Crate dbmd_core 

dbmd-core — the reference library for db.md, the open database in plain files.

db.md is one directory: raw evidence in sources/, atomic typed data plus curator-synthesized conclusions (meta-type: conclusion) in records/, and a single DB.md config file at the root. Records are markdown files with YAML frontmatter; relationships are wiki-links; the index is the derived, write-through index.md / index.jsonl catalog plus embedded ripgrep.

This crate owns all toolkit logic. The dbmd binary (dbmd-cli) is a thin wrapper that parses args, calls into here, and formats output. Any Rust tool wanting to be db.md-aware can cargo add dbmd-core and get the full library — the same shape as ripgrep, where the grep/ignore libs do the work and rg is a thin CLI.

Hard invariants this crate is built to uphold

• Zero AI/LLM dependencies. No provider SDKs, no API keys, no model calls, no embeddings, no vectors, no ANN — anywhere, ever. The agent driving dbmd is the semantic layer; dbmd is a deterministic tool.
• The interactive loop is O(changed), never O(store). Loop ops (graph::backlinks, validate::validate_working_set, index::Index::on_write, …) never call store::Store::walk on a non-empty changed set. The one documented exception is validate::validate_working_set, which falls back to a full sweep only when handed an empty changed set (the vacuous-pass guard). Whole-store walks otherwise belong only to SWEEP ops (validate::validate_all, index::Index::rebuild_all, stats).
• Wiki-links are full store-relative paths. A short-form wiki-link is a validation error (validate code WIKI_LINK_SHORT_FORM).
• Embedded ripgrep. Free-text body search uses the grep + ignore crates in-process; the toolkit never bundles or shells out to rg. Structured loop reads (graph::backlinks, query::Query) ride the index.jsonl sidecars instead, never a frontmatter tree scan.

Re-exports

pub use assets::AssetRecord;

pub use assets::Declaration;

pub use assets::ScanReport;

pub use assets::StatusReport;

pub use assets::VerifyReport;

pub use extract::ExtractError;

pub use extract::Extracted;

pub use extract::Format;

pub use extract::MetaValue;

pub use fsx::write_atomic;

pub use fsx::write_atomic_new;

pub use graph::ContextSlice;

pub use index::Index;

pub use index::IndexLevel;

pub use index::IndexRecord;

pub use log::Log;

pub use log::LogEntry;

pub use log::LogKind;

pub use parser::Config;

pub use parser::FieldSpec;

pub use parser::Frontmatter;

pub use parser::MarkdownLink;

pub use parser::ParseError;

pub use parser::Schema;

pub use parser::Section;

pub use parser::Shape;

pub use parser::WikiLink;

pub use query::Query;

pub use render::Outline;

pub use render::Tree;

pub use store::infer_type_from_path;

pub use store::layer_for_type;

pub use store::Layer;

pub use store::NotAStore;

pub use store::Store;

pub use store::StoreError;

pub use time::now;

pub use validate::Issue;

pub use validate::Severity;

Modules

assets

assets — the db.md asset layer.

extract

Document text extraction — the dbmd extract engine.

fsx

fsx — the one atomic, durable file write for db.md’s primary data.

graph

graph — the wiki-link relationship layer.

index

index — the hierarchical content catalog.

log

log — the append-only, month-rotating chronological log.

parser

parser — read and write db.md markdown files.

query

query — Dataview-style filters, sidecar-backed.

render

render — data structures for the structural views, no output formatting.

stats

stats — store overview, computed on demand (a SWEEP, like du — never a maintained or precomputed cache).

store

store — walk, locate, and shard a db.md store.

summary

summary — the deterministic default-summary composer.

time

Canonical wall-clock for write-surface timestamp seeding.

validate

validate — the validation engine.

Enums

Error

Top-level error for dbmd-core operations.

Type Aliases

Result

Crate-wide result alias over Error.