@tricoteuses/assemblee 1.10.5 → 2.0.0

package/lib/parsers/index.d.ts

@@ -1,3 +1,2 @@
-export { parseTexteLoi, retrieveTexteLoiParsed } from "./documents";
-export { parseTexte } from "./textes_lois";
-export { iterRechercheAmendements, iterRechercheRawAmendements, } from "./recherche_amendements";
+export { parseTexte } from './textes_lois';
+export { iterRechercheAmendements, iterRechercheRawAmendements, } from './recherche_amendements';

package/lib/parsers/plf.d.ts

@@ -0,0 +1 @@
+export declare function parseTexte(assembleeUrl: string, page: string): any;

package/lib/parsers/recherche_amendements.d.ts

@@ -1,5 +1,5 @@
-import { Amendement as RawAmendement } from "../raw_types/amendements";
-import { Amendement } from "../types/amendements";
+import { Amendement as RawAmendement } from '../raw_types/amendements';
+import { Amendement } from '../types/amendements';
 export declare function iterRechercheAmendements(amendementsSearchCacheDir: string, url: string, incremental: boolean, minDateDepot?: string | null, minDateExamen?: string | null, options?: {
     silent?: boolean;
     verbose?: boolean;

package/lib/parsers/textes_lois.d.ts

@@ -1,23 +1 @@
-export declare function parseTexte(assembleeUrl: string, page: string): {
-    error: {
-        code: number;
-        message: string;
-    };
-    html: string;
-    page: string;
-    subdivisions?: undefined;
-    url?: undefined;
-} | {
-    error: null;
-    html: string;
-    page: string;
-    subdivisions: {
-        id: string;
-        type: string;
-        niveau: number;
-        niveauRelatif: number;
-        titres: object[];
-        alineas: object[];
-    }[];
-    url: string;
-};
+export declare function parseTexte(assembleeUrl: string, page: string): any;

package/lib/parsers.js

@@ -0,0 +1,398 @@
+import e from "node:assert";
+import { JSDOM as z } from "jsdom";
+import { t as Q, a as v, p as k, q as J } from "./parse-Ccs6wcUg.js";
+import Z from "deep-equal";
+import q from "fs-extra";
+import K from "front-matter";
+import W from "js-yaml";
+import F from "node:path";
+import { parse as X } from "node-html-parser";
+import { d as G } from "./amendements-79bwpkvR.js";
+import { C as ee } from "./amendements-CV3s5a0M.js";
+function te(i) {
+  const t = [
+    { regex: /^(RAPPORT_)?ANNEXE(_|$)|^ETAT_/, level: 0, name: "Annexe" },
+    { regex: /^TOME_/, level: 1, name: "Tome" },
+    {
+      regex: /^PARTIE_|^(PREMIERE|SECONDE|DEUXIEME|TROISIEME|QUATRIEME)_PARTIE(_|$)/,
+      level: 2,
+      name: "Partie"
+    },
+    { regex: /^LIVRE_/, level: 3, name: "Livre" },
+    { regex: /^TITRE_/, level: 4, name: "Titre" },
+    { regex: /^SOUSTITRE_/, level: 5, name: "SousTitre" },
+    { regex: /^CHAPITRE_/, level: 6, name: "Chapitre" },
+    { regex: /^SECTION_/, level: 7, name: "Section" },
+    { regex: /^SOUSSECTION_/, level: 8, name: "SousSection" },
+    { regex: /^EXPOSE_DES_MOTIFS$/, level: 9, name: "ExposeMotifs" },
+    { regex: /^ARTICLES?(?!.*«)/, level: 9, name: "Article" },
+    { regex: /^(PROJET|PROPOSITION)_DE_LOI(_|$)/, level: -1, name: "PPL" },
+    { regex: /^PROPOSITION_DE_RESOLUTION(_|$)/, level: -1, name: "PR" },
+    {
+      regex: /^JEUX_OLYMPIQUES_ET_PARALYMPIQUES_DE_2024$/,
+      level: -1,
+      name: "JO"
+    },
+    {
+      regex: /^TEXTE_DE_LA_COMMISSION_MIXTE_PARITAIRE$/,
+      level: -1,
+      name: "CMP"
+    }
+  ];
+  for (const { regex: a, level: r, name: m } of t)
+    if (a.test(i))
+      return [r, m];
+  return [null, ""];
+}
+function P(i) {
+  return i?.replace(/[\n\t]+/g, "").trim() || "";
+}
+const ne = [
+  /^Délibéré en séance publique/,
+  /^Fait le/,
+  /^La Présidente,$/,
+  /^Le Président,$/,
+  /^Signé/
+];
+function ae(i) {
+  return ne.some((t) => t.test(i));
+}
+function he(i, t) {
+  const a = t.replace(/(<style[\w\W]+style>)/g, ""), { window: r } = new z(a), m = r.document;
+  e.strictEqual(m.children.length, 1);
+  const d = m.children[0];
+  e.strictEqual(d.children.length, 2);
+  const u = d.children[1];
+  let n = u.children?.[1];
+  if (u.children.length < 3 || n.tagName !== "BR")
+    return {
+      error: null,
+      subdivisions: [],
+      url: i
+    };
+  e.strictEqual(u.children[0].tagName, "DIV");
+  let o = null, s = !1, f = null, l = "section";
+  const c = [], S = [];
+  let p = [], g = [];
+  try {
+    for (; l !== null; )
+      switch (l) {
+        case "section":
+          n.nextElementSibling ? (n = n.nextElementSibling, n.tagName === "DIV" ? (o = n.firstElementChild, o && !n.id.includes("ftn") && (l = "alineaElement")) : e(
+            ["BR", "HR", "P"].includes(n.tagName),
+            `Unexpected tag name "${n.tagName}" for body child section`
+          )) : l = null;
+          break;
+        case "alineaElement":
+          if (!o) {
+            l = "nextAlineaElement";
+            break;
+          }
+          switch (o.tagName) {
+            case "DIV":
+            case "OL":
+            case "BR":
+            case "SPAN":
+              l = "nextAlineaElement";
+              break;
+            case "H4":
+            case "P":
+            case "TABLE":
+              break;
+            default:
+              return {
+                error: {
+                  code: -2,
+                  message: `Unexpected tag name for alinea element: ${o.tagName}`
+                }
+              };
+          }
+          const E = P(o.textContent).normalize("NFD").replace(/[\u0300-\u036f]/g, "").replace(/\(nouveau\)/, "").replace(/\(Pour coordination\)/, "").replace(/\(Supprimés?\)/, "").replace(/ /g, " ").replace(/[\-,.…]/g, "").trim().replace(/ {1,}/g, "_").toUpperCase(), [x, T] = te(E);
+          if (!E || x !== null && x < 0) {
+            l = "nextAlineaElement";
+            break;
+          }
+          const h = P(o.outerHTML), $ = P(o.textContent);
+          if (x === null) {
+            if (ae($)) {
+              l = "nextAlineaElement";
+              break;
+            }
+            g.length === 0 || s && p.length === 0 ? g.push({ texte: $, html: h }) : p.push({ texte: $, html: h }), s && g.length >= 2 && (s = !1);
+          } else {
+            for (f = x; c.length > 0 && f < c[c.length - 1]; )
+              c.pop();
+            (c.length === 0 || f > c[c.length - 1]) && c.push(f), p = [], g = [{ texte: $, html: h }], s = !E.match(/^ARTICLES?_/) && !E.match(/^EXPOSE_DES_MOTIFS$/), S.push({
+              id: `D_${E}`,
+              type: T,
+              niveau: f + 1,
+              niveauRelatif: c.length,
+              titres: g,
+              alineas: p
+            });
+          }
+          l = "nextAlineaElement";
+          break;
+        case "nextAlineaElement":
+          if (!o) {
+            l = "section";
+            break;
+          }
+          o = o.nextElementSibling, l = o ? "alineaElement" : "section";
+          break;
+        default:
+          throw new Error(`Unexpected state: ${l}`);
+      }
+    const w = {
+      error: null,
+      subdivisions: S,
+      url: i
+    };
+    return p = [], g = [], w;
+  } finally {
+    r.close();
+  }
+}
+function j(i, t) {
+  const a = Q(i, t?.in);
+  if (isNaN(+a))
+    throw new RangeError("Invalid time value");
+  const r = t?.format ?? "extended", m = t?.representation ?? "complete";
+  let d = "", u = "";
+  const n = r === "extended" ? "-" : "", o = r === "extended" ? ":" : "";
+  if (m !== "time") {
+    const s = v(a.getDate(), 2), f = v(a.getMonth() + 1, 2);
+    d = `${v(a.getFullYear(), 4)}${n}${f}${n}${s}`;
+  }
+  if (m !== "date") {
+    const s = a.getTimezoneOffset();
+    if (s !== 0) {
+      const g = Math.abs(s), w = v(Math.trunc(g / 60), 2), E = v(g % 60, 2);
+      u = `${s < 0 ? "+" : "-"}${w}:${E}`;
+    } else
+      u = "Z";
+    const f = v(a.getHours(), 2), l = v(a.getMinutes(), 2), c = v(a.getSeconds(), 2), S = d === "" ? "" : "T", p = [f, l, c].join(o);
+    d = `${d}${S}${p}${u}`;
+  }
+  return d;
+}
+async function* Se(i, t, a, r, m, d = {}) {
+  for await (const [
+    u,
+    n
+  ] of le(
+    i,
+    t,
+    a,
+    r,
+    m,
+    d
+  ))
+    G(n), yield [
+      u,
+      ee.toAmendement(JSON.stringify(n))
+    ];
+}
+async function* le(i, t, a, r, m, d = {}) {
+  await q.ensureDir(i), e.strictEqual(
+    /[?&]date_depot=/.exec(t),
+    null,
+    `URL ${t} already contains a deposit date`
+  ), e.strictEqual(
+    /[?&]order=/.exec(t),
+    null,
+    `URL ${t} already contains a sort order`
+  ), e.strictEqual(
+    /[?&]page=/.exec(t),
+    null,
+    `URL ${t} already contains a page number`
+  ), a && e.notStrictEqual(
+    /[?&]etat=/.exec(t),
+    null,
+    `In incremental mode, URL ${t} must contain an "etat" query parameter`
+  ), t += (t.includes("?") ? "&" : "?") + "order=date_depot,desc&page=1", r != null && e.notStrictEqual(
+    /\d{4}-\d{2}-\d{2}/.exec(r),
+    null,
+    `Invalid format for minimum date: ${r}`
+  );
+  let u = 0;
+  e: for (; ; ) {
+    d.verbose && console.log(`Fetching amendements search page at ${t}…`);
+    const n = await fetch(t);
+    e(
+      n.ok,
+      `Retrieval of search page at ${t} failed with error: ${n.status} ${n.statusText}`
+    );
+    const o = await n.text(), s = X(o);
+    if (s.querySelector("div.no-result") !== null)
+      break;
+    const f = s.querySelector("div.mirror-card-subtitle");
+    e.notStrictEqual(f, null);
+    const l = parseInt(/\d+/.exec(f.text)[0]), c = s.querySelector(
+      "div.amendement-list--results-table"
+    );
+    e.notStrictEqual(c, null);
+    const S = c.querySelector("table");
+    e.notStrictEqual(S, null);
+    const p = S.querySelector("tfoot > tr");
+    e.notStrictEqual(p, null);
+    const g = p.querySelectorAll("td, th").map((T) => T.text);
+    e(
+      Z(g, [
+        "",
+        "n°",
+        "Dossier législatif",
+        "Emplacement",
+        "Auteur",
+        "État",
+        "Sort",
+        "Date d'examen",
+        "Examiné par",
+        "Texte visé",
+        "Date de dépôt"
+      ]),
+      `Unexpected columns in ${JSON.stringify(g, null, 2)}`
+    );
+    const w = S.querySelectorAll("tbody > tr");
+    e.notStrictEqual(w.length, 0);
+    for (const T of w) {
+      if (u++, r != null || m != null) {
+        const b = T.querySelectorAll("td");
+        if (e.strictEqual(
+          b.length,
+          11,
+          `Unexpected number of columns in amendment row: ${T.outerHTML}`
+        ), r != null) {
+          const R = b[10];
+          if (j(
+            k(R.text, "d MMMM y", /* @__PURE__ */ new Date(), {
+              locale: J
+            }),
+            { representation: "date" }
+          ) < r)
+            break e;
+        }
+        if (m != null) {
+          const R = b[7];
+          if (R.text && j(
+            k(R.text, "d MMMM y", /* @__PURE__ */ new Date(), {
+              locale: J
+            }),
+            { representation: "date" }
+          ) < m)
+            continue;
+        }
+      }
+      const h = T.getAttribute("data-href");
+      e.notStrictEqual(h, void 0);
+      const $ = h.split("/");
+      e.strictEqual(
+        $[0],
+        "",
+        `Unexpected URL path for amendement: ${h}`
+      ), e.strictEqual(
+        $[1],
+        "dyn",
+        `Unexpected URL path for amendement: ${h}`
+      );
+      const A = F.join(
+        i,
+        ...$.slice(2)
+      ) + ".html";
+      await q.ensureDir(F.dirname(A));
+      const y = await q.pathExists(
+        A
+      ) ? await q.readFile(A, "utf8") : null, L = T.outerHTML, U = `---
+${W.dump(
+        {
+          position: l - u,
+          search: t.replace(/&page=[\d]+/, "")
+        },
+        {
+          sortKeys: !0
+        }
+      )}---
+
+${L}`;
+      if (U === y) {
+        if (a)
+          break e;
+        continue;
+      }
+      if (y !== null && K(y).body === L)
+        continue;
+      d.verbose && console.log(
+        y === null ? `Adding amendement search cache: ${A}…` : `Updating amendement search cache: ${A}…`
+      );
+      const _ = new URL(
+        h,
+        "https://www.assemblee-nationale.fr/"
+      ).toString(), O = await fetch(_);
+      if (O.status === 404) {
+        console.log(
+          `Amendement HTML page not found at ${_}. Skipping.`
+        );
+        continue;
+      }
+      e(
+        O.ok,
+        `Retrieval of amendement HTML page at ${_} failed with error: ${O.status} ${O.statusText}`
+      );
+      const V = await O.text(), N = X(V).querySelectorAll(
+        "li.mirror-card-header--options--content--item"
+      );
+      e.notStrictEqual(N.length, 0);
+      const I = N.find((b) => {
+        const R = b.querySelector("a > span");
+        return e.notStrictEqual(
+          R,
+          null,
+          `No <span> in <a> found in ${b.toString()}`
+        ), R.text === "Version JSON";
+      });
+      e.notStrictEqual(
+        I,
+        void 0,
+        `No JSON version found for amendement at ${_}`
+      );
+      const C = I.querySelector("a");
+      e.notStrictEqual(
+        C,
+        null,
+        `No <a> found in ${I.toString()}`
+      );
+      const H = C.getAttribute("href");
+      e.notStrictEqual(
+        H,
+        void 0,
+        `No URL found for JSON version of amendement: ${I.toString()}`
+      );
+      const D = new URL(
+        H,
+        "https://www.assemblee-nationale.fr/"
+      ).toString(), M = await fetch(D);
+      e(
+        M.ok,
+        `Retrieval of amendement JSON page at ${D} failed with error: ${M.status} ${M.statusText}`
+      );
+      const B = await M.json();
+      yield [h, B], await q.writeFile(
+        A,
+        U,
+        "utf8"
+      );
+    }
+    const E = s.querySelector(
+      "div.an-pagination--item > i.an-icons-chevron-right ~ a"
+    );
+    if (E === null)
+      break;
+    const x = E.getAttribute("href");
+    e.notStrictEqual(x, void 0), t = new URL(x, "https://www.assemblee-nationale.fr/").toString();
+  }
+}
+export {
+  Se as iterRechercheAmendements,
+  le as iterRechercheRawAmendements,
+  he as parseTexte
+};

package/lib/raw_types/debats.d.ts

@@ -17,7 +17,7 @@ export interface CompteRendu {
 }
 export interface Contenu {
     quantiemes: Quantiemes;
-    ouvertureSeance: Point[] | Point;
+    ouvertureSeance?: Point[] | Point;
     point: Point[] | Point;
     finSeance?: FinSeance;
     paragraphe?: Point[];
@@ -82,7 +82,7 @@ export interface Point {
     adt?: string;
     ssadt?: string;
     roledebat?: string;
-    changementPresidence?: Point;
+    changementPresidence?: Point[] | Point;
     type_debat?: string;
     decision?: Decision;
     ouvertureSeance?: Point;

package/lib/raw_types/dossiers_legislatifs.d.ts

@@ -1,4 +1,4 @@
-import { CodeActe } from "../shared_types/codes_actes";
+import { CodeActe } from '../shared_types/codes_actes';
 export interface Document {
     document: DocumentDocument | DocumentSupprime;
 }