diff --git a/doiwos-dumps-config.json b/doiwos-dumps-config.json
new file mode 100644
index 0000000..3fc9e72
--- /dev/null
+++ b/doiwos-dumps-config.json
@@ -0,0 +1,23 @@
+{
+  "files": {
+    "zip": "https://gitbucket.inist.fr/parmentf/web-dumps/archive/doiswos-dumps/doswos-dumps@1.0.0.zip"
+  },
+  "environnement": {
+    "CRON_VERBOSE": true,
+    "EZS_VERBOSE": false,
+    "NODE_OPTIONS": "--max_old_space_size=1024",
+    "WOS_API_KEY": "PUT HERE A VALID WOS API KEY"
+  },
+  "packages": [
+    "@ezs/core@2.1.0",
+    "@ezs/basics@1.22.3",
+    "@ezs/analytics@2.0.2"
+  ],
+  "tasks": [
+    {
+      "CronRule": "0 1 * * *",
+      "Target": "data/09-corpus-simple-cnrs.json",
+      "RunOnStartup": true
+    }
+  ]
+}
\ No newline at end of file
diff --git a/doiwos-dumps/00-dois.txt b/doiwos-dumps/00-dois.txt
new file mode 100644
index 0000000..7f1b201
--- /dev/null
+++ b/doiwos-dumps/00-dois.txt
@@ -0,0 +1,100 @@
+10.1332/0305573012501251
+10.1111/j.1540-6210.2011.02488.x
+https://doi.org/10.1016/j.gloenvcha.2015.01.004
+10.1038/ngeo1391
+10.1007/s10980-016-0365-y
+10.5194/gmd-10-3979-2017
+https://doi.org/10.1007/s11205-017-1650-0
+10.1111/j.1466- 8238.2010.00540.x
+10.1038/s41558-018-0156-3
+https://doi.org/10.3390/su4040656
+10.1257/jel.38.3.595
+10.1126/science.1234485
+10.1038/534320a
+https://doi.org/10.1016/j.envsci.2015.11.011
+10.1080/09640568.2016.1168288
+https://doi.org/10.1002/wcc.485
+10.1016/j.gloenvcha.2016.05.009
+10.1088/1748- 9326/9/6/064030
+10.1029/2011GL050087
+10.1175/BAMS-D-17-0157.1
+10.5194/gmd-10-4035-2017
+https://doi.org/10.1038/s41893-018-0021-4
+10.1007/s10712-019-09525-z
+10.1029/2007GC001784
+10.1093/ajcn/84.1.289
+https://doi.org/10.1016/j.cosust.2014.11.002
+10.1111/area.12189
+10.1007/s10113-018-1288-8
+10.1111/rec.12359
+10.1126/science.aap8826
+10.1007/s10584-013-0905-2
+http://dx.doi.org/10.1146/annurev.psych.52.1.141
+10.1210/jc.2008-1595
+https://doi.org/10.1186/s13021-018-0095-3
+10.1007/s10584-018-2317-9
+10.1038/nature08823
+10.1890/070062
+10.1016/j. landurbplan.2014.10.018
+10.1111/tgis.12265
+10.1111/psj.12212
+10.1016/j.ecolecon.2006.09.026
+doi:10.1073/pnas.0906974107
+10.1016/0967-0637(95)00012-U
+doi:10.1093/biosci/biu225
+10.1016/j.gloenvcha.2011.10.003
+10.1111/gcb.13988
+10.1111/j.1745- 6584.1998.tb02825.x.
+10.1023/A:1005582929198
+http://www.jstor.org/stable/27800489
+10.1007/s00035-011-0094-4
+10.5194/cp-12-663-2016
+10.17645/pag.v6i1.1169
+10.1007/s10584-011-0148-z
+10.1659/MRD-JOURNAL-D-17-00107.1
+10.1111/1365-2664.12643
+10.1371/journal.pone.0068037
+papers3://publication/doi/10.1016/j.foreco.2017.08.016
+10.5194/esd-9-1085-2018
+10.1016/j.palaeo.2011.04.005
+10.1007/s40808-018-0517-y
+10.1038/s41598-019-42792-9
+10.1126/science.aac7125
+10.1038/s41598-019-57280-3
+10.1111/ddi.12556
+10.1038/nature17155 http://www.nature.com/nature/journal/v531/n7594/abs/nature17155.html#supplementary-information
+http://dx.doi.org/10.1016/j.palaeo.2006.11.050
+10.15242/IICBE.C0415015
+10.1126/science.aac8083
+10.1007/s10531-016-1071-4
+10.1007/s10531-019-01827-3
+10.1038/s41467-017-00923-8
+10.1126/science.1081056
+10.1007/s00338-021-02170-2
+10.1098/rspb.2015.1211
+papers3://publication/doi/10.3390/f9100596
+papers3://publication/doi/10.1007/s40641-015-0002-x
+10.3390/sci2030067
+10.1111/gcb.12754
+10.1073/pnas.1621517114
+10.1098/rspb.2011.1897
+papers3://publication/doi/10.1002/eco.1997
+10.1371/journal.pone.0188142
+10.1146/annurev-environ-110615-085610
+10.1073/pnas.0804478105
+https://doi.org/10.1016/j.cub.2012.09.036
+http://dx.doi.org/10.1016/j.palaeo.2016.01.033
+doi.org/10.1111/2041-210X.13675
+10.3389/fmars.2019.00499
+10.1016/j.tree.2006.09.010
+doi:10.1111/j.1365-2699.2006.01482.x
+doi:10.1111/j.1466-822X.2005.00182.x
+https://doi.org/10.1016/j.palaeo.2009.10.001
+10.1038/ngeo905 https://www.nature.com/articles/ngeo905#supplementary-information
+10.1671/1952-4
+10.1029/2008gl035028
+http://dx.doi.org/10.1016/j.chemgeo.2013.10.012
+10.1002/ece3.2196
+10.1016/j.biocon.2016.10.007
+10.1016/j.biocon.2018.03.037
+10.1038/srep26513
diff --git a/doiwos-dumps/01-harvest-from-doi.ini b/doiwos-dumps/01-harvest-from-doi.ini
new file mode 100644
index 0000000..b3049d9
--- /dev/null
+++ b/doiwos-dumps/01-harvest-from-doi.ini
@@ -0,0 +1,85 @@
+; Usage: npx ezs harvest-from-doi.ini < dois.txt
+
+; Inspiré de https://gitbucket.inist.fr/tdm/web-services/blob/master/biblio-tools/v1/wos/works/expand.ini
+
+[use]
+plugin = @ezs/basics
+plugin = @ezs/analytics
+
+# Flow configuration
+[TXTParse]
+
+[env]
+; Maximum 90 (théoriquement 100, mais un DOI peut renvoyer plusieurs notices)
+; Parfois, on rencontre une erreur "Request Header Fields Too Large", sans doute
+; due à une URL trop longue (avec trop de DOI trop longs).
+path = size
+value = 50
+
+path = indent
+value = true
+
+path = token
+value = env('WOS_API_KEY')
+
+# Clean up DOIs
+# - lower case
+# - remove BOM
+# - from CRLF to LF
+# - remove - at then of a DOI
+# - replace middle point with normal point
+# - replace en dash with normal hyphen-minus
+# - remove soft hyphen (discretionary hyphen)
+# - remove "http" part of a DOI
+# - remove "doi:" part of a DOI
+# - remove URL following a DOI
+# - remove duplicates of a DOI
+# - remove PMID following DOI
+# - remove part preceding a DOI
+# - remove blanks
+# - remove ;subjmeta part of a DOI
+# - remove double quotes from DOIs
+[replace]
+path = value
+value = toLower().replace(/^\uFEFF/, "").replace(/\r/,"").replace(/-$/, "").replace(/\u00B7/g, ".").replace(/\u2013/g, "-").replace(/\u00AD/g, "").replace(/^.*doi(\.org)?\/(.+)/g, "$2").replace(/^.*doi[: ]+10\./g, "10.").replace(/^(10\.\d+\/.+\S) https?:\/\/.+/, "$1").replace(/^(10\.\d+\/\S+) \1.*/, "$1").replace(/^(10\.\d+\/\S+) pmid.+/, "$1").replace(/^.* (10\.\d+\/.+)/, "$1").replace(/ /g, "").replace(/;subjmeta=\d+$/,"").replace(/"+/g, "")
+
+[group]
+size = env('size')
+
+[replace]
+path = dois
+value = self().map('value').filter(Boolean).map(x => JSON.stringify(x)).join(' OR ')
+
+[replace]
+path = usrQuery
+value = fix('DO=(', self.dois ,')').join('')
+
+path = databaseId
+value = WOK
+
+path = count
+value = env('size')
+
+path = firstRecord
+value = 1
+
+; FR = Full Record
+path = optionView
+value = FR
+
+# Maximum 1 request by second
+[throttle]
+bySecond = 1
+
+; [debug]
+
+[URLStream]
+url = https://wos-api.clarivate.com/api/wos/
+header = env('token').prepend('X-ApiKey:')
+path = Data.Records.records.REC.*
+timeout = 50000
+noerror = true
+retries = 1
+
+[dump]
+indent = env('indent')
diff --git a/doiwos-dumps/02-extract-fields.ini b/doiwos-dumps/02-extract-fields.ini
new file mode 100644
index 0000000..f9f5087
--- /dev/null
+++ b/doiwos-dumps/02-extract-fields.ini
@@ -0,0 +1,55 @@
+# Extract few fields from a WoS file.
+# Usage:
+# npx ezs extract-fields.ini < corpus_WoS_vol1.json | jq
+
+[use]
+plugin = basics
+
+[JSONParse]
+
+[assign]
+path = tmp.reprint_addresses
+value = get('static_data.fullrecord_metadata.reprint_addresses.address_name', []).castArray().map(name => name.address_spec.full_address)
+
+path = tmp.addresses
+value = get('static_data.fullrecord_metadata.addresses.address_name', []).castArray().map(name => name.address_spec.full_address)
+
+[replace]
+path = uri
+value = get("UID")
+
+path = doi
+value = get("dynamic_data.cluster_related.identifiers.identifier").filter(i => i.type === "doi").map("value").uniq()
+
+path = title
+value = get('static_data.summary.titles.title').filter(title => title.type === "item").map(title => title.content).join('|')
+
+path = abstract
+value = get('static_data.fullrecord_metadata.abstracts.abstract.abstract_text.p', []).castArray().map(s => String(s).replace(/<[^>]*>/g, "").replace("Key Points", "")).join(" ; ")
+
+path = publication_year
+value = get('static_data.summary.pub_info.pubyear')
+
+path = source
+value = get('static_data.summary.titles.title').filter(title => title.type === "source").map(title => title.content).join('|')
+
+path = affiliations
+value = get('tmp.addresses', []).concat(_.get(self, 'tmp.reprint_addresses')).remove(null).uniq()
+
+path = countries
+value = get('static_data.fullrecord_metadata.addresses.address_name', []).castArray().map("address_spec.country", []).uniq()
+
+path = keywords
+value = get('static_data.item.keywords_plus.keyword')
+
+path = subjects
+value = get('static_data.fullrecord_metadata.category_info.subjects.subject').map('content').uniq()
+
+path = subheadings
+value = get('static_data.fullrecord_metadata.category_info.subheadings.subheading').castArray().uniq()
+
+path = headings
+value = get('static_data.fullrecord_metadata.category_info.headings.heading').castArray().uniq()
+
+[dump]
+indent = true
diff --git a/doiwos-dumps/03-enrich-rnsr.ini b/doiwos-dumps/03-enrich-rnsr.ini
new file mode 100644
index 0000000..90a896d
--- /dev/null
+++ b/doiwos-dumps/03-enrich-rnsr.ini
@@ -0,0 +1,33 @@
+# Usage: npx ezs enrich-rnsr.ini < data/corpus-simple.json > data/corpus-simple-rnsr.json
+
+# To see the enriched notices:
+# $ fx < data/corpus-simple-rnsr.json '.filter(o => o.ws.rnsr[0]?.length)'
+
+[use]
+# JSONParse URLConnect
+plugin = basics
+# aggregate exploding
+plugin = analytics
+
+[JSONParse]
+
+[assign]
+path = ws.rnsr
+value = get('affiliations').map(address => ({ year: _.get(self, 'publication_year'), address}))
+
+[expand]
+path = ws.rnsr
+size = 100
+
+[expand/exploding]
+
+[expand/URLConnect]
+url = https://affiliations-tools.services.inist.fr/v1/rnsr/info
+retries = 3
+timeout = 6000000
+noerror = true
+
+[expand/aggregate]
+
+[dump]
+indent = true
diff --git a/doiwos-dumps/04-enrich-etab.ini b/doiwos-dumps/04-enrich-etab.ini
new file mode 100644
index 0000000..be0062a
--- /dev/null
+++ b/doiwos-dumps/04-enrich-etab.ini
@@ -0,0 +1,19 @@
+# Usage: npx ezs enrich-etab.ini < data/corpus-simple-rnsr.json > data/corpus-simple-etab.json
+
+# To see the enriched notices:
+# $ fx < data/corpus-simple-etab.json '.filter(o => o.ws.rnsr[0]?.length)'
+
+
+[use]
+# JSONParse
+plugin = basics
+
+[JSONParse]
+
+# Concaténation des sigles et libellés des établissements associés
+[assign]
+path = ws.etab
+value = get('ws.rnsr', []).map(structs => Array.isArray(structs) ? structs.map(struct => `${struct.sigle}: ${struct.intitule}`) : []).flatten().uniq()
+
+[dump]
+indent = true
diff --git a/doiwos-dumps/05-enrich-institutes.ini b/doiwos-dumps/05-enrich-institutes.ini
new file mode 100644
index 0000000..dd04c73
--- /dev/null
+++ b/doiwos-dumps/05-enrich-institutes.ini
@@ -0,0 +1,48 @@
+# Usage: npx ezs enrich-institutes.ini < data/corpus-simple-etab.json > data/corpus-simple-instituts.json
+
+# To see the enriched notices:
+# $ fx < data/corpus-simple-instituts.json '.filter(o => o.ws.rnsr[0]?.length)'
+
+
+[use]
+# JSONParse URLConnect
+plugin = basics
+# expand aggregate exploding
+plugin = analytics
+
+[JSONParse]
+
+# - Instituts du CNRS
+
+[assign]
+path = ws.rnsr_id
+value = get("ws.rnsr").map(structs => Array.isArray(structs) ? structs.map(struct => struct.num_nat_struct) : "n/a").filter((id) => id !== "n/a").flatten().uniq()
+
+[assign]
+path = ws.instituts
+value = get("ws.rnsr_id")
+
+[expand]
+path = ws.instituts
+size = 1
+
+[expand/exploding]
+
+[expand/expand]
+path = value
+size = 100
+
+[expand/expand/URLConnect]
+url = https://mapping-tools.services.inist.fr/v1/rnsr/instituts-cnrs/json
+retries = 3
+timeout = 6000000
+noerror = true
+
+[expand/aggregate]
+
+[assign]
+path = ws.instituts
+value = get('ws.instituts').uniq()
+
+[dump]
+indent = true
diff --git a/doiwos-dumps/06-enrich-teeft-en.ini b/doiwos-dumps/06-enrich-teeft-en.ini
new file mode 100644
index 0000000..4e88690
--- /dev/null
+++ b/doiwos-dumps/06-enrich-teeft-en.ini
@@ -0,0 +1,33 @@
+# Usage: npx ezs enrich-teeft-en.ini < data/corpus-simple-instituts.json > data/corpus-simple-teeft-en.json
+
+# To see the enriched notices:
+# $ fx < data/corpus-simple-teeft-en.json '.filter(o => o.ws.rnsr[0]?.length)'
+
+
+[use]
+# JSONParse URLConnect
+plugin = basics
+# expand
+plugin = analytics
+
+[JSONParse]
+
+# - Teeft
+
+
+[assign]
+path = ws.teeft
+value = fix(_.get(self, "abstract", ""), _.get(self, "title")).join(" ; ")
+
+[expand]
+path = ws.teeft
+size = 100
+
+[expand/URLConnect]
+url = https://terms-extraction.services.inist.fr/v1/teeft/en?nb=10
+retries = 3
+timeout = 6000000
+noerror = true
+
+[dump]
+indent = true
diff --git a/doiwos-dumps/07-enrich-pascal.ini b/doiwos-dumps/07-enrich-pascal.ini
new file mode 100644
index 0000000..1fa07ff
--- /dev/null
+++ b/doiwos-dumps/07-enrich-pascal.ini
@@ -0,0 +1,34 @@
+# Usage: npx ezs enrich-pascal.ini < data/corpus-simple-teeft-en.json > data/corpus-simple-pascal.json
+
+
+# To see the enriched notices:
+# $ fx < data/corpus-simple-pascal.json '.filter(o => o.ws.rnsr[0]?.length)'
+
+
+[use]
+# JSONParse URLConnect
+plugin = basics
+# expand
+plugin = analytics
+
+[JSONParse]
+
+# - Pascal
+
+
+[assign]
+path = ws.pascal
+value = fix(_.get(self, "abstract", ""), _.get(self, "title")).join(" ; ")
+
+[expand]
+path = ws.pascal
+size = 100
+
+[expand/URLConnect]
+url = https://domains-classifier.services.inist.fr/v1/en/classify
+retries = 3
+timeout = 6000000
+noerror = true
+
+[dump]
+indent = true
diff --git a/doiwos-dumps/08-extract-pascal-label.ini b/doiwos-dumps/08-extract-pascal-label.ini
new file mode 100644
index 0000000..d33160b
--- /dev/null
+++ b/doiwos-dumps/08-extract-pascal-label.ini
@@ -0,0 +1,22 @@
+# Usage: npx ezs extract-pascal-label.ini < data/corpus-simple-pascal.json > data/corpus-simple-label-pascal.json
+
+
+# To see the enriched notices:
+# $ fx < data/corpus-simple-label-pascal.json '.filter(o => o.ws.rnsr[0]?.length)'
+
+
+[use]
+# JSONParse
+plugin = basics
+
+[JSONParse]
+
+# Action!
+
+[assign]
+path = ws.pascal
+value = fix(_.get(self, "ws.pascal.1.code.value", "").replace("BGI / Prodig", "Géographie"), _.get(self, "ws.pascal.2.code.value")).join(" - ")
+
+
+[dump]
+indent = true
diff --git a/doiwos-dumps/09-detect-cnrs.ini b/doiwos-dumps/09-detect-cnrs.ini
new file mode 100644
index 0000000..8a1252b
--- /dev/null
+++ b/doiwos-dumps/09-detect-cnrs.ini
@@ -0,0 +1,18 @@
+# Usage: npx ezs detect-cnrs.ini < data/corpus-simple-label-pascal.json > data/corpus-simple-cnrs.json
+
+# To see the enriched notices:
+# $ fx < data/corpus-simple-cnrs.json '.filter(o => o.ws.rnsr[0]?.length)'
+
+
+[use]
+# JSONParse
+plugin = basics
+
+[JSONParse]
+
+[assign]
+path = ws.cnrs
+value = get('affiliations').filter(a => a.includes('CNRS') || a.includes('UMR')).thru(affiliations => affiliations.length ? "oui" : "non")
+
+[dump]
+indent = true
diff --git a/doiwos-dumps/Makefile b/doiwos-dumps/Makefile
new file mode 100644
index 0000000..61cefae
--- /dev/null
+++ b/doiwos-dumps/Makefile
@@ -0,0 +1,31 @@
+# This ensures the next time you run Make, it’ll properly re-run the failed
+# rule, and guards against broken files.
+# See https://tech.davis-hansson.com/p/make/#change-some-make-defaults
+.DELETE_ON_ERROR:
+
+data/09-corpus-simple-cnrs.json: data/08-corpus-simple-label-pascal.json
+	npx ezs 09-detect-cnrs.ini < $< > $@
+
+data/08-corpus-simple-label-pascal.json: data/07-corpus-simple-pascal.json
+	npx ezs 08-extract-pascal-label.ini < $< > $@
+
+data/07-corpus-simple-pascal.json: data/06-corpus-simple-teeft-en.json
+	npx ezs 07-enrich-pascal.ini < $< > $@
+
+data/06-corpus-simple-teeft-en.json: data/05-corpus-simple-instituts.json
+	npx ezs 06-enrich-teeft-en.ini < $< > $@
+
+data/05-corpus-simple-instituts.json: data/04-corpus-simple-etab.json
+	npx ezs 05-enrich-institutes.ini < $< > $@
+
+data/04-corpus-simple-etab.json: data/03-corpus-simple-rnsr.json
+	npx ezs 04-enrich-etab.ini < $< > $@
+
+data/03-corpus-simple-rnsr.json: data/02-corpus-simple.json
+	npx ezs 03-enrich-rnsr.ini < $< > $@
+
+data/02-corpus-simple.json: data/01-corpus.json
+	npx ezs 02-extract-fields.ini < $< > $@
+
+data/01-corpus.json: 00-dois.txt
+	npx ezs 01-harvest-from-doi.ini < $< > $@
diff --git a/doiwos-dumps/data/.gitkeep b/doiwos-dumps/data/.gitkeep
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/doiwos-dumps/data/.gitkeep