Re: [PATCH v2 6/6] zoneconf.cgi: Improve VLAN & STP inputs

[Leo Hofmann <hofmann@leo-andres.de>]

[-- Attachment #1: Type: text/plain, Size: 5632 bytes --]

Hi Jonatan,

thank you for looking into this, I'll answer below!

Am 21.02.2021 um 20:06 schrieb Jonatan Schlag:
> Hi,
>
> first thanks for the patchset. Overall it looks quite good
>
> Am Sonntag, den 21.02.2021, 11:38 +0100 schrieb Leo Hofmann:
>> Hello Michael,
>>
>> thank you for looking into these patches. I'll answer below!
>>
>> Am 19.02.2021 um 20:22 schrieb Michael Tremer:
>>> Hello,
>>>
>>>> On 18 Feb 2021, at 14:30, Leo-Andres Hofmann <
>>>> hofmann(a)leo-andres.de> wrote:
>>>>
>>>> Add default values and mark fields as required.
>>>>
>>>> Signed-off-by: Leo-Andres Hofmann <hofmann(a)leo-andres.de>
>>>> ---
>>>> html/cgi-bin/zoneconf.cgi | 10 ++++++++--
>>>> 1 file changed, 8 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/html/cgi-bin/zoneconf.cgi b/html/cgi-
>>>> bin/zoneconf.cgi
>>>> index 9d01d06ce..bbd042ffc 100644
>>>> --- a/html/cgi-bin/zoneconf.cgi
>>>> +++ b/html/cgi-bin/zoneconf.cgi
>>>> @@ -478,6 +478,9 @@ END
>>>> 		if ($access_selected{"NONE"} eq "") {
>>>> 			$highlight = $_;
>>>> 		}
>>>> +		
>>>> +		# default VLAN tag if not configured
>>>> +		$zone_vlan_id = 1 unless
>>>> looks_like_number($zone_vlan_id);
>>> I am not sure if it is a good idea to add a default here.
>>>
>>> Isn’t there a danger that people will hit save prematurely and
>>> connect the wrong VLAN with another one?
> This was also my first thought.
>>> Usability issues like that cannot be prevented entirely, but I was
>>> wondering if this didn’t make it easier to run into that error.
> Thats definitly the case.
>> Agreed, this could happen if you are careless!
>> However, I hope you only enable VLAN if you are sure which ID you are
>> supposed to use. In that case, having a default value appear makes it
>> obvious where you need to put your desired ID.
>> But now I'm not sure about this one. I can revert this, of course.
> I would vote for without default value. There is just absolutey no save
> default value here and people could assume that this is something like,
> "I am clicking save and I am done". And thats not the case people have
> to think about vlan ids. I unterstand that field might be hard to spot,
> but that should not be solved by a default value.
>
> Another question
> which came to my mind while reviewing this: Is there any good error
> when the vlan id is wrong? Shure it is checked in the html, but i have
> not found any backend check ... (Not a problem with your patch, but i
> want to mention it.)
I just tried this on my test system. There is a range check but no error message.
Instead, it reverts the entire assignment, removes the NIC from the zone and saves the configuration like this. Well, now what...

I don't think this is very critical right now because the HTML range check will usually prevent this. But I think we should have a proper check in the backend at some point.
I'll put it on my list.

>
> Apart from this linter is happy too.
>
> Nice work.
>
> Jonat
> an
>
> PS.: Apart from this reading the file took 10 mins, till I found out
> that $uc has something todo with zones ... . At least a good learning
> how to not name variables :-)

$uc is the upper case zone name (RED, ..). This is already used everywhere, so I decided to keep it. I can put this on my list too.

Best regards
Leo

>
>
>>>> 		print <<END
>>>> 		<td class="$highlight">
>>>> @@ -486,7 +489,7 @@ END
>>>> 				<option value="NATIVE"
>>>> $access_selected{"NATIVE"}>$Lang::tr{"zoneconf access
>>>> native"}</option>
>>>> 				<option value="VLAN"
>>>> $access_selected{"VLAN"} $vlan_disabled>$Lang::tr{"zoneconf
>>>> access vlan"}</option>
>>>> 			</select>
>>>> -			<input type="number" class="vlanid" id="TAG-
>>>> $uc-$mac" name="TAG $uc $mac" min="1" max="4095"
>>>> value="$zone_vlan_id" $field_disabled>
>>>> +			<input type="number" class="vlanid" id="TAG-
>>>> $uc-$mac" name="TAG $uc $mac" min="1" max="4095"
>>>> value="$zone_vlan_id" required $field_disabled>
>>>> 		</td>
>>>> END
>>>> ;
>>>> @@ -513,6 +516,9 @@ foreach (@zones) { # load settings and
>>>> prepare form elements for each zone
>>>> 	my $stp_available = $ethsettings{"${uc}_MODE"} eq "bridge"; #
>>>> STP is only available in bridge mode
>>>> 	my $stp_enabled = $ethsettings{"${uc}_STP"} eq "on";
>>>> 	my $stp_priority = $ethsettings{"${uc}_STP_PRIORITY"};
>>>> +	
>>>> +	# set priority to default value if no numerical value is
>>>> configured
>>>> +	$stp_priority = 32768 unless looks_like_number($stp_priority);
>>> This is very good.
>> Thanks :)
>>> Since this is in this patchset and comes with the dependency to the
>>> other code above, I cannot pull this in with the STP patchset where
>>> it actually belongs.
>>>
>>> But I guess we should merge this all together anyways :)
>> I still have these commits in my local git. I could just submit these
>> last two changes if that makes your job easier (and you don't mind me
>> submitting everything multiple times)!
>>> -Michael
>>>
>>>> 	# form element modifiers
>>>> 	my $checked = "";
>>>> @@ -532,7 +538,7 @@ END
>>>> 	# priority input box HTML
>>>> 	my $row_2 = <<END
>>>> 		<td>
>>>> -			<input type="number" class="stp-priority"
>>>> id="STP-PRIORITY-$uc" name="STP-PRIORITY-$uc" min="1" max="65535"
>>>> value="$stp_priority" $disabled>
>>>> +			<input type="number" class="stp-priority"
>>>> id="STP-PRIORITY-$uc" name="STP-PRIORITY-$uc" min="1" max="65535"
>>>> value="$stp_priority" required $disabled>
>>>> 		</td>
>>>> END
>>>> ;
>>>> -- 
>>>> 2.27.0.windows.1
>>>>
>> Regards, Leo