我正在使用锐利的库来创建动态JPEG车牌图像。
基本上,我有一个PNG,它是一个没有数字的虚荣牌照。然后我在代码中创建一个svg,如此
const svg = new Buffer(
`<svg xmlns="http://www.w3.org/2000/svg" width="${width}" height="${height}" viewBox="${x} ${y} 500 40">
<defs>
<style type="text/css">
<![CDATA[
@font-face {
font-family: LicensePlate;
src: url('LicensePlate.ttf');
}
svg {
width: 100%;
height: 100%;
}
]]>
</style>
</defs>
<text x="0" y="0" font-family="LicensePlate" font-size="${fontsize}" letter-spacing="${letterspace}">
${platenumber.toUpperCase()}
</text>
</svg>`
);
传递所需的宽度,高度和车牌号码。然后我使用锐利的库将我的SVG叠加在牌照中间。这一切都很好。
但是,我已导入自定义牌照字体(LicensePlate.ttf)。为了调试我的代码内SVG图像,我制作了一个实际的svg图像文件,我在浏览器中打开它以确保它看起来都正确,它就是这样。
问题是,当创建最终的JPEG文件时,它不包含我的自定义字体。相反,它会落在Verdana上。
我的问题是,有什么方法可以在创建锐利的图像时保持SVG字体吗?
谢谢!
function createImage(platenumber) {
//Trying to create some sort of responsiveness
let fontsize = 80;
let letterspace = 10;
let width = 300;
let height = 90;
let x = 0;
let y = -45;
const inputlength = platenumber.length;
//Minumum Length
if (inputlength == 2) {
x = -200;
}
if (inputlength == 3) {
x = -150;
}
if (inputlength == 4) {
x = -130;
}
if (inputlength == 5) {
x = -105;
}
if (inputlength == 6) {
x = -65;
}
try {
console.log('stream is duplex, ', pipe instanceof stream.Duplex);
//Read the svg code into a buffer with a passed in plate number
const svg = new Buffer(
`<svg xmlns="http://www.w3.org/2000/svg" width="${width}" height="${height}" viewBox="${x} ${y} 500 40">
<defs>
<style type="text/css">
<![CDATA[
@font-face {
font-family: LicensePlate;
src: url('LicensePlate.ttf');
}
svg {
width: 100%;
height: 100%;
}
]]>
</style>
</defs>
<text x="0" y="0" font-family="LicensePlate" font-size="${fontsize}" letter-spacing="${letterspace}">
${platenumber.toUpperCase()}
</text>
</svg>`
);
const plateid = rand.generate(10);
//Create a write stream to a randomly generated file name
const write = new fs.createWriteStream(`plates/${plateid}.jpg`);
//Create the sharp pipeline
const pipeline = pipe
.overlayWith(svg, { gravity: sharp.gravity.center })//we center the svg image over the top of whatever image gets passed into the pipeline
.jpeg();//we convert to JPG because it is a compressed file format and will save space (we could also do webp if we really want to be slick about it)
//Create the read stream from the license plate template
const read = new fs.createReadStream('plate-2.png')
.pipe(pipeline)//pipe out sharp pipeline
.pipe(write);//add the write stream so that our sharp pipeline knows where to put the image
return plateid;
} catch (e) {
console.log(e);
return null;
}
}
<svg xmlns="http://www.w3.org/2000/svg" width="200" height="90" viewBox="0 -50 500 40">
<defs>
<style type="text/css">
<![CDATA[
@font-face {
font-family: LicensePlate;
src: url('LicensePlate.ttf');
}
]]>
</style>
</defs>
<text x="0" y="0" font-family="LicensePlate" font-size="150" letter-spacing="10">
Pl@T3#
</text>
</svg>
这是我使用的确切字体的链接
scipy.signal.welch
答案 0 :(得分:1)
我通过在操作系统中安装字体来解决此问题,在转换文件时,库只能访问操作系统字体。