commit 5b5949c05983a8d927bc44346b62fce5058064da
Author: Zsolt Tasnadi <rastasi@gmail.com>
Date:   Wed Mar 4 21:17:36 2026 +0100

    initial commit

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..55e02b8
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,4 @@
+.env
+BLOGPOST.md
+TRANSLATED_BLOGPOST.md
+SOURCE.md
diff --git a/INSTRUCTIONS.md b/INSTRUCTIONS.md
new file mode 100644
index 0000000..2ec75d2
--- /dev/null
+++ b/INSTRUCTIONS.md
@@ -0,0 +1 @@
+Írj egy kedves, pozitív hangvételű blogpost-ot ami tájékoztat arról, hogy épp min dolgozunk.
\ No newline at end of file
diff --git a/Makefile b/Makefile
new file mode 100644
index 0000000..afae97d
--- /dev/null
+++ b/Makefile
@@ -0,0 +1,23 @@
+ENV = export $(shell cat .env | grep -v '^\#' | grep -v '^$$' | xargs)
+
+.PHONY: fetch write translate upload all
+
+## Letölt egy wiki oldalt SOURCE.md-be
+## Használat: make fetch URL=/path/to/page
+fetch:
+	$(ENV) && python3 generator.py fetch $(URL)
+
+## Blogposztot ír SOURCE.md-ből → BLOGPOST.md
+write:
+	$(ENV) && python3 generator.py write
+
+## Lefordítja BLOGPOST.md → TRANSLATED_BLOGPOST.md
+translate:
+	$(ENV) && python3 generator.py translate
+
+## Feltölti TRANSLATED_BLOGPOST.md-t a wikire
+upload:
+	$(ENV) && python3 generator.py upload
+
+## Teljes pipeline: write → translate → upload
+all: write translate upload
\ No newline at end of file
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..1e86239
--- /dev/null
+++ b/README.md
@@ -0,0 +1,95 @@
+# Wiki.js Blog Writer Pipeline
+
+Automated pipeline that fetches a page from a [Wiki.js](https://js.wiki) instance, turns it into a blog post using Google Gemini, translates it, and publishes it back to the wiki under `/blog/`.
+
+## How it works
+
+```
+Wiki.js page  →  SOURCE.md  →  BLOGPOST.md  →  TRANSLATED_BLOGPOST.md  →  Wiki.js /blog/
+     fetch           write          translate             upload
+```
+
+1. **fetch** — Downloads a Wiki.js page as Markdown via GraphQL
+2. **write** — Gemini reads `INSTRUCTIONS.md` and writes a blog post from `SOURCE.md` in the configured language
+3. **translate** — Gemini translates the blog post into the target language
+4. **upload** — Publishes the translated post to Wiki.js under `/blog/{kebab-case-title}`; creates the page if it doesn't exist, updates it if it does
+
+## Requirements
+
+- Python 3.8+
+- A running Wiki.js instance with API access
+- A Google Gemini API key (get one free at [aistudio.google.com](https://aistudio.google.com))
+
+## Setup
+
+**1. Clone the repo and enter the directory**
+
+```bash
+git clone <your-repo>
+cd blog_writer
+```
+
+**2. Copy the example env file and fill in your values**
+
+```bash
+cp .env.example .env
+```
+
+| Variable | Description |
+|---|---|
+| `WIKI_BASE_DOMAIN` | Base URL of your Wiki.js instance, e.g. `https://wiki.example.com` |
+| `WIKI_TOKEN` | Wiki.js API bearer token (Admin → API Access) |
+| `GEMINI_API_KEY` | Google Gemini API key |
+| `ORIGINAL_LANG` | Language for the generated blog post (default: `Hungarian`) |
+| `TRANSLATE_LANG` | Target language for translation (default: `English`) |
+
+**3. Create an `INSTRUCTIONS.md` file**
+
+This file tells Gemini how to write the blog post — tone, structure, length, style, etc. Example:
+
+```markdown
+Write an engaging blog post aimed at a general technical audience.
+Use a friendly but professional tone. Keep it under 600 words.
+Start with a short hook, then summarize the key points, and end with a call to action.
+Do not use bullet points — write in flowing paragraphs.
+```
+
+## Usage
+
+```bash
+# Fetch a wiki page
+make fetch URL=/path/to/page
+
+# Generate the blog post
+make write
+
+# Translate it
+make translate
+
+# Upload to Wiki.js
+make upload
+
+# Run the full pipeline (write → translate → upload)
+make all
+```
+
+> `make fetch` must be run separately with a `URL` argument before `make all`.
+
+## File overview
+
+| File | Description |
+|---|---|
+| `tool.py` | Main Python script |
+| `Makefile` | Convenience wrapper around `tool.py` |
+| `.env` | Your local environment variables (never commit this) |
+| `.env.example` | Example env file safe to commit |
+| `INSTRUCTIONS.md` | Gemini prompt instructions for blog post style |
+| `SOURCE.md` | Raw Markdown fetched from Wiki.js |
+| `BLOGPOST.md` | Generated blog post in `ORIGINAL_LANG` |
+| `TRANSLATED_BLOGPOST.md` | Translated blog post in `TRANSLATE_LANG` |
+
+## Notes
+
+- The `# H1` title is extracted from `TRANSLATED_BLOGPOST.md` and used as both the Wiki.js page title and the URL slug — it is stripped from the content body to avoid duplication.
+- If a page already exists at the target path, it will be updated rather than duplicated.
+- Add `.env`, `SOURCE.md`, `BLOGPOST.md`, and `TRANSLATED_BLOGPOST.md` to your `.gitignore`.
\ No newline at end of file
diff --git a/env.example b/env.example
new file mode 100644
index 0000000..a06bae0
--- /dev/null
+++ b/env.example
@@ -0,0 +1,7 @@
+WIKI_BASE_DOMAIN=https://wiki.example.com
+WIKI_TOKEN=your_wikijs_bearer_token_here
+
+GEMINI_API_KEY=your_gemini_api_key_here
+
+ORIGINAL_LANG=Hungarian
+TRANSLATE_LANG=English
diff --git a/generator.py b/generator.py
new file mode 100644
index 0000000..878af91
--- /dev/null
+++ b/generator.py
@@ -0,0 +1,316 @@
+#!/usr/bin/env python3
+"""
+Wiki.js → Gemini Blog Post Pipeline
+
+Commands:
+  fetch     Download a Wiki.js page as Markdown via GraphQL
+  write     Generate a blog post from SOURCE.md using Gemini
+  translate Translate BLOGPOST.md using Gemini
+  upload    Upload TRANSLATED_BLOGPOST.md to Wiki.js under /blog/{kebab-title}
+
+Required environment variables:
+  WIKI_BASE_DOMAIN   e.g. https://wiki.example.com
+  WIKI_TOKEN         Bearer token for Wiki.js API
+  GEMINI_API_KEY     Google Gemini API key
+  ORIGINAL_LANG      Language for the blog post (default: Hungarian)
+  TRANSLATE_LANG     Target language for translation (default: English)
+"""
+
+import argparse
+import json
+import os
+import re
+import sys
+import urllib.request
+import urllib.error
+
+# ---------------------------------------------------------------------------
+# Config
+# ---------------------------------------------------------------------------
+
+SOURCE_FILE      = "SOURCE.md"
+BLOGPOST_FILE    = "BLOGPOST.md"
+TRANSLATED_FILE  = "TRANSLATED_BLOGPOST.md"
+INSTRUCTIONS_FILE = "INSTRUCTIONS.md"
+
+GEMINI_MODEL     = "gemini-flash-latest"
+GEMINI_BASE_URL  = "https://generativelanguage.googleapis.com/v1beta/models"
+
+
+# ---------------------------------------------------------------------------
+# Helpers
+# ---------------------------------------------------------------------------
+
+def require_env(name: str, default: str = None) -> str:
+    value = os.environ.get(name, default)
+    if not value:
+        print(f"ERROR: Environment variable '{name}' is required.", file=sys.stderr)
+        sys.exit(1)
+    return value
+
+
+def http_post(url: str, payload: dict, headers: dict) -> dict:
+    data = json.dumps(payload).encode("utf-8")
+    req = urllib.request.Request(url, data=data, headers=headers, method="POST")
+    try:
+        with urllib.request.urlopen(req) as resp:
+            return json.loads(resp.read().decode("utf-8"))
+    except urllib.error.HTTPError as e:
+        body = e.read().decode("utf-8")
+        print(f"ERROR: HTTP {e.code} from {url}\n{body}", file=sys.stderr)
+        sys.exit(1)
+    except urllib.error.URLError as e:
+        print(f"ERROR: Could not reach {url}: {e.reason}", file=sys.stderr)
+        sys.exit(1)
+
+
+def wiki_graphql(base: str, token: str, query: str, variables: dict = None) -> dict:
+    url = f"{base}/graphql"
+    payload = {"query": query}
+    if variables:
+        payload["variables"] = variables
+    headers = {
+        "Authorization": f"Bearer {token}",
+        "Content-Type": "application/json",
+    }
+    return http_post(url, payload, headers)
+
+
+def gemini_generate(api_key: str, prompt: str) -> str:
+    url = f"{GEMINI_BASE_URL}/{GEMINI_MODEL}:generateContent"
+    payload = {"contents": [{"parts": [{"text": prompt}]}]}
+    headers = {"Content-Type": "application/json", "X-goog-api-key": api_key}
+    resp = http_post(url, payload, headers)
+    try:
+        return resp["candidates"][0]["content"]["parts"][0]["text"]
+    except (KeyError, IndexError) as e:
+        print(f"ERROR: Unexpected Gemini response structure: {resp}", file=sys.stderr)
+        sys.exit(1)
+
+
+def to_kebab(text: str) -> str:
+    text = text.lower()
+    text = re.sub(r"[^a-z0-9\s-]", "", text)
+    text = re.sub(r"[\s-]+", "-", text)
+    return text.strip("-")
+
+
+def read_file(path: str) -> str:
+    if not os.path.exists(path):
+        print(f"ERROR: File not found: {path}", file=sys.stderr)
+        sys.exit(1)
+    with open(path, "r", encoding="utf-8") as f:
+        return f.read()
+
+
+def write_file(path: str, content: str) -> None:
+    with open(path, "w", encoding="utf-8") as f:
+        f.write(content)
+    print(f"✓ Saved to {path}")
+
+
+# ---------------------------------------------------------------------------
+# Commands
+# ---------------------------------------------------------------------------
+
+def cmd_fetch(args):
+    """Download a Wiki.js page as Markdown via GraphQL."""
+    base  = require_env("WIKI_BASE_DOMAIN")
+    token = require_env("WIKI_TOKEN")
+
+    # Strip base domain from URL if full URL was given, then strip leading slash
+    page_path = args.url.replace(base, "").lstrip("/")
+    print(f"→ Fetching wiki page: /{page_path}")
+
+    query = """
+    query ($path: String!) {
+      pages {
+        singleByPath(path: $path, locale: "en") {
+          id
+          title
+          description
+          content
+        }
+      }
+    }
+    """
+
+    resp = wiki_graphql(base, token, query, {"path": page_path})
+    page = resp.get("data", {}).get("pages", {}).get("singleByPath")
+
+    if not page:
+        errors = resp.get("errors", resp)
+        print(f"ERROR: Page not found at '{page_path}': {errors}", file=sys.stderr)
+        sys.exit(1)
+
+    write_file(SOURCE_FILE, page["content"])
+
+
+def cmd_write(args):
+    """Generate a blog post from SOURCE.md using Gemini."""
+    api_key       = require_env("GEMINI_API_KEY")
+    original_lang = require_env("ORIGINAL_LANG", "Hungarian")
+
+    instructions = read_file(INSTRUCTIONS_FILE)
+    source       = read_file(SOURCE_FILE)
+
+    print(f"→ Generating blog post in {original_lang} with Gemini...")
+
+    prompt = (
+        "Please read the following instructions carefully and follow them to write a blog post.\n\n"
+        "## INSTRUCTIONS\n\n"
+        f"{instructions}\n\n"
+        "## TASK\n\n"
+        f"Read the source content below and write a blog post from it in {original_lang} language. "
+        "Output only the blog post in Markdown format, with no additional commentary.\n\n"
+        "## SOURCE CONTENT\n\n"
+        f"{source}"
+    )
+
+    result = gemini_generate(api_key, prompt)
+    write_file(BLOGPOST_FILE, result)
+
+
+def cmd_translate(args):
+    """Translate BLOGPOST.md to TRANSLATED_BLOGPOST.md using Gemini."""
+    api_key        = require_env("GEMINI_API_KEY")
+    translate_lang = require_env("TRANSLATE_LANG", "English")
+
+    blogpost = read_file(BLOGPOST_FILE)
+
+    print(f"→ Translating blog post to {translate_lang} with Gemini...")
+
+    prompt = (
+        f"Translate the following Markdown blog post into {translate_lang}. "
+        "Preserve all Markdown formatting, headings, links, and code blocks exactly. "
+        "Output only the translated Markdown with no additional commentary.\n\n"
+        f"{blogpost}"
+    )
+
+    result = gemini_generate(api_key, prompt)
+    write_file(TRANSLATED_FILE, result)
+
+
+def cmd_upload(args):
+    """Upload TRANSLATED_BLOGPOST.md to Wiki.js under /blog/{kebab-title}."""
+    base  = require_env("WIKI_BASE_DOMAIN")
+    token = require_env("WIKI_TOKEN")
+
+    content = read_file(TRANSLATED_FILE)
+
+    # Extract H1 title
+    match = re.search(r"^#\s+(.+)", content, re.MULTILINE)
+    if not match:
+        print(f"ERROR: No H1 heading found in {TRANSLATED_FILE}", file=sys.stderr)
+        sys.exit(1)
+
+    title     = match.group(1).strip()
+    content   = re.sub(r"^#\s+.+\n?", "", content, count=1, flags=re.MULTILINE).lstrip("\n")
+    kebab     = to_kebab(title)
+    page_path = f"blog/{kebab}"
+
+    print(f"→ Uploading to Wiki.js")
+    print(f"  Title : {title}")
+    print(f"  Path  : /{page_path}")
+
+    # Check if page already exists
+    find_query = """
+    query ($path: String!) {
+      pages {
+        singleByPath(path: $path, locale: "en") {
+          id
+        }
+      }
+    }
+    """
+    find_resp = wiki_graphql(base, token, find_query, {"path": page_path})
+    existing  = find_resp.get("data", {}).get("pages", {}).get("singleByPath")
+    existing_id = existing.get("id") if existing else None
+
+    if existing_id:
+        print(f"  Found existing page id={existing_id}, updating...")
+        mutation = """
+        mutation ($id: Int!, $content: String!) {
+          pages {
+            update(id: $id, content: $content, tags: ["blog"]) {
+              responseResult { succeeded message }
+            }
+          }
+        }
+        """
+        variables = {"id": existing_id, "content": content}
+        resp = wiki_graphql(base, token, mutation, variables)
+        result = resp.get("data", {}).get("pages", {}).get("update", {}).get("responseResult", {})
+    else:
+        print("  Page not found, creating new...")
+        mutation = """
+        mutation ($path: String!, $title: String!, $content: String!) {
+          pages {
+            create(
+              path: $path
+              title: $title
+              content: $content
+              editor: "markdown"
+              locale: "en"
+              isPublished: true
+              isPrivate: false
+              tags: ["blog"]
+              description: ""
+            ) {
+              responseResult { succeeded message }
+              page { id }
+            }
+          }
+        }
+        """
+        variables = {"path": page_path, "title": title, "content": content}
+        resp = wiki_graphql(base, token, mutation, variables)
+        result = resp.get("data", {}).get("pages", {}).get("create", {}).get("responseResult", {})
+
+    errors = resp.get("errors")
+    if errors:
+        print(f"ERROR: {json.dumps(errors, indent=2)}", file=sys.stderr)
+        sys.exit(1)
+
+    if not result.get("succeeded"):
+        print(f"ERROR: Operation failed: {result.get('message')}", file=sys.stderr)
+        sys.exit(1)
+
+    print(f"✓ Successfully uploaded to {base}/{page_path}")
+
+
+# ---------------------------------------------------------------------------
+# Entry point
+# ---------------------------------------------------------------------------
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="Wiki.js → Gemini Blog Post Pipeline",
+        formatter_class=argparse.RawDescriptionHelpFormatter,
+        epilog=__doc__,
+    )
+    subparsers = parser.add_subparsers(dest="command", required=True)
+
+    # fetch
+    p_fetch = subparsers.add_parser("fetch", help="Download a Wiki.js page as Markdown")
+    p_fetch.add_argument("url", help="Page path or full URL, e.g. /my-page or https://wiki.example.com/my-page")
+    p_fetch.set_defaults(func=cmd_fetch)
+
+    # write
+    p_write = subparsers.add_parser("write", help=f"Generate blog post from {SOURCE_FILE} using Gemini")
+    p_write.set_defaults(func=cmd_write)
+
+    # translate
+    p_translate = subparsers.add_parser("translate", help=f"Translate {BLOGPOST_FILE} using Gemini")
+    p_translate.set_defaults(func=cmd_translate)
+
+    # upload
+    p_upload = subparsers.add_parser("upload", help=f"Upload {TRANSLATED_FILE} to Wiki.js")
+    p_upload.set_defaults(func=cmd_upload)
+
+    args = parser.parse_args()
+    args.func(args)
+
+
+if __name__ == "__main__":
+    main()
\ No newline at end of file