Here is the procedure I used. I hope it helps the next newbie who is in this situation.
Through a process of trial and error, I uninstalled other packages one at a time. Since I have 36 of them, this took some time.
After each uninstall (un-checking them in the Component -> Install Packages list box), I checked the box for TeeChart 7 Components until I found the correct conflicting package. (After uninstalling the conflicting package, the error message did not appear when I installed TeeChart.)
In my case, the conflicting package was TMS Component Pack. I looked at the documentation which came with it and found in the install.txt file:
Notes for TeeChart Pro users:
The TMS Component Pack comes with support for the default installed
TeeChart in Delphi & C++Builder. If you have installed another version
of TeeChart, open the TMS Component Pack package file, remove TEE*.*
from the requires list of the package an press Install. Delphi or
C++Builder will then automatically add the correct TeeChart library
reference to the TMS Component Pack package file.
At this point, I now had a TeeChart tab on BCB, and I had lost all of my 13-or so TMS tabs.
The next step was to reinstall the TMS components.
Just to be safe, I copied the entire TMSDIR subdirectory to another drive so that I would not ruin anything by experimenting. (This was unnecessary, but I felt better at the time.)
I opened the file TMSC5.BPK, removed TEE50.bpi from it, added TEE7C5.bpi to the package, clicked "Install," and the TMSC5.BPK package compiled and installed in BCB5.
I looked up and all of my TMS component tabs had reappeared.
Thank you again, Narcis, for your help in dealing with this problem.
I only wish that there were a better way for BCB to describe the conflict so that it would point to a solution. The fact that Tee50.bpi was the culprit and that it was hidden in TMS's .bpk package was a serious problem. It did not help that when I searched through my BCB subdirectories for files containing either "Tee50" or "Teeconst" I found that very many non-human-readable (non-expert-readable?) files contained one or both of them; enough to make that approach unfeasible.
There ought to be a better way than trial-and-error to locate this type of conflict. Still, I did find the path through the maze (this time), thanks to you, and the foresight of TMS (they make great stuff, by the way).
Keep up the good work!
-- Jim Gold --
jim@goha.com