On Thursday 26 February 2009 05:52:49 pm Ricardo Aráoz wrote: > johnf wrote: > > On Thursday 26 February 2009 10:03:24 am Ricardo Aráoz wrote: > >> Hi John, while waiting for my proposed change to the module to be > >> approved or rejected I had some time and did some regression testing. I > >> was under the impression that I had been able to use the wizard to > >> generate a form connecting to mssql. But it would seem I wasn't. > >> After many regressions I can report that I was not able to even connect > >> to my mssql server (through the cxnEditor) up till revision #3907 (you > >> removed the code supporting 'port') in which I'm able to connect but I > >> get the unicode error we've been talking about. > >> Have you tried the proposed changes in your system? Has anyone a reason > >> not to commit them? > >> I have no urgency as they are pretty straightforward and I can easily > >> modify the file after every svn update. But considering the suggested > >> change is really trivial and if there are no concrete objections to > >> doing it and there are no other proposed ways of solving the problem > >> (not even under research), then it would be nice to commit the change. > >> If there are concrete objections please let me know so that I can look > >> for some other way to correct the module. > >> > >> TIA > >> > >> Ricardo. > > > > I thought I told you that your patch does NOT work for me. > > No, I did not know that. > > > I had several > > errors from CD. I in fact asked my friend in Germany to also test your > > patch and it did NOT work. He is using unicode and encoding (german). > > But I'm not sure if his version of Dabo was up todate. The issue is about > > unicode and encoding. At the moment I don't have a solution that works. > > Do you remember the error you got? It's strange that the pymssql will > not parse a query if the name of the database is a string and only work > if it is unicode (and the opposite with me). Seems the module is more > fragile than I thought it was. > > > Paul, the patch is simple > > dbName=str(self.database) > > > > The trouble is it does not work on my standard install of sql2000 nor > > sql2005 (at least NOT cleanly). The encoding is "latin 1" for both SQL > > servers and I have tried from Linux and windows XP(32bit). BTW do you > > have a 64 bit install of windows? > > You seem to assume the trouble lies with the database server. But the > error message clearly shows that the query is not parsed correctly when > it reaches the server, so I am assuming the problem lies elsewhere. > > > I think you could help by discovering why the unicode/encoding is not > > being excepted by your database. You might try something like "print > > dbName.decode()" > > But if the trouble is that the unicode "is not being excepted by your > database" then why would I try printing dbName.decode()? If the problem > lies with the database this will accomplish nothing. > > > or anything else you can think of. You know most of what > > has been done in Dabo was to scratch a personal itch. Help by scratching > > your itch. You find the trouble, I'm sure we will get the patch into the > > code. > > No, don't worry. As I already told you those three lines scratch my > itch. Besides I'm not in production, just testing and trying to get the > thing to work. > Thanks anyway. > > Ricardo. First I got this from your email Returned mail: Requested action not taken: mailbox unavailable From: Mail Delivery Subsystem <[email protected]> To: [email protected] The original message was received at Thu, 26 Feb 2009 17:34:51 -0800 EST from wavecable.net [24.113.32.25]
----- The following addresses had permanent fatal errors ----- <[email protected]> ----- Transcript of session follows ----- >>> RCPT TO:<[email protected]> <<< 550 This account is not [email protected] unnamed unnamed The above email contained all the info I discovered from my testing. The error I get is "KeyError" I forget the rest. But I get it several times in the terminal when using CD. I don't get the error when NOT using your patch. I know all the tables have primary keys that I'm connecting to so it has to be something with the patch. > But if the trouble is that the unicode "is not being excepted by your > database" then why would I try printing dbName.decode()? If the problem > lies with the database this will accomplish nothing. I actually don't know the root cause. But if it has nothing to do with the encoding of the database why do my two SQL servers work with the existing code and yours does not. If not the database then it has to be what is being passed to the variable "dbName". Therefore, I would be trying to discover the difference. I noted the Ed got MsSQL working recently. Maybe he'll discover the root cause. On the other hand I don't use MsSQL very much. So if Paul or Ed wants your patch to go in - I don't have an objection. -- John Fabiani _______________________________________________ Post Messages to: [email protected] Subscription Maintenance: http://leafe.com/mailman/listinfo/dabo-users Searchable Archives: http://leafe.com/archives/search/dabo-users This message: http://leafe.com/archives/byMID/[email protected]
