Peter Jakobsson posted some extensive upgrade advice the other day. I wonder if any of these points might be related to the problems I'm seeing?
> [5] - the 4D 'Position' and 'Replace String' functions. If you use > these for any kind of string processing in the course of your > request > handling, better review the new behavior. The main 'gotchas' are > that > low level Ascii characters are ignored now by these functions. This could be part of the problem as it is searching through header text which might contain low level Ascii characters. Does any one have a replacement for Position? But maybe not, position wouldn't work differently compiled vs interpreted. > [6] - the Active 4D 'On Request' handler has a parameter change - > see > the docs We don't use this. > > [7] - the housekeeper call is now in a loop - I don't remember this > being the case in previous versions. If you use the 'out of the > box' > shell you won't have to worry about this > > [8] -you need to tell Active 4D what callback to use for processing > your requests submitted via 'post' - again, if you use the A4D > shell > 'out of the box then it's been updated for you My problem only appears to happen with POST requests Peter, if you are reading this can you please elaborate what you mean by 'again'? We are using NTK for the web server. Not 4D. I also noticed that when I upgraded the shell that the new NTK shell no longer contains the method A4D_NTK_StatusCallback? Was that an intentional ommission? Thanks, Brad On 8/31/10 10:33 AM, Brad Perkins wrote: > Up until now all v11.7/Active4Dv5 development testing has been done > using 4D Remote (client) running interpreted. > > 4D Remote v11.7 > Mac OS X Server (10.6.4) for client > Windows 2003 Server for server > Active4D v5.0r19 > NTK 1.1.3 as the web server > > I compiled the structure this morning and immediately hit a road block. > > When any POST requests are submitted I'm seeing the following error: > > Error when executing the method "A4D_HTTPD_RequestHandler" at line > number 97. > > Invalid parameters in an Execute command. > > The same requests seem be be fine when running interpreted. By > "fine" I mean I'm getting the expected results. > > Line 97 of our A4D_HTTPD_RequestHandler is a blank line so I'm in > the dark about the "Invalid Parameters" > > We have modified this method to contain a few A4D_HTTPD_RequestHook > calls. I just compared our method to the provided shell method with > a diff tool and those are the only differences with one exception. > The provided shell method has this code: > > For ($i;1;Get last table number) > $table:=Table($i) > > If (Is table number valid($i)) > If (Not(Read only state($table->))) > UNLOAD RECORD($table->) > READ ONLY($table->) > End if > End if > End for > > > I changed this slightly. I've logged this as a possible bug as ID > 0001090 in the issue tracker. Do you want to get a pointer on a > table number before testing if it is valid? > > For ($i;1;Get last table number) > `$table:=Table($i) `??? Shouldn't this be inside > the following if...end if ??? > > If (Is table number valid($i)) > $table:=Table($i) `11-Aug-2010, bdp, bug fix > If (Not(Read only state($table->))) > UNLOAD RECORD($table->) > READ ONLY($table->) > End if > End if > End for > > Any ideas? > > -- Brad Perkins _______________________________________________ Active4D-dev mailing list [email protected] http://mailman.aparajitaworld.com/mailman/listinfo/active4d-dev Archives: http://mailman.aparajitaworld.com/archive/active4d-dev/
