From 6eeed661b603c30e6dc32f66aa331480d215d543 Mon Sep 17 00:00:00 2001
From: David Linke <david.linke@catalysis.de>
Date: Mon, 14 Aug 2023 21:12:22 +0200
Subject: [PATCH 1/2] Fix modified date handling; add env.var support for
 version.

---
 CHANGELOG.md                                  |  12 +++
 src/voc4cat/models.py                         |  40 +++++----
 src/voc4cat/transform.py                      |  34 +++++++-
 .../043_exhaustive_example.xlsx               | Bin 51733 -> 51746 bytes
 .../043_exhaustive_example_perfect_output.ttl |   2 +-
 tests/test_models.py                          |  77 ++++++++++++++++--
 tests/test_template043.py                     |   4 +
 tests/test_transform.py                       |  70 ++++++++++++++++
 8 files changed, 211 insertions(+), 28 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 108379f..ad590ba 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,5 +1,17 @@
 # Change log
 
+## Release 0.6.x (2023-08-dd)
+
+New features:
+
+- Support for two environment variables was added. Both changes enable achieving consistent version/tag info in the gh-based vocabulary management (related [voc4cat-template #11](https://github.com/nfdi4cat/voc4cat-template/issues/11)). #148
+  - `VOC4CAT_MODIFIED` - If set to a truthy value, the current date is used as modified date; the date in the source is ignored.
+  - `VOC4CAT_VERSION` - version string; if present this version info has highest precedence. A leading "v" will be removed.
+
+Bug fixes:
+
+- Modified date of concept scheme was not transferred from xlsx to rdf. #147, #148
+
 ## Release 0.6.2 (2023-08-10)
 
 New features:
diff --git a/src/voc4cat/models.py b/src/voc4cat/models.py
index 44dab3c..e907251 100644
--- a/src/voc4cat/models.py
+++ b/src/voc4cat/models.py
@@ -1,5 +1,6 @@
 import datetime
 import logging
+import os
 from itertools import chain
 from typing import List, Union
 
@@ -169,6 +170,12 @@ def creator_must_be_from_list(cls, v):
             raise ValueError(msg)
         return v
 
+    @validator("modified")
+    def set_modified_date_if_missing(cls, v):
+        if os.getenv("CI") is not None:  # Don't track modified date in GitHub.
+            v = None
+        return v
+
     @validator("publisher")
     def publisher_must_be_from_list(cls, v):
         if v not in ORGANISATIONS:
@@ -176,6 +183,18 @@ def publisher_must_be_from_list(cls, v):
             raise ValueError(msg)
         return v
 
+    @validator("version")
+    def version_from_env(cls, v):
+        if os.getenv("CI") is not None:  # Don't track version in GitHub.
+            v = None
+        version_from_env = os.getenv("VOC4CAT_VERSION")
+        if version_from_env is not None:
+            if not version_from_env.startswith("v"):
+                msg = f'Invalid environment variable VOC4CAT_VERSION "{version_from_env}". Version must start with letter "v".'
+                raise ValueError(msg)
+            v = version_from_env
+        return v
+
     def to_graph(self):
         g = Graph()
         v = URIRef(self.uri)
@@ -187,25 +206,10 @@ def to_graph(self):
         g.add((v, SKOS.prefLabel, Literal(self.title, lang="en")))
         g.add((v, SKOS.definition, Literal(self.description, lang="en")))
         g.add((v, DCTERMS.created, Literal(self.created, datatype=XSD.date)))
-        if self.modified is not None:
-            g.add((v, DCTERMS.modified, Literal(self.created, datatype=XSD.date)))
-        else:
-            g.add(
-                (
-                    v,
-                    DCTERMS.modified,
-                    Literal(
-                        datetime.datetime.now(datetime.timezone.utc).strftime(
-                            "%Y-%m-%d"
-                        ),
-                        datatype=XSD.date,
-                    ),
-                )
-            )
+        g.add((v, DCTERMS.modified, Literal(self.modified, datatype=XSD.date))),
         g.add((v, DCTERMS.creator, ORGANISATIONS[self.creator]))
         g.add((v, DCTERMS.publisher, ORGANISATIONS[self.publisher]))
-        if self.version is not None:
-            g.add((v, OWL.versionInfo, Literal(self.version)))
+        g.add((v, OWL.versionInfo, Literal(self.version)))
         g.add((v, DCTERMS.provenance, Literal(self.provenance, lang="en")))
         if self.custodian is not None:
             g.add((v, DCAT.contactPoint, Literal(self.custodian)))
@@ -224,7 +228,7 @@ def to_excel(self, wb: Workbook):
         ws["B3"] = self.title
         ws["B4"] = self.description
         ws["B5"] = self.created.isoformat()
-        ws["B6"] = self.modified.isoformat()
+        ws["B6"] = None if self.modified is None else self.modified.isoformat()
         ws["B7"] = self.creator
         ws["B8"] = self.publisher
         ws["B9"] = self.version
diff --git a/src/voc4cat/transform.py b/src/voc4cat/transform.py
index 454e34a..5ce4404 100644
--- a/src/voc4cat/transform.py
+++ b/src/voc4cat/transform.py
@@ -1,4 +1,6 @@
+import datetime
 import logging
+import os
 import shutil
 from collections import defaultdict
 from itertools import count, zip_longest
@@ -7,8 +9,8 @@
 
 import openpyxl
 from openpyxl.styles import Alignment
-from rdflib import Graph
-from rdflib.namespace import SKOS
+from rdflib import Graph, Literal
+from rdflib.namespace import DCTERMS, OWL, RDF, SKOS, XSD
 
 from voc4cat import config
 from voc4cat.checks import Voc4catError
@@ -27,7 +29,7 @@
 def extract_numeric_id_from_iri(iri):
     iri_path = urlsplit(iri).path
     reverse_id = []
-    for char in reversed(iri_path):  # pragma: no cover
+    for char in reversed(iri_path):
         if char.isdigit():
             reverse_id.append(char)
         elif char == "/":
@@ -63,6 +65,27 @@ def write_split_turtle(vocab_graph: Graph, outdir: Path) -> None:
         logger.debug("-> wrote %i %ss-file(s).", len(qresults), skos_class)
 
 
+def autoversion_cs(graph: Graph) -> Graph:
+    """Set modified date and version if "requested" via environment variables."""
+    if any(graph.triples((None, RDF.type, SKOS.ConceptScheme))):
+        cs, _, _ = next(graph.triples((None, RDF.type, SKOS.ConceptScheme)))
+    if os.getenv("VOC4CAT_MODIFIED") is not None:
+        graph.remove((None, DCTERMS.modified, None))
+        date_modified = os.getenv("VOC4CAT_MODIFIED")
+        graph.add((cs, DCTERMS.modified, Literal(date_modified, datatype=XSD.date)))
+    if os.getenv("VOC4CAT_VERSION") is not None:
+        graph.remove((None, OWL.versionInfo, None))
+        version = os.getenv("VOC4CAT_VERSION")
+        if version is not None and not version.startswith("v"):
+            msg = 'Invalid environment variable VOC4CAT_VERSION "%s". Version must start with letter "v".'
+            logger.error(msg, version)
+            raise Voc4catError(msg % version)
+        graph.add(
+            (cs, OWL.versionInfo, Literal(version)),
+        )
+    return graph
+
+
 def join_split_turtle(vocab_dir: Path) -> Graph:
     # Search recursively all turtle files belonging to the concept scheme
     turtle_files = vocab_dir.rglob("*.ttl")
@@ -71,6 +94,11 @@ def join_split_turtle(vocab_dir: Path) -> Graph:
     # Load each turtle file into a separate graph and merge it into the concept scheme graph
     for file in turtle_files:
         graph = Graph().parse(file, format="turtle")
+        # Set modified date if "requested" via environment variable.
+        if file.name == "concept_scheme.ttl" or any(
+            graph.triples((None, RDF.type, SKOS.ConceptScheme))
+        ):
+            graph = autoversion_cs(graph)
         cs_graph += graph
     cs_graph.serialize(destination=vocab_dir.with_suffix(".ttl"), format="turtle")
     return cs_graph
diff --git a/tests/templ_versions/043_exhaustive_example.xlsx b/tests/templ_versions/043_exhaustive_example.xlsx
index 17a6fd46457bf9c923eb8574234260d4fb8f4ee4..a23968b026860f74e29d02799ea238082e7eb476 100644
GIT binary patch
delta 4414
zcmc(jXHZjH*T+MUa4jGxMWl!t6p$h&A@q)bASHAV5H%4ZNRwRp5v3?a$_YUtO?nYT
zLa%D1gD6CLM+KyY4kEnB^UU0NKfHI|56?6A%<TX9pWj-0t$k+BhrLf6kui<PSah11
z$#h%Z`6YOUSrj-Y%5yh8werGbV|)8YKPf5BJfyL(b(mcL958Ck0c-MgsiA0Cx61na
z*d}|7t<RqQ=Aearry-48igXqvYo5dP6CC^iGTEPzGwhDk__=iT$w;oo-fBr;cp=!&
z<|sW3B6um}YMRM{wHdn*Qm7%a*%1~S87*Q|M6}Lt284v%Vhr;z_hrlGEk*1T+%%TH
zp-$bR*t4@}nq(o3wb*oIjcx|S6xHN>?wUSbL2T~5f<7Pf9r6a3m|K2?rsDXuvn(O{
zPRF}>#)Y)UK`G3D6LpMsF>rb!hp36+xRrh%G9Y)_e_XjtJfc(0wIbr3r*@&x%(EFj
zX)~Qg5HRy*A#JdR%qw4|suYA>QaZE<It({TH&s**U)E2FqG`fA>tjsrhVecD)vn*l
za!^3}6mj_atgDc=D(>)QJe!hDwjV}4u+_FuS<t#+l)$dZ7$cH8(6XwWX&THAhiXR%
zOh?7)94Tewfc|+C9%Inc@F?d{T<r14w{`)NZviT{-E#d}{7QK+#~>#+`*@r4E!>Oe
zmFz}sNl#XtM9*$w0?U~q96y?8cS(@Mt76**HqQhJ#XO5=l9JnYtJymx>8iDj65+mf
zt|BFPBl@{k@na+d&xA2&r{M7lO6$3q$uFId(6ZqT*oAeUlimEQ>Xgx0817+`4qF7X
zJ^)PYK<{QuX+0u}%1WuIA7|n2@;^%k)y3bH*%faNN2ILJTv^+(&%xH0P^)?s5mk~S
zm1*(QOYe&<=0ZcB^1J?+_p5vzBq?TgINJGUb*j7HIvP88NUTxOKp7r$OMY9UCpF8)
zwIoS>vb}D}VU$C{`1sv*OL4}0fukX2h*lu4>N)mWSVhq72Vf4|bc_Y=?;o4kVDYnm
zWaTnXu!EASxc}wC&wncYP%qFxnok?mA9A>7o?V*1B3Yg4i8iYg_uok06%OfwN&ca=
z7HE9U<ZveK(R7($RlJtFe!gKzxk~*#zkITqJl9bj^xQ*%wb^{nV@n17-T6n-S(S)@
zLrJu@foUpp9|30dSb*mLGLI)OQ8l({c(-ri7+-hNMDEN)p!2h)<a-3FWA-4`{_>1I
zYHIC8kFvvuj*Gkwqj`kh@kgQsJ#_-Jg3`y)TBRogKUxLtvEj}2tM_#2>~(ONT8#I$
zU*@eP=UrqyW+Py^#3#WvUx6G&{j76hbh~NVB3i}4$MbZ-w(j%-8Ul_pa)5GFR?6~`
zKgF09#=43U!24(faO2$jylRIp{<&t=wF<m<AxGu2$|x%NyS`2O%^0In2lfxR#;NuM
zx*iVAEb@r^Q%HUlT;~)B!DoruzSdp-w({$y8uYfKjnPZIa0~mf%7l6VJ$R`W#S7sT
zRCm5s{Dxqz5o6CAO$)*^qkPP+>Eh|HZMAUOfo1}5O=2TJ7mg+NW?+R^QGEE)iDO)0
zs1x|XSO+#*S^#BS-s6_Mi8<Yrydc#JCL^AP_jIoa^#?1pB$!<-aB3z-SUJX9>Zi|~
z@xDg5o)-JVnZhebX9ca@-_3u{)`hl&g?bhsZI0B$sK**{@gW0@cUU?Pb+>8v>Sn#V
zTlMOzDH71(a`YN6EVa$;DmRsC@hx19bK1OGY-I4gH8VWKTxg5u&lHBjr@kK^pQ@%U
z>et!bQCi{O$(cfYv3PgSix%hzx9y{I1r5X>3e7<%(AOH})R8HgiiHz>HHBh<A*MP?
z@P=D*NmYp5W^-0wkFw)pHcF2nI0ozm_~uKZ9}X~idl78|b(RWwaI(?_KRb5Z>oNhO
z*sht@zRX?07}(^}B)n7|6**{2W=y`5e6L~y5n(Bj8z?VSM{m7<!|;SjQCeit-9}`4
z6aBKz%E(s6`k9vn>an4>9~3oO2>G734V)cMlaqLRneQG=AWNUy7<x3YQ^kG5=lw`F
zbNvD!&tmyJbIWUDvb5x)Cx4yhtxhj^A-_8@ScfYr@gp!+q^Q8!%Zk>cK8U@zwHIv}
z`k_=ktGhlp%<ow}0|c_S2l*|D!=v{(S~!`<%vA#*u1VV-KQzQJZ@Il$Gz_J=`y;uE
zbw-pBeGCS^7F&}?kV+D333lcNol^@xfZmjV-A($+@XyZ0(VE@)UZ8$+KKlh>WujYo
zbZ>cPwO%>$1-CM=J9D9@?KxUfX=k>3%5IMy9|i0UZZ#MW0|Q=r;|9R5#o;D{GfKPe
zKJ)9?p&IXB;|lF5os&Bcx()Uc1D0l1mls_Iyng-IK)<&0_sa49X?PO-TENyD*!UV(
z-SoOf&UvzzCGrw4xhif-Ro{Bc>-?1c`9<9U$LZIRb5*ar|FrY)iK<(xlF(f!B!2CU
z4R6%9_Rvd>mnH~&{nV#~B-nMWrxz3ss^8u~aYyp|)cabU7`<BQFjpea&ztYp;@ji>
zysamV->O>j>0na<KYnDmSK+?+1Q6XH7rR(OIQ_ll)p!zJq50*T0EfA{j4$z9$i^_2
zpP%9#Hx}x~((<XVs;e=}&MEbOIN8kP1&mp^3QT?Kv-hc~mE7vFP+oYt!F14fKPjnE
z|Fc5mEOxoj>vxVDX*m8yd$Gbg!Or9Jpc_e1+<TSPCWt`pH1N3xD1RFCPHY?5b|VkR
z@3^^apn$%??%2gOg4{8snR%qnJUH`vQPdBBubn%7Zs(2X&t<Raiaq*d^0s~2h>ggW
z^l`~$e0Dkw%zA%3+^$h})pJM8H>XnJ`x!E@Vff(l2)%pmTF1t-?fC(ohN!t{JD42O
zOXg6Ge@>;-5)j={9hw#nJKnv%VAWAB>&-J3hlm*o5PX|`@%*G~hS#9L$E*7Czj}=$
z<@m-%6*&|aT=)p%)t6Aee1F-<o^_;ud*Z%!Z0W(3jXjT+ODUV3I~|wK<Q%BgzIIEk
zw<XN4_{j5@rfxq_@CsPl`oNTDaPnB5fiIg$N-+@nBUJKOr*D1dsL0JZ<3hvhmM|<2
z5t}X6IU1ilx__6)_KURXTM}A}jY&zx3cwqm-JgoWZK{^6t?q`T$HjcaZ>eI;SlQo5
zzttT>0*e!g-!jB5Vr50zbpD0Hi*FFWrHOH4D}~!MEhSprjYz){7DEDbEH7Xae!IJP
zuiYtq4C!CsTb39aTPf0Z-SU2`yD{lS@euI<SrV~EBuz^$?Eg0Nr1wtcL|k>I%%J$U
z{`(@I2)a+ptiwu%$vi`+PY+q;-6b-{%AAZRTIY3OoJn|!$%{;%-P<WO0D|;mCW`M$
zmA&CnU6k0DA`>bG#g4bn4avj{)AWMLZwwLKc$j*}og5w(6y+sOhNc%vrW+#8;9(ja
zK{-6^C`1}go2K`aEM<hah=*x*WaRL0q9|!N3z}Xy+06(ciHBY7_>jZHgF>X^9B6ux
z<TpkLG#;ka@im7B;73u?ab7gNC^Fp$p^S%VcO1&)Ig3JM;GWR*V#rb$gccsA(*e)r
z5kgTiaL;LaAlVIrFu=pEb!g=Bh@cRexEz{Z9Qh3fVUCCCcHGJ35l2xnFIyk^8<DY#
zgmc&QnTh~7UHdYQ1DU$i|265&Y{Zh<`M+k?3^Y55w3DEcGFSe#;NQ3VzoS<Fm8u6t
z!Et5yiTSF``(?BaQdK_&ZT$nMJ6yPrdHDGkan!;7r2y&LmmLcZRO^AIWa*k*mPI^u
zaG)KeD+huCD*RxSiaeOA!Y@{R-|@LP3nkOO4B1W!>F;`r0$o4DZ=epxU+?k+oyOre
zWDm#dzb^${>2NSW<t(#Yftg--K>`?~BD-JUB6wNN3`|lL5Zy0C7Qar(bSy)5kS-Lq
zN9uY^4h&VupbI}1w0R6}4^^n33qM8bK6n8}son;YRhhvhwM(G0Iv=xsXKkFakR(V~
z6WxF27q%%cX(y5TyRKMUZSJ7`Abd&;tY;WPJ0DU*-_971%}Kd#^2zwIC4x8A5n!qs
zKS0FkI+p!&0{lO_6NWKl@y!1JtT<CLL8?X<_)wGc--8*zB#6iL|8oNTW8J?XoM*md
zA4^dV*zw!gUHOULW27Wsy+vRWgpDVvP?&1r3_wtfjmvvMj*!LpIc4NfIk6}Ayz_PG
z$+!!;J=nH|KltJ-gc-g!)t@n&AJQt;U9MA5m@?E`&wE;SGs(&GsdjAIA4r#ATK<Vz
z8pSK(+*AVBD@PP@8p3)#Z}7z<-d5|Q24$$5wTHNwVhO|-CkkrO9JiZ#Q=jXqsfW7u
zTQvZFG6S5_5iWg4TuE8#uihfwDNwa;I{BQFJI}Nj^PMWqui|=)MflT*eJtC~m$jR5
zKUdV^Z=m8<nNQ;6W^E}<%EA7=i+AjX<_Rqli^xaH?_9#SJkWTtH8*oUL(d@tYNB0`
zsf5szP<0g6S|qKK#w)zESix3vGGhq3BABMI1Xnz48{50%X4PpJmuV@Go&J2NcP5(k
z3ORD*=+0!o7sfCB<8~7=7ao&`J`+Ew258nm@;?6jS#F+kaMGZ-*M~Fy`+oqGieVT4
zO{rQ?Z5EL3%L!7bl2B2O1G`1#<U2Tvut6Y#O%TYb{oUkWmmT^bN%?~$R6_f}2JhQp
z-+g-;{PaN-+V}8)vNuRF@Peuhnhb7WXoCuaD%jj02aSjUQ)9Tm(~Y7~xg79w-YL+w
zQ4(64w{Q8uSNrsF!2w;`r?4XMY?A>41$1jNgtAe<m!-U*GUW&u*31TOHl2n#l^h_+
zW+XJSbl(c?M?@Ji!O~_$hP&X(zM6qjEs6|cpi9en-h&Tme{-x3g+SQ<dYyf@4HmR0
Q?XPmK<vi@+``@Ae0gFU*Hvj+t

delta 4394
zcmc(jcTf}i`iChZf<P=N(jx)_27)A^h!Uw9nh=@@C<sL9J%W@KR6M8@Sp*X#2uPPA
z9RZcl1Pn-3fdr6#LYE#1m&3U;cYgmkzqxZ~&d%(!pMBoh_nq0<o!RfEC$K+HU@sQp
z<apSWUX%igbIJj1F;4!c<f^5|nm>Q;2swN9SRT?$#^%pyr_!C!E1Dv`!CtkcIyU;9
zL3gi{{6^;K8e`MZZShgHUL(ST7m=~L<YlAb4JJRKa=+ZM)}v%yV@9*Oe00%$I6$3y
z5Msx*^q2(-XXxdbAHHfSn2kI&5iN6@Pu=Y#BHdnA#nKwMtRJtGY3@Aa>pobiME1?l
zJ1~wqakCD0R`f|=a+hr%G~PG;<1Zy=-S4lQmr&k<&x!`gPjxL>8Xx;)(axB#JNqb7
z$F<^*aaOK>(+=iBi!;AP*1H%-6g<gLI&NV4xcnXK?R@1eGT?4=BK3hiAb6i1XKT={
zlI0gzroWX6l(nI6Zi?G~V>%*vdh@D|TwK(MniTc$ds<H=`fS;n2|P>R&{Im0^@JMh
zqcAVhF~iej?Cf+ad$l}Seyo_jLFz%gsejht$93@2cH9{`zx3(L<J>uT#qBFD77npk
zR!Kd`N=qV!yPTb$?T)>q&qd3LmpS|%QVYTv_DglZK8OI<t<5ubpJ`KA_e9&q2XN(F
zLebI5W*vE&&YN3iT=fFnwek+1x!<2pz*pbydT!g_yaR<r1=Pa+FrB*;3W&kIFYV`*
zozj=6N{v}Kc{X;Ow|k${Wj>4Z56_QuD8)M8OG0Z8%{?6|^oUG5O7NV1fP?#A@at`O
zwr~^816)`TP(8H`J?@m>&CZVWAND>;6|g~+c;Smh-L01$ny)at33oMwiFG#y9H?}X
zAoaPpZI?kA;1DE%4{d1VQ(X;vH)N3vhndR872K+UV$qD`5+m;8V}(WH@mJrfZ0T4+
zv+TUfQbNZX-}`vn*q4<1=qTe&Vz@;thjzXoFkxY5=qIIH^3c&A%)#e<6u!KFOH$3K
zc&)VHr=nm?mwUvuW&0shnp)DMYf&y~UoyhDz9d>^XZCFh$ls=3${$Ls4OT!KzACFf
zNH+De*NGa?k}ZeU9)EBrNOG!-Ti1qTi`u2KW9zIiL!9f%2n#KbVSO(e@Xo>kS^wxr
zG%$5!CYReEG8mAb794ORuU_=C$plZl$lb~`>Wo^Ju*1mtd6zXt;Y@T!%-5!|uEr<G
zsnulVZ*GmNr-i2rEL>DZBjgJD!^M)JV&a8vZ`s3*ujwo*_IHSt?HDW`elMw7hgH4%
z9_zj0v4*U{u7xd^g+vt|6YC%^EkeDv9r^rWmqn0u($SVbrH~uE;3zwP=`{C47<T~r
z*~{mLSXiJfpgpf37^iX!fayH2xYL!#z!u<xgOod1j^hnI9vS2_gIkP0VNJ_#A8y1o
zPkkmV8D>>-%0`IZZm_V6mwbw0L75#ZvXa_d-gHou4u4x#m}7!U7(Wm>)Is}@JpY{b
z0(Fl!E=^g=+NMT6{ZldOva444L!k=>nNsB&Rz(t}eHYI_;wALQfIxmsn_IUM|CSJ6
zxvlShza}BBat{phfUD^T>z@T`Z$<O+H{^`gU(hj~gvKEH%x3(oR_eInnT$kbCw;ww
zvUb?%uJPFOQZ>@1nANQBd~tzKTIH)o)~0);!kwDN@+^r8`)Hp|w#ppj6Z#;_jJy>w
z<NTT~W!8_dvs{)V0+LUVvsG>dR(UHP7KY2DGoez#@1ME%g4Mn=>7y2yux9G5du~6(
zmY~fFt2N+)0Kq7f6Fu|Y9A&vEMDn*6iW0!rvZt*{)t~&6r!%CyqT>)H%QQA$Rk`Sl
z!7B&rcnIcfs-g7!Q}pul2HMLxb^`%8^q%!a0vA`9jMR-2z_G<vqoj*APft>H6cPqh
zH%^H?O)-b~J(DR7;arC<ZiwPM-ZQ>c&1I^MU?!_0InfO-CL)BYtbPa=%?@x#X>u40
z90<l4HvEznlXju0r}i;Vl?7a0UfeJG=B#Ba?GLMwyHz<ceu_LTLgkNgdzunHYK<V`
zhU!Sg>ofI548U0XIL;*uv9@$=0DQ+cEb;c%`6FNX)Y{0BPc9ixp}nul#5R;S@b)N*
zPGOj6r~Z>Zb8$1{UTG4kV_qyQEIT`UcuA$k@V6e`|HVr4Jd2?oQzx=XLU58?6t1|u
zMXu?O__{V&lPtz=*Yjr5o8!+YjKzwTwOweI93ALuNkxZj`0vd3)O;-pW2_E$hB0P>
zf|;{p=;7_<rIiop=&8;y=1R}ctMHo|z#8+0IC^_4_Cv@9W4HrO`jgqyT(b^r(SZeY
zNXX7;Qjgmm_>WDx)k0^(ug&uvS2`IpsXY)4(skx`fN;@uCUdK=eI0`G5_A2r+5X51
zplRHlU=CVcCsi!9M>a?E8d2MRLbD5g;ZIFDoPtZuAy<<0@OEEo>v~_(GK%U7>G2F<
zsm;vsyth65B1fk)qf<E5kQI<<JN*GBE}uUC<y+=t-dpX3^;m-+bQ!DY6nwU$<8ydm
z=Q)TU^;(&>TT*-#qoQ1!GH5`lEYJl+lxj?uQ~He?&J$6!2-0q|{=b@C3ng7~e6#Tm
z5=z2*!8xq#L~+#Z==dVWFAICRp4BRQOT<srN6maaW0cHsh)@l`Eeb!q&D3kXBc!hT
zQ?*c_wdP0AOKry%YrJeV(K+a)a}qJ17E?OwgW2gQ3?z2O-gQHN84Lnm(rH^hL_2+E
zRiyz%d~L02Jh(1}?|$rEbt5dp=^VuO_N?O4u&usvbPBp+{<;W6G)Wv7G(#Kq?%szR
zQ=@&7ZrK^V8Chh=lhhWqm4IKc;GAlWNpT{;F!ArBE%nUmQPyL&=K2MjVrS!0InN;j
zRrl2f<y3nv#!&_z0qObsAw6qMOUhTZAkmR%#4}nL^kd8EQ)Aw*1AD|e^o-TFlgy&e
ziHr<u@P`f#tIm?wRb02$w`zQ}x_|sUJTWOaGUGJAGZn9*v6Q-ztfFDE$1>7Ja<=2_
z6k;%Y6@SiEb9E+gapljSQ3uUarB9fru0eA&oFRRX8^8(DT3G4_vyb1hZPSP)JZ(xc
z?Z(!m5AX|hyGMLmiCJPw$!vmiHcZ_)h<S5$Wj)v=@qUR&!o({Wr?Z+&yQVc?o9|!h
zWx~X3*lA}q*>;SzahoqT5&T;TI9JQG>sl+c`I;s^DX~qMNQdpI*D$ceunka>ZGR2x
z>|I>xO!?dF3L|UJHea*E<dUI;i7c3kbM@b<M5a9*_M`;X_J6~m=)>>xG8FtJp9=N1
zQmW~zxtJp!clcdu5Qr)SvZ~a)Q-)L0M_NI)(mXkH0M>`gRW?}t)uIt}3)a^Ih^smI
zH@rHY=z2BJAvZG9Ric-S<u&Lt$sur~9Mf<Ly*!b`RuhDfD^8o@okQS5jiuo<dU^gJ
za+x9|TyZ*-xE#U}lw&&1xR>WK5o3ywbH(XWigO6ZQDf=2>%BZriBYBqC0Cpt<x37h
z2<4c8^X%n`A-0+#&bi_)0hGlYf*5Kn0~glI6G!C2BD7p_mnnyG2@)vB*SMHoo_Hb#
zi_mw)VJM2Z1ZmXRYaFSUCxIA+MPOZV`V^B~f*i^*6Ia;FlSpjEBCfgO3>+%>e*>TQ
zLKua>xB1x5iDY*`_AknISe3M@JAM@$eW>@JD+9})eaF6#=eX|e^xKBNANTvJ<?qY?
zZ55nHOCB8zFrd2UCd&tuW_niiQWD?URe}rX6Qbv6l*r1`(QWs;+R4<e$jY+O3m$=H
z!nD4~%5nn8M2CPT7bU>6nig<H8zDTkdzUza26vYi^1P*);cn8S>6j|lBMQl{>C)fH
zm@3~R3MmiVPLKY;fUho|<n+H*@Zz`)0wi5Tf)_MlAgFl@{G!P#x7!G5L&O>MxVsSL
zjcf_RNt4ax)jmF+wkbAAQx&V#KH;9WX~Id<%HUZoJFrue7cAD20zYaU+ueCNQ%J#Q
z0R8LWTP?|dJ3g5jPZxr`d}*e0McLw}*9MDf2%Cro8(%8xBS?5p_Q&szBa3R@8&Fkn
zNlOW|*A@e0DT&?g{~Un-=k&REnMfP`(*G?6a0cLnP9-?3V+1JK$dzXPa{&IS>3OFI
zy4z#X2Z4>n874LRPSFl$X_nqiJjZp1BY=7-Z=ED7@bsCT7a7VQxsWuAMh;ab;7u<L
z5Z#;47$tsoZl6CO^8Bg{D{gfdl8>jhG%2()mKR@YpS_TCZs2}ENtcua=CQ<=+QV2c
z{go18%Ta!{&NBOej5M7~-nXw&Y5YSbw5x$7(X4c-Dw(*ljG=_v^t!YY>hC$4Y>x<?
zdq8iN111}~Jc}JdT9ogVB1IiINL-VG8mx{zN4q#=17nfm74s|+jyam)Ykc5?k)lG*
z{Gn8VcTQi-yTy%O%%rP39Sq}Rtkt)^|5oAAJ(;4nczVO5I&fYuI#XlrF?*gdBbV^A
zy|}G>-;(zmoI6~%EbcBR2UAoa;!Z?@jlGx-zY6K~t$Md*BD%xU&)nH1v1LY8dOhD?
zXF-V=J$&%jSXdu>U(;x6BNyd>)aJcF4)c`tb*~}~*1bgr@ZkyjzkdTj8nCSW;LV0h
zY&yK)Qp4^`+yG~l1{WL7aO|x@KxpGJHXD__Qk)70*wk<cbZ8V}lfJyCZd~R7lN#mN
z`hxdPzcs?yzTDf3Voh+iqYw9@Qxlxk1x#qV$f^eRHJxL903K@AWK+rkQwvUjZq0DE
z^1QuRwi|;BcB3fB)*{EoECNq{5(Y06bAw5r4uOs>LTveO_Y|=O$#$T0FVb6(tWF^G
zlL_lX&}Ww^Ao<e;)*JDy7g&u!ja^EDA+4u`_Z}Qp78Y$b7M8=ipThsymce(e8oMnv
MTTkuxBmd?91^9qquK)l5

diff --git a/tests/templ_versions/043_exhaustive_example_perfect_output.ttl b/tests/templ_versions/043_exhaustive_example_perfect_output.ttl
index e033013..0349cf9 100644
--- a/tests/templ_versions/043_exhaustive_example_perfect_output.ttl
+++ b/tests/templ_versions/043_exhaustive_example_perfect_output.ttl
@@ -27,7 +27,7 @@ cs: a skos:ConceptScheme ;
     dcterms:created "2022-03-07"^^xsd:date ;
     dcterms:creator <https://linked.data.gov.au/org/cgi> ;
     dcterms:hasPart <http://example.org/example_collection_uri> ;
-    dcterms:modified "2022-03-07"^^xsd:date ;
+    dcterms:modified "2022-03-10"^^xsd:date ;
     dcterms:provenance "Example Provenance"@en ;
     dcterms:publisher <https://linked.data.gov.au/org/cgi> ;
     rdfs:seeAlso "1.2.3.4" ;
diff --git a/tests/test_models.py b/tests/test_models.py
index 3d304cb..d79eb3d 100644
--- a/tests/test_models.py
+++ b/tests/test_models.py
@@ -1,3 +1,7 @@
+import datetime
+import os
+from unittest import mock
+
 import pytest
 from pydantic.error_wrappers import ValidationError
 from rdflib import Graph
@@ -81,20 +85,81 @@ def test_check_uri_vs_config(datadir, temp_config):
 # === From here on: "old" tests from rdflib.vocexcel ===
 
 
+@mock.patch.dict(os.environ, clear=True)  # required to hide gh-action environment vars
 def test_vocabulary_valid():
-    ConceptScheme(
+    cs = ConceptScheme(
+        uri="https://linked.data.gov.au/def/borehole-start-point",
+        title="Borehole Start Point",
+        description="Indicates the nature of the borehole start point location",
+        created="2020-04-02",
+        modified="2020-04-04",
+        creator="GSQ",
+        publisher="GSQ",
+        version="1.0",
+        provenance="Derived from the 2011-09 version of CGI Borehole start point list",
+        custodian="Vance Kelly",
+        pid="http://pid.geoscience.gov.au/dataset/ga/114541",
+    )
+    assert cs.modified == datetime.date(2020, 4, 4)
+    assert cs.version == "1.0"
+
+
+@mock.patch.dict(os.environ, {"CI": ""})
+def test_vocabulary_valid_in_ci():
+    cs = ConceptScheme(
+        uri="https://linked.data.gov.au/def/borehole-start-point",
+        title="Borehole Start Point",
+        description="Indicates the nature of the borehole start point location",
+        created="2020-04-02",
+        modified="2020-04-04",
+        creator="GSQ",
+        publisher="GSQ",
+        version="1.0",
+        provenance="Derived from the 2011-09 version of CGI Borehole start point list",
+        custodian="Vance Kelly",
+        pid="http://pid.geoscience.gov.au/dataset/ga/114541",
+    )
+    assert cs.modified is None
+    assert cs.version is None
+
+
+@mock.patch.dict(os.environ, {"CI": "", "VOC4CAT_VERSION": "v2023-08-15"})
+def test_vocabulary_valid_version_via_envvar():
+    cs = ConceptScheme(
         uri="https://linked.data.gov.au/def/borehole-start-point",
         title="Borehole Start Point",
         description="Indicates the nature of the borehole start point location",
         created="2020-04-02",
-        modified="2020-04-02",
+        modified="2020-04-04",
         creator="GSQ",
         publisher="GSQ",
-        version="",
+        version="1.0",
         provenance="Derived from the 2011-09 version of CGI Borehole start point list",
         custodian="Vance Kelly",
         pid="http://pid.geoscience.gov.au/dataset/ga/114541",
     )
+    assert cs.modified is None
+    assert cs.version == "v2023-08-15"
+
+
+@mock.patch.dict(os.environ, {"CI": "", "VOC4CAT_VERSION": "2023-08-15"})
+def test_vocabulary_invalid_version_via_envvar():
+    with pytest.raises(
+        ValidationError, match="Invalid environment variable VOC4CAT_VERSION"
+    ):
+        ConceptScheme(
+            uri="https://linked.data.gov.au/def/borehole-start-point",
+            title="Borehole Start Point",
+            description="Indicates the nature of the borehole start point location",
+            created="2020-04-02",
+            modified="2020-04-04",
+            creator="GSQ",
+            publisher="GSQ",
+            version="1.0",
+            provenance="Derived from the 2011-09 version of CGI Borehole start point list",
+            custodian="Vance Kelly",
+            pid="http://pid.geoscience.gov.au/dataset/ga/114541",
+        )
 
 
 def test_vocabulary_invalid_uri():
@@ -104,7 +169,7 @@ def test_vocabulary_invalid_uri():
             title="Borehole Start Point",
             description="Indicates the nature of the borehole start point location",
             created="2020-04-02",
-            modified="02/042020",
+            modified=None,
             creator="GSQ",
             publisher="GSQ",
             version="",
@@ -121,7 +186,7 @@ def test_vocabulary_invalid_created_date():
             title="Borehole Start Point",
             description="Indicates the nature of the borehole start point location",
             created="2020-04",
-            modified="2020-04-02",
+            modified="2020-04-04",
             creator="GSQ",
             publisher="GSQ",
             version="",
@@ -138,7 +203,7 @@ def test_vocabulary_invalid_publisher():
             title="Borehole Start Point",
             description="Indicates the nature of the borehole start point location",
             created="2020-04-02",
-            modified="2020-04-02",
+            modified="2020-04-04",
             creator="GSQ",
             publisher="WHO",
             version="",
diff --git a/tests/test_template043.py b/tests/test_template043.py
index 8de5bb2..beebabc 100644
--- a/tests/test_template043.py
+++ b/tests/test_template043.py
@@ -1,4 +1,6 @@
+import os
 from pathlib import Path
+from unittest import mock
 
 import pytest
 import voc4cat
@@ -39,6 +41,7 @@ def test_simple():
     ) in g, "Provenance for vocab is not correct"
 
 
+@mock.patch.dict(os.environ, clear=True)  # required to hide gh-action environment vars
 def test_exhaustive_template_is_isomorphic():
     g1 = Graph().parse(
         Path(__file__).parent
@@ -52,6 +55,7 @@ def test_exhaustive_template_is_isomorphic():
     assert compare.isomorphic(g1, g2), "Graphs are not Isomorphic"
 
 
+@mock.patch.dict(os.environ, clear=True)  # required to hide gh-action environment vars
 def test_rdf_to_excel():
     g1 = Graph().parse(
         Path(__file__).parent
diff --git a/tests/test_transform.py b/tests/test_transform.py
index 5b0daa0..064d6d0 100644
--- a/tests/test_transform.py
+++ b/tests/test_transform.py
@@ -1,8 +1,12 @@
 import logging
+import os
 import shutil
+from unittest import mock
 
 import pytest
 from openpyxl import load_workbook
+from rdflib import Graph, Literal
+from rdflib.namespace import DCTERMS, OWL, SKOS, XSD
 from test_cli import (
     CS_CYCLES,
     CS_CYCLES_INDENT_DOT,
@@ -600,3 +604,69 @@ def test_join(monkeypatch, datadir, tmp_path, opt, caplog):
     assert (vocdir.parent / CS_SIMPLE_TURTLE).exists()
     if opt:
         assert not vocdir.exists()
+
+
+@mock.patch.dict(
+    os.environ, {"CI": "", "VOC4CAT_VERSION": "v2.0", "VOC4CAT_MODIFIED": "2023-08-15"}
+)
+def test_join_with_envvars(monkeypatch, datadir, tmp_path, caplog):
+    monkeypatch.chdir(tmp_path)
+    shutil.copy(datadir / CS_SIMPLE_TURTLE, tmp_path)
+    # create dir with split files
+    main_cli(["transform", "-v", "--split", "--inplace", str(tmp_path)])
+    # join files again as test
+    cmd = ["transform", "-v", "--join"]
+    cmd.append(str(tmp_path))
+    with caplog.at_level(logging.DEBUG):
+        main_cli(cmd)
+    assert "-> joined vocabulary into" in caplog.text
+    # Were version and modified date correctly set?
+    graph = Graph().parse(CS_SIMPLE_TURTLE, format="turtle")
+    cs_query = "SELECT ?iri WHERE {?iri a skos:ConceptScheme.}"
+    qresults = [*graph.query(cs_query, initNs={"skos": SKOS})]
+    assert len(qresults) == 1
+
+    cs_iri = qresults[0][0]
+    assert (
+        len(
+            [
+                *graph.triples(
+                    (cs_iri, OWL.versionInfo, Literal("v2.0")),
+                )
+            ]
+        )
+        == 1
+    )
+
+    assert (
+        len(
+            [
+                *graph.triples(
+                    (
+                        cs_iri,
+                        DCTERMS.modified,
+                        Literal("2023-08-15", datatype=XSD.date),
+                    ),
+                )
+            ]
+        )
+        == 1
+    )
+
+
+@mock.patch.dict(
+    os.environ, {"CI": "", "VOC4CAT_VERSION": "2.0", "VOC4CAT_MODIFIED": "2023-08-15"}
+)
+def test_join_with_invalid_envvar(monkeypatch, datadir, tmp_path, caplog):
+    monkeypatch.chdir(tmp_path)
+    shutil.copy(datadir / CS_SIMPLE_TURTLE, tmp_path)
+    # create dir with split files
+    main_cli(["transform", "-v", "--split", "--inplace", str(tmp_path)])
+    # join files again as test
+    cmd = ["transform", "-v", "--join"]
+    cmd.append(str(tmp_path))
+    with caplog.at_level(logging.ERROR), pytest.raises(
+        Voc4catError, match="Invalid environment variable VOC4CAT_VERSION"
+    ):
+        main_cli(cmd)
+    assert 'Invalid environment variable VOC4CAT_VERSION "2.0"' in caplog.text

From 1707153ac1a44f7ce4ea3da811ecfe1f06d04e71 Mon Sep 17 00:00:00 2001
From: David Linke <david.linke@catalysis.de>
Date: Tue, 15 Aug 2023 23:56:28 +0200
Subject: [PATCH 2/2] pylode docs: correct ttl-file location

---
 src/voc4cat/docs.py      | 8 ++++++++
 src/voc4cat/transform.py | 1 -
 2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/src/voc4cat/docs.py b/src/voc4cat/docs.py
index 2f94da2..bf39c08 100644
--- a/src/voc4cat/docs.py
+++ b/src/voc4cat/docs.py
@@ -32,6 +32,14 @@ def run_pylode(turtle_file: Path, output_path: Path) -> None:
         '<section id="overview">',
         '<section id="overview" style="display: none;">',
     )
+    content = content.replace(
+        "<dt>Ontology RDF</dt>",
+        "<dt>Vocabulary RDF</dt>",
+    )
+    content = content.replace(
+        f'<dd><a href="{filename.stem}.ttl">RDF (turtle)</a></dd>',
+        f'<dd><a href="../{filename.stem}.ttl">RDF (turtle)</a></dd>',
+    )
     with open(outfile, "w") as html_file:
         html_file.write(content)
 
diff --git a/src/voc4cat/transform.py b/src/voc4cat/transform.py
index 5ce4404..ec131c6 100644
--- a/src/voc4cat/transform.py
+++ b/src/voc4cat/transform.py
@@ -1,4 +1,3 @@
-import datetime
 import logging
 import os
 import shutil