sbt "scopes" its tasks and settings. Documentation talks about the "scope axes" One of them is the "configuration axis" which seems to be magically selected for the evaluation of a task. Evaluate "compile" and the configuration scope is "Compile". Evaluate "test" and the configuration scope is "Test".
I cannot find anything in the documentation on how this selection is done. The most specific description on this topic I found was in James Roper's "all that jazz" blog. It says:
When you don’t specify a scope, sbt will choose a default scope, in this case it has chosen the "compile" scope. The logic in how it makes that decision we won’t cover here.
Pretty terse, actually. So, my questions:
Is there anywhere else a coverage on that specific "logic"?
How does sbt choose the "configuration" scope it applies to the task/settings evaluation?
Which "configuration" scope values exist? Is this a fixed list? Where do the existing configuration scope values come from? How (and why) to create new ones?
Is there any "general idea" behind this whole scoping thing?
I feel that this is a key to understand sbt in general but I must admin that I currently have almost no clue on how sbt acts here and why it acts like this. Can anyone enlighten me?
Related
can anyone help me with this?
Actions on google deployment
I already removed this pattern and checked several times if it exists somewhere else. Not to fine.
Please help, thank you.
Kind regards
Check if in any of your attempts, you have correctly defined the names of the variables, more when working in languages, remember that all must have the same variable name when selecting an entity.
For example, I once had the same error because one of the variables was different from the other variables because it was not spelled correctly.
It is also the case that if instead of adding a training phrase, you add an entity directly, a common mistake is that this entity has no variable assigned to it.
It doesn't necessarily have to be the same name as the entity, but it does have to be in common with the phrases you're training or variables you're already using, otherwise dialogflow will throw you the error that that entity is not linked to any variable or query.
I'm writing powershell scripts to query data from Octopus, sooner or later I need to have a reference to a project or environment in the script. Is it better to use the Name, Id or slug in these situations? Readability and maintainability / brittleness of the scripts are concerns.
ID seems unambiguous, but is hard to read in the script as it has no semantic meaning (eg "Projects-197")
Name is very human readable, but seems likely to change for style reasons (eg "Public web app")
slug seems to be just a url-safe version of the name ("eg "public-web-app")
Many relative ID's can be looked up from other queries (the release-ID for a given deployment for example) but eventually we need to refer to one specific project.
In general, I would recommend using the ID as the name is not necessarily unique (variable names, for example) and the ID can't be changed.
When you say "is hard to read in the script as it has no semantic meaning" does that imply you are hard-coding the name/id into the script? Why not put a comment? For example,
$project-id = "Projects-123" # ID of "My Awesome Project"
I want to add a new section in control panel of liferay and within that section I want to have my custom-portlet. I did it using ext. However I want to do it with hook . Is it possible ?
I don't think it would be that easy with a hook, because of the following reasons:
You can't modify in a Hook - the class com.liferay.portal.util.PortletCategoryKeys which contains the keys for displaying the different sections. The different sections are hard-coded in this class in a String array ALL.
You can't modify the logic of PortalImpl#isControlPanelPortlet() which uses the PortletCategoryKeys#ALL to determine if the request in question is for a control panel portlet.
Then you also have another method which you can't modify with a Hook and is used extensively PortalImpl#getControlPanelCategory()
Doing it with a hook:
I have not tried this but I think if you need to do it with a hook you would have to change all those JSPs which make use of PortletCategoryKeys#ALL and the methods of PortalImpl as stated above, and provide your custom implementation for all these methods in the JSP.
I would really like to know how you implemented it with an EXT may be just the steps or the methods you have overridden. So that I can try to convert those in terms of a hook.
This is as far as my understanding goes. Hope this helps.
With the advent of Marketplace, ControlPanel has a new category named "Marketplace" and that section is introduced in a plugin. However, I never checked if 6.1 GA2 introduced a new section that this plugin just fills. Check the marketplace plugin if you can find a trace of this section implemented there.
On the other hand, nobody has yet named any section that definitely required a new section (though some have asked me how to solve the same problem). For this reason, you might want to re-think the requirement and choose one of the existing sections. If you don't, at least I'd be interested in the name and purpose of the new section - I might find a first one actually justifying this kind of implementation...
My toolchain (Realview) does not have implementation of unistd.h, I have a code that I need to use that is using unistd.h, some function like access, open.
So I need to implement this functions by myself.
My question is, is there another alternative or some reference (except the function description, found here) that I can use in order to port/implement these functions?
What is the correct way to do that? It's my first time posting.
The "function description" you link to has the page title "The Open Group Base Specifications Issue 6", which indicates that it's perhaps a bit more of a heavy-weight than just being a mere description.
If you implement a function that fulfills all the text on that page correctly, chances are you've managed to re-implement it in a compatible way, so that programs complying to the same standard will build and work.
Not sure what else you feel you would need.
Is it possible to modify the order of the intellisense options shown when I hit ctrl-space?
Specifically, I'd like to order them in scope, so that if I have a variable in my function that matches what I've typed so far then it goes to the top of the list. If there's a member in the class, that's next, etc. I'm just kind of sick of having to type enough that I don't match some random global symbol in Windows' crypto libraries or whatever.
Is this kind of this possible? Where do I start? I looked for an obvious option in vs2010, but didn't find anything.
My programming language is native c++.
Here's what I've found so far.
A walkthrough showing how to add items to the intellisense popup: http://msdn.microsoft.com/en-us/library/ee372314.aspx
The walkthrough uses the ICompletionSession interface:
http://msdn.microsoft.com/en-us/library/microsoft.visualstudio.language.intellisense.icompletionsession.aspx
It looks as if you might be able to clear the CompletionSets, and replace them with your own completion sets, but I'm not sure if there's any way to change the order, or if they are always sorted. There is the IIntellisensePresenter interface, but it looks like it's pretty much empty.
Also, from the walkthrough, it looks like you can fill the CompletionSets with a dumb list of strings, so you'd need some other way of interrogating the symbols to determine scope, which may not always be possible.
In short: Probably not possible at the moment.