Initial stab at building a tags index (#1)

* Initial stab at building a tags index
* Revise with Kirby's feedback

Author: i80and

Makefile

@@ -1,12 +1,14 @@
+PYTHON=python
 GIT_BRANCH=`git rev-parse --abbrev-ref HEAD`
 
-.PHONY: all serve help
+.PHONY: build serve help
 
 help: ## Show this help message
 	@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | sort | awk 'BEGIN {FS = ":.*?## "}; {printf "\033[36m%-20s\033[0m %s\n", $$1, $$2}'
 
-all: ## Build the documentation under build/<git branch>
+build: ## Build the documentation under build/<git branch>
 	hugo -d build/$(GIT_BRANCH)
+	$(PYTHON) -B tools/genindex.py --out build/$(GIT_BRANCH)/tags.json --config config.toml content/tutorials/
 
 serve: ## Host the documentation on port 1313
 	hugo serve

config.toml

@@ -2,3 +2,17 @@ baseurl = "http://example.org/"
 languageCode = "en-us"
 title = "MongoDB Tutorials"
 theme = "mongodb"
+
+[taxonomies]
+  category = "" 
+  tags = ""
+
+[tags]
+  python = "language"
+  ruby = "language"
+  compass = "product"
+  bi = "product"
+  mongodb = "product"
+  driver = "product"
+  beginner = "level"
+  crud = "topic"

content/index.mmark

@@ -1,7 +1,6 @@
 +++
 date = "2017-02-13T12:12:50-05:00"
 title = "index"
-tags = []
 +++
 
 # MongoDB Tutorials

tools/genindex.py

@@ -0,0 +1,70 @@
+#!/usr/bin/env python
+import argparse
+import json
+import os
+import re
+import sys
+import toml
+import logging
+
+logger = logging.getLogger('genindex')
+PAT_HEADMATTER = re.compile(r'^\+\+\+\n(.+)\n\+\+\+', re.DOTALL)
+
+
+def process_file(path):
+    with open(path, 'r') as f:
+        rawdata = f.read()
+
+    match = PAT_HEADMATTER.match(rawdata)
+    if not match:
+        raise ValueError('Couldn\'t find headmatter')
+
+    data = toml.loads(match.group(1))
+    title = data['title']
+    slug = data.get('slug', os.path.splitext(os.path.basename(path))[0])
+    tags = data['tags']
+
+    return (slug, title, tags)
+
+
+def main(args):
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument('source', help='Source directory under which to find pages.')
+    parser.add_argument('--out', metavar='PATH', help='Path in which to save the tag index.')
+    parser.add_argument('--config', metavar='PATH', help='Path to the project configuration file.')
+    args = parser.parse_args()
+
+    logging.basicConfig()
+    data = []
+
+    with open(args.config, 'r') as f:
+        tag_manifest = toml.load(f).get('tags', {})
+
+    error = False
+    for root, _, files in os.walk(args.source):
+        for filename in files:
+            filename = os.path.join(root, filename)
+            try:
+                slug, title, tags = process_file(filename)
+            except ValueError:
+                logger.exception('Error processing %s', filename)
+                continue
+
+            for tag in tags:
+                if not tag in tag_manifest:
+                    logger.fatal('Unknown tag "%s" in %s', tag, filename)
+                    error = True
+
+            data.append((slug, title, tags))
+
+    if error:
+        sys.exit(1)
+
+    with open(args.out, 'w') as f:
+        json.dump({
+            'tags': tag_manifest,
+            'pages': data
+        }, f)
+
+if __name__ == '__main__':
+    main(sys.argv)