init commit

.github/workflows/arche.yml

@@ -0,0 +1,43 @@
+name: ARCHE-Ingest
+
+on:
+  workflow_dispatch
+
+jobs:
+  build_pages:
+    name: Ingest Data into ARCHE
+    runs-on: ubuntu-latest
+    env:
+      ANT_OPTS: -Xmx5g
+    steps:
+    - name: Perform Checkout
+      uses: actions/checkout@v2
+    - name: Install Saxon, Ant and Fundament
+      run: |
+        apt-get update && apt-get install openjdk-11-jre-headless ant -y --no-install-recommend
+        ./script.sh
+    - name: Build
+      run: |
+        ant -f build_arche.xml
+    - name: ingestion dependencies
+      run: |
+        composer require "acdh-oeaw/arche-ingest:^1"
+    - name: ingest resources
+      run: |
+        vendor/bin/arche-import-metadata html/arche.rdf https://arche-dev.acdh-dev.oeaw.ac.at/api ${{secrets.ARCHE_LOGIN}} ${{secrets.ARCHE_PASSWORD}}--retriesOnConflict 25
+    - name: Copy XML Files to to_ingest 
+      run: |
+        mkdir $PWD/to_ingest
+        find -path "./data/editions/*.xml" -exec cp -prv '{}' './to_ingest' ';'
+        find -path "./data/indices/*.xml" -exec cp -prv '{}' './to_ingest' ';'
+        find -path "./data/meta/*.xml" -exec cp -prv '{}' './to_ingest' ';'
+    - name: install repo-file-checker
+      run: |
+        composer require "acdh-oeaw/repo-file-checker:^2"
+        mkdir -p filechecker/tmp filechecker/reports
+    - name: run repo-file-checker
+      run: |
+        php -f vendor/acdh-oeaw/repo-file-checker/index.php -- --tmpDir filechecker/tmp --reportDir filechecker/reports to_ingest 0
+    - name: ingest binaries
+      run: |
+        vendor/bin/arche-import-binary to_ingest https://id.acdh.oeaw.ac.at/rita-static https://arche-dev.acdh-dev.oeaw.ac.at/api ${{secrets.ARCHE_LOGIN}} ${{secrets.ARCHE_PASSWORD}} --skip not_exist

.github/workflows/build.yml

@@ -0,0 +1,33 @@
+name: Build and publish
+
+on: 
+  push:
+
+jobs:
+  build_pages:
+    name: Publish Edition as GitHub Pages Website
+    runs-on: ubuntu-latest
+    env:
+      ANT_OPTS: -Xmx5g
+    steps:
+    - name: Perform Checkout
+      uses: actions/checkout@v2
+    - name: Install Saxon, Ant and Fundament
+      run: |
+        apt-get update && apt-get install openjdk-11-jre-headless ant -y --no-install-recommend
+        ./script.sh
+    - name: Install Ant-Contrib
+      run: |
+        wget https://repo1.maven.org/maven2/ant-contrib/ant-contrib/1.0b3/ant-contrib-1.0b3.jar
+        mv ant-contrib-1.0b3.jar /usr/share/ant/lib
+    - name: Build
+      run: |
+        ant
+    - name: Build Index
+      run: |
+        ./build_index.sh
+    - name: Deploy
+      uses: peaceiris/actions-gh-pages@v3
+      with:
+        github_token: ${{secrets.GITHUB_TOKEN}}
+        publish_dir: ./html

.gitignore

@@ -0,0 +1,152 @@
+# project specific
+html/*.html
+html/search.html
+/saxon
+data/imprint.xml
+html/*.rdf
+html/*.xml
+html/*.txt
+/static-search
+html/static-search
+.env
+data/editions
+data/indices
+data/meta
+data/inventare
+data/rita1
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# Windows
+desktop.ini
+
+# MacOS
+.DS_Store

LICENSE

@@ -0,0 +1,2 @@
+CC BY 4.0
+https://creativecommons.org/licenses/by/4.0/legalcode

README.md

@@ -0,0 +1,4 @@
+# Privater Buchbesitz im katholisch dominierten Zentralalpenraum 1750–1800. Eine systematische Untersuchung anhand von Inventaren aus dem Pustertal und dem Stubaital
+
+
+* build with [DSE-Static-Cookiecutter](https://github.com/acdh-oeaw/dse-static-cookiecutter)

build.xml

@@ -0,0 +1,67 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project basedir="." name="tei2html">
+    <property name="index" value="./data/meta/about.xml"/>
+    <property name="target" value="./html"/>
+    <property name="target_xml" value="./html/xml-view"/>
+    <delete>
+        <fileset dir="${target}" includes="*.html"/>
+    </delete>
+    <delete dir="${target}/static-search"/>
+    <!-- <xslt style="./xslt/editions-xml.xsl"  basedir="./data/editions" destdir="${target_xml}" includes="*.xml">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt> -->
+    <xslt style="./xslt/editions.xsl"  basedir="./data/editions" destdir="${target}" includes="*.xml">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt>
+    <xslt style="./xslt/meta.xsl"  basedir="./data/meta" destdir="${target}" includes="*.xml">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt>
+    <xslt in="${index}" out="${target}/index.html" style="./xslt/index.xsl">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt>
+    <xslt in="${index}" out="${target}/toc.html" style="./xslt/toc.xsl">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt>
+    <xslt in="${index}" out="${target}/search.html" style="./xslt/search.xsl">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt>
+    <xslt in="./data/imprint.xml" out="${target}/imprint.html" style="./xslt/imprint.xsl">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt>
+    <xslt in="./data/indices/listperson.xml" out="${target}/listperson.html" style="./xslt/listperson.xsl">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt>
+    <xslt in="./data/indices/listplace.xml" out="${target}/listplace.html" style="./xslt/listplace.xsl">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt>
+    <xslt in="./data/indices/listorg.xml" out="${target}/listorg.html" style="./xslt/listorg.xsl">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt>
+    <xslt in="./data/meta/arche_constants.rdf" out="${target}/arche.rdf" style="./xslt/arche.xsl">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt>
+    <xslt in="./data/indices/listperson.xml" out="${target}/beacon.txt" style="./xslt/beacon.xsl">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt>
+    <replace dir="${target}" value="">
+        <include name="*.html"/>
+        <replacetoken> xmlns=""</replacetoken>
+    </replace>
+    <copy todir="${target}" flatten="true">
+        <fileset dir="./data/">
+            <include name="**/*.xml"/>
+        </fileset>
+    </copy>
+</project>

build_arche.xml

@@ -0,0 +1,8 @@
+
+<?xml version="1.0" encoding="UTF-8"?>
+<project basedir="." name="tei2html">
+    <xslt in="./data/meta/arche_constants.rdf" out="./html/arche.rdf" style="./xslt/arche.xsl">
+        <factory name="net.sf.saxon.TransformerFactoryImpl"/>
+        <classpath location="${basedir}/saxon/saxon9he.jar"/>
+    </xslt>
+</project>

build_index.sh

@@ -0,0 +1,2 @@
+echo "build index"
+ant -f ./static-search/build.xml -DssConfigFile=${PWD}/ss_config.xml

dl_fundament.sh

@@ -0,0 +1,8 @@
+# bin/bash
+
+echo "download and copy fundament"
+FUNDAMENT_VERSION=1.2.1
+FUNDAMENT_GIT=https://github.com/acdh-oeaw/fundament/archive/refs/tags/
+wget ${FUNDAMENT_GIT}v${FUNDAMENT_VERSION}.zip && unzip v${FUNDAMENT_VERSION}.zip && rm -rf v${FUNDAMENT_VERSION}.zip
+cp -r fundament-${FUNDAMENT_VERSION}/dist  html/
+rm -rf fundament-${FUNDAMENT_VERSION}

dl_imprint.sh

@@ -0,0 +1,9 @@
+# bin/bash
+
+REDMINE_ID=6930
+IMPRINT_XML=./data/imprint.xml
+rm ${IMPRINT_XML}
+echo '<?xml version="1.0" encoding="UTF-8"?>'
+echo "<root>" >> ${IMPRINT_XML}
+curl https://shared.acdh.oeaw.ac.at/acdh-common-assets/api/imprint.php?serviceID=${REDMINE_ID} >> ${IMPRINT_XML}
+echo "</root>" >> ${IMPRINT_XML}

dl_saxon.sh

@@ -0,0 +1,2 @@
+echo "downloading saxon"
+wget https://sourceforge.net/projects/saxon/files/Saxon-HE/9.9/SaxonHE9-9-1-7J.zip/download && unzip download -d saxon && rm -rf download

dl_staticsearch.sh

@@ -0,0 +1,11 @@
+# bin/bash
+
+echo "download static search"
+rm -rf ./static-search
+rm -rf ./tmp
+wget https://github.com/projectEndings/staticSearch/archive/refs/tags/v1.4.1.zip && unzip v1.4.1.zip -d tmp && rm v1.4.1.zip
+mv ./tmp/staticSearch-1.4.1 ./static-search && rm -rf ./tmp
+
+echo "get stopword list"
+wget https://raw.githubusercontent.com/stopwords-iso/stopwords-de/master/stopwords-de.txt -O stopwords.txt
+touch words.txt

fetch_data.sh

@@ -0,0 +1,24 @@
+rm -rf rita2-data-master && rm master.zip
+rm -rf data/editions && rm -rf data/indices && rm -rf data/meta
+rm -rf data/rita1 && rm -rf data/inventare
+wget https://github.com/reading-in-the-alps/rita2-data/archive/refs/heads/master.zip
+unzip master
+mv rita2-data-master/data/editions ./data/editions
+mv rita2-data-master/data/indices ./data/indices
+mv rita2-data-master/data/meta ./data/meta
+mv rita2-data-master/data/inventare ./data/inventare
+mv rita2-data-master/data/rita1 ./data/rita1
+
+
+./dl_imprint.sh
+rm -rf rita2-data-master && rm master.zip
+
+# echo "create calendar data"
+# python make_calendar_data.py
+add-attributes -g "./data/editions/*.xml" -b "https://id.acdh.oeaw.ac.at/rita/editions/"
+add-attributes -g "./data/meta/*.xml" -b "https://id.acdh.oeaw.ac.at/rita/meta/"
+add-attributes -g "./data/rita1/*.xml" -b "https://id.acdh.oeaw.ac.at/rita/rita1/"
+add-attributes -g "./data/inventare/*.xml" -b "https://id.acdh.oeaw.ac.at/rita/inventare/"
+add-attributes -g "./data/indices/*.xml" -b "https://id.acdh.oeaw.ac.at/rita/indices/"
+
+denormalize-indices -x './/tei:title[@type="short"]/text()' -i './data/indices/list*.xml' -f './data/editions/*.xml'