Zoom constraints not applied in Quagga.init

[myleftshoe]

Zoom only works when the constraint is applied to the active track after calling Quagga.init. I.e. this code works:

        Quagga.init(
            {
                inputStream: {
                    name: 'Live',
                    type: 'LiveStream',
                    target: video,
                    constraints: {
                        width: window.innerWidth,
                        height: 480,
                        facingMode: 'environment',
                    },
                    // area: {
                    //     // defines rectangle of the detection/localization area. Works with or without locator enabled
                    //     top: "40%",
                    //     right: "10%",
                    //     left: "10%",
                    //     bottom: "40%",
                    // },
                },
                locate: true,
                locator: {
                    halfSample: true,
                    patchSize: 'x-large',
                    willReadFrequently: true,
                },
                decoder: {
                    readers: ['code_128_reader'],
                },
            },
            function (err) {
                if (err) {
                    console.warn(err)
                    return
                }
                const track = Quagga.CameraAccess.getActiveTrack()
                track.applyConstraints({
                    advanced: [{ zoom: 2 }],
                })
                Quagga.start()
                Quagga.onDetected(handleScan)
            },
        )

But this code does not:

        Quagga.init(
            {
                inputStream: {
                    name: 'Live',
                    type: 'LiveStream',
                    target: video,
                    constraints: {
                        width: window.innerWidth,
                        height: 480,
                        facingMode: 'environment',
                        advanced: [{ zoom: 2 }],
                    },
                    // area: {
                    //     // defines rectangle of the detection/localization area. Works with or without locator enabled
                    //     top: "40%",
                    //     right: "10%",
                    //     left: "10%",
                    //     bottom: "40%",
                    // },
                },
                locate: true,
                locator: {
                    halfSample: true,
                    patchSize: 'x-large',
                    willReadFrequently: true,
                },
                decoder: {
                    readers: ['code_128_reader'],
                },
            },

I also tried using zoom: 2 directly, i.e. without advanced

This is a cosmetic issue really because you can see the zoom jump from the initial to zoom:2 rather than already zoomed on start.
(Other than that I am getting really quick and accurate detection results, much appreciation!)

[github-actions]

Thank you for filing an issue! Please be patient. :-)

[ericblade]

That's great that it works well! Thanks for the tip. I'd have to do a little research on it, probably try it out with a barebones test not using quagga at all, but I can think of two possible reasons.. either a) it doesn't pass the zoom parameter into the api, or b) the api is ignoring the zoom parameter until after the camera has been initialized. a should be easily fixable (i think :) ) b would probably mean it's more of a browser problem that I wouldn't really be able to do anything about....

I think the last time I looked at the zoom api, it wasn't implemented anywhere yet.. so.. good to take another look. If it's well supported, could add some functionality to the CameraAccess to manipulate it, too

[myleftshoe]

FYI I'm using it in a standalone Safari webapp on iPhone 11. It only works when specified using the advanced array. And to confuse things more, getSettings returns an object with { zoom: 1 }, not { zoom: 2 } nor advanced: [{ zoom: 2 }]. My point is that it is zoomed even though the settings indicate that it is not. Incidentally, getCapabilities returns zoom.min: 0.5 and zoom.max: 2.

The zoom really does make a massive difference - I don't get any detections without it, and I fiddled around a lot trying to make it work :)