newindex is expected to be run when the current working is the 'doc' subdirectory of a project. It will create a subdirectory '000.Biblio' in which it makes symbolic links to all 'tib' files in the directory under the current directory. This method (hopefully) avoids the problems of having too much string space used when building the command to build the index that was encountered in the earlier attempts a year ago.
lookup runs 'tiblook' on the most likely INDEX file. The search path used is basically '.:$PROJECT:$HOME'
For more information on tiblook, find some OLD Unix documentation. We don't seem to have it on-line.
$ pwd /home/feijoa/horton/380.1/doc $ newindex Tibdex -- version 2.2, released 12/15/89. 10 documents 72 distinct keys 129 key occurrences $ lookup Tiblook -- version 2.2, released 12/15/89. Enter keys separated by spaces for search, <return> to exit ... *discovery %T PVC Discovery %R /home/feijoa/horton/380.1/doc/005.Project_Deliverables/001.Research_Planning/ 008.PVCDiscovery %A horton %C NEC %D 03-May-94 %O Notes on uploading PVCs from switches into the NMS and presenting in the NMS map model. * $
PATH=$PATH:~horton/bin
Because of the searching for the 000.Biblio directory, lookup can be slow to start.