First commit

.github/workflows/pr.yml

@@ -0,0 +1,45 @@
+on:
+  push:
+    branches:
+      - 'ai-stuff'
+  pull_request:
+    branches:
+      - 'ai-stuff'
+
+name: Review
+
+jobs:
+  changelog:
+    runs-on: ubuntu-latest
+    name: Changelog should be updated
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 2
+
+      - name: Git fetch
+        run: git fetch
+
+      - name: Check that changelog has been updated.
+        run: git diff --exit-code origin/${{ github.base_ref }} -- CHANGELOG.md && exit 1 || exit 0
+
+  coding-standards-markdown:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Coding standards
+        run: |
+          docker run --rm --volume "$PWD:/md" peterdavehello/markdownlint markdownlint '**/*.md'
+
+  coding-standards-shellcheck:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Coding standards
+        run: |
+          docker run --rm --volume "$PWD:/mnt" koalaman/shellcheck:stable */*.sh

.markdownlint.jsonc

@@ -0,0 +1,18 @@
+{
+  "default": true,
+  // https://github.com/DavidAnson/markdownlint/blob/main/doc/md013.md
+  "line-length": {
+    "line_length": 120,
+    "code_blocks": false,
+    "tables": false
+  },
+  // https://github.com/DavidAnson/markdownlint/blob/main/doc/md024.md
+  "no-duplicate-heading": {
+    "siblings_only": true
+  },
+  // https://docs.github.com/en/get-started/writing-on-github/working-with-advanced-formatting/organizing-information-with-collapsed-sections#creating-a-collapsed-section
+  // https://github.com/DavidAnson/markdownlint/blob/main/doc/md033.md
+  "no-inline-html": {
+    "allowed_elements": ["details", "summary"]
+  }
+}

CHANGELOG.md

@@ -0,0 +1,14 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]
+
+### Added
+
+- Data export script
+
+[Unreleased]: https://github.com/itk-dev/os2loop/tree/ai-stuff

README.md

@@ -0,0 +1,14 @@
+# OS2Loop AI stuff
+
+``` shell
+git clone --branch ai-stuff https://github.com/itk-dev/os2loop os2loop-ai-stuff
+
+# Run a script, e.g.
+./os2loop-ai-stuff/data-export/export.sh
+```
+
+## Development
+
+``` shell
+task
+```

Taskfile.yml

@@ -0,0 +1,33 @@
+version: '3'
+
+tasks:
+  default:
+    cmds:
+        - task --list
+    silent: true
+
+  coding-standards:check:
+    desc: "Apply coding standards and run checks"
+    cmds:
+      - task: coding-standards:apply
+      - task: coding-standards:check:shellcheck
+
+  coding-standards:apply:
+    desc: "Apply coding standards"
+    cmds:
+      - task: coding-standards:apply:markdownlint
+
+  coding-standards:apply:markdownlint:
+    desc: "Run markdownlint-cli (https://github.com/igorshubovych/markdownlint-cli)"
+    cmds:
+      - docker run --rm --volume "$PWD:/md" peterdavehello/markdownlint markdownlint '**/*.md' --fix
+
+  coding-standards:check:markdownlint:
+    desc: "Run markdownlint-cli (https://github.com/igorshubovych/markdownlint-cli)"
+    cmds:
+      - docker run --rm --volume "$PWD:/md" peterdavehello/markdownlint markdownlint '**/*.md'
+
+  coding-standards:check:shellcheck:
+    desc: "Run ShellCheck (https://github.com/koalaman/shellcheck)"
+    cmds:
+      - docker run --rm --volume "$PWD:/mnt" koalaman/shellcheck:stable */*.sh

data-export/.gitignore

@@ -0,0 +1,2 @@
+*.csv
+*.json

data-export/export.sh

@@ -0,0 +1,74 @@
+#!/usr/bin/env bash
+set -o errexit -o errtrace -o noclobber -o nounset -o pipefail
+IFS=$'\n\t'
+
+script_dir=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+
+function usage() {
+    if [ -n "${1:-}" ]; then
+    >&2 cat <<EOF
+$1
+
+EOF
+    fi
+
+    >&2 cat <<EOF
+Usage: ${BASH_SOURCE[0]} project-dir site-uri
+
+EOF
+    exit 1
+}
+
+if (( $# < 2 )); then
+    usage "Too few arguments"
+fi
+
+project_dir="$1"
+uri="$2"
+
+if [ -z "$project_dir" ]; then
+    usage "Invalid project directory"
+fi
+
+if [ ! -d "$project_dir" ] ; then
+    (>&2 echo 'Project directory "'"$project_dir"'" does not exist')
+    exit 1
+fi
+
+if [ -z "$uri" ]; then
+    usage "Invalid site-uri"
+fi
+
+cd "$project_dir"
+
+filenames=("$script_dir"/export_*.sql)
+
+for filename in "${filenames[@]}"; do
+    echo "$filename"
+
+    # JSON
+
+    # https://tldp.org/LDP/abs/html/string-manipulation.html
+    output_filename=${filename/%.sql/.json}
+    # https://github.com/drush-ops/drush/issues/3071#issuecomment-347929777
+    vendor/bin/drush --uri="$uri" php:eval "return \Drupal::database()->query(file_get_contents('$filename'))->fetchAll()" --format=json >| "$output_filename" || true
+    echo "$output_filename"
+
+    # CSV
+
+    output_filename=${filename/%.sql/.csv}
+    # https://stackoverflow.com/a/22421445/2502647
+    vendor/bin/drush --uri="$uri" sql:cli < "$filename" | awk 'BEGIN { FS="\t"; OFS="," } {
+  rebuilt=0
+  for(i=1; i<=NF; ++i) {
+    if ($i ~ /,/ && $i !~ /^".*"$/) {
+      gsub("\"", "\"\"", $i)
+      $i = "\"" $i "\""
+      rebuilt=1
+    }
+  }
+  if (!rebuilt) { $1=$1 }
+  print
+}' >| "$output_filename" || true
+    echo "$output_filename"
+done

data-export/export_document_collections_with_meta.sql

@@ -0,0 +1,63 @@
+Select
+  n_fd.nid,
+  n_fd.type,
+  n_fd.title,
+  pa.`alias` as relative_url,
+  DATE_FORMAT(FROM_UNIXTIME(n_fd.created), '%Y-%m-%dT%H:%i:%s') as created,
+  DATE_FORMAT(FROM_UNIXTIME(n_fd.changed), '%Y-%m-%dT%H:%i:%s') as `changed`,
+  dci.document_node_ids,
+  dci.document_relative_urls,
+  docscol_content.os2loop_documents_dc_content_value as content, -- all format are rich text (html and div encoded)
+  docs_ib.os2loop_documents_info_box_value as info_box,
+  approval_date.os2loop_shared_approval_date_value as approval_date,
+  `subject`.`name` as `subject`,
+  tags.tags,
+  `owner`.os2loop_shared_owner_value as `owner`,
+  rev_date.os2loop_shared_rev_date_value as review_date,
+  `version`.os2loop_shared_version_value as `version`
+from (
+  SELECT nid,vid,type,uid,title,created,changed
+  FROM node_field_data
+  where type = 'os2loop_documents_collection'
+    -- the table os2loop_documents_collection_item associate document collections (their nid on collection_id)
+  -- to documents (document_id = nid) except for 20 document collections. Fx case /rammedelegation nid=3807 it is a collection of
+  -- links to sharepoint docs and /medicinhaandtering nid=3827 is a link to the collection
+    -- /instruks-korrekt-haandtering-af-medicin-i-sundhed-og-omsorg-mso nid 4188
+    -- of 805 documents 164 documents are not assigned to a document_collection
+  ) as n_fd
+left join path_alias as pa on CONCAT('/node/',n_fd.nid) = pa.path
+left join (
+  SELECT
+    doc_col_itm.collection_id,
+    json_arrayagg(doc_col_itm.document_id) as document_node_ids,
+    json_arrayagg(pa.`alias`) as document_relative_urls
+  from os2loop_documents_collection_item as doc_col_itm
+  left join path_alias as pa on CONCAT('/node/',doc_col_itm.document_id) = pa.path
+  group by doc_col_itm.collection_id
+  ) as dci on n_fd.nid = dci.collection_id
+left join node__os2loop_documents_dc_content as docscol_content on n_fd.nid = docscol_content.entity_id -- contains only records from bundle documents_collection (all delta 0, so top placement)
+left join (
+  SELECT
+  entity_id,
+  os2loop_documents_info_box_value
+  FROM node__os2loop_documents_info_box
+  WHERE bundle = 'os2loop_documents_collection') as docs_ib on n_fd.nid = docs_ib.entity_id  -- only from bundle document_collection
+left join node__os2loop_shared_approval_date as approval_date on n_fd.nid = approval_date.entity_id
+left join (
+  SELECT
+    n_ss.entity_id,
+    subject_tt_fd.name
+  FROM node__os2loop_shared_subject as n_ss
+  left join taxonomy_term_field_data as subject_tt_fd on n_ss.os2loop_shared_subject_target_id = subject_tt_fd.tid
+  where n_ss.bundle = 'os2loop_documents_collection') as `subject` on n_fd.nid = `subject`.entity_id
+left join node__os2loop_shared_owner as `owner` on n_fd.nid = `owner`.entity_id
+left join node__os2loop_shared_rev_date as rev_date on n_fd.nid = rev_date.entity_id
+left join (
+  SELECT
+    n_st.entity_id,
+        json_arrayagg(tt_fd.name) as tags
+  FROM node__os2loop_shared_tags  as n_st
+  left join taxonomy_term_field_data as tt_fd on n_st.os2loop_shared_tags_target_id = tt_fd.tid
+  where n_st.bundle = 'os2loop_documents_collection'
+  group by  n_st.entity_id) tags on n_fd.nid = tags.entity_id
+left join node__os2loop_shared_version as `version` on n_fd.nid = `version`.entity_id