src_small, src_big, images FROM photo WHERE album_object_id="10151088306597851" ORDER BY like_info DESC';
FB.api('/fql', {q: query}, function(r) {
console.log(r)
});
}

Related

Power BI Embedded - Get current filters with Javascript

I have a working app based on the App Owns Data documentation & Javascript sample:
https://learn.microsoft.com/en-us/power-bi/developer/embed-sample-for-customers
https://microsoft.github.io/PowerBI-JavaScript/demo/v2-demo/index.html#
I am able to get the data out of each visualisation but I also want to know what filters are affecting it.
I can get the slicer states, but cannot get any filters applied by clicking on the charts. If you click on one datapoint, that filters the entire page. In the dashboard I can then click on the filter icon next to each chart and see 'Filters and slicers affecting this visual'.
I would really like to get this information via the JS API, but can't find it in the page.getFilters(), report.getFilters() or visual.getFilters().
Is it possible to do, and how?
For getting the filters applied on the visual, you can use the following:
async function get_visual_filters() {
// Get all the pages of the report
const pages = await report.getPages();
// Find required page
const page = pages.find((p) => p.name === "ReportSection1"); // The page in which visual is present
// Get the required visual.
const visuals = await page.getVisuals();
// Find the visual using Guid
const visual = visuals.find((v) => v.name === "VisualContainer1");
// Retrieve the filter of visual:VisualContainer1
const filters = await visual.getFilters();
console.log("The filters applied on the visual are: ", filters);
}
To get the page level filters, you can use:
const pageFilters = await page.getFilters();
To get the report level filters, you can use:
const reportFilters = await report.getFilters();
Output:
[1]: https://i.stack.imgur.com/O5q6d.png
Please refer following links:
https://learn.microsoft.com/en-us/javascript/api/overview/powerbi/get-visuals
https://learn.microsoft.com/en-us/javascript/api/overview/powerbi/control-report-filters
https://learn.microsoft.com/en-us/javascript/api/overview/powerbi/control-report-filters#get-filters

Complex QueryBuilder expresion in createQuery() with Sonata Admin

I'm using Symfony 4.2 and Sonata Admin bundle and I'm trying to display a specific list in one of my admin list view.
I want to display the lastest user by group. I created the SQL query and I'm now trying to translate it to QueryBuilder.
SELECT * FROM users
WHERE (created_at, group_id) IN (
SELECT MAX(created_at), group_id
FROM users
GROUP BY group_id)
My createQuery function in my Sonata Admin class :
public function createQuery($context = 'list')
{
$container = $this->getConfigurationPool()->getContainer();
$em = $container->get('doctrine.orm.entity_manager');
$expr = $em->getExpressionBuilder();
$subQuery = $em->createQueryBuilder()
->select('MAX(c.createdAt), c.group')
->from('ApiBundle:Configuration', 'c')
->groupBy('c.group')
->getDQL();
$query = parent::createQuery($context);
$alias = $query->getRootAlias();
$query->where($expr->in('('.$alias.'.createdAt, '.$alias.'.group)', $subQuery));
}
Unfortunately it is not working and I'm not sure it's the right way to do it.
I could make it work with only selecting the createdAt in the in() expression but it's not what I want to have at the end.
Here is the error I got.
[Syntax Error] line 0, col 77: Error: Expected Doctrine\ORM\Query\Lexer::T_CLOSE_PARENTHESIS, got ','
How can I make it work using QueryBuilder ?
Thank you.

odoo: printing qweb report using wizard

i have two reports cout_materiel_facture_detail_report.xml (detailed report) and cout_materiel_facture_report.xml (simple report) and have a wizard where the user inputs some data and chooses whether he/she wants to print the simple report or the detailed one. Here is the wizard's class:
class CoutMaterielFactureWizard(models.TransientModel):
_name = 'gc.cout_materiel_facture_wizard'
directeur_parc_id = fields.Many2one('hr.employee', string='Directeur Parc')
procedure = fields.Char(string='Procedure')
version = fields.Char(string='Verion')
date_realisation = fields.Date(string='Date realisation')
# is_landscape = fields.Boolean(string='Mode paysage?')
is_detail = fields.Boolean(string='Version simplifiee?')
#api.multi
def do_toggle_print(self):
cout_materiel = self.env['gc.cout_materiel'].browse(self._context.get('active_id', False))
cout_materiel.write({
'directeur_parc_id': self.directeur_parc_id.id
})
# Print the simple report
if not self.is_detail:
return {
'type': 'ir.actions.report.xml',
'name': 'gestion_cout.cout_materiel_facture_report_template',
'report_name': 'gestion_cout.cout_materiel_facture_report_template',
}
# Print the detailed report
else:
sql = "SELECT SUM(h_sup)+SUM(h_exp),SUM(h_im),count(*),SUM(total), famille FROM gc_cout_materiel_line where " \
"cout_materiel_id =%s group by famille "
self.env.cr.execute(sql, (cout_materiel.id,))
results = self.env.cr.fetchall()
if len(results) > 0:
line_ids = []
for nbht, nbhim, qte, prix_total, famille in results:
line_ids.append((0, 0, {
'famille': famille,
'type': 'VA',
'qte': qte,
'nbr_heures': nbht,
'nbr_heures_im': nbhim,
'nbr_jours': 28,
'prix_unitaire': 'VA',
'prix_total': prix_total,
}))
self.env['gc.cout_materiel_facture_temp'].create({
'chantier_name': cout_materiel.chantier_id.name,
'mois_name': cout_materiel.mois_id.name,
'num_annexe': cout_materiel.num_annexe,
'expediteur': cout_materiel.expediteur,
'destinateur': cout_materiel.destinateur,
'application_date': cout_materiel.application_date,
'date_realisation': self.date_realisation,
'directeur_parc_name': self.directeur_parc_id.name,
'procedure': self.procedure,
'version': self.version,
'prix_total_global': cout_materiel.total_global,
'line_ids': line_ids,
})
return {
'type': 'ir.actions.report.xml',
'name': 'gestion_cout.gc_cout_materiel_facture_detail_report_template',
'report_name': 'gestion_cout.gc_cout_materiel_facture_detail_report_template',
}
But i get this error after i hit the print button
I checked out the database and found both reports are present there.
Any help? please!!
Finally i managed to solve my problem!!
Here is what i did:
I created a method in the wizard model which returns a list of objects that i whant to print and linked the wizard to the qweb report.
Then i called the method from the qweb report using object.my_mehtod() in a t-foreach loop, where object represents the wizard.
With this way i am able to create complex reports and print them easily. One can use this method to get data from several tables and organize the data and retur them as a list.
I hope that it will help someone.
Best regards!!

APEX 3.2 (Oracle): on change of an item dynamically populates the values of other items on the page

I am on APEX 3.2 and in a form used to insert fields f1 and f2 in table f, I would like to have a couple of text items that will automatically populate and display based on the data typed in f1 and from a query on table t. In other word, typing empno in f1, moving to the f2 box the user should automaticcaly see emp last_name and first_name displayed. This way she can verify the empno typed was correct, type other data in f2 and submit the form.
Any suggestion on how to reach this scope?
I guess I should use some javascript.
Many thanks.
Gabriele
I can see that you may already be going down the javascript road... I have worked with APEX 3.2 in the past on a similar problem and didn't know enough javascript to get this kind of dynamic functionality out of the platform. If you want to try an alternate method, you can tweak your user interface to provide the functionality you seek, but you'll have to check with the users if they can adapt to this style of approach, which you should be able to do using existing 3.2 functionality and no additional scripting:
Requirements:
Based on field inputs of field1 and field2, auto-populate an apex page with data from an existing table so that the user can confirm if the data inputted in field1 and field2 are correct. Use the data from the search results as parameters for input to a new process or form.
Suggested Design:
Page 1: start with blank page, then add three items: field1, field2 and a "submit" button.
add page process (after submit, conditional on submit button item) that redirects to Page 2 setting items P2_field1 and P2_field2 from the values on the Page 1 form.
Page 2: report page. set a query that will search your table of existing employees searching by partial or full string searches against the employee table. APEX 3.2 used to use the following convention to get around partial and case sensitive matching issues:
SELECT field1, field2, field3, field4, ...
FROM table1
WHERE instr(upper(field1), upper(:P2_field1), 1, 1) > 0
AND instr(upper(field2), upper(:P2_field2), 1, 1) > 0
Creating a report output "table" for a search effort helps when there may be multiple answers to the same search query... such as with common names like "Smith" or "John"...
Add an additional column to the report output or make the name column (field1 or field2) a linked column that goes to page 3 feeding the queried data from the employee table.
add the property to the column value on the report editor: redirect to "page 3", set P3_key = #KEY# where page 3 is your final target form or process and "key" is the index or primary key of the table you are searching from... or you can feed the n number of values needed already queried from the lookup table to populate the page 2 report.
I guess the important take-home is an appreciation for the limitations of older versions of Apex along with possible limitations with knowledge of advanced web scripting languages. Your requirements are still achievable even if all you are familiar with is the basics of your Apex version and a little SQL.
note: if your lookup table is HUGE (hundreds of thousands of records), you may need to consult your DBA for help with optimizing the table to handle the SQL search query. Adding UPPER and INSTR functions to the WHERE criteria isn't really the most efficient way to query large datasets without additional help in optimization.
I'm not a 3.2 wiz, i haven't worked with it. I do know there are no dynamic actions and no page on-demand process points, so i've worked with that in mind.
Application process: 'get_emp_details'
(I don't know the processing point you can/need to choose in 3.2. If you can pick on-demand, go for it.)
Code:
DECLARE
lEmp emp%rowtype;
BEGIN
SELECT *
INTO lEmp
FROM emp
WHERE empno = :P7_EMPNO;
htp.p('{"result":"ok", "emp":{"ename":"'||lEmp.ename||'","job":"'||lEmp.job||'"}}');
EXCEPTION
WHEN no_data_found THEN
htp.p('{"result":"nok", "emp":{}}');
END;
On your page (in my case i have used page 7, change the page numbers to your page number!) put this in the html headers section (or if you have the javascript section there, put the javascript code there without the script tags):
Edit: added in console.log lines. Mind that these will produce javascript errors in IE when the developer tools (F12) are not opened!
<script type="text/javascript">
$(document).ready(function(){
console.log('document is ready');
$("#P7_EMPNO").change(function(){
var oDBGet = new htmldb_Get(null, $('pFlowId').val(), "APPLICATION_PROCESS=get_emp_details", '0');
oDBGet.add('P7_EMPNO', $("#P7_EMPNO").val());
var oResult = oDBGet.get();
var lReturn = $.parseJSON(oResult);
if(lReturn.result=='ok'){
var lEmp = lReturn.emp;
$("#P7_ENAME").val(lEmp.ename);
} else {
alert('ename could not be found for this empno!');
};
});
});
</script>
This will do an ajax callback when the EMPNO changes, and change the ename of the employee. Since there are no temporary application items that can be used (apex_application.g_x##) i'm submitting the required page item to session state. In this case P7_EMPNO is added to the p_arg_names array and its value in the corresponding position in the p_arg_values array. This is necessary because the session state of P7_EMPNO is used in the application process.
When inspecting the console tab in Firebug:
Post values of ajax call:
Response values of ajax call:
Tom,
here is your code modified to work in my system:
-- On Demand Application Process get_emp_details:
DECLARE
lEmp emp%rowtype;
BEGIN
SELECT *
INTO lEmp
FROM emp
WHERE badge = :P2_BADGE;
htp.p('{"result":"ok", "emp":{"lastname":"'||lEmp.last_name||'","firstname":"'||lEmp.first_name||'"}}');
EXCEPTION
WHEN no_data_found THEN
htp.p('{"result":"nok", "emp":{}}');
END;
-- Javascript in page header:
<script type="text/javascript">
$(document).ready(function(){
$("#P2_BADGE").change(function(){
$.post('wwv_flow.show',
{"p_request" : "APPLICATION_PROCESS=get_emp_details",
"p_flow_id" : $v('pFlowId'),
"p_flow_step_id" : '0',
"p_instance" : $v('pInstance'),
"p_arg_names" : ['P2_BADGE'],
"p_arg_values" : [$('#P2_BADGE').val()]},
function(data){
if(data){
var lReturn = $.parseJSON(data);
if(lReturn.result=='ok'){
var lEmp = lReturn.emp;
$("#P2_LAST_NAME").val(lEmp.last_name);
} else {
alert('ename could not be found for this empno!');
};
};
}
);
});
});
</script>
I am probably missing something and it doesn't display last_name in P2_LAST_NAME when I type the badge number in P2_BADGE.
Thanks,
Gabriele
Tom, here is what I have in the watch right window of firebug for the function(data) step:
this
function()
arguments
[
"wwv_flow.show"
, Object { p_request=
"APPLICATION_PROCESS=get_emp_details"
, p_flow_id=
"120"
, p_flow_step_id=
"0"
, more...}, function()]
0
"wwv_flow.show"
1
Object { p_request=
"APPLICATION_PROCESS=get_emp_details"
, p_flow_id=
"120"
, p_flow_step_id=
"0"
, more...}
p_arg_names
[
"P2_BADGE"
]
p_arg_values
[
"155752"
]
p_flow_id
"120"
p_flow_step_id
"0"
p_instance
"2189517750670012"
p_request
"APPLICATION_PROCESS=get_emp_details"
2
function()
e
"wwv_flow.show"
r
Object { p_request=
"APPLICATION_PROCESS=get_emp_details"
, p_flow_id=
"120"
, p_flow_step_id=
"0"
, more...}
p_arg_names
[
"P2_BADGE"
]
p_arg_values
[
"155752"
]
p_flow_id
"120"
p_flow_step_id
"0"
p_instance
"2189517750670012"
p_request
"APPLICATION_PROCESS=get_emp_details"
s
undefined
i
function()
toString
function()
Closure Scope
Closure Scope { toString=function()}
Closure Scope
Closure Scope { toString=function()}
Window
Window f?p=120:2:2189517750670012::NO:::
From my firebug 1.10.6 console:
document is ready
f?p=12...::NO::: (line 29)
empno has changed
f?p=12...::NO::: (line 31)
performing a post with empno value: undefined
f?p=12...::NO::: (line 32)
POST
400 Bad Request
48ms
wwv_fl....min.js (line 2)
HeadersPostResponseHTMLCookies
<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN">
<HTML><HEAD>
<TITLE>400 Bad Request</TITLE>
</HEAD><BODY><H1>Bad Request</H1>
The HTTP client sent a request that this server could not understand.</BODY></HTML>
-- And the post tab:
Parametersapplication/x-www-form-urlencoded
p_arg_names[] P2_BADGE
p_arg_values[] 155751
p_flow_id 120
p_flow_step_id 0
p_instance 4466366911674565
p_request APPLICATION_PROCESS=get_emp_details
Source
p_request=APPLICATION_PROCESS%3Dget_emp_details&p_flow_id=120&p_flow_step_id=0&p_instance=4466366911674565&p_arg_names%5B%5D=P2_BADGE&p_arg_values%5B%5D=155751

How to get large photo URL in one API call?

I want to display large photos in my Facebook app's feed view immediately. Is it possible to get the large photo src URL from a stream/feed using one API call? The photo id is returned in the stream/feed and one can of course then supply this photo id in an additional FQL or graph API call to retrieve all the information about that photo. However, is there a way using multi-query or batch calls to get a larger photo src url using one API roundtrip?
I haven't tried this with stream/feed photos, but the generally accepted way of doing this is:
http://graph.facebook.com/{ID of object}/picture
If you want the "large" version, you would do:
http://graph.facebook.com/{ID of object}/picture?type=large
I'm not 100% sure if this would work for an actual photo (instead of a user profile picture or page profile pic), but I have a hunch it will - the only caveat is that you obviously must have a logged in user that is authorized to view the photo (unless it's public).
If anybody is looking to this and type large is not enough, I found other solutions.
Type large is kind of small anyway (close to 200px). You can get larger image by adding i.e. ?width=1000 or ?height=1000. Facebook will return picture closest to given dimension and preserve aspect ratio. When passing both dimenstions like ?width=1000&height=1000, facebook will cut image to given dimensions (in this case square).
Use Facebook UserId (Oject ID) to get the picture.
https://graph.facebook.com/173xxxx8635/picture?type=large&redirect=false
which returns JSON data with picture URL.
{
"data": {
"is_silhouette": false,
"url": "https://fbcdn-profile-a.akamaihd.net/xxx/xyz/1cc066a2cae3f301d"
}
}
A good trick with the new api is to get the pic_cover field from the event table and to process it according to the size you want to use
I found when I was having this trouble that it turned out to be the picture I was downloading rather than the size I was setting it.
If for example I downloaded all my photos with a request of
[FBRequestConnection startWithGraphPath:#"/me/photos?fields=created_time,name,picture&type=tagged" parameters:nil HTTPMethod:#"GET" completionHandler:^(FBRequestConnection * connection, id result, NSError *error) {
NSDictionary * userData = (NSDictionary *)result;
NSMutableArray * array = [[NSMutableArray alloc] initWithArray:userData[#"data"]];
for (NSDictionary * dict in eventsToAdd) {
UIImage * image = dict[#"picture"]
}
}];
I am using the dictionary key search "picture" as I want the picture.
This though will get me a lower quality picture than if I searched for "source" in this search:
[FBRequestConnection startWithGraphPath:#"/me/photos?fields=created_time,name,source&type=tagged" parameters:nil HTTPMethod:#"GET" completionHandler:^(FBRequestConnection * connection, id result, NSError *error) {
NSDictionary * userData = (NSDictionary *)result;
NSMutableArray * array = [[NSMutableArray alloc] initWithArray:userData[#"data"]];
for (NSDictionary * dict in eventsToAdd) {
UIImage * image = dict[#"source"]
}
}];
If you go on the Facebook API explorer and search for photos and then click on the picture and source jpg links you can see the difference in size and quality.
Since changing this method I have managed to get rid of using the type parameters as it doesn't seem to make a different.
Note: I am using iPhone and not iPad or a larger screen so I don't know how this affects bigger screens.
The answer by #streetlogics works fine but only on pictures that have {object_id}.
http://graph.facebook.com/{object_id}/picture
But I also wanted large pictures for the feed's shared links, which sometimes don't have {object_id}. I finally realized that the {picture} thumbnail URL contains the encoded URL for the original site's large image:
https://external.xx.fbcdn.net/safe_image.php?d=AQBe9UvGd0vPbAHP&w=130&h=130&url=http%3A%2F%2Fskift.com%2Fwp-content%2Fuploads%2F2015%2F12%2Fpollution.jpg&cfs=1
--> contains -->
http://skift.com/wp-content/uploads/2015/12/pollution.jpg
So I made a loop that checks for {object_id} and if not present then extracts the URL from {picture}:
if(isset($post['object_id'])) {
echo "http://graph.facebook.com/".$post['object_id']."/picture";
}
elseif(isset($post['picture'])) {
echo urldecode(preg_replace('/&cfs.*/', '', preg_replace('/.*url=/', '', $post['picture'])));
}
else {
echo "no_large_image";
}