Skip to content

chore(deps): review and reimplement runtime#18

Open
martsokha wants to merge 5 commits intomainfrom
feature/runtime
Open

chore(deps): review and reimplement runtime#18
martsokha wants to merge 5 commits intomainfrom
feature/runtime

Conversation

@martsokha
Copy link
Member

No description provided.

…es 0.3.32, uuid 1.21, jiff 0.2.20

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@martsokha martsokha self-assigned this Feb 15, 2026
@martsokha martsokha added docs improvements, updates or additions to docs feat request for or implementation of a new feature labels Feb 15, 2026
martsokha and others added 4 commits February 16, 2026 07:43
…les, extract text redaction primitives

Rename nvisy-ingest crate to nvisy-codec and update all workspace references.
Split render/image.rs into a folder-module (mod.rs, blur.rs, block.rs) with
improved documentation. Create render/text/ module and move text redaction
primitives (PendingRedaction → PendingReplacement, apply_text_redactions →
apply_replacements, apply_cell_redaction → mask_cell, hash_string) from
pipeline's apply.rs into nvisy-codec. Enable jpeg alongside png for the
pipeline's image-redaction feature. Remove old pipeline render/block.rs and
render/blur.rs modules.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
…ype safety

Move method() and replacement_value() to inner enums (TextRedactionOutput,
ImageRedactionOutput, AudioRedactionOutput, TextRedactionSpec, etc.) so
RedactionOutput and RedactionSpec delegate instead of duplicating match
logic. Drop Copy/Eq/Hash from AudioRedactionSpec for consistency. Add
audio_id to AudioLocation, type AnnotationLabel.scope as AnnotationScope
enum, make Redaction.original_value non-optional, and add confidence field.
Delete duplicated replacement_value() in nvisy-codec.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
…text/tabular redaction apply

Implement AsText trait on TxtHandler so apply_text_doc can use the
trait's redact() method instead of manually reconstructing content.
Narrow mask_cell to accept TextRedactionOutput directly instead of
RedactionOutput, removing the default_mask_char param. Simplify
apply_tabular_doc accordingly.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
…, tighten module visibility

Split AsText into AsText + AsRedactableText and AsImage into AsImage +
AsRedactableImage with blanket impls, mirroring the codec/redaction
separation. Add BoundingBoxU32 for pixel-exact rendering and use it in
blur/block/pixelate renderers. Make render submodules private, remove
internal re-exports from handler/mod.rs, and add pixelate renderer.

Also includes ontology improvements: strum Display derives, typed
AnnotationScope, non-optional Redaction.original_value, Entity helper
methods, and deduplicated spec/output method() delegation.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

docs improvements, updates or additions to docs feat request for or implementation of a new feature

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant