This chapter lists all the preferences that are pertinent to index server. Find a description of how to set, view, change or delete preferences, with the HELIOS utility programs “prefdump”, “prefvalue”, and “prefrestore”, in the HELIOS Base manual.
Make sure that preference keys DO NOT start or end with a slash (“/”) character, and note that they are case-sensitive! Also, if any preference key or preference value includes spaces, that key or value must be enclosed in quotes.
Key: Programs/indexsrv/<preference>
List of attributes that are not to be stored in the index database. For
example, setting ExcludedAttributes to “kMDItemInstructions”
would prevent the “instructions” from being indexed. Setting this
preference applies only to new or changed files, it does not remove the
attributes from the existing database.
If set to TRUE, access to the index server service port is
enabled to remote users (subject to the restrictions specified in the IP
access list).
Specifies the “telnet” port number.
Specifies the maximal number of open file descriptors.
(srvutil reconf indexsrv) to take effect:Key: Programs/indexsrv/<preference>
Specifies the time in seconds before the index database is flushed.
Specifies the number of changes in the database before the database is flushed.
This preference determines how long (in seconds) “indexsrv” plug-ins may run
before a time-out occurs. Specifying 0 disables the running
time (i.e. “run forever”).
Some slower plug-ins (“officeindexer”, “pdfindexer”) reset the timer
as long as they are continuously processing data, so that the total running
time can be longer than the value of this preference.
This preference determines how long (in seconds) “indexsrv” waits before the indexing plug-in process is killed. This happens only if RunTime has timed out.
Specifies the time interval (in seconds) after the file event is received before indexing is started.
This preference controls the maximum number of “indexsrv” processes that may run simultaneously. Default is the number of CPU cores.
It is recommended that the value for MaxProc be less or
equal to the number of CPU cores.
Specifies the file name of the IP access list that controls client access to the index server service port.
(srvutil reconf indexsrv), and
the volume to be mounted anew, to take effect:Key: Volumes/<UNIX directory>/<preference>
If the key contains a path, the “/” characters within the path must be escaped by preceding each with a “\” character, e.g.:
# prefvalue -k 'Volumes/\/data1\/demovol/Index' -t bool TRUE
Defines whether this volume supports a Spotlight index. This preference
corresponds to the Spotlight Index checkbox in the HELIOS Admin
Volume menu Spotlight tab.
This preferences allows switching off Spotlight indexing of text in PDF
documents. This preference corresponds to the PDF Text Indexing checkbox in the HELIOS Admin
Volume menu Spotlight tab.
This preferences allows switching off Spotlight indexing of text in Office
documents. This preference corresponds to the Office File Indexing checkbox in the HELIOS Admin
Volume menu Spotlight tab.
Automatically tracks changes to files in the specified volume. This preference
corresponds to the Automatic Indexing checkbox in the HELIOS Admin
Volume menu Spotlight tab.
Excludes one or more specified directories of indexed volumes from indexing.
This preference corresponds to the entries in the Omit Indexing
for Directories field in the HELIOS Admin Volume menu Spotlight tab.
Key: Programs/indexsrv/Plugins/<plug-in name>/<preference>
Defines the file suffixes that are subject to index, e.g. “txt”, “c” or “h” for the txtindexer plug-in, and “jpg”, “jpeg” or “png” for the oiindexer plug-in.
For this preference all extensions (i.e. also the default values) must be specified in addition to the desired extensions. The default values are:
oiindexer: tif, tiff, jpg, jpeg, eps, epsf, sct,
psd, psb, pct, pic, bmp, jb2, jbig2, jp2, jpf, jpx, png, raw, cr2, nef,
arw, dng, orf, raf, pef, sr2, crw, x3f, erf, mef, rw2, xpv
txtindexer: txt, c, h
pdfindexer: pdf
mediaindexer: mp3, wma, wmv, m4p, m4a, aax, mov
annotationindexer: annotation
officeindexer: doc, docx, xls, pps, ppt, pptx,
odt, ods, odp, rtf, pdf
zipindexer: zip
indesignindexer: indd, indt
Defines the file types that are subject to index. By default, the plug-ins use the file suffix rather than the file type.
If full-text is indexed, the amount of data that is stored in the database can be significant. So there are two additional preferences that can tune the behavior of the plug-in to the desired amount of data. These preferences are stored in:
Programs/indexsrv/Plugins/<Plugin-name>/<volumename>/<suffix /<ftype>
where <volumename>, <suffix> and
<ftype> can be empty. So if you wish to enable
FilterDuplicates for your plug-in by default, you can set
Programs/indexsrv/Plugins/txtindexer////FilterDuplicates to TRUE.
Determines whether duplicate words are stored for a specific file. Set
this preference to TRUE if you wish to ignore the second,
third, fourth, etc. instance of a word within a file when the database
is indexed during a full-text search. This has the advantage that the
database does not increase more than is necessary. The disadvantage is
that searches for phrases will not succeed. Only searches for one or
more separate words will return results, words with 1 or 2 letters are
not indexed.
Determines the path (must be absolute!) to an SQLite full-text database that contains in the “sWords” column keywords that should not be stored. As an example, in an English text it is not useful to store words like “the”, “than”, “also”, “although”, etc.
The advantage of using this filter is that text elements, which have no specific value in searches, e.g. conjunctions or articles, can be prevented from being stored in the database so that it does not increase more than is necessary. The disadvantage is that searches for exact phrases containing these words will not succeed.
See 5.2 “Create a word list database” for instructions on how to create and set up a word list database.
Allows increasing the Java memory allocation, where <size>
is the maximum Java heap size in megabytes. This may become necessary
when indexing large documents.
This preference is called with the following key:
Key: Programs/indexsrv/Plugins/officeindexer/TmpDir
The “officeindexer” plug-in creates for each job a separate directory (by default in “HELIOSDIR/var/tmp”), where all the temporary files are placed, and removes this directory when the indexing is done. This preference can be used to change the location of the temporary directory. The path must be specified with “/” as directory delimiter, even on Windows.
> prefvalue -k "Programs/indexsrv/Plugins/officeindexer /TmpDir" -t str "/d:/tika-tmpdir"
# prefvalue -k "Programs/indexsrv/Plugins/officeindexer /TmpDir" -t str "/data/tika-tmpdir"
This preference is called with the following key:
Key: Programs/indexsrv/Plugins/oiindexer/XMPAttributes/<URI>
The “oiindexer” plug-in can extract custom metadata from the XMP block of an image, and write them to the database. Read 5.1 “Custom XMP metadata attributes” for a detailled description of this feature.
This string list contains either KEY so that the corresponding value is stored in the database as a string under this KEY, or KEY;ENTRY, so the corresponding value is stored in the database in the ENTRY column. Or it contains KEY;ENTRY;TYPE, which makes the value being stored in the database in the ENTRY column, with the corresponding type. TYPE can either be “int”, “real”, “bool” or “date”.
prefvalue -k "Programs/indexsrv/Plugins/oiindexer
/XMPAttributes
/http:\/\/de.helios.samplenamespace\/"
-t strlist "Barcode;kMDItemBarcode;
real,Vendor;kMDItemVendor,GoodsGroup"