实习日志7

1.试试pdf发票识别

1.1.添加文件类型判断

//判断文件类型
if (getFileType(imgCodeCell.getValue()) === "jpg"||getFileType(imgCodeCell.getValue()) === "png"
    ||getFileType(imgCodeCell.getValue()) === "jpeg"||getFileType(imgCodeCell.getValue()) === "bmp") {
    url = "customapi/fapiaoapi/vatinvoicebyimg";
    var data = {
        //传入请求地址
        token: access_token, imageUrl: imageBase64Cell.getValue()
    };
} else if (getFileType(imgCodeCell.getValue()) === "pdf") {
    url = "customapi/fapiaoapi/vatinvoicebypdf";
    var data = {
        //传入请求地址
        token: access_token, url: imageBase64Cell.getValue()
    };
} else {
    url = "customapi/fapiaoapi/vatinvoicebyurl"
    var data = {
        //传入请求地址
        token: access_token, url: imageUrlCell.getValue()
    };
}

//直接发送百度AI识别请求
InvoiceIdentificationPost(url,data);

1.2.pdf文件上传报错

需要xhr来处理pdf上传

if (getFileType(imgArray[index]) === "pdf") {
            console.log("识别到pdf文件");
            image = new Image();
            var xhr = new XMLHttpRequest();
            xhr.open("GET", img, true);
            xhr.responseType = "blob"; // 设置响应类型为二进制数据

            xhr.onload = function () {
                console.log("PDF加载完成");
                // var blob = xhr.response;
                var base64 = getBase64Image(image);
                var name = getImgName(imgArray[index]);
                var code = imgArray[index];
                Forguncy.getTableData("image", {"code": code}, function (data) {
                    alert("重复上传文件:" + name);
                }, function (errorMessage) {
                    console.log("添加文件成功" + name);
                    Forguncy.modifyTablesData({
                        image: {
                            addRows: [{
                                name: name, code: imgArray[index], base64: base64, url: imageUrl, is_identify: "未识别"
                            }],
                        }
                    });
                })
                // image.src = URL.createObjectURL(blob);
            }
            xhr.send();
        }

1.3.base64编码错误

更改pdf的base64编码

改的有点捞，有优化空间，感觉发送的请求太多了，可能会出问题

//imgs=ffc31308-ec72-4268-a977-16f4c366a75f_whitepig.png|3e8582d5-7e4b-4544-8e51-446ba8f70905_blackpig.png
//img[1]=ffc31308-ec72-4268-a977-16f4c366a75f_whitepig.png
//img[2]=3e8582d5-7e4b-4544-8e51-446ba8f70905_blackpig.png
//....
const imgs = Forguncy.Page.getCell("img").getValue();
const imgArray = imgs.split("|");

var page = Forguncy.Page;

//遍历imgArray
for (let i = 0; i < imgArray.length; i++) {
    (function (index) {
        var img = `http://${window.location.host}/Forguncy/FileDownloadUpload/Download?file=` + imgArray[index];
        var imageUrl = `http://${window.location.host}/Forguncy/Upload/` + imgArray[index];
        console.log("第" + index + "轮次的文件:" + img);
        //加载image信息
        // 判断文件类型

        if (getFileType(imgArray[index]) === "jpg" || getFileType(imgArray[index]) === "png" || getFileType(imgArray[index]) === "jpeg" || getFileType(imgArray[index]) === "bmp") {
            console.log("识别到图像文件");
            var image = new Image();
            image.src = img;
            // 加载图像、PDF信息
            image.onload = function () {
                console.log("PDF加载完成");
                var base64 = getBase64Image(image);
                var name = getImgName(imgArray[index]);
                var code = imgArray[index];
                console.log("onload3");
                Forguncy.getTableData("image", {"code": code}, function (data) {
                    alert("重复上传文件:" + name);
                }, function (errorMessage) {
                    console.log("添加图片成功" + name);
                    Forguncy.modifyTablesData({
                        image: {
                            addRows: [{
                                name: name, code: imgArray[index], base64: base64, url: imageUrl, is_identify: "未识别"
                            }],
                        }
                    });
                });
            }
        } else if (getFileType(imgArray[index]) === "pdf") {
            console.log("识别到pdf文件");
            var xhr = new XMLHttpRequest();
            xhr.open("GET", img, true);
            xhr.responseType = "blob"; // 设置响应类型为二进制数据

            xhr.onload = function () {
                console.log("PDF加载完成");
                console.log(imageUrl);
                fetch(imageUrl)
                    .then(response => response.blob())
                    .then(blob => {
                        return new Promise((resolve, reject) => {
                            const reader = new FileReader();
                            reader.onloadend = () => resolve(reader.result.split(',')[1]);
                            reader.onerror = reject;
                            reader.readAsDataURL(blob);
                        });
                    })
                    .then(base64 => {
                        console.log(base64);
                        base64 = "data:application/pdf;base64," + base64;
                        //getPdfBase64返回base64
                        console.log("base64:" + base64);
                        var name = getImgName(imgArray[index]);
                        var code = imgArray[index];
                        Forguncy.getTableData("image", {"code": code}, function (data) {
                            alert("重复上传文件:" + name);
                        }, function (errorMessage) {
                            console.log("添加文件成功" + name);
                            Forguncy.modifyTablesData({
                                image: {
                                    addRows: [{
                                        name: name,
                                        code: imgArray[index],
                                        base64: base64,
                                        url: imageUrl,
                                        is_identify: "未识别"
                                    }],
                                }
                            });
                        })

                    })
                    .catch(error => {
                        console.error(error);
                        //getPdfBase64返回base64
                    });
            }
            xhr.send();
        } else {
            alert("识别到无效文件");
        }
    })(i);
}


function getBase64Image(img) {
    var canvas = document.createElement("canvas");
    canvas.width = img.width;
    canvas.height = img.height;
    var ctx = canvas.getContext("2d");
    ctx.drawImage(img, 0, 0, img.width, img.height);
    var ext = img.src.substring(img.src.lastIndexOf(".") + 1).toLowerCase();
    var dataURL = canvas.toDataURL("image/" + ext);
    return dataURL;
}


function getImgName(input) {
    var match = input.match(/_([^]*)/);
    var result = match && match[1];
    return result || null;
}

function getFileType(inputString) {
    let dotIndex = inputString.lastIndexOf('.');  // 找到最后一个句点的索引
    if (dotIndex !== -1) {
        // 获取句点后的子字符串
        return inputString.slice(dotIndex + 1);
    } else {
        return "文件名不合法";
    }
}

2.清空文件列表

2.1.js代码，活字格好像有点毛病，有时候js代码不生效

Forguncy.modifyTablesData({
    image: {
        editRows: [
            {
                primaryKey:
                    {
                        is_identify: "未识别"
                    },
                values: {
                    is_identify: "past_未识别"
                }
            },
            {
                primaryKey:
                    {
                        is_identify: "已识别"
                    },
                values: {
                    is_identify: "past_已识别"
                }
            },
            {
                primaryKey:
                    {
                        is_identify: "已验真"
                    },
                values: {
                    is_identify: "past_已验真"
                }
            },
            {
                primaryKey:
                    {
                        is_identify: "重复识别"
                    },
                values: {
                    is_identify: "past_重复识别"
                }
            },
        ]
    },
});