Fly2lunch

Why not just fetch the commercial (google/yelp/expedia/travelocity/4square) reviews? Seems like reinventing the wheel for the users.
After looking at the Fly2Lunch site, my goal was to not recreate anything. Like you, people can find restaurants through their favorite apps but there will likely be missing pilot-related information i.e. availability of courtesy cars, where's the best place to tie down, availability of Uber, etc. so once someone has found a restaurant, they can lookup that restaurant and see if there are any pilot things they should know about. Or, they go at it the other way. Lookup where pilots are stopping, and then click on the Google Maps link to read reviews and see what the general public has to say.

In the end, I gave myself this project to learn new stuff. If people think it's not worthwhile, no biggie. I've already accomplished much of my goal.
 
Last edited:
Are you saying I would need to provide my PoA userID and password to your wiki before I could access it? If not that, then what was all that stuff early on about PoA UID/PWD checking?

If yes...why?
 
Are you saying I would need to provide my PoA userID and password to your wiki before I could access it?...why?

Along the same lines of not recreating wheels, I wanted a way to keep the riffraff off my server but also not get into the business/complexity of UID/PWD management. One way to accomplish that is to only grant people access who also have valid POA credentials. I have a large disclaimer on the login page saying I do not store your PWD and do not enter it you reuse PWDs among accounts. However, I can see how this might turn off people and/or POA itself might take issue. I'm open to other ideas such as exploring how logging in w/ Google or FB might work but my thinking was only pilots would ever want to search this database b/c it only contains pilot stuff.

I briefly looked into standing up a Wiki. The more I thought about it, the more I realized a web-app would be better b/c of the flexibility it affords searching the data in different ways.
 
@Mahneuvers -
Cool idea, I like it. If you're looking for a project to pick up some skills that also would be helpful for pilots, I think this is a good one.
I'm confident I'd be a user both uploading new places I find, and using it as inspiration for new places to fly to.

FWIW, I would recommend against using creds for PoA for a few reasons:
- The pilot community is considerably larger than POA and this idea could appeal to a bigger base
- I've found the crowd here are more skeptical than your average Joe. I think the odds of getting existing POA members to plug in their creds on another page with a promise that it won't be stored/used maliciously could be an uphill battle.
- I'm not a POA admin, but I imagine that could run afoul of some TOS. That's just speculation. Not sure if you can piggyback off the existing session for those who are already logged into POA? (I'm not a webDev :))

Separately, I have quite a bit of experience in python, mysql/sql/etc, building web scrapers and developing information retrieval apps. If you want a tech-minded individual to bounce any ideas off of, plz feel free to DM me!
 
Took the day off to work on this thing that has become a obsession. Using MySQL Workbench. Spent the day writing a pretty large proc that extrapolates points on a course if the distance between airports is > 20 miles. Had it all working and was tidying it up. Workbench crashed during update. Apparently it crashed between the drop and create proc statements. Proc is gone :mad:
 
I just looked at Fly2Lunch.com for the first time. I liked it! It has a nice search feature and tells me what airports with restaurants are nearby my home airport or along my route. Additional information is also linked, so it is a one stop shop so to speak. Thanks for posting about it.
 
Is Fly2Lunch being updated?
Yes. But you have to wade through a lot stale data. That's what motivated me for this project. In my version, you'll be able to search based on # of comments and recency of comments to filter out old information. Also, users will be able mark restaurants as closed further eliminating noise.
 
...I think you be better off just having user provide airports as possible stops, then checking all airports within a certain radius and going from there...
I have that working. But I also have working identifying restaurants/airports w/in a number of miles of a course (of up to four airports). Maybe I'll add an option to ignore airports not explicitly noted as the course. Would make the query return faster.

I hope my airport database is okay. Spent an hour today chasing down why an airport I wasn't expecting was returned in a course search. Turns out, the coordinates for the airport were incorrect.
 
Yes. But you have to wade through a lot stale data. That's what motivated me for this project. In my version, you'll be able to search based on # of comments and recency of comments to filter out old information. Also, users will be able mark restaurants as closed further eliminating noise.

So the answer is sort of. :)

Data is updated, but stale data not purged.
 
PoA runs on XenForo which probably has an OAuth hook. That’d be the more secure way of handling logins.
Thanks for the tip. Best I can tell, there is an XenForo add-on which must first be configured for the POA site to act as a OAuth provider. I'm guessing that has not been done and there's no interest in doing so for my little project. XenForo also has an REST API with an auth/ endpoint which would be much cleaner than screen scraping to test valid UID/PWD combinations, but, I'm guessing that's also locked down w/ no appetite to change. Feel free to correct me if someone knows different.
 
There’s a site feedback and support sub-forum that is monitored by the admins. That’d change it from guessing to answers, likely quickly.
 
There’s a site feedback and support sub-forum that is monitored by the admins. That’d change it from guessing to answers, likely quickly.
Agreed. Still have some work to do and would like some beta tester feedback before bothering them.
 
or u could just not worry about PoA username/pwords or any username/pwords and move on. if this thing is as awesome as it sounds, people outside of PoA will prob use it. personally I think it should just require a restaurant receipt to be able to post, but that's just me.
 
or u could just not worry about PoA username/pwords or any username/pwords and move on..
At least initially, I'm hosting this on a machine inside my home network. I'm not a network/security expert, but the sound of opening a port through the firewall and putting a welcome mat out to the world is worrisome. I'll look into how much it costs to host, again, once I think this thing might have legs.
 
Looking for a couple more beta testers. Ping me if you're interested.
 
Good luck with the initial data population. The average (lazy) user is going to do just like I did: Try a couple of airports with a known restaurant nearby, see that they're not listed, and then realize, "Oh, so I've got to go populate everything?" and pass by.
 
Someone was able to trigger an error. I blame the beta-testers for not catching it :) Seriously, let me know specifically what you did and I'll look into it.
 
Good luck with the initial data population. The average (lazy) user is going to do just like I did: Try a couple of airports with a known restaurant nearby, see that they're not listed, and then realize, "Oh, so I've got to go populate everything?" and pass by.
Yep. Crowd sourcing is only as good as the crowd that is sourcing :) As I previously pointed out, I gave myself this project to learn stuff. If it's not adopted and this is as far as it goes, so be it.
 
I can't get the website to take a google map link.
 
I can't get the website to take a google map link.
Post the GM link here. I'm thinking it's huge...bigger than the DB field that holds it.

Edit: I just expanded the column to 1000 characters.

Edit 2: the link must start w/ "https://www.google.com/maps". If it doesn't, the site will not allow you to save it. I'm seeing some errors that looks like the link isn't even that long.
 
Last edited:
I got this when adding a new restaurant with a google maps link:

Whitelabel Error Page
This application has no explicit mapping for /error, so you are seeing this as a fallback.

Tue Aug 02 10:42:02 EDT 2022
There was an unexpected error (type=Internal Server Error, status=500).

then tried without the google maps link, same error

Whitelabel Error Page
This application has no explicit mapping for /error, so you are seeing this as a fallback.

Tue Aug 02 10:42:34 EDT 2022
There was an unexpected error (type=Internal Server Error, status=500).
 
Alrighty. Hold off for a bit adding restaurants. Something must have gone wrong w/ my deploy.
 
for the record, during beta testing, I submitted a tiny url for a google maps link and it worked. but that's too much work and I don't think what's going on here.

EDIT: it would not let me do a tiny url this time, since the switch to google maps only links.
 
Last edited:
So totally my fault in cleaning up bad code. Did not mean to dis the beta volunteers. It's fixed. Whoever added Jerry's Flyaway Kitchen and Iron Thunder, please try again.
 
Iron Thunder went thru this time with google maps link.

you have 'short drive' and 'long drive'. that may mean diff things to people. Maybe make them 'short drive <3 miles', long drive >3 miles' or something like that? I dunno, just a suggestion. OR pull the actual mileage from the google link, if possible, and auto-populate the mileage, like that 'other' place does:

upload_2022-8-2_12-15-46.png
 
Good luck with the initial data population. The average (lazy) user is going to do just like I did: Try a couple of airports with a known restaurant nearby, see that they're not listed, and then realize, "Oh, so I've got to go populate everything?" and pass by.

Yeah, it's a heavy lift to get it all populated. But hey, if you know of some places that definitely have restaurants but don't want to enter them, you can DM me and I'll research+add them. You'll be helping everyone in the long run!
 
Comment count of returned restaurants when searching by course is way overstated. I'll look into it when I have the time.
 
I got the Jerrys fly away to load. I will be added more when I get the time.
 
Can we put a link in the comments to make it easy for someone to look up?
 
Whoever just added Copper House Grill. I changed your link to this as it takes you directly to the restaurant information. If you enter a Google Maps Link based on lat/lon, other pilots will see the geographic location but not all the other helpful stuff.
 
Whoever just added Copper House Grill. I changed your link to this as it takes you directly to the restaurant information. If you enter a Google Maps Link based on lat/lon, other pilots will see the geographic location but not all the other helpful stuff.

I've actually been putting in the google link showing directions from the airport to the restaurant. should I not do that?
 
Can we put a link in the comments to make it easy for someone to look up?
Sure, but why is a Google Maps Link directly to the restaurant insufficient? W/ links in comments people can see the text the URL and then decide for themselves if it's safe.
 
I've actually been putting in the google link showing directions from the airport to the restaurant. should I not do that?
Nah. I personally like that and started doing that myself once I saw what you did. The issue w/ the Copper House Grill link was that it was lat/lon based so you still had to find the restaurant. Not a big deal since it was on airport and easy to find. Would have been more problematic if off airport.
 
Back
Top