@fllf/agent-sdk 0.1.1 → 0.1.3
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +169 -0
- package/README.md +844 -184
- package/dist/agent/Agent.d.ts +16 -5
- package/dist/agent/Agent.d.ts.map +1 -1
- package/dist/agent/Agent.js +283 -2
- package/dist/agent/Agent.js.map +1 -1
- package/dist/agent/index.d.ts +1 -1
- package/dist/agent/index.d.ts.map +1 -1
- package/dist/agent/types.d.ts +27 -3
- package/dist/agent/types.d.ts.map +1 -1
- package/dist/config/config.d.ts +22 -3
- package/dist/config/config.d.ts.map +1 -1
- package/dist/config/config.js +194 -47
- package/dist/config/config.js.map +1 -1
- package/dist/config/index.d.ts +1 -1
- package/dist/config/index.d.ts.map +1 -1
- package/dist/errors.d.ts +74 -0
- package/dist/errors.d.ts.map +1 -0
- package/dist/errors.js +186 -0
- package/dist/errors.js.map +1 -0
- package/dist/executors/base-executor.d.ts +14 -0
- package/dist/executors/base-executor.d.ts.map +1 -0
- package/dist/executors/base-executor.js +31 -0
- package/dist/executors/base-executor.js.map +1 -0
- package/dist/executors/base.d.ts +36 -5
- package/dist/executors/base.d.ts.map +1 -1
- package/dist/executors/chat-request-builder.d.ts +10 -0
- package/dist/executors/chat-request-builder.d.ts.map +1 -0
- package/dist/executors/chat-request-builder.js +96 -0
- package/dist/executors/chat-request-builder.js.map +1 -0
- package/dist/executors/index.d.ts +4 -1
- package/dist/executors/index.d.ts.map +1 -1
- package/dist/executors/index.js +6 -1
- package/dist/executors/index.js.map +1 -1
- package/dist/executors/rag-executor.js +1 -1
- package/dist/executors/rag-executor.js.map +1 -1
- package/dist/executors/simple-chat-executor.d.ts +4 -2
- package/dist/executors/simple-chat-executor.d.ts.map +1 -1
- package/dist/executors/simple-chat-executor.js +59 -57
- package/dist/executors/simple-chat-executor.js.map +1 -1
- package/dist/executors/tool-calling-executor.d.ts +20 -2
- package/dist/executors/tool-calling-executor.d.ts.map +1 -1
- package/dist/executors/tool-calling-executor.js +189 -91
- package/dist/executors/tool-calling-executor.js.map +1 -1
- package/dist/history/base.d.ts +55 -2
- package/dist/history/base.d.ts.map +1 -1
- package/dist/history/base.js +49 -0
- package/dist/history/base.js.map +1 -1
- package/dist/history/compression.d.ts +49 -0
- package/dist/history/compression.d.ts.map +1 -0
- package/dist/history/compression.js +53 -0
- package/dist/history/compression.js.map +1 -0
- package/dist/history/context-window.d.ts +33 -0
- package/dist/history/context-window.d.ts.map +1 -0
- package/dist/history/context-window.js +68 -0
- package/dist/history/context-window.js.map +1 -0
- package/dist/history/in-memory.d.ts +6 -4
- package/dist/history/in-memory.d.ts.map +1 -1
- package/dist/history/in-memory.js +25 -39
- package/dist/history/in-memory.js.map +1 -1
- package/dist/history/index.d.ts +9 -2
- package/dist/history/index.d.ts.map +1 -1
- package/dist/history/index.js +18 -1
- package/dist/history/index.js.map +1 -1
- package/dist/history/postgres/index.d.ts +4 -0
- package/dist/history/postgres/index.d.ts.map +1 -0
- package/dist/history/postgres/index.js +20 -0
- package/dist/history/postgres/index.js.map +1 -0
- package/dist/history/postgres/postgres-message-history-store.d.ts +35 -0
- package/dist/history/postgres/postgres-message-history-store.d.ts.map +1 -0
- package/dist/history/postgres/postgres-message-history-store.js +195 -0
- package/dist/history/postgres/postgres-message-history-store.js.map +1 -0
- package/dist/history/postgres/schema.d.ts +18 -0
- package/dist/history/postgres/schema.d.ts.map +1 -0
- package/dist/history/postgres/schema.js +46 -0
- package/dist/history/postgres/schema.js.map +1 -0
- package/dist/history/postgres/sql.d.ts +29 -0
- package/dist/history/postgres/sql.d.ts.map +1 -0
- package/dist/history/postgres/sql.js +102 -0
- package/dist/history/postgres/sql.js.map +1 -0
- package/dist/history/postgres/types.d.ts +20 -0
- package/dist/history/postgres/types.d.ts.map +1 -0
- package/dist/history/postgres/types.js +3 -0
- package/dist/history/postgres/types.js.map +1 -0
- package/dist/history/tool-pairing.d.ts +11 -0
- package/dist/history/tool-pairing.d.ts.map +1 -0
- package/dist/history/tool-pairing.js +52 -0
- package/dist/history/tool-pairing.js.map +1 -0
- package/dist/index.d.ts +2 -0
- package/dist/index.d.ts.map +1 -1
- package/dist/index.js +2 -0
- package/dist/index.js.map +1 -1
- package/dist/llm/LLM.d.ts +1 -0
- package/dist/llm/LLM.d.ts.map +1 -1
- package/dist/llm/LLM.js +136 -26
- package/dist/llm/LLM.js.map +1 -1
- package/dist/llm/errors.d.ts +11 -1
- package/dist/llm/errors.d.ts.map +1 -1
- package/dist/llm/errors.js +50 -1
- package/dist/llm/errors.js.map +1 -1
- package/dist/llm/factory.d.ts +14 -2
- package/dist/llm/factory.d.ts.map +1 -1
- package/dist/llm/factory.js +48 -9
- package/dist/llm/factory.js.map +1 -1
- package/dist/llm/index.d.ts +8 -3
- package/dist/llm/index.d.ts.map +1 -1
- package/dist/llm/index.js +10 -1
- package/dist/llm/index.js.map +1 -1
- package/dist/llm/providers/anthropic.d.ts +14 -0
- package/dist/llm/providers/anthropic.d.ts.map +1 -0
- package/dist/llm/providers/anthropic.js +336 -0
- package/dist/llm/providers/anthropic.js.map +1 -0
- package/dist/llm/providers/openai-compatible.d.ts +7 -0
- package/dist/llm/providers/openai-compatible.d.ts.map +1 -1
- package/dist/llm/providers/openai-compatible.js +151 -47
- package/dist/llm/providers/openai-compatible.js.map +1 -1
- package/dist/llm/stream-accumulator.d.ts +17 -0
- package/dist/llm/stream-accumulator.d.ts.map +1 -0
- package/dist/llm/stream-accumulator.js +83 -0
- package/dist/llm/stream-accumulator.js.map +1 -0
- package/dist/llm/tool-arguments.d.ts +6 -0
- package/dist/llm/tool-arguments.d.ts.map +1 -0
- package/dist/llm/tool-arguments.js +20 -0
- package/dist/llm/tool-arguments.js.map +1 -0
- package/dist/llm/types.d.ts +76 -4
- package/dist/llm/types.d.ts.map +1 -1
- package/dist/messages/content.d.ts +36 -0
- package/dist/messages/content.d.ts.map +1 -0
- package/dist/messages/content.js +70 -0
- package/dist/messages/content.js.map +1 -0
- package/dist/messages/index.d.ts +4 -2
- package/dist/messages/index.d.ts.map +1 -1
- package/dist/messages/index.js +10 -1
- package/dist/messages/index.js.map +1 -1
- package/dist/messages/message.d.ts +11 -9
- package/dist/messages/message.d.ts.map +1 -1
- package/dist/messages/message.js +69 -4
- package/dist/messages/message.js.map +1 -1
- package/dist/messages/types.d.ts +12 -0
- package/dist/messages/types.d.ts.map +1 -1
- package/dist/observability/collecting-observer.d.ts +8 -0
- package/dist/observability/collecting-observer.d.ts.map +1 -0
- package/dist/observability/collecting-observer.js +15 -0
- package/dist/observability/collecting-observer.js.map +1 -0
- package/dist/observability/index.d.ts +1 -0
- package/dist/observability/index.d.ts.map +1 -1
- package/dist/observability/index.js +3 -1
- package/dist/observability/index.js.map +1 -1
- package/dist/observability/observer.d.ts.map +1 -1
- package/dist/observability/observer.js +14 -1
- package/dist/observability/observer.js.map +1 -1
- package/dist/observability/types.d.ts +16 -1
- package/dist/observability/types.d.ts.map +1 -1
- package/dist/rag/chunking/auto-chunker.d.ts +7 -0
- package/dist/rag/chunking/auto-chunker.d.ts.map +1 -1
- package/dist/rag/chunking/auto-chunker.js +13 -0
- package/dist/rag/chunking/auto-chunker.js.map +1 -1
- package/dist/rag/chunking/chunker.d.ts.map +1 -1
- package/dist/rag/chunking/chunker.js +29 -5
- package/dist/rag/chunking/chunker.js.map +1 -1
- package/dist/rag/chunking/index.d.ts +2 -0
- package/dist/rag/chunking/index.d.ts.map +1 -1
- package/dist/rag/chunking/index.js +2 -0
- package/dist/rag/chunking/index.js.map +1 -1
- package/dist/rag/chunking/markdown-chunker.d.ts.map +1 -1
- package/dist/rag/chunking/markdown-chunker.js.map +1 -1
- package/dist/rag/chunking/qa-pair-chunker.d.ts +23 -0
- package/dist/rag/chunking/qa-pair-chunker.d.ts.map +1 -0
- package/dist/rag/chunking/qa-pair-chunker.js +162 -0
- package/dist/rag/chunking/qa-pair-chunker.js.map +1 -0
- package/dist/rag/chunking/semantic-chunker.d.ts +19 -0
- package/dist/rag/chunking/semantic-chunker.d.ts.map +1 -0
- package/dist/rag/chunking/semantic-chunker.js +291 -0
- package/dist/rag/chunking/semantic-chunker.js.map +1 -0
- package/dist/rag/embeddings/embedder.d.ts.map +1 -1
- package/dist/rag/embeddings/embedder.js +6 -0
- package/dist/rag/embeddings/embedder.js.map +1 -1
- package/dist/rag/embeddings/openai-compatible-embedder.d.ts +22 -4
- package/dist/rag/embeddings/openai-compatible-embedder.d.ts.map +1 -1
- package/dist/rag/embeddings/openai-compatible-embedder.js +108 -14
- package/dist/rag/embeddings/openai-compatible-embedder.js.map +1 -1
- package/dist/rag/generation/context-builder.d.ts +7 -0
- package/dist/rag/generation/context-builder.d.ts.map +1 -1
- package/dist/rag/generation/context-builder.js +4 -1
- package/dist/rag/generation/context-builder.js.map +1 -1
- package/dist/rag/ingestion/metadata.d.ts +6 -1
- package/dist/rag/ingestion/metadata.d.ts.map +1 -1
- package/dist/rag/ingestion/metadata.js +6 -2
- package/dist/rag/ingestion/metadata.js.map +1 -1
- package/dist/rag/pipeline.d.ts.map +1 -1
- package/dist/rag/pipeline.js +34 -11
- package/dist/rag/pipeline.js.map +1 -1
- package/dist/rag/retrieval/index.d.ts +1 -0
- package/dist/rag/retrieval/index.d.ts.map +1 -1
- package/dist/rag/retrieval/index.js +1 -0
- package/dist/rag/retrieval/index.js.map +1 -1
- package/dist/rag/retrieval/parent-child-expanding-retriever.d.ts +31 -0
- package/dist/rag/retrieval/parent-child-expanding-retriever.d.ts.map +1 -0
- package/dist/rag/retrieval/parent-child-expanding-retriever.js +194 -0
- package/dist/rag/retrieval/parent-child-expanding-retriever.js.map +1 -0
- package/dist/rag/stores/in-memory-keyword-store.d.ts.map +1 -1
- package/dist/rag/stores/in-memory-keyword-store.js +3 -8
- package/dist/rag/stores/in-memory-keyword-store.js.map +1 -1
- package/dist/rag/stores/keyword-tokenizer.d.ts +4 -0
- package/dist/rag/stores/keyword-tokenizer.d.ts.map +1 -0
- package/dist/rag/stores/keyword-tokenizer.js +113 -0
- package/dist/rag/stores/keyword-tokenizer.js.map +1 -0
- package/dist/rag/stores/postgres/pg-vector-store.d.ts.map +1 -1
- package/dist/rag/stores/postgres/pg-vector-store.js +21 -8
- package/dist/rag/stores/postgres/pg-vector-store.js.map +1 -1
- package/dist/rag/stores/postgres/postgres-document-store.d.ts.map +1 -1
- package/dist/rag/stores/postgres/postgres-document-store.js +26 -13
- package/dist/rag/stores/postgres/postgres-document-store.js.map +1 -1
- package/dist/rag/stores/postgres/postgres-keyword-store.d.ts.map +1 -1
- package/dist/rag/stores/postgres/postgres-keyword-store.js +43 -43
- package/dist/rag/stores/postgres/postgres-keyword-store.js.map +1 -1
- package/dist/rag/stores/postgres/sql.d.ts +22 -0
- package/dist/rag/stores/postgres/sql.d.ts.map +1 -1
- package/dist/rag/stores/postgres/sql.js +42 -0
- package/dist/rag/stores/postgres/sql.js.map +1 -1
- package/dist/rag/stores/types.d.ts +5 -0
- package/dist/rag/stores/types.d.ts.map +1 -1
- package/dist/rag/types.d.ts +6 -0
- package/dist/rag/types.d.ts.map +1 -1
- package/dist/testing/agent.d.ts +11 -0
- package/dist/testing/agent.d.ts.map +1 -0
- package/dist/testing/agent.js +45 -0
- package/dist/testing/agent.js.map +1 -0
- package/dist/testing/history-contract.d.ts +12 -0
- package/dist/testing/history-contract.d.ts.map +1 -0
- package/dist/testing/history-contract.js +111 -0
- package/dist/testing/history-contract.js.map +1 -0
- package/dist/testing/index.d.ts +11 -0
- package/dist/testing/index.d.ts.map +1 -0
- package/dist/testing/index.js +16 -0
- package/dist/testing/index.js.map +1 -0
- package/dist/testing/models.d.ts +34 -0
- package/dist/testing/models.d.ts.map +1 -0
- package/dist/testing/models.js +74 -0
- package/dist/testing/models.js.map +1 -0
- package/dist/testing/tool-schema.d.ts +6 -0
- package/dist/testing/tool-schema.d.ts.map +1 -0
- package/dist/testing/tool-schema.js +29 -0
- package/dist/testing/tool-schema.js.map +1 -0
- package/dist/testing/tools.d.ts +21 -0
- package/dist/testing/tools.d.ts.map +1 -0
- package/dist/testing/tools.js +43 -0
- package/dist/testing/tools.js.map +1 -0
- package/dist/tools/base.d.ts +38 -7
- package/dist/tools/base.d.ts.map +1 -1
- package/dist/tools/base.js +238 -11
- package/dist/tools/base.js.map +1 -1
- package/dist/tools/builtin/advancedSearchTool.d.ts.map +1 -1
- package/dist/tools/builtin/advancedSearchTool.js +30 -4
- package/dist/tools/builtin/advancedSearchTool.js.map +1 -1
- package/dist/tools/builtin/ragSearchTool.d.ts +6 -3
- package/dist/tools/builtin/ragSearchTool.d.ts.map +1 -1
- package/dist/tools/builtin/ragSearchTool.js +8 -6
- package/dist/tools/builtin/ragSearchTool.js.map +1 -1
- package/dist/tools/executor.d.ts +30 -2
- package/dist/tools/executor.d.ts.map +1 -1
- package/dist/tools/executor.js +83 -15
- package/dist/tools/executor.js.map +1 -1
- package/dist/tools/index.d.ts +3 -1
- package/dist/tools/index.d.ts.map +1 -1
- package/dist/tools/registry.d.ts +15 -3
- package/dist/tools/registry.d.ts.map +1 -1
- package/dist/tools/registry.js +21 -2
- package/dist/tools/registry.js.map +1 -1
- package/package.json +58 -4
|
@@ -0,0 +1,291 @@
|
|
|
1
|
+
"use strict";
|
|
2
|
+
Object.defineProperty(exports, "__esModule", { value: true });
|
|
3
|
+
exports.SemanticChunker = void 0;
|
|
4
|
+
const embeddings_1 = require("../embeddings");
|
|
5
|
+
const chunker_1 = require("./chunker");
|
|
6
|
+
const DEFAULT_SIMILARITY_THRESHOLD = 0.72;
|
|
7
|
+
const SENTENCE_ENDINGS = new Set(['。', '!', '?', '!', '?']);
|
|
8
|
+
class SemanticChunker {
|
|
9
|
+
embedder;
|
|
10
|
+
similarityThreshold;
|
|
11
|
+
maxChunkLength;
|
|
12
|
+
minChunkLength;
|
|
13
|
+
overlapLength;
|
|
14
|
+
constructor(options) {
|
|
15
|
+
this.embedder = options.embedder;
|
|
16
|
+
this.similarityThreshold = normalizeSimilarityThreshold(options.similarityThreshold, DEFAULT_SIMILARITY_THRESHOLD);
|
|
17
|
+
this.maxChunkLength = toPositiveInteger(options.maxChunkLength, chunker_1.DEFAULT_MAX_CHUNK_LENGTH);
|
|
18
|
+
this.minChunkLength = Math.min(toPositiveInteger(options.minChunkLength, chunker_1.DEFAULT_MIN_CHUNK_LENGTH), this.maxChunkLength);
|
|
19
|
+
this.overlapLength = Math.min(normalizeNonNegativeInteger(options.overlapLength, 0), Math.max(this.maxChunkLength - 1, 0));
|
|
20
|
+
}
|
|
21
|
+
async chunk(document) {
|
|
22
|
+
const units = createSemanticUnits(document.content, this.maxChunkLength);
|
|
23
|
+
if (units.length === 0) {
|
|
24
|
+
return [];
|
|
25
|
+
}
|
|
26
|
+
const embeddings = await this.embedder.embedDocuments(units.map((unit) => unit.content));
|
|
27
|
+
(0, embeddings_1.assertEmbeddingBatchSize)(embeddings, units.length);
|
|
28
|
+
embeddings.forEach((embedding, index) => {
|
|
29
|
+
(0, embeddings_1.assertEmbeddingDimensions)(embedding, this.embedder.dimensions, `semantic unit ${index} embedding`);
|
|
30
|
+
});
|
|
31
|
+
const groups = createSemanticGroups(units, embeddings, {
|
|
32
|
+
maxChunkLength: this.maxChunkLength,
|
|
33
|
+
minChunkLength: this.minChunkLength,
|
|
34
|
+
overlapLength: this.overlapLength,
|
|
35
|
+
similarityThreshold: this.similarityThreshold,
|
|
36
|
+
});
|
|
37
|
+
return groups.map((group, index) => {
|
|
38
|
+
const startOffset = group.units[0]?.startOffset ?? 0;
|
|
39
|
+
const endOffset = group.units.at(-1)?.endOffset ?? startOffset;
|
|
40
|
+
return (0, chunker_1.createChunk)({
|
|
41
|
+
document,
|
|
42
|
+
content: document.content.slice(startOffset, endOffset),
|
|
43
|
+
kind: 'text',
|
|
44
|
+
chunkIndex: index,
|
|
45
|
+
startOffset,
|
|
46
|
+
endOffset,
|
|
47
|
+
extra: createSemanticExtra({
|
|
48
|
+
embedder: this.embedder,
|
|
49
|
+
group,
|
|
50
|
+
similarityThreshold: this.similarityThreshold,
|
|
51
|
+
}),
|
|
52
|
+
});
|
|
53
|
+
});
|
|
54
|
+
}
|
|
55
|
+
async chunkMany(documents) {
|
|
56
|
+
return await (0, chunker_1.chunkManyDocuments)(this, documents);
|
|
57
|
+
}
|
|
58
|
+
}
|
|
59
|
+
exports.SemanticChunker = SemanticChunker;
|
|
60
|
+
function createSemanticUnits(text, maxChunkLength) {
|
|
61
|
+
const units = [];
|
|
62
|
+
for (const paragraph of splitParagraphs(text)) {
|
|
63
|
+
if (paragraph.content.length <= maxChunkLength) {
|
|
64
|
+
const sentences = splitSentences(paragraph.content, paragraph.startOffset);
|
|
65
|
+
units.push(...sentences.length > 0 ? sentences : [paragraph]);
|
|
66
|
+
continue;
|
|
67
|
+
}
|
|
68
|
+
for (const segment of (0, chunker_1.splitTextSegments)(paragraph.content, {
|
|
69
|
+
maxChunkLength,
|
|
70
|
+
minChunkLength: Math.min(chunker_1.DEFAULT_MIN_CHUNK_LENGTH, maxChunkLength),
|
|
71
|
+
overlapLength: 0,
|
|
72
|
+
})) {
|
|
73
|
+
units.push({
|
|
74
|
+
content: segment.content,
|
|
75
|
+
startOffset: paragraph.startOffset + segment.startOffset,
|
|
76
|
+
endOffset: paragraph.startOffset + segment.endOffset,
|
|
77
|
+
});
|
|
78
|
+
}
|
|
79
|
+
}
|
|
80
|
+
return units;
|
|
81
|
+
}
|
|
82
|
+
function splitParagraphs(text) {
|
|
83
|
+
const paragraphs = [];
|
|
84
|
+
const matches = text.matchAll(/\S(?:[\s\S]*?\S)?(?=\n\s*\n|$)/g);
|
|
85
|
+
for (const match of matches) {
|
|
86
|
+
const content = match[0];
|
|
87
|
+
const startOffset = match.index ?? 0;
|
|
88
|
+
const trimmedStartOffset = startOffset + countLeadingWhitespace(content);
|
|
89
|
+
const trimmedEndOffset = startOffset + content.length - countTrailingWhitespace(content);
|
|
90
|
+
if (trimmedStartOffset < trimmedEndOffset) {
|
|
91
|
+
paragraphs.push({
|
|
92
|
+
content: text.slice(trimmedStartOffset, trimmedEndOffset),
|
|
93
|
+
startOffset: trimmedStartOffset,
|
|
94
|
+
endOffset: trimmedEndOffset,
|
|
95
|
+
});
|
|
96
|
+
}
|
|
97
|
+
}
|
|
98
|
+
return paragraphs;
|
|
99
|
+
}
|
|
100
|
+
function splitSentences(paragraph, paragraphStartOffset) {
|
|
101
|
+
const units = [];
|
|
102
|
+
let start = 0;
|
|
103
|
+
for (let index = 0; index < paragraph.length; index += 1) {
|
|
104
|
+
const character = paragraph[index];
|
|
105
|
+
if (!isSentenceEnding(paragraph, index, character)) {
|
|
106
|
+
continue;
|
|
107
|
+
}
|
|
108
|
+
const end = index + 1;
|
|
109
|
+
appendUnit(units, paragraph, paragraphStartOffset, start, end);
|
|
110
|
+
start = skipWhitespace(paragraph, end);
|
|
111
|
+
}
|
|
112
|
+
appendUnit(units, paragraph, paragraphStartOffset, start, paragraph.length);
|
|
113
|
+
return units;
|
|
114
|
+
}
|
|
115
|
+
function appendUnit(units, text, baseOffset, start, end) {
|
|
116
|
+
const trimmed = trimRange(text, start, end);
|
|
117
|
+
if (trimmed.start >= trimmed.end) {
|
|
118
|
+
return;
|
|
119
|
+
}
|
|
120
|
+
units.push({
|
|
121
|
+
content: text.slice(trimmed.start, trimmed.end),
|
|
122
|
+
startOffset: baseOffset + trimmed.start,
|
|
123
|
+
endOffset: baseOffset + trimmed.end,
|
|
124
|
+
});
|
|
125
|
+
}
|
|
126
|
+
function createSemanticGroups(units, embeddings, options) {
|
|
127
|
+
const groups = [];
|
|
128
|
+
let current = [];
|
|
129
|
+
let pendingBreakSimilarity;
|
|
130
|
+
for (const [index, unit] of units.entries()) {
|
|
131
|
+
const previousEmbedding = index > 0 ? embeddings[index - 1] : undefined;
|
|
132
|
+
const currentEmbedding = embeddings[index];
|
|
133
|
+
const similarity = previousEmbedding !== undefined && currentEmbedding !== undefined
|
|
134
|
+
? cosineSimilarity(previousEmbedding, currentEmbedding)
|
|
135
|
+
: undefined;
|
|
136
|
+
const semanticBreak = (similarity !== undefined
|
|
137
|
+
&& similarity < options.similarityThreshold
|
|
138
|
+
&& getGroupLength(current) >= options.minChunkLength);
|
|
139
|
+
const lengthBreak = (current.length > 0
|
|
140
|
+
&& getRangeLength(current[0], unit) > options.maxChunkLength
|
|
141
|
+
&& getGroupLength(current) >= options.minChunkLength);
|
|
142
|
+
if (current.length > 0 && (semanticBreak || lengthBreak)) {
|
|
143
|
+
groups.push(createGroup(current, pendingBreakSimilarity));
|
|
144
|
+
current = createOverlapUnits(current, options.overlapLength, unit, options.maxChunkLength);
|
|
145
|
+
pendingBreakSimilarity = similarity;
|
|
146
|
+
}
|
|
147
|
+
current.push(unit);
|
|
148
|
+
}
|
|
149
|
+
if (current.length > 0) {
|
|
150
|
+
groups.push(createGroup(current, pendingBreakSimilarity));
|
|
151
|
+
}
|
|
152
|
+
return groups;
|
|
153
|
+
}
|
|
154
|
+
function createGroup(units, breakSimilarity) {
|
|
155
|
+
const group = {
|
|
156
|
+
units: [...units],
|
|
157
|
+
};
|
|
158
|
+
if (breakSimilarity !== undefined) {
|
|
159
|
+
group.breakSimilarity = breakSimilarity;
|
|
160
|
+
}
|
|
161
|
+
return group;
|
|
162
|
+
}
|
|
163
|
+
function createOverlapUnits(units, overlapLength, nextUnit, maxChunkLength) {
|
|
164
|
+
if (overlapLength <= 0) {
|
|
165
|
+
return [];
|
|
166
|
+
}
|
|
167
|
+
const overlap = [];
|
|
168
|
+
for (let index = units.length - 1; index >= 0; index -= 1) {
|
|
169
|
+
const unit = units[index];
|
|
170
|
+
if (unit === undefined) {
|
|
171
|
+
continue;
|
|
172
|
+
}
|
|
173
|
+
const nextOverlap = [unit, ...overlap];
|
|
174
|
+
if (getGroupLength(nextOverlap) > overlapLength) {
|
|
175
|
+
break;
|
|
176
|
+
}
|
|
177
|
+
if (getRangeLength(nextOverlap[0], nextUnit) > maxChunkLength) {
|
|
178
|
+
break;
|
|
179
|
+
}
|
|
180
|
+
overlap.unshift(unit);
|
|
181
|
+
}
|
|
182
|
+
return overlap;
|
|
183
|
+
}
|
|
184
|
+
function createSemanticExtra(input) {
|
|
185
|
+
const extra = {
|
|
186
|
+
strategy: 'semantic',
|
|
187
|
+
experimental: true,
|
|
188
|
+
embeddingModel: input.embedder.model,
|
|
189
|
+
similarityThreshold: input.similarityThreshold,
|
|
190
|
+
unitCount: input.group.units.length,
|
|
191
|
+
};
|
|
192
|
+
if (input.group.breakSimilarity !== undefined) {
|
|
193
|
+
extra.breakSimilarity = input.group.breakSimilarity;
|
|
194
|
+
}
|
|
195
|
+
return extra;
|
|
196
|
+
}
|
|
197
|
+
function cosineSimilarity(left, right) {
|
|
198
|
+
let dot = 0;
|
|
199
|
+
let leftMagnitude = 0;
|
|
200
|
+
let rightMagnitude = 0;
|
|
201
|
+
for (let index = 0; index < left.length; index += 1) {
|
|
202
|
+
const leftValue = left[index] ?? 0;
|
|
203
|
+
const rightValue = right[index] ?? 0;
|
|
204
|
+
dot += leftValue * rightValue;
|
|
205
|
+
leftMagnitude += leftValue * leftValue;
|
|
206
|
+
rightMagnitude += rightValue * rightValue;
|
|
207
|
+
}
|
|
208
|
+
if (leftMagnitude === 0 || rightMagnitude === 0) {
|
|
209
|
+
return 0;
|
|
210
|
+
}
|
|
211
|
+
return dot / (Math.sqrt(leftMagnitude) * Math.sqrt(rightMagnitude));
|
|
212
|
+
}
|
|
213
|
+
function getGroupLength(units) {
|
|
214
|
+
const first = units[0];
|
|
215
|
+
const last = units.at(-1);
|
|
216
|
+
if (first === undefined || last === undefined) {
|
|
217
|
+
return 0;
|
|
218
|
+
}
|
|
219
|
+
return getRangeLength(first, last);
|
|
220
|
+
}
|
|
221
|
+
function getRangeLength(first, last) {
|
|
222
|
+
return first === undefined ? last.content.length : last.endOffset - first.startOffset;
|
|
223
|
+
}
|
|
224
|
+
function isSentenceEnding(text, index, character) {
|
|
225
|
+
if (character === undefined) {
|
|
226
|
+
return false;
|
|
227
|
+
}
|
|
228
|
+
if (SENTENCE_ENDINGS.has(character)) {
|
|
229
|
+
return true;
|
|
230
|
+
}
|
|
231
|
+
if (character !== '.') {
|
|
232
|
+
return false;
|
|
233
|
+
}
|
|
234
|
+
const previous = text[index - 1];
|
|
235
|
+
const next = text[index + 1];
|
|
236
|
+
if (isAsciiAlphanumeric(previous) && isAsciiAlphanumeric(next)) {
|
|
237
|
+
return false;
|
|
238
|
+
}
|
|
239
|
+
return next === undefined || /\s/.test(next);
|
|
240
|
+
}
|
|
241
|
+
function isAsciiAlphanumeric(value) {
|
|
242
|
+
return value !== undefined && /[A-Za-z0-9]/.test(value);
|
|
243
|
+
}
|
|
244
|
+
function countLeadingWhitespace(text) {
|
|
245
|
+
const match = text.match(/^\s*/);
|
|
246
|
+
return match?.[0].length ?? 0;
|
|
247
|
+
}
|
|
248
|
+
function countTrailingWhitespace(text) {
|
|
249
|
+
const match = text.match(/\s*$/);
|
|
250
|
+
return match?.[0].length ?? 0;
|
|
251
|
+
}
|
|
252
|
+
function skipWhitespace(text, start) {
|
|
253
|
+
let index = start;
|
|
254
|
+
while (index < text.length && /\s/.test(text[index] ?? '')) {
|
|
255
|
+
index += 1;
|
|
256
|
+
}
|
|
257
|
+
return index;
|
|
258
|
+
}
|
|
259
|
+
function trimRange(text, start, end) {
|
|
260
|
+
let trimmedStart = start;
|
|
261
|
+
let trimmedEnd = end;
|
|
262
|
+
while (trimmedStart < trimmedEnd && /\s/.test(text[trimmedStart] ?? '')) {
|
|
263
|
+
trimmedStart += 1;
|
|
264
|
+
}
|
|
265
|
+
while (trimmedEnd > trimmedStart && /\s/.test(text[trimmedEnd - 1] ?? '')) {
|
|
266
|
+
trimmedEnd -= 1;
|
|
267
|
+
}
|
|
268
|
+
return {
|
|
269
|
+
start: trimmedStart,
|
|
270
|
+
end: trimmedEnd,
|
|
271
|
+
};
|
|
272
|
+
}
|
|
273
|
+
function normalizeSimilarityThreshold(value, fallback) {
|
|
274
|
+
if (value === undefined || !Number.isFinite(value)) {
|
|
275
|
+
return fallback;
|
|
276
|
+
}
|
|
277
|
+
return Math.max(-1, Math.min(1, value));
|
|
278
|
+
}
|
|
279
|
+
function toPositiveInteger(value, fallback) {
|
|
280
|
+
if (value === undefined || !Number.isFinite(value) || value <= 0) {
|
|
281
|
+
return fallback;
|
|
282
|
+
}
|
|
283
|
+
return Math.floor(value);
|
|
284
|
+
}
|
|
285
|
+
function normalizeNonNegativeInteger(value, fallback) {
|
|
286
|
+
if (value === undefined || !Number.isFinite(value) || value < 0) {
|
|
287
|
+
return fallback;
|
|
288
|
+
}
|
|
289
|
+
return Math.floor(value);
|
|
290
|
+
}
|
|
291
|
+
//# sourceMappingURL=semantic-chunker.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"semantic-chunker.js","sourceRoot":"","sources":["../../../src/rag/chunking/semantic-chunker.ts"],"names":[],"mappings":";;;AACA,8CAGuB;AAKvB,uCAMmB;AAMnB,MAAM,4BAA4B,GAAG,IAAI,CAAC;AAC1C,MAAM,gBAAgB,GAAG,IAAI,GAAG,CAAC,CAAC,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,CAAC,CAAC,CAAC;AAmB5D,MAAa,eAAe;IACP,QAAQ,CAAW;IACnB,mBAAmB,CAAS;IAC5B,cAAc,CAAS;IACvB,cAAc,CAAS;IACvB,aAAa,CAAS;IAEvC,YAAY,OAA+B;QACvC,IAAI,CAAC,QAAQ,GAAG,OAAO,CAAC,QAAQ,CAAC;QACjC,IAAI,CAAC,mBAAmB,GAAG,4BAA4B,CACnD,OAAO,CAAC,mBAAmB,EAC3B,4BAA4B,CAC/B,CAAC;QACF,IAAI,CAAC,cAAc,GAAG,iBAAiB,CACnC,OAAO,CAAC,cAAc,EACtB,kCAAwB,CAC3B,CAAC;QACF,IAAI,CAAC,cAAc,GAAG,IAAI,CAAC,GAAG,CAC1B,iBAAiB,CAAC,OAAO,CAAC,cAAc,EAAE,kCAAwB,CAAC,EACnE,IAAI,CAAC,cAAc,CACtB,CAAC;QACF,IAAI,CAAC,aAAa,GAAG,IAAI,CAAC,GAAG,CACzB,2BAA2B,CAAC,OAAO,CAAC,aAAa,EAAE,CAAC,CAAC,EACrD,IAAI,CAAC,GAAG,CAAC,IAAI,CAAC,cAAc,GAAG,CAAC,EAAE,CAAC,CAAC,CACvC,CAAC;IACN,CAAC;IAED,KAAK,CAAC,KAAK,CAAC,QAAqB;QAC7B,MAAM,KAAK,GAAG,mBAAmB,CAAC,QAAQ,CAAC,OAAO,EAAE,IAAI,CAAC,cAAc,CAAC,CAAC;QACzE,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;YACrB,OAAO,EAAE,CAAC;QACd,CAAC;QAED,MAAM,UAAU,GAAG,MAAM,IAAI,CAAC,QAAQ,CAAC,cAAc,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,CAAC;QACzF,IAAA,qCAAwB,EAAC,UAAU,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC;QACnD,UAAU,CAAC,OAAO,CAAC,CAAC,SAAS,EAAE,KAAK,EAAE,EAAE;YACpC,IAAA,sCAAyB,EACrB,SAAS,EACT,IAAI,CAAC,QAAQ,CAAC,UAAU,EACxB,iBAAiB,KAAK,YAAY,CACrC,CAAC;QACN,CAAC,CAAC,CAAC;QAEH,MAAM,MAAM,GAAG,oBAAoB,CAAC,KAAK,EAAE,UAAU,EAAE;YACnD,cAAc,EAAE,IAAI,CAAC,cAAc;YACnC,cAAc,EAAE,IAAI,CAAC,cAAc;YACnC,aAAa,EAAE,IAAI,CAAC,aAAa;YACjC,mBAAmB,EAAE,IAAI,CAAC,mBAAmB;SAChD,CAAC,CAAC;QAEH,OAAO,MAAM,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,KAAK,EAAE,EAAE;YAC/B,MAAM,WAAW,GAAG,KAAK,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,WAAW,IAAI,CAAC,CAAC;YACrD,MAAM,SAAS,GAAG,KAAK,CAAC,KAAK,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,EAAE,SAAS,IAAI,WAAW,CAAC;YAE/D,OAAO,IAAA,qBAAW,EAAC;gBACf,QAAQ;gBACR,OAAO,EAAE,QAAQ,CAAC,OAAO,CAAC,KAAK,CAAC,WAAW,EAAE,SAAS,CAAC;gBACvD,IAAI,EAAE,MAAM;gBACZ,UAAU,EAAE,KAAK;gBACjB,WAAW;gBACX,SAAS;gBACT,KAAK,EAAE,mBAAmB,CAAC;oBACvB,QAAQ,EAAE,IAAI,CAAC,QAAQ;oBACvB,KAAK;oBACL,mBAAmB,EAAE,IAAI,CAAC,mBAAmB;iBAChD,CAAC;aACL,CAAC,CAAC;QACP,CAAC,CAAC,CAAC;IACP,CAAC;IAED,KAAK,CAAC,SAAS,CAAC,SAAwB;QACpC,OAAO,MAAM,IAAA,4BAAkB,EAAC,IAAI,EAAE,SAAS,CAAC,CAAC;IACrD,CAAC;CACJ;AAzED,0CAyEC;AAED,SAAS,mBAAmB,CAAC,IAAY,EAAE,cAAsB;IAC7D,MAAM,KAAK,GAAmB,EAAE,CAAC;IAEjC,KAAK,MAAM,SAAS,IAAI,eAAe,CAAC,IAAI,CAAC,EAAE,CAAC;QAC5C,IAAI,SAAS,CAAC,OAAO,CAAC,MAAM,IAAI,cAAc,EAAE,CAAC;YAC7C,MAAM,SAAS,GAAG,cAAc,CAAC,SAAS,CAAC,OAAO,EAAE,SAAS,CAAC,WAAW,CAAC,CAAC;YAC3E,KAAK,CAAC,IAAI,CAAC,GAAG,SAAS,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC,CAAC;YAC9D,SAAS;QACb,CAAC;QAED,KAAK,MAAM,OAAO,IAAI,IAAA,2BAAiB,EAAC,SAAS,CAAC,OAAO,EAAE;YACvD,cAAc;YACd,cAAc,EAAE,IAAI,CAAC,GAAG,CAAC,kCAAwB,EAAE,cAAc,CAAC;YAClE,aAAa,EAAE,CAAC;SACnB,CAAC,EAAE,CAAC;YACD,KAAK,CAAC,IAAI,CAAC;gBACP,OAAO,EAAE,OAAO,CAAC,OAAO;gBACxB,WAAW,EAAE,SAAS,CAAC,WAAW,GAAG,OAAO,CAAC,WAAW;gBACxD,SAAS,EAAE,SAAS,CAAC,WAAW,GAAG,OAAO,CAAC,SAAS;aACvD,CAAC,CAAC;QACP,CAAC;IACL,CAAC;IAED,OAAO,KAAK,CAAC;AACjB,CAAC;AAED,SAAS,eAAe,CAAC,IAAY;IACjC,MAAM,UAAU,GAAmB,EAAE,CAAC;IACtC,MAAM,OAAO,GAAG,IAAI,CAAC,QAAQ,CAAC,iCAAiC,CAAC,CAAC;IAEjE,KAAK,MAAM,KAAK,IAAI,OAAO,EAAE,CAAC;QAC1B,MAAM,OAAO,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC;QACzB,MAAM,WAAW,GAAG,KAAK,CAAC,KAAK,IAAI,CAAC,CAAC;QACrC,MAAM,kBAAkB,GAAG,WAAW,GAAG,sBAAsB,CAAC,OAAO,CAAC,CAAC;QACzE,MAAM,gBAAgB,GAAG,WAAW,GAAG,OAAO,CAAC,MAAM,GAAG,uBAAuB,CAAC,OAAO,CAAC,CAAC;QAEzF,IAAI,kBAAkB,GAAG,gBAAgB,EAAE,CAAC;YACxC,UAAU,CAAC,IAAI,CAAC;gBACZ,OAAO,EAAE,IAAI,CAAC,KAAK,CAAC,kBAAkB,EAAE,gBAAgB,CAAC;gBACzD,WAAW,EAAE,kBAAkB;gBAC/B,SAAS,EAAE,gBAAgB;aAC9B,CAAC,CAAC;QACP,CAAC;IACL,CAAC;IAED,OAAO,UAAU,CAAC;AACtB,CAAC;AAED,SAAS,cAAc,CAAC,SAAiB,EAAE,oBAA4B;IACnE,MAAM,KAAK,GAAmB,EAAE,CAAC;IACjC,IAAI,KAAK,GAAG,CAAC,CAAC;IAEd,KAAK,IAAI,KAAK,GAAG,CAAC,EAAE,KAAK,GAAG,SAAS,CAAC,MAAM,EAAE,KAAK,IAAI,CAAC,EAAE,CAAC;QACvD,MAAM,SAAS,GAAG,SAAS,CAAC,KAAK,CAAC,CAAC;QACnC,IAAI,CAAC,gBAAgB,CAAC,SAAS,EAAE,KAAK,EAAE,SAAS,CAAC,EAAE,CAAC;YACjD,SAAS;QACb,CAAC;QAED,MAAM,GAAG,GAAG,KAAK,GAAG,CAAC,CAAC;QACtB,UAAU,CAAC,KAAK,EAAE,SAAS,EAAE,oBAAoB,EAAE,KAAK,EAAE,GAAG,CAAC,CAAC;QAC/D,KAAK,GAAG,cAAc,CAAC,SAAS,EAAE,GAAG,CAAC,CAAC;IAC3C,CAAC;IAED,UAAU,CAAC,KAAK,EAAE,SAAS,EAAE,oBAAoB,EAAE,KAAK,EAAE,SAAS,CAAC,MAAM,CAAC,CAAC;IAE5E,OAAO,KAAK,CAAC;AACjB,CAAC;AAED,SAAS,UAAU,CACf,KAAqB,EACrB,IAAY,EACZ,UAAkB,EAClB,KAAa,EACb,GAAW;IAEX,MAAM,OAAO,GAAG,SAAS,CAAC,IAAI,EAAE,KAAK,EAAE,GAAG,CAAC,CAAC;IAC5C,IAAI,OAAO,CAAC,KAAK,IAAI,OAAO,CAAC,GAAG,EAAE,CAAC;QAC/B,OAAO;IACX,CAAC;IAED,KAAK,CAAC,IAAI,CAAC;QACP,OAAO,EAAE,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,KAAK,EAAE,OAAO,CAAC,GAAG,CAAC;QAC/C,WAAW,EAAE,UAAU,GAAG,OAAO,CAAC,KAAK;QACvC,SAAS,EAAE,UAAU,GAAG,OAAO,CAAC,GAAG;KACtC,CAAC,CAAC;AACP,CAAC;AAED,SAAS,oBAAoB,CACzB,KAAqB,EACrB,UAAsB,EACtB,OAKC;IAED,MAAM,MAAM,GAAoB,EAAE,CAAC;IACnC,IAAI,OAAO,GAAmB,EAAE,CAAC;IACjC,IAAI,sBAA0C,CAAC;IAE/C,KAAK,MAAM,CAAC,KAAK,EAAE,IAAI,CAAC,IAAI,KAAK,CAAC,OAAO,EAAE,EAAE,CAAC;QAC1C,MAAM,iBAAiB,GAAG,KAAK,GAAG,CAAC,CAAC,CAAC,CAAC,UAAU,CAAC,KAAK,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC;QACxE,MAAM,gBAAgB,GAAG,UAAU,CAAC,KAAK,CAAC,CAAC;QAC3C,MAAM,UAAU,GAAG,iBAAiB,KAAK,SAAS,IAAI,gBAAgB,KAAK,SAAS;YAChF,CAAC,CAAC,gBAAgB,CAAC,iBAAiB,EAAE,gBAAgB,CAAC;YACvD,CAAC,CAAC,SAAS,CAAC;QAChB,MAAM,aAAa,GAAG,CAClB,UAAU,KAAK,SAAS;eACrB,UAAU,GAAG,OAAO,CAAC,mBAAmB;eACxC,cAAc,CAAC,OAAO,CAAC,IAAI,OAAO,CAAC,cAAc,CACvD,CAAC;QACF,MAAM,WAAW,GAAG,CAChB,OAAO,CAAC,MAAM,GAAG,CAAC;eACf,cAAc,CAAC,OAAO,CAAC,CAAC,CAAC,EAAE,IAAI,CAAC,GAAG,OAAO,CAAC,cAAc;eACzD,cAAc,CAAC,OAAO,CAAC,IAAI,OAAO,CAAC,cAAc,CACvD,CAAC;QAEF,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,IAAI,CAAC,aAAa,IAAI,WAAW,CAAC,EAAE,CAAC;YACvD,MAAM,CAAC,IAAI,CAAC,WAAW,CAAC,OAAO,EAAE,sBAAsB,CAAC,CAAC,CAAC;YAC1D,OAAO,GAAG,kBAAkB,CAAC,OAAO,EAAE,OAAO,CAAC,aAAa,EAAE,IAAI,EAAE,OAAO,CAAC,cAAc,CAAC,CAAC;YAC3F,sBAAsB,GAAG,UAAU,CAAC;QACxC,CAAC;QAED,OAAO,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IACvB,CAAC;IAED,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QACrB,MAAM,CAAC,IAAI,CAAC,WAAW,CAAC,OAAO,EAAE,sBAAsB,CAAC,CAAC,CAAC;IAC9D,CAAC;IAED,OAAO,MAAM,CAAC;AAClB,CAAC;AAED,SAAS,WAAW,CAAC,KAAqB,EAAE,eAAmC;IAC3E,MAAM,KAAK,GAAkB;QACzB,KAAK,EAAE,CAAC,GAAG,KAAK,CAAC;KACpB,CAAC;IAEF,IAAI,eAAe,KAAK,SAAS,EAAE,CAAC;QAChC,KAAK,CAAC,eAAe,GAAG,eAAe,CAAC;IAC5C,CAAC;IAED,OAAO,KAAK,CAAC;AACjB,CAAC;AAED,SAAS,kBAAkB,CACvB,KAAqB,EACrB,aAAqB,EACrB,QAAsB,EACtB,cAAsB;IAEtB,IAAI,aAAa,IAAI,CAAC,EAAE,CAAC;QACrB,OAAO,EAAE,CAAC;IACd,CAAC;IAED,MAAM,OAAO,GAAmB,EAAE,CAAC;IACnC,KAAK,IAAI,KAAK,GAAG,KAAK,CAAC,MAAM,GAAG,CAAC,EAAE,KAAK,IAAI,CAAC,EAAE,KAAK,IAAI,CAAC,EAAE,CAAC;QACxD,MAAM,IAAI,GAAG,KAAK,CAAC,KAAK,CAAC,CAAC;QAC1B,IAAI,IAAI,KAAK,SAAS,EAAE,CAAC;YACrB,SAAS;QACb,CAAC;QAED,MAAM,WAAW,GAAG,CAAC,IAAI,EAAE,GAAG,OAAO,CAAC,CAAC;QACvC,IAAI,cAAc,CAAC,WAAW,CAAC,GAAG,aAAa,EAAE,CAAC;YAC9C,MAAM;QACV,CAAC;QACD,IAAI,cAAc,CAAC,WAAW,CAAC,CAAC,CAAC,EAAE,QAAQ,CAAC,GAAG,cAAc,EAAE,CAAC;YAC5D,MAAM;QACV,CAAC;QAED,OAAO,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC;IAC1B,CAAC;IAED,OAAO,OAAO,CAAC;AACnB,CAAC;AAED,SAAS,mBAAmB,CAAC,KAI5B;IACG,MAAM,KAAK,GAA4B;QACnC,QAAQ,EAAE,UAAU;QACpB,YAAY,EAAE,IAAI;QAClB,cAAc,EAAE,KAAK,CAAC,QAAQ,CAAC,KAAK;QACpC,mBAAmB,EAAE,KAAK,CAAC,mBAAmB;QAC9C,SAAS,EAAE,KAAK,CAAC,KAAK,CAAC,KAAK,CAAC,MAAM;KACtC,CAAC;IAEF,IAAI,KAAK,CAAC,KAAK,CAAC,eAAe,KAAK,SAAS,EAAE,CAAC;QAC5C,KAAK,CAAC,eAAe,GAAG,KAAK,CAAC,KAAK,CAAC,eAAe,CAAC;IACxD,CAAC;IAED,OAAO,KAAK,CAAC;AACjB,CAAC;AAED,SAAS,gBAAgB,CAAC,IAAuB,EAAE,KAAwB;IACvE,IAAI,GAAG,GAAG,CAAC,CAAC;IACZ,IAAI,aAAa,GAAG,CAAC,CAAC;IACtB,IAAI,cAAc,GAAG,CAAC,CAAC;IAEvB,KAAK,IAAI,KAAK,GAAG,CAAC,EAAE,KAAK,GAAG,IAAI,CAAC,MAAM,EAAE,KAAK,IAAI,CAAC,EAAE,CAAC;QAClD,MAAM,SAAS,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;QACnC,MAAM,UAAU,GAAG,KAAK,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;QAErC,GAAG,IAAI,SAAS,GAAG,UAAU,CAAC;QAC9B,aAAa,IAAI,SAAS,GAAG,SAAS,CAAC;QACvC,cAAc,IAAI,UAAU,GAAG,UAAU,CAAC;IAC9C,CAAC;IAED,IAAI,aAAa,KAAK,CAAC,IAAI,cAAc,KAAK,CAAC,EAAE,CAAC;QAC9C,OAAO,CAAC,CAAC;IACb,CAAC;IAED,OAAO,GAAG,GAAG,CAAC,IAAI,CAAC,IAAI,CAAC,aAAa,CAAC,GAAG,IAAI,CAAC,IAAI,CAAC,cAAc,CAAC,CAAC,CAAC;AACxE,CAAC;AAED,SAAS,cAAc,CAAC,KAAqB;IACzC,MAAM,KAAK,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC;IACvB,MAAM,IAAI,GAAG,KAAK,CAAC,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC;IAC1B,IAAI,KAAK,KAAK,SAAS,IAAI,IAAI,KAAK,SAAS,EAAE,CAAC;QAC5C,OAAO,CAAC,CAAC;IACb,CAAC;IAED,OAAO,cAAc,CAAC,KAAK,EAAE,IAAI,CAAC,CAAC;AACvC,CAAC;AAED,SAAS,cAAc,CAAC,KAA+B,EAAE,IAAkB;IACvE,OAAO,KAAK,KAAK,SAAS,CAAC,CAAC,CAAC,IAAI,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC,CAAC,IAAI,CAAC,SAAS,GAAG,KAAK,CAAC,WAAW,CAAC;AAC1F,CAAC;AAED,SAAS,gBAAgB,CACrB,IAAY,EACZ,KAAa,EACb,SAA6B;IAE7B,IAAI,SAAS,KAAK,SAAS,EAAE,CAAC;QAC1B,OAAO,KAAK,CAAC;IACjB,CAAC;IACD,IAAI,gBAAgB,CAAC,GAAG,CAAC,SAAS,CAAC,EAAE,CAAC;QAClC,OAAO,IAAI,CAAC;IAChB,CAAC;IACD,IAAI,SAAS,KAAK,GAAG,EAAE,CAAC;QACpB,OAAO,KAAK,CAAC;IACjB,CAAC;IAED,MAAM,QAAQ,GAAG,IAAI,CAAC,KAAK,GAAG,CAAC,CAAC,CAAC;IACjC,MAAM,IAAI,GAAG,IAAI,CAAC,KAAK,GAAG,CAAC,CAAC,CAAC;IAC7B,IAAI,mBAAmB,CAAC,QAAQ,CAAC,IAAI,mBAAmB,CAAC,IAAI,CAAC,EAAE,CAAC;QAC7D,OAAO,KAAK,CAAC;IACjB,CAAC;IAED,OAAO,IAAI,KAAK,SAAS,IAAI,IAAI,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;AACjD,CAAC;AAED,SAAS,mBAAmB,CAAC,KAAyB;IAClD,OAAO,KAAK,KAAK,SAAS,IAAI,aAAa,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;AAC5D,CAAC;AAED,SAAS,sBAAsB,CAAC,IAAY;IACxC,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC;IAEjC,OAAO,KAAK,EAAE,CAAC,CAAC,CAAC,CAAC,MAAM,IAAI,CAAC,CAAC;AAClC,CAAC;AAED,SAAS,uBAAuB,CAAC,IAAY;IACzC,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC;IAEjC,OAAO,KAAK,EAAE,CAAC,CAAC,CAAC,CAAC,MAAM,IAAI,CAAC,CAAC;AAClC,CAAC;AAED,SAAS,cAAc,CAAC,IAAY,EAAE,KAAa;IAC/C,IAAI,KAAK,GAAG,KAAK,CAAC;IAClB,OAAO,KAAK,GAAG,IAAI,CAAC,MAAM,IAAI,IAAI,CAAC,IAAI,CAAC,IAAI,CAAC,KAAK,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC;QACzD,KAAK,IAAI,CAAC,CAAC;IACf,CAAC;IAED,OAAO,KAAK,CAAC;AACjB,CAAC;AAED,SAAS,SAAS,CAAC,IAAY,EAAE,KAAa,EAAE,GAAW;IACvD,IAAI,YAAY,GAAG,KAAK,CAAC;IACzB,IAAI,UAAU,GAAG,GAAG,CAAC;IAErB,OAAO,YAAY,GAAG,UAAU,IAAI,IAAI,CAAC,IAAI,CAAC,IAAI,CAAC,YAAY,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC;QACtE,YAAY,IAAI,CAAC,CAAC;IACtB,CAAC;IACD,OAAO,UAAU,GAAG,YAAY,IAAI,IAAI,CAAC,IAAI,CAAC,IAAI,CAAC,UAAU,GAAG,CAAC,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC;QACxE,UAAU,IAAI,CAAC,CAAC;IACpB,CAAC;IAED,OAAO;QACH,KAAK,EAAE,YAAY;QACnB,GAAG,EAAE,UAAU;KAClB,CAAC;AACN,CAAC;AAED,SAAS,4BAA4B,CAAC,KAAyB,EAAE,QAAgB;IAC7E,IAAI,KAAK,KAAK,SAAS,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,KAAK,CAAC,EAAE,CAAC;QACjD,OAAO,QAAQ,CAAC;IACpB,CAAC;IAED,OAAO,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,KAAK,CAAC,CAAC,CAAC;AAC5C,CAAC;AAED,SAAS,iBAAiB,CAAC,KAAyB,EAAE,QAAgB;IAClE,IAAI,KAAK,KAAK,SAAS,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,KAAK,CAAC,IAAI,KAAK,IAAI,CAAC,EAAE,CAAC;QAC/D,OAAO,QAAQ,CAAC;IACpB,CAAC;IAED,OAAO,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC;AAC7B,CAAC;AAED,SAAS,2BAA2B,CAAC,KAAyB,EAAE,QAAgB;IAC5E,IAAI,KAAK,KAAK,SAAS,IAAI,CAAC,MAAM,CAAC,QAAQ,CAAC,KAAK,CAAC,IAAI,KAAK,GAAG,CAAC,EAAE,CAAC;QAC9D,OAAO,QAAQ,CAAC;IACpB,CAAC;IAED,OAAO,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC;AAC7B,CAAC"}
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"embedder.d.ts","sourceRoot":"","sources":["../../../src/rag/embeddings/embedder.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EACR,KAAK,EACL,aAAa,EAChB,MAAM,UAAU,CAAC;AAElB,MAAM,WAAW,QAAQ;IACrB,QAAQ,CAAC,KAAK,EAAE,MAAM,CAAC;IACvB,QAAQ,CAAC,UAAU,EAAE,MAAM,CAAC;IAC5B,cAAc,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,OAAO,CAAC,MAAM,EAAE,EAAE,CAAC,CAAC;IACrD,UAAU,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,EAAE,CAAC,CAAC;CAC/C;AAED,qBAAa,uBAAwB,SAAQ,KAAK;IAC9C,QAAQ,CAAC,kBAAkB,EAAE,MAAM,CAAC;IACpC,QAAQ,CAAC,gBAAgB,EAAE,MAAM,CAAC;gBAEtB,kBAAkB,EAAE,MAAM,EAAE,gBAAgB,EAAE,MAAM,EAAE,KAAK,SAAc;CAQxF;AAED,qBAAa,uBAAwB,SAAQ,KAAK;IAC9C,QAAQ,CAAC,YAAY,EAAE,MAAM,CAAC;IAC9B,QAAQ,CAAC,UAAU,EAAE,MAAM,CAAC;gBAEhB,YAAY,EAAE,MAAM,EAAE,UAAU,EAAE,MAAM;CAMvD;AAED,wBAAsB,WAAW,CAC7B,QAAQ,EAAE,QAAQ,EAClB,MAAM,EAAE,KAAK,EAAE,GAChB,OAAO,CAAC,aAAa,EAAE,CAAC,
|
|
1
|
+
{"version":3,"file":"embedder.d.ts","sourceRoot":"","sources":["../../../src/rag/embeddings/embedder.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EACR,KAAK,EACL,aAAa,EAChB,MAAM,UAAU,CAAC;AAElB,MAAM,WAAW,QAAQ;IACrB,QAAQ,CAAC,KAAK,EAAE,MAAM,CAAC;IACvB,QAAQ,CAAC,UAAU,EAAE,MAAM,CAAC;IAC5B,cAAc,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,OAAO,CAAC,MAAM,EAAE,EAAE,CAAC,CAAC;IACrD,UAAU,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,EAAE,CAAC,CAAC;CAC/C;AAED,qBAAa,uBAAwB,SAAQ,KAAK;IAC9C,QAAQ,CAAC,kBAAkB,EAAE,MAAM,CAAC;IACpC,QAAQ,CAAC,gBAAgB,EAAE,MAAM,CAAC;gBAEtB,kBAAkB,EAAE,MAAM,EAAE,gBAAgB,EAAE,MAAM,EAAE,KAAK,SAAc;CAQxF;AAED,qBAAa,uBAAwB,SAAQ,KAAK;IAC9C,QAAQ,CAAC,YAAY,EAAE,MAAM,CAAC;IAC9B,QAAQ,CAAC,UAAU,EAAE,MAAM,CAAC;gBAEhB,YAAY,EAAE,MAAM,EAAE,UAAU,EAAE,MAAM;CAMvD;AAED,wBAAsB,WAAW,CAC7B,QAAQ,EAAE,QAAQ,EAClB,MAAM,EAAE,KAAK,EAAE,GAChB,OAAO,CAAC,aAAa,EAAE,CAAC,CA2B1B;AAED,wBAAgB,yBAAyB,CACrC,SAAS,EAAE,SAAS,MAAM,EAAE,EAC5B,kBAAkB,EAAE,MAAM,EAC1B,KAAK,CAAC,EAAE,MAAM,GACf,IAAI,CAIN;AAED,wBAAgB,wBAAwB,CACpC,UAAU,EAAE,SAAS,OAAO,EAAE,EAC9B,YAAY,EAAE,MAAM,GACrB,IAAI,CAIN;AAED,wBAAgB,cAAc,CAAC,SAAS,EAAE,SAAS,MAAM,EAAE,GAAG,MAAM,EAAE,CAErE;AAED,wBAAgB,kBAAkB,CAAC,SAAS,EAAE,SAAS,MAAM,EAAE,GAAG,MAAM,EAAE,CAUzE;AAED,wBAAgB,8BAA8B,CAAC,UAAU,EAAE,MAAM,GAAG,MAAM,CAMzE"}
|
|
@@ -30,6 +30,12 @@ class EmbeddingBatchSizeError extends Error {
|
|
|
30
30
|
}
|
|
31
31
|
exports.EmbeddingBatchSizeError = EmbeddingBatchSizeError;
|
|
32
32
|
async function embedChunks(embedder, chunks) {
|
|
33
|
+
// 空批次直接返回:replaceExisting 全部 skip 时没有新 chunk,
|
|
34
|
+
// 不应该对远程 embedding 服务发起一次空请求。
|
|
35
|
+
if (chunks.length === 0) {
|
|
36
|
+
return [];
|
|
37
|
+
}
|
|
38
|
+
// TODO 当前只取 content 去 embed,chunk的metadata里还有 headingPath、title等信息,需考虑是否做处理
|
|
33
39
|
const embeddings = await embedder.embedDocuments(chunks.map((chunk) => chunk.content));
|
|
34
40
|
assertEmbeddingBatchSize(embeddings, chunks.length);
|
|
35
41
|
return chunks.map((chunk, index) => {
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"embedder.js","sourceRoot":"","sources":["../../../src/rag/embeddings/embedder.ts"],"names":[],"mappings":";;;AAsCA,
|
|
1
|
+
{"version":3,"file":"embedder.js","sourceRoot":"","sources":["../../../src/rag/embeddings/embedder.ts"],"names":[],"mappings":";;;AAsCA,kCA8BC;AAED,8DAQC;AAED,4DAOC;AAED,wCAEC;AAED,gDAUC;AAED,wEAMC;AAnGD,MAAa,uBAAwB,SAAQ,KAAK;IACrC,kBAAkB,CAAS;IAC3B,gBAAgB,CAAS;IAElC,YAAY,kBAA0B,EAAE,gBAAwB,EAAE,KAAK,GAAG,WAAW;QACjF,KAAK,CACD,GAAG,KAAK,kCAAkC,kBAAkB,SAAS,gBAAgB,GAAG,CAC3F,CAAC;QACF,IAAI,CAAC,IAAI,GAAG,yBAAyB,CAAC;QACtC,IAAI,CAAC,kBAAkB,GAAG,kBAAkB,CAAC;QAC7C,IAAI,CAAC,gBAAgB,GAAG,gBAAgB,CAAC;IAC7C,CAAC;CACJ;AAZD,0DAYC;AAED,MAAa,uBAAwB,SAAQ,KAAK;IACrC,YAAY,CAAS;IACrB,UAAU,CAAS;IAE5B,YAAY,YAAoB,EAAE,UAAkB;QAChD,KAAK,CAAC,2CAA2C,YAAY,SAAS,UAAU,GAAG,CAAC,CAAC;QACrF,IAAI,CAAC,IAAI,GAAG,yBAAyB,CAAC;QACtC,IAAI,CAAC,YAAY,GAAG,YAAY,CAAC;QACjC,IAAI,CAAC,UAAU,GAAG,UAAU,CAAC;IACjC,CAAC;CACJ;AAVD,0DAUC;AAEM,KAAK,UAAU,WAAW,CAC7B,QAAkB,EAClB,MAAe;IAEf,8CAA8C;IAC9C,8BAA8B;IAC9B,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QACtB,OAAO,EAAE,CAAC;IACd,CAAC;IAED,4EAA4E;IAC5E,MAAM,UAAU,GAAG,MAAM,QAAQ,CAAC,cAAc,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC;IACvF,wBAAwB,CAAC,UAAU,EAAE,MAAM,CAAC,MAAM,CAAC,CAAC;IAEpD,OAAO,MAAM,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,KAAK,EAAE,EAAE;QAC/B,MAAM,SAAS,GAAG,UAAU,CAAC,KAAK,CAAC,CAAC;QACpC,IAAI,SAAS,KAAK,SAAS,EAAE,CAAC;YAC1B,MAAM,IAAI,uBAAuB,CAAC,MAAM,CAAC,MAAM,EAAE,UAAU,CAAC,MAAM,CAAC,CAAC;QACxE,CAAC;QACD,yBAAyB,CACrB,SAAS,EACT,QAAQ,CAAC,UAAU,EACnB,SAAS,KAAK,CAAC,EAAE,YAAY,CAChC,CAAC;QAEF,OAAO;YACH,KAAK;YACL,SAAS,EAAE,cAAc,CAAC,SAAS,CAAC;SACvC,CAAC;IACN,CAAC,CAAC,CAAC;AACP,CAAC;AAED,SAAgB,yBAAyB,CACrC,SAA4B,EAC5B,kBAA0B,EAC1B,KAAc;IAEd,IAAI,SAAS,CAAC,MAAM,KAAK,kBAAkB,EAAE,CAAC;QAC1C,MAAM,IAAI,uBAAuB,CAAC,kBAAkB,EAAE,SAAS,CAAC,MAAM,EAAE,KAAK,CAAC,CAAC;IACnF,CAAC;AACL,CAAC;AAED,SAAgB,wBAAwB,CACpC,UAA8B,EAC9B,YAAoB;IAEpB,IAAI,UAAU,CAAC,MAAM,KAAK,YAAY,EAAE,CAAC;QACrC,MAAM,IAAI,uBAAuB,CAAC,YAAY,EAAE,UAAU,CAAC,MAAM,CAAC,CAAC;IACvE,CAAC;AACL,CAAC;AAED,SAAgB,cAAc,CAAC,SAA4B;IACvD,OAAO,CAAC,GAAG,SAAS,CAAC,CAAC;AAC1B,CAAC;AAED,SAAgB,kBAAkB,CAAC,SAA4B;IAC3D,MAAM,SAAS,GAAG,IAAI,CAAC,IAAI,CACvB,SAAS,CAAC,MAAM,CAAC,CAAC,GAAG,EAAE,KAAK,EAAE,EAAE,CAAC,GAAG,GAAG,KAAK,GAAG,KAAK,EAAE,CAAC,CAAC,CAC3D,CAAC;IAEF,IAAI,SAAS,KAAK,CAAC,EAAE,CAAC;QAClB,OAAO,CAAC,GAAG,SAAS,CAAC,CAAC;IAC1B,CAAC;IAED,OAAO,SAAS,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,GAAG,SAAS,CAAC,CAAC;AACvD,CAAC;AAED,SAAgB,8BAA8B,CAAC,UAAkB;IAC7D,IAAI,CAAC,MAAM,CAAC,SAAS,CAAC,UAAU,CAAC,IAAI,UAAU,IAAI,CAAC,EAAE,CAAC;QACnD,MAAM,IAAI,UAAU,CAAC,kDAAkD,CAAC,CAAC;IAC7E,CAAC;IAED,OAAO,UAAU,CAAC;AACtB,CAAC"}
|
|
@@ -1,6 +1,10 @@
|
|
|
1
|
-
import OpenAI from 'openai';
|
|
2
1
|
import type { Embedder } from './embedder';
|
|
3
|
-
|
|
2
|
+
export interface OpenAIEmbeddingRequestBody {
|
|
3
|
+
model: string;
|
|
4
|
+
input: string[];
|
|
5
|
+
encoding_format: 'float';
|
|
6
|
+
dimensions?: number;
|
|
7
|
+
}
|
|
4
8
|
export interface OpenAIEmbeddingResponseLike {
|
|
5
9
|
data: Array<{
|
|
6
10
|
embedding: unknown;
|
|
@@ -9,9 +13,21 @@ export interface OpenAIEmbeddingResponseLike {
|
|
|
9
13
|
}
|
|
10
14
|
export interface OpenAIEmbeddingsClientLike {
|
|
11
15
|
embeddings: {
|
|
12
|
-
create(body:
|
|
16
|
+
create(body: OpenAIEmbeddingRequestBody): Promise<OpenAIEmbeddingResponseLike>;
|
|
13
17
|
};
|
|
14
18
|
}
|
|
19
|
+
export type OpenAIEmbeddingFetchLike = (input: string, init?: {
|
|
20
|
+
method?: string;
|
|
21
|
+
headers?: Record<string, string>;
|
|
22
|
+
body?: string;
|
|
23
|
+
signal?: AbortSignal;
|
|
24
|
+
}) => Promise<OpenAIEmbeddingHttpResponseLike>;
|
|
25
|
+
export interface OpenAIEmbeddingHttpResponseLike {
|
|
26
|
+
ok: boolean;
|
|
27
|
+
status: number;
|
|
28
|
+
statusText: string;
|
|
29
|
+
text(): Promise<string>;
|
|
30
|
+
}
|
|
15
31
|
export interface OpenAICompatibleEmbedderOptions {
|
|
16
32
|
apiKey?: string;
|
|
17
33
|
model: string;
|
|
@@ -19,6 +35,9 @@ export interface OpenAICompatibleEmbedderOptions {
|
|
|
19
35
|
baseURL?: string;
|
|
20
36
|
batchSize?: number;
|
|
21
37
|
sendDimensions?: boolean;
|
|
38
|
+
timeoutMs?: number;
|
|
39
|
+
headers?: Record<string, string>;
|
|
40
|
+
fetch?: OpenAIEmbeddingFetchLike;
|
|
22
41
|
client?: OpenAIEmbeddingsClientLike;
|
|
23
42
|
}
|
|
24
43
|
export interface OpenAIEmbedderOptions extends Omit<OpenAICompatibleEmbedderOptions, 'model'> {
|
|
@@ -42,5 +61,4 @@ export declare class OpenAICompatibleEmbedder implements Embedder {
|
|
|
42
61
|
export declare class OpenAIEmbedder extends OpenAICompatibleEmbedder {
|
|
43
62
|
constructor(options?: OpenAIEmbedderOptions);
|
|
44
63
|
}
|
|
45
|
-
export {};
|
|
46
64
|
//# sourceMappingURL=openai-compatible-embedder.d.ts.map
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"openai-compatible-embedder.d.ts","sourceRoot":"","sources":["../../../src/rag/embeddings/openai-compatible-embedder.ts"],"names":[],"mappings":"
|
|
1
|
+
{"version":3,"file":"openai-compatible-embedder.d.ts","sourceRoot":"","sources":["../../../src/rag/embeddings/openai-compatible-embedder.ts"],"names":[],"mappings":"AAMA,OAAO,KAAK,EACR,QAAQ,EACX,MAAM,YAAY,CAAC;AAEpB,MAAM,WAAW,0BAA0B;IACvC,KAAK,EAAE,MAAM,CAAC;IACd,KAAK,EAAE,MAAM,EAAE,CAAC;IAChB,eAAe,EAAE,OAAO,CAAC;IACzB,UAAU,CAAC,EAAE,MAAM,CAAC;CACvB;AAED,MAAM,WAAW,2BAA2B;IACxC,IAAI,EAAE,KAAK,CAAC;QACR,SAAS,EAAE,OAAO,CAAC;QACnB,KAAK,CAAC,EAAE,MAAM,CAAC;KAClB,CAAC,CAAC;CACN;AAED,MAAM,WAAW,0BAA0B;IACvC,UAAU,EAAE;QACR,MAAM,CAAC,IAAI,EAAE,0BAA0B,GAAG,OAAO,CAAC,2BAA2B,CAAC,CAAC;KAClF,CAAC;CACL;AAED,MAAM,MAAM,wBAAwB,GAAG,CACnC,KAAK,EAAE,MAAM,EACb,IAAI,CAAC,EAAE;IACH,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,OAAO,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;IACjC,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,MAAM,CAAC,EAAE,WAAW,CAAC;CACxB,KACA,OAAO,CAAC,+BAA+B,CAAC,CAAC;AAE9C,MAAM,WAAW,+BAA+B;IAC5C,EAAE,EAAE,OAAO,CAAC;IACZ,MAAM,EAAE,MAAM,CAAC;IACf,UAAU,EAAE,MAAM,CAAC;IACnB,IAAI,IAAI,OAAO,CAAC,MAAM,CAAC,CAAC;CAC3B;AAED,MAAM,WAAW,+BAA+B;IAC5C,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,KAAK,EAAE,MAAM,CAAC;IACd,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB,OAAO,CAAC,EAAE,MAAM,CAAC;IACjB,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB,cAAc,CAAC,EAAE,OAAO,CAAC;IACzB,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB,OAAO,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;IACjC,KAAK,CAAC,EAAE,wBAAwB,CAAC;IACjC,MAAM,CAAC,EAAE,0BAA0B,CAAC;CACvC;AAED,MAAM,WAAW,qBAAsB,SAAQ,IAAI,CAAC,+BAA+B,EAAE,OAAO,CAAC;IACzF,KAAK,CAAC,EAAE,MAAM,CAAC;CAClB;AAYD;;;GAGG;AACH,qBAAa,wBAAyB,YAAW,QAAQ;IACrD,QAAQ,CAAC,KAAK,EAAE,MAAM,CAAC;IACvB,QAAQ,CAAC,UAAU,EAAE,MAAM,CAAC;IAC5B,OAAO,CAAC,QAAQ,CAAC,MAAM,CAA6B;IACpD,OAAO,CAAC,QAAQ,CAAC,SAAS,CAAS;IACnC,OAAO,CAAC,QAAQ,CAAC,cAAc,CAAU;gBAE7B,OAAO,EAAE,+BAA+B;IAQ9C,cAAc,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,OAAO,CAAC,MAAM,EAAE,EAAE,CAAC;IAcpD,UAAU,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,EAAE,CAAC;YAUnC,gBAAgB;CAkBjC;AAED,qBAAa,cAAe,SAAQ,wBAAwB;gBAC5C,OAAO,GAAE,qBAA0B;CAgBlD"}
|
|
@@ -1,13 +1,11 @@
|
|
|
1
1
|
"use strict";
|
|
2
|
-
var __importDefault = (this && this.__importDefault) || function (mod) {
|
|
3
|
-
return (mod && mod.__esModule) ? mod : { "default": mod };
|
|
4
|
-
};
|
|
5
2
|
Object.defineProperty(exports, "__esModule", { value: true });
|
|
6
3
|
exports.OpenAIEmbedder = exports.OpenAICompatibleEmbedder = void 0;
|
|
7
|
-
const openai_1 = __importDefault(require("openai"));
|
|
8
4
|
const embedder_1 = require("./embedder");
|
|
9
5
|
const DEFAULT_OPENAI_EMBEDDING_MODEL = 'text-embedding-3-small';
|
|
6
|
+
const DEFAULT_OPENAI_BASE_URL = 'https://api.openai.com/v1';
|
|
10
7
|
const DEFAULT_BATCH_SIZE = 100;
|
|
8
|
+
const DEFAULT_TIMEOUT_MS = 30000;
|
|
11
9
|
const DEFAULT_OPENAI_EMBEDDING_DIMENSIONS = {
|
|
12
10
|
'text-embedding-3-small': 1536,
|
|
13
11
|
'text-embedding-3-large': 3072,
|
|
@@ -27,9 +25,8 @@ class OpenAICompatibleEmbedder {
|
|
|
27
25
|
this.model = options.model;
|
|
28
26
|
this.dimensions = resolveEmbeddingDimensions(options.model, options.dimensions);
|
|
29
27
|
this.batchSize = resolveBatchSize(options.batchSize);
|
|
30
|
-
this.sendDimensions = options.sendDimensions
|
|
31
|
-
|
|
32
|
-
this.client = options.client ?? createOpenAIClient(options);
|
|
28
|
+
this.sendDimensions = options.sendDimensions ?? options.dimensions !== undefined;
|
|
29
|
+
this.client = options.client ?? createHttpEmbeddingsClient(options);
|
|
33
30
|
}
|
|
34
31
|
async embedDocuments(texts) {
|
|
35
32
|
const embeddings = [];
|
|
@@ -56,6 +53,7 @@ class OpenAICompatibleEmbedder {
|
|
|
56
53
|
const body = {
|
|
57
54
|
model: this.model,
|
|
58
55
|
input: texts,
|
|
56
|
+
encoding_format: 'float',
|
|
59
57
|
};
|
|
60
58
|
if (this.sendDimensions) {
|
|
61
59
|
body.dimensions = this.dimensions;
|
|
@@ -83,17 +81,22 @@ class OpenAIEmbedder extends OpenAICompatibleEmbedder {
|
|
|
83
81
|
}
|
|
84
82
|
}
|
|
85
83
|
exports.OpenAIEmbedder = OpenAIEmbedder;
|
|
86
|
-
function
|
|
84
|
+
function createHttpEmbeddingsClient(options) {
|
|
87
85
|
if (!options.apiKey) {
|
|
88
86
|
throw new Error('OpenAI-compatible embedder requires an apiKey.');
|
|
89
87
|
}
|
|
90
88
|
const clientOptions = {
|
|
91
89
|
apiKey: options.apiKey,
|
|
90
|
+
baseURL: options.baseURL ?? DEFAULT_OPENAI_BASE_URL,
|
|
91
|
+
timeoutMs: options.timeoutMs ?? DEFAULT_TIMEOUT_MS,
|
|
92
92
|
};
|
|
93
|
-
if (options.
|
|
94
|
-
clientOptions.
|
|
93
|
+
if (options.headers !== undefined) {
|
|
94
|
+
clientOptions.headers = options.headers;
|
|
95
|
+
}
|
|
96
|
+
if (options.fetch !== undefined) {
|
|
97
|
+
clientOptions.fetch = options.fetch;
|
|
95
98
|
}
|
|
96
|
-
return new
|
|
99
|
+
return new HttpOpenAIEmbeddingsClient(clientOptions);
|
|
97
100
|
}
|
|
98
101
|
function resolveEmbeddingDimensions(model, dimensions) {
|
|
99
102
|
if (dimensions !== undefined) {
|
|
@@ -114,9 +117,6 @@ function resolveBatchSize(batchSize) {
|
|
|
114
117
|
}
|
|
115
118
|
return batchSize;
|
|
116
119
|
}
|
|
117
|
-
function supportsDimensionsParam(model) {
|
|
118
|
-
return model === 'text-embedding-3-small' || model === 'text-embedding-3-large';
|
|
119
|
-
}
|
|
120
120
|
function toOrderedEmbeddings(response, expectedSize, expectedDimensions) {
|
|
121
121
|
(0, embedder_1.assertEmbeddingBatchSize)(response.data, expectedSize);
|
|
122
122
|
const ordered = new Array(expectedSize);
|
|
@@ -142,4 +142,98 @@ function readEmbedding(value) {
|
|
|
142
142
|
}
|
|
143
143
|
return value;
|
|
144
144
|
}
|
|
145
|
+
class HttpOpenAIEmbeddingsClient {
|
|
146
|
+
embeddings = {
|
|
147
|
+
create: async (body) => {
|
|
148
|
+
const controller = this.timeoutMs > 0 ? new AbortController() : undefined;
|
|
149
|
+
const timeoutId = controller === undefined
|
|
150
|
+
? undefined
|
|
151
|
+
: setTimeout(() => controller.abort(), this.timeoutMs);
|
|
152
|
+
try {
|
|
153
|
+
const request = {
|
|
154
|
+
method: 'POST',
|
|
155
|
+
headers: this.createHeaders(),
|
|
156
|
+
body: JSON.stringify(body),
|
|
157
|
+
};
|
|
158
|
+
if (controller !== undefined) {
|
|
159
|
+
request.signal = controller.signal;
|
|
160
|
+
}
|
|
161
|
+
const response = await this.fetchImpl(this.url, request);
|
|
162
|
+
const text = await response.text();
|
|
163
|
+
if (!response.ok) {
|
|
164
|
+
throw new Error(`Embedding API failed: ${response.status} ${response.statusText}${formatBody(text)}`);
|
|
165
|
+
}
|
|
166
|
+
return parseEmbeddingResponse(text);
|
|
167
|
+
}
|
|
168
|
+
finally {
|
|
169
|
+
if (timeoutId !== undefined) {
|
|
170
|
+
clearTimeout(timeoutId);
|
|
171
|
+
}
|
|
172
|
+
}
|
|
173
|
+
},
|
|
174
|
+
};
|
|
175
|
+
url;
|
|
176
|
+
apiKey;
|
|
177
|
+
timeoutMs;
|
|
178
|
+
headers;
|
|
179
|
+
fetchImpl;
|
|
180
|
+
constructor(options) {
|
|
181
|
+
this.url = joinURL(options.baseURL, '/embeddings');
|
|
182
|
+
this.apiKey = options.apiKey;
|
|
183
|
+
this.timeoutMs = options.timeoutMs;
|
|
184
|
+
this.headers = { ...(options.headers ?? {}) };
|
|
185
|
+
this.fetchImpl = options.fetch ?? getGlobalFetch();
|
|
186
|
+
}
|
|
187
|
+
createHeaders() {
|
|
188
|
+
const headers = {
|
|
189
|
+
'content-type': 'application/json',
|
|
190
|
+
...this.headers,
|
|
191
|
+
};
|
|
192
|
+
const hasAuthorization = Object.keys(headers).some((key) => key.toLowerCase() === 'authorization');
|
|
193
|
+
if (!hasAuthorization) {
|
|
194
|
+
headers.authorization = `Bearer ${this.apiKey}`;
|
|
195
|
+
}
|
|
196
|
+
return headers;
|
|
197
|
+
}
|
|
198
|
+
}
|
|
199
|
+
function parseEmbeddingResponse(text) {
|
|
200
|
+
const parsed = JSON.parse(text);
|
|
201
|
+
if (!isRecord(parsed) || !Array.isArray(parsed.data)) {
|
|
202
|
+
throw new TypeError('Embedding API response must contain a data array.');
|
|
203
|
+
}
|
|
204
|
+
return {
|
|
205
|
+
data: parsed.data.map((item) => {
|
|
206
|
+
if (!isRecord(item)) {
|
|
207
|
+
throw new TypeError('Embedding API response data items must be objects.');
|
|
208
|
+
}
|
|
209
|
+
const embedding = item.embedding;
|
|
210
|
+
const result = { embedding };
|
|
211
|
+
if (item.index !== undefined) {
|
|
212
|
+
if (typeof item.index !== 'number') {
|
|
213
|
+
throw new TypeError('Embedding API response item index must be a number.');
|
|
214
|
+
}
|
|
215
|
+
result.index = item.index;
|
|
216
|
+
}
|
|
217
|
+
return result;
|
|
218
|
+
}),
|
|
219
|
+
};
|
|
220
|
+
}
|
|
221
|
+
function joinURL(baseURL, endpoint) {
|
|
222
|
+
return `${baseURL.replace(/\/+$/, '')}/${endpoint.replace(/^\/+/, '')}`;
|
|
223
|
+
}
|
|
224
|
+
function getGlobalFetch() {
|
|
225
|
+
if (typeof fetch !== 'function') {
|
|
226
|
+
throw new Error('OpenAI-compatible embedder requires fetch support or a custom fetch option.');
|
|
227
|
+
}
|
|
228
|
+
return fetch;
|
|
229
|
+
}
|
|
230
|
+
function isRecord(value) {
|
|
231
|
+
return typeof value === 'object' && value !== null && !Array.isArray(value);
|
|
232
|
+
}
|
|
233
|
+
function formatBody(text) {
|
|
234
|
+
if (text.length === 0) {
|
|
235
|
+
return '';
|
|
236
|
+
}
|
|
237
|
+
return `: ${text}`;
|
|
238
|
+
}
|
|
145
239
|
//# sourceMappingURL=openai-compatible-embedder.js.map
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"openai-compatible-embedder.js","sourceRoot":"","sources":["../../../src/rag/embeddings/openai-compatible-embedder.ts"],"names":[],"mappings":"
|
|
1
|
+
{"version":3,"file":"openai-compatible-embedder.js","sourceRoot":"","sources":["../../../src/rag/embeddings/openai-compatible-embedder.ts"],"names":[],"mappings":";;;AAAA,yCAKoB;AA2DpB,MAAM,8BAA8B,GAAG,wBAAwB,CAAC;AAChE,MAAM,uBAAuB,GAAG,2BAA2B,CAAC;AAC5D,MAAM,kBAAkB,GAAG,GAAG,CAAC;AAC/B,MAAM,kBAAkB,GAAG,KAAK,CAAC;AACjC,MAAM,mCAAmC,GAA2B;IAChE,wBAAwB,EAAE,IAAI;IAC9B,wBAAwB,EAAE,IAAI;IAC9B,wBAAwB,EAAE,IAAI;CACjC,CAAC;AAEF;;;GAGG;AACH,MAAa,wBAAwB;IACxB,KAAK,CAAS;IACd,UAAU,CAAS;IACX,MAAM,CAA6B;IACnC,SAAS,CAAS;IAClB,cAAc,CAAU;IAEzC,YAAY,OAAwC;QAChD,IAAI,CAAC,KAAK,GAAG,OAAO,CAAC,KAAK,CAAC;QAC3B,IAAI,CAAC,UAAU,GAAG,0BAA0B,CAAC,OAAO,CAAC,KAAK,EAAE,OAAO,CAAC,UAAU,CAAC,CAAC;QAChF,IAAI,CAAC,SAAS,GAAG,gBAAgB,CAAC,OAAO,CAAC,SAAS,CAAC,CAAC;QACrD,IAAI,CAAC,cAAc,GAAG,OAAO,CAAC,cAAc,IAAI,OAAO,CAAC,UAAU,KAAK,SAAS,CAAC;QACjF,IAAI,CAAC,MAAM,GAAG,OAAO,CAAC,MAAM,IAAI,0BAA0B,CAAC,OAAO,CAAC,CAAC;IACxE,CAAC;IAED,KAAK,CAAC,cAAc,CAAC,KAAe;QAChC,MAAM,UAAU,GAAe,EAAE,CAAC;QAElC,KAAK,IAAI,KAAK,GAAG,CAAC,EAAE,KAAK,GAAG,KAAK,CAAC,MAAM,EAAE,KAAK,IAAI,IAAI,CAAC,SAAS,EAAE,CAAC;YAChE,MAAM,KAAK,GAAG,KAAK,CAAC,KAAK,CAAC,KAAK,EAAE,KAAK,GAAG,IAAI,CAAC,SAAS,CAAC,CAAC;YACzD,MAAM,eAAe,GAAG,MAAM,IAAI,CAAC,gBAAgB,CAAC,KAAK,CAAC,CAAC;YAC3D,UAAU,CAAC,IAAI,CAAC,GAAG,eAAe,CAAC,CAAC;QACxC,CAAC;QAED,IAAA,mCAAwB,EAAC,UAAU,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC;QAEnD,OAAO,UAAU,CAAC,GAAG,CAAC,CAAC,SAAS,EAAE,EAAE,CAAC,IAAA,yBAAc,EAAC,SAAS,CAAC,CAAC,CAAC;IACpE,CAAC;IAED,KAAK,CAAC,UAAU,CAAC,IAAY;QACzB,MAAM,UAAU,GAAG,MAAM,IAAI,CAAC,cAAc,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC;QACrD,MAAM,SAAS,GAAG,UAAU,CAAC,CAAC,CAAC,CAAC;QAChC,IAAI,SAAS,KAAK,SAAS,EAAE,CAAC;YAC1B,MAAM,IAAI,KAAK,CAAC,+DAA+D,CAAC,CAAC;QACrF,CAAC;QAED,OAAO,IAAA,yBAAc,EAAC,SAAS,CAAC,CAAC;IACrC,CAAC;IAEO,KAAK,CAAC,gBAAgB,CAAC,KAAe;QAC1C,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;YACrB,OAAO,EAAE,CAAC;QACd,CAAC;QAED,MAAM,IAAI,GAA+B;YACrC,KAAK,EAAE,IAAI,CAAC,KAAK;YACjB,KAAK,EAAE,KAAK;YACZ,eAAe,EAAE,OAAO;SAC3B,CAAC;QACF,IAAI,IAAI,CAAC,cAAc,EAAE,CAAC;YACtB,IAAI,CAAC,UAAU,GAAG,IAAI,CAAC,UAAU,CAAC;QACtC,CAAC;QAED,MAAM,QAAQ,GAAG,MAAM,IAAI,CAAC,MAAM,CAAC,UAAU,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC;QAE3D,OAAO,mBAAmB,CAAC,QAAQ,EAAE,KAAK,CAAC,MAAM,EAAE,IAAI,CAAC,UAAU,CAAC,CAAC;IACxE,CAAC;CACJ;AAzDD,4DAyDC;AAED,MAAa,cAAe,SAAQ,wBAAwB;IACxD,YAAY,UAAiC,EAAE;QAC3C,MAAM,eAAe,GAAoC;YACrD,GAAG,OAAO;YACV,KAAK,EAAE,OAAO,CAAC,KAAK,IAAI,OAAO,CAAC,GAAG,CAAC,sBAAsB,IAAI,8BAA8B;SAC/F,CAAC;QACF,MAAM,MAAM,GAAG,OAAO,CAAC,MAAM,IAAI,OAAO,CAAC,GAAG,CAAC,cAAc,CAAC;QAC5D,IAAI,MAAM,KAAK,SAAS,EAAE,CAAC;YACvB,eAAe,CAAC,MAAM,GAAG,MAAM,CAAC;QACpC,CAAC;QACD,MAAM,OAAO,GAAG,OAAO,CAAC,OAAO,IAAI,OAAO,CAAC,GAAG,CAAC,eAAe,CAAC;QAC/D,IAAI,OAAO,KAAK,SAAS,EAAE,CAAC;YACxB,eAAe,CAAC,OAAO,GAAG,OAAO,CAAC;QACtC,CAAC;QAED,KAAK,CAAC,eAAe,CAAC,CAAC;IAC3B,CAAC;CACJ;AAjBD,wCAiBC;AAED,SAAS,0BAA0B,CAAC,OAAwC;IACxE,IAAI,CAAC,OAAO,CAAC,MAAM,EAAE,CAAC;QAClB,MAAM,IAAI,KAAK,CAAC,gDAAgD,CAAC,CAAC;IACtE,CAAC;IAED,MAAM,aAAa,GAAsC;QACrD,MAAM,EAAE,OAAO,CAAC,MAAM;QACtB,OAAO,EAAE,OAAO,CAAC,OAAO,IAAI,uBAAuB;QACnD,SAAS,EAAE,OAAO,CAAC,SAAS,IAAI,kBAAkB;KACrD,CAAC;IACF,IAAI,OAAO,CAAC,OAAO,KAAK,SAAS,EAAE,CAAC;QAChC,aAAa,CAAC,OAAO,GAAG,OAAO,CAAC,OAAO,CAAC;IAC5C,CAAC;IACD,IAAI,OAAO,CAAC,KAAK,KAAK,SAAS,EAAE,CAAC;QAC9B,aAAa,CAAC,KAAK,GAAG,OAAO,CAAC,KAAK,CAAC;IACxC,CAAC;IAED,OAAO,IAAI,0BAA0B,CAAC,aAAa,CAAC,CAAC;AACzD,CAAC;AAED,SAAS,0BAA0B,CAAC,KAAa,EAAE,UAA8B;IAC7E,IAAI,UAAU,KAAK,SAAS,EAAE,CAAC;QAC3B,OAAO,IAAA,yCAA8B,EAAC,UAAU,CAAC,CAAC;IACtD,CAAC;IAED,MAAM,eAAe,GAAG,mCAAmC,CAAC,KAAK,CAAC,CAAC;IACnE,IAAI,eAAe,KAAK,SAAS,EAAE,CAAC;QAChC,OAAO,eAAe,CAAC;IAC3B,CAAC;IAED,MAAM,IAAI,KAAK,CACX,+EAA+E,KAAK,IAAI,CAC3F,CAAC;AACN,CAAC;AAED,SAAS,gBAAgB,CAAC,SAA6B;IACnD,IAAI,SAAS,KAAK,SAAS,EAAE,CAAC;QAC1B,OAAO,kBAAkB,CAAC;IAC9B,CAAC;IACD,IAAI,CAAC,MAAM,CAAC,SAAS,CAAC,SAAS,CAAC,IAAI,SAAS,IAAI,CAAC,EAAE,CAAC;QACjD,MAAM,IAAI,UAAU,CAAC,iDAAiD,CAAC,CAAC;IAC5E,CAAC;IAED,OAAO,SAAS,CAAC;AACrB,CAAC;AAED,SAAS,mBAAmB,CACxB,QAAqC,EACrC,YAAoB,EACpB,kBAA0B;IAE1B,IAAA,mCAAwB,EAAC,QAAQ,CAAC,IAAI,EAAE,YAAY,CAAC,CAAC;IAEtD,MAAM,OAAO,GAAG,IAAI,KAAK,CAAuB,YAAY,CAAC,CAAC;IAC9D,QAAQ,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,IAAI,EAAE,MAAM,EAAE,EAAE;QACnC,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,IAAI,MAAM,CAAC;QACnC,IAAI,CAAC,MAAM,CAAC,SAAS,CAAC,KAAK,CAAC,IAAI,KAAK,GAAG,CAAC,IAAI,KAAK,IAAI,YAAY,EAAE,CAAC;YACjE,MAAM,IAAI,UAAU,CAAC,0CAA0C,MAAM,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;QACrF,CAAC;QAED,MAAM,SAAS,GAAG,aAAa,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC;QAChD,IAAA,oCAAyB,EAAC,SAAS,EAAE,kBAAkB,EAAE,2BAA2B,KAAK,EAAE,CAAC,CAAC;QAC7F,OAAO,CAAC,KAAK,CAAC,GAAG,IAAA,yBAAc,EAAC,SAAS,CAAC,CAAC;IAC/C,CAAC,CAAC,CAAC;IAEH,OAAO,OAAO,CAAC,GAAG,CAAC,CAAC,SAAS,EAAE,KAAK,EAAE,EAAE;QACpC,IAAI,SAAS,KAAK,SAAS,EAAE,CAAC;YAC1B,MAAM,IAAI,KAAK,CAAC,4CAA4C,KAAK,GAAG,CAAC,CAAC;QAC1E,CAAC;QAED,OAAO,IAAA,yBAAc,EAAC,SAAS,CAAC,CAAC;IACrC,CAAC,CAAC,CAAC;AACP,CAAC;AAED,SAAS,aAAa,CAAC,KAAc;IACjC,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,KAAK,CAAC,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,OAAO,IAAI,KAAK,QAAQ,CAAC,EAAE,CAAC;QAC5E,MAAM,IAAI,SAAS,CAAC,iEAAiE,CAAC,CAAC;IAC3F,CAAC;IAED,OAAO,KAAK,CAAC;AACjB,CAAC;AAUD,MAAM,0BAA0B;IACnB,UAAU,GAAG;QAClB,MAAM,EAAE,KAAK,EAAE,IAAgC,EAAwC,EAAE;YACrF,MAAM,UAAU,GAAG,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,eAAe,EAAE,CAAC,CAAC,CAAC,SAAS,CAAC;YAC1E,MAAM,SAAS,GAAG,UAAU,KAAK,SAAS;gBACtC,CAAC,CAAC,SAAS;gBACX,CAAC,CAAC,UAAU,CAAC,GAAG,EAAE,CAAC,UAAU,CAAC,KAAK,EAAE,EAAE,IAAI,CAAC,SAAS,CAAC,CAAC;YAE3D,IAAI,CAAC;gBACD,MAAM,OAAO,GAKT;oBACA,MAAM,EAAE,MAAM;oBACd,OAAO,EAAE,IAAI,CAAC,aAAa,EAAE;oBAC7B,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC,IAAI,CAAC;iBAC7B,CAAC;gBACF,IAAI,UAAU,KAAK,SAAS,EAAE,CAAC;oBAC3B,OAAO,CAAC,MAAM,GAAG,UAAU,CAAC,MAAM,CAAC;gBACvC,CAAC;gBAED,MAAM,QAAQ,GAAG,MAAM,IAAI,CAAC,SAAS,CAAC,IAAI,CAAC,GAAG,EAAE,OAAO,CAAC,CAAC;gBACzD,MAAM,IAAI,GAAG,MAAM,QAAQ,CAAC,IAAI,EAAE,CAAC;gBACnC,IAAI,CAAC,QAAQ,CAAC,EAAE,EAAE,CAAC;oBACf,MAAM,IAAI,KAAK,CACX,yBAAyB,QAAQ,CAAC,MAAM,IAAI,QAAQ,CAAC,UAAU,GAAG,UAAU,CAAC,IAAI,CAAC,EAAE,CACvF,CAAC;gBACN,CAAC;gBAED,OAAO,sBAAsB,CAAC,IAAI,CAAC,CAAC;YACxC,CAAC;oBAAS,CAAC;gBACP,IAAI,SAAS,KAAK,SAAS,EAAE,CAAC;oBAC1B,YAAY,CAAC,SAAS,CAAC,CAAC;gBAC5B,CAAC;YACL,CAAC;QACL,CAAC;KACJ,CAAC;IAEe,GAAG,CAAS;IACZ,MAAM,CAAS;IACf,SAAS,CAAS;IAClB,OAAO,CAAyB;IAChC,SAAS,CAA2B;IAErD,YAAY,OAA0C;QAClD,IAAI,CAAC,GAAG,GAAG,OAAO,CAAC,OAAO,CAAC,OAAO,EAAE,aAAa,CAAC,CAAC;QACnD,IAAI,CAAC,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC;QAC7B,IAAI,CAAC,SAAS,GAAG,OAAO,CAAC,SAAS,CAAC;QACnC,IAAI,CAAC,OAAO,GAAG,EAAE,GAAG,CAAC,OAAO,CAAC,OAAO,IAAI,EAAE,CAAC,EAAE,CAAC;QAC9C,IAAI,CAAC,SAAS,GAAG,OAAO,CAAC,KAAK,IAAI,cAAc,EAAE,CAAC;IACvD,CAAC;IAEO,aAAa;QACjB,MAAM,OAAO,GAA2B;YACpC,cAAc,EAAE,kBAAkB;YAClC,GAAG,IAAI,CAAC,OAAO;SAClB,CAAC;QAEF,MAAM,gBAAgB,GAAG,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,IAAI,CAC9C,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,WAAW,EAAE,KAAK,eAAe,CACjD,CAAC;QACF,IAAI,CAAC,gBAAgB,EAAE,CAAC;YACpB,OAAO,CAAC,aAAa,GAAG,UAAU,IAAI,CAAC,MAAM,EAAE,CAAC;QACpD,CAAC;QAED,OAAO,OAAO,CAAC;IACnB,CAAC;CACJ;AAED,SAAS,sBAAsB,CAAC,IAAY;IACxC,MAAM,MAAM,GAAY,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;IACzC,IAAI,CAAC,QAAQ,CAAC,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,MAAM,CAAC,IAAI,CAAC,EAAE,CAAC;QACnD,MAAM,IAAI,SAAS,CAAC,mDAAmD,CAAC,CAAC;IAC7E,CAAC;IAED,OAAO;QACH,IAAI,EAAE,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE;YAC3B,IAAI,CAAC,QAAQ,CAAC,IAAI,CAAC,EAAE,CAAC;gBAClB,MAAM,IAAI,SAAS,CAAC,oDAAoD,CAAC,CAAC;YAC9E,CAAC;YAED,MAAM,SAAS,GAAG,IAAI,CAAC,SAAS,CAAC;YACjC,MAAM,MAAM,GAAgD,EAAE,SAAS,EAAE,CAAC;YAC1E,IAAI,IAAI,CAAC,KAAK,KAAK,SAAS,EAAE,CAAC;gBAC3B,IAAI,OAAO,IAAI,CAAC,KAAK,KAAK,QAAQ,EAAE,CAAC;oBACjC,MAAM,IAAI,SAAS,CAAC,qDAAqD,CAAC,CAAC;gBAC/E,CAAC;gBACD,MAAM,CAAC,KAAK,GAAG,IAAI,CAAC,KAAK,CAAC;YAC9B,CAAC;YAED,OAAO,MAAM,CAAC;QAClB,CAAC,CAAC;KACL,CAAC;AACN,CAAC;AAED,SAAS,OAAO,CAAC,OAAe,EAAE,QAAgB;IAC9C,OAAO,GAAG,OAAO,CAAC,OAAO,CAAC,MAAM,EAAE,EAAE,CAAC,IAAI,QAAQ,CAAC,OAAO,CAAC,MAAM,EAAE,EAAE,CAAC,EAAE,CAAC;AAC5E,CAAC;AAED,SAAS,cAAc;IACnB,IAAI,OAAO,KAAK,KAAK,UAAU,EAAE,CAAC;QAC9B,MAAM,IAAI,KAAK,CAAC,6EAA6E,CAAC,CAAC;IACnG,CAAC;IAED,OAAO,KAAK,CAAC;AACjB,CAAC;AAED,SAAS,QAAQ,CAAC,KAAc;IAC5B,OAAO,OAAO,KAAK,KAAK,QAAQ,IAAI,KAAK,KAAK,IAAI,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC;AAChF,CAAC;AAED,SAAS,UAAU,CAAC,IAAY;IAC5B,IAAI,IAAI,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QACpB,OAAO,EAAE,CAAC;IACd,CAAC;IAED,OAAO,KAAK,IAAI,EAAE,CAAC;AACvB,CAAC"}
|
|
@@ -6,11 +6,18 @@ export interface DefaultContextBuilderOptions {
|
|
|
6
6
|
maxContextTokens?: number;
|
|
7
7
|
maxChunkCharacters?: number;
|
|
8
8
|
includeScores?: boolean;
|
|
9
|
+
/**
|
|
10
|
+
* 融合分数尺度上的过滤阈值(如 RRF)。不要复用检索的 minScore:
|
|
11
|
+
* 那是 retriever 原始分数尺度(cosine / TF-IDF),量级和融合分完全不同。
|
|
12
|
+
* 默认 undefined,不过滤。
|
|
13
|
+
*/
|
|
14
|
+
minFusedScore?: number;
|
|
9
15
|
}
|
|
10
16
|
export declare class DefaultContextBuilder implements ContextBuilder {
|
|
11
17
|
private readonly maxContextTokens;
|
|
12
18
|
private readonly maxChunkCharacters;
|
|
13
19
|
private readonly includeScores;
|
|
20
|
+
private readonly minFusedScore;
|
|
14
21
|
constructor(options?: DefaultContextBuilderOptions);
|
|
15
22
|
build(query: string, retrieved: RetrievedChunk[], options?: AnswerOptions): Promise<RagContext>;
|
|
16
23
|
}
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"context-builder.d.ts","sourceRoot":"","sources":["../../../src/rag/generation/context-builder.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EACR,aAAa,EAEb,UAAU,EAEV,cAAc,EACjB,MAAM,UAAU,CAAC;AAElB,MAAM,WAAW,cAAc;IAC3B,KAAK,CACD,KAAK,EAAE,MAAM,EACb,SAAS,EAAE,cAAc,EAAE,EAC3B,OAAO,CAAC,EAAE,aAAa,GACxB,OAAO,CAAC,UAAU,CAAC,CAAC;CAC1B;AAED,MAAM,WAAW,4BAA4B;IACzC,gBAAgB,CAAC,EAAE,MAAM,CAAC;IAC1B,kBAAkB,CAAC,EAAE,MAAM,CAAC;IAC5B,aAAa,CAAC,EAAE,OAAO,CAAC;
|
|
1
|
+
{"version":3,"file":"context-builder.d.ts","sourceRoot":"","sources":["../../../src/rag/generation/context-builder.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EACR,aAAa,EAEb,UAAU,EAEV,cAAc,EACjB,MAAM,UAAU,CAAC;AAElB,MAAM,WAAW,cAAc;IAC3B,KAAK,CACD,KAAK,EAAE,MAAM,EACb,SAAS,EAAE,cAAc,EAAE,EAC3B,OAAO,CAAC,EAAE,aAAa,GACxB,OAAO,CAAC,UAAU,CAAC,CAAC;CAC1B;AAED,MAAM,WAAW,4BAA4B;IACzC,gBAAgB,CAAC,EAAE,MAAM,CAAC;IAC1B,kBAAkB,CAAC,EAAE,MAAM,CAAC;IAC5B,aAAa,CAAC,EAAE,OAAO,CAAC;IACxB;;;;OAIG;IACH,aAAa,CAAC,EAAE,MAAM,CAAC;CAC1B;AAED,qBAAa,qBAAsB,YAAW,cAAc;IACxD,OAAO,CAAC,QAAQ,CAAC,gBAAgB,CAAS;IAC1C,OAAO,CAAC,QAAQ,CAAC,kBAAkB,CAAqB;IACxD,OAAO,CAAC,QAAQ,CAAC,aAAa,CAAU;IACxC,OAAO,CAAC,QAAQ,CAAC,aAAa,CAAqB;gBAEvC,OAAO,GAAE,4BAAiC;IAOhD,KAAK,CACP,KAAK,EAAE,MAAM,EACb,SAAS,EAAE,cAAc,EAAE,EAC3B,OAAO,GAAE,aAAkB,GAC5B,OAAO,CAAC,UAAU,CAAC;CAyDzB;AAqHD;;;GAGG;AACH,wBAAgB,cAAc,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,CAKnD"}
|