Merge pull request #416 from python/create-issues-by-pageviews

humitos · web-flow · commit 82bf0f09b6b8 · 2020-06-20T15:15:39.000+02:00
diff --git a/scripts/create_issue.py b/scripts/create_issue.py
@@ -1,3 +1,6 @@
+# Use together with `pageviews.py`
+# python scripts/pageviews.py | head -n 150 | grep -v whats | cut -d ' ' -f 2 | sed 's/\.html/\.po/g' | xargs -I '{}' python scripts/create_issue.py '{}'
+
 import os
 import sys
 from pathlib import Path
@@ -17,17 +20,29 @@
 repo = g.get_repo('PyCampES/python-docs-es')
 
 
-issues = repo.get_issues(state='open')
+issues = repo.get_issues(state='all')
 for issue in issues:
     if pofilename in issue.title:
+
+        print(f'Skipping {pofilename}. There is a similar issue already created at {issue.html_url}')
+        sys.exit(1)
+
         msg = f'There is a similar issue already created at {issue.html_url}.\nDo you want to create it anyways? [y/N] '
         answer = input(msg)
         if answer != 'y':
             sys.exit(1)
 
+if any([
+    pofile.translated_nb == pofile.po_file_size,
+    pofile.untranslated_nb == 0,
+]):
+    print(f'Skipping {pofilename}. The file is 100% translated already.')
+    sys.exit(1)
+
 # https://pygithub.readthedocs.io/en/latest/github_objects/Repository.html#github.Repository.Repository.create_issue
+title = f'Translate `{pofilename}`'
 issue = repo.create_issue(
-    title=f'Translate `{pofilename}`',
+    title=title,
     body=f'''This needs to reach 100% translated.
 
 Current stats for `{pofilename}`:
@@ -41,4 +56,4 @@
 
 Remember to follow the steps in our [Contributing Guide](https://python-docs-es.readthedocs.io/page/CONTRIBUTING.html).''',
 )
-print(f'Issue created at {issue.html_url}')
+print(f'Issue "{title}" created at {issue.html_url}')
diff --git a/scripts/pageviews.py b/scripts/pageviews.py
@@ -0,0 +1,18 @@
+from pprint import pprint
+
+results = {}
+
+# page-requests-20200504-20200510.txt
+# is a file containing stats from pageviews on the official Python documentation
+# (including different versions and languages)
+# grep -E '\.html$' page-requests-20200504-20200510.txt | grep -v tutorial | sed 's/3\..\///g' | sed 's/3\///g' | sed 's/2\///g' > pageviews.txt
+pages = open('pageviews.txt').readlines()[:-1]
+for p in pages:
+    count, key = int(p.split()[0]), p.split()[-1].strip()
+    if key in results:
+        results[key] += count
+    else:
+        results[key] = count
+
+for p in sorted(list(results.items()), key=lambda x: x[1], reverse=True)[50:100]:
+    print(p[1], p[0][1:])
