esaltzm / photocast Goto Github PK

PhotoCast is an app that lets users upload their photos and see weather data from the time/place they took those photos. Users can select a weather parameter (ex. wind speed) and sort their photos by that parameter, as well as view where the photos were taken on a map. Additionally, map markers can be color coded to a gradient showing the value of that parameter at each marker. This project idea was inspired by my thru hike of the Colorado Trail this summer - I have a lot of photos and there was some crazy weather during the trip, and I did not keep a journal or anything! I would love to use this app on my own photos and make it available for other people to help remember their trips (hiking or otherwise).

• GitHub Repo
• Deployed App
• Demo Recording

• Iconify
• React-Progress-Bar
• Axios
• Google-Map-React
• TZ-Lookup

• ☑ Can upload photos, display photos on the page
• ☑ Extract EXIF data from each photo, call historical weather API for each location and time
• ☑ Sort photos based on chosen attribute

• ☑ Render photos as location pins on a Google Maps component
• ☑ Upon hovering over photo, highlight location pin for that photo
• ☑ After sorting photos by weather attribute, color code location pins with a gradient for the range of values present
• ☑ On hovering over location pin, show thumbnail of image represented by the pin

• ☐ Implement drop zone for draggable file upload
• ☐ Dynamically ender center and zoom of Google Maps component based on location pins present (right now they are default values)
• ☐ Find a different historical weather API that goes back further than one year and is not limited by a free trial

I was kind of at a loss about how to implement this, but ended up changing the photoURLs state from an array of URLs to an array of objects, one key containing the url, and one key containing all the metadata and weather information. That way, each time the user sorts, the setPhotoURLs function can be called to replace the array with a new array sorted by that parameter

const sortColors = () => {
    const lightest = photoURLs[photoURLs.length - 1].data[paramKeys[param]]
    const darkest = photoURLs[0].data[paramKeys[param]]
    const diff = darkest - lightest
    const newColors = [...photoURLs.map((photo, i) => {
        const rScale = 255 - 255 * Math.abs((photo.data[paramKeys[param]] - darkest) / diff)
        const z = 1000 - i
        return {
            rgb: [rScale, 0, 0],
            z: z
        }
    })]
    setColors(newColors)
}

This function creates a list of RGB colors and Z-indices based on the sort parameter - it renders the highest or lowest location pin (depending on user input) as red and brings that pin to the foreground, with the other side of the array rendering as black. I would like to implement a more interesting gradient but I have to research color theory for non-linear changes to RGB values!