@zvec/zvec-grep 0.1.0
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/LICENSE +201 -0
- package/README.md +222 -0
- package/README_CN.md +222 -0
- package/dist/cli/args.d.ts +2 -0
- package/dist/cli/args.js +611 -0
- package/dist/cli/args.js.map +1 -0
- package/dist/cli/commands.d.ts +2 -0
- package/dist/cli/commands.js +270 -0
- package/dist/cli/commands.js.map +1 -0
- package/dist/cli/errors.d.ts +6 -0
- package/dist/cli/errors.js +97 -0
- package/dist/cli/errors.js.map +1 -0
- package/dist/cli/format/context.d.ts +5 -0
- package/dist/cli/format/context.js +490 -0
- package/dist/cli/format/context.js.map +1 -0
- package/dist/cli/format/debug.d.ts +4 -0
- package/dist/cli/format/debug.js +42 -0
- package/dist/cli/format/debug.js.map +1 -0
- package/dist/cli/format/highlight.d.ts +3 -0
- package/dist/cli/format/highlight.js +106 -0
- package/dist/cli/format/highlight.js.map +1 -0
- package/dist/cli/format/progress.d.ts +9 -0
- package/dist/cli/format/progress.js +92 -0
- package/dist/cli/format/progress.js.map +1 -0
- package/dist/cli/format/range.d.ts +3 -0
- package/dist/cli/format/range.js +24 -0
- package/dist/cli/format/range.js.map +1 -0
- package/dist/cli/format/status.d.ts +7 -0
- package/dist/cli/format/status.js +200 -0
- package/dist/cli/format/status.js.map +1 -0
- package/dist/cli/help.d.ts +1 -0
- package/dist/cli/help.js +88 -0
- package/dist/cli/help.js.map +1 -0
- package/dist/cli/index.d.ts +2 -0
- package/dist/cli/index.js +32 -0
- package/dist/cli/index.js.map +1 -0
- package/dist/cli/types.d.ts +56 -0
- package/dist/cli/types.js +10 -0
- package/dist/cli/types.js.map +1 -0
- package/dist/cli/version.d.ts +1 -0
- package/dist/cli/version.js +11 -0
- package/dist/cli/version.js.map +1 -0
- package/dist/engine/collection/index.d.ts +50 -0
- package/dist/engine/collection/index.js +388 -0
- package/dist/engine/collection/index.js.map +1 -0
- package/dist/engine/errors/details.d.ts +6 -0
- package/dist/engine/errors/details.js +27 -0
- package/dist/engine/errors/details.js.map +1 -0
- package/dist/engine/errors/index.d.ts +14 -0
- package/dist/engine/errors/index.js +16 -0
- package/dist/engine/errors/index.js.map +1 -0
- package/dist/engine/models/catalog.d.ts +47 -0
- package/dist/engine/models/catalog.js +68 -0
- package/dist/engine/models/catalog.js.map +1 -0
- package/dist/engine/models/embeddings.d.ts +25 -0
- package/dist/engine/models/embeddings.js +100 -0
- package/dist/engine/models/embeddings.js.map +1 -0
- package/dist/engine/models/factory.d.ts +5 -0
- package/dist/engine/models/factory.js +66 -0
- package/dist/engine/models/factory.js.map +1 -0
- package/dist/engine/models/index.d.ts +6 -0
- package/dist/engine/models/index.js +5 -0
- package/dist/engine/models/index.js.map +1 -0
- package/dist/engine/models/providers/index.d.ts +2 -0
- package/dist/engine/models/providers/index.js +3 -0
- package/dist/engine/models/providers/index.js.map +1 -0
- package/dist/engine/models/providers/llama-cpp/embedding.d.ts +94 -0
- package/dist/engine/models/providers/llama-cpp/embedding.js +459 -0
- package/dist/engine/models/providers/llama-cpp/embedding.js.map +1 -0
- package/dist/engine/models/providers/llama-cpp/index.d.ts +1 -0
- package/dist/engine/models/providers/llama-cpp/index.js +2 -0
- package/dist/engine/models/providers/llama-cpp/index.js.map +1 -0
- package/dist/engine/models/providers/qwen/embedding.d.ts +38 -0
- package/dist/engine/models/providers/qwen/embedding.js +322 -0
- package/dist/engine/models/providers/qwen/embedding.js.map +1 -0
- package/dist/engine/models/providers/qwen/index.d.ts +1 -0
- package/dist/engine/models/providers/qwen/index.js +2 -0
- package/dist/engine/models/providers/qwen/index.js.map +1 -0
- package/dist/engine/models/ranking.d.ts +15 -0
- package/dist/engine/models/ranking.js +2 -0
- package/dist/engine/models/ranking.js.map +1 -0
- package/dist/engine/models/types.d.ts +30 -0
- package/dist/engine/models/types.js +2 -0
- package/dist/engine/models/types.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/adapter.d.ts +19 -0
- package/dist/engine/pipeline/indexing/extractor/code/adapter.js +24 -0
- package/dist/engine/pipeline/indexing/extractor/code/adapter.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/extractor.d.ts +14 -0
- package/dist/engine/pipeline/indexing/extractor/code/extractor.js +594 -0
- package/dist/engine/pipeline/indexing/extractor/code/extractor.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/families/c-family.d.ts +2 -0
- package/dist/engine/pipeline/indexing/extractor/code/families/c-family.js +116 -0
- package/dist/engine/pipeline/indexing/extractor/code/families/c-family.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/families/js-ts.d.ts +12 -0
- package/dist/engine/pipeline/indexing/extractor/code/families/js-ts.js +112 -0
- package/dist/engine/pipeline/indexing/extractor/code/families/js-ts.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/families/metadata.d.ts +8 -0
- package/dist/engine/pipeline/indexing/extractor/code/families/metadata.js +83 -0
- package/dist/engine/pipeline/indexing/extractor/code/families/metadata.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/families/name-field.d.ts +3 -0
- package/dist/engine/pipeline/indexing/extractor/code/families/name-field.js +22 -0
- package/dist/engine/pipeline/indexing/extractor/code/families/name-field.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/c.d.ts +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/c.js +15 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/c.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/cpp.d.ts +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/cpp.js +18 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/cpp.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/go.d.ts +2 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/go.js +72 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/go.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/java.d.ts +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/java.js +17 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/java.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/javascript.d.ts +2 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/javascript.js +25 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/javascript.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/python.d.ts +2 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/python.js +53 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/python.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/rust.d.ts +2 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/rust.js +29 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/rust.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/typescript.d.ts +2 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/typescript.js +37 -0
- package/dist/engine/pipeline/indexing/extractor/code/languages/typescript.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/index.d.ts +44 -0
- package/dist/engine/pipeline/indexing/extractor/index.js +281 -0
- package/dist/engine/pipeline/indexing/extractor/index.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/markdown/extractor.d.ts +13 -0
- package/dist/engine/pipeline/indexing/extractor/markdown/extractor.js +339 -0
- package/dist/engine/pipeline/indexing/extractor/markdown/extractor.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/tree-sitter/grammar.d.ts +4 -0
- package/dist/engine/pipeline/indexing/extractor/tree-sitter/grammar.js +49 -0
- package/dist/engine/pipeline/indexing/extractor/tree-sitter/grammar.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/tree-sitter/nodes.d.ts +3 -0
- package/dist/engine/pipeline/indexing/extractor/tree-sitter/nodes.js +26 -0
- package/dist/engine/pipeline/indexing/extractor/tree-sitter/nodes.js.map +1 -0
- package/dist/engine/pipeline/indexing/extractor/tree-sitter/parser.d.ts +2 -0
- package/dist/engine/pipeline/indexing/extractor/tree-sitter/parser.js +31 -0
- package/dist/engine/pipeline/indexing/extractor/tree-sitter/parser.js.map +1 -0
- package/dist/engine/pipeline/indexing/ids.d.ts +1 -0
- package/dist/engine/pipeline/indexing/ids.js +5 -0
- package/dist/engine/pipeline/indexing/ids.js.map +1 -0
- package/dist/engine/pipeline/indexing/index.d.ts +12 -0
- package/dist/engine/pipeline/indexing/index.js +538 -0
- package/dist/engine/pipeline/indexing/index.js.map +1 -0
- package/dist/engine/pipeline/indexing/root-paths.d.ts +7 -0
- package/dist/engine/pipeline/indexing/root-paths.js +151 -0
- package/dist/engine/pipeline/indexing/root-paths.js.map +1 -0
- package/dist/engine/pipeline/indexing/scanner/index.d.ts +5 -0
- package/dist/engine/pipeline/indexing/scanner/index.js +504 -0
- package/dist/engine/pipeline/indexing/scanner/index.js.map +1 -0
- package/dist/engine/pipeline/search/index.d.ts +12 -0
- package/dist/engine/pipeline/search/index.js +761 -0
- package/dist/engine/pipeline/search/index.js.map +1 -0
- package/dist/engine/service/index.d.ts +2 -0
- package/dist/engine/service/index.js +2 -0
- package/dist/engine/service/index.js.map +1 -0
- package/dist/engine/service/lexical.d.ts +29 -0
- package/dist/engine/service/lexical.js +482 -0
- package/dist/engine/service/lexical.js.map +1 -0
- package/dist/engine/service/root.d.ts +15 -0
- package/dist/engine/service/root.js +44 -0
- package/dist/engine/service/root.js.map +1 -0
- package/dist/engine/service/types.d.ts +157 -0
- package/dist/engine/service/types.js +2 -0
- package/dist/engine/service/types.js.map +1 -0
- package/dist/engine/service/zvec-grep.d.ts +2 -0
- package/dist/engine/service/zvec-grep.js +935 -0
- package/dist/engine/service/zvec-grep.js.map +1 -0
- package/dist/engine/storage/index.d.ts +42 -0
- package/dist/engine/storage/index.js +2 -0
- package/dist/engine/storage/index.js.map +1 -0
- package/dist/engine/storage/zvec.d.ts +51 -0
- package/dist/engine/storage/zvec.js +706 -0
- package/dist/engine/storage/zvec.js.map +1 -0
- package/dist/engine/types.d.ts +266 -0
- package/dist/engine/types.js +14 -0
- package/dist/engine/types.js.map +1 -0
- package/dist/engine/utils/glob.d.ts +6 -0
- package/dist/engine/utils/glob.js +110 -0
- package/dist/engine/utils/glob.js.map +1 -0
- package/dist/engine/utils/hash.d.ts +2 -0
- package/dist/engine/utils/hash.js +8 -0
- package/dist/engine/utils/hash.js.map +1 -0
- package/dist/engine/utils/json.d.ts +4 -0
- package/dist/engine/utils/json.js +45 -0
- package/dist/engine/utils/json.js.map +1 -0
- package/dist/engine/utils/lock.d.ts +19 -0
- package/dist/engine/utils/lock.js +254 -0
- package/dist/engine/utils/lock.js.map +1 -0
- package/dist/engine/utils/path.d.ts +4 -0
- package/dist/engine/utils/path.js +18 -0
- package/dist/engine/utils/path.js.map +1 -0
- package/dist/index.d.ts +6 -0
- package/dist/index.js +4 -0
- package/dist/index.js.map +1 -0
- package/package.json +47 -0
- package/skills/zvec-grep/SKILL.md +141 -0
- package/skills/zvec-grep/agents/openai.yaml +4 -0
|
@@ -0,0 +1,339 @@
|
|
|
1
|
+
import { makeEntityId } from "../../ids.js";
|
|
2
|
+
const DEFAULT_MARKDOWN_CHUNK_CHARS = 3600;
|
|
3
|
+
const DEFAULT_MARKDOWN_CHUNK_OVERLAP_CHARS = 540;
|
|
4
|
+
export class MarkdownExtractor {
|
|
5
|
+
maxChunkChars;
|
|
6
|
+
chunkOverlapChars;
|
|
7
|
+
constructor(options = {}) {
|
|
8
|
+
this.maxChunkChars = options.maxChunkChars ?? DEFAULT_MARKDOWN_CHUNK_CHARS;
|
|
9
|
+
this.chunkOverlapChars = options.chunkOverlapChars ?? DEFAULT_MARKDOWN_CHUNK_OVERLAP_CHARS;
|
|
10
|
+
}
|
|
11
|
+
supports(source) {
|
|
12
|
+
return source.kind === "text" && source.file.format === "markdown";
|
|
13
|
+
}
|
|
14
|
+
async extract(source) {
|
|
15
|
+
if (!this.supports(source) || source.kind !== "text") {
|
|
16
|
+
return [];
|
|
17
|
+
}
|
|
18
|
+
const lines = source.text.split("\n");
|
|
19
|
+
const headings = scanHeadings(lines);
|
|
20
|
+
if (headings.length === 0) {
|
|
21
|
+
return [];
|
|
22
|
+
}
|
|
23
|
+
const lineOffsets = computeLineOffsets(lines);
|
|
24
|
+
const fenceLines = computeFenceLines(lines);
|
|
25
|
+
const sections = buildSections(headings, lines);
|
|
26
|
+
const fragments = [];
|
|
27
|
+
for (const section of sections) {
|
|
28
|
+
const metadata = markdownMetadata(section);
|
|
29
|
+
const windows = splitMarkdownSection({
|
|
30
|
+
lines,
|
|
31
|
+
lineOffsets,
|
|
32
|
+
fenceLines,
|
|
33
|
+
section,
|
|
34
|
+
maxChars: this.maxChunkChars,
|
|
35
|
+
overlapChars: this.chunkOverlapChars,
|
|
36
|
+
});
|
|
37
|
+
if (windows.length > 1) {
|
|
38
|
+
const id = makeEntityId(source.file.id, fragments.length);
|
|
39
|
+
fragments.push({
|
|
40
|
+
id,
|
|
41
|
+
group: id,
|
|
42
|
+
fileId: source.file.id,
|
|
43
|
+
range: linesToWindow(lines, lineOffsets, section.startIndex, section.endIndex).range,
|
|
44
|
+
content: {
|
|
45
|
+
kind: "text",
|
|
46
|
+
text: markdownOutline(metadata),
|
|
47
|
+
},
|
|
48
|
+
metadata,
|
|
49
|
+
});
|
|
50
|
+
for (const window of windows) {
|
|
51
|
+
fragments.push(markdownWindowToFragment(source, metadata, window, fragments.length, id));
|
|
52
|
+
}
|
|
53
|
+
continue;
|
|
54
|
+
}
|
|
55
|
+
for (const window of windows) {
|
|
56
|
+
fragments.push({
|
|
57
|
+
id: makeEntityId(source.file.id, fragments.length),
|
|
58
|
+
fileId: source.file.id,
|
|
59
|
+
range: window.range,
|
|
60
|
+
content: {
|
|
61
|
+
kind: "text",
|
|
62
|
+
text: window.text,
|
|
63
|
+
},
|
|
64
|
+
metadata,
|
|
65
|
+
});
|
|
66
|
+
}
|
|
67
|
+
}
|
|
68
|
+
return fragments;
|
|
69
|
+
}
|
|
70
|
+
}
|
|
71
|
+
function markdownWindowToFragment(source, metadata, window, index, group) {
|
|
72
|
+
return {
|
|
73
|
+
id: makeEntityId(source.file.id, index),
|
|
74
|
+
group,
|
|
75
|
+
fileId: source.file.id,
|
|
76
|
+
range: window.range,
|
|
77
|
+
content: {
|
|
78
|
+
kind: "text",
|
|
79
|
+
text: window.text,
|
|
80
|
+
},
|
|
81
|
+
metadata,
|
|
82
|
+
};
|
|
83
|
+
}
|
|
84
|
+
function markdownOutline(metadata) {
|
|
85
|
+
return metadata.heading ?? "markdown section";
|
|
86
|
+
}
|
|
87
|
+
function scanHeadings(lines) {
|
|
88
|
+
const headings = [];
|
|
89
|
+
let fence = null;
|
|
90
|
+
for (let index = 0; index < lines.length; index++) {
|
|
91
|
+
const line = lines[index];
|
|
92
|
+
const trimmed = line.trimStart();
|
|
93
|
+
if (fence) {
|
|
94
|
+
if (trimmed.startsWith(fence)) {
|
|
95
|
+
fence = null;
|
|
96
|
+
}
|
|
97
|
+
continue;
|
|
98
|
+
}
|
|
99
|
+
if (trimmed.startsWith("```")) {
|
|
100
|
+
fence = "```";
|
|
101
|
+
continue;
|
|
102
|
+
}
|
|
103
|
+
if (trimmed.startsWith("~~~")) {
|
|
104
|
+
fence = "~~~";
|
|
105
|
+
continue;
|
|
106
|
+
}
|
|
107
|
+
const atx = line.match(/^(#{1,6})\s+(.+?)\s*#*\s*$/);
|
|
108
|
+
if (atx) {
|
|
109
|
+
headings.push({
|
|
110
|
+
level: atx[1].length,
|
|
111
|
+
text: atx[2].trim(),
|
|
112
|
+
lineIndex: index,
|
|
113
|
+
});
|
|
114
|
+
continue;
|
|
115
|
+
}
|
|
116
|
+
const next = lines[index + 1]?.trim();
|
|
117
|
+
if (line.trim().length > 0 && next && /^(=+|-+)\s*$/.test(next)) {
|
|
118
|
+
headings.push({
|
|
119
|
+
level: next.startsWith("=") ? 1 : 2,
|
|
120
|
+
text: line.trim(),
|
|
121
|
+
lineIndex: index,
|
|
122
|
+
});
|
|
123
|
+
index++;
|
|
124
|
+
}
|
|
125
|
+
}
|
|
126
|
+
return headings;
|
|
127
|
+
}
|
|
128
|
+
function buildSections(headings, lines) {
|
|
129
|
+
const stack = [];
|
|
130
|
+
const sections = [];
|
|
131
|
+
const firstHeading = headings[0];
|
|
132
|
+
if (firstHeading.lineIndex > 0 && lines.slice(0, firstHeading.lineIndex).join("\n").trim().length > 0) {
|
|
133
|
+
sections.push({
|
|
134
|
+
heading: null,
|
|
135
|
+
startIndex: 0,
|
|
136
|
+
endIndex: firstHeading.lineIndex - 1,
|
|
137
|
+
breadcrumb: [],
|
|
138
|
+
});
|
|
139
|
+
}
|
|
140
|
+
for (let index = 0; index < headings.length; index++) {
|
|
141
|
+
const heading = headings[index];
|
|
142
|
+
while (stack.length > 0 && stack[stack.length - 1].level >= heading.level) {
|
|
143
|
+
stack.pop();
|
|
144
|
+
}
|
|
145
|
+
sections.push({
|
|
146
|
+
heading,
|
|
147
|
+
startIndex: heading.lineIndex,
|
|
148
|
+
endIndex: index + 1 < headings.length
|
|
149
|
+
? headings[index + 1].lineIndex - 1
|
|
150
|
+
: lines.length - 1,
|
|
151
|
+
breadcrumb: stack.map((item) => item.text),
|
|
152
|
+
});
|
|
153
|
+
stack.push(heading);
|
|
154
|
+
}
|
|
155
|
+
return sections;
|
|
156
|
+
}
|
|
157
|
+
function splitMarkdownSection(input) {
|
|
158
|
+
const windows = [];
|
|
159
|
+
let startIndex = input.section.startIndex;
|
|
160
|
+
while (startIndex <= input.section.endIndex) {
|
|
161
|
+
if (input.lines[startIndex].length + 1 > input.maxChars) {
|
|
162
|
+
windows.push(...splitLongLine(input.lines[startIndex], startIndex, input.lineOffsets[startIndex], input.maxChars));
|
|
163
|
+
startIndex++;
|
|
164
|
+
continue;
|
|
165
|
+
}
|
|
166
|
+
let endIndex = startIndex;
|
|
167
|
+
let usedChars = 0;
|
|
168
|
+
while (endIndex <= input.section.endIndex) {
|
|
169
|
+
const lineLength = input.lines[endIndex].length + 1;
|
|
170
|
+
if (usedChars + lineLength > input.maxChars && endIndex > startIndex) {
|
|
171
|
+
break;
|
|
172
|
+
}
|
|
173
|
+
usedChars += lineLength;
|
|
174
|
+
endIndex++;
|
|
175
|
+
}
|
|
176
|
+
if (endIndex <= input.section.endIndex && endIndex - startIndex > 1) {
|
|
177
|
+
endIndex = chooseMarkdownBreak(input.lines, input.fenceLines, startIndex, endIndex);
|
|
178
|
+
}
|
|
179
|
+
windows.push(linesToWindow(input.lines, input.lineOffsets, startIndex, endIndex - 1));
|
|
180
|
+
if (endIndex > input.section.endIndex) {
|
|
181
|
+
break;
|
|
182
|
+
}
|
|
183
|
+
const overlapLines = computeMarkdownOverlapLines(input.lines, startIndex, endIndex, input.overlapChars);
|
|
184
|
+
const nextStart = endIndex - overlapLines;
|
|
185
|
+
startIndex = nextStart > startIndex ? nextStart : endIndex;
|
|
186
|
+
}
|
|
187
|
+
return windows.filter((window) => window.text.trim().length > 0);
|
|
188
|
+
}
|
|
189
|
+
function chooseMarkdownBreak(lines, fenceLines, startIndex, endIndex) {
|
|
190
|
+
const minBreak = startIndex + Math.max(1, Math.floor((endIndex - startIndex) * 0.7));
|
|
191
|
+
let bestBreak = endIndex;
|
|
192
|
+
let bestScore = markdownBreakScore(lines, fenceLines, endIndex);
|
|
193
|
+
for (let index = minBreak; index <= endIndex; index++) {
|
|
194
|
+
const score = markdownBreakScore(lines, fenceLines, index);
|
|
195
|
+
if (score > bestScore) {
|
|
196
|
+
bestBreak = index;
|
|
197
|
+
bestScore = score;
|
|
198
|
+
}
|
|
199
|
+
}
|
|
200
|
+
return bestBreak;
|
|
201
|
+
}
|
|
202
|
+
function markdownBreakScore(lines, fenceLines, breakIndex) {
|
|
203
|
+
if (breakIndex <= 0 || breakIndex >= lines.length || fenceLines[breakIndex]) {
|
|
204
|
+
return 0;
|
|
205
|
+
}
|
|
206
|
+
const current = lines[breakIndex].trim();
|
|
207
|
+
const previous = lines[breakIndex - 1].trim();
|
|
208
|
+
if (/^#{1,6}\s+/.test(current)) {
|
|
209
|
+
return 100;
|
|
210
|
+
}
|
|
211
|
+
if (previous === "" && current === "") {
|
|
212
|
+
return 70;
|
|
213
|
+
}
|
|
214
|
+
if (previous === "") {
|
|
215
|
+
return 60;
|
|
216
|
+
}
|
|
217
|
+
if (/^([-*+]|\d+\.)\s+/.test(current)) {
|
|
218
|
+
return 35;
|
|
219
|
+
}
|
|
220
|
+
if (/^>\s+/.test(current)) {
|
|
221
|
+
return 25;
|
|
222
|
+
}
|
|
223
|
+
return 10;
|
|
224
|
+
}
|
|
225
|
+
function splitLongLine(line, lineIndex, lineOffset, maxChars) {
|
|
226
|
+
const windows = [];
|
|
227
|
+
let offset = 0;
|
|
228
|
+
while (offset < line.length) {
|
|
229
|
+
const remaining = line.length - offset;
|
|
230
|
+
const sliceLength = remaining <= maxChars
|
|
231
|
+
? remaining
|
|
232
|
+
: findLineCut(line.slice(offset), maxChars);
|
|
233
|
+
const text = line.slice(offset, offset + sliceLength);
|
|
234
|
+
windows.push({
|
|
235
|
+
text,
|
|
236
|
+
range: {
|
|
237
|
+
kind: "text",
|
|
238
|
+
startLine: lineIndex + 1,
|
|
239
|
+
endLine: lineIndex + 1,
|
|
240
|
+
startOffset: lineOffset + offset,
|
|
241
|
+
endOffset: lineOffset + offset + text.length,
|
|
242
|
+
},
|
|
243
|
+
});
|
|
244
|
+
offset += sliceLength;
|
|
245
|
+
}
|
|
246
|
+
return windows;
|
|
247
|
+
}
|
|
248
|
+
function linesToWindow(lines, lineOffsets, startIndex, endIndex) {
|
|
249
|
+
return {
|
|
250
|
+
text: lines.slice(startIndex, endIndex + 1).join("\n"),
|
|
251
|
+
range: {
|
|
252
|
+
kind: "text",
|
|
253
|
+
startLine: startIndex + 1,
|
|
254
|
+
endLine: endIndex + 1,
|
|
255
|
+
startOffset: lineOffsets[startIndex],
|
|
256
|
+
endOffset: lineOffsets[endIndex] + lines[endIndex].length,
|
|
257
|
+
},
|
|
258
|
+
};
|
|
259
|
+
}
|
|
260
|
+
function computeLineOffsets(lines) {
|
|
261
|
+
const offsets = [];
|
|
262
|
+
let offset = 0;
|
|
263
|
+
for (const line of lines) {
|
|
264
|
+
offsets.push(offset);
|
|
265
|
+
offset += line.length + 1;
|
|
266
|
+
}
|
|
267
|
+
return offsets;
|
|
268
|
+
}
|
|
269
|
+
function computeFenceLines(lines) {
|
|
270
|
+
const inFence = new Array(lines.length).fill(false);
|
|
271
|
+
let fence = null;
|
|
272
|
+
for (let index = 0; index < lines.length; index++) {
|
|
273
|
+
const trimmed = lines[index].trimStart();
|
|
274
|
+
if (fence) {
|
|
275
|
+
inFence[index] = true;
|
|
276
|
+
if (trimmed.startsWith(fence)) {
|
|
277
|
+
fence = null;
|
|
278
|
+
}
|
|
279
|
+
continue;
|
|
280
|
+
}
|
|
281
|
+
if (trimmed.startsWith("```")) {
|
|
282
|
+
fence = "```";
|
|
283
|
+
}
|
|
284
|
+
else if (trimmed.startsWith("~~~")) {
|
|
285
|
+
fence = "~~~";
|
|
286
|
+
}
|
|
287
|
+
}
|
|
288
|
+
return inFence;
|
|
289
|
+
}
|
|
290
|
+
function computeMarkdownOverlapLines(lines, startIndex, endIndex, overlapChars) {
|
|
291
|
+
if (overlapChars <= 0) {
|
|
292
|
+
return 0;
|
|
293
|
+
}
|
|
294
|
+
let chars = 0;
|
|
295
|
+
let count = 0;
|
|
296
|
+
for (let index = endIndex - 1; index > startIndex; index--) {
|
|
297
|
+
chars += lines[index].length + 1;
|
|
298
|
+
if (chars > overlapChars) {
|
|
299
|
+
break;
|
|
300
|
+
}
|
|
301
|
+
count++;
|
|
302
|
+
}
|
|
303
|
+
return Math.min(count, Math.floor((endIndex - startIndex) / 2));
|
|
304
|
+
}
|
|
305
|
+
function findLineCut(line, maxChars) {
|
|
306
|
+
const minPosition = Math.floor(maxChars * 0.7);
|
|
307
|
+
let bestPosition = -1;
|
|
308
|
+
let bestScore = 0;
|
|
309
|
+
for (let index = minPosition; index < maxChars; index++) {
|
|
310
|
+
const character = line[index];
|
|
311
|
+
let score = 0;
|
|
312
|
+
if (character === "." || character === "!" || character === "?") {
|
|
313
|
+
score = 4;
|
|
314
|
+
}
|
|
315
|
+
else if (character === "," || character === ";" || character === ":") {
|
|
316
|
+
score = 3;
|
|
317
|
+
}
|
|
318
|
+
else if (character === " " || character === "\t") {
|
|
319
|
+
score = 2;
|
|
320
|
+
}
|
|
321
|
+
else if (character === "-" || character === "/" || character === "\\") {
|
|
322
|
+
score = 1;
|
|
323
|
+
}
|
|
324
|
+
if (score >= bestScore && score > 0) {
|
|
325
|
+
bestScore = score;
|
|
326
|
+
bestPosition = index + 1;
|
|
327
|
+
}
|
|
328
|
+
}
|
|
329
|
+
return bestPosition > 0 ? bestPosition : maxChars;
|
|
330
|
+
}
|
|
331
|
+
function markdownMetadata(section) {
|
|
332
|
+
return {
|
|
333
|
+
kind: "markdown",
|
|
334
|
+
heading: section.heading?.text ?? null,
|
|
335
|
+
level: section.heading?.level ?? null,
|
|
336
|
+
scope: section.breadcrumb.length > 0 ? section.breadcrumb.join("::") : null,
|
|
337
|
+
};
|
|
338
|
+
}
|
|
339
|
+
//# sourceMappingURL=extractor.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"extractor.js","sourceRoot":"","sources":["../../../../../../src/engine/pipeline/indexing/extractor/markdown/extractor.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,YAAY,EAAE,MAAM,cAAc,CAAC;AAU5C,MAAM,4BAA4B,GAAG,IAAI,CAAC;AAC1C,MAAM,oCAAoC,GAAG,GAAG,CAAC;AAwBjD,MAAM,OAAO,iBAAiB;IACX,aAAa,CAAS;IACtB,iBAAiB,CAAS;IAG3C,YAAY,UAAoC,EAAE;QAChD,IAAI,CAAC,aAAa,GAAG,OAAO,CAAC,aAAa,IAAI,4BAA4B,CAAC;QAC3E,IAAI,CAAC,iBAAiB,GAAG,OAAO,CAAC,iBAAiB,IAAI,oCAAoC,CAAC;IAC7F,CAAC;IAGD,QAAQ,CAAC,MAAc;QACrB,OAAO,MAAM,CAAC,IAAI,KAAK,MAAM,IAAI,MAAM,CAAC,IAAI,CAAC,MAAM,KAAK,UAAU,CAAC;IACrE,CAAC;IAGD,KAAK,CAAC,OAAO,CAAC,MAAc;QAC1B,IAAI,CAAC,IAAI,CAAC,QAAQ,CAAC,MAAM,CAAC,IAAI,MAAM,CAAC,IAAI,KAAK,MAAM,EAAE,CAAC;YACrD,OAAO,EAAE,CAAC;QACZ,CAAC;QAED,MAAM,KAAK,GAAG,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;QACtC,MAAM,QAAQ,GAAG,YAAY,CAAC,KAAK,CAAC,CAAC;QACrC,IAAI,QAAQ,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;YAC1B,OAAO,EAAE,CAAC;QACZ,CAAC;QAED,MAAM,WAAW,GAAG,kBAAkB,CAAC,KAAK,CAAC,CAAC;QAC9C,MAAM,UAAU,GAAG,iBAAiB,CAAC,KAAK,CAAC,CAAC;QAC5C,MAAM,QAAQ,GAAG,aAAa,CAAC,QAAQ,EAAE,KAAK,CAAC,CAAC;QAChD,MAAM,SAAS,GAAqB,EAAE,CAAC;QAEvC,KAAK,MAAM,OAAO,IAAI,QAAQ,EAAE,CAAC;YAC/B,MAAM,QAAQ,GAAG,gBAAgB,CAAC,OAAO,CAAC,CAAC;YAC3C,MAAM,OAAO,GAAG,oBAAoB,CAAC;gBACnC,KAAK;gBACL,WAAW;gBACX,UAAU;gBACV,OAAO;gBACP,QAAQ,EAAE,IAAI,CAAC,aAAa;gBAC5B,YAAY,EAAE,IAAI,CAAC,iBAAiB;aACrC,CAAC,CAAC;YAEH,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACvB,MAAM,EAAE,GAAG,YAAY,CAAC,MAAM,CAAC,IAAI,CAAC,EAAE,EAAE,SAAS,CAAC,MAAM,CAAC,CAAC;gBAC1D,SAAS,CAAC,IAAI,CAAC;oBACb,EAAE;oBACF,KAAK,EAAE,EAAE;oBACT,MAAM,EAAE,MAAM,CAAC,IAAI,CAAC,EAAE;oBACtB,KAAK,EAAE,aAAa,CAAC,KAAK,EAAE,WAAW,EAAE,OAAO,CAAC,UAAU,EAAE,OAAO,CAAC,QAAQ,CAAC,CAAC,KAAK;oBACpF,OAAO,EAAE;wBACP,IAAI,EAAE,MAAM;wBACZ,IAAI,EAAE,eAAe,CAAC,QAAQ,CAAC;qBAChC;oBACD,QAAQ;iBACT,CAAC,CAAC;gBAEH,KAAK,MAAM,MAAM,IAAI,OAAO,EAAE,CAAC;oBAC7B,SAAS,CAAC,IAAI,CAAC,wBAAwB,CACrC,MAAM,EACN,QAAQ,EACR,MAAM,EACN,SAAS,CAAC,MAAM,EAChB,EAAE,CACH,CAAC,CAAC;gBACL,CAAC;gBAED,SAAS;YACX,CAAC;YAED,KAAK,MAAM,MAAM,IAAI,OAAO,EAAE,CAAC;gBAC7B,SAAS,CAAC,IAAI,CAAC;oBACb,EAAE,EAAE,YAAY,CAAC,MAAM,CAAC,IAAI,CAAC,EAAE,EAAE,SAAS,CAAC,MAAM,CAAC;oBAClD,MAAM,EAAE,MAAM,CAAC,IAAI,CAAC,EAAE;oBACtB,KAAK,EAAE,MAAM,CAAC,KAAK;oBACnB,OAAO,EAAE;wBACP,IAAI,EAAE,MAAM;wBACZ,IAAI,EAAE,MAAM,CAAC,IAAI;qBAClB;oBACD,QAAQ;iBACT,CAAC,CAAC;YACL,CAAC;QACH,CAAC;QAED,OAAO,SAAS,CAAC;IACnB,CAAC;CACF;AAGD,SAAS,wBAAwB,CAC/B,MAAkB,EAClB,QAAgC,EAChC,MAAsB,EACtB,KAAa,EACb,KAAa;IAEb,OAAO;QACL,EAAE,EAAE,YAAY,CAAC,MAAM,CAAC,IAAI,CAAC,EAAE,EAAE,KAAK,CAAC;QACvC,KAAK;QACL,MAAM,EAAE,MAAM,CAAC,IAAI,CAAC,EAAE;QACtB,KAAK,EAAE,MAAM,CAAC,KAAK;QACnB,OAAO,EAAE;YACP,IAAI,EAAE,MAAM;YACZ,IAAI,EAAE,MAAM,CAAC,IAAI;SAClB;QACD,QAAQ;KACT,CAAC;AACJ,CAAC;AAGD,SAAS,eAAe,CAAC,QAAgC;IACvD,OAAO,QAAQ,CAAC,OAAO,IAAI,kBAAkB,CAAC;AAChD,CAAC;AAGD,SAAS,YAAY,CAAC,KAAwB;IAC5C,MAAM,QAAQ,GAAc,EAAE,CAAC;IAC/B,IAAI,KAAK,GAAyB,IAAI,CAAC;IAEvC,KAAK,IAAI,KAAK,GAAG,CAAC,EAAE,KAAK,GAAG,KAAK,CAAC,MAAM,EAAE,KAAK,EAAE,EAAE,CAAC;QAClD,MAAM,IAAI,GAAG,KAAK,CAAC,KAAK,CAAC,CAAC;QAC1B,MAAM,OAAO,GAAG,IAAI,CAAC,SAAS,EAAE,CAAC;QAEjC,IAAI,KAAK,EAAE,CAAC;YACV,IAAI,OAAO,CAAC,UAAU,CAAC,KAAK,CAAC,EAAE,CAAC;gBAC9B,KAAK,GAAG,IAAI,CAAC;YACf,CAAC;YACD,SAAS;QACX,CAAC;QAED,IAAI,OAAO,CAAC,UAAU,CAAC,KAAK,CAAC,EAAE,CAAC;YAC9B,KAAK,GAAG,KAAK,CAAC;YACd,SAAS;QACX,CAAC;QAED,IAAI,OAAO,CAAC,UAAU,CAAC,KAAK,CAAC,EAAE,CAAC;YAC9B,KAAK,GAAG,KAAK,CAAC;YACd,SAAS;QACX,CAAC;QAED,MAAM,GAAG,GAAG,IAAI,CAAC,KAAK,CAAC,4BAA4B,CAAC,CAAC;QACrD,IAAI,GAAG,EAAE,CAAC;YACR,QAAQ,CAAC,IAAI,CAAC;gBACZ,KAAK,EAAE,GAAG,CAAC,CAAC,CAAC,CAAC,MAAM;gBACpB,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,EAAE;gBACnB,SAAS,EAAE,KAAK;aACjB,CAAC,CAAC;YACH,SAAS;QACX,CAAC;QAED,MAAM,IAAI,GAAG,KAAK,CAAC,KAAK,GAAG,CAAC,CAAC,EAAE,IAAI,EAAE,CAAC;QACtC,IAAI,IAAI,CAAC,IAAI,EAAE,CAAC,MAAM,GAAG,CAAC,IAAI,IAAI,IAAI,cAAc,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC;YAChE,QAAQ,CAAC,IAAI,CAAC;gBACZ,KAAK,EAAE,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;gBACnC,IAAI,EAAE,IAAI,CAAC,IAAI,EAAE;gBACjB,SAAS,EAAE,KAAK;aACjB,CAAC,CAAC;YACH,KAAK,EAAE,CAAC;QACV,CAAC;IACH,CAAC;IAED,OAAO,QAAQ,CAAC;AAClB,CAAC;AAGD,SAAS,aAAa,CAAC,QAA4B,EAAE,KAAwB;IAC3E,MAAM,KAAK,GAAc,EAAE,CAAC;IAC5B,MAAM,QAAQ,GAAc,EAAE,CAAC;IAC/B,MAAM,YAAY,GAAG,QAAQ,CAAC,CAAC,CAAC,CAAC;IAEjC,IAAI,YAAY,CAAC,SAAS,GAAG,CAAC,IAAI,KAAK,CAAC,KAAK,CAAC,CAAC,EAAE,YAAY,CAAC,SAAS,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,IAAI,EAAE,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QACtG,QAAQ,CAAC,IAAI,CAAC;YACZ,OAAO,EAAE,IAAI;YACb,UAAU,EAAE,CAAC;YACb,QAAQ,EAAE,YAAY,CAAC,SAAS,GAAG,CAAC;YACpC,UAAU,EAAE,EAAE;SACf,CAAC,CAAC;IACL,CAAC;IAED,KAAK,IAAI,KAAK,GAAG,CAAC,EAAE,KAAK,GAAG,QAAQ,CAAC,MAAM,EAAE,KAAK,EAAE,EAAE,CAAC;QACrD,MAAM,OAAO,GAAG,QAAQ,CAAC,KAAK,CAAC,CAAC;QAEhC,OAAO,KAAK,CAAC,MAAM,GAAG,CAAC,IAAI,KAAK,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,KAAK,IAAI,OAAO,CAAC,KAAK,EAAE,CAAC;YAC1E,KAAK,CAAC,GAAG,EAAE,CAAC;QACd,CAAC;QAED,QAAQ,CAAC,IAAI,CAAC;YACZ,OAAO;YACP,UAAU,EAAE,OAAO,CAAC,SAAS;YAC7B,QAAQ,EAAE,KAAK,GAAG,CAAC,GAAG,QAAQ,CAAC,MAAM;gBACnC,CAAC,CAAC,QAAQ,CAAC,KAAK,GAAG,CAAC,CAAC,CAAC,SAAS,GAAG,CAAC;gBACnC,CAAC,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC;YACpB,UAAU,EAAE,KAAK,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,CAAC,IAAI,CAAC;SAC3C,CAAC,CAAC;QACH,KAAK,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC;IACtB,CAAC;IAED,OAAO,QAAQ,CAAC;AAClB,CAAC;AAGD,SAAS,oBAAoB,CAAC,KAO7B;IACC,MAAM,OAAO,GAAqB,EAAE,CAAC;IACrC,IAAI,UAAU,GAAG,KAAK,CAAC,OAAO,CAAC,UAAU,CAAC;IAE1C,OAAO,UAAU,IAAI,KAAK,CAAC,OAAO,CAAC,QAAQ,EAAE,CAAC;QAC5C,IAAI,KAAK,CAAC,KAAK,CAAC,UAAU,CAAC,CAAC,MAAM,GAAG,CAAC,GAAG,KAAK,CAAC,QAAQ,EAAE,CAAC;YACxD,OAAO,CAAC,IAAI,CAAC,GAAG,aAAa,CAC3B,KAAK,CAAC,KAAK,CAAC,UAAU,CAAC,EACvB,UAAU,EACV,KAAK,CAAC,WAAW,CAAC,UAAU,CAAC,EAC7B,KAAK,CAAC,QAAQ,CACf,CAAC,CAAC;YACH,UAAU,EAAE,CAAC;YACb,SAAS;QACX,CAAC;QAED,IAAI,QAAQ,GAAG,UAAU,CAAC;QAC1B,IAAI,SAAS,GAAG,CAAC,CAAC;QAElB,OAAO,QAAQ,IAAI,KAAK,CAAC,OAAO,CAAC,QAAQ,EAAE,CAAC;YAC1C,MAAM,UAAU,GAAG,KAAK,CAAC,KAAK,CAAC,QAAQ,CAAC,CAAC,MAAM,GAAG,CAAC,CAAC;YACpD,IAAI,SAAS,GAAG,UAAU,GAAG,KAAK,CAAC,QAAQ,IAAI,QAAQ,GAAG,UAAU,EAAE,CAAC;gBACrE,MAAM;YACR,CAAC;YACD,SAAS,IAAI,UAAU,CAAC;YACxB,QAAQ,EAAE,CAAC;QACb,CAAC;QAED,IAAI,QAAQ,IAAI,KAAK,CAAC,OAAO,CAAC,QAAQ,IAAI,QAAQ,GAAG,UAAU,GAAG,CAAC,EAAE,CAAC;YACpE,QAAQ,GAAG,mBAAmB,CAAC,KAAK,CAAC,KAAK,EAAE,KAAK,CAAC,UAAU,EAAE,UAAU,EAAE,QAAQ,CAAC,CAAC;QACtF,CAAC;QAED,OAAO,CAAC,IAAI,CAAC,aAAa,CAAC,KAAK,CAAC,KAAK,EAAE,KAAK,CAAC,WAAW,EAAE,UAAU,EAAE,QAAQ,GAAG,CAAC,CAAC,CAAC,CAAC;QAEtF,IAAI,QAAQ,GAAG,KAAK,CAAC,OAAO,CAAC,QAAQ,EAAE,CAAC;YACtC,MAAM;QACR,CAAC;QAED,MAAM,YAAY,GAAG,2BAA2B,CAC9C,KAAK,CAAC,KAAK,EACX,UAAU,EACV,QAAQ,EACR,KAAK,CAAC,YAAY,CACnB,CAAC;QACF,MAAM,SAAS,GAAG,QAAQ,GAAG,YAAY,CAAC;QAC1C,UAAU,GAAG,SAAS,GAAG,UAAU,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC,CAAC,QAAQ,CAAC;IAC7D,CAAC;IAED,OAAO,OAAO,CAAC,MAAM,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,MAAM,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;AACnE,CAAC;AAGD,SAAS,mBAAmB,CAC1B,KAAwB,EACxB,UAA8B,EAC9B,UAAkB,EAClB,QAAgB;IAEhB,MAAM,QAAQ,GAAG,UAAU,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,KAAK,CAAC,CAAC,QAAQ,GAAG,UAAU,CAAC,GAAG,GAAG,CAAC,CAAC,CAAC;IACrF,IAAI,SAAS,GAAG,QAAQ,CAAC;IACzB,IAAI,SAAS,GAAG,kBAAkB,CAAC,KAAK,EAAE,UAAU,EAAE,QAAQ,CAAC,CAAC;IAEhE,KAAK,IAAI,KAAK,GAAG,QAAQ,EAAE,KAAK,IAAI,QAAQ,EAAE,KAAK,EAAE,EAAE,CAAC;QACtD,MAAM,KAAK,GAAG,kBAAkB,CAAC,KAAK,EAAE,UAAU,EAAE,KAAK,CAAC,CAAC;QAC3D,IAAI,KAAK,GAAG,SAAS,EAAE,CAAC;YACtB,SAAS,GAAG,KAAK,CAAC;YAClB,SAAS,GAAG,KAAK,CAAC;QACpB,CAAC;IACH,CAAC;IAED,OAAO,SAAS,CAAC;AACnB,CAAC;AAGD,SAAS,kBAAkB,CACzB,KAAwB,EACxB,UAA8B,EAC9B,UAAkB;IAElB,IAAI,UAAU,IAAI,CAAC,IAAI,UAAU,IAAI,KAAK,CAAC,MAAM,IAAI,UAAU,CAAC,UAAU,CAAC,EAAE,CAAC;QAC5E,OAAO,CAAC,CAAC;IACX,CAAC;IAED,MAAM,OAAO,GAAG,KAAK,CAAC,UAAU,CAAC,CAAC,IAAI,EAAE,CAAC;IACzC,MAAM,QAAQ,GAAG,KAAK,CAAC,UAAU,GAAG,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;IAE9C,IAAI,YAAY,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,CAAC;QAC/B,OAAO,GAAG,CAAC;IACb,CAAC;IAED,IAAI,QAAQ,KAAK,EAAE,IAAI,OAAO,KAAK,EAAE,EAAE,CAAC;QACtC,OAAO,EAAE,CAAC;IACZ,CAAC;IAED,IAAI,QAAQ,KAAK,EAAE,EAAE,CAAC;QACpB,OAAO,EAAE,CAAC;IACZ,CAAC;IAED,IAAI,mBAAmB,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,CAAC;QACtC,OAAO,EAAE,CAAC;IACZ,CAAC;IAED,IAAI,OAAO,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,CAAC;QAC1B,OAAO,EAAE,CAAC;IACZ,CAAC;IAED,OAAO,EAAE,CAAC;AACZ,CAAC;AAGD,SAAS,aAAa,CACpB,IAAY,EACZ,SAAiB,EACjB,UAAkB,EAClB,QAAgB;IAEhB,MAAM,OAAO,GAAqB,EAAE,CAAC;IACrC,IAAI,MAAM,GAAG,CAAC,CAAC;IAEf,OAAO,MAAM,GAAG,IAAI,CAAC,MAAM,EAAE,CAAC;QAC5B,MAAM,SAAS,GAAG,IAAI,CAAC,MAAM,GAAG,MAAM,CAAC;QACvC,MAAM,WAAW,GAAG,SAAS,IAAI,QAAQ;YACvC,CAAC,CAAC,SAAS;YACX,CAAC,CAAC,WAAW,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC,EAAE,QAAQ,CAAC,CAAC;QAC9C,MAAM,IAAI,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,EAAE,MAAM,GAAG,WAAW,CAAC,CAAC;QAEtD,OAAO,CAAC,IAAI,CAAC;YACX,IAAI;YACJ,KAAK,EAAE;gBACL,IAAI,EAAE,MAAM;gBACZ,SAAS,EAAE,SAAS,GAAG,CAAC;gBACxB,OAAO,EAAE,SAAS,GAAG,CAAC;gBACtB,WAAW,EAAE,UAAU,GAAG,MAAM;gBAChC,SAAS,EAAE,UAAU,GAAG,MAAM,GAAG,IAAI,CAAC,MAAM;aAC7C;SACF,CAAC,CAAC;QACH,MAAM,IAAI,WAAW,CAAC;IACxB,CAAC;IAED,OAAO,OAAO,CAAC;AACjB,CAAC;AAGD,SAAS,aAAa,CACpB,KAAwB,EACxB,WAA8B,EAC9B,UAAkB,EAClB,QAAgB;IAEhB,OAAO;QACL,IAAI,EAAE,KAAK,CAAC,KAAK,CAAC,UAAU,EAAE,QAAQ,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC;QACtD,KAAK,EAAE;YACL,IAAI,EAAE,MAAM;YACZ,SAAS,EAAE,UAAU,GAAG,CAAC;YACzB,OAAO,EAAE,QAAQ,GAAG,CAAC;YACrB,WAAW,EAAE,WAAW,CAAC,UAAU,CAAC;YACpC,SAAS,EAAE,WAAW,CAAC,QAAQ,CAAC,GAAG,KAAK,CAAC,QAAQ,CAAC,CAAC,MAAM;SAC1D;KACF,CAAC;AACJ,CAAC;AAGD,SAAS,kBAAkB,CAAC,KAAwB;IAClD,MAAM,OAAO,GAAa,EAAE,CAAC;IAC7B,IAAI,MAAM,GAAG,CAAC,CAAC;IAEf,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;QACzB,OAAO,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC;QACrB,MAAM,IAAI,IAAI,CAAC,MAAM,GAAG,CAAC,CAAC;IAC5B,CAAC;IAED,OAAO,OAAO,CAAC;AACjB,CAAC;AAGD,SAAS,iBAAiB,CAAC,KAAwB;IACjD,MAAM,OAAO,GAAG,IAAI,KAAK,CAAU,KAAK,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;IAC7D,IAAI,KAAK,GAAyB,IAAI,CAAC;IAEvC,KAAK,IAAI,KAAK,GAAG,CAAC,EAAE,KAAK,GAAG,KAAK,CAAC,MAAM,EAAE,KAAK,EAAE,EAAE,CAAC;QAClD,MAAM,OAAO,GAAG,KAAK,CAAC,KAAK,CAAC,CAAC,SAAS,EAAE,CAAC;QAEzC,IAAI,KAAK,EAAE,CAAC;YACV,OAAO,CAAC,KAAK,CAAC,GAAG,IAAI,CAAC;YACtB,IAAI,OAAO,CAAC,UAAU,CAAC,KAAK,CAAC,EAAE,CAAC;gBAC9B,KAAK,GAAG,IAAI,CAAC;YACf,CAAC;YACD,SAAS;QACX,CAAC;QAED,IAAI,OAAO,CAAC,UAAU,CAAC,KAAK,CAAC,EAAE,CAAC;YAC9B,KAAK,GAAG,KAAK,CAAC;QAChB,CAAC;aAAM,IAAI,OAAO,CAAC,UAAU,CAAC,KAAK,CAAC,EAAE,CAAC;YACrC,KAAK,GAAG,KAAK,CAAC;QAChB,CAAC;IACH,CAAC;IAED,OAAO,OAAO,CAAC;AACjB,CAAC;AAGD,SAAS,2BAA2B,CAClC,KAAwB,EACxB,UAAkB,EAClB,QAAgB,EAChB,YAAoB;IAEpB,IAAI,YAAY,IAAI,CAAC,EAAE,CAAC;QACtB,OAAO,CAAC,CAAC;IACX,CAAC;IAED,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,IAAI,KAAK,GAAG,CAAC,CAAC;IAEd,KAAK,IAAI,KAAK,GAAG,QAAQ,GAAG,CAAC,EAAE,KAAK,GAAG,UAAU,EAAE,KAAK,EAAE,EAAE,CAAC;QAC3D,KAAK,IAAI,KAAK,CAAC,KAAK,CAAC,CAAC,MAAM,GAAG,CAAC,CAAC;QACjC,IAAI,KAAK,GAAG,YAAY,EAAE,CAAC;YACzB,MAAM;QACR,CAAC;QACD,KAAK,EAAE,CAAC;IACV,CAAC;IAED,OAAO,IAAI,CAAC,GAAG,CAAC,KAAK,EAAE,IAAI,CAAC,KAAK,CAAC,CAAC,QAAQ,GAAG,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC;AAClE,CAAC;AAGD,SAAS,WAAW,CAAC,IAAY,EAAE,QAAgB;IACjD,MAAM,WAAW,GAAG,IAAI,CAAC,KAAK,CAAC,QAAQ,GAAG,GAAG,CAAC,CAAC;IAC/C,IAAI,YAAY,GAAG,CAAC,CAAC,CAAC;IACtB,IAAI,SAAS,GAAG,CAAC,CAAC;IAElB,KAAK,IAAI,KAAK,GAAG,WAAW,EAAE,KAAK,GAAG,QAAQ,EAAE,KAAK,EAAE,EAAE,CAAC;QACxD,MAAM,SAAS,GAAG,IAAI,CAAC,KAAK,CAAC,CAAC;QAC9B,IAAI,KAAK,GAAG,CAAC,CAAC;QAEd,IAAI,SAAS,KAAK,GAAG,IAAI,SAAS,KAAK,GAAG,IAAI,SAAS,KAAK,GAAG,EAAE,CAAC;YAChE,KAAK,GAAG,CAAC,CAAC;QACZ,CAAC;aAAM,IAAI,SAAS,KAAK,GAAG,IAAI,SAAS,KAAK,GAAG,IAAI,SAAS,KAAK,GAAG,EAAE,CAAC;YACvE,KAAK,GAAG,CAAC,CAAC;QACZ,CAAC;aAAM,IAAI,SAAS,KAAK,GAAG,IAAI,SAAS,KAAK,IAAI,EAAE,CAAC;YACnD,KAAK,GAAG,CAAC,CAAC;QACZ,CAAC;aAAM,IAAI,SAAS,KAAK,GAAG,IAAI,SAAS,KAAK,GAAG,IAAI,SAAS,KAAK,IAAI,EAAE,CAAC;YACxE,KAAK,GAAG,CAAC,CAAC;QACZ,CAAC;QAED,IAAI,KAAK,IAAI,SAAS,IAAI,KAAK,GAAG,CAAC,EAAE,CAAC;YACpC,SAAS,GAAG,KAAK,CAAC;YAClB,YAAY,GAAG,KAAK,GAAG,CAAC,CAAC;QAC3B,CAAC;IACH,CAAC;IAED,OAAO,YAAY,GAAG,CAAC,CAAC,CAAC,CAAC,YAAY,CAAC,CAAC,CAAC,QAAQ,CAAC;AACpD,CAAC;AAGD,SAAS,gBAAgB,CAAC,OAAgB;IACxC,OAAO;QACL,IAAI,EAAE,UAAU;QAChB,OAAO,EAAE,OAAO,CAAC,OAAO,EAAE,IAAI,IAAI,IAAI;QACtC,KAAK,EAAE,OAAO,CAAC,OAAO,EAAE,KAAK,IAAI,IAAI;QACrC,KAAK,EAAE,OAAO,CAAC,UAAU,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,UAAU,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,IAAI;KAC5E,CAAC;AACJ,CAAC"}
|
|
@@ -0,0 +1,49 @@
|
|
|
1
|
+
import { createRequire } from "node:module";
|
|
2
|
+
import { join } from "node:path";
|
|
3
|
+
import Parser from "web-tree-sitter";
|
|
4
|
+
const require = createRequire(import.meta.url);
|
|
5
|
+
const LANGUAGE_WASM_MAP = {
|
|
6
|
+
c: "c",
|
|
7
|
+
cpp: "cpp",
|
|
8
|
+
go: "go",
|
|
9
|
+
java: "java",
|
|
10
|
+
javascript: "javascript",
|
|
11
|
+
jsx: "javascript",
|
|
12
|
+
python: "python",
|
|
13
|
+
rust: "rust",
|
|
14
|
+
tsx: "tsx",
|
|
15
|
+
typescript: "typescript",
|
|
16
|
+
};
|
|
17
|
+
let parserReady = false;
|
|
18
|
+
const grammarCache = new Map();
|
|
19
|
+
export async function ensureParser() {
|
|
20
|
+
if (!parserReady) {
|
|
21
|
+
await Parser.init();
|
|
22
|
+
parserReady = true;
|
|
23
|
+
}
|
|
24
|
+
}
|
|
25
|
+
export async function loadGrammar(format) {
|
|
26
|
+
const cached = grammarCache.get(format);
|
|
27
|
+
if (cached) {
|
|
28
|
+
return cached;
|
|
29
|
+
}
|
|
30
|
+
const wasmName = LANGUAGE_WASM_MAP[format];
|
|
31
|
+
if (!wasmName) {
|
|
32
|
+
return null;
|
|
33
|
+
}
|
|
34
|
+
await ensureParser();
|
|
35
|
+
try {
|
|
36
|
+
const wasmsDir = join(require.resolve("tree-sitter-wasms/package.json"), "..", "out");
|
|
37
|
+
const wasmPath = join(wasmsDir, `tree-sitter-${wasmName}.wasm`);
|
|
38
|
+
const grammar = await Parser.Language.load(wasmPath);
|
|
39
|
+
grammarCache.set(format, grammar);
|
|
40
|
+
return grammar;
|
|
41
|
+
}
|
|
42
|
+
catch {
|
|
43
|
+
return null;
|
|
44
|
+
}
|
|
45
|
+
}
|
|
46
|
+
export function hasGrammar(format) {
|
|
47
|
+
return format in LANGUAGE_WASM_MAP;
|
|
48
|
+
}
|
|
49
|
+
//# sourceMappingURL=grammar.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"grammar.js","sourceRoot":"","sources":["../../../../../../src/engine/pipeline/indexing/extractor/tree-sitter/grammar.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,aAAa,EAAE,MAAM,aAAa,CAAC;AAC5C,OAAO,EAAE,IAAI,EAAE,MAAM,WAAW,CAAC;AACjC,OAAO,MAAM,MAAM,iBAAiB,CAAC;AAGrC,MAAM,OAAO,GAAG,aAAa,CAAC,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;AAG/C,MAAM,iBAAiB,GAA2B;IAChD,CAAC,EAAE,GAAG;IACN,GAAG,EAAE,KAAK;IACV,EAAE,EAAE,IAAI;IACR,IAAI,EAAE,MAAM;IACZ,UAAU,EAAE,YAAY;IACxB,GAAG,EAAE,YAAY;IACjB,MAAM,EAAE,QAAQ;IAChB,IAAI,EAAE,MAAM;IACZ,GAAG,EAAE,KAAK;IACV,UAAU,EAAE,YAAY;CACzB,CAAC;AAGF,IAAI,WAAW,GAAG,KAAK,CAAC;AACxB,MAAM,YAAY,GAAG,IAAI,GAAG,EAA2B,CAAC;AAGxD,MAAM,CAAC,KAAK,UAAU,YAAY;IAChC,IAAI,CAAC,WAAW,EAAE,CAAC;QACjB,MAAM,MAAM,CAAC,IAAI,EAAE,CAAC;QACpB,WAAW,GAAG,IAAI,CAAC;IACrB,CAAC;AACH,CAAC;AAGD,MAAM,CAAC,KAAK,UAAU,WAAW,CAAC,MAAc;IAC9C,MAAM,MAAM,GAAG,YAAY,CAAC,GAAG,CAAC,MAAM,CAAC,CAAC;IACxC,IAAI,MAAM,EAAE,CAAC;QACX,OAAO,MAAM,CAAC;IAChB,CAAC;IAED,MAAM,QAAQ,GAAG,iBAAiB,CAAC,MAAM,CAAC,CAAC;IAC3C,IAAI,CAAC,QAAQ,EAAE,CAAC;QACd,OAAO,IAAI,CAAC;IACd,CAAC;IAED,MAAM,YAAY,EAAE,CAAC;IAErB,IAAI,CAAC;QACH,MAAM,QAAQ,GAAG,IAAI,CAAC,OAAO,CAAC,OAAO,CAAC,gCAAgC,CAAC,EAAE,IAAI,EAAE,KAAK,CAAC,CAAC;QACtF,MAAM,QAAQ,GAAG,IAAI,CAAC,QAAQ,EAAE,eAAe,QAAQ,OAAO,CAAC,CAAC;QAChE,MAAM,OAAO,GAAG,MAAM,MAAM,CAAC,QAAQ,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAC;QACrD,YAAY,CAAC,GAAG,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC;QAElC,OAAO,OAAO,CAAC;IACjB,CAAC;IAAC,MAAM,CAAC;QACP,OAAO,IAAI,CAAC;IACd,CAAC;AACH,CAAC;AAGD,MAAM,UAAU,UAAU,CAAC,MAAc;IACvC,OAAO,MAAM,IAAI,iBAAiB,CAAC;AACrC,CAAC"}
|
|
@@ -0,0 +1,26 @@
|
|
|
1
|
+
export function findIdentifierLeaf(node) {
|
|
2
|
+
const wrappers = new Set([
|
|
3
|
+
"array_declarator",
|
|
4
|
+
"function_declarator",
|
|
5
|
+
"init_declarator",
|
|
6
|
+
"parenthesized_declarator",
|
|
7
|
+
"pointer_declarator",
|
|
8
|
+
"reference_declarator",
|
|
9
|
+
]);
|
|
10
|
+
let current = node;
|
|
11
|
+
for (let depth = 0; current && depth < 16; depth++) {
|
|
12
|
+
if (current.type === "identifier" || current.type.endsWith("_identifier")) {
|
|
13
|
+
return current;
|
|
14
|
+
}
|
|
15
|
+
if (current.type === "destructor_name" || current.type === "operator_name") {
|
|
16
|
+
return current;
|
|
17
|
+
}
|
|
18
|
+
if (wrappers.has(current.type)) {
|
|
19
|
+
current = current.childForFieldName("declarator");
|
|
20
|
+
continue;
|
|
21
|
+
}
|
|
22
|
+
return null;
|
|
23
|
+
}
|
|
24
|
+
return null;
|
|
25
|
+
}
|
|
26
|
+
//# sourceMappingURL=nodes.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"nodes.js","sourceRoot":"","sources":["../../../../../../src/engine/pipeline/indexing/extractor/tree-sitter/nodes.ts"],"names":[],"mappings":"AAMA,MAAM,UAAU,kBAAkB,CAAC,IAAY;IAC7C,MAAM,QAAQ,GAAG,IAAI,GAAG,CAAC;QACvB,kBAAkB;QAClB,qBAAqB;QACrB,iBAAiB;QACjB,0BAA0B;QAC1B,oBAAoB;QACpB,sBAAsB;KACvB,CAAC,CAAC;IAEH,IAAI,OAAO,GAAkB,IAAI,CAAC;IAElC,KAAK,IAAI,KAAK,GAAG,CAAC,EAAE,OAAO,IAAI,KAAK,GAAG,EAAE,EAAE,KAAK,EAAE,EAAE,CAAC;QACnD,IAAI,OAAO,CAAC,IAAI,KAAK,YAAY,IAAI,OAAO,CAAC,IAAI,CAAC,QAAQ,CAAC,aAAa,CAAC,EAAE,CAAC;YAC1E,OAAO,OAAO,CAAC;QACjB,CAAC;QAED,IAAI,OAAO,CAAC,IAAI,KAAK,iBAAiB,IAAI,OAAO,CAAC,IAAI,KAAK,eAAe,EAAE,CAAC;YAC3E,OAAO,OAAO,CAAC;QACjB,CAAC;QAED,IAAI,QAAQ,CAAC,GAAG,CAAC,OAAO,CAAC,IAAI,CAAC,EAAE,CAAC;YAC/B,OAAO,GAAG,OAAO,CAAC,iBAAiB,CAAC,YAAY,CAAC,CAAC;YAClD,SAAS;QACX,CAAC;QAED,OAAO,IAAI,CAAC;IACd,CAAC;IAED,OAAO,IAAI,CAAC;AACd,CAAC"}
|
|
@@ -0,0 +1,31 @@
|
|
|
1
|
+
import Parser from "web-tree-sitter";
|
|
2
|
+
import { ensureParser, loadGrammar } from "./grammar.js";
|
|
3
|
+
export async function withParser(source, format, fn) {
|
|
4
|
+
await ensureParser();
|
|
5
|
+
const grammar = await loadGrammar(format);
|
|
6
|
+
if (!grammar) {
|
|
7
|
+
return null;
|
|
8
|
+
}
|
|
9
|
+
const parser = new Parser();
|
|
10
|
+
parser.setLanguage(grammar);
|
|
11
|
+
let tree;
|
|
12
|
+
try {
|
|
13
|
+
tree = parser.parse(source);
|
|
14
|
+
}
|
|
15
|
+
catch {
|
|
16
|
+
parser.delete();
|
|
17
|
+
return null;
|
|
18
|
+
}
|
|
19
|
+
if (!tree) {
|
|
20
|
+
parser.delete();
|
|
21
|
+
return null;
|
|
22
|
+
}
|
|
23
|
+
try {
|
|
24
|
+
return await fn(tree);
|
|
25
|
+
}
|
|
26
|
+
finally {
|
|
27
|
+
tree.delete();
|
|
28
|
+
parser.delete();
|
|
29
|
+
}
|
|
30
|
+
}
|
|
31
|
+
//# sourceMappingURL=parser.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"parser.js","sourceRoot":"","sources":["../../../../../../src/engine/pipeline/indexing/extractor/tree-sitter/parser.ts"],"names":[],"mappings":"AAAA,OAAO,MAAM,MAAM,iBAAiB,CAAC;AACrC,OAAO,EAAE,YAAY,EAAE,WAAW,EAAE,MAAM,cAAc,CAAC;AAGzD,MAAM,CAAC,KAAK,UAAU,UAAU,CAC9B,MAAc,EACd,MAAc,EACd,EAAyC;IAEzC,MAAM,YAAY,EAAE,CAAC;IAErB,MAAM,OAAO,GAAG,MAAM,WAAW,CAAC,MAAM,CAAC,CAAC;IAC1C,IAAI,CAAC,OAAO,EAAE,CAAC;QACb,OAAO,IAAI,CAAC;IACd,CAAC;IAED,MAAM,MAAM,GAAG,IAAI,MAAM,EAAE,CAAC;IAC5B,MAAM,CAAC,WAAW,CAAC,OAAO,CAAC,CAAC;IAE5B,IAAI,IAAwB,CAAC;IAE7B,IAAI,CAAC;QACH,IAAI,GAAG,MAAM,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC;IAC9B,CAAC;IAAC,MAAM,CAAC;QACP,MAAM,CAAC,MAAM,EAAE,CAAC;QAChB,OAAO,IAAI,CAAC;IACd,CAAC;IAED,IAAI,CAAC,IAAI,EAAE,CAAC;QACV,MAAM,CAAC,MAAM,EAAE,CAAC;QAChB,OAAO,IAAI,CAAC;IACd,CAAC;IAED,IAAI,CAAC;QACH,OAAO,MAAM,EAAE,CAAC,IAAI,CAAC,CAAC;IACxB,CAAC;YAAS,CAAC;QACT,IAAI,CAAC,MAAM,EAAE,CAAC;QACd,MAAM,CAAC,MAAM,EAAE,CAAC;IAClB,CAAC;AACH,CAAC"}
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
export declare function makeEntityId(fileId: string, index: number): string;
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"ids.js","sourceRoot":"","sources":["../../../../src/engine/pipeline/indexing/ids.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,UAAU,EAAE,MAAM,qBAAqB,CAAC;AAGjD,MAAM,UAAU,YAAY,CAAC,MAAc,EAAE,KAAa;IACxD,OAAO,UAAU,CAAC,GAAG,MAAM,KAAK,KAAK,EAAE,CAAC,CAAC;AAC3C,CAAC"}
|
|
@@ -0,0 +1,12 @@
|
|
|
1
|
+
import type { EmbeddingModel } from "../../models/embeddings.js";
|
|
2
|
+
import type { CollectionStorage } from "../../storage/index.js";
|
|
3
|
+
import type { CollectionIndexStatus, CollectionInfo, FileInfo, IndexProgress, IndexResult } from "../../types.js";
|
|
4
|
+
export type IndexContext = {
|
|
5
|
+
collection: CollectionInfo;
|
|
6
|
+
storage: CollectionStorage;
|
|
7
|
+
embeddingModel: EmbeddingModel;
|
|
8
|
+
embeddingConcurrency?: number;
|
|
9
|
+
onProgress?: (progress: IndexProgress) => void;
|
|
10
|
+
};
|
|
11
|
+
export declare function indexCollection(ctx: IndexContext): Promise<IndexResult>;
|
|
12
|
+
export declare function getCollectionIndexStatus(collection: CollectionInfo, storedFiles: readonly FileInfo[]): Promise<CollectionIndexStatus>;
|