Overview
QuikData uses dtSearch as its full-text indexing and search engine. When documents are loaded into a review room, dtSearch indexes their content so reviewers can run keyword and phrase searches across the document set.
How dtSearch indexes content is controlled by three configuration files that are automatically generated when a room is created. These files live on the file server — they are not exposed in the QuikData application interface. This article explains what each file does and how to modify them when needed.
Configuration Files
When a new room is created, QuikData generates three dtSearch configuration files and stores them in a dedicated dtSearch folder within that room's directory on the file server:
| File | Purpose |
|---|---|
noise.dat |
The stop word list — words excluded from the full-text index |
default.abc |
The alphabet file — defines how characters are treated during indexing |
filetype.xml |
File type rules — controls how dtSearch parses specific file formats |
noise.dat — Stop Words
The noise.dat file defines the stop word list — common words that dtSearch skips entirely during indexing because they are too frequent to be useful in search results. Stop words cannot be searched in full-text search, even if they appear in document content.
When a phrase contains a stop word, dtSearch skips over it. For example, searching for "statue of liberty" retrieves documents containing "statue", any intervening word, and "liberty" — because "of" is a stop word.
Default Stop Words
default.abc — Alphabet and Character Rules
The default.abc file defines how dtSearch treats individual characters during indexing. This determines what counts as part of a word, what separates words, and what gets stripped out entirely.
| Category | Characters | Behavior |
|---|---|---|
| Letters | A–Z, a–z, 0–9, underscore (_) |
Indexed as part of words; case is normalized |
| Hyphens | - |
Word connector — hyphenated terms are indexed both joined and as individual words |
| Spaces | ! " # $ & ' ( ) * + , / : ; < = > ? @ [ \ ] ^ ` { | } ~ |
Break words apart during indexing |
| Ignored | % |
Stripped and ignored entirely |
% character is the fuzzy search operator in QuikData search syntax. It is ignored at the indexing level by design. The file also includes CJK (Chinese, Japanese, Korean) character ranges for auto-detection and word-breaking in those scripts.filetype.xml — File Type Parsing Rules
The filetype.xml file controls how dtSearch handles specific file formats during indexing. Each entry specifies whether dtSearch uses a built-in parser, an IFilter, or bypasses processing for that file type.
| File Type | Parser | Status |
|---|---|---|
.MPP, .ODB
|
ANSI text parser | Enabled |
.eml |
Legacy dtSearch 6.40 parser | Disabled |
| CSV | - | Disabled |
Microsoft Access (.mdb) |
- | Disabled |
.vsd, .jtd, .jtdc, .jtt, .jttc
|
Installed IFilter | Enabled |
.msg (Outlook messages) |
Container parser | Disabled |
Modifying dtSearch Configuration Files
Configuration changes — such as adding or removing stop words, adjusting character rules, or enabling a file type parser — must be made by editing the relevant file directly on the file server. After any changes, the affected documents must be reindexed for the new configuration to take effect.
For SaaS Clients
SaaS clients do not have direct access to the file server. To request a dtSearch configuration change:
- Contact your QuikData Project Manager, or
- Email support@quikdata.com with details of the change needed and which room(s) are affected
For On-Premises Clients
On-premises clients manage their own file servers. A site admin with file server access can edit the configuration files directly.
- Navigate to the room's dtSearch folder on the file server.
- Open and edit the relevant file (
noise.dat,default.abc, orfiletype.xml). - Save the file.
- Reindex the room using the steps in the next section.
Reindexing After Configuration Changes
Changes to dtSearch configuration files only take effect after the affected documents are reindexed. To reindex a room:
- In QuikData, navigate to Manage > Reports > Upload Session Report.
- Select all Upload Sessions using the checkbox in the top-left.
- Right-click and choose Submit for Indexing.
Notes
- Configuration files are generated automatically at room creation using QuikData defaults. If your organization requires a custom configuration applied to all new rooms, contact your QuikData Project Manager.
- These files are not versioned or backed up automatically. Keep a record of any customizations made, including which room and what was changed.
- For a complete reference on how the dtSearch index affects search behavior — including noise words, wildcards, and alphabet customization — see the Full Text Search Syntax article.
Comments
0 comments
Please sign in to leave a comment.