Add summaries

Add README

Name by packages

Author: jmandel

.gitignore

@@ -0,0 +1,175 @@
+# Based on https://raw.githubusercontent.com/github/gitignore/main/Node.gitignore
+
+# Logs
+
+logs
+_.log
+npm-debug.log_
+yarn-debug.log*
+yarn-error.log*
+lerna-debug.log*
+.pnpm-debug.log*
+
+# Caches
+
+.cache
+
+# Diagnostic reports (https://nodejs.org/api/report.html)
+
+report.[0-9]_.[0-9]_.[0-9]_.[0-9]_.json
+
+# Runtime data
+
+pids
+_.pid
+_.seed
+*.pid.lock
+
+# Directory for instrumented libs generated by jscoverage/JSCover
+
+lib-cov
+
+# Coverage directory used by tools like istanbul
+
+coverage
+*.lcov
+
+# nyc test coverage
+
+.nyc_output
+
+# Grunt intermediate storage (https://gruntjs.com/creating-plugins#storing-task-files)
+
+.grunt
+
+# Bower dependency directory (https://bower.io/)
+
+bower_components
+
+# node-waf configuration
+
+.lock-wscript
+
+# Compiled binary addons (https://nodejs.org/api/addons.html)
+
+build/Release
+
+# Dependency directories
+
+node_modules/
+jspm_packages/
+
+# Snowpack dependency directory (https://snowpack.dev/)
+
+web_modules/
+
+# TypeScript cache
+
+*.tsbuildinfo
+
+# Optional npm cache directory
+
+.npm
+
+# Optional eslint cache
+
+.eslintcache
+
+# Optional stylelint cache
+
+.stylelintcache
+
+# Microbundle cache
+
+.rpt2_cache/
+.rts2_cache_cjs/
+.rts2_cache_es/
+.rts2_cache_umd/
+
+# Optional REPL history
+
+.node_repl_history
+
+# Output of 'npm pack'
+
+*.tgz
+
+# Yarn Integrity file
+
+.yarn-integrity
+
+# dotenv environment variable files
+
+.env
+.env.development.local
+.env.test.local
+.env.production.local
+.env.local
+
+# parcel-bundler cache (https://parceljs.org/)
+
+.parcel-cache
+
+# Next.js build output
+
+.next
+out
+
+# Nuxt.js build / generate output
+
+.nuxt
+dist
+
+# Gatsby files
+
+# Comment in the public line in if your project uses Gatsby and not Next.js
+
+# https://nextjs.org/blog/next-9-1#public-directory-support
+
+# public
+
+# vuepress build output
+
+.vuepress/dist
+
+# vuepress v2.x temp and cache directory
+
+.temp
+
+# Docusaurus cache and generated files
+
+.docusaurus
+
+# Serverless directories
+
+.serverless/
+
+# FuseBox cache
+
+.fusebox/
+
+# DynamoDB Local files
+
+.dynamodb/
+
+# TernJS port file
+
+.tern-port
+
+# Stores VSCode versions used for testing VSCode extensions
+
+.vscode-test
+
+# yarn v2
+
+.yarn/cache
+.yarn/unplugged
+.yarn/build-state.yml
+.yarn/install-state.gz
+.pnp.*
+
+# IntelliJ based IDEs
+.idea
+
+# Finder (MacOS) folder config
+.DS_Store

README.md

@@ -0,0 +1,90 @@
+# HL7 FHIR Implementation Guide Analyzer
+
+This project contains scripts to fetch, analyze, and summarize HL7 FHIR Implementation Guides (IGs).
+
+## Prerequisites
+
+- [bun](https://bun.sh/) (latest version)
+- Git
+- Google Cloud SDK (for Vertex AI)
+- Anthropic API key (for Claude)
+
+## Setup
+
+1. Clone this repository:
+   ```
+   git clone <repository-url>
+   cd <repository-name>
+   ```
+
+2. Install dependencies:
+   ```
+   bun install
+   ```
+
+3. Set up Google Cloud credentials:
+   ```
+   gcloud auth application-default login
+   ```
+
+4. Set your Anthropic API key as an environment variable:
+   ```
+   export ANTHROPIC_API_KEY=your_api_key_here
+   ```
+
+## Usage
+
+### 1. Fetch HL7 Repositories
+
+Run the `hl7repo.ts` script to fetch HL7 repositories:
+
+```
+bun run hl7repo.ts
+```
+
+This script will:
+- Fetch a list of HL7 repositories from https://build.fhir.org/ig/HL7/
+- Create a `repos` directory
+- Shallow clone each repository into the `repos` directory
+
+### 2. Process Repositories
+
+After fetching the repositories, you can process them using the `process_repo.ts` script:
+
+```
+bun run process_repo.ts repos/<repository-name>
+```
+
+Replace `<repository-name>` with the name of the repository you want to process.
+
+This script will:
+- Concatenate relevant files from the repository's `input` directory
+- Generate an initial analysis using Google's Vertex AI
+- Refine the analysis using Anthropic's Claude AI
+- Save the results in the following directories:
+  - `prompts`: Contains the prompts used for AI analysis
+  - `analysis`: Contains the initial AI analysis
+  - `summaries`: Contains the final refined summaries
+
+## Output
+
+- `repos/`: Contains the cloned HL7 repositories
+- `prompts/`: Contains the prompts used for AI analysis
+- `analysis/`: Contains the initial AI analysis for each IG
+- `summaries/`: Contains the final refined summaries for each IG
+
+Each summary in the `summaries/` directory is a markdown file named after the IG, containing a concise, plain-language description of the IG's purpose, features, and relevance.
+
+## Notes
+
+- The scripts use AI models, so results will vary between runs.
+- Ensure you have sufficient permissions and API quota for Google Cloud and Anthropic services.
+- Processing large repositories may take some time and consume API resources.
+
+## Contributing
+
+Contributions are welcome! Please feel free to submit a Pull Request.
+
+## License
+
+This project is licensed under the MIT License.

bun.lockb

@@ -0,0 +1,30 @@
+import { writeFile } from 'fs/promises';
+
+interface IGInfo {
+  repo: string;
+  'package-id': string;
+}
+
+async function fetchAndProcessIGs() {
+  try {
+    const response = await fetch('https://build.fhir.org/ig/qas.json');
+    const data: IGInfo[] = await response.json();
+
+    const repoToPackage: Record<string, string> = {};
+
+    data.forEach((ig) => {
+      const repoPath = ig.repo.split('/');
+      if (repoPath[0] === 'HL7') {
+        const repoName = repoPath[1];
+        repoToPackage[repoName] = ig['package-id'];
+      }
+    });
+
+    await writeFile('repository_to_package.json', JSON.stringify(repoToPackage, null, 2));
+    console.log('repository_to_package.json has been created successfully.');
+  } catch (error) {
+    console.error('Error fetching or processing data:', error);
+  }
+}
+
+fetchAndProcessIGs();

generate-repository-to-package.ts

@@ -0,0 +1,50 @@
+import { parse } from 'node-html-parser';
+import { exec } from 'child_process';
+import { promises as fs } from 'fs';
+import path from 'path';
+
+const url = 'https://build.fhir.org/ig/HL7/';
+const reposDir = 'repos';
+
+async function fetchHL7Repos() {
+  try {
+    const response = await fetch(url);
+    const html = await response.text();
+    const root = parse(html);
+
+    const repos = root.querySelectorAll('tr.file')
+      .map(tr => tr.querySelector('.name'))
+      .filter(nameEl => nameEl !== null)
+      .map(nameEl => nameEl.textContent.trim())
+      .filter(name => name.endsWith('/'))
+      .map(name => name.slice(0, -1)) // Remove trailing slash
+      .filter(name => !name.toLowerCase().includes('template')); // Filter out repos with 'template' in the name
+
+    const githubUrls = repos.map(repo => `https://github.com/HL7/${repo}`);
+
+    // Create repos directory if it doesn't exist
+    await fs.mkdir(reposDir, { recursive: true });
+
+    for (const url of githubUrls) {
+      const repoName = url.split('/').pop();
+      const repoPath = path.join(reposDir, repoName);
+
+      console.log(`Shallow cloning ${url}...`);
+      await new Promise((resolve, reject) => {
+        exec(`git clone --depth 1 ${url} ${repoPath}`, (error, stdout, stderr) => {
+          if (error) {
+            console.error(`Error cloning ${url}:`, error);
+            reject(error);
+          } else {
+            console.log(`Successfully shallow cloned ${url}`);
+            resolve(stdout);
+          }
+        });
+      });
+    }
+  } catch (error) {
+    console.error('Error fetching or parsing the page:', error);
+  }
+}
+
+fetchHL7Repos();

hl7repo.ts

@@ -9,6 +9,7 @@
     "typescript": "^5.0.0"
   },
   "dependencies": {
+    "@anthropic-ai/sdk": "^0.27.3",
     "@google-cloud/vertexai": "^1.7.0",
     "node-html-parser": "^6.1.13"
   }