From 1b5ba6032b3dfadcbde4fd49ab36060d663996a7 Mon Sep 17 00:00:00 2001
From: AnaelKremer <170732172+AnaelKremer@users.noreply.github.com>
Date: Thu, 25 Jul 2024 10:04:52 +0200
Subject: [PATCH 1/2] feat: add query conditor for hal cnrs loader

---
 .../loaders/query-conditor-for-halcnrs.ini    | 162 ++++++++++++++++++
 1 file changed, 162 insertions(+)
 create mode 100644 workers/loaders/query-conditor-for-halcnrs.ini

diff --git a/workers/loaders/query-conditor-for-halcnrs.ini b/workers/loaders/query-conditor-for-halcnrs.ini
new file mode 100644
index 000000000..8f0d8d55b
--- /dev/null
+++ b/workers/loaders/query-conditor-for-halcnrs.ini
@@ -0,0 +1,162 @@
+append = pack
+label = query-conditor
+extension = json
+mimeType = application/json
+
+# load some plugins to activate some statements
+[use]
+plugin = conditor
+plugin = basics
+
+# Toggle ezs traces (see server stderr log)
+[debug]
+ezs = false
+
+# {{{
+[TXTConcat]
+
+[replace]
+path = q
+value = self().trim()
+
+[CORHALFetch]
+url = https://corhal-api.inist.fr
+retries = 3
+timeout = 60000
+
+[assign]
+path = uri
+value = get('business.sourceUidChain')
+# }}}
+
+[assign]
+path = typologie
+value = get("business.duplicateGenre")
+
+path = titre
+value = get("title.default")
+
+path = resume
+value = get("abstract.default")
+
+path = auteurs
+value = get("authors").map("fullname")
+
+path = identifiantsAuteurs
+value = get("authors").map(author => _.pick(author, ['orcId', 'idRef', 'idHal','viaf']))
+
+path = affiliations
+value = get("authors").map( author =>_.flatMap(author.affiliations, 'address'))
+
+path = identifiantsAffiliations
+value = get("authors").map("rnsr")
+
+path = domaine
+value = get("classifications")
+
+path = keywords
+value = get("keywords")
+
+path = ppn
+value=get("sourceUids").filter( id => id.startsWith('sudoc')).replace("sudoc-theses$","")
+
+path = nnt
+value = get("fulltextUrl").replace(/^(?!http:\/\/www\.theses\.fr\/).*$/, '').replace("http://www.theses.fr/","").replace("/document","")
+
+[assign]
+path = autresIdentifiants
+value = fix({pmcid: self.pmcId,arxiv : self.arxiv,pmid: self.pmId,nnt: self.nnt,ppn : self.ppn})
+
+path = funders
+value = get("funders").map("fullname")
+
+path = publicationDate
+value = get("host.publicationDate")
+
+path = electronicPublicationDate
+value = get("host.electronicPublicationDate")
+
+path = volume
+value = get("host.volume")
+
+path = issue
+value = get("host.issue")
+
+path = pages
+value = get("host.pages.range")
+
+path = nomConference
+value = get("host.conference.name")
+
+path = dateDebutConference
+value = get("host.conference.date")
+
+path = villeConference
+value = get("host.conference.place")
+
+path = langueDocument
+value = get("host.language")
+
+path = issn
+value = get("host.issn")
+
+path = nomRevue
+value = get("host.title")
+
+path = publisherRevue
+value = get("host.publisher")
+
+path = isbn
+value = get("host.isbn")
+
+path = editors
+value = get("host.editors").map("fullname")
+
+path = sourceUids
+value = get("sourceUids")
+
+path = halID
+value = get("halId")
+
+path = doublonsHal
+value = get("sourceUids").filter(uid => uid.includes('hal')).size().gt(1).replace(true,"Oui").replace(false,"Non")
+
+[assign]
+path=isHal
+value=get("business.sourceUidChain").replace(/\$.*?!/g,"!").split("!").compact().some(item=>(/hal/).test(item)).replace(false,"Non").replace(true,"Oui")
+
+path = fulltextURL
+value = get("fulltextUrl")
+
+# Ensures that each object contains an identification key (required by lodex)
+[swing]
+test = pick(['URI', 'uri']).pickBy(_.identity).isEmpty()
+[swing/identify]
+
+# Ignore objects with duplicate URI
+[dedupe]
+ignore = true
+
+# 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]
+
+[exchange]
+value = omit(["abstract","authors","classifications","business","pii","arxiv","inspire","localRef","pmcId","articleNumber","nnt","ppn","origins","technical","halId","title","originalGenre","pmId","fulltextUrl","enrichments","host"])
+
+# 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')
+path = uri
+value = get('uri').trim()
+

From 2463e69f7ed776654b9bcf01ab5843fe9ce2ca5a Mon Sep 17 00:00:00 2001
From: AnaelKremer <170732172+AnaelKremer@users.noreply.github.com>
Date: Fri, 26 Jul 2024 11:36:58 +0200
Subject: [PATCH 2/2] Update query-conditor-for-halcnrs.ini

---
 workers/loaders/query-conditor-for-halcnrs.ini | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/workers/loaders/query-conditor-for-halcnrs.ini b/workers/loaders/query-conditor-for-halcnrs.ini
index 8f0d8d55b..273181960 100644
--- a/workers/loaders/query-conditor-for-halcnrs.ini
+++ b/workers/loaders/query-conditor-for-halcnrs.ini
@@ -1,5 +1,5 @@
 append = pack
-label = query-conditor
+label = query-conditor-for-halcnrs
 extension = json
 mimeType = application/json