Fine polish clean recent changes

Author: cbeauchesne

campbot/__main__.py

@@ -2,9 +2,9 @@
 CampBot, Python bot framework for camptocamp.org
 
 Usage:
-  campbot clean_rc <days> [--login=<login>] [--password=<password>] [--delay=<seconds>] [--batch]
+  campbot clean_rc <days> <lang> <thread_url> [--login=<login>] [--password=<password>] [--delay=<seconds>] [--batch]
   campbot report_rc <days> <lang> <thread_url> [--login=<login>] [--password=<password>] [--delay=<seconds>]
-  campbot clean <url_or_file> <langs> [--login=<login>] [--password=<password>] [--delay=<seconds>] [--batch] [--bbcode]
+  campbot clean <url_or_file> <lang> <thread_url> [--login=<login>] [--password=<password>] [--delay=<seconds>] [--batch] [--bbcode]
   campbot report <url_or_file> <lang> [--login=<login>] [--password=<password>] [--delay=<seconds>]
   campbot contribs [--out=<filename>] [--starts=<start_date>] [--ends=<end_date>] [--delay=<seconds>]
   campbot export <url> [--out=<filename>] [--delay=<seconds>]
@@ -28,7 +28,7 @@
                 filename is also accepted, and must be like : 
                 123 | r
                 456 | w
-                <langs> is comma-separated lang identifiers, like fr,de for french and german.
+                <lang> is a lang identifier, like fr for french.
   report        Make quality report on documents.
   contribs      Export all contribution in a CSV file. <start_date> and <end_date> are like 2018-05-12
   export        Export all documents in a CSV file.
@@ -91,7 +91,10 @@ def main(args):
 
     elif args["clean_rc"]:
         get_campbot(args).clean_recent_changes(
-            days=int(args["<days>"]), lang="fr", ask_before_saving=not args["--batch"],
+            days=float(args["<days>"]),
+            lang=args["<lang>"],
+            ask_before_saving=not args["--batch"],
+            thread_url=args["<thread_url>"],
         )
 
     elif args["report"]:
@@ -102,8 +105,9 @@ def main(args):
     elif args["clean"]:
         get_campbot(args).clean(
             args["<url_or_file>"],
-            langs=args["<langs>"].split(","),
+            lang=args["<lang>"],
             ask_before_saving=not args["--batch"],
+            thread_url=args["<thread_url>"],
             clean_bbcode=args["--bbcode"],
         )
 

campbot/checkers.py

@@ -202,7 +202,6 @@ def report_recent_changes(bot, days, lang, thread_url):
     tests = get_fixed_tests(lang)
     tests += get_re_tests(bot.forum.get_post(url=thread_url), lang)
 
-    logging.info(f"Get modified documents from {oldest_date} to {newest_date}")
     items = bot.get_modified_documents(
         lang=lang, oldest_date=oldest_date, newest_date=newest_date
     ).values()

campbot/core.py

@@ -7,7 +7,7 @@
 import requests
 from datetime import datetime, timedelta
 from dateutil import parser
-from collections import OrderedDict
+from collections import OrderedDict, defaultdict
 import pytz
 import logging
 import time
@@ -523,24 +523,24 @@ def _get_documents_from_url(self, url):
         constructor, filters = _parse_filter(url)
         return self.wiki.get_documents(filters, constructor=constructor)
 
-    def clean(self, url_or_filename, langs, ask_before_saving=True, clean_bbcode=False):
+    def clean(
+        self, url_or_filename, lang, ask_before_saving, thread_url, clean_bbcode=False
+    ):
         """
             Clean a set of document.
 
             :param url_or_filename: Camptocamp.org URL, or filename
-            :param langs: comma-separated list of lang identifiers
+            :param lang: lang identifier
             :param ask_before_saving: Boolean
             :param clean_bbcode: Boolean
 
         """
 
-        assert len(langs) != 0
-
         documents = self.get_documents(url_or_filename)
-        processors = get_automatic_replacments(self, clean_bbcode)
+        report_header = f"Clean documents from `{url_or_filename}`"
 
         self._process_documents(
-            documents, processors, langs, ask_before_saving, excluded_ids=[996571,]
+            documents, lang, ask_before_saving, report_header, thread_url, clean_bbcode
         )
 
     def report(self, url_or_filename, lang):
@@ -581,48 +581,77 @@ def report(self, url_or_filename, lang):
         print("\n".join(stdout_report))
 
     def _process_documents(
-        self, documents, processors, langs, ask_before_saving=True, excluded_ids=None
+        self,
+        documents,
+        lang,
+        ask_before_saving,
+        report_header,
+        thread_url,
+        clean_bbcode=False,
     ):
 
-        for document in documents:
+        excluded_document_ids = [
+            996571,  # article with all automatic corrections
+        ]
+
+        processors = get_automatic_replacments(self, clean_bbcode=clean_bbcode)
 
+        report = defaultdict(int)
+
+        for document in documents:
             if "redirects_to" in document:
-                pass  # document id is not available...
+                continue  # document id is not available...
+
+            document_url = document.get_url()
+            report[f"Inspected"] += 1
 
-            elif excluded_ids is not None and document.document_id in excluded_ids:
+            if document.document_id in excluded_document_ids:
                 pass
 
             elif document.get("protected", False) and not self.moderator:
-                print("{} is a protected".format(document.get_url()))
+                logging.info(f"{document_url} is protected")
+                report["Skipped because protected"] += 1
 
             elif document.is_personal() and not self.moderator:
-                print("{} is a personal".format(document.get_url()))
+                logging.info(f"{document_url} is a personal document")
+                report["Skipped because is not CC-BY-SA"] += 1
 
             elif not document.is_valid():
-                print(
-                    "{} : {}".format(
-                        document.get_url(), document.get_invalidity_reason()
-                    )
-                )
+                reason = document.get_invalidity_reason()
+                logging.info(f"{document_url} : {reason}")
+                report[f"Skipped because {reason}"] += 1
 
             else:
                 messages = []
                 must_save = False
 
                 for processor in processors:
                     if processor.ready_for_production:
-                        if processor(document, langs):
+                        if processor(document, [lang,]):
                             messages.append(processor.comment)
                             must_save = True
 
                 if must_save:
                     comment = ", ".join(messages)
                     try:
-                        document.save(comment, ask_before_saving=ask_before_saving)
-                    except Exception as e:
-                        print(
-                            "Error while saving {} :\n{}".format(document.get_url(), e)
+                        new_document = document.save(
+                            comment, ask_before_saving=ask_before_saving
                         )
+                    except Exception as e:
+                        report["Unexpcted error"] += 1
+                        logging.error(f"Error while saving {document_url} :\n{e}")
+                    else:
+                        if new_document is None:
+                            report["Skipped by bot owner"] += 1
+                        else:
+                            report["Corrected"] += 1
+
+        log_report = "\n".join(
+            [f"* `{bucket}`: {count}" for bucket, count in report.items()]
+        )
+        self.forum.post_message(f"### {report_header}\n\n{log_report}", thread_url)
+
+        return report
 
     def export(self, url, filename=None):
         """
@@ -719,6 +748,10 @@ def write(**kwargs):
     def get_modified_documents(
         self, lang, oldest_date=None, newest_date=None, excluded_users=()
     ):
+        logging.info(
+            f"Get modified documents from {oldest_date} to {newest_date} in lang:{lang}"
+        )
+
         result = OrderedDict()
         for contrib in self.wiki.get_contributions(
             oldest_date=oldest_date, newest_date=newest_date
@@ -738,15 +771,11 @@ def get_modified_documents(
 
         return result
 
-    def clean_recent_changes(self, days, lang, ask_before_saving):
+    def clean_recent_changes(self, days, lang, ask_before_saving, thread_url):
         newest_date = utils.today().replace(hour=0, minute=0, second=0, microsecond=0)
         oldest_date = newest_date - timedelta(days=days)
 
-        excluded_ids = [
-            996571,
-        ]
-
-        processors = get_automatic_replacments(self)
+        report_header = f"Clean recent change from `{oldest_date}` to `{newest_date}`"
 
         def get_documents():
 
@@ -761,12 +790,11 @@ def get_documents():
                     document_id, document_type=document_type
                 )
 
-                if document_id not in excluded_ids:
-                    yield document
+                yield document
 
-        print("Fix recent changes")
-        self._process_documents(get_documents(), processors, [lang,], ask_before_saving)
-        print("Fix recent changes finished")
+        self._process_documents(
+            get_documents(), lang, ask_before_saving, report_header, thread_url
+        )
 
     def get_new_contributors(self, contrib_threshold=20, outings_threshold=15):
         with open("contributors.txt", "r") as f:

campbot/objects.py

@@ -25,6 +25,7 @@
 from __future__ import print_function, unicode_literals, division
 
 import re
+import logging
 from .differ import get_diff_report
 
 
@@ -289,7 +290,7 @@ def save(self, message, ask_before_saving=True):
             if _input("Save {} : {}, y/[n] ?\n".format(self.get_url(), message)) != "y":
                 return None
         else:
-            print("Saving {} : {}".format(self.get_url(), message))
+            logging.info(f"Saving {self.get_url()} : {message}")
 
         return self._campbot.wiki.put(
             "/{}/{}".format(self.url_path, self.document_id),

docs/CLI/clean.rst

@@ -9,13 +9,13 @@ Command line
 
 .. code-block:: bash
 
-    campbot clean <url> <langs> --login=<login> --password=<password> [--delay=<seconds>] [--bbcode]
+    campbot clean <url> <lang> --login=<login> --password=<password> [--delay=<seconds>] [--bbcode]
 
 Options and arguments 
 ---------------------
 
 * ``<url>`` is like https://www.camptocamp.org/routes#w=940468 : all routes associated to waypoint 940468 will be cleaned. Shorthand ``routes#w=940468`` is accepted.
-* ``<langs>`` is a comma-saprated list of langs, like fr,de. Clean procedure will impacts only this langs.
+* ``<lang>`` is a lang identifier, like fr or de. Clean procedure will impacts only this lang.
 
 Clean processors
 ----------------

docs/CLI/index.rst

@@ -8,9 +8,9 @@ Here isthe help you get by typing ``campbot`` in you command line tool :
     CampBot, Python bot framework for camptocamp.org
 
     Usage:
-      campbot clean_rc <days> [--login=<login>] [--password=<password>] [--delay=<seconds>] [--batch]
-      campbot report_rc <days> [--login=<login>] [--password=<password>] [--delay=<seconds>] [--batch]
-      campbot clean <url_or_file> <langs> [--login=<login>] [--password=<password>] [--delay=<seconds>] [--batch] [--bbcode]
+      campbot clean_rc <days> <lang> [--login=<login>] [--password=<password>] [--delay=<seconds>] [--batch]
+      campbot report_rc <days> <lang> [--login=<login>] [--password=<password>] [--delay=<seconds>] [--batch]
+      campbot clean <url_or_file> <lang> [--login=<login>] [--password=<password>] [--delay=<seconds>] [--batch] [--bbcode]
       campbot contribs [--out=<filename>] [--starts=<start_date>] [--ends=<end_date>] [--delay=<seconds>]
       campbot export <url> [--out=<filename>] [--delay=<seconds>]
 
@@ -33,7 +33,7 @@ Here isthe help you get by typing ``campbot`` in you command line tool :
                     filename is also accepted, and must be like : 
                     123 | r
                     456 | w
-                    <langs> is comma-separated lang identifiers, like fr,de for french and german.
+                    <lang> is a lang identifiers, like fr for french.
       contribs      Export all contribution in a CSV file. <start_date> and <end_date> are like 2018-05-12
       export        Export all documents in a CSV file.
                     <url> is like https://www.camptocamp.org/outings#u=2, or, simplier, outings#u=2