-
Notifications
You must be signed in to change notification settings - Fork 1
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
1 parent
d7cb55a
commit dfe47c3
Showing
22 changed files
with
686 additions
and
5 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,276 @@ | ||
| { | ||
| "cells": [ | ||
| { | ||
| "cell_type": "markdown", | ||
| "id": "45495455", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "Le but de ce notebook est de générer des structures et services pour l'offre de service de l'agefiph.\n", | ||
| "\n", | ||
| "Ce notebook prend 2 sources de données :\n", | ||
| "\n", | ||
| "* une liste de strucures issu d'un tableau grist maintenu par data.inclusion,\n", | ||
| "* une liste de services issu de l'api de l'agefiph.\n" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "id": "517ed275", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "import requests\n", | ||
| "import numpy as np\n", | ||
| "import pandas as pd\n", | ||
| "import trafilatura\n", | ||
| "import hashlib\n", | ||
| "from uuid import UUID\n", | ||
| "\n", | ||
| "\n", | ||
| "def html_to_markdown(s: str):\n", | ||
| " if s is None or s == \"\":\n", | ||
| " return s\n", | ||
| " return trafilatura.extract(trafilatura.load_html(\"<html>\" + s + \"</html>\"))\n", | ||
| "\n", | ||
| "\n", | ||
| "# https://grist.incubateur.net/o/datainclusion/bWqnEafQaLgc/Partage-de-donnes-AGEFIPH-Mars-Avril-2023/p/4\n", | ||
| "STRUCTURES_TABLE_URL = \"https://grist.incubateur.net/o/datainclusion/api/docs/bWqnEafQaLgcTvFv7rv6hF/download/csv?tableId=Structures\"\n", | ||
| "SERVICES_API_URL = \"https://www.agefiph.fr/jsonapi/node/aide_service\"\n", | ||
| "\n", | ||
| "HEADERS = {\"User-Agent\": \"data-inclusion\"}\n" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "id": "5d20e16b", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "raw_structures_df = pd.read_csv(STRUCTURES_TABLE_URL, dtype=str).replace([np.nan, \"\"], None)\n", | ||
| "raw_services_df = pd.json_normalize(\n", | ||
| " requests.get(SERVICES_API_URL, headers=HEADERS).json()[\"data\"]\n", | ||
| ").replace([np.nan, \"\"], None)\n" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "id": "15a0ca22", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "raw_structures_df = raw_structures_df\n", | ||
| "\n", | ||
| "raw_structures_df.info()\n" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "id": "7bd7e018", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "raw_services_df = raw_services_df[\n", | ||
| " [\n", | ||
| " \"id\",\n", | ||
| " \"attributes.created\",\n", | ||
| " \"attributes.changed\",\n", | ||
| " \"attributes.title\",\n", | ||
| " \"attributes.field_titre_card_employeur\",\n", | ||
| " \"attributes.field_essentiel_ph.processed\",\n", | ||
| " \"attributes.field_essentiel_employeur.processed\",\n", | ||
| " \"attributes.field_texte_brut_long\",\n", | ||
| " \"relationships.field_thematique.data\",\n", | ||
| " ]\n", | ||
| "]\n", | ||
| "\n", | ||
| "raw_services_df.info()\n" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "id": "d5ad9896", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "pd.json_normalize(\n", | ||
| " raw_services_df.rename(columns={\"id\": \"service_id\"}).to_dict(orient=\"records\"),\n", | ||
| " record_path=\"relationships.field_thematique.data\",\n", | ||
| " meta=\"service_id\",\n", | ||
| ").info()\n" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "id": "e2ad9881", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "DI_THEMATIQUES_BY_AGEFIPH_THEMATIQUE_ID = {\n", | ||
| " # Source https://grist.incubateur.net/o/datainclusion/uVsB8pabQGoe/Thmatiques/p/13\n", | ||
| " \"4e08047f-b0ed-431a-9182-61e8e61b1486\": \"handicap--favoriser-le-retour-et-le-maintien-dans-lemploi\",\n", | ||
| " \"11618ce3-e59b-404f-8eb2-5763215464f2\": \"handicap--favoriser-le-retour-et-le-maintien-dans-lemploi\",\n", | ||
| " \"60c25ci7-61sc-89a9-ny54-126hslf808a2\": \"handicap--connaissance-des-droits-des-travailleurs\",\n", | ||
| " \"51be0003-13d8-4ffa-9923-248e7aa4a227\": None,\n", | ||
| " \"ddf0fa87-2ee0-481c-a258-96985b7826c3\": None,\n", | ||
| " \"cb2c9fec-c190-4e2f-aeee-6da818109bf8\": \"handicap--favoriser-le-retour-et-le-maintien-dans-lemploi\",\n", | ||
| " \"78b28acb-803e-4b06-ab77-58dabfbd8571\": \"handicap--adaptation-au-poste-de-travail\",\n", | ||
| " \"366eb399-1e6c-4609-8066-d1504fae2a8e\": None,\n", | ||
| " \"907a8c33-5c56-49d3-bd64-a736a9ceac76\": None,\n", | ||
| " \"5d8c88d8-db03-4f27-b517-d7016896b01a\": None,\n", | ||
| " \"fb5e6180-290b-4216-ba68-624d25defa3a\": \"handicap--favoriser-le-retour-et-le-maintien-dans-lemploi\",\n", | ||
| " \"03228d62-2a59-49d8-8443-b25cb2e684b9\": \"accompagnement-social-et-professionnel-personnalise--definition-du-projet-professionnel\",\n", | ||
| " \"f9ab3e06-af51-463a-aaf7-7b04a28e047f\": \"se-former--trouver-sa-formation\",\n", | ||
| " \"aeab1d68-4e89-4e2a-a612-d8645e3999d8\": \"creation-activite--definir-son-projet-de-creation-dentreprise\",\n", | ||
| " \"f4551558-8315-4708-8357-5ecc89751bc6\": \"handicap--faire-reconnaitre-un-handicap\",\n", | ||
| " \"4b8b0473-52c2-4a21-956d-d7d68a7053b5\": None,\n", | ||
| "}\n", | ||
| "\n", | ||
| "\n", | ||
| "def map_service(row) -> dict:\n", | ||
| " service = {}\n", | ||
| " service[\"id\"] = row[\"id\"]\n", | ||
| " service[\"date_creation\"] = row[\"attributes.created\"]\n", | ||
| " service[\"date_maj\"] = row[\"attributes.changed\"]\n", | ||
| " service[\"nom\"] = row[\"attributes.title\"]\n", | ||
| " service[\"contact_public\"] = True\n", | ||
| " service[\"presentation_resume\"] = row[\"attributes.field_titre_card_employeur\"]\n", | ||
| "\n", | ||
| " service[\"presentation_detail\"] = \"\"\n", | ||
| " if row[\"attributes.field_essentiel_ph.processed\"] is not None:\n", | ||
| " service[\"presentation_detail\"] += (\n", | ||
| " \"<p>Pour la personne handicapée :</p>\"\n", | ||
| " + row[\"attributes.field_essentiel_ph.processed\"]\n", | ||
| " )\n", | ||
| " if row[\"attributes.field_essentiel_employeur.processed\"] is not None:\n", | ||
| " service[\"presentation_detail\"] += (\n", | ||
| " \"<p>Pour l'employeur :</p>\"\n", | ||
| " + row[\"attributes.field_essentiel_employeur.processed\"]\n", | ||
| " )\n", | ||
| " if row[\"attributes.field_texte_brut_long\"] is not None:\n", | ||
| " service[\"presentation_detail\"] = (\n", | ||
| " row[\"attributes.field_texte_brut_long\"] + service[\"presentation_detail\"]\n", | ||
| " )\n", | ||
| " service[\"presentation_detail\"] = html_to_markdown(service[\"presentation_detail\"])\n", | ||
| " service[\"presentation_detail\"] = service[\"presentation_detail\"] or None\n", | ||
| "\n", | ||
| " service[\"thematiques\"] = list(\n", | ||
| " set(\n", | ||
| " [\n", | ||
| " v\n", | ||
| " for v in [\n", | ||
| " DI_THEMATIQUES_BY_AGEFIPH_THEMATIQUE_ID[\n", | ||
| " agefiph_thematique_data[\"id\"]\n", | ||
| " ]\n", | ||
| " for agefiph_thematique_data in row[\n", | ||
| " \"relationships.field_thematique.data\"\n", | ||
| " ]\n", | ||
| " ]\n", | ||
| " if v is not None\n", | ||
| " ]\n", | ||
| " )\n", | ||
| " )\n", | ||
| "\n", | ||
| " return service\n" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "id": "9f98dd4d", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "template_services_df = raw_services_df.apply(map_service, axis=1, result_type=\"expand\")\n", | ||
| "\n", | ||
| "template_services_df.info()\n" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "id": "e0680158", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "# Cartesian products\n", | ||
| "services_df = (\n", | ||
| " raw_structures_df[\n", | ||
| " [\n", | ||
| " \"id\",\n", | ||
| " \"courriel\",\n", | ||
| " \"telephone\",\n", | ||
| " \"adresse\",\n", | ||
| " \"commune\",\n", | ||
| " \"code_postal\",\n", | ||
| " \"code_insee\",\n", | ||
| " ]\n", | ||
| " ]\n", | ||
| " .rename(columns={\"id\": \"structure_id\"})\n", | ||
| " .join(template_services_df, how=\"cross\")\n", | ||
| ")\n", | ||
| "\n", | ||
| "# Making service id unique across all regions\n", | ||
| "services_df = services_df.assign(\n", | ||
| " id=services_df.apply(\n", | ||
| " lambda row: str(\n", | ||
| " UUID(\n", | ||
| " hex=hashlib.md5((row[\"structure_id\"] + row[\"id\"]).encode()).hexdigest()\n", | ||
| " )\n", | ||
| " ),\n", | ||
| " axis=1,\n", | ||
| " )\n", | ||
| ")\n", | ||
| "\n", | ||
| "services_df.info()\n" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "id": "ecfc54c4", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "services_df.to_json(\"services.json\", orient=\"records\", force_ascii=False)\n", | ||
| "raw_structures_df.to_json(\"structures.json\", orient=\"records\", force_ascii=False)\n" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": null, | ||
| "id": "85c609c1", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "services_df\n" | ||
| ] | ||
| } | ||
| ], | ||
| "metadata": { | ||
| "kernelspec": { | ||
| "display_name": "Python 3 (ipykernel)", | ||
| "language": "python", | ||
| "name": "python3" | ||
| }, | ||
| "language_info": { | ||
| "codemirror_mode": { | ||
| "name": "ipython", | ||
| "version": 3 | ||
| }, | ||
| "file_extension": ".py", | ||
| "mimetype": "text/x-python", | ||
| "name": "python", | ||
| "nbconvert_exporter": "python", | ||
| "pygments_lexer": "ipython3", | ||
| "version": "3.10.11" | ||
| } | ||
| }, | ||
| "nbformat": 4, | ||
| "nbformat_minor": 5 | ||
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.