[Bug 4620] New: saying no to record adds hunk anyway

[mercurial-bugs at selenic.com]

http://bz.selenic.com/show_bug.cgi?id=4620

          Priority: normal
            Bug ID: 4620
                CC: mercurial-devel at selenic.com
          Assignee: bugzilla at selenic.com
           Summary: saying no to record adds hunk anyway
          Severity: bug
    Classification: Unclassified
                OS: All
          Reporter: gregory.szorc at gmail.com
          Hardware: All
            Status: UNCONFIRMED
           Version: 3.4-rc
         Component: record
           Product: Mercurial

I moved a file, made a change, then did a `hg record`. I said to examine the
moved file, but said no to the single hunk that changed. When the commit was
made, the hunk was added anyway.

$ hg st
M ansible/deploy-mozreview-prod.yml
A ansible/library/apache_worker_drain.py
R ansible/library/apache_worker_wait.py

$ hg record
diff --git a/ansible/deploy-mozreview-prod.yml
b/ansible/deploy-mozreview-prod.yml
  1 rename
1 hunks, 1 lines changed
examine changes to 'ansible/deploy-mozreview-prod.yml'? [Ynesfdaq?] n

diff --git a/ansible/library/apache_worker_wait.py
b/ansible/library/apache_worker_drain.py
rename from ansible/library/apache_worker_wait.py
rename to ansible/library/apache_worker_drain.py
1 hunks, 1 lines changed
examine changes to 'ansible/library/apache_worker_wait.py' and
'ansible/library/apache_worker_drain.py'? [Ynesfdaq?] y

@@ -3,17 +3,17 @@
 # GNU General Public License version 2 or any later version.

 import requests
 import time


 DOCUMENTATION = '''
 ---
-module: apache_worker_wait
+module: apache_worker_drain
 short_description: Waits for Apache workers to drain
 description:
      - When doing a rolling update, you often want to wait for active requests
        and connections to drain. Ansible's built-in wait_for module can look
        for TCP connections. However, it counts TIME_WAIT as an active
        connection, which adds significant wait times during rolling updates.
        Waiting for Apache workers to flush is accurate and faster.
     - This module polls Apache's server-status and waits for the worker count
record change 2/2 to 'ansible/library/apache_worker_drain.py'? [Ynesfdaq?] n

gps at gps-mbp:~/src/hgcustom/version-control-tools (@)$ hg record
diff --git a/ansible/deploy-mozreview-prod.yml
b/ansible/deploy-mozreview-prod.yml
1 hunks, 1 lines changed
examine changes to 'ansible/deploy-mozreview-prod.yml'? [Ynesfdaq?] n

no changes to record

$ hg export
# HG changeset patch
# User Gregory Szorc <gps at mozilla.com>
# Date 1429728532 25200
#      Wed Apr 22 11:48:52 2015 -0700
# Node ID 0023fa583432585a458cfeb1e380240b5db73288
# Parent  a755ba4cd23f94112978a297e30492adb78c70f6
rename

diff --git ansible/library/apache_worker_wait.py
ansible/library/apache_worker_drain.py
rename from ansible/library/apache_worker_wait.py
rename to ansible/library/apache_worker_drain.py
--- ansible/library/apache_worker_wait.py
+++ ansible/library/apache_worker_drain.py
@@ -3,17 +3,17 @@
 # GNU General Public License version 2 or any later version.

 import requests
 import time


 DOCUMENTATION = '''
 ---
-module: apache_worker_wait
+module: apache_worker_drain
 short_description: Waits for Apache workers to drain
 description:
      - When doing a rolling update, you often want to wait for active requests
        and connections to drain. Ansible's built-in wait_for module can look
        for TCP connections. However, it counts TIME_WAIT as an active
        connection, which adds significant wait times during rolling updates.
        Waiting for Apache workers to flush is accurate and faster.
     - This module polls Apache's server-status and waits for the worker count

-- 
You are receiving this mail because:
You are on the CC list for the bug.