Re: New patch for bugs in GPDB csv format log handling
От | Chuck McDevitt |
---|---|
Тема | Re: New patch for bugs in GPDB csv format log handling |
Дата | |
Msg-id | 2106D8DC89010842BABA5CD03FEA4061CA55CEC0@EXVMBX018-10.exch018.msoutlookonline.net обсуждение исходный текст |
Ответ на | Re: New patch for bugs in GPDB csv format log handling (Dave Page <dpage@pgadmin.org>) |
Ответы |
Re: New patch for bugs in GPDB csv format log
handling
Re: New patch for bugs in GPDB csv format log handling Re: New patch for bugs in GPDB csv format log handling |
Список | pgadmin-hackers |
Thanks Dave. Some questions: > Hi, > > Some issues with this, 2 minor, one less so: > > - Please rearrange the code in frmStatus.h to remove the class > definition. We try to keep the headers purely for declarations and > only include one-line functions in them. I would suggest a new file in > utils/ given that it's a generally useful class. I would have done this already, but was worried you wouldn't accept new files this late in the cycle. I will put the class definition in its own file. Should I also move the class declaration to its own .h file? I think that would be cleaner. > > - Status bar messages such as _("Reading log from server.") should > generally end with ... to indicate the task is ongoing. Will do. > > - Messages passed to wxLogError should be translatable, and explain > the problem to the user without being phrased as a question (wxT("Log > line does not start with timestamp?: %s \n")). Messages passed to > wxLogNotice needn't be translated however. Will do. Hopefully, this can only be hit if a log file is corrupted. Or should this just be a wxLogNotice, since the code is supposed to just recover any continue without any user interaction. I'd made it a LogError mostly to help with my debugging when I was working on the code. > > Regards, Dave. > > -- > Dave Page > EnterpriseDB UK: http://www.enterprisedb.com
В списке pgadmin-hackers по дате отправления: