brazilian-courts-scrappers/dist/scrappers/Pje1gTjbaAndamentosScrapper.js

Data scrapper para fazer raspagem de dados de processos judiciais dos portais de tribunais do Brasil.

"use strict";
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
    function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
    return new (P || (P = Promise))(function (resolve, reject) {
        function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
        function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
        function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
        step((generator = generator.apply(thisArg, _arguments || [])).next());
    });
};
var __importDefault = (this && this.__importDefault) || function (mod) {
    return (mod && mod.__esModule) ? mod : { "default": mod };
};
Object.defineProperty(exports, "__esModule", { value: true });
const ScrappedAndamento_1 = __importDefault(require("../data-structures/ScrappedAndamento"));
const utils_1 = require("../utils");
const AndamentosScrapper_1 = __importDefault(require("./AndamentosScrapper"));
const Pje1gTjbaProcessoScrapper_1 = __importDefault(require("./Pje1gTjbaProcessoScrapper"));
class Pje1gTjbaAndamentosScrapper extends AndamentosScrapper_1.default {
    constructor(doc) {
        super(doc);
        this.doc = doc;
    }
    fetchAndamentosInfo() {
        const _super = Object.create(null, {
            fetchAndamentosInfo: { get: () => super.fetchAndamentosInfo }
        });
        return __awaiter(this, void 0, void 0, function* () {
            return yield _super.fetchAndamentosInfo.call(this);
        });
    }
    loadPageCheckpoints() {
        this.divTimeline = this.doc.getElementById("divTimeLine:eventosTimeLineElement");
    }
    getAndamentos() {
        var _a, _b, _c;
        return __awaiter(this, void 0, void 0, function* () {
            if (this.divTimeline == null)
                return [];
            const andamentosMediaDivs = this.divTimeline.querySelectorAll(":scope > .media:not(.data)");
            const andamentos = [];
            for (const mediaDiv of Array.from(andamentosMediaDivs)) {
                const andamentoPjeType = Array.from(mediaDiv.classList).includes("tipo-D")
                    ? "documento"
                    : "movimentação";
                const bodyBoxDiv = mediaDiv.querySelector(".media-body");
                if (!bodyBoxDiv)
                    continue;
                const horaAndamento = (_c = (_b = (_a = bodyBoxDiv.querySelector("small.text-muted")) === null || _a === void 0 ? void 0 : _a.textContent) === null || _b === void 0 ? void 0 : _b.trim()) !== null && _c !== void 0 ? _c : "";
                const [id, cancelado, docName, docContent] = yield this.getDocumentInfo(bodyBoxDiv, andamentoPjeType);
                const data = this.getDate(mediaDiv, horaAndamento);
                const nomeOriginalSistemaJustica = this.getNome(bodyBoxDiv, docName);
                const fullObservacao = `${docName !== null && docName !== void 0 ? docName : ""}\n${docContent !== null && docContent !== void 0 ? docContent : ""}`;
                const observacao = (0, utils_1.stripBlankLines)(fullObservacao);
                const andamento = new ScrappedAndamento_1.default(nomeOriginalSistemaJustica, data, id, observacao, undefined, cancelado);
                andamentos.unshift(andamento);
            }
            return andamentos;
        });
    }
    getDocumentInfo(bodyBoxDiv, andamentoType) {
        return __awaiter(this, void 0, void 0, function* () {
            if (andamentoType !== "documento") {
                return new Promise(resolve => resolve([]));
            }
            let mainDocumentNameString;
            let docContent = "";
            const isCancelado = this.isCancelado(bodyBoxDiv);
            if (isCancelado) {
                mainDocumentNameString = bodyBoxDiv
                    .querySelector(".anexos > .anexos-inativos > span")
                    .textContent.trim();
            }
            else {
                const mainDocumentA = bodyBoxDiv.querySelector(".anexos > a:first-child");
                mainDocumentNameString = mainDocumentA
                    .querySelector(":scope > span")
                    .textContent.trim();
                docContent = yield this.getDocumentTextContentIfExists(mainDocumentA);
            }
            const id = this.getIdFromDocString(mainDocumentNameString);
            const docName = this.getNameFromDocString(mainDocumentNameString);
            return [id, isCancelado, docName, docContent];
        });
    }
    getIdFromDocString(docString) {
        const idRegex = /^\d+(?= -)/;
        const match = docString.match(idRegex);
        return match ? match[0] : "";
    }
    getNameFromDocString(docString) {
        const idRegex = /(?<=^\d+ - ).+/;
        const match = docString.match(idRegex);
        return match ? match[0] : "";
    }
    getDate(mediaDiv, horaAndamento) {
        const dateString = this.findDateSibling(mediaDiv).textContent.trim();
        return Pje1gTjbaProcessoScrapper_1.default.getDateFromPjeTjbaDateString(dateString, horaAndamento);
    }
    findDateSibling(mediaDiv) {
        const prevSibling = mediaDiv.previousElementSibling;
        if (Array.from(prevSibling.classList).includes("data"))
            return prevSibling;
        return this.findDateSibling(prevSibling);
    }
    getNome(bodyBoxDiv, filename = "") {
        if (filename !== "") {
            const matchParenthesisResults = filename.match(/.+(?= (\(.+\))$)/);
            return matchParenthesisResults
                ? matchParenthesisResults[0].trim()
                : filename;
        }
        const titleDivs = bodyBoxDiv.querySelectorAll(":scope > :not(.anexos, .col-sm-12)");
        const titles = Array.from(titleDivs).map(titleDiv => titleDiv.querySelector("span").textContent.trim());
        return titles[0];
    }
    isCancelado(bodyBoxDiv) {
        const inativosDiv = bodyBoxDiv.querySelector(".anexos > .anexos-inativos");
        return !!inativosDiv;
        // TODO: buscar exemplo de andamento cancelado SEM DOCUMENTO no PJe
    }
    getDocumentTextContentIfExists(documentAnchor) {
        return __awaiter(this, void 0, void 0, function* () {
            if (this.isPdf(documentAnchor))
                return new Promise(resolve => resolve(""));
            documentAnchor.click();
            const contentDoc = yield this.waitPageLoad("#frameHtml");
            yield new Promise(resolve => setTimeout(resolve, 250));
            if (this.isPjeDocumentLandingPageBug(contentDoc))
                return new Promise(resolve => resolve(""));
            else
                return this.getDocumentTextContent(contentDoc);
        });
    }
    isPdf(documentAnchor) {
        const icon = documentAnchor.querySelector(":scope > i:first-child");
        return Array.from(icon.classList).includes("fa-file-pdf-o");
    }
    waitPageLoad(iframeDocSelector) {
        var _a, _b, _c;
        return __awaiter(this, void 0, void 0, function* () {
            let readystateComplete = false;
            let bodyLengthChanged = false;
            let bodyLength = undefined;
            let iframe;
            while (!readystateComplete || bodyLengthChanged) {
                yield new Promise(resolve => setTimeout(resolve, 350));
                iframe = this.doc.querySelector(iframeDocSelector);
                readystateComplete = ((_a = iframe === null || iframe === void 0 ? void 0 : iframe.contentDocument) === null || _a === void 0 ? void 0 : _a.readyState) === "complete";
                bodyLengthChanged =
                    bodyLength !== ((_b = iframe === null || iframe === void 0 ? void 0 : iframe.contentDocument) === null || _b === void 0 ? void 0 : _b.body.children.length);
                bodyLength = (_c = iframe === null || iframe === void 0 ? void 0 : iframe.contentDocument) === null || _c === void 0 ? void 0 : _c.body.children.length;
            }
            return iframe.contentDocument;
        });
    }
    isPjeDocumentLandingPageBug(htmlDoc) {
        if (htmlDoc.head.children.length > 3)
            return true;
        else
            return false;
    }
    getDocumentTextContent(contentDoc) {
        const noScriptInnerHtml = (0, utils_1.stripScriptTagsFromHtmlString)(contentDoc.body.innerHTML);
        return (0, utils_1.getTextContent)(noScriptInnerHtml, contentDoc);
    }
}
exports.default = Pje1gTjbaAndamentosScrapper;
//# sourceMappingURL=Pje1gTjbaAndamentosScrapper.js.map