erdavis1 / RoadColors

Plot US roads, coloring each by its designation (e.g. Street, Road, Avenue)
160 stars 37 forks source link

Easy to use web application? #2

Open Pjaerr opened 4 years ago

Pjaerr commented 4 years ago

I haven't looked into specifically how this is working but I absolutely love it.

I am wondering if it is possible to create an easy to use web application for those that aren't developers to select a city and generate the image for them? I'm not sure how much of this is purely manual and can't be automated though.

I am happy to tackle this in the future (maybe a port to JavaScript or use of web assembly) but just opening to share/discuss ideas.

erdavis1 commented 4 years ago

Thank you! Making a web app would be really cool if this feels like something you want to work on (my skills are limited to R/SQL, unfortunately). Right now, I'm manually choosing street types and their colors, but that could be automated to pick road types by length or popularity and assign colors from there. The legend is also manually created, but I'm sure that doesn't need to be the case.

On Thu, Oct 3, 2019 at 3:13 AM Joshua Jackson notifications@github.com wrote:

I haven't looked into specifically how this is working but I absolutely love it.

I am wondering if it is possible to create an easy to use web application for those that aren't developers to select a city and generate the image for them? I'm not sure how much of this is purely manual and can't be automated though.

I am happy to tackle this in the future (maybe a port to JavaScript or use of web assembly) but just opening to share/discuss ideas.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/erdavis1/RoadColors/issues/2?email_source=notifications&email_token=AK43WPY63DITH5OFG3W3MADQMXAVLA5CNFSM4I5BC4DKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HPLQ4UQ, or mute the thread https://github.com/notifications/unsubscribe-auth/AK43WP7BDE3CHDOYWT7JP4LQMXAVLANCNFSM4I5BC4DA .

Pjaerr commented 4 years ago

@erdavis1 I would love to take this on, I need to finish my current data visualisation project/article and then I think I'll come back to this and try and build a web app so everyone can make use of this awesome project!

Thanks for your response and Ill update this when I do come back to it!

Pjaerr commented 4 years ago

Just experimenting with this. Instead of me porting the code to JavaScript, I could instead just keep your R code as is and use something like Plumber (https://www.rplumber.io/) to expose the functionality as API endpoints for a web application.

Will look more into this. (Let me know if this is spamming you)

erdavis1 commented 4 years ago

Hey, sorry for the delayed response. I had no idea something like Plumber existed--that looks really cool! I'd probably want to clean up the code a bit knowing some things I do now, but feel free to keep me informed. Three emails in three months is hardly spam, haha

On Sun, Jan 5, 2020 at 2:59 AM Joshua Jackson notifications@github.com wrote:

Just experimenting with this. Instead of me porting the code to JavaScript, I could instead just keep your R code as is and use something like Plumber (https://www.rplumber.io/) to expose the functionality as API endpoints for a web application.

Will look more into this. (Let me know if this is spamming you)

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/erdavis1/RoadColors/issues/2?email_source=notifications&email_token=AK43WPZAJHADYQQTSYUJOTDQ4G4Q7A5CNFSM4I5BC4DKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEIDT5QY#issuecomment-570900163, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK43WPYEUJLPUAAAZEQSJZ3Q4G4Q7ANCNFSM4I5BC4DA .

Pjaerr commented 4 years ago

So I got your code sort of working locally, having never used R or R Studio I had no idea what I was doing but it seemed the code ran really slowly and took over 5 minutes on the plotting and saving of an image, is this expected?

If that was just an issue with my machine or the way I set up R then I think using Plumber and exposing to the a web frontend would be extremely viable and is definitely something I'd want to do (subject to you cleaning up the code if that's what you feel like doing) to make this awesome project easily accessible.

Let me know what you think!

Thanks, Josh

erdavis1 commented 4 years ago

Again, sorry for the long delay in response! Unfortunately, R/RStudio does take a pretty long time to render complicated networks.

I'm pretty sure it can be done faster/more elegantly with OSMnx https://geoffboeing.com/2016/11/osmnx-python-street-networks/, but I've never worked with it (or anything Python, for that matter). I'm sure you probably saw this https://anvaka.github.io/city-roads/around the web--the creator has the code available on Github, which might be useful.

On Wed, Jan 8, 2020 at 8:13 AM Joshua Jackson notifications@github.com wrote:

So I got your code sort of working locally, having never used R or R Studio I had no idea what I was doing but it seemed the code ran really slowly and took over 5 minutes on the plotting and saving of an image, is this expected?

If that was just an issue with my machine or the way I set up R then I think using Plumber and exposing to the a web frontend would be extremely viable and is definitely something I'd want to do (subject to you cleaning up the code if that's what you feel like doing) to make this awesome project easily accessible.

Let me know what you think!

Thanks, Josh

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/erdavis1/RoadColors/issues/2?email_source=notifications&email_token=AK43WP5PURSAF6CRQDF3763Q4X3S5A5CNFSM4I5BC4DKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEINC75I#issuecomment-572141557, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK43WPYYKVC3AJJSJFHK6ETQ4X3S5ANCNFSM4I5BC4DA .

Pjaerr commented 4 years ago

@erdavis1 That's not a problem!

Okay so I think re-writing this code in something like Python would be the best option. I don't really know python too well but I think the issue would be more of understanding and porting over your R code as I absolutely don't know R.

A lot of my projects recently have been "comfortable" for me so I think it's definitely time for a challenge and I am happy to tackle rewriting this in tech that can be ran on the web!

I don't want to take up any of your time unnecessarily so if I have any questions should I just ask them here?

Thanks, Josh