naming-conventions

Sometimes I see variable names with double underscore in the beginning and the end. For example: Article.__elasticsearch__ Is there some naming convention related to double underscores in Ruby variable names? An initial underscore or double underscore basically indicates "special/avoid overwrite" --meaning it's meant to reduce the likelihood that someone else might define a method/attribute of the same name. The most common occurrence is __send__ . From Ruby Forum The author of the ElasticSearch gem made the wrong call IMO. At the end of the thread, Avdi Grimm, who is well-known in the Ruby

In JRE, Sun's internal packages are prefixed with 2 top-level domains (sun and com). For example, com.sun.security.jgss sun.security.jgss It seems pretty random to me which prefix they choose. I am curious what rules Sun uses for this. The "com.sun" convention is the more preferable format because it follows the "naming conventions" that have been established for naming Java packages. http://java.sun.com/docs/codeconv/html/CodeConventions.doc8.html You're supposed to use your unique company or personal website URL as the first few words in the package to guarantee uniqueness in the namespace.

I am relatively new to ASP.NET MVC, and am very impressed with the clarity of the platform so far. However, there is one aspect that I find uncomfortable. At first, I accepted the fact that when I say return View(); I am calling a helper method that returns an ActionResult, and makes some assumptions about which view to present, route values, etc. But lately I have been writing code that looks more like this: return View("Index", new { id = myID }) because it is immediately clear to me what's happening by reading that single line of code. Lately I have been struggling with the fact that I can

If I have user id variable, what is correct java naming convention for it? userId or userID I think I read somewhere that the best guideline is to always capitalize only the first letter of an acronym (i.e., if you have a user TTL where TTL is some random acronym in your project, then it's best to write userTtl). This makes sense, since it solves some problems. For example, say you want a user id counter. Readability of userIDCounter is worse than userIdCounter. For a longer acronym, it gets even worse. This is a tough decision, even Sun isn't so sure what to do. Look at classes in JRE, java

Java's standard libraries seem to use camelCase for method names. Native functions like nanoTime() are no exceptions. If so, why is System.arraycopy not camelCased? Is there something special about System.arraycopy ? It's been in Java for before v1.0 was released - so my guess is that it predates the naming conventions, and it was missed in a sweep of the API when the naming conventions were decided. (In other news, NullPointerException should be called NullReferenceException .) 来源： https://stackoverflow.com/questions/8421103/why-is-system-arraycopy-not-camelcased

I'll prefix by saying that I understand that both Code Analysis and StyleCop are meant as guidelines, and many people chose to ignore these anyway. But having said that, I'd like to see what the general consensus is with regard to these two rules. Rule CA1500 says don't make parameter names and private field names the same. Rule SA1309 , on the other hand, says don't prefix members with underscore or "m_". This leaves us with little options for distinguishing private backing fields from their corresponding parameters. Take these examples. SA1309 complains: class SomeClass { int _someField;