[Laura]

Thanks Leah for the response.

I’m able to set/not set password to the local analyst instance. My question was for connecting to another remote server (requires auth). For remote server connections, I was asking if there were alternate methods to specify password once and its re-used everytime.

Inspector in new tab is helpful.

I’m looking at ways to reduce number of things I need to do to get csv/excel output. Also, it could randomly choose a file name.

[Laura]

I’ve chunked with .Q.fs[{`trade insert flip colnames!(“**********”;”,”)0:x}]`:filename and it runs until it crashed.

Did some more research and thought it could be a gc issue, so I added a gc call but that didn’t help me either.

Dumb question, is this bc I’m using w32 instead of w64?

[Laura]

Hey Sam,
Thanks (again) for checking the logs. I can now confirm that there was a sync error on my vm. Same error repeated today and after I reloaded my vm I could see new data on HDB.

That said, I have one final question:
How do we manually EOD under this new setup? Does calling end from a TP scratchpad still works?

Cheers,
Adriano

[Laura]

Hi Tom,

Try using pastebin for pasting the xml, thats how I got around attaching my logs on one of my threads

[Laura]

Hi Tom!

At a glance, is it possible that the Selected Min/Max values actually constrain the range of the slider? (e.g. if the range doesn’t include all values in the data source the component doesn’t work correctly)

If not, my first step in debugging this would be to replicate it in a new dashboard (i.e. just setup a range slider or two with with the same data sources) as this can sometimes make the issue very apparent when you’re setting it up the second time. If it DOES still occur post the xml for that test dashboard here and someone might be able to debug the issue from that.

Kind regards,

Josh

[Laura]

Hi Sam,
I have attached pictures of the configuration as requested and put links for the logs below.
(The portal does not accept .log files as attachments )

I put logs for 01/07 (when i was originally facing issues) and today.
Funny thing: i noticed today that data flowed (!?!?). So if nothing catches your eyes, i will probably consider my initial issue as a synchronization error on my vm setup.

Bonus:
Log File List on LR
ENV=HOME=/TP/uiCore/uiCore_tp_*%DATE%
ENV=HOME=/TP/uiCore/uiCore_tp_*%DATE%_H_[0-9][0-9][0-9][0-9].[0-9][0-9].[0-9][0-9].[0-9][0-9][0-9][0-9][0-9][0-9]

LR Logs

2021-07-01D05:14: https://pastebin.com/gjVvGU5s
2021-07-07D02:30: https://pastebin.com/XpBTt1JM

2021-07-07D02:30: https://pastebin.com/NUFqRxCQ

 

 

 

 

[Laura]

Hi Kate and David,

Both of your tips were great, thank you very much

It seems I will have to adopt the solution for on-disk tables that Kate suggested because the function doSomething is routed to the IHDB and performs its operations on on-disk tables.

So, when people is an on-disk table, if i try pass by value:

people: `name xkey people

then the IHDB throws a int error message
if i do:

`name xkey select from people

then the IHDB is happy. Since I won’t dare to change the key of my table on the IHDB, it looks like pulling the values to the memory first is the safest option.

Thanks again!

[Laura]

Thanks for the reply Josh,

I was able to replicate the issue in a blank Dashboard with Min/Max Selected view states within the range of the data source too.

I cant attach the XML as the website is saying the file type is not supported. Do you have an email address or something I can send it to?

 

 

 

[Laura]

Hi Adriano,

We’ve had a look at your example and for in-memory tables your first solution appears to work perfectly as long as you are passing a reference into your function.

for example:

 

doSomething[`tablename]

 

However if you were working with an on-disk table rather than an in-memory table your solution could be simplified to

 

q)r:`sym xkey select from tableName

 

using the select function in order to pull your table into memory first.

Your functional form solution works fine however most people will use q-SQL as it is a more readable format. A where clause can be used alongside the select query to avoid pulling in a large amount of data from disk.

Cheers, 

Kate

[Laura]

ANSWER: table.update(‘col’,col=py list)

[Laura]

Ah ha!
I knew it would be something silly like that.
My user is in a group that i created called wombatDevTest. Adding my user to the ActionTrackerUsers group did the trick and now I can see the Action Tracker dashboard.

In case you are wondering what AdminAPIs are applied to wombatDevTest, here they are:

Thanks

[Laura]

so play along with me:

1) have tried table[`col] or table.col assignment — no go

2) have tried “update” which naturally won’t work given that im not trying to create a new col from data existing within the table itself

3) have tried to upsert the column as its own table into the existing table (it should work given they’re of the same length) but thusfar no go.

what’s the pyq-onic way to do this?

[Laura]

For others looking for PyQ documentation. This is the most up to date pdf I was able to find 

[Laura]

 

There is no inbuilt weighted correlation, but you can easily implement Pearson Weighted Correlation as follows

wcov: {[x;y;w] w wavg (x-w wavg x)*y-w wavg y} // weighted covariance

wcor: {[x;y;w] wcov[x;y;w]%sqrt wcov[x;x;w]*wcov[y;y;w]} // weighted correlation

-Ajay