/**
* Content analyzer for atomic note system
*
* Analyzes markdown content to determine if it exceeds atomic note limits.
* Extracts sections, detects sub-concepts, and provides metrics.
*/
import type { AtomicConfig } from '../../types/index.js';
import type { ContentAnalysis, SectionInfo, SubConcept } from '../../types/atomic.js';
import { stripWikiLinks } from '../../utils/markdown.js';
/**
* Default atomic configuration limits
* Phase 018: Removed hub_filename - hub names are now derived from title
*/
const DEFAULT_LIMITS: AtomicConfig = {
max_lines: 200,
max_sections: 6,
section_max_lines: 50,
auto_split: true,
};
/**
* Analyze markdown content for atomic note metrics
*/
export function analyzeContent(
content: string,
config: Partial<AtomicConfig> = {}
): ContentAnalysis {
const limits = { ...DEFAULT_LIMITS, ...config };
// Separate frontmatter from body
const { body, frontmatterLines } = separateFrontmatter(content);
const lines = body.split('\n');
const lineCount = lines.length;
const wordCount = countWords(body);
// Extract sections
const sections = extractSections(lines);
const sectionCount = sections.length;
// Find large sections
const largeSections = sections.filter(
(s) => s.lineCount > (limits.section_max_lines ?? 50)
);
// Detect sub-concepts (H3+ with substantial content)
const subConcepts = detectSubConcepts(lines, sections);
// Identify code blocks
const codeBlocks = extractCodeBlocks(lines);
const codeBlockLines = codeBlocks.reduce((sum, cb) => sum + cb.lineCount, 0);
// Calculate content-only lines (excluding code blocks)
const contentLines = lineCount - codeBlockLines;
return {
lineCount,
sectionCount,
wordCount,
contentLines,
frontmatterLines,
sections,
largeSections: largeSections.map((s) => s.title),
subConcepts,
codeBlocks: codeBlocks.map((cb) => ({
language: cb.language,
lineCount: cb.lineCount,
startLine: cb.startLine,
})),
limits,
};
}
/**
* Separate YAML frontmatter from markdown body
*/
function separateFrontmatter(content: string): { body: string; frontmatterLines: number } {
const lines = content.split('\n');
if (lines[0] !== '---') {
return { body: content, frontmatterLines: 0 };
}
// Find closing ---
let endIndex = -1;
for (let i = 1; i < lines.length; i++) {
if (lines[i] === '---') {
endIndex = i;
break;
}
}
if (endIndex === -1) {
return { body: content, frontmatterLines: 0 };
}
const frontmatterLines = endIndex + 1;
const body = lines.slice(endIndex + 1).join('\n').trim();
return { body, frontmatterLines };
}
/**
* Count words in content (excluding code blocks and links)
*/
function countWords(content: string): number {
// Remove code blocks
let text = content.replace(/```[\s\S]*?```/g, '');
// Remove inline code
text = text.replace(/`[^`]+`/g, '');
// Remove wiki-links but keep display text
text = text.replace(/\[\[([^\]|]+)\|([^\]]+)\]\]/g, '$2');
text = text.replace(/\[\[([^\]]+)\]\]/g, '$1');
// Remove markdown links
text = text.replace(/\[([^\]]+)\]\([^)]+\)/g, '$1');
// Count words
const words = text.split(/\s+/).filter((w) => w.length > 0);
return words.length;
}
/**
* Extract H2 sections from content
*/
function extractSections(lines: string[]): SectionInfo[] {
const sections: SectionInfo[] = [];
let currentSection: SectionInfo | null = null;
for (let i = 0; i < lines.length; i++) {
const line = lines[i] ?? '';
if (line.startsWith('## ') && !line.startsWith('### ')) {
// End previous section
if (currentSection) {
currentSection.endLine = i - 1;
currentSection.lineCount = currentSection.endLine - currentSection.startLine + 1;
sections.push(currentSection);
}
// Start new section (strip wiki-links from title)
currentSection = {
title: stripWikiLinks(line.replace(/^##\s+/, '').trim()),
startLine: i,
endLine: lines.length - 1,
lineCount: 0,
level: 2,
};
}
}
// Close last section
if (currentSection) {
currentSection.endLine = lines.length - 1;
currentSection.lineCount = currentSection.endLine - currentSection.startLine + 1;
sections.push(currentSection);
}
return sections;
}
/**
* Detect sub-concepts within sections (H3+ headings with content)
*/
function detectSubConcepts(lines: string[], sections: SectionInfo[]): SubConcept[] {
const subConcepts: SubConcept[] = [];
const minContentLines = 5; // Minimum lines to be considered a sub-concept
let currentSubConcept: SubConcept | null = null;
for (let i = 0; i < lines.length; i++) {
const line = lines[i] ?? '';
// Check for H3+ headings
const headingMatch = line.match(/^(#{3,6})\s+(.+)$/);
if (headingMatch) {
// End previous sub-concept
if (currentSubConcept) {
currentSubConcept.endLine = i - 1;
currentSubConcept.lineCount = currentSubConcept.endLine - currentSubConcept.startLine + 1;
if (currentSubConcept.lineCount >= minContentLines) {
// Find parent section
const parentSection = sections.find(
(s) => s.startLine < currentSubConcept!.startLine && s.endLine >= currentSubConcept!.endLine
);
if (parentSection) {
currentSubConcept.parentSection = parentSection.title;
}
subConcepts.push(currentSubConcept);
}
}
// Start new sub-concept (strip wiki-links from title)
currentSubConcept = {
title: stripWikiLinks(headingMatch[2]?.trim() ?? ''),
level: headingMatch[1]?.length ?? 3,
startLine: i,
endLine: lines.length - 1,
lineCount: 0,
};
}
}
// Close last sub-concept
if (currentSubConcept) {
currentSubConcept.endLine = lines.length - 1;
currentSubConcept.lineCount = currentSubConcept.endLine - currentSubConcept.startLine + 1;
if (currentSubConcept.lineCount >= minContentLines) {
const parentSection = sections.find(
(s) => s.startLine < currentSubConcept!.startLine && s.endLine >= currentSubConcept!.endLine
);
if (parentSection) {
currentSubConcept.parentSection = parentSection.title;
}
subConcepts.push(currentSubConcept);
}
}
return subConcepts;
}
/**
* Extract code blocks with their sizes
*/
interface CodeBlockInfo {
language: string;
startLine: number;
endLine: number;
lineCount: number;
}
function extractCodeBlocks(lines: string[]): CodeBlockInfo[] {
const codeBlocks: CodeBlockInfo[] = [];
let inCodeBlock = false;
let currentBlock: CodeBlockInfo | null = null;
for (let i = 0; i < lines.length; i++) {
const line = lines[i] ?? '';
if (line.startsWith('```')) {
if (!inCodeBlock) {
// Start code block
inCodeBlock = true;
const language = line.slice(3).trim();
currentBlock = {
language,
startLine: i,
endLine: -1,
lineCount: 0,
};
} else {
// End code block
inCodeBlock = false;
if (currentBlock) {
currentBlock.endLine = i;
currentBlock.lineCount = currentBlock.endLine - currentBlock.startLine + 1;
codeBlocks.push(currentBlock);
currentBlock = null;
}
}
}
}
return codeBlocks;
}
/**
* Check if content is primarily code
*/
export function isCodeHeavy(analysis: ContentAnalysis): boolean {
const codeLines = analysis.codeBlocks.reduce((sum, cb) => sum + cb.lineCount, 0);
return codeLines > analysis.lineCount * 0.5;
}
/**
* Get the title from content (first H1 heading)
* Strips wiki-link syntax from the extracted title
*/
export function extractTitle(content: string): string | null {
const lines = content.split('\n');
for (const line of lines) {
if (line.startsWith('# ') && !line.startsWith('## ')) {
const rawTitle = line.replace(/^#\s+/, '').trim();
return stripWikiLinks(rawTitle);
}
}
return null;
}
/**
* Extract all wiki-links from content
*/
export function extractWikiLinks(content: string): string[] {
const linkRegex = /\[\[([^\]|]+)(?:\|[^\]]+)?\]\]/g;
const links: string[] = [];
let match;
while ((match = linkRegex.exec(content)) !== null) {
if (match[1]) {
links.push(match[1]);
}
}
return [...new Set(links)];
}
