@eagleoutice/flowr/statistics/features/supported/used-packages/post-process.js

Static Dataflow Analyzer and Program Slicer for the R Programming Language

"use strict";
var __importDefault = (this && this.__importDefault) || function (mod) {
    return (mod && mod.__esModule) ? mod : { "default": mod };
};
Object.defineProperty(exports, "__esModule", { value: true });
exports.postProcess = postProcess;
const post_processing_1 = require("../../post-processing");
const used_packages_1 = require("./used-packages");
const fs_1 = __importDefault(require("fs"));
const path_1 = __importDefault(require("path"));
const summarizer_1 = require("../../../../util/summarizer");
const files_1 = require("../../../../util/files");
const strings_1 = require("../../../../util/strings");
const arrays_1 = require("../../../../util/arrays");
function postProcess(featureRoot, info, outputPath, config) {
    const collected = {};
    for (const [filepath, data] of info.entries()) {
        const value = data.usedPackages;
        for (const [key, val] of Object.entries(value)) {
            let get = collected[key];
            if (!get) {
                get = (0, post_processing_1.emptySummarizedWithProject)();
                collected[key] = get;
            }
            get.count.push(val);
            if (val > 0) {
                (0, post_processing_1.recordFilePath)(get, filepath, config);
            }
        }
    }
    const variablesOutStream = fs_1.default.createWriteStream(path_1.default.join(outputPath, 'loading-functions.csv'));
    variablesOutStream.write(`kind,unique-projects,unique-files,${(0, summarizer_1.summarizedMeasurement2CsvHeader)()}\n`);
    for (const [key, val] of Object.entries(collected)) {
        const data = val;
        const sum = (0, summarizer_1.summarizeMeasurement)(data.count);
        variablesOutStream.write(`${JSON.stringify(key)},${data.uniqueProjects.size},${data.uniqueFiles.size},${(0, summarizer_1.summarizedMeasurement2Csv)(sum)}\n`);
    }
    variablesOutStream.close();
    // now we want to collect the names of the loaded libraries,
    // we collect and store them separately (per kind) but also, we want store the summarized results in the end!
    const loadedLibraries = new Map();
    for (const key of Object.keys(used_packages_1.initialUsedPackageInfos)) {
        const data = retrieveDataForLoad(key, featureRoot, outputPath, config);
        for (const [name, val] of data.entries()) {
            let get = loadedLibraries.get(name);
            if (!get) {
                get = (0, post_processing_1.emptySummarizedWithProject)();
                loadedLibraries.set(name, get);
            }
            get.count.push(...val.count);
            for (const uniqueFile of val.uniqueFiles) {
                get.uniqueFiles.add(uniqueFile);
            }
            for (const uniqueProject of val.uniqueProjects) {
                get.uniqueProjects.add(uniqueProject);
            }
        }
    }
    const output = path_1.default.join(outputPath, 'all-operators.csv');
    const out = fs_1.default.createWriteStream(output);
    out.write(`kind,unique-projects,unique-files,${(0, summarizer_1.summarizedMeasurement2CsvHeader)()}\n`);
    for (const [key, val] of loadedLibraries.entries()) {
        const { count, uniqueProjects, uniqueFiles } = val;
        const sum = (0, summarizer_1.summarizeMeasurement)(count);
        out.write(`${JSON.stringify(key)},${uniqueProjects.size},${uniqueFiles.size},${(0, summarizer_1.summarizedMeasurement2Csv)(sum)}\n`);
    }
    out.close();
}
// directly writes the results to the output path
function retrieveDataForLoad(operator, readFromPath, outputPath, config) {
    const input = path_1.default.join(readFromPath, `${operator}.txt`);
    if (!fs_1.default.existsSync(input)) {
        // if there is nothing with this, just return :)
        return new Map();
    }
    const collected = new Map();
    (0, files_1.readLineByLineSync)(input, (line, lineNumber) => {
        if (line.length === 0) {
            return;
        }
        if (lineNumber % 2_500 === 0) {
            console.log(`    Processing line ${lineNumber} from ${input}`);
        }
        const [packages, context] = JSON.parse(line.toString());
        // first, we have to collect what this file gives us
        // we normalize surrounding quotation marks
        const bag = (0, arrays_1.array2bag)(packages.map(p => {
            if ((0, strings_1.startAndEndsWith)(p, '"') || (0, strings_1.startAndEndsWith)(p, "'") || (0, strings_1.startAndEndsWith)(p, '`')) {
                return p.slice(1, -1);
            }
            else {
                return p;
            }
        }));
        // now we merge it into the global map (oh gosh this is so horrible
        for (const [name, count] of bag.entries()) {
            let get = collected.get(name);
            if (!get) {
                get = (0, post_processing_1.emptySummarizedWithProject)();
                collected.set(name, get);
            }
            get.count.push(count);
            if (count > 0) {
                (0, post_processing_1.recordFilePath)(get, context, config);
            }
        }
    });
    const output = path_1.default.join(outputPath, `${operator}.csv`);
    const out = fs_1.default.createWriteStream(output);
    out.write(`kind,unique-projects,unique-files,${(0, summarizer_1.summarizedMeasurement2CsvHeader)()}\n`);
    for (const [key, val] of collected.entries()) {
        const { count, uniqueProjects, uniqueFiles } = val;
        const sum = (0, summarizer_1.summarizeMeasurement)(count);
        out.write(`${JSON.stringify(key)},${uniqueProjects.size},${uniqueFiles.size},${(0, summarizer_1.summarizedMeasurement2Csv)(sum)}\n`);
    }
    out.close();
    return collected;
}
//# sourceMappingURL=post-process.js.map