|
| 1 | +{ |
| 2 | + "translatorID": "86c090c5-ad3f-4c54-a1f3-9870942014ac", |
| 3 | + "label": "Library of Congress", |
| 4 | + "creator": "Sebastian Karcher", |
| 5 | + "target": "^https?://search\\.catalog\\.loc\\.gov/", |
| 6 | + "minVersion": "5.0", |
| 7 | + "maxVersion": "", |
| 8 | + "priority": 100, |
| 9 | + "inRepository": true, |
| 10 | + "translatorType": 4, |
| 11 | + "browserSupport": "gcsibv", |
| 12 | + "lastUpdated": "2025-08-23 18:51:50" |
| 13 | +} |
| 14 | + |
| 15 | +/* |
| 16 | + ***** BEGIN LICENSE BLOCK ***** |
| 17 | +
|
| 18 | + Copyright © 2025 Sebastian Karcher |
| 19 | +
|
| 20 | + This file is part of Zotero. |
| 21 | +
|
| 22 | + Zotero is free software: you can redistribute it and/or modify |
| 23 | + it under the terms of the GNU Affero General Public License as published by |
| 24 | + the Free Software Foundation, either version 3 of the License, or |
| 25 | + (at your option) any later version. |
| 26 | +
|
| 27 | + Zotero is distributed in the hope that it will be useful, |
| 28 | + but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 29 | + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 30 | + GNU Affero General Public License for more details. |
| 31 | +
|
| 32 | + You should have received a copy of the GNU Affero General Public License |
| 33 | + along with Zotero. If not, see <http://www.gnu.org/licenses/>. |
| 34 | +
|
| 35 | + ***** END LICENSE BLOCK ***** |
| 36 | +*/ |
| 37 | + |
| 38 | + |
| 39 | +function detectWeb(doc, url) { |
| 40 | + if (url.includes('/instances/')) { |
| 41 | + let type = text(doc, 'div[class*="__metadata-type"]') |
| 42 | + Z.debug(type) |
| 43 | + switch (type) { |
| 44 | + case "Book": |
| 45 | + return "book"; |
| 46 | + case "Cartographic Material": |
| 47 | + return "map"; |
| 48 | + case "Audio": |
| 49 | + return "audioRecording"; |
| 50 | + default: |
| 51 | + return "book"; |
| 52 | + } |
| 53 | + } |
| 54 | + else if (getSearchResults(doc, true)) { |
| 55 | + return 'multiple'; |
| 56 | + } |
| 57 | + return false; |
| 58 | +} |
| 59 | + |
| 60 | +function getSearchResults(doc, checkOnly) { |
| 61 | + var items = {}; |
| 62 | + var found = false; |
| 63 | + var rows = doc.querySelectorAll('h2 > a[href*="/instances/"]'); |
| 64 | + for (let row of rows) { |
| 65 | + let href = row.href; |
| 66 | + let title = ZU.trimInternal(row.textContent); |
| 67 | + if (!href || !title) continue; |
| 68 | + if (checkOnly) return true; |
| 69 | + found = true; |
| 70 | + items[href] = title; |
| 71 | + } |
| 72 | + return found ? items : false; |
| 73 | +} |
| 74 | + |
| 75 | + |
| 76 | +function getLCCN(doc){ |
| 77 | + let lccn = text(doc, '[data-auto="record-lccn-content"]'); |
| 78 | + return lccn; |
| 79 | +} |
| 80 | + |
| 81 | +async function doWeb(doc, url) { |
| 82 | + if (detectWeb(doc, url) == 'multiple') { |
| 83 | + let items = await Zotero.selectItems(getSearchResults(doc, false)); |
| 84 | + if (!items) return; |
| 85 | + for (let url of Object.keys(items)) { |
| 86 | + // There's an internal API, but I'm getting consistent 401 responses |
| 87 | + // url = url.replace("/instances/", "/api/opac-inventory/instances/").replace(/\?.*/, ""); |
| 88 | + // Z.debug(url) |
| 89 | + await scrape(await requestDocument(url)); |
| 90 | + } |
| 91 | + } |
| 92 | + else { |
| 93 | + let lccn = getLCCN(doc); |
| 94 | + let marcxmlurl = "https://lccn.loc.gov/" + lccn + "/marcxml" |
| 95 | + await scrapeMARC(marcxmlurl); |
| 96 | + } |
| 97 | +} |
| 98 | + |
| 99 | +async function scrapeMARC(marcxmlurl) { |
| 100 | + let marcxml = await requestText(marcxmlurl); |
| 101 | + Z.debug(marcxml) |
| 102 | + let translate = Zotero.loadTranslator('import'); |
| 103 | + translate.setTranslator('edd87d07-9194-42f8-b2ad-997c4c7deefd'); |
| 104 | + translate.setString(marcxml); |
| 105 | + translate.setHandler('itemDone', (_, item) => {}); |
| 106 | + [item] = await translate.translate(); |
| 107 | + // item.itemType = detectWeb(doc, url); |
| 108 | + item.complete(); |
| 109 | +} |
| 110 | + |
| 111 | +async function scrape(doc) { |
| 112 | + let lccn = getLCCN(doc); |
| 113 | + let marcxmlurl = "https://lccn.loc.gov/" + lccn + "/marcxml" |
| 114 | + await scrapeMARC(marcxmlurl); |
| 115 | +} |
| 116 | + |
| 117 | +/** BEGIN TEST CASES **/ |
| 118 | +var testCases = [ |
| 119 | + { |
| 120 | + "type": "web", |
| 121 | + "url": "https://search.catalog.loc.gov/instances/5f9c256e-151d-5615-ad15-3d2749318cac?option=keyword&query=karcher", |
| 122 | + "defer": true, |
| 123 | + "items": [ |
| 124 | + { |
| 125 | + "itemType": "book", |
| 126 | + "title": "Praxis des Beton- und Stahlbetonbaus: Wissensgrundlagen für die Baustelle und das Ingenieurbüro", |
| 127 | + "creators": [ |
| 128 | + { |
| 129 | + "firstName": "Gustav", |
| 130 | + "lastName": "Kärcher", |
| 131 | + "creatorType": "author" |
| 132 | + } |
| 133 | + ], |
| 134 | + "date": "1952", |
| 135 | + "callNumber": "TA681 .K25", |
| 136 | + "libraryCatalog": "Library of Congress", |
| 137 | + "numPages": "200", |
| 138 | + "place": "Stuttgart", |
| 139 | + "publisher": "Franckh", |
| 140 | + "shortTitle": "Praxis des Beton- und Stahlbetonbaus", |
| 141 | + "attachments": [], |
| 142 | + "tags": [ |
| 143 | + { |
| 144 | + "tag": "Concrete construction" |
| 145 | + }, |
| 146 | + { |
| 147 | + "tag": "Reinforced concrete construction" |
| 148 | + } |
| 149 | + ], |
| 150 | + "notes": [], |
| 151 | + "seeAlso": [] |
| 152 | + } |
| 153 | + ] |
| 154 | + } |
| 155 | +] |
| 156 | +/** END TEST CASES **/ |
0 commit comments