Hi folks. Is possible that APRS Stations List show only the stations in the defined radius range?
TIA
|
Re: Modernizing the HTML control panel
The old system will be kept whatever happens and I expect to keep it
as the default, though I'm very happy for people to have something
different if they prefer it.
My current idea is to have an option to install the new version if
you want it. I'm also old, started programming with Fortran but in
1969 and earned my living writing cobol for a few years before
moving on to writing networking software, and find the current
system works fine for me. But I don't need to access from Android.
73, John
On 03/10/2024 15:55, Chris Curtis
wrote:
toggle quoted message
Show quoted text
my apologies.
?
i had a feeling, i couldn't express myself in a way that
wouldn't stir the pot.
?
in reply, i see the same thing.
?
i do not see shortcomings where others do.
?
i do not see 'better' in the different where others do.
screen readers, one example, would break whenever something
'better' was introduced.
?
i am old, i understand that.
my masters is from the era of cobol and fortran 77.
?
my hobby was turbo pascal, writing doors for bbs systems.
?
i have no relevance in the actual writing of today's programs
but as a user, i wanted to express my desire to keep a 'look and
feel' that works for me.
?
again, my apologies, i'm QRT
:)
kb0wlf
|
Linbpq, Winlink Express, and Kenwood D710 won't connect
I am testing a Linbpq with a Direwolf and soundcard. Connecting to the node is okay with my Kenwood D710 and Ipserial. But when I try Winlink Express, the connection starts and then fails. It keeps retrying and sometimes ends up with node info or an unexpected disconnect.
?
The RMS works fine when I connect from Winlink on a PC and Direwolf on a Pi 4 with a Kiss connection. I was having a problem with an incoming message so I changed MAXFRAME to 2 and have been creeping TXDELAY higher.?
?
Not sure if this is the right place but the Linbpq Jedi Knights are here.
?
Here is my Port definition for Linbpq
?
PORT
?PORTNUM=2
?ID=Direwolf 1200 Baud 145.53 MHz
?TYPE=ASYNC
?PROTOCOL=KISS
?IPADDR=127.0.0.1
?TCPPORT=8001
?CHANNEL=A
?PACLEN=128
?TXDELAY=575 ;was 550
?FRACK=7000
?RESPTIME=1500
?MAXFRAME=2 ;was 4
?RETRIES=6
?FULLDUP=0
?SLOTTIME=100
?TXTAIL=30
?PERSIST=63
?WL2KREPORT PUBLIC, api.winlink.org, 80, K9TSM-10, EN71CP, 22-23, 145530000, PKT1200, 60, 100, 3, 0
ENDPORT
?
?
Dave Menges
|
Re: Modernizing the HTML control panel
No need to apologise, I didn't see your comments as "stirring the pot", I just wasn't sure what you meant.
?
One of the improvements I have made is to add labels to all the form inputs, which will be correctly read out by screen readers. I've very much had accessibility in mind - we're all getting older with eyes getting worse, so at very least need to bump the text size up a bit! 16px is the minimum suggested size in modern guidelines such as .
?
As I mentioned above, I'm mostly concerned with making the layout work across different devices, rather than making any sweeping changes to how things work.
|
Re: Modernizing the HTML control panel
my apologies.
?
i had a feeling, i couldn't express myself in a way that wouldn't stir the pot.
?
in reply, i see the same thing.
?
i do not see shortcomings where others do.
?
i do not see 'better' in the different where others do.
screen readers, one example, would break whenever something 'better' was introduced.
?
i am old, i understand that.
my masters is from the era of cobol and fortran 77.
?
my hobby was turbo pascal, writing doors for bbs systems.
?
i have no relevance in the actual writing of today's programs but as a user, i wanted to express my desire to keep a 'look and feel' that works for me.
?
again, my apologies, i'm QRT
:)
kb0wlf
|
Re: Modernizing the HTML control panel
Hi Chris,
?
I'm not sure what you mean. This is what I see on Android in Firefox, and Chrome:
?
?
The redesign scales the content to make it much more readable and usable:
|
Re: Modernizing the HTML control panel
just so i don't grumble at myself for not 'saying something' when i had the chance;
?
:)
?
if anything does change, please make the current 'look and feel' the standard, rather than an option to 'roll back to'.
(the proposed formatting changes would obviously be 'good' but 'good' ought to be tempered by necessity)
?
when i visit my node's 'page', either desktop browser or android browser, i don't think 'hey, this looks so old it must not work anymore.? i wish it looked new and shiny like youtube.com/@soloscout '
?
;)
?
good luck on the idea
appreciate the BPQ32 !
?
kb0wlf
?
|
Re: Modernizing the HTML control panel
This looks really good, Rik.
+1 on the option to use HTTP or HTTPS, if possible.
73 MT
--
Thanks. 73 Michael GM5AUG
SYSOP GB7AUG, MB7NKL
|
Different IDTEXT per port
Good afternoon everyone,
?
(I've asked this in the OARC Discord that a number of you will be part of - so apologies for any deja vu.)
?
I was wondering whether it is possible to assign a different IDTEXT value to each port in LinBPQ? At the moment I configure IDTEXT globally and use it as an APRS location ident. This is then iGated by another VHF station ~20 miles away. However I don't want to clog up the HF bands with my APRS location information as it is quite a busy back-channel for mail forwarding. The obvious answer would, I know, be to use the ports' Beacon functions, and configure a different beacon file in each, but I'm already utilising those for some live data using the beacon 'from file' option.
?
* Is there a way in which I could set a different IDTEXT on a port-by-port basis?
?
* If not, is there a way I can suppress the IDTEXT on certain ports?
?
Many thanks.
73 Michael
?
--
Thanks. 73 Michael GM5AUG
SYSOP GB7AUG, MB7NKL
|
Re: Modernizing the HTML control panel
I've used the mail forwarding page as an example, mainly because it is one of the most complex pages in terms of layout. As this is just a prototype, I haven't spent much time on styling the page, but I have:
?
- Used a grid system for layout instead of tables.
- Tweaked the grid settings so that components take up varying amounts of space depending on the screen size.
- Added proper labels to form inputs (important for screen readers).
- Used proper input types (e.g. number instead of text, where appropriate).
- Removed a lot of superfluous tags such as <font> (styling is all handled in CSS).
- Made the main menu collapsible on screens < desktop.
?
Again, this is just a demo, so it doesn't actually function, and there is no JS functionality either at the moment (aside from the main menu).
?
Extract the .zip, open forwarding.html and resize your browser window from very narrow (like a mobile) up to full screen to see the content reflow.
|
Re: Modernizing the HTML control panel
Re: https://
?
While https:// is necessary over the Internet these days, unencrypted http:// is still necessary for pages delivered over Amateur Radio, e.g, AREDN, HamNet, HamWan. So, in the quest for updating the web presence, please remember to give the sysop a choice of how his pages are to be delivered.
?
73, Mark, N2MH
AREDN Ambassador
BPQ Sysop
?
|
Re: Modernizing the HTML control panel
Oh, that's excellent, I had no idea that already existed. With that in mind, expanding it to cover configuration options too is definitely the way to go. To some extent, REST APIs are self documenting. The root endpoint (/api) should return an object with a _links property, letting the user know of the potential paths they could take.
?
I'll put together some static example pages. Even if they don't actually function to change settings, they will give you an idea of how the layout could work at different screen sizes.
|
|
|
Re: Modernizing the HTML control panel
Rik,
There is a basic apt that returns some node info, but nothing for
the BBS
/api/ports
/api/nodes
/api/users
/api/info
/api/links
/api/mheardport
This could be extended, but would need a fair bit of work for me.
The simplest for me is to change the code to look for an external
file before using the internal copy. The downside of this is that
the template is used directly by the C code and introduces the
problem of compatibility if I change my pages and the risk of
instability if there are any problems with the template.
Perhaps it would be interesting if you built a few pages using
the existing api functions and see how they look. I'll need to
document the api calls but I should do that anyway.
73,
John
On 02/10/2024 11:13, Rik M7GMT wrote:
toggle quoted message
Show quoted text
Hi John,
?
Thanks for the explanation. While I know HTML, CSS, and JS
intimately, C is above my pay grade! That does put a small
spanner in the works as I was hoping to be able to do this
without too much work for you, by simply editing the templates.
Spitballing some potential approaches:
?
- Modify the existing template strings in the C source
code. This seems like it would be very awkward -
having the HTML and C code mixed together seems like a recipe
for a lot of headaches.
- Go back to having the HTML in separate files.
The C code would load the template from disk, all the data
could simply be printed into an object in a script tag, then
assembled into the page using Handlebars (on the client). This
would avoid having to do the string replacement in C, which I
imagine is quite fragile. You can see a basic example of how
this would work .
Handlebars also has a concept called "Partials", so things
like the main menu can be written once, then reused on
multiple pages. You would also get proper syntax highlighting
and formatting in your editor, rather than just bunched up
strings.
- Leave everything as is, and build a new REST API.
Provide endpoints such as /api/main-config
and /api/node-list
which return a JSON object containing the data, which can then
be used by any kind of custom front end. This would be very
versatile, especially for things like the node list and link
status, as it could also be used in scripting, not just in a
web front end. I would probably build a separate program using
NodeJS that provides a custom front end, but others could do
something similar using PHP, Python, or any other language.
I'm sure other people would appreciate being able to get at
this data for use in monitoring scripts, Discord bots, or
other innovative things like that.
- Use
to create a hybrid approach. This would involve
creating new endpoints that return small chunks of HTML,
rather than whole pages (e.g. /api/main-config would return
just the form). This means the chunks could remain in the C
code (although it would be easier to manage them in dedicated
files) and use most of your existing logic.
- "Gaffer tape" scraper. Add a single JS file
that essentially scraped the data out of the existing compiled
HTML (as it's delivered to the client), and then rebuilt the
page using this data. This would mean no extra work on your
part at all, but would be a bit more fragile if you did change
anything, and isn't a very efficient way of doing things.
I think my personal preference of order would be 3, 2, 4, 5,
1. As I mentioned, a REST API would be very versatile for all
sorts of uses, not just the web front end. Each approach has
it's pros and cons, and would involve varying amounts of work
for you, so I would be interested to hear your thoughts.
|
Re: Modernizing the HTML control panel
Hi John,
?
Thanks for the explanation. While I know HTML, CSS, and JS intimately, C is above my pay grade! That does put a small spanner in the works as I was hoping to be able to do this without too much work for you, by simply editing the templates. Spitballing some potential approaches:
?
- Modify the existing template strings in the C source code. This seems like it would be very awkward - having the HTML and C code mixed together seems like a recipe for a lot of headaches.
- Go back to having the HTML in separate files. The C code would load the template from disk, all the data could simply be printed into an object in a script tag, then assembled into the page using Handlebars (on the client). This would avoid having to do the string replacement in C, which I imagine is quite fragile. You can see a basic example of how this would work . Handlebars also has a concept called "Partials", so things like the main menu can be written once, then reused on multiple pages. You would also get proper syntax highlighting and formatting in your editor, rather than just bunched up strings.
- Leave everything as is, and build a new REST API. Provide endpoints such as /api/main-config and /api/node-list which return a JSON object containing the data, which can then be used by any kind of custom front end. This would be very versatile, especially for things like the node list and link status, as it could also be used in scripting, not just in a web front end. I would probably build a separate program using NodeJS that provides a custom front end, but others could do something similar using PHP, Python, or any other language. I'm sure other people would appreciate being able to get at this data for use in monitoring scripts, Discord bots, or other innovative things like that.
- Use to create a hybrid approach. This would involve creating new endpoints that return small chunks of HTML, rather than whole pages (e.g. /api/main-config would return just the form). This means the chunks could remain in the C code (although it would be easier to manage them in dedicated files) and use most of your existing logic.
- "Gaffer tape" scraper. Add a single JS file that essentially scraped the data out of the existing compiled HTML (as it's delivered to the client), and then rebuilt the page using this data. This would mean no extra work on your part at all, but would be a bit more fragile if you did change anything, and isn't a very efficient way of doing things.
I think my personal preference of order would be 3, 2, 4, 5, 1. As I mentioned, a REST API would be very versatile for all sorts of uses, not just the web front end. Each approach has it's pros and cons, and would involve varying amounts of work for you, so I would be interested to hear your thoughts.
|
Re: Adding SSID to APPLICATION TALK?
No, you can use any ssid not used for other purposes.
73,
John
On 02/10/2024 01:01, Brian Pasternak
wrote:
toggle quoted message
Show quoted text
Hello everyone. Are there any issues with adding a
SSID to the TALK Node Application? Such as KA3VSP-2?
Best,
Brian, KA3VSP
|
Re: BPQ32 Connect Script Problems
I just recently ran into a situation where I needed to do the same thing, i.e. change frequencies via ELSE. I tend to use 30 second pauses whenever something fails and has to move on with an ELSE, so things worked for me right off the bat. I think I see why at least a 20 second pause is needed. After a failed attempt to connect using the first frequency, a detach evidently happens automatically, and there's a delay before scanning resumes. I expect it's the same delay seen when a detach is done manually. It looks like the script needs to pause in order to allow scanning to resume. What's after the ELSE essentially starts over, so things need to be back to square one, i.e. back to scanning, before attaching again. That's what it looks like to me, anyway.
If you have a yaesu 857d using a digrig mobile sound card this is what work for me to get it to change frequencies?
TIMES 0000-2359
ATT 5
RADIO 7.0945 USB?
C (callsign)
ELSE
PAUSE 20
ATT 5
RADIO 7.100 USB?
C (callsign)
73, Lee K5DAT
|
Re: Changing the VARA mode
On Tue, Oct 1, 2024 at 10:05?AM John G8BPQ via <john.wiseman= [email protected]> wrote:
Yes, you can add BW commands to the connect script, after attach
and before connect. Format is BW500, BW2300 or BW2750 and for
VARAFM FM1200 or FM9600.
If you include a BW command in the port config the TNC will start
in that mode and return to it after a session.
73, John
Thank-you Sir!? Also I see one can override the port default with the VN, VW, etc. parameters in the scan lines. That's great as I have some frequencies where I don't want to accept VARA wide and others that I do.
73, Lee
_._,_._,_
|
Adding SSID to APPLICATION TALK?
Hello everyone. Are there any issues with adding a SSID to the TALK Node Application? Such as KA3VSP-2?
Best, Brian, KA3VSP
|
Re: VARA port default setting
Thank-you for the details John.
Lee K5DAT?
|
Mauro Veiga
Michael_GM5AUG
Mark Herson, N2MH
Brian Pasternak