From 415232b56a9d3b6550870db921fa7c88a525affc Mon Sep 17 00:00:00 2001
From: Christian Tacke <58549698+ChristianTackeGSI@users.noreply.github.com>
Date: Tue, 23 Feb 2021 12:32:43 +0100
Subject: [PATCH] Add codemeta_update.py

This tool can take the data from AUTHORS and CONTRIBUTORS
and merge it into the appropriate codemeta.json sections.
This is really a merge: If things already exist, they will
be updated.

Also apply it the first time.
---
 codemeta.json      | 98 +++++++++++++++++++++++++++++++++++++++++++++-
 codemeta_update.py | 83 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 180 insertions(+), 1 deletion(-)
 create mode 100755 codemeta_update.py

diff --git a/codemeta.json b/codemeta.json
index 8788d92f..2da3b3c9 100644
--- a/codemeta.json
+++ b/codemeta.json
@@ -8,5 +8,101 @@
   "developmentStatus": "active",
   "codeRepository": "https://github.com/FairRootGroup/FairMQ/",
   "issueTracker": "https://github.com/FairRootGroup/FairMQ/issues",
-  "identifier": "https://doi.org/10.5281/zenodo.1689985"
+  "identifier": "https://doi.org/10.5281/zenodo.1689985",
+  "author": [
+    {
+      "@type": "Person",
+      "givenName": "Mohammad",
+      "familyName": "Al-Turany"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Dennis",
+      "familyName": "Klein"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Thorsten",
+      "familyName": "Kollegger"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Alexey",
+      "familyName": "Rybalchenko"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Nicolas",
+      "familyName": "Winckler"
+    }
+  ],
+  "contributor": [
+    {
+      "@type": "Person",
+      "givenName": "Laurent",
+      "familyName": "Aphecetche"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Sebastien",
+      "familyName": "Binet"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Giulio",
+      "familyName": "Eulisse"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Radoslaw",
+      "familyName": "Karabowicz"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Matthias",
+      "familyName": "Kretz",
+      "email": "kretz@kde.org"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Mikolaj",
+      "familyName": "Krzewicki"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Andrey",
+      "familyName": "Lebedev"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Teo",
+      "familyName": "Mrnjavac",
+      "email": "teo.m@cern.ch"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Gvozden",
+      "familyName": "Neskovic"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Matthias",
+      "familyName": "Richter"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Christian",
+      "familyName": "Tacke"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Florian",
+      "familyName": "Uhlig"
+    },
+    {
+      "@type": "Person",
+      "givenName": "Sandro",
+      "familyName": "Wenzel"
+    }
+  ]
 }
diff --git a/codemeta_update.py b/codemeta_update.py
new file mode 100755
index 00000000..65c1c95c
--- /dev/null
+++ b/codemeta_update.py
@@ -0,0 +1,83 @@
+#! /usr/bin/env python3
+# Copyright (C) 2021 GSI Helmholtzzentrum fuer Schwerionenforschung GmbH
+#
+# SPDX-License-Identifier: LGPL-3.0-or-later
+
+
+import argparse
+import json
+import re
+from collections import OrderedDict
+
+
+class CodeMetaManipulator(object):
+    def load(self, filename='codemeta.json'):
+        with open(filename, 'rb') as fp:
+            self.data = json.load(fp, object_pairs_hook=OrderedDict)
+
+    @staticmethod
+    def _dict_entry_cmp(dict1, dict2, field):
+        if (field in dict1) and (field in dict2):
+            return dict1[field] == dict2[field]
+        else:
+            return False
+
+    @classmethod
+    def find_person_entry(cls, person_list, matchdict):
+        for entry in person_list:
+            if cls._dict_entry_cmp(entry, matchdict, 'email'):
+                return entry
+            if cls._dict_entry_cmp(entry, matchdict, 'familyName') \
+                    and cls._dict_entry_cmp(entry, matchdict, 'givenName'):
+                return entry
+        return None
+
+    @staticmethod
+    def update_person_entry(entry, matchdict):
+        if entry is None:
+            entry = OrderedDict()
+            entry['@type'] = 'Person'
+        for field in ('givenName', 'familyName', 'email'):
+            val = matchdict.get(field, None)
+            if val is not None:
+                entry[field] = val
+        return entry
+
+    def handle_person_list_file(self, filename, cm_field_name):
+        fp = open(filename, 'r', encoding='utf8')
+        findregex = re.compile(r'^(?P<familyName>[-\w\s]*[-\w]),\s*'
+                               r'(?P<givenName>[-\w\s]*[-\w])\s*'
+                               r'(?:<(?P<email>\S+@\S+)>)?$')
+        person_list = self.data.setdefault(cm_field_name, [])
+        for line in fp:
+            line = line.strip()
+            m = findregex.match(line)
+            if m is None:
+                raise RuntimeError("Could not analyze line %r" % line)
+            found_entry = self.find_person_entry(person_list, m.groupdict())
+            entry = self.update_person_entry(found_entry, m.groupdict())
+            if found_entry is None:
+                person_list.append(entry)
+
+    def save(self, filename='codemeta.json'):
+        with open('codemeta.json', 'w', encoding='utf8') as fp:
+            json.dump(self.data, fp, indent=2)
+            fp.write('\n')
+
+
+def main():
+    parser = argparse.ArgumentParser(description='Update codemeta.json')
+    parser.add_argument('--set-version', dest='newversion')
+    args = parser.parse_args()
+
+    cm = CodeMetaManipulator()
+    cm.load()
+    if args.newversion is not None:
+        cm.data['softwareVersion'] = args.newversion
+    cm.handle_person_list_file('AUTHORS', 'author')
+    cm.handle_person_list_file('CONTRIBUTORS', 'contributor')
+    cm.save()
+
+
+if __name__ == '__main__':
+    main()