@tricoteuses/senat 2.22.11 → 2.22.13

package/lib/model/texte.js

@@ -0,0 +1,228 @@
+import { JSDOM } from "jsdom";
+import { AKN_IDENTIFICATION_STRUCTURE_REGEXP, AKN_WORKFLOW_IDENTIFICATION_STRUCTURE_REGEXP } from "../scripts/datautil";
+import { DivisionType, } from "../types/texte";
+function buildWorklow(metaElement) {
+    const stepElements = metaElement.querySelectorAll("workflow step");
+    const steps = [];
+    for (const stepElement of stepElements) {
+        const identification = stepElement.getAttribute("href") ?? "";
+        const identificationParts = AKN_WORKFLOW_IDENTIFICATION_STRUCTURE_REGEXP.exec(identification)?.groups;
+        steps.push({
+            eId: stepElement.getAttribute("eId"),
+            date: stepElement.getAttribute("date") ? new Date(stepElement.getAttribute("date") ?? "") : null,
+            type: identificationParts?.["type"] || null,
+            session: identificationParts?.["session"] || null,
+            numero: identificationParts?.["numTexte"] || null,
+            version: identificationParts?.["version"] ? identificationParts["version"] : null,
+            outcome: stepElement.getAttribute("outcome"),
+        });
+    }
+    return steps;
+}
+function buildDivision(node, index) {
+    const eId = node.getAttribute("eId");
+    const tag = node.nodeName;
+    const level = DivisionType[tag];
+    const titleNode = node.querySelector("num");
+    const subtitleNode = node.querySelector("heading");
+    const headings = [
+        ...(titleNode
+            ? [
+                {
+                    text: titleNode.textContent?.trim() ?? null,
+                    html: titleNode.innerHTML?.trim() ?? null,
+                },
+            ]
+            : []),
+        ...(subtitleNode
+            ? [
+                {
+                    text: subtitleNode.textContent?.trim() ?? null,
+                    html: subtitleNode.innerHTML?.trim() ?? null,
+                },
+            ]
+            : []),
+    ];
+    const division = {
+        index,
+        eId,
+        tag,
+        level,
+        headings,
+    };
+    if (tag === "article") {
+        ;
+        division.alineas = [];
+    }
+    return division;
+}
+function buildAlinea(contentNode, alineaNode) {
+    const eId = alineaNode.getAttribute("eId");
+    const heading = {
+        text: alineaNode.querySelector("num")?.textContent ?? null,
+    };
+    const pastille = alineaNode.getAttribute("data:pastille") ?? null;
+    return {
+        eId,
+        heading,
+        text: contentNode.textContent?.trim() ?? null,
+        html: contentNode.innerHTML?.trim() ?? null,
+        pastille,
+    };
+}
+function buildEmptyArticle(index) {
+    return {
+        index: index,
+        eId: "",
+        tag: "article",
+        level: DivisionType["article"],
+        headings: [],
+        alineas: [],
+    };
+}
+function flattenTexte(texteContentRoot) {
+    const divisions = [];
+    let divisionIndex = 0;
+    const iter = (node) => {
+        if (node.nodeName === "content") {
+            return;
+        }
+        switch (node.nodeName) {
+            case "tome":
+            case "part":
+            case "book":
+            case "title":
+            case "subtitle":
+            case "chapter":
+            case "section":
+            case "subsection":
+            case "paragraph":
+            case "article":
+                divisions.push(buildDivision(node, divisionIndex++));
+                break;
+        }
+        if (node.nodeName === "alinea") {
+            Array.from(node.childNodes)
+                // Find direct content children programmatically
+                // because `:scope` selector does not work
+                // https://github.com/jsdom/jsdom/issues/2998
+                .filter((alineaChildNode) => alineaChildNode.nodeName === "content")
+                .forEach((alineaContentNode) => {
+                // Hypothesis: alineas should always be enclosed in articles
+                let lastArticle = divisions.findLast((division) => division.tag === "article");
+                if (!lastArticle) {
+                    lastArticle = buildEmptyArticle(divisionIndex++);
+                    divisions.push(lastArticle);
+                }
+                lastArticle.alineas.push(buildAlinea(alineaContentNode, node));
+            });
+        }
+        if (node.hasChildNodes()) {
+            node.childNodes.forEach((childNode) => iter(childNode));
+        }
+    };
+    iter(texteContentRoot);
+    return divisions;
+}
+export function transformTexte(document) {
+    const metaElement = document.querySelector("meta");
+    const preambleElement = document.querySelector("preamble");
+    const identification = metaElement?.querySelector("FRBRExpression FRBRuri")?.getAttribute("value") ?? "";
+    const identificationParts = AKN_IDENTIFICATION_STRUCTURE_REGEXP.exec(identification)?.groups;
+    const bodyElement = document.querySelector("body");
+    const sessionYears = identificationParts?.["session"]?.split("-") || null;
+    const datePresentation = metaElement?.querySelector("FRBRdate[name='#presentation']")?.getAttribute("date");
+    const dateDepot = metaElement?.querySelector("FRBRdate[name='#depot']")?.getAttribute("date");
+    const datePublicationXml = metaElement?.querySelector("FRBRdate[name='#publication-xml']")?.getAttribute("date");
+    return {
+        titre: preambleElement?.querySelector("docTitle")?.textContent || null,
+        titreCourt: metaElement?.querySelector("FRBRalias[name='intitule-court']")?.getAttribute("value") || null,
+        signetDossier: metaElement?.querySelector("FRBRalias[name='signet-dossier-legislatif-senat']")?.getAttribute("value") || null,
+        urlDossierSenat: metaElement?.querySelector("FRBRalias[name='url-senat']")?.getAttribute("value") || null,
+        urlDossierAssemblee: metaElement?.querySelector("FRBRalias[name='url-AN']")?.getAttribute("value") || null,
+        type: identificationParts?.["type"] || null,
+        session: sessionYears && sessionYears.length > 0 ? sessionYears[0] : null,
+        numero: identificationParts?.["numTexte"] ? parseInt(identificationParts["numTexte"]) : null,
+        datePresentation: datePresentation ? new Date(datePresentation) : null,
+        dateDepot: dateDepot ? new Date(dateDepot) : null,
+        datePublicationXml: datePublicationXml ? new Date(datePublicationXml) : null,
+        version: identificationParts?.["version"] ? identificationParts["version"] : null,
+        workflow: metaElement ? buildWorklow(metaElement) : [],
+        divisions: bodyElement ? flattenTexte(bodyElement) : [],
+    };
+}
+export function transformExposeDesMotifs(document) {
+    const sectionElements = document.querySelectorAll("section");
+    const exposeDesMotifsRegexp = new RegExp("EXPOS.{1,2}[\\n\\s]DES[\\n\\s]MOTIFS");
+    for (const sectionElement of sectionElements) {
+        const firstParagraph = sectionElement.querySelector("p:first-of-type");
+        const secondParagraph = sectionElement.querySelector("p:nth-of-type(2)");
+        if (!firstParagraph) {
+            continue;
+        }
+        const firstParagraphContent = firstParagraph.textContent;
+        const secondParagraphContent = secondParagraph?.textContent;
+        if (!firstParagraphContent || !exposeDesMotifsRegexp.test(firstParagraphContent.toUpperCase())) {
+            if (!secondParagraphContent || !exposeDesMotifsRegexp.test(secondParagraphContent.toUpperCase())) {
+                continue;
+            }
+            else {
+                secondParagraph.remove();
+            }
+        }
+        firstParagraph.remove();
+        return {
+            text: sectionElement.textContent?.trim() ?? null,
+            html: sectionElement.innerHTML?.trim() ?? null,
+        };
+    }
+    return null;
+}
+export function parseTexte(texteXml) {
+    try {
+        const { document } = new JSDOM(texteXml, {
+            contentType: "text/xml",
+        }).window;
+        return transformTexte(document);
+    }
+    catch (error) {
+        console.error(`Could not parse texte with error ${error}`);
+    }
+    return null;
+}
+// Prevent from memory leak
+// https://github.com/jsdom/jsdom/issues/2583#issuecomment-559520814
+export async function parseTexteFromFile(xmlFilePath) {
+    try {
+        const { document } = (await JSDOM.fromFile(xmlFilePath, { contentType: "text/xml" })).window;
+        return transformTexte(document);
+    }
+    catch (error) {
+        console.error(`Could not parse texte with error ${error}`);
+    }
+    return null;
+}
+export function parseExposeDesMotifs(exposeDesMotifsHtml) {
+    try {
+        const { document } = new JSDOM(exposeDesMotifsHtml, {
+            contentType: "text/html",
+        }).window;
+        return transformExposeDesMotifs(document);
+    }
+    catch (error) {
+        console.error(`Could not parse exposé des motifs with error ${error}`);
+    }
+    return null;
+}
+// Prevent from memory leak
+// https://github.com/jsdom/jsdom/issues/2583#issuecomment-559520814
+export async function parseExposeDesMotifsFromFile(htmlFilePath) {
+    try {
+        const { document } = (await JSDOM.fromFile(htmlFilePath, { contentType: "text/html" })).window;
+        return transformExposeDesMotifs(document);
+    }
+    catch (error) {
+        console.error(`Could not parse exposé des motifs with error ${error}`);
+    }
+    return null;
+}

package/lib/model/util.d.ts

@@ -0,0 +1,9 @@
+import { Expression } from "kysely";
+export declare function concat(...exprs: Expression<number | string | null | undefined>[]): import("kysely").RawBuilder<string>;
+export declare function expandToRows(expr: Expression<string | null | undefined>, regexp: Expression<string>): import("kysely").RawBuilder<unknown>;
+export declare function nullIf(expr: Expression<number | string | null | undefined>): import("kysely").RawBuilder<unknown>;
+export declare function removeSubstring(expr: Expression<string | null | undefined>, pattern: Expression<string>): import("kysely").RawBuilder<string>;
+export declare function replace(expr: Expression<string | null | undefined>, pattern: Expression<string>, replacement: Expression<string>): import("kysely").RawBuilder<string>;
+export declare function rtrim(expr: Expression<string | null | undefined>): import("kysely").RawBuilder<string>;
+export declare function toDateString(expr: Expression<Date | null | undefined>, format?: Expression<string>): import("kysely").RawBuilder<string>;
+export declare function toCRDate(dateISO: string, startTime?: string | null): string;

package/lib/model/util.js

@@ -0,0 +1,38 @@
+import { sql } from "kysely";
+import { STANDARD_DATE_FORMAT } from "../scripts/datautil";
+export function concat(...exprs) {
+    return sql.join(exprs, sql `||`).$castTo();
+}
+export function expandToRows(expr, regexp) {
+    return sql `unnest(regexp_matches(${expr}, ${regexp}, 'g'))`;
+}
+export function nullIf(expr) {
+    return sql `nullif(trim(${expr}), '')`;
+}
+export function removeSubstring(expr, pattern) {
+    return replace(expr, pattern, sql.val(""));
+}
+export function replace(expr, pattern, replacement) {
+    return sql `replace(${expr}, ${pattern}, ${replacement})`;
+}
+export function rtrim(expr) {
+    return sql `rtrim(${expr})`;
+}
+export function toDateString(expr, format = sql.val(STANDARD_DATE_FORMAT)) {
+    return sql `to_char(${expr}, ${format})`;
+}
+export function toCRDate(dateISO, startTime) {
+    const yyyymmdd = dateISO.replace(/-/g, ""); // "20250716"
+    let hh = "00", mm = "00", ss = "00", SSS = "000";
+    if (startTime) {
+        // accepte "HH:MM:SS", "HH:MM:SS.mmm", "HH:MM:SS.mmm+02:00"
+        const m = startTime.match(/(\d{2}):(\d{2}):(\d{2})(?:\.(\d{3}))?/);
+        if (m) {
+            hh = m[1];
+            mm = m[2];
+            ss = m[3];
+            SSS = m[4] || "000";
+        }
+    }
+    return `${yyyymmdd}${hh}${mm}${ss}${SSS}`;
+}

package/lib/parsers/texte.d.ts

@@ -0,0 +1,7 @@
+import { ExposeDesMotifs, FlatTexte } from "../types/texte";
+export declare function transformTexte(document: Document): FlatTexte | null;
+export declare function transformExposeDesMotifs(document: Document): ExposeDesMotifs | null;
+export declare function parseTexte(texteXml: string): FlatTexte | null;
+export declare function parseTexteFromFile(xmlFilePath: string): Promise<FlatTexte | null>;
+export declare function parseExposeDesMotifs(exposeDesMotifsHtml: string): ExposeDesMotifs | null;
+export declare function parseExposeDesMotifsFromFile(htmlFilePath: string): Promise<ExposeDesMotifs | null>;

package/lib/parsers/texte.js

@@ -0,0 +1,228 @@
+import { JSDOM } from "jsdom";
+import { AKN_IDENTIFICATION_STRUCTURE_REGEXP, AKN_WORKFLOW_IDENTIFICATION_STRUCTURE_REGEXP } from "../scripts/datautil";
+import { DivisionType, } from "../types/texte";
+function buildWorklow(metaElement) {
+    const stepElements = metaElement.querySelectorAll("workflow step");
+    const steps = [];
+    for (const stepElement of stepElements) {
+        const identification = stepElement.getAttribute("href") ?? "";
+        const identificationParts = AKN_WORKFLOW_IDENTIFICATION_STRUCTURE_REGEXP.exec(identification)?.groups;
+        steps.push({
+            eId: stepElement.getAttribute("eId"),
+            date: stepElement.getAttribute("date") ? new Date(stepElement.getAttribute("date") ?? "") : null,
+            type: identificationParts?.["type"] || null,
+            session: identificationParts?.["session"] || null,
+            numero: identificationParts?.["numTexte"] || null,
+            version: identificationParts?.["version"] ? identificationParts["version"] : null,
+            outcome: stepElement.getAttribute("outcome"),
+        });
+    }
+    return steps;
+}
+function buildDivision(node, index) {
+    const eId = node.getAttribute("eId");
+    const tag = node.nodeName;
+    const level = DivisionType[tag];
+    const titleNode = node.querySelector("num");
+    const subtitleNode = node.querySelector("heading");
+    const headings = [
+        ...(titleNode
+            ? [
+                {
+                    text: titleNode.textContent?.trim() ?? null,
+                    html: titleNode.innerHTML?.trim() ?? null,
+                },
+            ]
+            : []),
+        ...(subtitleNode
+            ? [
+                {
+                    text: subtitleNode.textContent?.trim() ?? null,
+                    html: subtitleNode.innerHTML?.trim() ?? null,
+                },
+            ]
+            : []),
+    ];
+    const division = {
+        index,
+        eId,
+        tag,
+        level,
+        headings,
+    };
+    if (tag === "article") {
+        ;
+        division.alineas = [];
+    }
+    return division;
+}
+function buildAlinea(contentNode, alineaNode) {
+    const eId = alineaNode.getAttribute("eId");
+    const heading = {
+        text: alineaNode.querySelector("num")?.textContent ?? null,
+    };
+    const pastille = alineaNode.getAttribute("data:pastille") ?? null;
+    return {
+        eId,
+        heading,
+        text: contentNode.textContent?.trim() ?? null,
+        html: contentNode.innerHTML?.trim() ?? null,
+        pastille,
+    };
+}
+function buildEmptyArticle(index) {
+    return {
+        index: index,
+        eId: "",
+        tag: "article",
+        level: DivisionType["article"],
+        headings: [],
+        alineas: [],
+    };
+}
+function flattenTexte(texteContentRoot) {
+    const divisions = [];
+    let divisionIndex = 0;
+    const iter = (node) => {
+        if (node.nodeName === "content") {
+            return;
+        }
+        switch (node.nodeName) {
+            case "tome":
+            case "part":
+            case "book":
+            case "title":
+            case "subtitle":
+            case "chapter":
+            case "section":
+            case "subsection":
+            case "paragraph":
+            case "article":
+                divisions.push(buildDivision(node, divisionIndex++));
+                break;
+        }
+        if (node.nodeName === "alinea") {
+            Array.from(node.childNodes)
+                // Find direct content children programmatically
+                // because `:scope` selector does not work
+                // https://github.com/jsdom/jsdom/issues/2998
+                .filter((alineaChildNode) => alineaChildNode.nodeName === "content")
+                .forEach((alineaContentNode) => {
+                // Hypothesis: alineas should always be enclosed in articles
+                let lastArticle = divisions.findLast((division) => division.tag === "article");
+                if (!lastArticle) {
+                    lastArticle = buildEmptyArticle(divisionIndex++);
+                    divisions.push(lastArticle);
+                }
+                lastArticle.alineas.push(buildAlinea(alineaContentNode, node));
+            });
+        }
+        if (node.hasChildNodes()) {
+            node.childNodes.forEach((childNode) => iter(childNode));
+        }
+    };
+    iter(texteContentRoot);
+    return divisions;
+}
+export function transformTexte(document) {
+    const metaElement = document.querySelector("meta");
+    const preambleElement = document.querySelector("preamble");
+    const identification = metaElement?.querySelector("FRBRExpression FRBRuri")?.getAttribute("value") ?? "";
+    const identificationParts = AKN_IDENTIFICATION_STRUCTURE_REGEXP.exec(identification)?.groups;
+    const bodyElement = document.querySelector("body");
+    const sessionYears = identificationParts?.["session"]?.split("-") || null;
+    const datePresentation = metaElement?.querySelector("FRBRdate[name='#presentation']")?.getAttribute("date");
+    const dateDepot = metaElement?.querySelector("FRBRdate[name='#depot']")?.getAttribute("date");
+    const datePublicationXml = metaElement?.querySelector("FRBRdate[name='#publication-xml']")?.getAttribute("date");
+    return {
+        titre: preambleElement?.querySelector("docTitle")?.textContent || null,
+        titreCourt: metaElement?.querySelector("FRBRalias[name='intitule-court']")?.getAttribute("value") || null,
+        signetDossier: metaElement?.querySelector("FRBRalias[name='signet-dossier-legislatif-senat']")?.getAttribute("value") || null,
+        urlDossierSenat: metaElement?.querySelector("FRBRalias[name='url-senat']")?.getAttribute("value") || null,
+        urlDossierAssemblee: metaElement?.querySelector("FRBRalias[name='url-AN']")?.getAttribute("value") || null,
+        type: identificationParts?.["type"] || null,
+        session: sessionYears && sessionYears.length > 0 ? sessionYears[0] : null,
+        numero: identificationParts?.["numTexte"] ? parseInt(identificationParts["numTexte"]) : null,
+        datePresentation: datePresentation ? new Date(datePresentation) : null,
+        dateDepot: dateDepot ? new Date(dateDepot) : null,
+        datePublicationXml: datePublicationXml ? new Date(datePublicationXml) : null,
+        version: identificationParts?.["version"] ? identificationParts["version"] : null,
+        workflow: metaElement ? buildWorklow(metaElement) : [],
+        divisions: bodyElement ? flattenTexte(bodyElement) : [],
+    };
+}
+export function transformExposeDesMotifs(document) {
+    const sectionElements = document.querySelectorAll("section");
+    const exposeDesMotifsRegexp = new RegExp("EXPOS.{1,2}[\\n\\s]DES[\\n\\s]MOTIFS");
+    for (const sectionElement of sectionElements) {
+        const firstParagraph = sectionElement.querySelector("p:first-of-type");
+        const secondParagraph = sectionElement.querySelector("p:nth-of-type(2)");
+        if (!firstParagraph) {
+            continue;
+        }
+        const firstParagraphContent = firstParagraph.textContent;
+        const secondParagraphContent = secondParagraph?.textContent;
+        if (!firstParagraphContent || !exposeDesMotifsRegexp.test(firstParagraphContent.toUpperCase())) {
+            if (!secondParagraphContent || !exposeDesMotifsRegexp.test(secondParagraphContent.toUpperCase())) {
+                continue;
+            }
+            else {
+                secondParagraph.remove();
+            }
+        }
+        firstParagraph.remove();
+        return {
+            text: sectionElement.textContent?.trim() ?? null,
+            html: sectionElement.innerHTML?.trim() ?? null,
+        };
+    }
+    return null;
+}
+export function parseTexte(texteXml) {
+    try {
+        const { document } = new JSDOM(texteXml, {
+            contentType: "text/xml",
+        }).window;
+        return transformTexte(document);
+    }
+    catch (error) {
+        console.error(`Could not parse texte with error ${error}`);
+    }
+    return null;
+}
+// Prevent from memory leak
+// https://github.com/jsdom/jsdom/issues/2583#issuecomment-559520814
+export async function parseTexteFromFile(xmlFilePath) {
+    try {
+        const { document } = (await JSDOM.fromFile(xmlFilePath, { contentType: "text/xml" })).window;
+        return transformTexte(document);
+    }
+    catch (error) {
+        console.error(`Could not parse texte with error ${error}`);
+    }
+    return null;
+}
+export function parseExposeDesMotifs(exposeDesMotifsHtml) {
+    try {
+        const { document } = new JSDOM(exposeDesMotifsHtml, {
+            contentType: "text/html",
+        }).window;
+        return transformExposeDesMotifs(document);
+    }
+    catch (error) {
+        console.error(`Could not parse exposé des motifs with error ${error}`);
+    }
+    return null;
+}
+// Prevent from memory leak
+// https://github.com/jsdom/jsdom/issues/2583#issuecomment-559520814
+export async function parseExposeDesMotifsFromFile(htmlFilePath) {
+    try {
+        const { document } = (await JSDOM.fromFile(htmlFilePath, { contentType: "text/html" })).window;
+        return transformExposeDesMotifs(document);
+    }
+    catch (error) {
+        console.error(`Could not parse exposé des motifs with error ${error}`);
+    }
+    return null;
+}