argustack 0.1.0

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Ievgen
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -0,0 +1,433 @@
+# Argustack
+
+**RAG engine for project intelligence — ask AI about your Jira, Git, and databases**
+
+[RAG](https://en.wikipedia.org/wiki/Retrieval-augmented_generation) (Retrieval-Augmented Generation) — an architecture where AI answers questions based on **your** data, not its training set:
+
+1. **Retrieval** — pull and index your project data locally (PostgreSQL + pgvector)
+2. **Augmentation** — inject relevant context into the AI prompt
+3. **Generation** — LLM generates answers grounded in your actual Jira tickets, code, and databases
+
+Argustack builds this knowledge base from your project's sources of truth:
+
+```
+     YOUR SOURCES                                YOUR MACHINE
+  ┌──────────────────────┐        ┌─────────────────────────────────────┐
+  │  Jira Cloud / Server │        │                                     │
+  │  (issues, comments,  │  pull  │  PostgreSQL (Docker, localhost)     │
+  │   changelogs, etc.)  │ ──────►│  ├── issues          (all fields)  │
+  └──────────────────────┘        │  ├── issue_comments  (discussions) │
+                                  │  ├── issue_changelogs (history)    │
+  ┌──────────────────────┐        │  ├── issue_worklogs  (time logs)  │
+  │  Git repository      │  pull  │  ├── issue_links     (relations)  │
+  │  (commits, files,    │ ──────►│  ├── commits         (history)    │
+  │   diffs, authors)    │        │  ├── commit_files    (per-file Δ) │
+  └──────────────────────┘        │  ├── commit_issue_refs (cross-ref)│
+                                  │  │                                  │
+  ┌──────────────────────┐        │  │  GitHub (optional, via token)    │
+  │  GitHub API          │  pull  │  ├── pull_requests   (PRs + meta) │
+  │  (PRs, reviews,      │ ──────►│  ├── pr_reviews      (approvals)  │
+  │   releases)          │        │  ├── pr_comments     (discussions) │
+  └──────────────────────┘        │  ├── pr_issue_refs   (PR↔Jira)    │
+                                  │  └── releases        (tags)        │
+  ┌──────────────────────┐        │                                     │
+  │  Database (planned)  │        │  MCP Server (localhost, stdio)      │
+  └──────────────────────┘        │  └── queries DB ──► Claude / LLM   │
+                                  │                                     │
+                                  │  .env (credentials — never leaves) │
+                                  └─────────────────────────────────────┘
+```
+
+> *Is this bug still relevant or already fixed in code?*
+> *Was the feature implemented as described in the ticket?*
+> *Who worked on this module and what changed last month?*
+> *Which commits and PRs reference ticket PAP-123?*
+> *Who approved the PR and what was the review feedback?*
+
+## How it works
+
+**Retrieval** — pulls all data from Jira, Git, and GitHub into local PostgreSQL with pgvector. Every field, every comment, every changelog entry, every commit with per-file diffs, every PR with reviews and approvals. Raw JSON preserved as-is. Nothing is filtered or lost.
+
+**Augmentation** — MCP server gives Claude Desktop / Claude Code direct access to your local database. Full-text search, semantic search (pgvector embeddings), filters, raw SQL, aggregate statistics, cross-source timeline, cross-reference between Jira issues, Git commits, and GitHub PRs — all without leaving your machine.
+
+**Generation** — ask questions in natural language. Claude queries your local data and answers with full project context.
+
+## Quick Start
+
+### Prerequisites
+
+- [Node.js](https://nodejs.org/) >= 20
+- [Docker](https://www.docker.com/) (or [OrbStack](https://orbstack.dev/) / Podman)
+
+### Install & setup
+
+```bash
+npm i -g argustack
+argustack init
+```
+
+Argustack supports **two modes** of initialization:
+
+#### Interactive mode (default)
+
+Run `argustack init` and follow the prompts:
+
+```
+? Workspace directory: ~/projects/my-team
+? Sources: ✔ Jira — issues, comments, changelogs
+           ✔ Git — commits, diffs, authors
+           ✔ GitHub — PRs, reviews, releases
+? Jira URL: https://your-team.atlassian.net
+? Email: you@company.com
+? API Token: ****
+
+Testing connection... Connected! Found 3 projects: MKT, BRAND, WEB
+
+? Projects to pull: ✔ MKT  ✔ BRAND
+? Where is your Git repository?
+  ● Clone from GitHub — select repos using your token
+? GitHub token (PAT): ****
+
+Fetching repositories... Found 12 repos.
+
+? Repositories to clone: ✔ your-org/frontend  ✔ your-org/backend
+? GitHub source: Auto-configured from clone step
+
+? Start database and sync now? Yes
+
+✔ Database running!
+✔ Jira sync complete!
+  MKT: 506 issues (150/506 ████░░░░ 30% → 506/506 done)
+✔ Git sync complete!
+  your-org/frontend: 735 commits (400/735 ██████░░ 54% → 735/735 done)
+✔ GitHub sync complete!
+  66 PRs (30/66 ██████░░ 45% → 66/66 done), 124 reviews, 3 releases
+```
+
+The interactive setup will:
+
+1. Ask which sources you have (Jira, Git, GitHub, Database)
+2. Collect credentials and test connections
+3. For Git — choose local path, clone from GitHub (multi-select repos), or clone from URL
+4. For GitHub — auto-configured if you cloned from GitHub in the previous step, or enter a Personal Access Token (see [GitHub token setup](#github-token-setup))
+5. Create a workspace with Docker config
+6. Start PostgreSQL + pgweb automatically
+7. Pull all your data
+
+#### Non-interactive mode (for AI agents and CI/CD)
+
+Pass `--no-interactive` with all values as CLI flags — no prompts, no terminal needed:
+
+```bash
+argustack init \
+  --no-interactive \
+  --dir ~/projects/my-team \
+  --source jira,git,github \
+  --jira-url "https://your-team.atlassian.net" \
+  --jira-email "you@company.com" \
+  --jira-token "ATATT3x..." \
+  --jira-projects PAP,MKT \
+  --git-repo /path/to/repo1,/path/to/repo2 \
+  --github-token "github_pat_..." \
+  --github-owner your-org \
+  --github-repo your-repo
+```
+
+Then start the database and sync:
+
+```bash
+cd ~/projects/my-team
+docker compose up -d
+argustack sync
+```
+
+All available flags:
+
+| Flag | Description |
+|------|-------------|
+| `--no-interactive` | Run without prompts — all values from flags |
+| `-d, --dir <path>` | Workspace directory (default: current) |
+| `-s, --source <list>` | Comma-separated: `jira,git,github,csv,db` |
+| `--jira-url <url>` | Jira instance URL |
+| `--jira-email <email>` | Jira user email |
+| `--jira-token <token>` | Jira API token |
+| `--jira-projects <keys>` | Comma-separated project keys, or `all` |
+| `--git-repo <paths>` | Git repo paths, comma-separated |
+| `--github-token <token>` | GitHub Personal Access Token |
+| `--github-owner <owner>` | GitHub repo owner |
+| `--github-repo <repo>` | GitHub repo name |
+| `--csv-file <path>` | Path to Jira CSV export file |
+| `--db-port <port>` | Argustack PostgreSQL port (default: `5434`) |
+| `--pgweb-port <port>` | pgweb UI port (default: `8086`) |
+
+This mode is ideal for:
+- **AI agents** — a project manager tells their AI agent "set up argustack" and it runs everything autonomously
+- **CI/CD pipelines** — automated workspace provisioning
+- **Scripted setups** — reproducible environment creation
+
+Browse your data at [localhost:8086](http://localhost:8086) — pgweb UI for running SQL queries and exploring tables in your browser.
+
+### Connect to Claude
+
+```bash
+argustack mcp install
+```
+
+Adds Argustack as an MCP server to Claude Desktop. Now you can ask Claude questions about your project data directly.
+
+## Commands
+
+```bash
+argustack init                       # create workspace (interactive prompts)
+argustack init --no-interactive ...  # create workspace from CLI flags (no prompts)
+argustack sync                       # pull data from all configured sources
+argustack sync jira                  # pull Jira only
+argustack sync git                   # pull Git commits only
+argustack sync github                # pull GitHub PRs, reviews, releases
+argustack sync csv                   # import from Jira CSV export
+argustack sync csv -f /path/to.csv   # import specific CSV file
+argustack sync -p PROJ               # pull specific Jira project
+argustack sync --since 2025-01-01    # incremental pull (only new/updated)
+argustack sources                    # list configured sources
+argustack status                     # workspace info
+argustack embed                      # generate embeddings for semantic search
+argustack mcp install                # connect to Claude Desktop
+```
+
+## What gets stored
+
+All data goes into local PostgreSQL in Docker on your machine (nothing leaves `localhost`):
+
+### Jira tables
+
+| Table | Content |
+|-------|---------|
+| `issues` | All issues — typed columns + `custom_fields` JSONB + full `raw_json` |
+| `issue_comments` | Comments with authors and timestamps |
+| `issue_changelogs` | Every field change in history |
+| `issue_worklogs` | Time tracking entries |
+| `issue_links` | Issue-to-issue relationships |
+
+Every custom field is preserved exactly as Jira returns it. 500 custom fields? All stored. Zero filtering, zero data loss.
+
+### Git tables
+
+| Table | Content |
+|-------|---------|
+| `commits` | Commit hash, message, author, email, date, full-text search |
+| `commit_files` | Per-file changes — path, status, additions, deletions |
+| `commit_issue_refs` | Cross-reference: commit ↔ Jira issue (extracted from commit messages) |
+
+Commit messages mentioning issue keys like `PAP-123` or `PROJ-45` are automatically linked to Jira issues.
+
+### GitHub tables
+
+Select "GitHub" during `argustack init` or add later with `argustack source add github`:
+
+| Table | Content |
+|-------|---------|
+| `pull_requests` | PRs — state, author, reviewers, additions/deletions, merge info, full-text search |
+| `pr_reviews` | Review approvals and change requests |
+| `pr_comments` | Inline review comments with file paths and line numbers |
+| `pr_files` | Per-file changes in each PR |
+| `pr_issue_refs` | Cross-reference: PR ↔ Jira issue (extracted from PR title and body) |
+| `releases` | GitHub releases with tags, notes, and full-text search |
+
+PR titles and bodies mentioning issue keys like `PAP-123` are automatically linked to Jira issues — just like commits.
+
+## MCP Tools
+
+When connected to Claude, these tools are available:
+
+### Jira tools
+
+| Tool | What it does |
+|------|-------------|
+| `query_issues` | Search issues — full-text, filters, or raw SQL |
+| `get_issue` | Full issue details with comments, changelogs, custom fields |
+| `issue_stats` | Aggregate stats — by status, type, assignee, project |
+| `pull_jira` | Sync latest data from Jira |
+| `list_projects` | List available Jira projects |
+
+### Git tools
+
+| Tool | What it does |
+|------|-------------|
+| `query_commits` | Search commits by text, author, date, file path, or raw SQL. Optional `repo_path` filter for multi-repo workspaces |
+| `issue_commits` | Cross-reference: find all commits mentioning a Jira issue key. Optional `repo_path` filter |
+| `commit_stats` | Aggregate stats — top authors, most changed files, linked issues. Optional `repo_path` filter |
+
+### GitHub tools
+
+| Tool | What it does |
+|------|-------------|
+| `query_prs` | Search PRs — full-text, state, author, base branch, or raw SQL |
+| `issue_prs` | Cross-reference: find all PRs mentioning a Jira issue key with reviews |
+| `query_releases` | List releases with full-text search |
+
+### Cross-source tools
+
+| Tool | What it does |
+|------|-------------|
+| `issue_timeline` | Chronological timeline — changelogs + commits + PRs for one issue |
+| `semantic_search` | Find similar issues by meaning (pgvector embeddings) |
+| `estimate` | Predict effort for new tasks — finds similar completed tasks, analyzes developer profiles, cycle/coding time, bug rate |
+
+### System tools
+
+| Tool | What it does |
+|------|-------------|
+| `workspace_info` | Current workspace configuration |
+
+## Embeddings & Semantic Search
+
+**Embeddings** turn issue text into numerical vectors that capture meaning. Two issues about "login not working" and "SSO authentication fails" will have similar vectors — even though they share zero keywords.
+
+How it works:
+
+1. `argustack embed` sends each issue's `summary + description` to OpenAI API (`text-embedding-3-small` model)
+2. Returns a 1536-dimensional vector per issue, stored in PostgreSQL via pgvector
+3. `semantic_search` MCP tool embeds your question, then finds issues with the closest vectors using cosine similarity
+
+```bash
+# Generate embeddings for all issues (requires OPENAI_API_KEY in .env)
+argustack embed
+```
+
+After embedding, ask Claude: *"Find issues similar to payment timeout errors"* — it will search by meaning, not keywords.
+
+**Costs:** text-embedding-3-small costs ~$0.02 per 1M tokens. 10,000 issues ≈ $0.05-0.10.
+
+**Optional:** Embeddings require an OpenAI API key. All other features work without it.
+
+## GitHub token setup
+
+Argustack only reads data from GitHub — it never writes anything. You need a **fine-grained Personal Access Token** with read-only permissions.
+
+1. Go to [github.com/settings/personal-access-tokens/new](https://github.com/settings/personal-access-tokens/new)
+2. **Token name** — anything (e.g. `argustack`)
+3. **Description** — optional, can leave empty
+4. **Resource owner** — your account, or the organization that owns the repo
+5. **Expiration** — "No expiration" recommended. Token is read-only and stays in your local `.env`
+6. **Repository access** — pick one:
+   - **Only select repositories** (recommended) — pick specific repos, max 50
+   - **All repositories** — all your current and future repos
+   - **Public repositories** — read-only access to public repos only
+5. **Permissions** → Repository permissions (3 total):
+
+| Permission | Access | Why |
+|---|---|---|
+| **Contents** | Read-only | Releases, downloads, tags |
+| **Metadata** | Read-only | Repository info (auto-selected, required) |
+| **Pull requests** | Read-only | PRs, reviews, comments, files |
+
+7. Click "Generate token" and copy it
+
+During `argustack init`, select "GitHub" as a source and paste the token when asked. Or add to `.env` manually:
+
+```bash
+GITHUB_TOKEN=github_pat_...
+GITHUB_OWNER=your-org
+GITHUB_REPO=your-repo
+```
+
+Then run `argustack source add github` and `argustack sync github`.
+
+## Multiple workspaces
+
+Each data source = separate workspace (like git repos):
+
+```
+~/projects/
+├── client-alpha/       # argustack init → Alpha's Jira + Git
+│   ├── .argustack/
+│   ├── .env            # Alpha credentials
+│   └── docker-compose.yml
+│
+├── client-beta/        # argustack init → Beta's Jira + Git
+│   ├── .argustack/
+│   ├── .env            # Beta credentials
+│   └── docker-compose.yml
+```
+
+## Security & Credentials
+
+**Argustack is a CLI tool. It has no backend, no cloud, no accounts.** Everything runs on your machine.
+
+When you run `argustack init`, it creates a `.env` file in your workspace with your credentials:
+
+```bash
+# .env — YOUR file, on YOUR machine, never uploaded anywhere
+
+# === Jira ===
+JIRA_URL=https://your-team.atlassian.net
+JIRA_EMAIL=you@company.com
+JIRA_API_TOKEN=your-api-token-here
+JIRA_PROJECTS=PROJ,OTHER
+
+# === Git ===
+GIT_REPO_PATHS=/path/to/repo1,/path/to/repo2
+
+# === GitHub ===
+GITHUB_TOKEN=github_pat_...
+GITHUB_OWNER=your-org
+GITHUB_REPO=your-repo
+
+# === Argustack internal PostgreSQL (match docker-compose.yml) ===
+DB_HOST=localhost
+DB_PORT=5434
+DB_USER=argustack
+DB_PASSWORD=argustack_local
+DB_NAME=argustack
+
+# === OpenAI embeddings (optional, for semantic search) ===
+# OPENAI_API_KEY=sk-...
+```
+
+**Where credentials go:**
+
+| What | Where | Who can see |
+|------|-------|-------------|
+| Jira token | `.env` on your disk | Only you |
+| GitHub token | `.env` on your disk | Only you |
+| Jira data | PostgreSQL in Docker on `localhost:5434` | Only you |
+| Git + GitHub data | PostgreSQL in Docker on `localhost:5434` | Only you |
+| Database password | `.env` on your disk | Only you |
+| Source code (this repo) | GitHub | Everyone — **no secrets here** |
+
+**What Argustack does NOT do:**
+- Does not send your data to any external server
+- Does not have analytics, telemetry, or tracking
+- Does not store credentials anywhere except your local `.env`
+- Does not require registration or accounts
+
+**`.env` is in `.gitignore`** — if you accidentally run `git add .`, your credentials won't be committed.
+
+## Tech Stack
+
+- TypeScript / Node.js
+- Commander.js — CLI
+- jira.js — Jira REST API
+- Octokit — GitHub REST API (PRs, reviews, releases)
+- es-git — native Git bindings (N-API, powered by libgit2)
+- PostgreSQL 16 + pgvector — storage + vector search
+- MCP SDK — Claude integration
+- Docker — database infrastructure
+
+## Roadmap
+
+- [x] Jira pull (all fields, comments, changelogs, worklogs, links)
+- [x] Git pull (commits, per-file diffs, issue cross-references)
+- [x] GitHub pull (PRs, reviews, comments, files, releases, Jira cross-references)
+- [x] MCP server for Claude Desktop / Claude Code (15 tools)
+- [x] Embeddings + semantic search (OpenAI text-embedding-3-small, pgvector)
+- [x] Cross-source timeline (issue_timeline: changelogs + commits + PRs)
+- [x] Multi-repo Git support (multiple repos per workspace, `GIT_REPO_PATHS`)
+- [x] Progress indicators during sync (e.g. `150/506 issues (30%)`)
+- [ ] Database adapter (schema, sample data)
+- [ ] Cross-source analysis (Jira ticket vs actual code vs DB state)
+- [x] CSV import (Jira export without API token)
+
+## License
+
+MIT

package/dist/adapters/csv/index.d.ts

@@ -0,0 +1,4 @@
+export { CsvProvider } from './provider.js';
+export { mapCsvRow, type CsvRowResult } from './mapper.js';
+export { detectSchema, parseJiraDate, type CsvSchema } from './parser.js';
+//# sourceMappingURL=index.d.ts.map

package/dist/adapters/csv/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/adapters/csv/index.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,WAAW,EAAE,MAAM,eAAe,CAAC;AAC5C,OAAO,EAAE,SAAS,EAAE,KAAK,YAAY,EAAE,MAAM,aAAa,CAAC;AAC3D,OAAO,EAAE,YAAY,EAAE,aAAa,EAAE,KAAK,SAAS,EAAE,MAAM,aAAa,CAAC"}

package/dist/adapters/csv/index.js

@@ -0,0 +1,4 @@
+export { CsvProvider } from './provider.js';
+export { mapCsvRow } from './mapper.js';
+export { detectSchema, parseJiraDate } from './parser.js';
+//# sourceMappingURL=index.js.map

package/dist/adapters/csv/index.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../../../src/adapters/csv/index.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,WAAW,EAAE,MAAM,eAAe,CAAC;AAC5C,OAAO,EAAE,SAAS,EAAqB,MAAM,aAAa,CAAC;AAC3D,OAAO,EAAE,YAAY,EAAE,aAAa,EAAkB,MAAM,aAAa,CAAC"}

package/dist/adapters/csv/mapper.d.ts

@@ -0,0 +1,10 @@
+import type { Issue, IssueComment, IssueWorklog, IssueLink } from '../../core/types/index.js';
+import type { CsvSchema } from './parser.js';
+export interface CsvRowResult {
+    issue: Issue;
+    comments: IssueComment[];
+    worklogs: IssueWorklog[];
+    links: IssueLink[];
+}
+export declare function mapCsvRow(row: string[], schema: CsvSchema): CsvRowResult;
+//# sourceMappingURL=mapper.d.ts.map

package/dist/adapters/csv/mapper.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"mapper.d.ts","sourceRoot":"","sources":["../../../src/adapters/csv/mapper.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EACV,KAAK,EACL,YAAY,EACZ,YAAY,EACZ,SAAS,EACV,MAAM,2BAA2B,CAAC;AACnC,OAAO,KAAK,EAAE,SAAS,EAAE,MAAM,aAAa,CAAC;AAG7C,MAAM,WAAW,YAAY;IAC3B,KAAK,EAAE,KAAK,CAAC;IACb,QAAQ,EAAE,YAAY,EAAE,CAAC;IACzB,QAAQ,EAAE,YAAY,EAAE,CAAC;IACzB,KAAK,EAAE,SAAS,EAAE,CAAC;CACpB;AAcD,wBAAgB,SAAS,CAAC,GAAG,EAAE,MAAM,EAAE,EAAE,MAAM,EAAE,SAAS,GAAG,YAAY,CAsCxE"}

package/dist/adapters/csv/mapper.js

@@ -0,0 +1,179 @@
+import { parseJiraDate } from './parser.js';
+function cell(row, index) {
+    if (index === undefined) {
+        return null;
+    }
+    const value = row[index]?.trim() ?? '';
+    return value === '' ? null : value;
+}
+function cellStr(row, index) {
+    return cell(row, index) ?? '';
+}
+export function mapCsvRow(row, schema) {
+    const sf = schema.standardFields;
+    const issueKey = cellStr(row, sf.get('Issue key'));
+    const projectKey = issueKey.split('-')[0] ?? '';
+    const issue = {
+        key: issueKey,
+        id: cellStr(row, sf.get('Issue id')),
+        projectKey,
+        summary: cellStr(row, sf.get('Summary')),
+        description: cell(row, sf.get('Description')),
+        issueType: cell(row, sf.get('Issue Type')),
+        status: cell(row, sf.get('Status')),
+        statusCategory: cell(row, sf.get('Status Category')),
+        priority: cell(row, sf.get('Priority')),
+        resolution: cell(row, sf.get('Resolution')),
+        assignee: cell(row, sf.get('Assignee')),
+        reporter: cell(row, sf.get('Reporter')),
+        created: parseJiraDate(cell(row, sf.get('Created'))),
+        updated: parseJiraDate(cell(row, sf.get('Updated'))),
+        resolved: parseJiraDate(cell(row, sf.get('Resolved'))),
+        dueDate: cell(row, sf.get('Due date')),
+        labels: extractRepeated(row, schema, 'Labels'),
+        components: extractRepeated(row, schema, 'Components'),
+        fixVersions: extractRepeated(row, schema, 'Fix versions'),
+        parentKey: cell(row, sf.get('Parent key')),
+        sprint: null,
+        storyPoints: parseFloat(cell(row, sf.get('Original estimate')) ?? '') || null,
+        customFields: extractCustomFields(row, schema),
+        rawJson: buildRawJson(row, schema),
+    };
+    const comments = extractComments(issueKey, row, schema);
+    const worklogs = extractWorklogs(issueKey, row, schema);
+    const links = extractLinks(issueKey, row, schema);
+    return { issue, comments, worklogs, links };
+}
+function extractRepeated(row, schema, name) {
+    const group = schema.repeatedGroups.get(name);
+    if (!group) {
+        return [];
+    }
+    const result = [];
+    for (let i = group.startIndex; i < group.startIndex + group.count; i++) {
+        const value = row[i]?.trim();
+        if (value) {
+            result.push(value);
+        }
+    }
+    return result;
+}
+function extractCustomFields(row, schema) {
+    const result = {};
+    for (const cf of schema.customFields) {
+        const value = cell(row, cf.columnIndex);
+        if (value !== null) {
+            result[cf.name] = value;
+        }
+    }
+    return result;
+}
+function buildRawJson(row, schema) {
+    const raw = {};
+    for (const [name, index] of schema.standardFields) {
+        raw[name] = cell(row, index);
+    }
+    for (const cf of schema.customFields) {
+        raw[`Custom field (${cf.name})`] = cell(row, cf.columnIndex);
+    }
+    return raw;
+}
+/**
+ * Jira CSV comment format: `timestamp;user_id;comment_text`
+ * Split on first 2 semicolons — comment body may contain semicolons.
+ */
+function extractComments(issueKey, row, schema) {
+    const group = schema.repeatedGroups.get('Comment');
+    if (!group) {
+        return [];
+    }
+    const result = [];
+    for (let i = group.startIndex; i < group.startIndex + group.count; i++) {
+        const raw = row[i]?.trim();
+        if (!raw) {
+            continue;
+        }
+        const firstSemi = raw.indexOf(';');
+        if (firstSemi === -1) {
+            continue;
+        }
+        const secondSemi = raw.indexOf(';', firstSemi + 1);
+        if (secondSemi === -1) {
+            continue;
+        }
+        const timestamp = raw.slice(0, firstSemi);
+        const author = raw.slice(firstSemi + 1, secondSemi);
+        const body = raw.slice(secondSemi + 1);
+        result.push({
+            issueKey,
+            commentId: `csv-${issueKey}-comment-${i}`,
+            author: author || null,
+            body: body || null,
+            created: parseJiraDate(timestamp),
+            updated: null,
+        });
+    }
+    return result;
+}
+/**
+ * Jira CSV worklog format: `description;timestamp;user;minutes`
+ * Minutes × 60 = seconds.
+ */
+function extractWorklogs(issueKey, row, schema) {
+    const group = schema.repeatedGroups.get('Log Work');
+    if (!group) {
+        return [];
+    }
+    const result = [];
+    for (let i = group.startIndex; i < group.startIndex + group.count; i++) {
+        const raw = row[i]?.trim();
+        if (!raw) {
+            continue;
+        }
+        const parts = raw.split(';');
+        if (parts.length < 4) {
+            continue;
+        }
+        const description = parts[0] ?? '';
+        const timestamp = parts[1] ?? '';
+        const author = parts[2] ?? '';
+        const minutesStr = parts[parts.length - 1] ?? '0';
+        const minutes = parseInt(minutesStr, 10);
+        result.push({
+            issueKey,
+            author: author || null,
+            timeSpent: isNaN(minutes) ? null : `${minutes}m`,
+            timeSpentSeconds: isNaN(minutes) ? null : minutes * 60,
+            comment: description || null,
+            started: parseJiraDate(timestamp),
+        });
+    }
+    return result;
+}
+function extractLinks(issueKey, row, schema) {
+    const result = [];
+    for (const link of schema.issueLinks) {
+        const targetKey = row[link.columnIndex]?.trim();
+        if (!targetKey) {
+            continue;
+        }
+        if (link.direction === 'outward') {
+            result.push({
+                sourceKey: issueKey,
+                targetKey,
+                linkType: link.linkType,
+                direction: 'outward',
+            });
+        }
+        else {
+            result.push({
+                sourceKey: targetKey,
+                targetKey: issueKey,
+                linkType: link.linkType,
+                direction: 'inward',
+            });
+        }
+    }
+    return result;
+}
+//# sourceMappingURL=mapper.js.map

package/dist/adapters/csv/mapper.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"mapper.js","sourceRoot":"","sources":["../../../src/adapters/csv/mapper.ts"],"names":[],"mappings":"AAOA,OAAO,EAAE,aAAa,EAAE,MAAM,aAAa,CAAC;AAS5C,SAAS,IAAI,CAAC,GAAa,EAAE,KAAyB;IACpD,IAAI,KAAK,KAAK,SAAS,EAAE,CAAC;QACxB,OAAO,IAAI,CAAC;IACd,CAAC;IACD,MAAM,KAAK,GAAG,GAAG,CAAC,KAAK,CAAC,EAAE,IAAI,EAAE,IAAI,EAAE,CAAC;IACvC,OAAO,KAAK,KAAK,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,KAAK,CAAC;AACrC,CAAC;AAED,SAAS,OAAO,CAAC,GAAa,EAAE,KAAyB;IACvD,OAAO,IAAI,CAAC,GAAG,EAAE,KAAK,CAAC,IAAI,EAAE,CAAC;AAChC,CAAC;AAED,MAAM,UAAU,SAAS,CAAC,GAAa,EAAE,MAAiB;IACxD,MAAM,EAAE,GAAG,MAAM,CAAC,cAAc,CAAC;IAEjC,MAAM,QAAQ,GAAG,OAAO,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,WAAW,CAAC,CAAC,CAAC;IACnD,MAAM,UAAU,GAAG,QAAQ,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;IAEhD,MAAM,KAAK,GAAU;QACnB,GAAG,EAAE,QAAQ;QACb,EAAE,EAAE,OAAO,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,UAAU,CAAC,CAAC;QACpC,UAAU;QACV,OAAO,EAAE,OAAO,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC;QACxC,WAAW,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,aAAa,CAAC,CAAC;QAC7C,SAAS,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC;QAC1C,MAAM,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC;QACnC,cAAc,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,iBAAiB,CAAC,CAAC;QACpD,QAAQ,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,UAAU,CAAC,CAAC;QACvC,UAAU,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC;QAC3C,QAAQ,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,UAAU,CAAC,CAAC;QACvC,QAAQ,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,UAAU,CAAC,CAAC;QACvC,OAAO,EAAE,aAAa,CAAC,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC,CAAC;QACpD,OAAO,EAAE,aAAa,CAAC,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC,CAAC;QACpD,QAAQ,EAAE,aAAa,CAAC,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,UAAU,CAAC,CAAC,CAAC;QACtD,OAAO,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,UAAU,CAAC,CAAC;QACtC,MAAM,EAAE,eAAe,CAAC,GAAG,EAAE,MAAM,EAAE,QAAQ,CAAC;QAC9C,UAAU,EAAE,eAAe,CAAC,GAAG,EAAE,MAAM,EAAE,YAAY,CAAC;QACtD,WAAW,EAAE,eAAe,CAAC,GAAG,EAAE,MAAM,EAAE,cAAc,CAAC;QACzD,SAAS,EAAE,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC;QAC1C,MAAM,EAAE,IAAI;QACZ,WAAW,EAAE,UAAU,CAAC,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,GAAG,CAAC,mBAAmB,CAAC,CAAC,IAAI,EAAE,CAAC,IAAI,IAAI;QAC7E,YAAY,EAAE,mBAAmB,CAAC,GAAG,EAAE,MAAM,CAAC;QAC9C,OAAO,EAAE,YAAY,CAAC,GAAG,EAAE,MAAM,CAAC;KACnC,CAAC;IAEF,MAAM,QAAQ,GAAG,eAAe,CAAC,QAAQ,EAAE,GAAG,EAAE,MAAM,CAAC,CAAC;IACxD,MAAM,QAAQ,GAAG,eAAe,CAAC,QAAQ,EAAE,GAAG,EAAE,MAAM,CAAC,CAAC;IACxD,MAAM,KAAK,GAAG,YAAY,CAAC,QAAQ,EAAE,GAAG,EAAE,MAAM,CAAC,CAAC;IAElD,OAAO,EAAE,KAAK,EAAE,QAAQ,EAAE,QAAQ,EAAE,KAAK,EAAE,CAAC;AAC9C,CAAC;AAED,SAAS,eAAe,CAAC,GAAa,EAAE,MAAiB,EAAE,IAAY;IACrE,MAAM,KAAK,GAAG,MAAM,CAAC,cAAc,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC;IAC9C,IAAI,CAAC,KAAK,EAAE,CAAC;QACX,OAAO,EAAE,CAAC;IACZ,CAAC;IACD,MAAM,MAAM,GAAa,EAAE,CAAC;IAC5B,KAAK,IAAI,CAAC,GAAG,KAAK,CAAC,UAAU,EAAE,CAAC,GAAG,KAAK,CAAC,UAAU,GAAG,KAAK,CAAC,KAAK,EAAE,CAAC,EAAE,EAAE,CAAC;QACvE,MAAM,KAAK,GAAG,GAAG,CAAC,CAAC,CAAC,EAAE,IAAI,EAAE,CAAC;QAC7B,IAAI,KAAK,EAAE,CAAC;YACV,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;QACrB,CAAC;IACH,CAAC;IACD,OAAO,MAAM,CAAC;AAChB,CAAC;AAED,SAAS,mBAAmB,CAAC,GAAa,EAAE,MAAiB;IAC3D,MAAM,MAAM,GAA4B,EAAE,CAAC;IAC3C,KAAK,MAAM,EAAE,IAAI,MAAM,CAAC,YAAY,EAAE,CAAC;QACrC,MAAM,KAAK,GAAG,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,WAAW,CAAC,CAAC;QACxC,IAAI,KAAK,KAAK,IAAI,EAAE,CAAC;YACnB,MAAM,CAAC,EAAE,CAAC,IAAI,CAAC,GAAG,KAAK,CAAC;QAC1B,CAAC;IACH,CAAC;IACD,OAAO,MAAM,CAAC;AAChB,CAAC;AAED,SAAS,YAAY,CAAC,GAAa,EAAE,MAAiB;IACpD,MAAM,GAAG,GAA4B,EAAE,CAAC;IACxC,KAAK,MAAM,CAAC,IAAI,EAAE,KAAK,CAAC,IAAI,MAAM,CAAC,cAAc,EAAE,CAAC;QAClD,GAAG,CAAC,IAAI,CAAC,GAAG,IAAI,CAAC,GAAG,EAAE,KAAK,CAAC,CAAC;IAC/B,CAAC;IACD,KAAK,MAAM,EAAE,IAAI,MAAM,CAAC,YAAY,EAAE,CAAC;QACrC,GAAG,CAAC,iBAAiB,EAAE,CAAC,IAAI,GAAG,CAAC,GAAG,IAAI,CAAC,GAAG,EAAE,EAAE,CAAC,WAAW,CAAC,CAAC;IAC/D,CAAC;IACD,OAAO,GAAG,CAAC;AACb,CAAC;AAED;;;GAGG;AACH,SAAS,eAAe,CAAC,QAAgB,EAAE,GAAa,EAAE,MAAiB;IACzE,MAAM,KAAK,GAAG,MAAM,CAAC,cAAc,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC;IACnD,IAAI,CAAC,KAAK,EAAE,CAAC;QACX,OAAO,EAAE,CAAC;IACZ,CAAC;IACD,MAAM,MAAM,GAAmB,EAAE,CAAC;IAClC,KAAK,IAAI,CAAC,GAAG,KAAK,CAAC,UAAU,EAAE,CAAC,GAAG,KAAK,CAAC,UAAU,GAAG,KAAK,CAAC,KAAK,EAAE,CAAC,EAAE,EAAE,CAAC;QACvE,MAAM,GAAG,GAAG,GAAG,CAAC,CAAC,CAAC,EAAE,IAAI,EAAE,CAAC;QAC3B,IAAI,CAAC,GAAG,EAAE,CAAC;YACT,SAAS;QACX,CAAC;QACD,MAAM,SAAS,GAAG,GAAG,CAAC,OAAO,CAAC,GAAG,CAAC,CAAC;QACnC,IAAI,SAAS,KAAK,CAAC,CAAC,EAAE,CAAC;YACrB,SAAS;QACX,CAAC;QACD,MAAM,UAAU,GAAG,GAAG,CAAC,OAAO,CAAC,GAAG,EAAE,SAAS,GAAG,CAAC,CAAC,CAAC;QACnD,IAAI,UAAU,KAAK,CAAC,CAAC,EAAE,CAAC;YACtB,SAAS;QACX,CAAC;QAED,MAAM,SAAS,GAAG,GAAG,CAAC,KAAK,CAAC,CAAC,EAAE,SAAS,CAAC,CAAC;QAC1C,MAAM,MAAM,GAAG,GAAG,CAAC,KAAK,CAAC,SAAS,GAAG,CAAC,EAAE,UAAU,CAAC,CAAC;QACpD,MAAM,IAAI,GAAG,GAAG,CAAC,KAAK,CAAC,UAAU,GAAG,CAAC,CAAC,CAAC;QAEvC,MAAM,CAAC,IAAI,CAAC;YACV,QAAQ;YACR,SAAS,EAAE,OAAO,QAAQ,YAAY,CAAC,EAAE;YACzC,MAAM,EAAE,MAAM,IAAI,IAAI;YACtB,IAAI,EAAE,IAAI,IAAI,IAAI;YAClB,OAAO,EAAE,aAAa,CAAC,SAAS,CAAC;YACjC,OAAO,EAAE,IAAI;SACd,CAAC,CAAC;IACL,CAAC;IACD,OAAO,MAAM,CAAC;AAChB,CAAC;AAED;;;GAGG;AACH,SAAS,eAAe,CAAC,QAAgB,EAAE,GAAa,EAAE,MAAiB;IACzE,MAAM,KAAK,GAAG,MAAM,CAAC,cAAc,CAAC,GAAG,CAAC,UAAU,CAAC,CAAC;IACpD,IAAI,CAAC,KAAK,EAAE,CAAC;QACX,OAAO,EAAE,CAAC;IACZ,CAAC;IACD,MAAM,MAAM,GAAmB,EAAE,CAAC;IAClC,KAAK,IAAI,CAAC,GAAG,KAAK,CAAC,UAAU,EAAE,CAAC,GAAG,KAAK,CAAC,UAAU,GAAG,KAAK,CAAC,KAAK,EAAE,CAAC,EAAE,EAAE,CAAC;QACvE,MAAM,GAAG,GAAG,GAAG,CAAC,CAAC,CAAC,EAAE,IAAI,EAAE,CAAC;QAC3B,IAAI,CAAC,GAAG,EAAE,CAAC;YACT,SAAS;QACX,CAAC;QACD,MAAM,KAAK,GAAG,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;QAC7B,IAAI,KAAK,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACrB,SAAS;QACX,CAAC;QAED,MAAM,WAAW,GAAG,KAAK,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;QACnC,MAAM,SAAS,GAAG,KAAK,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;QACjC,MAAM,MAAM,GAAG,KAAK,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;QAC9B,MAAM,UAAU,GAAG,KAAK,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAC,IAAI,GAAG,CAAC;QAClD,MAAM,OAAO,GAAG,QAAQ,CAAC,UAAU,EAAE,EAAE,CAAC,CAAC;QAEzC,MAAM,CAAC,IAAI,CAAC;YACV,QAAQ;YACR,MAAM,EAAE,MAAM,IAAI,IAAI;YACtB,SAAS,EAAE,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,GAAG,OAAO,GAAG;YAChD,gBAAgB,EAAE,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,OAAO,GAAG,EAAE;YACtD,OAAO,EAAE,WAAW,IAAI,IAAI;YAC5B,OAAO,EAAE,aAAa,CAAC,SAAS,CAAC;SAClC,CAAC,CAAC;IACL,CAAC;IACD,OAAO,MAAM,CAAC;AAChB,CAAC;AAED,SAAS,YAAY,CAAC,QAAgB,EAAE,GAAa,EAAE,MAAiB;IACtE,MAAM,MAAM,GAAgB,EAAE,CAAC;IAC/B,KAAK,MAAM,IAAI,IAAI,MAAM,CAAC,UAAU,EAAE,CAAC;QACrC,MAAM,SAAS,GAAG,GAAG,CAAC,IAAI,CAAC,WAAW,CAAC,EAAE,IAAI,EAAE,CAAC;QAChD,IAAI,CAAC,SAAS,EAAE,CAAC;YACf,SAAS;QACX,CAAC;QACD,IAAI,IAAI,CAAC,SAAS,KAAK,SAAS,EAAE,CAAC;YACjC,MAAM,CAAC,IAAI,CAAC;gBACV,SAAS,EAAE,QAAQ;gBACnB,SAAS;gBACT,QAAQ,EAAE,IAAI,CAAC,QAAQ;gBACvB,SAAS,EAAE,SAAS;aACrB,CAAC,CAAC;QACL,CAAC;aAAM,CAAC;YACN,MAAM,CAAC,IAAI,CAAC;gBACV,SAAS,EAAE,SAAS;gBACpB,SAAS,EAAE,QAAQ;gBACnB,QAAQ,EAAE,IAAI,CAAC,QAAQ;gBACvB,SAAS,EAAE,QAAQ;aACpB,CAAC,CAAC;QACL,CAAC;IACH,CAAC;IACD,OAAO,MAAM,CAAC;AAChB,CAAC"}