Capture and report GenAI stats (#55)

Author: DavidSouther

aws_doc_sdk_examples_tools/doc_gen.py

@@ -4,7 +4,9 @@
 import yaml
 import json
 
+from collections import defaultdict
 from dataclasses import dataclass, field, is_dataclass, asdict
+from functools import reduce
 from pathlib import Path
 from typing import Dict, Iterable, Optional, Set
 
@@ -236,6 +238,18 @@ def validate(self):
         )
 
     def stats(self):
+        values = self.examples.values()
+        initial = defaultdict(int)
+
+        def count_genai(d: Dict[str, int], e: Example):
+            for lang in e.languages.values():
+                for version in lang.versions:
+                    for excerpt in version.excerpts:
+                        d[excerpt.genai] += 1
+            return d
+
+        genai = reduce(count_genai, values, initial)
+
         return {
             "sdks": len(self.sdks),
             "services": len(self.services),
@@ -245,6 +259,7 @@ def stats(self):
                 for e in self.examples.values()
             ),
             "snippets": len(self.snippets) + len(self.snippet_files),
+            "genai": genai,
         }
 
 

aws_doc_sdk_examples_tools/stats.py

@@ -1,5 +1,6 @@
 from typing import List
 from pathlib import Path
+from pprint import pformat
 
 from .doc_gen import DocGen
 
@@ -17,6 +18,8 @@ def main(roots: List[str]):
         print(f"Examples	{stats['examples']}")
         print(f"Version	{stats['versions']}")
         print(f"Snippets	{stats['snippets']}")
+        genai = pformat(dict(stats["genai"]))
+        print(f"GenAI	{genai}")
 
 
 if __name__ == "__main__":