mdouchin/AuditQGISProjects.py

## README.md

      
    Raw
  

              README.md
            
          
    QGIS projects audit

Presentation

This QGIS processing algorithm scans a local directory,
finds all the QGIS projects,
fetches a bunch of information for each project,
and save them in a CSV (Comma Separated Values) file.
It also can generate 2 other optional CSV files:

one containing the distinct layers datasources (QGIS internal layers definition) and their referencing projects.
one containing the distinct PostgreSQL tables in the format "schema_name"."table_name" and their referencing projects

It is only usable with a QGIS version >= 3.16.
The properties fetched are:
        fieldnames = [
            'path', 'basename', 'last_modified', 'crs',
            'layer_count', 'invalid_layer_count', 'memory_usage_mb',
            'print_layout_count', 'print_layout_pictures_paths', 'print_layout_pictures_sizes',
            'print_layout_pictures_total_size', 'print_layout_pictures_total_size_mb',
            'trust_option_active', 'last_save_version'
        ]
Installation

To install it in your computer, you need to:

download the script AuditQGISProjects.py manually or with a CLI tool

# Get the algorithm
wget "https://gist.githubusercontent.com/mdouchin/cd89a259d3560635d4bb8708ad430caa/raw/AuditQGISProjects.py" \
     -O ~/.local/share/QGIS/QGIS3/profiles/default/processing/scripts/AuditQGISProjects.py

put in in your user profile, inside the processing/scripts/ folder,
and then restart QGIS Desktop.

NB: The script tries to calculate the memory used by QGIS to load each project. We rely on the python module psutil. This is not very robust but can give an rough idea of the heavy projects.
Usage

Use inside QGIS desktop

Open the processing toolbox, and search for "audit", or find the algorithm inside the Scripts entry, named Audit QGIS projects

Use it in command line with qgis_process

In the recent version of QGIS, you can use a command line binary called qgis_process to run a processing algorithms (or script).
This is very handy ! In the example below, replace the fake projects root directory and output csv path with real paths.
# Run the script
qgis_process run script:audit_qgis_projects -- \
    PROJECTS_ROOT_DIRECTORY="~/The_directory/you_want/to_audit/" \
    OUTPUT_CSV_PROJECTS_PATH="/tmp/output_projects.csv" \
    OUTPUT_CSV_LAYERS_PATH="/tmp/output_layers.csv" \
    OUTPUT_CSV_TABLES_PATH="/tmp/output_postgresql_tables.csv" \    
    DONT_RESOLVE_LAYERS=True \
    GET_LIZMAP_CONFIG_PROPERTIES=True
Use "q" tool to easily query the output CSV file with SQL

q is a Python CLI tool which allows to query data contained in a CSV file. We can use it to have a closer look to the data. Obviously, you can open it with LibreOffice or any other software capable of reading CSV files (but it is less fun). More info on q: http://harelba.github.io/q/
# Install
sudo apt install -y python3-q-text-as-data

# Go to the directory containing the output CSV file (output.csv)
cd /tmp/

# Count the QGIS projects
q -H -d "," "SELECT count(*) FROM output.csv" -T -O -b

# Find projects by name (here with a name containing cats)
q -H -d "," "SELECT path, layer_count FROM output.csv WHERE basename LIKE '%cats%' " -O -T -b

# Count the projects with no Lizmap configuration file
q -H -d "," "SELECT count(*) FROM output.csv WHERE has_lizmap_config = 'False'" -T -O -b

# Count the projects group by the QGIS version
q -H -d "," "SELECT substr(last_save_version,1,3) AS version, count(*) AS nb FROM output.csv WHERE has_lizmap_config = 'True' GROUP BY version ORDER BY version" -T -O -b

# Find the projects modified before the 1st of january
q -H -d "," "SELECT path FROM output.csv WHERE last_modified > '2021-01-01' " -O -T -b

# Count the project by trust option status
q -H -d "," "SELECT trust_option_active, count(path) AS nb FROM output.csv GROUP BY trust_option_active" -O -T -b

# Find the 50 projects containing the higher layers count, ordered by the layers count (descending)
q -H -d "," "SELECT path, layer_count FROM output.csv WHERE has_lizmap_config = 'True' ORDER BY layer_count DESC LIMIT 50" -O -T -b

# Show the project by memory usage in descending order
q -H -d "," "SELECT path, memory_usage_mb FROM output.csv WHERE has_lizmap_config = 'True' ORDER BY memory_usage_mb DESC LIMIT 50" -O -T -b

# Show the potential full memory usage if every project is loaded. 
# You can multiply by a number of QGIS Server instances (here 8 instances)
q -H -d "," "SELECT sum(memory_usage_mb) AS total_one, sum(memory_usage_mb) * 8 AS total_full FROM output.csv WHERE has_lizmap_config = 'True'" -O -T -b

# Find the projects containing heavy images in the print composers (layouts)
# See: https://github.com/qgis/QGIS/issues/43787
q -H -d "," "SELECT path, print_layout_pictures_total_size_mb FROM output.csv WHERE has_lizmap_config = 'True' AND print_layout_pictures_total_size_mb > 1.0 ORDER BY print_layout_pictures_total_size_mb DESC" -O -T -b

  
## AuditQGISProjects.py
# -*- coding: utf-8 -*-
"""
***************************************************************************
*                                                                         *
*   This program is free software; you can redistribute it and/or modify  *
*   it under the terms of the GNU General Public License as published by  *
*   the Free Software Foundation; either version 2 of the License, or     *
*   (at your option) any later version.                                   *
*                                                                         *
***************************************************************************
"""

__author__ = 'Michaël Douchin'

import os
from pathlib import Path
from sys import platform
import re
import csv
import json
from glob import glob
from xml.etree import cElementTree as et
from xml.etree.ElementTree import ParseError as et_error

from qgis.PyQt.QtCore import QCoreApplication
from qgis.core import (
    QgsProject,
    QgsDataSourceUri,
    QgsLayoutItemPicture,
    QgsSettings,
    QgsProcessingAlgorithm,
    QgsProcessingParameterString,
    QgsProcessingParameterFile,
    QgsProcessingParameterFileDestination,
    QgsProcessingParameterBoolean
)


class AuditQGISProjects(QgsProcessingAlgorithm):
    """
    List all the QGIS projects of a directory and fetch key datas
    """
    PROJECTS_ROOT_DIRECTORY = 'PROJECTS_ROOT_DIRECTORY'
    OUTPUT_CSV_PROJECTS_PATH = 'OUTPUT_CSV_PROJECTS_PATH'
    OUTPUT_CSV_LAYERS_PATH = 'OUTPUT_CSV_LAYERS_PATH'
    OUTPUT_CSV_TABLES_PATH = 'OUTPUT_CSV_TABLES_PATH'
    DONT_RESOLVE_LAYERS = 'DONT_RESOLVE_LAYERS'
    LAYER_NAME_FILTER = 'LAYER_NAME_FILTER'
    LAYER_DATASOURCE_FILTER = 'LAYER_DATASOURCE_FILTER'
    GET_LIZMAP_CONFIG_PROPERTIES = 'GET_LIZMAP_CONFIG_PROPERTIES'

    def tr(self, string):
        """
        Returns a translatable string with the self.tr() function.
        """
        return QCoreApplication.translate('Processing', string)

    def createInstance(self):
        return AuditQGISProjects()

    def name(self):
        return 'audit_qgis_projects'

    def displayName(self):
        return self.tr('Audit QGIS projects')

    def group(self):
        return self.tr('Projects')

    def groupId(self):
        return 'projects'

    def shortHelpString(self):
        return self.tr(
            'Fetch the properties of all the QGIS *.qgs projects'
            ' found in a directory and its subdirectories.'
            '\n'
            'You can also find specific projects matching layers name or datasources.'
            '\n'
            'This algorithm creates CSV file(s) containing the list of projects properties'
            ' and optionally the list of layers (unique datasources and related projects)'
        )

    def helpUrl(self):
        url = 'https://gist.github.com/mdouchin/cd89a259d3560635d4bb8708ad430caa#file-readme-md'
        return url

    def initAlgorithm(self, config=None):
        # Settings
        s = QgsSettings()

        # Root directory
        projects_root_directory = s.value("lizmap/audit_projects_root_directory", Path.home())
        input_param = QgsProcessingParameterFile(
            self.PROJECTS_ROOT_DIRECTORY,
            self.tr('Local directory containing projects'),
            defaultValue=projects_root_directory,
            behavior=QgsProcessingParameterFile.Folder,
            optional=False
        )
        input_param.setHelp(self.tr(
            'The QGIS projects contained in this directory and sub-directories'
            ' will be analysed to gather their key parameters.'
        ))
        self.addParameter(input_param)

        # CSV file to create
        output_csv_projects_path = s.value(
            "lizmap/audit_output_csv_projects_path",
            os.path.join(Path.home(), 'qgis_projects_audit.csv')
        )
        input_param = QgsProcessingParameterFileDestination(
            self.OUTPUT_CSV_PROJECTS_PATH,
            self.tr('Output CSV file with the list of projects and their properties'),
            fileFilter='*.csv',
            optional=False,
            defaultValue=output_csv_projects_path
        )
        input_param.setHelp(self.tr(
            'This algorithm will create a Comma Separated Values (CSV) file'
            ' containing the list of projects with their properties.'
            ' Please choose the path.'
        ))
        self.addParameter(input_param)

        # CSV file containing unique datasources
        output_csv_layers_path = s.value(
            "lizmap/audit_output_csv_layers_path",
            os.path.join(Path.home(), 'qgis_layers_audit.csv')
        )
        input_param = QgsProcessingParameterFileDestination(
            self.OUTPUT_CSV_LAYERS_PATH,
            self.tr('Output CSV file with the list of distinct layers datasources'),
            fileFilter='*.csv',
            optional=True,
            defaultValue=output_csv_layers_path
        )
        input_param.setHelp(self.tr(
            'This algorithm will create an optional Comma Separated Values (CSV) file'
            ' containing the list of distinct datasources of layers and their projects'
        ))
        self.addParameter(input_param)

        # CSV file containing unique PostgreSQL tables
        output_csv_tables_path = s.value(
            "lizmap/audit_output_csv_tables_path",
            os.path.join(Path.home(), 'qgis_tables_audit.csv')
        )
        input_param = QgsProcessingParameterFileDestination(
            self.OUTPUT_CSV_TABLES_PATH,
            self.tr('Output CSV file with the list of distinct PostgreSQL tables'),
            fileFilter='*.csv',
            optional=True,
            defaultValue=output_csv_tables_path
        )
        input_param.setHelp(self.tr(
            'This algorithm will create an optional Comma Separated Values (CSV) file'
            ' containing the list of distinct PostgreSQL tables and their projects'
        ))
        self.addParameter(input_param)

        # Choose to resolve layers, to find invalid layers
        # If set to True, this would be much longer
        # And will not work out of QGIS Desktop
        # See https://github.com/qgis/QGIS/issues/34408
        dont_resolve_layers = s.value("lizmap/audit_dont_resolve_layers", True)
        if dont_resolve_layers in ('True', 'False'):
            dont_resolve_layers = eval(dont_resolve_layers)
        input_param = QgsProcessingParameterBoolean(
            self.DONT_RESOLVE_LAYERS,
            self.tr('Do not resolve layers'),
            defaultValue=dont_resolve_layers,
            optional=False
        )
        input_param.setHelp(self.tr(
            'If checked, the algorithm will load every layer of every project'
            ' to find invalid layers.'
            ' This could be very long depending on the number of projects,'
            ' and layers, and should be used only in QGIS Desktop context'
        ))
        self.addParameter(input_param)

        # Filter projects containing layer names
        layer_name_filter = s.value("lizmap/audit_layer_name_filter", '')
        input_param = QgsProcessingParameterString(
            self.LAYER_NAME_FILTER,
            self.tr('Filter projects with layers names like...'),
            defaultValue=layer_name_filter,
            optional=True
        )
        input_param.setHelp(self.tr(
            'If this input is not empty, the output CSV will list only'
            ' the projects containing layers corresponding to the written text.'
            ' You can write a full name or an extract.'
        ))
        self.addParameter(input_param)

        # Filter projects containing layer datasources
        layer_datasource_filter = s.value("lizmap/audit_layer_datasource_filter", '')
        input_param = QgsProcessingParameterString(
            self.LAYER_DATASOURCE_FILTER,
            self.tr('Filter projects with layers XML definition containing...'),
            defaultValue=layer_datasource_filter,
            optional=True
        )
        input_param.setHelp(self.tr(
            'If this input is not empty, the output CSV will list only'
            ' the projects containing layers corresponding to the written datasource.'
            ' You can write a full datasource or an extract.'
        ))
        self.addParameter(input_param)

        # Lizmap related data
        # If set to True, the script will look for a */qgs.cfg file
        # and fetch more properties from it
        get_lizmap_config_properties = s.value("lizmap/audit_get_lizmap_config_properties", False)
        if get_lizmap_config_properties in ('True', 'False'):
            get_lizmap_config_properties = eval(get_lizmap_config_properties)
        input_param = QgsProcessingParameterBoolean(
            self.GET_LIZMAP_CONFIG_PROPERTIES,
            self.tr('Get Lizmap configuration properties (from Lizmap *.cfg file'),
            defaultValue=get_lizmap_config_properties,
            optional=False
        )
        input_param.setHelp(self.tr(
            'If checked, the algorithm will search for a Lizmap *.cfg file'
            ' and fetch more properties from it.'
        ))
        self.addParameter(input_param)

    def getMemoryUsage(self):
        has_psutil = True
        used_memory = None
        used_memory_mb = None
        # Use psutil if available
        # else on linux use quick&dirty os call
        try:
            import psutil
        except ImportError:
            has_psutil = False

        if has_psutil:
            used_memory = psutil.virtual_memory().used
        else:
            if platform == "linux" or platform == "linux2":
                total_memory, used_memory, free_memory = map(
                    int,
                    os.popen('free -t -b').readlines()[-1].split()[1:]
                )
        if used_memory:
            used_memory_mb = float(used_memory / 1024 / 1024)

        return used_memory_mb

    def processAlgorithm(self, parameters, context, feedback):

        # Parameters
        projects_root_directory = parameters[self.PROJECTS_ROOT_DIRECTORY]
        output_csv_projects_path = parameters[self.OUTPUT_CSV_PROJECTS_PATH]
        output_csv_layers_path = parameters[self.OUTPUT_CSV_LAYERS_PATH]
        output_csv_tables_path = parameters[self.OUTPUT_CSV_TABLES_PATH]
        dont_resolve_layers = self.parameterAsBool(parameters, self.DONT_RESOLVE_LAYERS, context)
        layer_name_filter = self.parameterAsString(parameters, self.LAYER_NAME_FILTER, context).strip()
        layer_datasource_filter = self.parameterAsString(parameters, self.LAYER_DATASOURCE_FILTER, context).strip()
        get_lizmap_config_properties = self.parameterAsBool(parameters, self.GET_LIZMAP_CONFIG_PROPERTIES, context)

        # Store settings
        s = QgsSettings()
        s.setValue("lizmap/audit_projects_root_directory", projects_root_directory)
        s.setValue("lizmap/audit_output_csv_projects_path", output_csv_projects_path)
        s.setValue("lizmap/audit_output_csv_layers_path", output_csv_layers_path)
        s.setValue("lizmap/audit_output_csv_tables_path", output_csv_tables_path)
        s.setValue("lizmap/audit_dont_resolve_layers", str(dont_resolve_layers))
        s.setValue("lizmap/audit_layer_name_filter", str(layer_name_filter))
        s.setValue("lizmap/audit_layer_datasource_filter", str(layer_datasource_filter))
        s.setValue("lizmap/audit_get_lizmap_config_properties", str(get_lizmap_config_properties))

        # List project files
        feedback.pushInfo(self.tr('Fetching the list of QGIS project files...'))
        qgis_project_files = [y for x in os.walk(projects_root_directory) for y in glob(os.path.join(x[0], '*.qg?'))]

        # Do nothing if there are no projects
        if not qgis_project_files:
            feedback.pushInfo(self.tr('-> No project found in the given directory !'))
            return {}

        # Sort project paths alphabetically
        def sort_case_insensitive(e):
            return e.lower()

        qgis_project_files.sort(key=sort_case_insensitive)

        # Report the number of projects found
        feedback.pushInfo(
            self.tr('-> {} projects have been found in the directory and its subdirectories.'.format(
                len(qgis_project_files)
            ))
        )

        # Objects used to store the projects and layers properties
        projects = []
        exported_datasources = {}
        unique_postgresql_tables = {}
        filtered_project_number = 0

        # Loop for each project
        feedback.pushInfo(self.tr('Parsing each project file...'))
        total = len(qgis_project_files)
        for current, project_file in enumerate(qgis_project_files):
            # Stop the algorithm if the cancel button has been clicked
            if feedback.isCanceled():
                break

            # Get system memory used before reading the project
            start_memory = self.getMemoryUsage()

            # Load project
            p = QgsProject()
            if not dont_resolve_layers:
                p.read(
                    project_file
                )
            else:
                p.read(
                    project_file,
                    QgsProject.ReadFlag.FlagDontResolveLayers
                )

            # Get memory after reading project
            end_memory = self.getMemoryUsage()

            # Projet info object
            project = {}

            project_relative_file = project_file.replace(projects_root_directory, '')
            # Print project relative path
            feedback.pushInfo(
                '* {} - {}'.format(
                    str(current).zfill(len(str(total))),
                    project_relative_file
                )
            )
            # Progress bar
            feedback.setProgress(int(100 * current / total))

            # Parse layers
            name_matches = []
            datasource_matches = []
            for lid in p.mapLayers():
                # Get layer
                layer = p.mapLayer(lid)
                xml_properties = layer.originalXmlProperties()
                try:
                    root = et.fromstring(xml_properties)
                    datasource = root.find('datasource').text
                except et_error:
                    datasource = 'GHOST'
                    feedback.reportError(
                        self.tr('Project: {} - Layer "{}" ({}) - Error while parsing datasource: ghost layer ?').format(
                            project_relative_file,
                            layer.name(),
                            layer.id()
                        ),
                        False
                    )

                datasource = datasource.strip()
                match_name = True
                match_datasource = True

                # Check if the layer name matches the optional filter
                layer_name = layer.name().strip()
                if layer_name_filter:
                    if layer_name_filter.lower() in layer_name.lower():
                        name_matches.append(layer.id())
                    else:
                        match_name = False

                # Check if the datasource matches the optional filter
                if layer_datasource_filter:
                    if layer_datasource_filter in datasource:
                        datasource_matches.append(layer.id())
                    else:
                        match_datasource = False

                # Do not add the layer to the exported layers if there is no match
                if not match_name or not match_datasource:
                    continue

                # Do not add the layer datasource in the exported datasources object
                if not datasource:
                    continue

                # Add PostgreSQL schema and table if needed
                if output_csv_tables_path and layer.providerType() == 'postgres':
                    try:
                        uri = QgsDataSourceUri(datasource)
                        if uri:
                            table_name = '"{}"."{}"'.format(
                                uri.schema(),
                                uri.table(),
                            )
                            if table_name not in unique_postgresql_tables:
                                unique_postgresql_tables[table_name] = []
                            if project_relative_file not in unique_postgresql_tables[table_name]:
                                unique_postgresql_tables[table_name].append(project_relative_file)

                    except Exception:
                        print(self.tr('Error while parsing the PostgreSQL layer datasource'))

                # Compute the layer properties to keep
                if output_csv_layers_path:
                    if datasource not in exported_datasources:
                        exported_datasources[datasource] = {
                            'names': [],
                            'projects': [],
                        }
                    if layer_name not in exported_datasources[datasource]['names']:
                        exported_datasources[datasource]['names'].append(layer_name)
                    if project_relative_file not in exported_datasources[datasource]['projects']:
                        exported_datasources[datasource]['projects'].append(project_relative_file)

            # Do not continue if the project does not contain filtered layers
            if layer_name_filter and not name_matches:
                feedback.pushInfo(self.tr('-> No layers for this project matches the given name filter. Skipping.'))
                continue
            if layer_datasource_filter and not datasource_matches:
                feedback.pushInfo(self.tr('-> No layers for this project matches the given XML filter. Skipping.'))
                continue

            filtered_project_number += 1

            # File name
            project['path'] = project_file
            project['basename'] = p.baseName()

            # Last modified date
            project['last_modified'] = p.lastModified().toString('yyyy-MM-dd hh:mm:ss')

            # Projection
            project['crs'] = p.crs().authid()

            # Layer count
            project['layer_count'] = p.count()

            # Memory usage in MB
            if start_memory and end_memory:
                memory_usage_mb = end_memory - start_memory
                m = round(memory_usage_mb, 2)
            else:
                m = ''
            project['memory_usage_mb'] = m

            # Bad layers
            if dont_resolve_layers:
                project['invalid_layer_count'] = ''
            else:
                project['invalid_layer_count'] = p.count() - p.validCount()

            # Composer count
            layout_manager = p.layoutManager()
            layouts = layout_manager.layouts()
            layout_number = len(layouts)
            project['print_layout_count'] = layout_number

            # Todo composer names in a list

            # Count pictures in composer and sum up size
            picture_paths = []
            picture_sizes = []
            pictures_total_size = 0
            image_re = re.compile('(bmp|gif|png|jpg|jpeg|tif)$', re.IGNORECASE)
            for lay in layouts:
                layout = layout_manager.layoutByName(lay.name())
                # avoid reports
                if not hasattr(layout, 'items'):
                    continue
                for item in layout.items():
                    if isinstance(item, QgsLayoutItemPicture):
                        picture_path = item.evaluatedPath()
                        if not os.path.isfile(picture_path):
                            picture_path = item.picturePath()
                        if not os.path.isfile(picture_path):
                            continue
                        # Keep only interesting images
                        if not image_re.search(picture_path):
                            continue

                        # Add to the list
                        # picture_paths.append(picture_path)
                        picture_added = False
                        if picture_path not in picture_paths:
                            picture_paths.append(picture_path)
                            picture_added = True

                        # Get size
                        picture_size = os.path.getsize(picture_path)
                        if not picture_size:
                            picture_size = '0'
                        # picture_sizes.append(str(picture_size))
                        if picture_added:
                            picture_sizes.append(str(picture_size))
                        pictures_total_size += int(picture_size)

            project['print_layout_pictures_paths'] = '|'.join(picture_paths)
            project['print_layout_pictures_sizes'] = '|'.join(picture_sizes)
            project['print_layout_pictures_total_size'] = pictures_total_size
            pictures_total_size_mb = float(pictures_total_size / 1024 / 1024)
            project['print_layout_pictures_total_size_mb'] = round(pictures_total_size_mb, 2)

            # Trust option
            project['trust_option_active'] = p.trustLayerMetadata()

            # Project version
            p_version = p.lastSaveVersion()
            project['last_save_version'] = '{}{}{}'.format(
                p_version.majorVersion(),
                str(p_version.minorVersion()).zfill(2),
                str(p_version.subVersion()).zfill(2)
            )

            # Lizmap plugin and Lizmap Web Client related properties
            if get_lizmap_config_properties:
                cfg_path = project_file + '.cfg'
                has_lizmap_config = os.path.isfile(cfg_path)
                project['has_lizmap_config'] = str(has_lizmap_config)
                qgis_desktop_version = ''
                lizmap_plugin_version = ''
                lizmap_web_client_target_version = ''
                project_valid = ''
                if has_lizmap_config:
                    with open(cfg_path) as cfg:
                        data = json.load(cfg)
                        if 'metadata' in data:
                            if 'qgis_desktop_version' in data['metadata']:
                                qgis_desktop_version = data['metadata']['qgis_desktop_version']
                            if 'lizmap_plugin_version' in data['metadata']:
                                lizmap_plugin_version = data['metadata']['lizmap_plugin_version']
                            if 'lizmap_web_client_target_version' in data['metadata']:
                                lizmap_web_client_target_version = data['metadata']['lizmap_web_client_target_version']
                            if 'project_valid' in data['metadata']:
                                project_valid = data['metadata']['project_valid']
                project['qgis_desktop_version'] = qgis_desktop_version
                project['lizmap_plugin_version'] = lizmap_plugin_version
                project['lizmap_web_client_target_version'] = lizmap_web_client_target_version
                project['project_valid'] = project_valid

            # Add project to the list
            projects.append(project)

        # Feedback
        feedback.pushInfo('')
        feedback.pushInfo(self.tr('-> The project files have been parsed !'))
        if total != filtered_project_number:
            feedback.pushInfo(
                self.tr('{} filtered projects among the {} listed projects !').format(
                    filtered_project_number,
                    total
                )
            )

        # Write CSV output files

        # projects CSV
        feedback.pushInfo('')
        feedback.pushInfo(self.tr('Writing the result to the CSV output file for projects...'))
        fieldnames = [
            'path', 'basename', 'last_modified', 'crs',
            'layer_count', 'invalid_layer_count', 'memory_usage_mb',
            'print_layout_count', 'print_layout_pictures_paths', 'print_layout_pictures_sizes',
            'print_layout_pictures_total_size', 'print_layout_pictures_total_size_mb',
            'trust_option_active', 'last_save_version'
        ]
        if get_lizmap_config_properties:
            fieldnames += [
                'has_lizmap_config', 'qgis_desktop_version', 'lizmap_plugin_version',
                'lizmap_web_client_target_version', 'project_valid',
            ]
        with open(output_csv_projects_path, mode='w') as csv_file:
            writer = csv.DictWriter(
                csv_file,
                fieldnames=fieldnames,
                delimiter=',', quotechar='"', quoting=csv.QUOTE_ALL
            )
            writer.writeheader()
            writer.writerows(projects)

        feedback.pushInfo(
            self.tr(
                '-> The results have been written to the output CSV file "{}" !'.format(output_csv_projects_path)
            )
        )

        # Optional layers CSV
        if output_csv_layers_path:
            feedback.pushInfo('')
            feedback.pushInfo(self.tr('Writing the result to the CSV output file for layers...'))
            layers = []
            for datasource in sorted(exported_datasources.keys()):
                item = exported_datasources[datasource]
                layer_item = {
                    'datasource': datasource,
                    'layer_names': '|'.join(item['names']),
                    'projects': '|'.join(item['projects']),
                }
                layers.append(layer_item)

            fieldnames = [
                'datasource', 'layer_names', 'projects'
            ]
            with open(output_csv_layers_path, mode='w') as csv_file:
                writer = csv.DictWriter(
                    csv_file,
                    fieldnames=fieldnames,
                    delimiter=',', quotechar='"', quoting=csv.QUOTE_ALL
                )
                writer.writeheader()
                writer.writerows(layers)

            feedback.pushInfo(
                self.tr(
                    '-> The results have been written to the output CSV file "{}" !'.format(output_csv_layers_path)
                )
            )

        # Print unique_postgresql_tables
        if output_csv_tables_path:
            feedback.pushInfo('')
            feedback.pushInfo(self.tr('Writing the PostgreSQL tables listed in the projects...'))
            feedback.pushInfo('')
            tables = []
            for table in sorted(unique_postgresql_tables.keys()):
                table_projects = unique_postgresql_tables[table]
                table_item = {
                    'table': table,
                    'projects': '|'.join(table_projects),
                }
                tables.append(table_item)

            fieldnames = [
                'table', 'projects'
            ]
            with open(output_csv_tables_path, mode='w') as csv_file:
                writer = csv.DictWriter(
                    csv_file,
                    fieldnames=fieldnames,
                    delimiter=',', quotechar='"', quoting=csv.QUOTE_ALL
                )
                writer.writeheader()
                writer.writerows(tables)

            feedback.pushInfo(
                self.tr(
                    '-> The results have been written to the output CSV file "{}" !'.format(output_csv_tables_path)
                )
            )

        return {}
	# -- coding: utf-8 --
	"""
	***************************************************************************
	* *
	* This program is free software; you can redistribute it and/or modify *
	* it under the terms of the GNU General Public License as published by *
	* the Free Software Foundation; either version 2 of the License, or *
	* (at your option) any later version. *
	* *
	***************************************************************************
	"""

	__author__ = 'Michaël Douchin'

	import os
	from pathlib import Path
	from sys import platform
	import re
	import csv
	import json
	from glob import glob
	from xml.etree import cElementTree as et
	from xml.etree.ElementTree import ParseError as et_error

	from qgis.PyQt.QtCore import QCoreApplication
	from qgis.core import (
	QgsProject,
	QgsDataSourceUri,
	QgsLayoutItemPicture,
	QgsSettings,
	QgsProcessingAlgorithm,
	QgsProcessingParameterString,
	QgsProcessingParameterFile,
	QgsProcessingParameterFileDestination,
	QgsProcessingParameterBoolean
	)


	class AuditQGISProjects(QgsProcessingAlgorithm):
	"""
	List all the QGIS projects of a directory and fetch key datas
	"""
	PROJECTS_ROOT_DIRECTORY = 'PROJECTS_ROOT_DIRECTORY'
	OUTPUT_CSV_PROJECTS_PATH = 'OUTPUT_CSV_PROJECTS_PATH'
	OUTPUT_CSV_LAYERS_PATH = 'OUTPUT_CSV_LAYERS_PATH'
	OUTPUT_CSV_TABLES_PATH = 'OUTPUT_CSV_TABLES_PATH'
	DONT_RESOLVE_LAYERS = 'DONT_RESOLVE_LAYERS'
	LAYER_NAME_FILTER = 'LAYER_NAME_FILTER'
	LAYER_DATASOURCE_FILTER = 'LAYER_DATASOURCE_FILTER'
	GET_LIZMAP_CONFIG_PROPERTIES = 'GET_LIZMAP_CONFIG_PROPERTIES'

	def tr(self, string):
	"""
	Returns a translatable string with the self.tr() function.
	"""
	return QCoreApplication.translate('Processing', string)

	def createInstance(self):
	return AuditQGISProjects()

	def name(self):
	return 'audit_qgis_projects'

	def displayName(self):
	return self.tr('Audit QGIS projects')

	def group(self):
	return self.tr('Projects')

	def groupId(self):
	return 'projects'

	def shortHelpString(self):
	return self.tr(
	'Fetch the properties of all the QGIS *.qgs projects'
	' found in a directory and its subdirectories.'
	'\n'
	'You can also find specific projects matching layers name or datasources.'
	'\n'
	'This algorithm creates CSV file(s) containing the list of projects properties'
	' and optionally the list of layers (unique datasources and related projects)'
	)

	def helpUrl(self):
	url = 'https://gist.github.com/mdouchin/cd89a259d3560635d4bb8708ad430caa#file-readme-md'
	return url

	def initAlgorithm(self, config=None):
	# Settings
	s = QgsSettings()

	# Root directory
	projects_root_directory = s.value("lizmap/audit_projects_root_directory", Path.home())
	input_param = QgsProcessingParameterFile(
	self.PROJECTS_ROOT_DIRECTORY,
	self.tr('Local directory containing projects'),
	defaultValue=projects_root_directory,
	behavior=QgsProcessingParameterFile.Folder,
	optional=False
	)
	input_param.setHelp(self.tr(
	'The QGIS projects contained in this directory and sub-directories'
	' will be analysed to gather their key parameters.'
	))
	self.addParameter(input_param)

	# CSV file to create
	output_csv_projects_path = s.value(
	"lizmap/audit_output_csv_projects_path",
	os.path.join(Path.home(), 'qgis_projects_audit.csv')
	)
	input_param = QgsProcessingParameterFileDestination(
	self.OUTPUT_CSV_PROJECTS_PATH,
	self.tr('Output CSV file with the list of projects and their properties'),
	fileFilter='*.csv',
	optional=False,
	defaultValue=output_csv_projects_path
	)
	input_param.setHelp(self.tr(
	'This algorithm will create a Comma Separated Values (CSV) file'
	' containing the list of projects with their properties.'
	' Please choose the path.'
	))
	self.addParameter(input_param)

	# CSV file containing unique datasources
	output_csv_layers_path = s.value(
	"lizmap/audit_output_csv_layers_path",
	os.path.join(Path.home(), 'qgis_layers_audit.csv')
	)
	input_param = QgsProcessingParameterFileDestination(
	self.OUTPUT_CSV_LAYERS_PATH,
	self.tr('Output CSV file with the list of distinct layers datasources'),
	fileFilter='*.csv',
	optional=True,
	defaultValue=output_csv_layers_path
	)
	input_param.setHelp(self.tr(
	'This algorithm will create an optional Comma Separated Values (CSV) file'
	' containing the list of distinct datasources of layers and their projects'
	))
	self.addParameter(input_param)

	# CSV file containing unique PostgreSQL tables
	output_csv_tables_path = s.value(
	"lizmap/audit_output_csv_tables_path",
	os.path.join(Path.home(), 'qgis_tables_audit.csv')
	)
	input_param = QgsProcessingParameterFileDestination(
	self.OUTPUT_CSV_TABLES_PATH,
	self.tr('Output CSV file with the list of distinct PostgreSQL tables'),
	fileFilter='*.csv',
	optional=True,
	defaultValue=output_csv_tables_path
	)
	input_param.setHelp(self.tr(
	'This algorithm will create an optional Comma Separated Values (CSV) file'
	' containing the list of distinct PostgreSQL tables and their projects'
	))
	self.addParameter(input_param)

	# Choose to resolve layers, to find invalid layers
	# If set to True, this would be much longer
	# And will not work out of QGIS Desktop
	# See https://github.com/qgis/QGIS/issues/34408
	dont_resolve_layers = s.value("lizmap/audit_dont_resolve_layers", True)
	if dont_resolve_layers in ('True', 'False'):
	dont_resolve_layers = eval(dont_resolve_layers)
	input_param = QgsProcessingParameterBoolean(
	self.DONT_RESOLVE_LAYERS,
	self.tr('Do not resolve layers'),
	defaultValue=dont_resolve_layers,
	optional=False
	)
	input_param.setHelp(self.tr(
	'If checked, the algorithm will load every layer of every project'
	' to find invalid layers.'
	' This could be very long depending on the number of projects,'
	' and layers, and should be used only in QGIS Desktop context'
	))
	self.addParameter(input_param)

	# Filter projects containing layer names
	layer_name_filter = s.value("lizmap/audit_layer_name_filter", '')
	input_param = QgsProcessingParameterString(
	self.LAYER_NAME_FILTER,
	self.tr('Filter projects with layers names like...'),
	defaultValue=layer_name_filter,
	optional=True
	)
	input_param.setHelp(self.tr(
	'If this input is not empty, the output CSV will list only'
	' the projects containing layers corresponding to the written text.'
	' You can write a full name or an extract.'
	))
	self.addParameter(input_param)

	# Filter projects containing layer datasources
	layer_datasource_filter = s.value("lizmap/audit_layer_datasource_filter", '')
	input_param = QgsProcessingParameterString(
	self.LAYER_DATASOURCE_FILTER,
	self.tr('Filter projects with layers XML definition containing...'),
	defaultValue=layer_datasource_filter,
	optional=True
	)
	input_param.setHelp(self.tr(
	'If this input is not empty, the output CSV will list only'
	' the projects containing layers corresponding to the written datasource.'
	' You can write a full datasource or an extract.'
	))
	self.addParameter(input_param)

	# Lizmap related data
	# If set to True, the script will look for a */qgs.cfg file
	# and fetch more properties from it
	get_lizmap_config_properties = s.value("lizmap/audit_get_lizmap_config_properties", False)
	if get_lizmap_config_properties in ('True', 'False'):
	get_lizmap_config_properties = eval(get_lizmap_config_properties)
	input_param = QgsProcessingParameterBoolean(
	self.GET_LIZMAP_CONFIG_PROPERTIES,
	self.tr('Get Lizmap configuration properties (from Lizmap *.cfg file'),
	defaultValue=get_lizmap_config_properties,
	optional=False
	)
	input_param.setHelp(self.tr(
	'If checked, the algorithm will search for a Lizmap *.cfg file'
	' and fetch more properties from it.'
	))
	self.addParameter(input_param)

	def getMemoryUsage(self):
	has_psutil = True
	used_memory = None
	used_memory_mb = None
	# Use psutil if available
	# else on linux use quick&dirty os call
	try:
	import psutil
	except ImportError:
	has_psutil = False

	if has_psutil:
	used_memory = psutil.virtual_memory().used
	else:
	if platform == "linux" or platform == "linux2":
	total_memory, used_memory, free_memory = map(
	int,
	os.popen('free -t -b').readlines()[-1].split()[1:]
	)
	if used_memory:
	used_memory_mb = float(used_memory / 1024 / 1024)

	return used_memory_mb

	def processAlgorithm(self, parameters, context, feedback):

	# Parameters
	projects_root_directory = parameters[self.PROJECTS_ROOT_DIRECTORY]
	output_csv_projects_path = parameters[self.OUTPUT_CSV_PROJECTS_PATH]
	output_csv_layers_path = parameters[self.OUTPUT_CSV_LAYERS_PATH]
	output_csv_tables_path = parameters[self.OUTPUT_CSV_TABLES_PATH]
	dont_resolve_layers = self.parameterAsBool(parameters, self.DONT_RESOLVE_LAYERS, context)
	layer_name_filter = self.parameterAsString(parameters, self.LAYER_NAME_FILTER, context).strip()
	layer_datasource_filter = self.parameterAsString(parameters, self.LAYER_DATASOURCE_FILTER, context).strip()
	get_lizmap_config_properties = self.parameterAsBool(parameters, self.GET_LIZMAP_CONFIG_PROPERTIES, context)

	# Store settings
	s = QgsSettings()
	s.setValue("lizmap/audit_projects_root_directory", projects_root_directory)
	s.setValue("lizmap/audit_output_csv_projects_path", output_csv_projects_path)
	s.setValue("lizmap/audit_output_csv_layers_path", output_csv_layers_path)
	s.setValue("lizmap/audit_output_csv_tables_path", output_csv_tables_path)
	s.setValue("lizmap/audit_dont_resolve_layers", str(dont_resolve_layers))
	s.setValue("lizmap/audit_layer_name_filter", str(layer_name_filter))
	s.setValue("lizmap/audit_layer_datasource_filter", str(layer_datasource_filter))
	s.setValue("lizmap/audit_get_lizmap_config_properties", str(get_lizmap_config_properties))

	# List project files
	feedback.pushInfo(self.tr('Fetching the list of QGIS project files...'))
	qgis_project_files = [y for x in os.walk(projects_root_directory) for y in glob(os.path.join(x[0], '*.qg?'))]

	# Do nothing if there are no projects
	if not qgis_project_files:
	feedback.pushInfo(self.tr('-> No project found in the given directory !'))
	return {}

	# Sort project paths alphabetically
	def sort_case_insensitive(e):
	return e.lower()

	qgis_project_files.sort(key=sort_case_insensitive)

	# Report the number of projects found
	feedback.pushInfo(
	self.tr('-> {} projects have been found in the directory and its subdirectories.'.format(
	len(qgis_project_files)
	))
	)

	# Objects used to store the projects and layers properties
	projects = []
	exported_datasources = {}
	unique_postgresql_tables = {}
	filtered_project_number = 0

	# Loop for each project
	feedback.pushInfo(self.tr('Parsing each project file...'))
	total = len(qgis_project_files)
	for current, project_file in enumerate(qgis_project_files):
	# Stop the algorithm if the cancel button has been clicked
	if feedback.isCanceled():
	break

	# Get system memory used before reading the project
	start_memory = self.getMemoryUsage()

	# Load project
	p = QgsProject()
	if not dont_resolve_layers:
	p.read(
	project_file
	)
	else:
	p.read(
	project_file,
	QgsProject.ReadFlag.FlagDontResolveLayers
	)

	# Get memory after reading project
	end_memory = self.getMemoryUsage()

	# Projet info object
	project = {}

	project_relative_file = project_file.replace(projects_root_directory, '')
	# Print project relative path
	feedback.pushInfo(
	'* {} - {}'.format(
	str(current).zfill(len(str(total))),
	project_relative_file
	)
	)
	# Progress bar
	feedback.setProgress(int(100 * current / total))

	# Parse layers
	name_matches = []
	datasource_matches = []
	for lid in p.mapLayers():
	# Get layer
	layer = p.mapLayer(lid)
	xml_properties = layer.originalXmlProperties()
	try:
	root = et.fromstring(xml_properties)
	datasource = root.find('datasource').text
	except et_error:
	datasource = 'GHOST'
	feedback.reportError(
	self.tr('Project: {} - Layer "{}" ({}) - Error while parsing datasource: ghost layer ?').format(
	project_relative_file,
	layer.name(),
	layer.id()
	),
	False
	)

	datasource = datasource.strip()
	match_name = True
	match_datasource = True

	# Check if the layer name matches the optional filter
	layer_name = layer.name().strip()
	if layer_name_filter:
	if layer_name_filter.lower() in layer_name.lower():
	name_matches.append(layer.id())
	else:
	match_name = False

	# Check if the datasource matches the optional filter
	if layer_datasource_filter:
	if layer_datasource_filter in datasource:
	datasource_matches.append(layer.id())
	else:
	match_datasource = False

	# Do not add the layer to the exported layers if there is no match
	if not match_name or not match_datasource:
	continue

	# Do not add the layer datasource in the exported datasources object
	if not datasource:
	continue

	# Add PostgreSQL schema and table if needed
	if output_csv_tables_path and layer.providerType() == 'postgres':
	try:
	uri = QgsDataSourceUri(datasource)
	if uri:
	table_name = '"{}"."{}"'.format(
	uri.schema(),
	uri.table(),
	)
	if table_name not in unique_postgresql_tables:
	unique_postgresql_tables[table_name] = []
	if project_relative_file not in unique_postgresql_tables[table_name]:
	unique_postgresql_tables[table_name].append(project_relative_file)

	except Exception:
	print(self.tr('Error while parsing the PostgreSQL layer datasource'))

	# Compute the layer properties to keep
	if output_csv_layers_path:
	if datasource not in exported_datasources:
	exported_datasources[datasource] = {
	'names': [],
	'projects': [],
	}
	if layer_name not in exported_datasources[datasource]['names']:
	exported_datasources[datasource]['names'].append(layer_name)
	if project_relative_file not in exported_datasources[datasource]['projects']:
	exported_datasources[datasource]['projects'].append(project_relative_file)

	# Do not continue if the project does not contain filtered layers
	if layer_name_filter and not name_matches:
	feedback.pushInfo(self.tr('-> No layers for this project matches the given name filter. Skipping.'))
	continue
	if layer_datasource_filter and not datasource_matches:
	feedback.pushInfo(self.tr('-> No layers for this project matches the given XML filter. Skipping.'))
	continue

	filtered_project_number += 1

	# File name
	project['path'] = project_file
	project['basename'] = p.baseName()

	# Last modified date
	project['last_modified'] = p.lastModified().toString('yyyy-MM-dd hh:mm:ss')

	# Projection
	project['crs'] = p.crs().authid()

	# Layer count
	project['layer_count'] = p.count()

	# Memory usage in MB
	if start_memory and end_memory:
	memory_usage_mb = end_memory - start_memory
	m = round(memory_usage_mb, 2)
	else:
	m = ''
	project['memory_usage_mb'] = m

	# Bad layers
	if dont_resolve_layers:
	project['invalid_layer_count'] = ''
	else:
	project['invalid_layer_count'] = p.count() - p.validCount()

	# Composer count
	layout_manager = p.layoutManager()
	layouts = layout_manager.layouts()
	layout_number = len(layouts)
	project['print_layout_count'] = layout_number

	# Todo composer names in a list

	# Count pictures in composer and sum up size
	picture_paths = []
	picture_sizes = []
	pictures_total_size = 0
	image_re = re.compile('(bmp\|gif\|png\|jpg\|jpeg\|tif)$', re.IGNORECASE)
	for lay in layouts:
	layout = layout_manager.layoutByName(lay.name())
	# avoid reports
	if not hasattr(layout, 'items'):
	continue
	for item in layout.items():
	if isinstance(item, QgsLayoutItemPicture):
	picture_path = item.evaluatedPath()
	if not os.path.isfile(picture_path):
	picture_path = item.picturePath()
	if not os.path.isfile(picture_path):
	continue
	# Keep only interesting images
	if not image_re.search(picture_path):
	continue

	# Add to the list
	# picture_paths.append(picture_path)
	picture_added = False
	if picture_path not in picture_paths:
	picture_paths.append(picture_path)
	picture_added = True

	# Get size
	picture_size = os.path.getsize(picture_path)
	if not picture_size:
	picture_size = '0'
	# picture_sizes.append(str(picture_size))
	if picture_added:
	picture_sizes.append(str(picture_size))
	pictures_total_size += int(picture_size)

	project['print_layout_pictures_paths'] = '\|'.join(picture_paths)
	project['print_layout_pictures_sizes'] = '\|'.join(picture_sizes)
	project['print_layout_pictures_total_size'] = pictures_total_size
	pictures_total_size_mb = float(pictures_total_size / 1024 / 1024)
	project['print_layout_pictures_total_size_mb'] = round(pictures_total_size_mb, 2)

	# Trust option
	project['trust_option_active'] = p.trustLayerMetadata()

	# Project version
	p_version = p.lastSaveVersion()
	project['last_save_version'] = '{}{}{}'.format(
	p_version.majorVersion(),
	str(p_version.minorVersion()).zfill(2),
	str(p_version.subVersion()).zfill(2)
	)

	# Lizmap plugin and Lizmap Web Client related properties
	if get_lizmap_config_properties:
	cfg_path = project_file + '.cfg'
	has_lizmap_config = os.path.isfile(cfg_path)
	project['has_lizmap_config'] = str(has_lizmap_config)
	qgis_desktop_version = ''
	lizmap_plugin_version = ''
	lizmap_web_client_target_version = ''
	project_valid = ''
	if has_lizmap_config:
	with open(cfg_path) as cfg:
	data = json.load(cfg)
	if 'metadata' in data:
	if 'qgis_desktop_version' in data['metadata']:
	qgis_desktop_version = data['metadata']['qgis_desktop_version']
	if 'lizmap_plugin_version' in data['metadata']:
	lizmap_plugin_version = data['metadata']['lizmap_plugin_version']
	if 'lizmap_web_client_target_version' in data['metadata']:
	lizmap_web_client_target_version = data['metadata']['lizmap_web_client_target_version']
	if 'project_valid' in data['metadata']:
	project_valid = data['metadata']['project_valid']
	project['qgis_desktop_version'] = qgis_desktop_version
	project['lizmap_plugin_version'] = lizmap_plugin_version
	project['lizmap_web_client_target_version'] = lizmap_web_client_target_version
	project['project_valid'] = project_valid

	# Add project to the list
	projects.append(project)

	# Feedback
	feedback.pushInfo('')
	feedback.pushInfo(self.tr('-> The project files have been parsed !'))
	if total != filtered_project_number:
	feedback.pushInfo(
	self.tr('{} filtered projects among the {} listed projects !').format(
	filtered_project_number,
	total
	)
	)

	# Write CSV output files

	# projects CSV
	feedback.pushInfo('')
	feedback.pushInfo(self.tr('Writing the result to the CSV output file for projects...'))
	fieldnames = [
	'path', 'basename', 'last_modified', 'crs',
	'layer_count', 'invalid_layer_count', 'memory_usage_mb',
	'print_layout_count', 'print_layout_pictures_paths', 'print_layout_pictures_sizes',
	'print_layout_pictures_total_size', 'print_layout_pictures_total_size_mb',
	'trust_option_active', 'last_save_version'
	]
	if get_lizmap_config_properties:
	fieldnames += [
	'has_lizmap_config', 'qgis_desktop_version', 'lizmap_plugin_version',
	'lizmap_web_client_target_version', 'project_valid',
	]
	with open(output_csv_projects_path, mode='w') as csv_file:
	writer = csv.DictWriter(
	csv_file,
	fieldnames=fieldnames,
	delimiter=',', quotechar='"', quoting=csv.QUOTE_ALL
	)
	writer.writeheader()
	writer.writerows(projects)

	feedback.pushInfo(
	self.tr(
	'-> The results have been written to the output CSV file "{}" !'.format(output_csv_projects_path)
	)
	)

	# Optional layers CSV
	if output_csv_layers_path:
	feedback.pushInfo('')
	feedback.pushInfo(self.tr('Writing the result to the CSV output file for layers...'))
	layers = []
	for datasource in sorted(exported_datasources.keys()):
	item = exported_datasources[datasource]
	layer_item = {
	'datasource': datasource,
	'layer_names': '\|'.join(item['names']),
	'projects': '\|'.join(item['projects']),
	}
	layers.append(layer_item)

	fieldnames = [
	'datasource', 'layer_names', 'projects'
	]
	with open(output_csv_layers_path, mode='w') as csv_file:
	writer = csv.DictWriter(
	csv_file,
	fieldnames=fieldnames,
	delimiter=',', quotechar='"', quoting=csv.QUOTE_ALL
	)
	writer.writeheader()
	writer.writerows(layers)

	feedback.pushInfo(
	self.tr(
	'-> The results have been written to the output CSV file "{}" !'.format(output_csv_layers_path)
	)
	)

	# Print unique_postgresql_tables
	if output_csv_tables_path:
	feedback.pushInfo('')
	feedback.pushInfo(self.tr('Writing the PostgreSQL tables listed in the projects...'))
	feedback.pushInfo('')
	tables = []
	for table in sorted(unique_postgresql_tables.keys()):
	table_projects = unique_postgresql_tables[table]
	table_item = {
	'table': table,
	'projects': '\|'.join(table_projects),
	}
	tables.append(table_item)

	fieldnames = [
	'table', 'projects'
	]
	with open(output_csv_tables_path, mode='w') as csv_file:
	writer = csv.DictWriter(
	csv_file,
	fieldnames=fieldnames,
	delimiter=',', quotechar='"', quoting=csv.QUOTE_ALL
	)
	writer.writeheader()
	writer.writerows(tables)

	feedback.pushInfo(
	self.tr(
	'-> The results have been written to the output CSV file "{}" !'.format(output_csv_tables_path)
	)
	)

	return {}