From a5391a0f80f17abcea01502ef825d6ddf7bb1e2f Mon Sep 17 00:00:00 2001
From: fordmadox <fordmadox@gmail.com>
Date: Sun, 20 Mar 2022 17:54:51 -0400
Subject: [PATCH] Schematron tests (#286)

* use ietf regardless of script encoding choice

* check dates filter added

* a few more date refinements

* update schematron slightly

* update intro

* update build before generating new beta release

* Revert "update build before generating new beta release"

This reverts commit 7212b79fcbea69e376acf3011f8c7db5ead466c9.

* update schematron tests in correct file

* update build before new test release
---
 source/modules/schematron/shared-rules.sch | 51 +++++++++++++++-------
 source/release-info/intro.xml              |  6 ++-
 source/release-info/metadata.json          |  2 +-
 xml-schemas/eac-cpf/eac.rng                |  6 ++-
 xml-schemas/eac-cpf/eac.xsd                |  6 ++-
 xml-schemas/eac-cpf/schematron/eac.sch     | 51 +++++++++++++++-------
 6 files changed, 88 insertions(+), 34 deletions(-)

diff --git a/source/modules/schematron/shared-rules.sch b/source/modules/schematron/shared-rules.sch
index d18582d..6a4cf61 100644
--- a/source/modules/schematron/shared-rules.sch
+++ b/source/modules/schematron/shared-rules.sch
@@ -12,22 +12,23 @@
         <language value="iso639-1" filename="iso639-1.rdf"/>
         <language value="iso639-2b" filename="iso639-2.rdf"/>
         <language value="iso639-3" filename="iso639-3.xml"/>
-        <!-- ietf-bcp-47 -->
     </xsl:variable>
     
-    <!-- VARIABLE $language-code-key: the EAD3 document's /ead:ead/ead:control/@langencoding, with iso639-2b as a default value. -->
-    <sch:let name="active-language-code-key" value="(*/*:control/@languageEncoding[.=$supported-language-codes/@value], 'iso639-2b')[1]"/>
+    <!-- VARIABLE $language-code-key: the EAD3 document's /ead:ead/ead:control/@langencoding  -->
+    <sch:let name="active-language-code-key" value="(*/*:control/@languageEncoding[.=$supported-language-codes/@value])"/>
     
     <!-- still need to add something here to distinguish between active and deprecated codes -->
     <!-- and will also need a functional way, or ability to hit an API endpoint, to test for ietf-bcp-47 codes, most likely -->
     <sch:let name="valid-language-codes" value="document($supported-language-codes[@value = $active-language-code-key]/@filename)//(madsrdf:code | iso_639_3_entries/iso_639_3_entry/@id)"/>
     
     <!-- until we have a better way with dealing with "other" as a value, etc.-->
-    <sch:let name="check-language-codes" value="if (*/*:control/@languageEncoding = ('otherLanguageEncoding', 'ietf-bcp-47')) then false() else true()"/>
-    <sch:let name="check-ietf-codes" value="if (*/*:control[@languageEncoding eq 'ietf-bcp-47'][not(@scriptEncoding)]) then true() else false()"/>
+    <sch:let name="check-language-codes" value="if (*/*:control/@languageEncoding = ('iso639-1', 'iso639-2b', 'iso639-3')) then true() else false()"/>
+    <sch:let name="check-ietf-codes" value="if (*/*:control[@languageEncoding eq 'ietf-bcp-47']) then true() else false()"/>
     <sch:let name="check-country-codes" value="if (*/*:control/@countryEncoding eq 'otherCountryEncoding') then false() else true()"/>
     <sch:let name="check-script-codes" value="if (*/*:control/@scriptEncoding eq 'otherScriptEncoding') then false() else true()"/>
     <sch:let name="check-repository-codes" value="if (*/*:control/@repositoryEncoding eq 'otherRepositoryEncoding') then false() else true()"/>
+    <sch:let name="check-date-attributes" value="if (*/*:control/@dateEncoding eq 'otherDateEncoding') then false() else true()"/>
+    
     
     <!-- VARIABLE iso15511Pattern -->
     <sch:let name="iso15511Pattern" value="'(^([A-Z]{2})|([a-zA-Z]{1})|([a-zA-Z]{3,4}))(-[a-zA-Z0-9:/\-]{1,11})$'"/>
@@ -47,7 +48,7 @@
             <sch:assert test="every $l in (@languageCode | @languageOfElement) satisfies matches(normalize-space($l), $ietfPattern)">The <sch:name/> element's lang or langcode attribute should contain a value from the 'ietf-bcp-47' codelist.</sch:assert>
         </sch:rule>
     </sch:pattern>
-     
+    
     <!-- COUNTRY CODES (in process) -->
     <sch:pattern>
         <sch:let name="countryCodes" value="document('iso_3166.xml')"/>
@@ -121,7 +122,7 @@
             </sch:assert>
         </sch:rule>
     </sch:pattern>
-       
+    
     <!-- CO-OCCURENCE CONSTRAINTS -->
     <sch:pattern id="maintenanceAgency-constraints">
         <sch:rule context="*:maintenanceAgency[*:agencyCode[not(normalize-space())]] | *:maintenanceAgency[not(*:agencyCode)]">
@@ -131,7 +132,7 @@
             <sch:assert test="*:agencyCode[normalize-space()]">The maintenanceAgency element requires either an agencyCode or agencyName element that cannot be empty.</sch:assert>
         </sch:rule>
     </sch:pattern>
-  
+    
     <sch:pattern id="eventDateTime">
         <sch:rule context="/*/*:control/*:maintenanceHistory/*:maintenanceEvent/*:eventDateTime[not(@standardDateTime)]">
             <sch:assert test="normalize-space()">The eventDateTime element requires either a standardDateTime attribute or text.</sch:assert>
@@ -146,16 +147,36 @@
     </sch:pattern>
     
     <!-- DATE NORMALIZATION -->
-    <!-- will need to update considerably, still.  iso 8601 2019 possiblities are quite different...
+    <!-- will need to update considerably.  iso 8601 2019 possiblities are quite different...
         also, we will need to be clear that we don't support all values of iso 8601.  
+        for instance, would need to add support for decade, week, dayo, dayk, etc.
+        as well as additional qualifiers, such as 
+            ~, %, X, ?
+        but likely should never try to add full support, unless we change the attribute names?  
+        also...
+        
+        Ranges (which negate our notAfter / notBefore attributes):
+            1978..1984
+        
+        etc.:
+            2052Y1MX*D
+        -->
+    <!-- might still want to add optional Y indicator, plus support for Years > and < 4 digits...
+        also need to state whether we support both basic and extended formats. currently we try to, but it does introduce invalid options right now.
         -->
     <sch:pattern id="dates">
-        <!-- without using the ETDF parts of ISO 8601 2019, years are capped as such, it looks like...-->
-        <sch:let name="isoYYYY" value="'\-?(0|1|2)([0-9]{3})'"/>
-        <sch:let name="isoMM" value="'\-?(01|02|03|04|05|06|07|08|09|10|11|12)'"/>
-        <sch:let name="isoDD" value="'\-?((0[1-9])|((1|2)[0-9])|(3[0-1]))'"/>
-        <sch:let name="isoPattern" value="concat('^', $isoYYYY, '$','|', '^', $isoYYYY, $isoMM,'$', '|', '^', $isoYYYY, $isoMM, $isoDD,'$')"/>
-        <sch:rule context="*:date[exists(@notBefore | @notAfter | @standardDate)] | *:toDate[exists(@notBefore | @notAfter | @standardDate)] | *:fromDate[exists(@notBefore | @notAfter | @standardDate)]">
+        <sch:let name="Months" value="1 to 12"/>
+        <sch:let name="Seasons" value="21 to 41"/>
+        <sch:let name="Y" value="'[+-]?([0-9u]{1}[0-9ux]{3})'"/>
+        <sch:let name="M" value="'-?(' || string-join(for $x in ($Months) return format-number($x, '00'), '|') || ')'"/>
+        <sch:let name="M_S" value="'-?(' || string-join(for $x in ($Months, $Seasons) return format-number($x, '00'), '|') || ')'"/>
+        <sch:let name="D" value="'-?((0[1-9])|((1|2)[0-9])|(3[0-1]))'"/>
+        <sch:let name="isoPattern" value="concat(
+            '^', $Y, '$','|'
+            , '^', $Y, $M_S,'$', '|'
+            , '^', $Y, $M, $D,'$'
+            )"/>
+        <sch:rule context="*:date[$check-date-attributes][exists(@notBefore | @notAfter | @standardDate)] | *:toDate[$check-date-attributes][exists(@notBefore | @notAfter | @standardDate)] | *:fromDate[$check-date-attributes][exists(@notBefore | @notAfter | @standardDate)]">
             <sch:assert test="every $d in (@notBefore, @notAfter, @standardDate) satisfies matches($d, $isoPattern)">The <sch:emph>notBefore</sch:emph>, <sch:emph>notAfter</sch:emph>, and <sch:emph>standardDate</sch:emph> attributes of <sch:name/> must be a iso8601 date.</sch:assert>
         </sch:rule>
     </sch:pattern>
diff --git a/source/release-info/intro.xml b/source/release-info/intro.xml
index cd9d3d8..be772b7 100644
--- a/source/release-info/intro.xml
+++ b/source/release-info/intro.xml
@@ -12,7 +12,7 @@
   *                                                               *                
   *                            EAC-CPF 2.0                        *
   *                                                               *
-  *                           2021 November 2                     *          
+  *                           2022 March 20                       *          
   *                                                               *
   *****************************************************************
 
@@ -37,6 +37,10 @@
   
   Comments, questions, and suggestions may be addressed to: 
   ts-eas@archivists.org
+  
+  EAC-CPF Team Lead: Silke Jagodzinski
+  
+  Schema Team Lead: Mark Custer
 
   *****************************************************************
   *****************************************************************
diff --git a/source/release-info/metadata.json b/source/release-info/metadata.json
index 3916ae2..3299862 100644
--- a/source/release-info/metadata.json
+++ b/source/release-info/metadata.json
@@ -1,7 +1,7 @@
 [
     {
         "name": "EAC-CPF",
-        "date": "2021-11-02",
+        "date": "2022-03-20",
         "lead": "Silke Jagodzinski",
         "team": [],
         "schema-team": [],
diff --git a/xml-schemas/eac-cpf/eac.rng b/xml-schemas/eac-cpf/eac.rng
index 42ba62b..dfa72bc 100644
--- a/xml-schemas/eac-cpf/eac.rng
+++ b/xml-schemas/eac-cpf/eac.rng
@@ -10,7 +10,7 @@
   *                                                               *                
   *                            EAC-CPF 2.0                        *
   *                                                               *
-  *                           2021 November 2                     *          
+  *                           2022 March 20                       *          
   *                                                               *
   *****************************************************************
 
@@ -35,6 +35,10 @@
   
   Comments, questions, and suggestions may be addressed to: 
   ts-eas@archivists.org
+  
+  EAC-CPF Team Lead: Silke Jagodzinski
+  
+  Schema Team Lead: Mark Custer
 
   *****************************************************************
   *****************************************************************
diff --git a/xml-schemas/eac-cpf/eac.xsd b/xml-schemas/eac-cpf/eac.xsd
index 9091d23..b0417bd 100644
--- a/xml-schemas/eac-cpf/eac.xsd
+++ b/xml-schemas/eac-cpf/eac.xsd
@@ -10,7 +10,7 @@
   *                                                               *                
   *                            EAC-CPF 2.0                        *
   *                                                               *
-  *                           2021 November 2                     *          
+  *                           2022 March 20                       *          
   *                                                               *
   *****************************************************************
   
@@ -36,6 +36,10 @@
   Comments, questions, and suggestions may be addressed to: 
   ts-eas@archivists.org
   
+  EAC-CPF Team Lead: Silke Jagodzinski
+  
+  Schema Team Lead: Mark Custer
+  
   *****************************************************************
   *****************************************************************
   
diff --git a/xml-schemas/eac-cpf/schematron/eac.sch b/xml-schemas/eac-cpf/schematron/eac.sch
index d18582d..6a4cf61 100644
--- a/xml-schemas/eac-cpf/schematron/eac.sch
+++ b/xml-schemas/eac-cpf/schematron/eac.sch
@@ -12,22 +12,23 @@
         <language value="iso639-1" filename="iso639-1.rdf"/>
         <language value="iso639-2b" filename="iso639-2.rdf"/>
         <language value="iso639-3" filename="iso639-3.xml"/>
-        <!-- ietf-bcp-47 -->
     </xsl:variable>
     
-    <!-- VARIABLE $language-code-key: the EAD3 document's /ead:ead/ead:control/@langencoding, with iso639-2b as a default value. -->
-    <sch:let name="active-language-code-key" value="(*/*:control/@languageEncoding[.=$supported-language-codes/@value], 'iso639-2b')[1]"/>
+    <!-- VARIABLE $language-code-key: the EAD3 document's /ead:ead/ead:control/@langencoding  -->
+    <sch:let name="active-language-code-key" value="(*/*:control/@languageEncoding[.=$supported-language-codes/@value])"/>
     
     <!-- still need to add something here to distinguish between active and deprecated codes -->
     <!-- and will also need a functional way, or ability to hit an API endpoint, to test for ietf-bcp-47 codes, most likely -->
     <sch:let name="valid-language-codes" value="document($supported-language-codes[@value = $active-language-code-key]/@filename)//(madsrdf:code | iso_639_3_entries/iso_639_3_entry/@id)"/>
     
     <!-- until we have a better way with dealing with "other" as a value, etc.-->
-    <sch:let name="check-language-codes" value="if (*/*:control/@languageEncoding = ('otherLanguageEncoding', 'ietf-bcp-47')) then false() else true()"/>
-    <sch:let name="check-ietf-codes" value="if (*/*:control[@languageEncoding eq 'ietf-bcp-47'][not(@scriptEncoding)]) then true() else false()"/>
+    <sch:let name="check-language-codes" value="if (*/*:control/@languageEncoding = ('iso639-1', 'iso639-2b', 'iso639-3')) then true() else false()"/>
+    <sch:let name="check-ietf-codes" value="if (*/*:control[@languageEncoding eq 'ietf-bcp-47']) then true() else false()"/>
     <sch:let name="check-country-codes" value="if (*/*:control/@countryEncoding eq 'otherCountryEncoding') then false() else true()"/>
     <sch:let name="check-script-codes" value="if (*/*:control/@scriptEncoding eq 'otherScriptEncoding') then false() else true()"/>
     <sch:let name="check-repository-codes" value="if (*/*:control/@repositoryEncoding eq 'otherRepositoryEncoding') then false() else true()"/>
+    <sch:let name="check-date-attributes" value="if (*/*:control/@dateEncoding eq 'otherDateEncoding') then false() else true()"/>
+    
     
     <!-- VARIABLE iso15511Pattern -->
     <sch:let name="iso15511Pattern" value="'(^([A-Z]{2})|([a-zA-Z]{1})|([a-zA-Z]{3,4}))(-[a-zA-Z0-9:/\-]{1,11})$'"/>
@@ -47,7 +48,7 @@
             <sch:assert test="every $l in (@languageCode | @languageOfElement) satisfies matches(normalize-space($l), $ietfPattern)">The <sch:name/> element's lang or langcode attribute should contain a value from the 'ietf-bcp-47' codelist.</sch:assert>
         </sch:rule>
     </sch:pattern>
-     
+    
     <!-- COUNTRY CODES (in process) -->
     <sch:pattern>
         <sch:let name="countryCodes" value="document('iso_3166.xml')"/>
@@ -121,7 +122,7 @@
             </sch:assert>
         </sch:rule>
     </sch:pattern>
-       
+    
     <!-- CO-OCCURENCE CONSTRAINTS -->
     <sch:pattern id="maintenanceAgency-constraints">
         <sch:rule context="*:maintenanceAgency[*:agencyCode[not(normalize-space())]] | *:maintenanceAgency[not(*:agencyCode)]">
@@ -131,7 +132,7 @@
             <sch:assert test="*:agencyCode[normalize-space()]">The maintenanceAgency element requires either an agencyCode or agencyName element that cannot be empty.</sch:assert>
         </sch:rule>
     </sch:pattern>
-  
+    
     <sch:pattern id="eventDateTime">
         <sch:rule context="/*/*:control/*:maintenanceHistory/*:maintenanceEvent/*:eventDateTime[not(@standardDateTime)]">
             <sch:assert test="normalize-space()">The eventDateTime element requires either a standardDateTime attribute or text.</sch:assert>
@@ -146,16 +147,36 @@
     </sch:pattern>
     
     <!-- DATE NORMALIZATION -->
-    <!-- will need to update considerably, still.  iso 8601 2019 possiblities are quite different...
+    <!-- will need to update considerably.  iso 8601 2019 possiblities are quite different...
         also, we will need to be clear that we don't support all values of iso 8601.  
+        for instance, would need to add support for decade, week, dayo, dayk, etc.
+        as well as additional qualifiers, such as 
+            ~, %, X, ?
+        but likely should never try to add full support, unless we change the attribute names?  
+        also...
+        
+        Ranges (which negate our notAfter / notBefore attributes):
+            1978..1984
+        
+        etc.:
+            2052Y1MX*D
+        -->
+    <!-- might still want to add optional Y indicator, plus support for Years > and < 4 digits...
+        also need to state whether we support both basic and extended formats. currently we try to, but it does introduce invalid options right now.
         -->
     <sch:pattern id="dates">
-        <!-- without using the ETDF parts of ISO 8601 2019, years are capped as such, it looks like...-->
-        <sch:let name="isoYYYY" value="'\-?(0|1|2)([0-9]{3})'"/>
-        <sch:let name="isoMM" value="'\-?(01|02|03|04|05|06|07|08|09|10|11|12)'"/>
-        <sch:let name="isoDD" value="'\-?((0[1-9])|((1|2)[0-9])|(3[0-1]))'"/>
-        <sch:let name="isoPattern" value="concat('^', $isoYYYY, '$','|', '^', $isoYYYY, $isoMM,'$', '|', '^', $isoYYYY, $isoMM, $isoDD,'$')"/>
-        <sch:rule context="*:date[exists(@notBefore | @notAfter | @standardDate)] | *:toDate[exists(@notBefore | @notAfter | @standardDate)] | *:fromDate[exists(@notBefore | @notAfter | @standardDate)]">
+        <sch:let name="Months" value="1 to 12"/>
+        <sch:let name="Seasons" value="21 to 41"/>
+        <sch:let name="Y" value="'[+-]?([0-9u]{1}[0-9ux]{3})'"/>
+        <sch:let name="M" value="'-?(' || string-join(for $x in ($Months) return format-number($x, '00'), '|') || ')'"/>
+        <sch:let name="M_S" value="'-?(' || string-join(for $x in ($Months, $Seasons) return format-number($x, '00'), '|') || ')'"/>
+        <sch:let name="D" value="'-?((0[1-9])|((1|2)[0-9])|(3[0-1]))'"/>
+        <sch:let name="isoPattern" value="concat(
+            '^', $Y, '$','|'
+            , '^', $Y, $M_S,'$', '|'
+            , '^', $Y, $M, $D,'$'
+            )"/>
+        <sch:rule context="*:date[$check-date-attributes][exists(@notBefore | @notAfter | @standardDate)] | *:toDate[$check-date-attributes][exists(@notBefore | @notAfter | @standardDate)] | *:fromDate[$check-date-attributes][exists(@notBefore | @notAfter | @standardDate)]">
             <sch:assert test="every $d in (@notBefore, @notAfter, @standardDate) satisfies matches($d, $isoPattern)">The <sch:emph>notBefore</sch:emph>, <sch:emph>notAfter</sch:emph>, and <sch:emph>standardDate</sch:emph> attributes of <sch:name/> must be a iso8601 date.</sch:assert>
         </sch:rule>
     </sch:pattern>