> There's already code that checks for an existing lyxpipe. If another > instance of lyx is connected to the other end, a message is printed > to that effect, otherwise it is a stale pipe (maybe a left over from > a previous crash) and is removed. > > So, instead of printing a message to the console, the other instance > could be instructed to open a new document. Re-use of code, no new > dependencies on external libraries, everything under our control. >
Enrico, You're right and I'm not fighting. I just took a solution which was 'ready' and asked for comments. I had no intention to add dependencies, to overdo your work or to add new features in the beta phase or anything else without proper evaluation and discussion of the to-be-added stuff. The last thing I wanted to do is to spark a discussion between the conservative people and the Qt-aholics (with all do respect). I'd like you to guide me a bit in with respect to using the pipes then. Detecting whether a pipe exists is indeed easy. However, I need to add some public functions to the Server interface that relays to LyXComm. Moreover, sending a new command over the pipe to the existing application needs some of the client code.. right ? If you can share your thoughts with me, I'd appreciate it. Vincent
