toenuff / pshodata Goto Github PK

Looks like < /get is being dropped and < delete is there twice.

Having an issue pasting the xml data

Consider protecting the psobject in the module and creating accessor functions:

Get-OdataClass |new-odataendpoint

Not sure if this is really needed. There's not much harm in allowing people to access the objects natively.

Get a working example of update. I'm sure there will be issues.

When multiple FieldParameters are used with Delete, the entire odata service does not run.

Additionally, parameters in options are not usable when delete is called.

Need to understand whether Delete is only usable with PK or if there is a way to get it to work with parameters or fieldparameters.

Multiple fieldname/parametername are being set in . There should be one for each param.

Consider validation cmdlet that ensures that other verbs will work with PK, i.e., they should have a fieldparametermap value for pk or at least some fieldparametermap

Currently we only support strings. We need to support switch as well. This should be automatically determined by inspecting the function (cmdlet property) and its parameters.

Need to ensure that the Case that is applied is the exact case of the parameter from get-command. If the wrong case is used, it should fix and supply a warning to the end user.

Hi,

The URL for the Management OData Schema Designer download, doesn't go anywhere and no current reference on the archive page lists it. I also did many Google searches (though possibly for the wrong thing) and every link references the broken link right now. Do you have a better link that works or the files that you could post?

http://archive.msdn.microsoft.com/mgmtODataWebServ/Release/ProjectReleases.aspx?ReleaseId=5877

-Adam

Need to create a create sample and create issues that I know will need to be created in order to support it.

This should be automatically determined by inspecting the function (cmdlet property) and its parameters. However, I'm unsure how needed it is. Strings are working fine, but the schema designer allows other types. It probably makes sense to support this.

Modify psd1 to only export the 3 user functions.

We should be able to create and publish directly from the tools rather than copying to IIS manually. This is one thing the schema designer does really well.

Lines like this:
$text += " "_14 + ("{0}" -f $parameterset.name)
Should be written without parens:
$text += " "_14 + "{0}" -f $parameterset.name

In other words, I should be able to use something like this in within schema.xml:
"get-eventlog -log application"

I believe this is supported. It needs to be tested and ensure that it works with add-pshodatamethod.

Psh! Why Not?!?!

A neat feature in the odata engine is the ability to use FieldParameterMap (the section that works with querying via filters direct to using appropriate parameters) with a mapping. Therefore, someone could query via filter (blah eq 'blahblah') and blah could be mapped to the name parameter of a cmdlet.

It would be easy to support this, but we have to give some thought to how the data is represented in the class.verb object.

This is built into the odata engine. We only need to expose it by grabbing all of the possible parameter sets that are possible from the cmdlet. This is the way the schema designer currently works.

Ideally, this should be intelligent enough to only include parameter sets that are needed for the possible parameters allowed by the class.

Too much hardcoded pathing. Need to fix.

Need documentation for the exported functions.
Need a readme.md with an overview.
Create some pages on the twiki with examples of how to deploy the code manually to IIS with odata extensions installed.
Finally, set up the website with a better overview than readme.md is :)

Oh, and a really open license that gives me nothing but good feelings.

Should test to see if you can remove params that are not used in the parameterset section of schema.xml. If it works, the autogenerator should not use them in the parameterset section.

Need to give consideration to how this works: If one parameter is used and it is only in one parameterset, we need to be sure that the schema.xml does NOT include a bunch of parametersets with no parameters in them.

Cannot see diffs in github when the file changes.