@glion/parser 0.14.1

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2023 Rethink Health
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -0,0 +1,153 @@
+# HL7v2 Parser
+
+This package is a utility that takes a raw HL7v2 message as input and turns it into an HL7v2 syntax tree following the [@glion/ast](../hl7v2-ast/README.md) definition.
+
+This utility is a low level project. It’s used in [@glion/hl7v2](../hl7v2/), which focusses on making it easier to transform content by abstracting these internals away.
+
+## When should I use this?
+
+If you want to handle syntax trees manually, use this. For an easier time processing content, use the [@glion/hl7v2](../hl7v2/) ecosystem instead.
+
+## Features
+
+- **Speed-first**: pull-based tokenizer, single pass, minimal object allocations.
+- **Spec-aware**: auto-detects delimiters from MSH-1 and MSH-2.
+- **AST design**: outputs a unist-compatible tree for use in the HL7v2 unified ecosystem.
+- **Composable**: works directly as a unified parser plugin or as a standalone function.
+- **Streaming-friendly**: pull-based design is ready for streaming use cases.
+
+## Install
+
+This package is [ESM only](https://gist.github.com/sindresorhus/a39789f98801d908bbc7ff3ecc99d99c). In Node.js (version 16+), install with npm:
+
+```bash
+npm install @glion/parser
+```
+
+## Use
+
+### Basic Usage
+
+```typescript
+import { unified } from "unified";
+import { hl7v2Parser } from "@glion/parser";
+
+const message = `MSH|^~\\&|SENDING_APP|SENDING_FAC|...`;
+
+const tree = unified().use(hl7v2Parser).parse(message);
+
+console.log(tree);
+```
+
+### Custom Delimiters
+
+You can specify custom delimiters for parsing non-standard HL7v2 messages. The `delimiters` option accepts a partial object, so you only need to specify the delimiters you want to override:
+
+```typescript
+import { unified } from "unified";
+import { hl7v2Parser } from "@glion/parser";
+
+// Override only the segment delimiter
+const tree = unified()
+  .use(hl7v2Parser, {
+    delimiters: {
+      segment: "\n", // Use newline instead of carriage return
+    },
+  })
+  .parse(message);
+
+// Override multiple delimiters
+const customTree = unified()
+  .use(hl7v2Parser, {
+    delimiters: {
+      field: "$",
+      component: "%",
+      segment: "\n",
+    },
+  })
+  .parse(customMessage);
+```
+
+The default delimiters are:
+
+- `field`: `|`
+- `component`: `^`
+- `repetition`: `~`
+- `subcomponent`: `&`
+- `escape`: `\`
+- `segment`: `\r`
+
+### Experimental Features
+
+The parser supports experimental features through the `experimental` option. These features are subject to change but provide opt-in access to upcoming behaviors.
+
+#### Empty Array Mode
+
+By default, the parser represents empty fields with full scaffolding (Field → FieldRepetition → Component → Subcomponent with `value: ""`). The `emptyMode: 'empty'` option changes this behavior to use empty children arrays instead, making the AST more compact and easier to work with.
+
+```typescript
+import { unified } from "unified";
+import { hl7v2Parser } from "@glion/parser";
+
+// With empty-array mode (new behavior)
+const tree = unified()
+  .use(hl7v2Parser, {
+    experimental: {
+      emptyMode: "empty-array",
+    },
+  })
+  .parse("PID|1||");
+
+// PID.2 (empty field) will have: { type: 'field', children: [] }
+// Instead of: Field → Rep → Comp → Sub with value: ""
+```
+
+**Spec Rules:**
+
+- **Leaf nodes** (Subcomponent, SegmentHeader) with no value: `value: ""`
+- **Parent nodes** (Field, FieldRepetition, Component) with no content: `children: []`
+- **Presence vs Value**:
+  - Node exists in parent's children array = position exists
+  - Node has empty children array = no value at that position
+
+**Examples:**
+
+| Wire Format     | Legacy Mode                                    | Empty-Array Mode                       |
+| --------------- | ---------------------------------------------- | -------------------------------------- |
+| `PID\|1\|\|`    | Field → Rep → Comp → Sub("")                   | Field(children: [])                    |
+| `PID\|1\|^\|`   | Field → Rep → [Comp → Sub(""), Comp → Sub("")] | Field → Rep → [Comp[], Comp[]]         |
+| `PID\|1\|~\|`   | Field → [Rep → Comp → Sub(""), Rep → ...]      | Field → [Rep[], Rep[]]                 |
+| `PID\|1\|ABC\|` | Field → Rep → Comp → Sub("ABC")                | Field → Rep → Comp → Sub("ABC") (same) |
+
+**Benefits:**
+
+- 37-63% reduction in node count for messages with empty fields
+- ~11% performance improvement for sparse messages
+- Cleaner AST structure for visitor patterns and transformations
+- Easier to distinguish "field exists but is empty" from "field has content"
+
+**Note:** This feature will become the default behavior in future versions. The legacy mode will be deprecated.
+
+## Contributing
+
+We welcome contributions! Please see our [Contributing Guide][github-contributing] for more details.
+
+1. Fork the repository
+2. Create your feature branch (`git checkout -b feature/amazing-feature`)
+3. Commit your changes (`git commit -m 'Add some amazing feature'`)
+4. Push to the branch (`git push origin feature/amazing-feature`)
+5. Open a Pull Request
+
+## Code of Conduct
+
+To ensure a welcoming and positive environment, we have a [Code of Conduct][github-code-of-conduct] that all contributors and participants are expected to adhere to.
+
+## License
+
+Copyright 2025 Rethink Health, SUARL. All rights reserved.
+
+This program is licensed to you under the terms of the [MIT License](https://opensource.org/licenses/MIT). This program is distributed WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the [LICENSE][github-license] file for details.
+
+[github-code-of-conduct]: https://github.com/rethinkhealth/glion/blob/main/CODE_OF_CONDUCT.md
+[github-license]: https://github.com/rethinkhealth/glion/blob/main/LICENSE
+[github-contributing]: https://github.com/rethinkhealth/glion/blob/main/CONTRIBUTING.md

package/dist/index.d.ts

@@ -0,0 +1,6 @@
+export { default as hl7v2Parser, parseHL7v2 } from "./parser";
+export * from "./preprocessor";
+export { parseHL7v2FromIterator } from "./processor";
+export * from "./tokenizer";
+export * from "./types";
+//# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,OAAO,IAAI,WAAW,EAAE,UAAU,EAAE,MAAM,UAAU,CAAC;AAC9D,cAAc,gBAAgB,CAAC;AAC/B,OAAO,EAAE,sBAAsB,EAAE,MAAM,aAAa,CAAC;AACrD,cAAc,aAAa,CAAC;AAC5B,cAAc,SAAS,CAAC"}

package/dist/index.js

@@ -0,0 +1,579 @@
+import { DEFAULT_DELIMITERS, isEmptyNode } from "@glion/utils";
+//#region src/preprocessor.ts
+/**
+* Step: strip UTF-8 BOM.
+*/
+const stripBOM = (ctx) => {
+	if (ctx.input.codePointAt(0) === 65279) ctx.input = ctx.input.slice(1);
+	return ctx;
+};
+/**
+* Step: normalize newlines to the default delimiters.
+*/
+const normalizeNewlines = (ctx) => {
+	ctx.input = ctx.input.replaceAll(/\r?\n/g, ctx.delimiters.segment);
+	return ctx;
+};
+/**
+* Step: auto-detect delimiters from MSH.
+*/
+const detectDelimiters = (ctx) => {
+	if (ctx.input.startsWith("MSH")) {
+		const field = ctx.input.charAt(3) || DEFAULT_DELIMITERS.field;
+		const enc = ctx.input.slice(4, 8);
+		const component = enc.charAt(0) || DEFAULT_DELIMITERS.component;
+		const repetition = enc.charAt(1) || DEFAULT_DELIMITERS.repetition;
+		const _escape = enc.charAt(2) || DEFAULT_DELIMITERS.escape;
+		const subcomponent = enc.charAt(3) || DEFAULT_DELIMITERS.subcomponent;
+		const { segment } = DEFAULT_DELIMITERS;
+		ctx.delimiters = {
+			component,
+			escape: _escape,
+			field,
+			repetition,
+			segment,
+			subcomponent
+		};
+	}
+	return ctx;
+};
+/**
+* Default preprocessing pipeline.
+*/
+const defaultPreprocessors = [
+	stripBOM,
+	normalizeNewlines,
+	detectDelimiters
+];
+/**
+* Run the pipeline to initialize ParserContext.
+*/
+function runPreprocessors(ctx, steps) {
+	for (const step of steps) ctx = step(ctx);
+	return ctx;
+}
+//#endregion
+//#region src/processor.ts
+/**
+* Check if a field has structural content beyond a single empty value.
+* A field with multiple repetitions, multiple components, or multiple
+* subcomponents has structural content even if all values are empty.
+* This is used to prevent dropping trailing fields like |^| or |&|
+* that have meaningful delimiter structure.
+*/
+function hasStructuralContent(field) {
+	if (field.children.length > 1) return true;
+	const rep = field.children[0];
+	if (rep && rep.children.length > 1) return true;
+	const comp = rep?.children[0];
+	return comp !== void 0 && comp.children.length > 1;
+}
+function createSubcomponent(start) {
+	return {
+		position: {
+			end: start,
+			start
+		},
+		type: "subcomponent",
+		value: ""
+	};
+}
+function createComponent(start, mode) {
+	return {
+		children: mode === "empty" ? [] : [createSubcomponent(start)],
+		position: {
+			end: start,
+			start
+		},
+		type: "component"
+	};
+}
+function createFieldRepetition(start, mode) {
+	return {
+		children: mode === "empty" ? [] : [createComponent(start, mode)],
+		position: {
+			end: start,
+			start
+		},
+		type: "field-repetition"
+	};
+}
+function createField(start, mode) {
+	return {
+		children: mode === "empty" ? [] : [createFieldRepetition(start, mode)],
+		position: {
+			end: start,
+			start
+		},
+		type: "field"
+	};
+}
+function createParserCore(ctx) {
+	const mode = ctx.emptyMode;
+	const root = {
+		children: [],
+		type: "root"
+	};
+	let seg = null;
+	let field = null;
+	let rep = null;
+	let comp = null;
+	let currentSub = null;
+	let segmentHasContent = false;
+	let lastContentEnd = null;
+	let documentEnd = null;
+	let expectingSegmentName = true;
+	let justSetSegmentName = false;
+	let rootStartSet = false;
+	const resetState = () => {
+		seg = null;
+		field = null;
+		rep = null;
+		comp = null;
+		currentSub = null;
+		segmentHasContent = false;
+		lastContentEnd = null;
+		expectingSegmentName = true;
+	};
+	const openSegment = (name, position) => {
+		seg = {
+			children: [],
+			name,
+			position: {
+				end: position.end,
+				start: position.start
+			},
+			type: "segment"
+		};
+		root.children.push(seg);
+		field = null;
+		rep = null;
+		comp = null;
+		currentSub = null;
+		segmentHasContent = false;
+		justSetSegmentName = true;
+		expectingSegmentName = false;
+	};
+	/**
+	* Ensure there is an active segment. If delimiter tokens arrive before
+	* any segment name, lazily open a segment with an empty name.
+	*/
+	const ensureSegment = (start) => {
+		if (!seg) {
+			openSegment("", {
+				start,
+				end: start
+			});
+			justSetSegmentName = false;
+		}
+	};
+	const openField = (start) => {
+		ensureSegment(start);
+		field = createField(start, mode);
+		seg.children.push(field);
+		if (mode === "empty") {
+			rep = null;
+			comp = null;
+			currentSub = null;
+		} else {
+			rep = field.children[0] ?? null;
+			comp = rep?.children[0] ?? null;
+			currentSub = comp?.children[0] ?? null;
+		}
+		segmentHasContent = true;
+	};
+	const openRepetition = (start) => {
+		if (!field) {
+			openField(start);
+			if (mode === "empty") {
+				const emptyRep = createFieldRepetition(start, mode);
+				field.children.push(emptyRep);
+			}
+		}
+		if (mode === "empty" && field.children.length === 0) {
+			const emptyRep = createFieldRepetition(start, mode);
+			field.children.push(emptyRep);
+		}
+		rep = createFieldRepetition(start, mode);
+		field.children.push(rep);
+		if (mode === "empty") {
+			comp = null;
+			currentSub = null;
+		} else {
+			comp = rep.children[0] ?? null;
+			currentSub = comp?.children[0] ?? null;
+		}
+		segmentHasContent = true;
+	};
+	const ensureComponent = (start) => {
+		if (!field) openField(start);
+		if (!rep) {
+			rep = createFieldRepetition(start, mode);
+			field.children.push(rep);
+		}
+		if (!comp) {
+			comp = createComponent(start, mode);
+			rep.children.push(comp);
+		}
+		segmentHasContent = true;
+	};
+	const openComponent = (start) => {
+		if (!field) {
+			openField(start);
+			if (mode === "empty") {
+				rep = createFieldRepetition(start, mode);
+				field.children.push(rep);
+				const emptyComp = createComponent(start, mode);
+				rep.children.push(emptyComp);
+			}
+		}
+		if (!rep) {
+			rep = createFieldRepetition(start, mode);
+			field.children.push(rep);
+			if (mode === "empty") {
+				const emptyComp = createComponent(start, mode);
+				rep.children.push(emptyComp);
+			}
+		} else if (mode === "empty" && rep.children.length === 0) {
+			const emptyComp = createComponent(start, mode);
+			rep.children.push(emptyComp);
+		}
+		comp = createComponent(start, mode);
+		rep.children.push(comp);
+		currentSub = mode === "empty" ? null : comp.children[0] ?? null;
+		segmentHasContent = true;
+	};
+	const ensureForText = (start) => {
+		if (!field) {
+			openField(start);
+			if (mode !== "empty") return;
+		}
+		if (!rep) {
+			if (mode !== "empty") {
+				openRepetition(start);
+				return;
+			}
+			rep = createFieldRepetition(start, mode);
+			field.children.push(rep);
+		}
+		if (!comp) {
+			if (mode !== "empty") {
+				openComponent(start);
+				return;
+			}
+			comp = createComponent(start, mode);
+			rep.children.push(comp);
+		}
+		if (!currentSub) {
+			currentSub = createSubcomponent(start);
+			comp.children.push(currentSub);
+			segmentHasContent = true;
+		}
+	};
+	const updatePositionsToEnd = (endPos) => {
+		if (currentSub?.position) currentSub.position.end = endPos;
+		if (comp?.position) comp.position.end = endPos;
+		if (rep?.position) rep.position.end = endPos;
+		if (field?.position) field.position.end = endPos;
+		if (seg?.position) seg.position.end = endPos;
+	};
+	const processToken = (tok) => {
+		switch (tok.type) {
+			case "SEGMENT_END":
+				updatePositionsToEnd(lastContentEnd || tok.position.start);
+				dropTrailingEmptyFieldIfPresent();
+				resetState();
+				return;
+			case "FIELD_DELIM":
+				lastContentEnd = tok.position.end;
+				documentEnd = tok.position.end;
+				if (justSetSegmentName) {
+					justSetSegmentName = false;
+					return;
+				}
+				if (!field) openField(tok.position.start);
+				openField(tok.position.end);
+				return;
+			case "REPETITION_DELIM":
+				lastContentEnd = tok.position.end;
+				documentEnd = tok.position.end;
+				if (!field) openField(tok.position.start);
+				openRepetition(tok.position.end);
+				return;
+			case "COMPONENT_DELIM":
+				lastContentEnd = tok.position.end;
+				documentEnd = tok.position.end;
+				openComponent(tok.position.end);
+				return;
+			case "SUBCOMP_DELIM":
+				lastContentEnd = tok.position.end;
+				documentEnd = tok.position.end;
+				ensureComponent(tok.position.start);
+				if (mode === "empty" && comp.children.length === 0) {
+					const emptySub = createSubcomponent(tok.position.start);
+					comp.children.push(emptySub);
+				}
+				currentSub = createSubcomponent(tok.position.end);
+				comp?.children.push(currentSub);
+				segmentHasContent = true;
+				return;
+			case "TEXT": {
+				const val = tok.value ?? "";
+				if (!rootStartSet) {
+					root.position = {
+						end: tok.position.start,
+						start: tok.position.start
+					};
+					rootStartSet = true;
+				}
+				if (expectingSegmentName) {
+					openSegment(val, tok.position);
+					lastContentEnd = tok.position.end;
+					documentEnd = tok.position.end;
+					return;
+				}
+				ensureForText(tok.position.start);
+				if (currentSub) currentSub.value += val;
+				updatePositionsToEnd(tok.position.end);
+				lastContentEnd = tok.position.end;
+				documentEnd = tok.position.end;
+				segmentHasContent = true;
+				return;
+			}
+			default: throw new Error(`Unexpected token type: ${tok.type}`);
+		}
+	};
+	const finalize = () => {
+		if (lastContentEnd && seg) updatePositionsToEnd(lastContentEnd);
+		dropTrailingEmptyFieldIfPresent();
+		if (seg && !segmentHasContent) root.children.pop();
+		if (root.position && documentEnd) root.position.end = documentEnd;
+		else if (!root.position) root.position = {
+			end: {
+				column: 1,
+				line: 1,
+				offset: 0
+			},
+			start: {
+				column: 1,
+				line: 1,
+				offset: 0
+			}
+		};
+		return root;
+	};
+	/**
+	* Drop only the phantom trailing empty field created by a terminal field
+	* delimiter (e.g. the last `|` in `PID|foo|`). Without this, every segment
+	* ending with `|` would gain an extra empty field on each parse cycle.
+	*/
+	function dropTrailingEmptyFieldIfPresent() {
+		if (!seg || seg.children.length === 0) return;
+		const lastChild = seg.children.at(-1);
+		if (lastChild?.type !== "field") return;
+		const lastField = lastChild;
+		if (isEmptyNode(lastField) && !hasStructuralContent(lastField)) seg.children.pop();
+	}
+	return {
+		finalize,
+		processToken,
+		root
+	};
+}
+function parseHL7v2FromIterator(tokens, ctx) {
+	const core = createParserCore(ctx);
+	for (const tok of tokens) core.processToken(tok);
+	return core.finalize();
+}
+//#endregion
+//#region src/tokenizer.ts
+const MSH_SEGMENT_START = 0;
+const MSH_SEGMENT_END = 3;
+const MSH_FIELD_SEPERATOR_START = 3;
+const MSH_FIELD_SEPERATOR_END = 4;
+const MSH_FIELD_DELIMITER_START = 4;
+const MSH_FIELD_DELIMITER_END = 8;
+var HL7v2Tokenizer = class {
+	input = "";
+	i = 0;
+	line = 1;
+	col = 1;
+	delims;
+	didMshBootstrap = false;
+	pendingBootstrap = null;
+	reset(ctx) {
+		this.input = ctx.input;
+		this.delims = ctx.delimiters;
+		this.i = 0;
+		this.line = 1;
+		this.col = 1;
+		this.didMshBootstrap = false;
+		this.pendingBootstrap = null;
+		if (this.input.startsWith("MSH")) {
+			const msh = this._slice(MSH_SEGMENT_START, MSH_SEGMENT_END);
+			const msh1 = this._slice(MSH_FIELD_SEPERATOR_START, MSH_FIELD_SEPERATOR_END);
+			const msh2 = this._slice(MSH_FIELD_DELIMITER_START, MSH_FIELD_DELIMITER_END);
+			this.pendingBootstrap = [
+				{
+					advance: msh.length,
+					kind: "TEXT",
+					value: msh
+				},
+				{
+					advance: 0,
+					kind: "FIELD_DELIM"
+				},
+				{
+					advance: msh1.length,
+					kind: "TEXT",
+					value: msh1
+				},
+				{
+					advance: 0,
+					kind: "FIELD_DELIM"
+				},
+				{
+					advance: msh2.length,
+					kind: "TEXT",
+					value: msh2
+				}
+			];
+		}
+	}
+	next() {
+		const s = this.input;
+		const n = s.length;
+		if (this.i >= n && !this.pendingBootstrap?.length) return null;
+		const start = {
+			column: this.col,
+			line: this.line,
+			offset: this.i
+		};
+		if (!this.didMshBootstrap && this.pendingBootstrap) {
+			const step = this.pendingBootstrap.shift();
+			if (step) {
+				if (step.kind === "TEXT") {
+					const take = Math.min(step.advance, n - this.i);
+					const out = step.value.slice(0, take);
+					this._fastAdvance(out);
+					if (this.pendingBootstrap.length === 0) {
+						this.pendingBootstrap = null;
+						this.didMshBootstrap = true;
+					}
+					return this._tok("TEXT", out, start);
+				}
+				this._advance(Math.min(step.advance, n - this.i));
+				if (this.pendingBootstrap.length === 0) {
+					this.pendingBootstrap = null;
+					this.didMshBootstrap = true;
+				}
+				return this._tok("FIELD_DELIM", void 0, start);
+			}
+		}
+		if (this.i >= n) return null;
+		const ch = s.charAt(this.i);
+		if (ch === this.delims.segment) {
+			this._advance(1, true);
+			return this._tok("SEGMENT_END", void 0, start);
+		}
+		if (ch === this.delims.field) {
+			this._advance(1);
+			return this._tok("FIELD_DELIM", void 0, start);
+		}
+		if (ch === this.delims.repetition) {
+			this._advance(1);
+			return this._tok("REPETITION_DELIM", void 0, start);
+		}
+		if (ch === this.delims.component) {
+			this._advance(1);
+			return this._tok("COMPONENT_DELIM", void 0, start);
+		}
+		if (ch === this.delims.subcomponent) {
+			this._advance(1);
+			return this._tok("SUBCOMP_DELIM", void 0, start);
+		}
+		let j = this.i;
+		const cmp = this.delims.component, fld = this.delims.field, rep = this.delims.repetition, seg = this.delims.segment, sub = this.delims.subcomponent;
+		while (j < s.length) {
+			const c = s.charAt(j);
+			if (c === seg || c === fld || c === rep || c === cmp || c === sub) break;
+			j += 1;
+		}
+		const val = s.slice(this.i, j);
+		this._fastAdvance(val);
+		return this._tok("TEXT", val, start);
+	}
+	_slice(start, end) {
+		return this.input.slice(start, Math.min(end, this.input.length));
+	}
+	_advance(n, isNewline = false) {
+		this.i += n;
+		if (isNewline) {
+			this.line += 1;
+			this.col = 1;
+		} else this.col += n;
+	}
+	_fastAdvance(chunk) {
+		const parts = chunk.split(this.delims.segment);
+		if (parts.length > 1) {
+			this.line += parts.length - 1;
+			this.col = (parts.at(-1)?.length ?? 0) + 1;
+		} else this.col += chunk.length;
+		this.i += chunk.length;
+	}
+	_tok(type, value, start) {
+		return {
+			position: {
+				end: {
+					column: this.col,
+					line: this.line,
+					offset: this.i
+				},
+				start
+			},
+			type,
+			value
+		};
+	}
+	[Symbol.iterator]() {
+		return { next: () => {
+			const t = this.next();
+			if (!t) return {
+				done: true,
+				value: void 0
+			};
+			return {
+				done: false,
+				value: t
+			};
+		} };
+	}
+};
+//#endregion
+//#region src/parser.ts
+function* iterateTokenizerSync(t) {
+	for (let tok = t.next(); tok; tok = t.next()) yield tok;
+}
+function parseHL7v2(input, opts = {}, settings) {
+	let ctx = {
+		delimiters: {
+			...DEFAULT_DELIMITERS,
+			...settings?.delimiters ?? {}
+		},
+		emptyMode: settings?.experimental?.emptyMode,
+		input
+	};
+	ctx = runPreprocessors(ctx, opts.preprocess || defaultPreprocessors);
+	const tokenizer = new HL7v2Tokenizer();
+	tokenizer.reset(ctx);
+	return parseHL7v2FromIterator(iterateTokenizerSync(tokenizer), ctx);
+}
+const hl7v2Parser = function hl7v2Parser(options = {}) {
+	this.parser = (document) => {
+		return parseHL7v2(document, options, this.data("settings"));
+	};
+};
+//#endregion
+export { HL7v2Tokenizer, defaultPreprocessors, detectDelimiters, hl7v2Parser, normalizeNewlines, parseHL7v2, parseHL7v2FromIterator, runPreprocessors, stripBOM };
+
+//# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.js","names":[],"sources":["../src/preprocessor.ts","../src/processor.ts","../src/tokenizer.ts","../src/parser.ts"],"sourcesContent":["import { DEFAULT_DELIMITERS } from \"@glion/utils\";\n\nimport type { ParserContext, PreprocessorStep } from \"./types\";\n\n// PreprocessorStep is declared in `types.ts` to avoid circular imports\n\n/**\n * Step: strip UTF-8 BOM.\n */\nexport const stripBOM: PreprocessorStep = (ctx) => {\n  if (ctx.input.codePointAt(0) === 0xfe_ff) {\n    ctx.input = ctx.input.slice(1);\n  }\n  return ctx;\n};\n\n/**\n * Step: normalize newlines to the default delimiters.\n */\nexport const normalizeNewlines: PreprocessorStep = (ctx) => {\n  ctx.input = ctx.input.replaceAll(/\\r?\\n/g, ctx.delimiters.segment);\n  return ctx;\n};\n\n/**\n * Step: auto-detect delimiters from MSH.\n */\nexport const detectDelimiters: PreprocessorStep = (ctx) => {\n  if (ctx.input.startsWith(\"MSH\")) {\n    const field = ctx.input.charAt(3) || DEFAULT_DELIMITERS.field;\n    const enc = ctx.input.slice(4, 8);\n    const component = enc.charAt(0) || DEFAULT_DELIMITERS.component;\n    const repetition = enc.charAt(1) || DEFAULT_DELIMITERS.repetition;\n    const _escape = enc.charAt(2) || DEFAULT_DELIMITERS.escape;\n    const subcomponent = enc.charAt(3) || DEFAULT_DELIMITERS.subcomponent;\n    const { segment } = DEFAULT_DELIMITERS;\n\n    ctx.delimiters = {\n      component,\n      escape: _escape,\n      field,\n      repetition,\n      segment,\n      subcomponent,\n    };\n  }\n  return ctx;\n};\n\n/**\n * Default preprocessing pipeline.\n */\nexport const defaultPreprocessors: PreprocessorStep[] = [\n  stripBOM,\n  normalizeNewlines,\n  detectDelimiters,\n];\n\n/**\n * Run the pipeline to initialize ParserContext.\n */\nexport function runPreprocessors(\n  ctx: ParserContext,\n  steps: PreprocessorStep[]\n): ParserContext {\n  for (const step of steps) {\n    // oxlint-disable-next-line no-param-reassign\n    ctx = step(ctx);\n  }\n\n  return ctx;\n}\n","import type {\n  Component,\n  Field,\n  FieldRepetition,\n  Root,\n  Segment,\n  Subcomponent,\n} from \"@glion/ast\";\nimport { isEmptyNode } from \"@glion/utils\";\nimport type { Position } from \"unist\";\n\nimport type { ParserContext, Token } from \"./types\";\n\n/**\n * Check if a field has structural content beyond a single empty value.\n * A field with multiple repetitions, multiple components, or multiple\n * subcomponents has structural content even if all values are empty.\n * This is used to prevent dropping trailing fields like |^| or |&|\n * that have meaningful delimiter structure.\n */\nfunction hasStructuralContent(field: Field): boolean {\n  if (field.children.length > 1) {\n    return true;\n  }\n  const rep = field.children[0];\n  if (rep && rep.children.length > 1) {\n    return true;\n  }\n  const comp = rep?.children[0];\n  return comp !== undefined && comp.children.length > 1;\n}\n\n// Helper to create an empty subcomponent at a given position\nfunction createSubcomponent(start: Position[\"start\"]): Subcomponent {\n  return {\n    position: { end: start, start },\n    type: \"subcomponent\",\n    value: \"\",\n  };\n}\n\n// Helper to create a component (with or without children based on mode)\nfunction createComponent(\n  start: Position[\"start\"],\n  mode: \"legacy\" | \"empty\" | undefined\n): Component {\n  return {\n    children: mode === \"empty\" ? [] : [createSubcomponent(start)],\n    position: { end: start, start },\n    type: \"component\",\n  };\n}\n\n// Helper to create a field repetition (with or without children based on mode)\nfunction createFieldRepetition(\n  start: Position[\"start\"],\n  mode: \"legacy\" | \"empty\" | undefined\n): FieldRepetition {\n  return {\n    children: mode === \"empty\" ? [] : [createComponent(start, mode)],\n    position: { end: start, start },\n    type: \"field-repetition\",\n  };\n}\n\n// Helper to create a field (with or without children based on mode)\nfunction createField(\n  start: Position[\"start\"],\n  mode: \"legacy\" | \"empty\" | undefined\n): Field {\n  return {\n    children: mode === \"empty\" ? [] : [createFieldRepetition(start, mode)],\n    position: { end: start, start },\n    type: \"field\",\n  };\n}\n\n// Shared core: process a single token into mutable parse state\nfunction createParserCore(ctx: ParserContext) {\n  const mode = ctx.emptyMode;\n  const root: Root = {\n    children: [],\n    type: \"root\",\n  };\n\n  let seg: Segment | null = null;\n  let field: Field | null = null;\n  let rep: FieldRepetition | null = null;\n  let comp: Component | null = null;\n  let currentSub: Subcomponent | null = null;\n  let segmentHasContent = false;\n  let lastContentEnd: Position[\"end\"] | null = null;\n  let documentEnd: Position[\"end\"] | null = null; // Track document-level end position\n  let expectingSegmentName = true; // Start expecting a segment name\n  let justSetSegmentName = false; // Track if we just set the segment name\n  let rootStartSet = false; // Track if we've set the root start position\n\n  const resetState = () => {\n    seg = null;\n    field = null;\n    rep = null;\n    comp = null;\n    currentSub = null;\n    segmentHasContent = false;\n    lastContentEnd = null;\n    expectingSegmentName = true;\n  };\n\n  const openSegment = (name: string, position: Position) => {\n    seg = {\n      children: [],\n      name,\n      position: { end: position.end, start: position.start },\n      type: \"segment\",\n    };\n    root.children.push(seg);\n    // Reset field-level state\n    field = null;\n    rep = null;\n    comp = null;\n    currentSub = null;\n    segmentHasContent = false;\n    justSetSegmentName = true;\n    expectingSegmentName = false;\n  };\n\n  /**\n   * Ensure there is an active segment. If delimiter tokens arrive before\n   * any segment name, lazily open a segment with an empty name.\n   */\n  const ensureSegment = (start: Position[\"start\"]) => {\n    if (!seg) {\n      openSegment(\"\", {\n        start,\n        end: start,\n      });\n      // Reset justSetSegmentName so the first FIELD_DELIM isn't skipped\n      // (there's no real segment name to separate from fields)\n      justSetSegmentName = false;\n    }\n  };\n\n  const openField = (start: Position[\"start\"]) => {\n    ensureSegment(start);\n    field = createField(start, mode);\n    seg!.children.push(field);\n    if (mode === \"empty\") {\n      rep = null;\n      comp = null;\n      currentSub = null;\n    } else {\n      rep = field.children[0] ?? null;\n      comp = rep?.children[0] ?? null;\n      currentSub = comp?.children[0] ?? null;\n    }\n    segmentHasContent = true;\n  };\n\n  const openRepetition = (start: Position[\"start\"]) => {\n    if (!field) {\n      openField(start);\n      // In empty-array mode, openField creates empty field\n      // We need to add an empty first repetition before the delimiter\n      if (mode === \"empty\") {\n        const emptyRep = createFieldRepetition(start, mode);\n        field!.children.push(emptyRep);\n      }\n      // Fall through to create the repetition after the delimiter\n    }\n    // In empty-array mode, if field has no children yet, add empty first rep\n    if (mode === \"empty\" && field!.children.length === 0) {\n      const emptyRep = createFieldRepetition(start, mode);\n      field!.children.push(emptyRep);\n    }\n    rep = createFieldRepetition(start, mode);\n    field!.children.push(rep);\n    if (mode === \"empty\") {\n      comp = null;\n      currentSub = null;\n    } else {\n      comp = rep.children[0] ?? null;\n      currentSub = comp?.children[0] ?? null;\n    }\n    segmentHasContent = true;\n  };\n\n  // Ensure there's a component to add subcomponents to (for SUBCOMP_DELIM)\n  // Does NOT add empty sibling components\n  const ensureComponent = (start: Position[\"start\"]) => {\n    if (!field) {\n      openField(start);\n    }\n    if (!rep) {\n      rep = createFieldRepetition(start, mode);\n      field!.children.push(rep);\n    }\n    if (!comp) {\n      comp = createComponent(start, mode);\n      rep.children.push(comp);\n    }\n    segmentHasContent = true;\n  };\n\n  // Handle COMPONENT_DELIM - adds empty component before delimiter if needed\n  const openComponent = (start: Position[\"start\"]) => {\n    if (!field) {\n      openField(start);\n      // In empty-array mode, openField creates empty field\n      // We need to create rep and add empty first component\n      if (mode === \"empty\") {\n        rep = createFieldRepetition(start, mode);\n        field!.children.push(rep);\n        const emptyComp = createComponent(start, mode);\n        rep.children.push(emptyComp);\n      }\n      // Fall through to create the component after the delimiter\n    }\n    if (!rep) {\n      rep = createFieldRepetition(start, mode);\n      field!.children.push(rep);\n      // In empty-array mode, need to add empty first component before the delimiter\n      if (mode === \"empty\") {\n        const emptyComp = createComponent(start, mode);\n        rep.children.push(emptyComp);\n      }\n    } else if (mode === \"empty\" && rep.children.length === 0) {\n      // If rep exists but has no children, add empty first component\n      const emptyComp = createComponent(start, mode);\n      rep.children.push(emptyComp);\n    }\n    comp = createComponent(start, mode);\n    rep.children.push(comp);\n    currentSub = mode === \"empty\" ? null : (comp.children[0] ?? null);\n    segmentHasContent = true;\n  };\n\n  const ensureForText = (start: Position[\"start\"]) => {\n    if (!field) {\n      openField(start);\n      if (mode !== \"empty\") {\n        return; // Everything is already set up by openField in legacy mode\n      }\n      // In empty-array mode, openField created empty field, need to build structure\n    }\n    if (!rep) {\n      if (mode !== \"empty\") {\n        openRepetition(start);\n        return; // Everything is already set up by openRepetition in legacy mode\n      }\n      // In empty-array mode, need to create repetition\n      rep = createFieldRepetition(start, mode);\n      field!.children.push(rep);\n    }\n    if (!comp) {\n      if (mode !== \"empty\") {\n        openComponent(start);\n        return; // Everything is already set up by openComponent in legacy mode\n      }\n      // In empty-array mode, need to create component\n      comp = createComponent(start, mode);\n      rep!.children.push(comp);\n    }\n    if (!currentSub) {\n      currentSub = createSubcomponent(start);\n      comp!.children.push(currentSub);\n      segmentHasContent = true;\n    }\n  };\n\n  const updatePositionsToEnd = (endPos: Position[\"end\"]) => {\n    if (currentSub?.position) {\n      currentSub.position.end = endPos;\n    }\n    if (comp?.position) {\n      comp.position.end = endPos;\n    }\n    if (rep?.position) {\n      rep.position.end = endPos;\n    }\n    if (field?.position) {\n      field.position.end = endPos;\n    }\n    if (seg?.position) {\n      seg.position.end = endPos;\n    }\n  };\n\n  const processToken = (tok: Token) => {\n    switch (tok.type) {\n      case \"SEGMENT_END\": {\n        // Use the last content position instead of the segment delimiter position\n        const endPos = lastContentEnd || tok.position.start;\n        updatePositionsToEnd(endPos);\n        dropTrailingEmptyFieldIfPresent();\n        resetState();\n        return;\n      }\n      case \"FIELD_DELIM\": {\n        lastContentEnd = tok.position.end;\n        documentEnd = tok.position.end;\n        // Skip the first field delimiter after setting segment name (it separates segment name from fields)\n        if (justSetSegmentName) {\n          justSetSegmentName = false;\n          return;\n        }\n        // Leading field delimiter implies an empty first field\n        if (!field) {\n          // Open an empty first field (already has empty subcomponent from openField)\n          openField(tok.position.start);\n        }\n        openField(tok.position.end);\n        return;\n      }\n      case \"REPETITION_DELIM\": {\n        lastContentEnd = tok.position.end;\n        documentEnd = tok.position.end;\n        if (!field) {\n          openField(tok.position.start);\n        }\n        openRepetition(tok.position.end);\n        return;\n      }\n      case \"COMPONENT_DELIM\": {\n        lastContentEnd = tok.position.end;\n        documentEnd = tok.position.end;\n        openComponent(tok.position.end);\n        return;\n      }\n      case \"SUBCOMP_DELIM\": {\n        lastContentEnd = tok.position.end;\n        documentEnd = tok.position.end;\n        // Ensure there's a component to add subcomponents to\n        ensureComponent(tok.position.start);\n        // In empty-array mode, if comp has no children yet, add empty first subcomponent\n        if (mode === \"empty\" && comp!.children.length === 0) {\n          const emptySub = createSubcomponent(tok.position.start);\n          comp!.children.push(emptySub);\n        }\n        currentSub = createSubcomponent(tok.position.end);\n        comp?.children.push(currentSub);\n        segmentHasContent = true;\n        return;\n      }\n      case \"TEXT\": {\n        const val = tok.value ?? \"\";\n\n        // Set root start position on first token\n        if (!rootStartSet) {\n          root.position = {\n            end: tok.position.start,\n            start: tok.position.start,\n          };\n          rootStartSet = true;\n        }\n\n        // If we're expecting a segment name, this TEXT is the segment name\n        if (expectingSegmentName) {\n          // `tok.position` is always available for TEXT tokens\n          openSegment(val, tok.position);\n          lastContentEnd = tok.position.end;\n          documentEnd = tok.position.end;\n          return;\n        }\n\n        // Otherwise, it's regular field content\n        ensureForText(tok.position.start);\n        if (currentSub) {\n          currentSub.value += val;\n        }\n        updatePositionsToEnd(tok.position.end);\n        lastContentEnd = tok.position.end;\n        documentEnd = tok.position.end;\n        segmentHasContent = true;\n        return;\n      }\n      default: {\n        throw new Error(`Unexpected token type: ${tok.type}`);\n      }\n    }\n  };\n\n  // Do not pre-open a segment; lazily open upon first non-SEGMENT_END token.\n\n  const finalize = () => {\n    // Handle input that ends without an explicit segment delimiter as above.\n    if (lastContentEnd && seg) {\n      updatePositionsToEnd(lastContentEnd);\n    }\n    dropTrailingEmptyFieldIfPresent();\n    if (seg && !segmentHasContent) {\n      // Drop trailing empty segment if no content was added\n      root.children.pop();\n    }\n\n    // Set root end position once at finalization using document-level end\n    if (root.position && documentEnd) {\n      root.position.end = documentEnd;\n    } else if (!root.position) {\n      // Empty document - set default position\n      root.position = {\n        end: { column: 1, line: 1, offset: 0 },\n        start: { column: 1, line: 1, offset: 0 },\n      };\n    }\n\n    return root;\n  };\n\n  /**\n   * Drop only the phantom trailing empty field created by a terminal field\n   * delimiter (e.g. the last `|` in `PID|foo|`). Without this, every segment\n   * ending with `|` would gain an extra empty field on each parse cycle.\n   */\n  function dropTrailingEmptyFieldIfPresent() {\n    if (!seg || seg.children.length === 0) {\n      return;\n    }\n    const lastChild = seg.children.at(-1);\n    if (lastChild?.type !== \"field\") {\n      return;\n    }\n    const lastField = lastChild as Field;\n    // Both conditions are needed: a field like `|^|` is isEmpty (no leaf data)\n    // but has structural content (component separators) and must be preserved.\n    if (isEmptyNode(lastField) && !hasStructuralContent(lastField)) {\n      seg.children.pop();\n    }\n  }\n\n  return { finalize, processToken, root };\n}\n\n// Sync convenience wrapper over a sync Iterable token source\nexport function parseHL7v2FromIterator(\n  tokens: Iterable<Token>,\n  ctx: ParserContext\n): Root {\n  const core = createParserCore(ctx);\n  for (const tok of tokens) {\n    core.processToken(tok);\n  }\n  return core.finalize();\n}\n","// src/tokenizer.ts\nimport type { Delimiters } from \"@glion/ast\";\nimport type { Position } from \"unist\";\n\nimport type { ParserContext, Token, Tokenizer, TokenType } from \"./types\";\n\nconst MSH_SEGMENT_START = 0;\nconst MSH_SEGMENT_END = 3;\nconst MSH_FIELD_SEPERATOR_START = 3;\nconst MSH_FIELD_SEPERATOR_END = 4;\nconst MSH_FIELD_DELIMITER_START = 4;\nconst MSH_FIELD_DELIMITER_END = 8;\n\nexport class HL7v2Tokenizer implements Tokenizer, Iterable<Token> {\n  private input = \"\";\n  private i = 0;\n  private line = 1;\n  private col = 1;\n  private delims!: Delimiters;\n\n  // Only-run-once MSH bootstrap at the start of the file\n  private didMshBootstrap = false;\n  private pendingBootstrap: null | Array<\n    | { kind: \"TEXT\"; value: string; advance: number }\n    | { kind: \"FIELD_DELIM\"; advance: number }\n  > = null; // queue to emit TEXT('MSH'), FIELD_DELIM, TEXT('^~\\\\&')\n\n  reset(ctx: ParserContext) {\n    this.input = ctx.input;\n    this.delims = ctx.delimiters;\n    this.i = 0;\n    this.line = 1;\n    this.col = 1;\n    this.didMshBootstrap = false;\n    this.pendingBootstrap = null;\n\n    // Prepare a one-time bootstrap if file starts with MSH\n    if (this.input.startsWith(\"MSH\")) {\n      // Precompute MSH and MSH.2; MSH.1 is the field delimiter char at index 3\n      const msh = this._slice(MSH_SEGMENT_START, MSH_SEGMENT_END);\n      const msh1 = this._slice(\n        MSH_FIELD_SEPERATOR_START,\n        MSH_FIELD_SEPERATOR_END\n      ); // the field delimiter char at index 3\n      const msh2 = this._slice(\n        MSH_FIELD_DELIMITER_START,\n        MSH_FIELD_DELIMITER_END\n      ); // may be shorter than 4 if truncated\n      this.pendingBootstrap = [\n        { advance: msh.length, kind: \"TEXT\", value: msh },\n        { advance: 0, kind: \"FIELD_DELIM\" },\n        { advance: msh1.length, kind: \"TEXT\", value: msh1 },\n        { advance: 0, kind: \"FIELD_DELIM\" },\n        { advance: msh2.length, kind: \"TEXT\", value: msh2 },\n      ];\n      // NOTE: we have not advanced indices yet; we will as we emit tokens\n    }\n  }\n\n  // oxlint-disable-next-line complexity\n  next(): Token | null {\n    const s = this.input;\n    const n = s.length;\n    if (this.i >= n && !this.pendingBootstrap?.length) {\n      return null;\n    }\n\n    const start = { column: this.col, line: this.line, offset: this.i };\n\n    // ---- One-time MSH bootstrap at file start ----\n    if (!this.didMshBootstrap && this.pendingBootstrap) {\n      const step = this.pendingBootstrap.shift();\n      if (step) {\n        if (step.kind === \"TEXT\") {\n          const take = Math.min(step.advance, n - this.i);\n          const out = step.value.slice(0, take);\n          this._fastAdvance(out);\n          if (this.pendingBootstrap.length === 0) {\n            this.pendingBootstrap = null;\n            this.didMshBootstrap = true;\n          }\n          return this._tok(\"TEXT\", out, start);\n        }\n        // FIELD_DELIM: advance exactly one char (the delimiter) and emit a FIELD_DELIM token\n        this._advance(Math.min(step.advance, n - this.i));\n        if (this.pendingBootstrap.length === 0) {\n          this.pendingBootstrap = null;\n          this.didMshBootstrap = true;\n        }\n        return this._tok(\"FIELD_DELIM\", undefined, start);\n      }\n    }\n\n    // ---- Normal tokenization (delimiters + text) ----\n    if (this.i >= n) {\n      return null;\n    }\n    const ch = s.charAt(this.i);\n\n    if (ch === this.delims.segment) {\n      this._advance(1, true);\n      return this._tok(\"SEGMENT_END\", undefined, start);\n    }\n    if (ch === this.delims.field) {\n      this._advance(1);\n      return this._tok(\"FIELD_DELIM\", undefined, start);\n    }\n    if (ch === this.delims.repetition) {\n      this._advance(1);\n      return this._tok(\"REPETITION_DELIM\", undefined, start);\n    }\n    if (ch === this.delims.component) {\n      this._advance(1);\n      return this._tok(\"COMPONENT_DELIM\", undefined, start);\n    }\n    if (ch === this.delims.subcomponent) {\n      this._advance(1);\n      return this._tok(\"SUBCOMP_DELIM\", undefined, start);\n    }\n\n    // TEXT until next delimiter or end\n    let j = this.i;\n    const cmp = this.delims.component,\n      fld = this.delims.field,\n      rep = this.delims.repetition,\n      seg = this.delims.segment,\n      sub = this.delims.subcomponent;\n\n    while (j < s.length) {\n      const c = s.charAt(j);\n      if (c === seg || c === fld || c === rep || c === cmp || c === sub) {\n        break;\n      }\n      j += 1;\n    }\n\n    const val = s.slice(this.i, j);\n    this._fastAdvance(val);\n    return this._tok(\"TEXT\", val, start);\n  }\n\n  // ---- helpers ----\n  private _slice(start: number, end: number): string {\n    return this.input.slice(start, Math.min(end, this.input.length));\n  }\n\n  private _advance(n: number, isNewline = false) {\n    this.i += n;\n    if (isNewline) {\n      this.line += 1;\n      this.col = 1;\n    } else {\n      this.col += n;\n    }\n  }\n\n  private _fastAdvance(chunk: string) {\n    const parts = chunk.split(this.delims.segment);\n    if (parts.length > 1) {\n      this.line += parts.length - 1;\n      this.col = (parts.at(-1)?.length ?? 0) + 1;\n    } else {\n      this.col += chunk.length;\n    }\n    this.i += chunk.length;\n  }\n\n  private _tok(\n    type: TokenType,\n    value: string | undefined,\n    start: Position[\"start\"]\n  ): Token {\n    const end = { column: this.col, line: this.line, offset: this.i };\n    return { position: { end, start }, type, value };\n  }\n\n  // Iterable protocol (sync)\n  [Symbol.iterator](): Iterator<Token> {\n    return {\n      next: () => {\n        const t = this.next();\n        if (!t) {\n          return { done: true, value: undefined as unknown as Token };\n        }\n        return { done: false, value: t };\n      },\n    };\n  }\n}\n","import type { Root } from \"@glion/ast\";\nimport type { HL7v2Settings } from \"@glion/config\";\nimport { DEFAULT_DELIMITERS } from \"@glion/utils\";\nimport type { Plugin, Processor } from \"unified\";\n\nimport { defaultPreprocessors, runPreprocessors } from \"./preprocessor\";\nimport { parseHL7v2FromIterator } from \"./processor\";\nimport { HL7v2Tokenizer } from \"./tokenizer\";\nimport type { ParseOptions, ParserContext, Token, Tokenizer } from \"./types\";\n\nfunction* iterateTokenizerSync(t: Tokenizer): Iterable<Token> {\n  for (let tok = t.next(); tok; tok = t.next()) {\n    yield tok;\n  }\n}\n\nexport function parseHL7v2(\n  input: string,\n  opts: ParseOptions = {},\n  settings?: HL7v2Settings\n): Root {\n  // Merge settings: defaults < provided settings\n  // Note: MSH auto-detection (via detectDelimiters preprocessor) runs after this\n  // and will override delimiters if the message starts with MSH\n\n  let ctx: ParserContext = {\n    delimiters: {\n      ...DEFAULT_DELIMITERS,\n      ...(settings?.delimiters ?? {}),\n    },\n    emptyMode: settings?.experimental?.emptyMode,\n    input,\n  };\n  // Run preprocessing\n  ctx = runPreprocessors(ctx, opts.preprocess || defaultPreprocessors);\n  // Run tokenizer\n  const tokenizer = new HL7v2Tokenizer();\n  // Reset tokenizer.\n  tokenizer.reset(ctx);\n  // Parse\n  return parseHL7v2FromIterator(iterateTokenizerSync(tokenizer), ctx);\n}\n\nconst hl7v2Parser: Plugin<[ParseOptions?], string, Root> = function hl7v2Parser(\n  this: Processor,\n  options = {}\n) {\n  this.parser = (document: string) => {\n    // Get settings from processor (populated by unified-args from config)\n    const settings = this.data(\"settings\");\n\n    return parseHL7v2(document, options, settings);\n  };\n};\n\nexport default hl7v2Parser;\n"],"mappings":";;;;;AASA,MAAa,YAA8B,QAAQ;AACjD,KAAI,IAAI,MAAM,YAAY,EAAE,KAAK,MAC/B,KAAI,QAAQ,IAAI,MAAM,MAAM,EAAE;AAEhC,QAAO;;;;;AAMT,MAAa,qBAAuC,QAAQ;AAC1D,KAAI,QAAQ,IAAI,MAAM,WAAW,UAAU,IAAI,WAAW,QAAQ;AAClE,QAAO;;;;;AAMT,MAAa,oBAAsC,QAAQ;AACzD,KAAI,IAAI,MAAM,WAAW,MAAM,EAAE;EAC/B,MAAM,QAAQ,IAAI,MAAM,OAAO,EAAE,IAAI,mBAAmB;EACxD,MAAM,MAAM,IAAI,MAAM,MAAM,GAAG,EAAE;EACjC,MAAM,YAAY,IAAI,OAAO,EAAE,IAAI,mBAAmB;EACtD,MAAM,aAAa,IAAI,OAAO,EAAE,IAAI,mBAAmB;EACvD,MAAM,UAAU,IAAI,OAAO,EAAE,IAAI,mBAAmB;EACpD,MAAM,eAAe,IAAI,OAAO,EAAE,IAAI,mBAAmB;EACzD,MAAM,EAAE,YAAY;AAEpB,MAAI,aAAa;GACf;GACA,QAAQ;GACR;GACA;GACA;GACA;GACD;;AAEH,QAAO;;;;;AAMT,MAAa,uBAA2C;CACtD;CACA;CACA;CACD;;;;AAKD,SAAgB,iBACd,KACA,OACe;AACf,MAAK,MAAM,QAAQ,MAEjB,OAAM,KAAK,IAAI;AAGjB,QAAO;;;;;;;;;;;AClDT,SAAS,qBAAqB,OAAuB;AACnD,KAAI,MAAM,SAAS,SAAS,EAC1B,QAAO;CAET,MAAM,MAAM,MAAM,SAAS;AAC3B,KAAI,OAAO,IAAI,SAAS,SAAS,EAC/B,QAAO;CAET,MAAM,OAAO,KAAK,SAAS;AAC3B,QAAO,SAAS,KAAA,KAAa,KAAK,SAAS,SAAS;;AAItD,SAAS,mBAAmB,OAAwC;AAClE,QAAO;EACL,UAAU;GAAE,KAAK;GAAO;GAAO;EAC/B,MAAM;EACN,OAAO;EACR;;AAIH,SAAS,gBACP,OACA,MACW;AACX,QAAO;EACL,UAAU,SAAS,UAAU,EAAE,GAAG,CAAC,mBAAmB,MAAM,CAAC;EAC7D,UAAU;GAAE,KAAK;GAAO;GAAO;EAC/B,MAAM;EACP;;AAIH,SAAS,sBACP,OACA,MACiB;AACjB,QAAO;EACL,UAAU,SAAS,UAAU,EAAE,GAAG,CAAC,gBAAgB,OAAO,KAAK,CAAC;EAChE,UAAU;GAAE,KAAK;GAAO;GAAO;EAC/B,MAAM;EACP;;AAIH,SAAS,YACP,OACA,MACO;AACP,QAAO;EACL,UAAU,SAAS,UAAU,EAAE,GAAG,CAAC,sBAAsB,OAAO,KAAK,CAAC;EACtE,UAAU;GAAE,KAAK;GAAO;GAAO;EAC/B,MAAM;EACP;;AAIH,SAAS,iBAAiB,KAAoB;CAC5C,MAAM,OAAO,IAAI;CACjB,MAAM,OAAa;EACjB,UAAU,EAAE;EACZ,MAAM;EACP;CAED,IAAI,MAAsB;CAC1B,IAAI,QAAsB;CAC1B,IAAI,MAA8B;CAClC,IAAI,OAAyB;CAC7B,IAAI,aAAkC;CACtC,IAAI,oBAAoB;CACxB,IAAI,iBAAyC;CAC7C,IAAI,cAAsC;CAC1C,IAAI,uBAAuB;CAC3B,IAAI,qBAAqB;CACzB,IAAI,eAAe;CAEnB,MAAM,mBAAmB;AACvB,QAAM;AACN,UAAQ;AACR,QAAM;AACN,SAAO;AACP,eAAa;AACb,sBAAoB;AACpB,mBAAiB;AACjB,yBAAuB;;CAGzB,MAAM,eAAe,MAAc,aAAuB;AACxD,QAAM;GACJ,UAAU,EAAE;GACZ;GACA,UAAU;IAAE,KAAK,SAAS;IAAK,OAAO,SAAS;IAAO;GACtD,MAAM;GACP;AACD,OAAK,SAAS,KAAK,IAAI;AAEvB,UAAQ;AACR,QAAM;AACN,SAAO;AACP,eAAa;AACb,sBAAoB;AACpB,uBAAqB;AACrB,yBAAuB;;;;;;CAOzB,MAAM,iBAAiB,UAA6B;AAClD,MAAI,CAAC,KAAK;AACR,eAAY,IAAI;IACd;IACA,KAAK;IACN,CAAC;AAGF,wBAAqB;;;CAIzB,MAAM,aAAa,UAA6B;AAC9C,gBAAc,MAAM;AACpB,UAAQ,YAAY,OAAO,KAAK;AAChC,MAAK,SAAS,KAAK,MAAM;AACzB,MAAI,SAAS,SAAS;AACpB,SAAM;AACN,UAAO;AACP,gBAAa;SACR;AACL,SAAM,MAAM,SAAS,MAAM;AAC3B,UAAO,KAAK,SAAS,MAAM;AAC3B,gBAAa,MAAM,SAAS,MAAM;;AAEpC,sBAAoB;;CAGtB,MAAM,kBAAkB,UAA6B;AACnD,MAAI,CAAC,OAAO;AACV,aAAU,MAAM;AAGhB,OAAI,SAAS,SAAS;IACpB,MAAM,WAAW,sBAAsB,OAAO,KAAK;AACnD,UAAO,SAAS,KAAK,SAAS;;;AAKlC,MAAI,SAAS,WAAW,MAAO,SAAS,WAAW,GAAG;GACpD,MAAM,WAAW,sBAAsB,OAAO,KAAK;AACnD,SAAO,SAAS,KAAK,SAAS;;AAEhC,QAAM,sBAAsB,OAAO,KAAK;AACxC,QAAO,SAAS,KAAK,IAAI;AACzB,MAAI,SAAS,SAAS;AACpB,UAAO;AACP,gBAAa;SACR;AACL,UAAO,IAAI,SAAS,MAAM;AAC1B,gBAAa,MAAM,SAAS,MAAM;;AAEpC,sBAAoB;;CAKtB,MAAM,mBAAmB,UAA6B;AACpD,MAAI,CAAC,MACH,WAAU,MAAM;AAElB,MAAI,CAAC,KAAK;AACR,SAAM,sBAAsB,OAAO,KAAK;AACxC,SAAO,SAAS,KAAK,IAAI;;AAE3B,MAAI,CAAC,MAAM;AACT,UAAO,gBAAgB,OAAO,KAAK;AACnC,OAAI,SAAS,KAAK,KAAK;;AAEzB,sBAAoB;;CAItB,MAAM,iBAAiB,UAA6B;AAClD,MAAI,CAAC,OAAO;AACV,aAAU,MAAM;AAGhB,OAAI,SAAS,SAAS;AACpB,UAAM,sBAAsB,OAAO,KAAK;AACxC,UAAO,SAAS,KAAK,IAAI;IACzB,MAAM,YAAY,gBAAgB,OAAO,KAAK;AAC9C,QAAI,SAAS,KAAK,UAAU;;;AAIhC,MAAI,CAAC,KAAK;AACR,SAAM,sBAAsB,OAAO,KAAK;AACxC,SAAO,SAAS,KAAK,IAAI;AAEzB,OAAI,SAAS,SAAS;IACpB,MAAM,YAAY,gBAAgB,OAAO,KAAK;AAC9C,QAAI,SAAS,KAAK,UAAU;;aAErB,SAAS,WAAW,IAAI,SAAS,WAAW,GAAG;GAExD,MAAM,YAAY,gBAAgB,OAAO,KAAK;AAC9C,OAAI,SAAS,KAAK,UAAU;;AAE9B,SAAO,gBAAgB,OAAO,KAAK;AACnC,MAAI,SAAS,KAAK,KAAK;AACvB,eAAa,SAAS,UAAU,OAAQ,KAAK,SAAS,MAAM;AAC5D,sBAAoB;;CAGtB,MAAM,iBAAiB,UAA6B;AAClD,MAAI,CAAC,OAAO;AACV,aAAU,MAAM;AAChB,OAAI,SAAS,QACX;;AAIJ,MAAI,CAAC,KAAK;AACR,OAAI,SAAS,SAAS;AACpB,mBAAe,MAAM;AACrB;;AAGF,SAAM,sBAAsB,OAAO,KAAK;AACxC,SAAO,SAAS,KAAK,IAAI;;AAE3B,MAAI,CAAC,MAAM;AACT,OAAI,SAAS,SAAS;AACpB,kBAAc,MAAM;AACpB;;AAGF,UAAO,gBAAgB,OAAO,KAAK;AACnC,OAAK,SAAS,KAAK,KAAK;;AAE1B,MAAI,CAAC,YAAY;AACf,gBAAa,mBAAmB,MAAM;AACtC,QAAM,SAAS,KAAK,WAAW;AAC/B,uBAAoB;;;CAIxB,MAAM,wBAAwB,WAA4B;AACxD,MAAI,YAAY,SACd,YAAW,SAAS,MAAM;AAE5B,MAAI,MAAM,SACR,MAAK,SAAS,MAAM;AAEtB,MAAI,KAAK,SACP,KAAI,SAAS,MAAM;AAErB,MAAI,OAAO,SACT,OAAM,SAAS,MAAM;AAEvB,MAAI,KAAK,SACP,KAAI,SAAS,MAAM;;CAIvB,MAAM,gBAAgB,QAAe;AACnC,UAAQ,IAAI,MAAZ;GACE,KAAK;AAGH,yBADe,kBAAkB,IAAI,SAAS,MAClB;AAC5B,qCAAiC;AACjC,gBAAY;AACZ;GAEF,KAAK;AACH,qBAAiB,IAAI,SAAS;AAC9B,kBAAc,IAAI,SAAS;AAE3B,QAAI,oBAAoB;AACtB,0BAAqB;AACrB;;AAGF,QAAI,CAAC,MAEH,WAAU,IAAI,SAAS,MAAM;AAE/B,cAAU,IAAI,SAAS,IAAI;AAC3B;GAEF,KAAK;AACH,qBAAiB,IAAI,SAAS;AAC9B,kBAAc,IAAI,SAAS;AAC3B,QAAI,CAAC,MACH,WAAU,IAAI,SAAS,MAAM;AAE/B,mBAAe,IAAI,SAAS,IAAI;AAChC;GAEF,KAAK;AACH,qBAAiB,IAAI,SAAS;AAC9B,kBAAc,IAAI,SAAS;AAC3B,kBAAc,IAAI,SAAS,IAAI;AAC/B;GAEF,KAAK;AACH,qBAAiB,IAAI,SAAS;AAC9B,kBAAc,IAAI,SAAS;AAE3B,oBAAgB,IAAI,SAAS,MAAM;AAEnC,QAAI,SAAS,WAAW,KAAM,SAAS,WAAW,GAAG;KACnD,MAAM,WAAW,mBAAmB,IAAI,SAAS,MAAM;AACvD,UAAM,SAAS,KAAK,SAAS;;AAE/B,iBAAa,mBAAmB,IAAI,SAAS,IAAI;AACjD,UAAM,SAAS,KAAK,WAAW;AAC/B,wBAAoB;AACpB;GAEF,KAAK,QAAQ;IACX,MAAM,MAAM,IAAI,SAAS;AAGzB,QAAI,CAAC,cAAc;AACjB,UAAK,WAAW;MACd,KAAK,IAAI,SAAS;MAClB,OAAO,IAAI,SAAS;MACrB;AACD,oBAAe;;AAIjB,QAAI,sBAAsB;AAExB,iBAAY,KAAK,IAAI,SAAS;AAC9B,sBAAiB,IAAI,SAAS;AAC9B,mBAAc,IAAI,SAAS;AAC3B;;AAIF,kBAAc,IAAI,SAAS,MAAM;AACjC,QAAI,WACF,YAAW,SAAS;AAEtB,yBAAqB,IAAI,SAAS,IAAI;AACtC,qBAAiB,IAAI,SAAS;AAC9B,kBAAc,IAAI,SAAS;AAC3B,wBAAoB;AACpB;;GAEF,QACE,OAAM,IAAI,MAAM,0BAA0B,IAAI,OAAO;;;CAO3D,MAAM,iBAAiB;AAErB,MAAI,kBAAkB,IACpB,sBAAqB,eAAe;AAEtC,mCAAiC;AACjC,MAAI,OAAO,CAAC,kBAEV,MAAK,SAAS,KAAK;AAIrB,MAAI,KAAK,YAAY,YACnB,MAAK,SAAS,MAAM;WACX,CAAC,KAAK,SAEf,MAAK,WAAW;GACd,KAAK;IAAE,QAAQ;IAAG,MAAM;IAAG,QAAQ;IAAG;GACtC,OAAO;IAAE,QAAQ;IAAG,MAAM;IAAG,QAAQ;IAAG;GACzC;AAGH,SAAO;;;;;;;CAQT,SAAS,kCAAkC;AACzC,MAAI,CAAC,OAAO,IAAI,SAAS,WAAW,EAClC;EAEF,MAAM,YAAY,IAAI,SAAS,GAAG,GAAG;AACrC,MAAI,WAAW,SAAS,QACtB;EAEF,MAAM,YAAY;AAGlB,MAAI,YAAY,UAAU,IAAI,CAAC,qBAAqB,UAAU,CAC5D,KAAI,SAAS,KAAK;;AAItB,QAAO;EAAE;EAAU;EAAc;EAAM;;AAIzC,SAAgB,uBACd,QACA,KACM;CACN,MAAM,OAAO,iBAAiB,IAAI;AAClC,MAAK,MAAM,OAAO,OAChB,MAAK,aAAa,IAAI;AAExB,QAAO,KAAK,UAAU;;;;ACnbxB,MAAM,oBAAoB;AAC1B,MAAM,kBAAkB;AACxB,MAAM,4BAA4B;AAClC,MAAM,0BAA0B;AAChC,MAAM,4BAA4B;AAClC,MAAM,0BAA0B;AAEhC,IAAa,iBAAb,MAAkE;CAChE,QAAgB;CAChB,IAAY;CACZ,OAAe;CACf,MAAc;CACd;CAGA,kBAA0B;CAC1B,mBAGI;CAEJ,MAAM,KAAoB;AACxB,OAAK,QAAQ,IAAI;AACjB,OAAK,SAAS,IAAI;AAClB,OAAK,IAAI;AACT,OAAK,OAAO;AACZ,OAAK,MAAM;AACX,OAAK,kBAAkB;AACvB,OAAK,mBAAmB;AAGxB,MAAI,KAAK,MAAM,WAAW,MAAM,EAAE;GAEhC,MAAM,MAAM,KAAK,OAAO,mBAAmB,gBAAgB;GAC3D,MAAM,OAAO,KAAK,OAChB,2BACA,wBACD;GACD,MAAM,OAAO,KAAK,OAChB,2BACA,wBACD;AACD,QAAK,mBAAmB;IACtB;KAAE,SAAS,IAAI;KAAQ,MAAM;KAAQ,OAAO;KAAK;IACjD;KAAE,SAAS;KAAG,MAAM;KAAe;IACnC;KAAE,SAAS,KAAK;KAAQ,MAAM;KAAQ,OAAO;KAAM;IACnD;KAAE,SAAS;KAAG,MAAM;KAAe;IACnC;KAAE,SAAS,KAAK;KAAQ,MAAM;KAAQ,OAAO;KAAM;IACpD;;;CAML,OAAqB;EACnB,MAAM,IAAI,KAAK;EACf,MAAM,IAAI,EAAE;AACZ,MAAI,KAAK,KAAK,KAAK,CAAC,KAAK,kBAAkB,OACzC,QAAO;EAGT,MAAM,QAAQ;GAAE,QAAQ,KAAK;GAAK,MAAM,KAAK;GAAM,QAAQ,KAAK;GAAG;AAGnE,MAAI,CAAC,KAAK,mBAAmB,KAAK,kBAAkB;GAClD,MAAM,OAAO,KAAK,iBAAiB,OAAO;AAC1C,OAAI,MAAM;AACR,QAAI,KAAK,SAAS,QAAQ;KACxB,MAAM,OAAO,KAAK,IAAI,KAAK,SAAS,IAAI,KAAK,EAAE;KAC/C,MAAM,MAAM,KAAK,MAAM,MAAM,GAAG,KAAK;AACrC,UAAK,aAAa,IAAI;AACtB,SAAI,KAAK,iBAAiB,WAAW,GAAG;AACtC,WAAK,mBAAmB;AACxB,WAAK,kBAAkB;;AAEzB,YAAO,KAAK,KAAK,QAAQ,KAAK,MAAM;;AAGtC,SAAK,SAAS,KAAK,IAAI,KAAK,SAAS,IAAI,KAAK,EAAE,CAAC;AACjD,QAAI,KAAK,iBAAiB,WAAW,GAAG;AACtC,UAAK,mBAAmB;AACxB,UAAK,kBAAkB;;AAEzB,WAAO,KAAK,KAAK,eAAe,KAAA,GAAW,MAAM;;;AAKrD,MAAI,KAAK,KAAK,EACZ,QAAO;EAET,MAAM,KAAK,EAAE,OAAO,KAAK,EAAE;AAE3B,MAAI,OAAO,KAAK,OAAO,SAAS;AAC9B,QAAK,SAAS,GAAG,KAAK;AACtB,UAAO,KAAK,KAAK,eAAe,KAAA,GAAW,MAAM;;AAEnD,MAAI,OAAO,KAAK,OAAO,OAAO;AAC5B,QAAK,SAAS,EAAE;AAChB,UAAO,KAAK,KAAK,eAAe,KAAA,GAAW,MAAM;;AAEnD,MAAI,OAAO,KAAK,OAAO,YAAY;AACjC,QAAK,SAAS,EAAE;AAChB,UAAO,KAAK,KAAK,oBAAoB,KAAA,GAAW,MAAM;;AAExD,MAAI,OAAO,KAAK,OAAO,WAAW;AAChC,QAAK,SAAS,EAAE;AAChB,UAAO,KAAK,KAAK,mBAAmB,KAAA,GAAW,MAAM;;AAEvD,MAAI,OAAO,KAAK,OAAO,cAAc;AACnC,QAAK,SAAS,EAAE;AAChB,UAAO,KAAK,KAAK,iBAAiB,KAAA,GAAW,MAAM;;EAIrD,IAAI,IAAI,KAAK;EACb,MAAM,MAAM,KAAK,OAAO,WACtB,MAAM,KAAK,OAAO,OAClB,MAAM,KAAK,OAAO,YAClB,MAAM,KAAK,OAAO,SAClB,MAAM,KAAK,OAAO;AAEpB,SAAO,IAAI,EAAE,QAAQ;GACnB,MAAM,IAAI,EAAE,OAAO,EAAE;AACrB,OAAI,MAAM,OAAO,MAAM,OAAO,MAAM,OAAO,MAAM,OAAO,MAAM,IAC5D;AAEF,QAAK;;EAGP,MAAM,MAAM,EAAE,MAAM,KAAK,GAAG,EAAE;AAC9B,OAAK,aAAa,IAAI;AACtB,SAAO,KAAK,KAAK,QAAQ,KAAK,MAAM;;CAItC,OAAe,OAAe,KAAqB;AACjD,SAAO,KAAK,MAAM,MAAM,OAAO,KAAK,IAAI,KAAK,KAAK,MAAM,OAAO,CAAC;;CAGlE,SAAiB,GAAW,YAAY,OAAO;AAC7C,OAAK,KAAK;AACV,MAAI,WAAW;AACb,QAAK,QAAQ;AACb,QAAK,MAAM;QAEX,MAAK,OAAO;;CAIhB,aAAqB,OAAe;EAClC,MAAM,QAAQ,MAAM,MAAM,KAAK,OAAO,QAAQ;AAC9C,MAAI,MAAM,SAAS,GAAG;AACpB,QAAK,QAAQ,MAAM,SAAS;AAC5B,QAAK,OAAO,MAAM,GAAG,GAAG,EAAE,UAAU,KAAK;QAEzC,MAAK,OAAO,MAAM;AAEpB,OAAK,KAAK,MAAM;;CAGlB,KACE,MACA,OACA,OACO;AAEP,SAAO;GAAE,UAAU;IAAE,KADT;KAAE,QAAQ,KAAK;KAAK,MAAM,KAAK;KAAM,QAAQ,KAAK;KAAG;IACvC;IAAO;GAAE;GAAM;GAAO;;CAIlD,CAAC,OAAO,YAA6B;AACnC,SAAO,EACL,YAAY;GACV,MAAM,IAAI,KAAK,MAAM;AACrB,OAAI,CAAC,EACH,QAAO;IAAE,MAAM;IAAM,OAAO,KAAA;IAA+B;AAE7D,UAAO;IAAE,MAAM;IAAO,OAAO;IAAG;KAEnC;;;;;AChLL,UAAU,qBAAqB,GAA+B;AAC5D,MAAK,IAAI,MAAM,EAAE,MAAM,EAAE,KAAK,MAAM,EAAE,MAAM,CAC1C,OAAM;;AAIV,SAAgB,WACd,OACA,OAAqB,EAAE,EACvB,UACM;CAKN,IAAI,MAAqB;EACvB,YAAY;GACV,GAAG;GACH,GAAI,UAAU,cAAc,EAAE;GAC/B;EACD,WAAW,UAAU,cAAc;EACnC;EACD;AAED,OAAM,iBAAiB,KAAK,KAAK,cAAc,qBAAqB;CAEpE,MAAM,YAAY,IAAI,gBAAgB;AAEtC,WAAU,MAAM,IAAI;AAEpB,QAAO,uBAAuB,qBAAqB,UAAU,EAAE,IAAI;;AAGrE,MAAM,cAAqD,SAAS,YAElE,UAAU,EAAE,EACZ;AACA,MAAK,UAAU,aAAqB;AAIlC,SAAO,WAAW,UAAU,SAFX,KAAK,KAAK,WAAW,CAEQ"}

package/dist/parser.d.ts

@@ -0,0 +1,8 @@
+import type { Root } from "@glion/ast";
+import type { HL7v2Settings } from "@glion/config";
+import type { Plugin } from "unified";
+import type { ParseOptions } from "./types";
+export declare function parseHL7v2(input: string, opts?: ParseOptions, settings?: HL7v2Settings): Root;
+declare const hl7v2Parser: Plugin<[ParseOptions?], string, Root>;
+export default hl7v2Parser;
+//# sourceMappingURL=parser.d.ts.map

package/dist/parser.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"parser.d.ts","sourceRoot":"","sources":["../src/parser.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,IAAI,EAAE,MAAM,YAAY,CAAC;AACvC,OAAO,KAAK,EAAE,aAAa,EAAE,MAAM,eAAe,CAAC;AAEnD,OAAO,KAAK,EAAE,MAAM,EAAa,MAAM,SAAS,CAAC;AAKjD,OAAO,KAAK,EAAE,YAAY,EAAmC,MAAM,SAAS,CAAC;AAQ7E,wBAAgB,UAAU,CACxB,KAAK,EAAE,MAAM,EACb,IAAI,GAAE,YAAiB,EACvB,QAAQ,CAAC,EAAE,aAAa,GACvB,IAAI,CAqBN;AAED,QAAA,MAAM,WAAW,EAAE,MAAM,CAAC,CAAC,YAAY,CAAC,CAAC,EAAE,MAAM,EAAE,IAAI,CAUtD,CAAC;AAEF,eAAe,WAAW,CAAC"}

package/dist/preprocessor.d.ts

@@ -0,0 +1,22 @@
+import type { ParserContext, PreprocessorStep } from "./types";
+/**
+ * Step: strip UTF-8 BOM.
+ */
+export declare const stripBOM: PreprocessorStep;
+/**
+ * Step: normalize newlines to the default delimiters.
+ */
+export declare const normalizeNewlines: PreprocessorStep;
+/**
+ * Step: auto-detect delimiters from MSH.
+ */
+export declare const detectDelimiters: PreprocessorStep;
+/**
+ * Default preprocessing pipeline.
+ */
+export declare const defaultPreprocessors: PreprocessorStep[];
+/**
+ * Run the pipeline to initialize ParserContext.
+ */
+export declare function runPreprocessors(ctx: ParserContext, steps: PreprocessorStep[]): ParserContext;
+//# sourceMappingURL=preprocessor.d.ts.map

package/dist/preprocessor.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"preprocessor.d.ts","sourceRoot":"","sources":["../src/preprocessor.ts"],"names":[],"mappings":"AAEA,OAAO,KAAK,EAAE,aAAa,EAAE,gBAAgB,EAAE,MAAM,SAAS,CAAC;AAI/D;;GAEG;AACH,eAAO,MAAM,QAAQ,EAAE,gBAKtB,CAAC;AAEF;;GAEG;AACH,eAAO,MAAM,iBAAiB,EAAE,gBAG/B,CAAC;AAEF;;GAEG;AACH,eAAO,MAAM,gBAAgB,EAAE,gBAoB9B,CAAC;AAEF;;GAEG;AACH,eAAO,MAAM,oBAAoB,EAAE,gBAAgB,EAIlD,CAAC;AAEF;;GAEG;AACH,wBAAgB,gBAAgB,CAC9B,GAAG,EAAE,aAAa,EAClB,KAAK,EAAE,gBAAgB,EAAE,GACxB,aAAa,CAOf"}

package/dist/processor.d.ts

@@ -0,0 +1,4 @@
+import type { Root } from "@glion/ast";
+import type { ParserContext, Token } from "./types";
+export declare function parseHL7v2FromIterator(tokens: Iterable<Token>, ctx: ParserContext): Root;
+//# sourceMappingURL=processor.d.ts.map

package/dist/processor.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"processor.d.ts","sourceRoot":"","sources":["../src/processor.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAIV,IAAI,EAGL,MAAM,YAAY,CAAC;AAIpB,OAAO,KAAK,EAAE,aAAa,EAAE,KAAK,EAAE,MAAM,SAAS,CAAC;AAsapD,wBAAgB,sBAAsB,CACpC,MAAM,EAAE,QAAQ,CAAC,KAAK,CAAC,EACvB,GAAG,EAAE,aAAa,GACjB,IAAI,CAMN"}

package/dist/tokenizer.d.ts

@@ -0,0 +1,18 @@
+import type { ParserContext, Token, Tokenizer } from "./types";
+export declare class HL7v2Tokenizer implements Tokenizer, Iterable<Token> {
+    private input;
+    private i;
+    private line;
+    private col;
+    private delims;
+    private didMshBootstrap;
+    private pendingBootstrap;
+    reset(ctx: ParserContext): void;
+    next(): Token | null;
+    private _slice;
+    private _advance;
+    private _fastAdvance;
+    private _tok;
+    [Symbol.iterator](): Iterator<Token>;
+}
+//# sourceMappingURL=tokenizer.d.ts.map

package/dist/tokenizer.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"tokenizer.d.ts","sourceRoot":"","sources":["../src/tokenizer.ts"],"names":[],"mappings":"AAIA,OAAO,KAAK,EAAE,aAAa,EAAE,KAAK,EAAE,SAAS,EAAa,MAAM,SAAS,CAAC;AAS1E,qBAAa,cAAe,YAAW,SAAS,EAAE,QAAQ,CAAC,KAAK,CAAC;IAC/D,OAAO,CAAC,KAAK,CAAM;IACnB,OAAO,CAAC,CAAC,CAAK;IACd,OAAO,CAAC,IAAI,CAAK;IACjB,OAAO,CAAC,GAAG,CAAK;IAChB,OAAO,CAAC,MAAM,CAAc;IAG5B,OAAO,CAAC,eAAe,CAAS;IAChC,OAAO,CAAC,gBAAgB,CAGf;IAET,KAAK,CAAC,GAAG,EAAE,aAAa;IAiCxB,IAAI,IAAI,KAAK,GAAG,IAAI;IAkFpB,OAAO,CAAC,MAAM;IAId,OAAO,CAAC,QAAQ;IAUhB,OAAO,CAAC,YAAY;IAWpB,OAAO,CAAC,IAAI;IAUZ,CAAC,MAAM,CAAC,QAAQ,CAAC,IAAI,QAAQ,CAAC,KAAK,CAAC;CAWrC"}

package/dist/types.d.ts

@@ -0,0 +1,26 @@
+import type { Delimiters } from "@glion/ast";
+import type { Position } from "unist";
+export type PreprocessorStep = (ctx: ParserContext) => ParserContext;
+export interface ParseOptions {
+    /**
+     * Optional preprocessing steps to apply to the input before parsing.
+     */
+    preprocess?: PreprocessorStep[];
+}
+export interface ParserContext {
+    input: string;
+    delimiters: Delimiters;
+    metadata?: Record<string, unknown>;
+    emptyMode?: "legacy" | "empty";
+}
+export interface Token {
+    type: TokenType;
+    value?: string;
+    position: Position;
+}
+export type TokenType = "SEGMENT_END" | "FIELD_DELIM" | "REPETITION_DELIM" | "COMPONENT_DELIM" | "SUBCOMP_DELIM" | "TEXT";
+export interface Tokenizer {
+    reset(ctx: ParserContext): void;
+    next(): Token | null;
+}
+//# sourceMappingURL=types.d.ts.map

package/dist/types.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"types.d.ts","sourceRoot":"","sources":["../src/types.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,YAAY,CAAC;AAC7C,OAAO,KAAK,EAAE,QAAQ,EAAE,MAAM,OAAO,CAAC;AAItC,MAAM,MAAM,gBAAgB,GAAG,CAAC,GAAG,EAAE,aAAa,KAAK,aAAa,CAAC;AAErE,MAAM,WAAW,YAAY;IAC3B;;OAEG;IACH,UAAU,CAAC,EAAE,gBAAgB,EAAE,CAAC;CACjC;AAED,MAAM,WAAW,aAAa;IAC5B,KAAK,EAAE,MAAM,CAAC;IACd,UAAU,EAAE,UAAU,CAAC;IACvB,QAAQ,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC;IACnC,SAAS,CAAC,EAAE,QAAQ,GAAG,OAAO,CAAC;CAChC;AAGD,MAAM,WAAW,KAAK;IACpB,IAAI,EAAE,SAAS,CAAC;IAChB,KAAK,CAAC,EAAE,MAAM,CAAC;IACf,QAAQ,EAAE,QAAQ,CAAC;CACpB;AAGD,MAAM,MAAM,SAAS,GACjB,aAAa,GACb,aAAa,GACb,kBAAkB,GAClB,iBAAiB,GACjB,eAAe,GACf,MAAM,CAAC;AAEX,MAAM,WAAW,SAAS;IACxB,KAAK,CAAC,GAAG,EAAE,aAAa,GAAG,IAAI,CAAC;IAChC,IAAI,IAAI,KAAK,GAAG,IAAI,CAAC;CACtB"}

package/package.json

@@ -0,0 +1,59 @@
+{
+  "name": "@glion/parser",
+  "version": "0.14.1",
+  "description": "hl7v2 plugin to parse hl7v2 messages",
+  "keywords": [
+    "health",
+    "healthcare",
+    "hl7",
+    "hl7v2",
+    "nodejs",
+    "typescript"
+  ],
+  "homepage": "https://glion.dev",
+  "license": "MIT",
+  "author": {
+    "name": "Melek Somai",
+    "email": "melek@rethinkhealth.io"
+  },
+  "repository": "rethinkhealth/glion.git",
+  "files": [
+    "dist"
+  ],
+  "type": "module",
+  "types": "./dist/index.d.ts",
+  "exports": {
+    ".": "./dist/index.js"
+  },
+  "publishConfig": {
+    "access": "public"
+  },
+  "dependencies": {
+    "unified": "11.0.5",
+    "@glion/utils": "0.14.1"
+  },
+  "devDependencies": {
+    "@types/node": "^25.5.0",
+    "@types/unist": "^3.0.3",
+    "@vitest/coverage-v8": "4.1.2",
+    "tsdown": "0.21.7",
+    "typescript": "^5.9.3",
+    "unist": "^0.0.1",
+    "vitest": "4.1.2",
+    "@glion/testing": "0.0.2",
+    "@glion/ast": "0.14.1",
+    "@glion/config": "0.14.1",
+    "@glion/tsconfig": "0.0.1"
+  },
+  "engines": {
+    "node": ">=18"
+  },
+  "scripts": {
+    "bench": "vitest bench --run",
+    "build": "tsdown && tsc --emitDeclarationOnly",
+    "check-types": "tsc --noEmit",
+    "test": "vitest run",
+    "test:coverage": "vitest run --coverage",
+    "test:watch": "vitest"
+  }
+}