I think workspace sensitivity is to cater for linux/unix users. It's theorectically possible to use to directories that only differs in cases on those systems, though in reality this is hardly a case.

_If_ we want to get rid of case sensitivity in file names, then my 2 cents is that: (1) Store file names in a case-sensitive way (no change to sensor and persistence mechanism) (2) Change WorkspaceFile.getFileName() to return either all lower case or all upper case file names.

The advantage is that if we change our mind later, it's easy to go back to the case-sensitive way.

Another option is to only make workspace root case-insensitive, but relative file name case-sensitive.

Of couse, there is a third option, status-quo.

By the way, file pattern matcher, used extensively in telemetry, is case-sensitive.

Cheers,

Cedric



Philip Johnson wrote:

I was looking at my daily email from Hackystat this morning and my active time did not seem correct. Upon inspection, I discovered that the problem was due to the fact that while "C:\svn" was set as a workspace root, "c:\svn" was not. In other words, workspaces are case sensitive.

I have always thought this was a mistake, and I'd like to raise this issue for discussion once again. So, here are some questions:

(1) Is there any reason why workspaces should be case-sensitive?

(2) Aren't there significant advantages to case-insensitive workspaces, such as: - increased robustness (example: my problem this morning would not have occurred)
   - reduced numbers of workspaces

(3) How would case-insensitivity be implemented?
   - on the sensor-side, when providing file paths to the sensor shell?
   - on the client-side, upon receipt of sensor data but before storage?
   - during Workspace instance construction?
   - during Workspace instance comparison?

(4) Are there implementation issues we need to be aware of?
   - Hongbing is best suited to answer this question.

(5) Is this somehow a problem unique to me?
- Do others observe/require multiple workspace roots with different cases?

Posted to Jira as <http://hackydev.ics.hawaii.edu:8080/browse/HACK-471>, so this doesn't get lost in the holiday shuffle.

Cheers,
Philip

Reply via email to