Perceptive Search can read PKZIP files. ZIP files are treated by Perceptive Search as if they are sub-directories. That is, if only one entry in a file is changed or created, only that entry is indexed and the entire ZIP file is not Reindexed. Entries in a ZIP file are considered to have the same name as the ZIP file, with the component file name following in parentheses.
For example, F:\PUBLIC\ARCHIVE\Y1993.ZIP (PRUDACK.WP5) identifies the document PRUDACK.WP5 that has been included into the ZIP file Y1993.ZIP.
The documents within a ZIP can be of different formats. However, when Perceptive Search does the scanning part of its Update, it has to decompress the first kilobyte of each document to determine its format. If Perceptive Search indexes the document, it does not have to decompress the first kilobyte next time it does an Update (because the document is already in the index). However, if the document does not index for whatever reason, the next time you do an Update, Perceptive Search will decompress the first kilobyte again to try and determine the format once again.
Therefore, explicit rules for ZIP files are preferable. When using AUTO there will be a performance degradation in the scanning phase during the first indexing run and for all subsequent Updates if several documents cannot be indexed. Query performance is not affected in either case.
If you activate a document within a ZIP file, Perceptive Search will extract the document to a temporary file and open that file into the originating application. You can view, but cannot change documents in ZIP files through Perceptive Search. The temporary file, however, can be saved under a new name.