Overview
QuikData's Autoredaction workflow can be used to search and redact simple words, but also has two main pattern matching syntax rules for identifying and redacting content using either DTSearch's pattern matching syntax or Regular Expressions.
DTSearch Pattern Matching Syntax
DTSearch can be used to identify and redact based on simple patterns, for example searching for a US Social Security number.
Important: To enable DTSearch pattern matching, the search term must be prefaced with the $$ symbol to enable the system to recognize it as a DTSearch-enabled pattern.
Note: By default, DTSearch does not index non-alphanumeric characters like "-", or other punctuation -- they are treated as spaces.
Examples
| US Social Security Number | $$=== == ==== |
| Credit-card-like sequences | $$==== ==== ==== ==== |
| US Phone 10 digit phone numbers | $$=== === ==== |
Regular Expressions Pattern Matching Syntax
Regular expressions can be used when the patterns being search for are not accommodated through DTSearch or for more complex patterns.
Important: To enable regular expressions, the search term must be prefaced with the ## symbol to enable the system to recognize it as a regular expression pattern.
What flavor we use
Engine: .NET Framework regex (
System.Text.RegularExpressions) — the classic .NET flavor (not PCRE/RE2/ECMAScript).
See: Microsoft’s Regex language quick reference, options, and grouping docs. Microsoft Learn+2Microsoft Learn+2Default options in code:
RegexOptions.IgnoreCase | RegexOptions.Multiline.
Inline flags like(?i),(?m),(?x)are supported..NET supports lookbehind, but variable-length lookbehind has restrictions; keep it fixed-length. Prefer forward “scoped” matches instead of complex lookbehind.
-
Capture Groups:
Our redactor follows a groups-first rule:
If your regex has capturing groups, only captured spans are boxed (Group 1..N).
If no groups are present, the whole match is boxed.
Examples
| Email Address | ##[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,}\b |
| US Phone Number | ##\b(?:\+?1\D*)?(?:\(?\d{3}\)?\D*)?\d{3}\D*\d{4}\b |
| SSN | ##\b\d{3}[-\s]?\d{2}[-\s]?\d{4}\b |
| Tax ID / EIN (##-#######) | ##\b\d{2}[-\s]?\d{7}\b |
Need help?
If a pattern feels brittle (OCR variance, multilingual text, odd page layouts), or you want us to encode a complex rule as a reusable recipe, contact QuikData Support and include:
2–3 sample pages (sanitized if needed),
the rule you want (in plain English),
which portion must be redacted (label vs value),
whether it should be limited to a specific block/section (e.g., “Complainant”).
Comments
0 comments
Please sign in to leave a comment.