Update app.py

app.py

@@ -7,17 +7,24 @@ df = pd.read_csv("./stackv2_languages_freq.csv")
 langs = sorted(list(df["language"].unique()))
 exts = list(df["extension"].unique())
 
-
-lang_to_ext = defaultdict(set)
-
-for lang, ext in zip(df["language"], df["extension"]):
-    if isinstance(ext, str):
-        lang_to_ext[lang].update([ext])
-
 def compute(lang):
-    
-    report = f"## Summary:\n\n The `{lang}` language has {len(lang_to_ext[lang])} extensions: {', '.join(sorted(['`'+ext+'`' for ext in lang_to_ext[lang]]))}.\n\n"
     df_lang = df[df["language"]==lang]
+    # clean up weird exts
+    df_lang = df_lang[df_lang["ext_fraction_per_lang"] > 0.0001].reset_index()
+    
+    df_lang_uniq = df_lang.groupby("extension").first().reset_index()
+    
+    report = f"## Summary:\n\n The `{lang}` language has {df_lang_uniq.shape[0]} extensions: \n\n"
+
+    for i, (ext, ext_fraction, gen_fraction, vend_fraction) in enumerate(zip(df_lang_uniq["extension"], df_lang_uniq["ext_fraction_per_lang"], df_lang_uniq["generated_fraction"], df_lang_uniq["vendor_fraction"])):
+        fractions_string = f"{ext_fraction*100:.2f}%"
+        if gen_fraction > 0.5:
+            fractions_string += f", autogenerated: {gen_fraction*100:.2f}%"
+        if vend_fraction > 0.5:
+            fractions_string += f", vendor files: {vend_fraction*100:.2f}%"
+        
+        report += f"`{ext}` ({fractions_string}), \n\n"
+    report = report[:-2] + "\n\n\n\n"
 
     for i, (ext, example) in enumerate(zip(df_lang["extension"], df_lang["content"])):
         example_string = f"**Example {i+1} (extension=`{ext}`):**\n\n ```\n{example}\n```\n\n"