Open jinujoseph opened 8 years ago
One interesting point about this: what types of characters do we want to support between classes as part of the file name? Do we want to just do +
and .
, or others as well? One immediate idea that comes to mind is the following algorithm:
Making 2 smart enough is going to be the interesting part. One initial thought I have is to make it support any valid path character that is not a valid identifier character for the language. It should also handle cases like:
Another point: should ordering matter? Ie, should I be able to have a file named foo+bar+baz.cs with the following class definitions:
class baz {
class bar {
}
class foo {
}
}
rename it to test+bar+baz.cs, and expect to see:
class baz {
class bar {
}
class test {
}
}
yes , we do just + and . only i had some notes here https://microsoft.sharepoint.com/teams/managedlanguages/_layouts/15/WopiFrame.aspx?sourcedoc={79b652be-6aa1-4feb-8d23-fa9127483ce9}&action=edit&wd=target%28%2F%2FProject%20System.one%7C89e7c9d2-f741-4060-a593-54327019c089%2FFileRename%7Ce2d3b337-7322-4ae0-b02a-5421594fa134%2F%29
Another complex scenario:
class c1 {
class c2 {
class c3 {
}
class c5 {
}
}
class c4 {
}
}
Do we need to specify how the name of the file is created for this? I can see arguments for naming this either c1+c2+c4+c3+c5.cs
(breadth-first) or c1+c2+c3+c5+c4.cs
(depth-first). Additionally, what if it's named c1+c4+c2+c5+c4.cs
(reversed order of classes).
Related to #109