azure / azurekusto Goto Github PK

{dbplyr}, {rlang}, and a few other dependencies have evolved significantly since this package was written. Need to rewrite the dplyr translation backend, probably taking a direct dependency on {dbplyr} instead of reimplementing much of its internals.

There are important files that Microsoft projects should all have that are not present in this repository. A pull request has been opened to add the missing file(s). When the pr is merged this issue will be closed automatically.

Microsoft teams can learn more about this effort and share feedback within the open source guidance available internally.

Merge this pull request

New issue since GH doesn't allow transferring across orgs

Some specific use cases require username/password rather than an AAD token

Hi AzureKusto R package authors.

Firstly, thank you for the wonderful package!

Forgive me if I am missing something obvious, but is it possible to use dbWriteTable to write a table into a specific folder in Kusto? I cannot seem to figure out how to do this. Help is appreciated!

If i try to create
const kustoClient = new KustoClient(kcb);

It fails with source undefined from inside the pacakge
cancelToken: import("axios").CancelTokenSource;

Please help me to addess the issue.

Versions used in my pacakge.json
"axios": "^1.6.7",
"azure-kusto-data": "^5.2.3",

I think AzureKusto could be a great tool for data scientist using R. But I have trouble to set it up correctly. After I installed AzureKusto library, I run the following code but got an error:

testpoint<-kusto_database_endpoint(server="https://kusto.xxxx.microsoft.com",database="test")
res2<-run_query(testpoint, "buildsetcalculator | take 20")

The second line throws out an error as follows:

res2<-run_query(testpoint, "buildsetcalculator | take 20")
Error: $ operator is invalid for atomic vectors

Even if I run an incorrect query as follows, the error is same
res2<-run_query(testpoint, "dfasqrwe")

Unlike the server address used in example, this cluster server address has the name of cluster 'xxxx' behind 'kusto'. It may be the reason.

Thanks for developing AzureKusto!

Hello there!

Suggested change to wrap the param names in square brackets in the method build_param_list. This would enable us to auto-create tables with periods in the column names (i.e. for the sample R "iris" dataset for instance). However, I don't know if this would have adverse effects elsewhere in the code, so leaving it to you to figure out whether this is the right fix, thanks!

build_param_list <- function(query_params)
{
ps <- mapply(function(name, value)
{
type <- switch(class(value)[1],
"logical"="bool",
"numeric"="real",
"integer64"="long",
"integer"="int",
"Date"="datetime",
"POSIXct"="datetime",
"string"
)
paste(name, type, sep=":")
}, paste0('[', names(query_params), ']'), query_params)

paste0("(", paste(ps, collapse=", "), ")")

}

I received an error: Kusto.Cloud.Platform.Storage.PersistentStorage.PersistentStoragePathDoesNotExistException when querying a Kusto cluster with the default options "queryconsistency": "weakconsistency".

We should remove this from the default options, as weakconsistency can result in query failures under certain cluster configurations.

Also, it appears that passing query options as arguments to run_query is not correctly overriding the default options.

Thanks for your work on this package, its really helpful!

I'm having trouble getting around the default timeout settings when running KQL through AzureKusto.

It doesn't look like norequesttimeout = true is having any impact when inside KQL query. I'm wondering if there is a way to specify this at connection?

the is.kql function doesn't seem to exist, so %in% errors. %in~ seems to work (for vectors) however, as a workaround.

New issue since GH doesn't allow transferring across orgs