Re: [paho-dev] Status of paho-mqtt for python

[Peter Silva <peter@xxxxxxxxxxxxxxx>]

right. summarizing:

* a patch breaking the build has been the HEAD of the master branch for about 10 months.

* there is no other activity on that branch.

How could nobody have complained about this for 10 months? 

This leads me to believe that development work is going on somewhere else, and somebody is 

updating this branch from time to time, but not using it at all...

Should I just open an issue on the breakage in the master branch?  

Or is there somewhere else where branches are more active?

On Sun, Oct 1, 2023 at 7:52 AM Greg Troxel <gdt@xxxxxxxxxx> wrote:

Peter Silva via paho-dev <paho-dev@xxxxxxxxxxx> writes:

> I normally use ubuntu packages, but I wanted to make a PR for a feature I
> need.  To start work, I need to get a working
> source tree.  Is the master branch on github supposed to be usable?

Of course it should be....

> I cloned the master branch of
> https://github.com/eclipse/paho.mqtt.python and the try using it by
> installing with "pip3 install -e . " on my linux workstation.

>
> I try using it in my application,
>     self.reconnect()
>   File "/home/peter/src/paho.mqtt.python/src/paho/mqtt/client.py", line
> 1045, in reconnect
>     _on_pre_connect_ = self.on_pre_connect
>   File "/home/peter/src/paho.mqtt.python/src/paho/mqtt/client.py", line
> 1863, in on_pre_connect
>     return self._on_pre_connect
> AttributeError: 'Client' object has no attribute '_on_pre_connect'. Did you
> mean: 'on_pre_connect'?

I see this too.


With some excessive punctuation, on
  NetBSD 10 amd64
  python 3.10

I get

SKIPPED [21] .eggs/pytest-7.4.2-py3.11.egg/_pytest/unittest.py:371: paho.mqtt.testing not present.
==== 26 failed, 18 passed, 21 skipped, 8 warnings in 251.32s (0:04:11) ====

> Does this make sense?  Should I be working from some other source?

This is the right repo, according to starting with the paho homepage.



If I back up to the v1.6.1 tag:

  commit a4cb435ca2864d073ea3e0e18b0407e4bbe85b16 (origin/master, origin/HEAD, master)
  Author: Roger Light <roger@xxxxxxxxxx>
  Date:   Sat Jan 21 00:12:33 2023 +0000

      Add on_pre_connect() callback

      This is called immediately before a connection attempt is made.

  commit 9782ab81fe7ee3a05e74c7f3e1d03d5611ea4be4 (HEAD, tag: v1.6.1)
  Merge: 6dfdeba 1f3cd09
  Author: Roger A. Light <roger@xxxxxxxxxx>
  Date:   Thu Oct 21 11:05:05 2021 +0100

      Merge branch '1.6.x'

then tests show (again punctuation remediated):

SKIPPED [21] .eggs/pytest-7.4.2-py3.11.egg/_pytest/unittest.py:371: paho.mqtt.testing not present.
==== 44 passed, 21 skipped in 11.25s ====


I think the addition of the preconnect callback has a few issues.
With this, it passes:


diff --git a/src/paho/mqtt/client.py b/src/paho/mqtt/client.py
index 54405a4..5515af7 100644
--- a/src/paho/mqtt/client.py
+++ b/src/paho/mqtt/client.py
@@ -615,6 +615,7 @@ class Client(object):
         self._registered_write = False
         # No default callbacks
         self._on_log = None
+        self._on_pre_connect = None
         self._on_connect = None
         self._on_connect_fail = None
         self._on_subscribe = None
@@ -1881,7 +1882,7 @@ class Client(object):

     def pre_connect_callback(self):
         def decorator(func):
-            self._on_pre_connect_ = func
+            self._on_pre_connect = func
             return func
         return decorator