Specifications

Page 16 DocSTAR Level 2 Service Training Workbook
X1 The horizontal coordinates of the beginning of the uncertainty on
the scanned image file. (Used to draw the highlight box on the
image).
Y1 The vertical coordinates of the beginning of the uncertainty on the
scanned image file. (Used to draw the highlight box on the image).
X2 The horizontal coordinates of the end of the uncertainty on the
scanned image file. (Used to draw the highlight box on the image).
Y2 The vertical coordinates of the end of the uncertainty on the
scanned image file. (Used to draw the highlight box on the image).
Unindex table
DOCID The DOCID of a document that is in the queue for having the
indexed information removed from the Full-Text Index. (The
Unindex queue is processed the next time that a document is filed,
the index information will be removed after the Unindex queue is
processed and a backup of the database occurs).
What is the Zylab ZyIndex Full-Text Index?
The Zylab ZyIndex Full-Text Index is basically an alpha-numeric listing of words, built
from the document title, keywords, and OCR generated text (if the document was
‘READ’), that were found in the documents that have been scanned and filed in a
DocSTAR. These words have a pointer associated with them. The pointer is actually the
document number (DOCID) that these words were seen in. The pointer allows the search
engine to very quickly locate documents that contain these words in the DocSTAR
database without performing a database search, which can be very lengthy as the
database grows in size.
The index or indices for each table in the database, which was described earlier in the
database section, should not be confused with the Full-Text Index used in DocSTAR. The
Full-Text Index (Zylab ZyIndex) used in DocSTAR is not found inside the database. The
Full-Text Index is located in the C:\DOCSTAR\DATABASE\MAINIDX directory and is
comprised of a number of files. There is no utility to view or edit the Full-Text Index. The
relationship between the database and the full-text index can be seen in Figure 2.3.
Figure 2.3: Relationship between Database & Full-Text Index
One of the files found in the C:\DOCSTAR\DATABASE\MAINIDX\INDEX directory is the
Mainidx.noi file, the ‘Noise Word’ file. This file contains a list of letters and words that will
not be indexed since they are not unique to any document, in most cases, and would just
cause the Full-Text Index size to grow tremendously.