scripts/query.sh

   1 #!/bin/bash
   2
   3 # query.sh - Set of Bash functions to work with wdef files.
   4 # Copyright (C) 2022  Pierre Choffet
   5 #
   6 # This program is free software: you can redistribute it and/or modify
   7 # it under the terms of version 3 of the GNU General Public License as
   8 # published by the Free Software Foundation.
   9 #
  10 # This program is distributed in the hope that it will be useful,
  11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  13 # GNU General Public License for more details.
  14 #
  15 # You should have received a copy of the GNU General Public License
  16 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
  17
  18 set -euo pipefail
  19
  20 readonly WIKIDATA_SPARQL_ENDPOINT='https://query.wikidata.org/bigdata/namespace/wdq/sparql'
  21
  22 readonly CACHE_DIR=${CACHE_DIR:-"${HOME}/.cache/wdef_tools/"}
  23 readonly QUERY_CACHE_DIR="${CACHE_DIR}/queries/"
  24
  25 # Send query and return a path to the result into cache dir
  26 # Parameter:
  27 #   $1: Path to file containing query
  28 # Output:
  29 #   Path to the file containing result, in cache dir
  30 function query() {
  31         local -r query_path="${1}"
  32
  33         if [ ! -s "${query_path}" ]
  34         then
  35                 echo "Query not found at path ${query_path}. Exiting." >&2
  36                 return 1
  37         fi
  38
  39         local -r query_checksum=$(md5sum "${query_path}" | cut -d ' ' -f 1)
  40         local -r result_path="${QUERY_CACHE_DIR}/${query_checksum}"
  41
  42         # Prepare query
  43         local -r query=$(cat "${query_path}" | jq -sRr @uri)
  44
  45         # Send query, cache result
  46         mkdir -p "${QUERY_CACHE_DIR}"
  47         wget -O "${result_path}" "${WIKIDATA_SPARQL_ENDPOINT}?query=${query}"
  48
  49         echo "${result_path}"
  50 }
  51
  52 # Build query from template and a set of variables
  53 # Params:
  54 #   $1: Path to file containing query template
  55 #   $2: Associative array with variables values, indexed on their names
  56 # Output:
  57 #   Path to the file containing result, in cache dir
  58 function queryVariables() {
  59         local -r template_path="${1}"
  60         local -nr variables=${2}
  61
  62         # Check parameters
  63         if [ ! -s "${template_path}" ]
  64         then
  65                 echo "Query template not found at path ${template_path}. Exiting." >&2
  66                 return 1
  67         fi
  68         if [ ${#variables[@]} -eq 0 ]
  69         then
  70                 echo "Missing query variables. Exiting." >&2
  71                 return 1
  72         fi
  73
  74         local -r template_checksum=$(md5sum "${template_path}" | cut -d ' ' -f 1)
  75         local -r variables_checksum=$(echo "$(typeset -p ${2})" | md5sum - | cut -d ' ' -f 1)
  76
  77         # Build query
  78         local query=$(cat "${template_path}")
  79         for var_name in ${!variables[@]}
  80         do
  81                 query=${query//\%${var_name}\%/${variables[${var_name}]}}
  82         done
  83
  84         # Save query into a temporary file
  85         local -r query_path="$(mktemp)"
  86         echo "${query}" > "${query_path}"
  87
  88         # Check at least one variable has been substituted (file name collision otherwise)
  89         if [ "$(md5sum "${query_path}" | cut -d ' ' -f 1)" == "${template_checksum}" ]
  90         then
  91                 echo "No variable substituted. Exiting." >&2
  92                 return 1
  93         fi
  94
  95         # Send query
  96         local -r query_result_path=$(query "${query_path}")
  97
  98         # Generate query result cache path, create dir if needed
  99         local -r query_cache_path="${QUERY_CACHE_DIR}/${template_checksum}/${variables_checksum}"
 100         mkdir -p "$(dirname "${query_cache_path}")"
 101
 102         # Link result
 103         if [ ! -f "${query_cache_path}" ]
 104         then
 105                 ln "${query_result_path}" "${query_cache_path}"
 106         fi
 107
 108         # Remove query
 109         rm "${query_path}"
 110
 111         # Return result path
 112         echo "${query_cache_path}"
 113 }