chain-modules/text-anonymize
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

feat: text.anonymize v0.1.0 — regex-based PII redaction

Browse source 
First release. Pure-Rust, in-WASM, regex-only, declares no
permissions. Suitable as a first redaction pass after
text.extract before any cloud-LLM step.

Detection categories:

  EMAIL          RFC-5321-ish local@domain, IDN-aware.
  PHONE          International (+CC …) and DE national
                 (030 …, 0151-…) shapes, 7..20 raw digits.
  IBAN           Word-bounded [A-Z]{2}\d{2}[A-Z0-9]{11,30}.
                 Structural only — MOD-97 checksum
                 deliberately skipped so partial / truncated
                 tokens in running text still get redacted.
  BIC            8 or 11 uppercase alnum.
  IPV4           Four 0..255 octets, dot-separated.
  GERMAN_TAX_ID  11 consecutive digits, word-bounded.
  CUSTOM         Operator-supplied bare terms from the
                 newline-separated `custom_terms` input,
                 matched whole-word case-insensitive.

Token shape: ⟦TYPE_N⟧ — U+27E6 / U+27E7 mathematical white
square brackets. Distinct from any plain ASCII `[…]` already
present in source text (Markdown links, legal citations,
code blocks) so a reviewer never has to guess which `[…]`
is a redaction.

Outputs:

  anonymized  text  Input with PII replaced by ⟦TYPE_N⟧.
                    Counter restarts at 1 per type so the
                    tokens stay operator-readable.
  report      json  { redactions: [{type, token, original,
                    offset}…], counts: { TYPE: n, … } }.
                    Full original-text reconstruction is
                    possible from this — the GDPR
                    Art. 32(1)(a) "ability to undo"
                    requirement.

Quality bar (7 unit tests):
  * email round-trip
  * IBAN + BIC don't eat each other
  * three phone-number shapes redact
  * IPv4 only matches valid 0..255 octets
  * custom_terms case-insensitive
  * no double-redaction on overlapping patterns
  * per-category counter resets correctly

Built artefact: target/wasm32-wasip2/release/text_anonymize.wasm
(~180 KiB stripped).

NER for free-text names / organisations / locations is the
v0.2.0 plan once a benchmarked ONNX model is selected; the
operator's `custom_terms` field is the v0.1.0 escape hatch.

Signed-off-by: flemming-it <sf@flemming.it>
This commit is contained in:
flemming-it 2026-05-25 20:53:24 +02:00
commit 734d8f6e6f
8 changed files with 1164 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

75
MODULE.md Normal file
View file

@ -0,0 +1,75 @@
# text.anonymize
Regex-based PII anonymization for plain text. Detects and
replaces personally-identifiable patterns with stable tokens
of the shape `⟦TYPE_N⟧`, and emits a JSON report describing
every redaction so a downstream verify step can audit
completeness.
## Capability
- `text.anonymize@0.1.0`
## Inputs
| Name | Type | Description |
| -------------- | ---- | ------------------------------------------------------ |
| `text` | text | UTF-8 string to redact. |
| `custom_terms` | text | Optional. Newline-separated list of extra bare terms to redact whole-word (case-insensitive). |
## Outputs
| Name | Type | Description |
| ------------ | ---- | -------------------------------------------------------------------------------------------- |
| `anonymized` | text | The input text with PII replaced by `⟦TYPE_N⟧` tokens. |
| `report` | json | `{ redactions: [{ type, token, original, offset }…], counts: { TYPE: n, … } }`. |
## Categories detected
| Type | Pattern |
| --------------- | ---------------------------------------------------------------------- |
| `EMAIL` | RFC-5321-ish `local@domain`, IDN domains OK. |
| `PHONE` | International (`+49 …`) or German national (`030 …`, `0151-…`) shape. |
| `IBAN` | Word-bounded `[A-Z]{2}\d{2}[A-Z0-9]{11,30}`. Structural only — checksum deliberately not computed. |
| `BIC` | Word-bounded uppercase alnum, 8 or 11 chars. |
| `IPV4` | Four 0..255 octets, dot-separated. |
| `GERMAN_TAX_ID` | 11 consecutive digits, word-bounded. |
| `CUSTOM` | Every line of `custom_terms` matched whole-word, case-insensitive. |
No NER yet. v0.2.0 will add a small ONNX model for name /
organisation detection once we have a benchmarked candidate.
## Permissions
None. Pure-Rust, regex-only, in-WASM. No filesystem, no network,
no LLM call.
## Example flow
```yaml
name: redact-case-file
inputs:
document: bytes
steps:
- id: extract
use: text.extract@^0
with:
document: $inputs.document
- id: redact
use: text.anonymize@^0
with:
text: $extract.extracted.pages[*].text
custom_terms: |
Mustermann
Musterfrau
outputs:
redacted: $redact.anonymized
audit_report: $redact.report
```
## Build
```bash
cargo build --release --target wasm32-wasip2
# Output: target/wasm32-wasip2/release/text_anonymize.wasm
```