Update Inspire Loader

workers/loaders/json-inspirehep.ini

@@ -1,94 +1,77 @@
-#loader json-inspirehep.ini (version novembre 2022 - Inist-CNRS)
-
-append = pack
-label = json-inspirehep
-
-# load some plugins to activate some statements
-[use]
-plugin = basics
-
-# Toggle ezs traces (see server stderr log)
-[debug]
-ezs = false
-
-[JSONParse]
-separator = hits.hits.*
-
-[replace]
-path = id
-value = get('id')
-
-path = Titre
-value = get('metadata.titles[0].title')
-
-path = Type de document
-value = get('metadata.document_type')
-
-path = Année de publication
-value = get('metadata.publication_info[0].year')
-
-path = Revue
-value = get('metadata.publication_info[0].journal_title')
-
-path = Issue
-value = get('metadata.publication_info[0].journal_issue')
-
-path = N° de volume
-value = get('metadata.publication_info[0].journal_volume')
-
-path = N° conférence
-value = get('metadata.publication_info[0].cnum')
-
-path = Auteurs
-value = get('metadata.authors').map(author => ({ full_name: author.full_name, affiliations: author.affiliations ? author.affiliations.map(aff => ({ label: aff.value || "", url: ( aff.record ? aff.record["$ref"] : "q" ) }) ) : [] } ))
-
-path = Catégorie inspire
-value = get('metadata.inspire_categories').map(cat => cat.term).uniq()
-
-path = Expériences
-value = get('metadata.accelerator_experiments').map(equ => equ.legacy_name).uniq()
-
-path = Collaborations
-value = get('metadata.collaborations').map(col => col.value).uniq()
-
-path = Résumé
-value = get('metadata.abstracts[0].value')
-
-path = DOI
-value = get('metadata.dois').map(doi => doi.value).uniq()
-
-[assign]
-path = UrlsLabo
-value = get('Auteurs').map(author => author.affiliations.map(aff => aff.url)).flatten().uniq()
-
-[assign]
-path = codesLabos
-value = get("UrlsLabo").map(url => String(url).split("/").slice(-1)[0])
-
-[assign]
-path = uri
-value = get('id')
-
-# Ensures that each object contains an identification key (required by lodex)
-[swing]
-test = pick(['URI', 'uri']).pickBy(_.identity).isEmpty()
-[swing/identify]
-
-# Prevent keys from containing dot path notation (which is forbidden by nodejs mongoDB driver)
-[OBJFlatten]
-separator = fix('.')
-reverse = true
-safe = true
-
-# Uncomment to see each data sent to the database
-#[debug]
-
-# Add contextual metadata related to the import
-[assign]
-path = lodexStamp.importedDate
-value = fix(new Date()).thru(d => d.toDateString())
-path = lodexStamp.usedParser
-value = env('parser')
-path = lodexStamp.uploadedFilename
-value = env('source')
-
+#loader json-inspirehep.ini (version juin 2023 - Inist-CNRS)
+
+append = dump
+label = json-inspirehep
+
+[use]
+plugin = basics
+plugin = lodex
+
+[JSONParse]
+separator = hits.hits.*
+
+[replace]
+path = id
+value = get('id')
+
+path = Titre
+value = get('metadata.titles[0].title')
+#1e occurrence
+
+path = Type de document
+value = get('metadata.document_type')
+
+path = Année de publication
+value = get('metadata.publication_info[0].year')
+#1e occurrence
+
+path = Revue
+value = get('metadata.publication_info[0].journal_title')
+#1e occurrence
+
+path = Issue
+value = get('metadata.publication_info[0].journal_issue')
+#1e occurrence
+
+path = N° de volume
+value = get('metadata.publication_info[0].journal_volume')
+#1e occurrence
+
+path = N° conférence
+value = get('metadata.publication_info[0].cnum')
+#1e occurrence
+
+path = Auteurs
+value = get('metadata.authors').map(author => ({ full_name: author.full_name, affiliations: author.affiliations ? author.affiliations.map(aff => ({ label: aff.value || "", url: ( aff.record ? aff.record["$ref"] : "empty" ) }) ) : [] } ))
+
+path = Catégorie inspire
+value = get('metadata.inspire_categories').map(cat => cat.term).uniq()
+
+path = Expériences
+value = get('metadata.accelerator_experiments').map(equ => equ.legacy_name).uniq()
+
+path = Collaborations
+value = get('metadata.collaborations').map(col => col.value).uniq()
+
+path = Résumé
+value = get('metadata.abstracts[0].value')
+#1e occurrence
+
+path = DOI
+value = get('metadata.dois').map(doi => doi.value).uniq()
+
+[assign]
+path = UrlsLabo
+value = get('Auteurs').map(author => author.affiliations.filter(aff => aff.url!=="empty").map(aff => aff.url)).flatten().uniq()
+
+[assign]
+path = codesLabos
+value = get("UrlsLabo").map(url => url.split("/").slice(-1)[0])
+
+[OBJFlatten]
+separator = /
+
+[assign]
+path = uri
+value = get('id')
+#constitution identifiant URI LOdex à partir de l'ID