From 05e373d29b91cf929d0ac2ad74a90dce264df022 Mon Sep 17 00:00:00 2001
From: Brion Vibber <brion@pobox.com>
Date: Thu, 8 Apr 2010 17:05:02 -0700
Subject: scripts/strip_geo.php to remove geodata from notices by a given
 user/profile.

May be slow or run out of memory if run on particularly prolific posters -- not yet optimized for that case.

Note that geodata that has already been sent out to other services (via ostatus, omb, twitter, etc) will not be removed from them.

(fixed version -- previous had accidentally undone another commit)
---
 scripts/strip_geo.php | 116 ++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 116 insertions(+)
 create mode 100755 scripts/strip_geo.php

(limited to 'scripts/strip_geo.php')

diff --git a/scripts/strip_geo.php b/scripts/strip_geo.php
new file mode 100755
index 000000000..010fb31f5
--- /dev/null
+++ b/scripts/strip_geo.php
@@ -0,0 +1,116 @@
+#!/usr/bin/env php
+<?php
+/*
+ * StatusNet - a distributed open-source microblogging tool
+ * Copyright (C) 2009-2010, StatusNet, Inc.
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+define('INSTALLDIR', realpath(dirname(__FILE__) . '/..'));
+
+$shortoptions = 'i::n::y';
+$longoptions = array('id=', 'nickname=', 'yes', 'dry-run');
+
+$helptext = <<<END_OF_HELP
+strip_geo.php [options]
+Removes geolocation info from the given user's notices.
+
+  -i --id       ID of the user (may be a remote profile)
+  -n --nickname nickname of the user
+  -y --yes      do not wait for confirmation
+     --dry-run  list affected notices without deleting
+
+END_OF_HELP;
+
+require_once INSTALLDIR.'/scripts/commandline.inc';
+
+if (have_option('i', 'id')) {
+    $id = get_option_value('i', 'id');
+    $profile = Profile::staticGet('id', $id);
+    if (empty($profile)) {
+        print "Can't find local or remote profile with ID $id\n";
+        exit(1);
+    }
+} else if (have_option('n', 'nickname')) {
+    $nickname = get_option_value('n', 'nickname');
+    $user = User::staticGet('nickname', $nickname);
+    if (empty($user)) {
+        print "Can't find local user with nickname '$nickname'\n";
+        exit(1);
+    }
+    $profile = $user->getProfile();
+} else {
+    print "You must provide either an ID or a nickname.\n\n";
+    show_help();
+    exit(1);
+}
+
+if (!have_option('y', 'yes') && !have_option('--dry-run')) {
+    print "About to PERMANENTLY remove geolocation data from user '{$profile->nickname}' ({$profile->id})'s notices. Are you sure? [y/N] ";
+    $response = fgets(STDIN);
+    if (strtolower(trim($response)) != 'y') {
+        print "Aborting.\n";
+        exit(0);
+    }
+}
+
+// @fixme for a very prolific poster this could be too many.
+print "Finding notices with geolocation data...";
+$notice = new Notice();
+$notice->profile_id = $profile->id;
+$notice->whereAdd("lat != ''");
+$notice->find();
+
+if ($notice->N) {
+    print " $notice->N found.\n";
+    while ($notice->fetch()) {
+        print "notice id $notice->id ";
+        if (have_option('v') || have_option('--verbose')) {
+            print "({$notice->lat},{$notice->lon}) ";
+            if ($notice->location_ns) {
+                print "ns {$notice->location_ns} id {$notice->location_id} ";
+            }
+        }
+        if (have_option('--dry-run')) {
+            // sucka
+            echo "(skipped)";
+        } else {
+            // note: setting fields to null and calling update() doesn't save the nulled fields
+            $orig = clone($notice);
+            $update = clone($notice);
+
+            // In theory we could hit a chunk of notices at once in the UPDATE,
+            // but we're going to have to decache them individually anyway and
+            // it doesn't hurt to make sure we don't hold up replication with
+            // what might be a very slow single UPDATE.
+            $query = sprintf('UPDATE notice ' .
+                             'SET lat=NULL,lon=NULL,location_ns=NULL,location_id=NULL ' .
+                             'WHERE id=%d', $notice->id);
+            $ok = $update->query($query);
+            if ($ok) {
+                // And now we decache him manually, as query() doesn't know what we're doing...
+                $orig->blow();
+                echo "(removed)";
+            } else {
+                echo "(failed?)";
+            }
+        }
+        print "\n";
+    }
+} else {
+    print " none found.\n";
+}
+
+print "DONE.\n";
-- 
cgit v1.2.3-54-g00ecf


From b47fc9c0bcd9f37438d2aaeb7b5bc98183b554ea Mon Sep 17 00:00:00 2001
From: Brion Vibber <brion@pobox.com>
Date: Fri, 9 Apr 2010 11:36:02 -0700
Subject: Fix for strip_geo's decaching; also added --all option to run over
 all notices by given profile to help in fixing up cache inconsistencies

---
 scripts/strip_geo.php | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

(limited to 'scripts/strip_geo.php')

diff --git a/scripts/strip_geo.php b/scripts/strip_geo.php
index 010fb31f5..b3f27be61 100755
--- a/scripts/strip_geo.php
+++ b/scripts/strip_geo.php
@@ -21,7 +21,7 @@
 define('INSTALLDIR', realpath(dirname(__FILE__) . '/..'));
 
 $shortoptions = 'i::n::y';
-$longoptions = array('id=', 'nickname=', 'yes', 'dry-run');
+$longoptions = array('id=', 'nickname=', 'yes', 'dry-run', 'all');
 
 $helptext = <<<END_OF_HELP
 strip_geo.php [options]
@@ -31,6 +31,8 @@ Removes geolocation info from the given user's notices.
   -n --nickname nickname of the user
   -y --yes      do not wait for confirmation
      --dry-run  list affected notices without deleting
+     --all      run over and decache all messages, even if they don't
+                have geo data now (helps to fix cache bugs)
 
 END_OF_HELP;
 
@@ -67,10 +69,14 @@ if (!have_option('y', 'yes') && !have_option('--dry-run')) {
 }
 
 // @fixme for a very prolific poster this could be too many.
-print "Finding notices with geolocation data...";
 $notice = new Notice();
 $notice->profile_id = $profile->id;
-$notice->whereAdd("lat != ''");
+if (have_option('--all')) {
+    print "Finding all notices by $profile->nickname...";
+} else {
+    print "Finding notices by $profile->nickname with geolocation data...";
+    $notice->whereAdd("lat != ''");
+}
 $notice->find();
 
 if ($notice->N) {
@@ -101,10 +107,10 @@ if ($notice->N) {
             $ok = $update->query($query);
             if ($ok) {
                 // And now we decache him manually, as query() doesn't know what we're doing...
-                $orig->blow();
+                $orig->decache();
                 echo "(removed)";
             } else {
-                echo "(failed?)";
+                echo "(unchanged?)";
             }
         }
         print "\n";
-- 
cgit v1.2.3-54-g00ecf