Razzak: Thank you for addressing this matter.
Please be assured I am NOT implying an inherent R:BASE anomaly; but rather, anomalous behaviour in a code construct which has worked very well within this and other databases. I will take your phrase: 'bogus computed columns' as my Blues Clue, and following your steps, understand how I created the problem. As is my established practice, any anomaly I can reliably demonstrate to be a bug will be submitted privately to RDCC for the DreamTeam to address. Yours, Bruce Chitiea SafeSectors, Inc. 909.238.9012 cell -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of A. Razzak Memon Sent: Thursday, February 26, 2015 5:51 PM To: RBASE-L Mailing List Subject: [RBASE-L] - Re: ALTER TABLE Anomaly At 08:24 PM 2/26/2015, Bruce A. Chitiea wrote: >Within an RMD file, attempting to add a column to a table with a >construct which works in other contexts: > >ALTER TABLE &vWDIncidentTotals_t + >ADD COLUMN UnitCount INTEGER > >I receive the error: > >"-ERROR- As column BuildingNUM is used in another table, its type >cannot be redefined. (2942)" > >The existing table columns are: 'BuildingNUM' and 'IncidentCOUNT'. > >'UnitCOUNT' is NOT a computed column. I'm not touching 'BuildingNUM'. > >What's up? Your lack of knowledge in constructing a proper text string, and not properly evaluating the actual table with bogus computed columns, shall NOT constitute an anomaly in R:BASE. FYI ... When altering a TABLE and Adding/Removing column(s), the entire table and the entire database relational integrity is re-evaluated. Thus, the result of such -ERROR- message. That's what's up, if you would like to know. Try to add the same column using the GUI RBDEFINE and see what find. At the R> prompt, LIST COLUMN BuildingNUM should provide you with complete details and all associated tables. Re-Building the database using UNLOAD ALL technique should also illustrate any issues with you database concerning the relational integrity. Very Best R:egards, Razzak.
