sondes bathym\u00e9triques<\/div>

<\/div><\/div>

<\/p>\n\n\n\n

-oo SPLIT_MULTIPOINT=ON -oo ADD_SOUNDG_DEPTH=ON<\/em><\/strong><\/p>\n\n\n\n

<\/div><\/div>\n\n\n\n

Retrieving \u00ab\u00a0parent\u00a0\u00bb identifiers<\/p>\n\n\n\n

Some objects can be grouped by a non-geographic parent object. For example, fire sectors are coded with one sector per record in the \u00ab\u00a0LIGHTS\u00a0\u00bb table. The different sectors of the same fire have no particular attribute that could indicate that they correspond to the same entity. This information is contained in a \u00ab\u00a0parent\u00a0\u00bb record. To retrieve this identifier as an attribute from the LIGTHS table, you need to add an option to the command line:<\/p>\n\n\n\n

<\/div>
identifiants parents<\/div>
<\/div><\/div>
<\/p>\n\n\n\n
-oo RETURN_LINKAGES=O<\/strong><\/p>\n\n\n\n
<\/div><\/div>\n\n\n\n
This option creates a name_rcid attribute that is common to all sectors of the same fire, but also creates a series of fields (name_rcnm,ORNT,USAG,MASK).<\/p>\n\n\n\n
\u00ab\u00a0List\u00a0\u00bb type processing<\/h2>\n\n\n\n
In addition to the classic field types (integer, real, text), the S57 format uses a special type: lists of character strings or integers. These field types are found in PostgreSQL, but not in shapefiles and geopackages.<\/p>\n\n\n\n
In the absence of specific options, there will be warning messages indicating that the format is not supported, and the content of these fields will be ignored. To avoid losing content, add the :<\/p>\n\n\n\n
<\/div>
listes<\/div>
<\/div><\/div>
<\/p>\n\n\n\n
-mapFieldType StringList=String,IntegerList=String<\/em><\/strong><\/p>\n\n\n\n
<\/div><\/div>\n\n\n\n
This option takes a list {..,\u2026} and generates a text string with a first number indicating the number of elements in the list, a colon, then the list elements separated by commas:<\/p>\n\n\n\n
{3} -> ‘(1:3)’<\/p>\n\n\n\n
{3,1} -> ‘(2:3,1)’<\/p>\n\n\n\n
This inevitably complicates the processing of these fields, but the information is not lost. On the other hand, the GDAL driver presents a problem when it comes to handling the resulting String format.<\/p>\n\n\n\n
First of all, there’s no problem with the attributes of S57 objects. The field in the geopackage table is of type ‘Text’, which allows the creation of a character string of any length.
However, for \u00ab\u00a0parent identifiers\u00a0\u00bb, the definition used by the driver is ‘Text()’, i.e. a character string with a maximum length. This length is, a priori, determined by the content of the first record processed. In other words, it’s only by chance that you have a field with the length needed to integrate all the values in the file.<\/p>\n\n\n\n
You’ll then see messages like:<\/p>\n\n\n\n
WARNING : Value of field ‘NAME_RCID’ has 371 characters, whereas maximum allowed is 30<\/em><\/strong><\/p>\n\n\n\n
The solution isn’t complicated, but it does involve a fair amount of work. Assume that for the classic display of ENC cards, this is not a problem and you can simply ignore these warning messages.<\/p>\n\n\n\n
On the other hand, if you need more advanced processing including this information, here’s the only way I’ve found yet:<\/p>\n\n\n\n
<\/div>
r\u00e9cup\u00e9ration des values<\/div>
<\/div><\/div>
<\/p>\n\n\n\n
\n
Run the ogr2ogr command to create the import geopackage<\/li>\n\n\n\n
Locate the maximum value indicated for each field type<\/li>\n\n\n\n
For the output table(s) from which you wish to retrieve all information, open the QGis database manager<\/li>\n\n\n\n
Delete all table records (DELETE FROM table_name)<\/li>\n\n\n\n
Go to the Table->Modify Table\u2026 menu.<\/li>\n\n\n\n
Select the relevant field from the list of fields, then click on Edit a column<\/li>\n\n\n\n
Enter the desired value in the Length field, then click OK<\/li>\n\n\n\n
Do the same for all other fields of interest<\/li>\n\n\n\n
Delete all other tables in the import geopackage that do not pose a problem<\/li>\n\n\n\n
Re-execute the ogr2ogr import command.<\/li>\n<\/ol>\n\n\n\n
<\/div><\/div>\n\n\n\n
Workflow for loading a single S57 file<\/h1>\n\n\n\n
Creating the ENC geopackage<\/h2>\n\n\n\n
We’re going to create an empty geopackage that will receive the imported tables, then successive updates when other S57 files are loaded.<\/p>\n\n\n\n
In the QGis Explorer panel, right-click on Geopackages <\/em><\/strong>and select Create database<\/em><\/strong><\/p>\n\n\n\n
$\"\"$ <\/a><\/figure>\n\n\n\n
In the window that opens, enter the geopackage name, leave the default option for table name and select No geometry<\/em><\/strong> in Geometry type<\/em><\/strong>.<\/p>\n\n\n\n
$\"\"$ <\/a><\/figure>\n\n\n\n
Click OK to have the new empty geopackage added to the QGis geopackage connections. It is advisable to delete the empty ENC table created when the geopackage was created. To do this, open the QGis database manager from the main menu, position yourself on the ENC table of the ENC.gpkg geopackage and right-click to select Delete\u2026<\/p>\n\n\n\n
ogr2ogr commands for creating import geopackages<\/h2>\n\n\n\n
With all this in mind, here are the three lines of ogr2ogr commands for creating tables in import geopackages:<\/p>\n\n\n\n
<\/div>
Commandes ogr2ogr<\/div>
<\/div><\/div>
<\/p>\n\n\n\n
ogr2ogr -f GPKG -skipfailures -append -update -where \"OGR_GEOMETRY='POINT' or OGR_GEOMETRY='MULTIPOINT'\" -oo SPLIT_MULTIPOINT=ON -oo RETURN_LINKAGES=ON -oo LNAM_REFS=ON -oo ADD_SOUNDG_DEPTH=ON -mapFieldType StringList=String,IntegerList=String C:\/testgpkg\/pointsENC.gpkg C:\\testgpkg\\US1GC09M\\US1GC09M.000\n\nogr2ogr -f GPKG -skipfailures -append -update -where \"OGR_GEOMETRY='LINESTRING' or OGR_GEOMETRY='MULTILINESTRING'\" -oo RETURN_LINKAGES=ON -oo LNAM_REFS=ON -mapFieldType StringList=String,IntegerList=String C:\/testgpkg\/LinesENC.gpkg C:\\testgpkg\\US1GC09M\\US1GC09M.000\n\nogr2ogr -f GPKG -skipfailures -append -update -where \"OGR_GEOMETRY='POLYGON' or OGR_GEOMETRY='MULTIPOLYGON'\" -oo RETURN_LINKAGES=ON -oo LNAM_REFS=ON -mapFieldType StringList=String,IntegerList=String C:\/testgpkg\/PolysENC.gpkg C:\\testgpkg\\US1GC09M\\US1GC09M.000<\/code><\/pre>\n\n\n\n <\/div><\/div>\n\n\n\n o execute these command lines, simply open the OSGeo4W shell window:<\/p>\n\n\n\n <\/a><\/figure>\n\n\n\nThe Shell window opens<\/p>\n\n\n\n<\/a><\/figure>\n\n\n\nEnter the command lines, taking care to enter the path and filename with the .000 extension.<\/p>\n\n\n\n The result will be a series of tables created in each import geopackage. On the other hand, some tables will be completely empty. Indeed, if the geometry type is possible for an object class, the table will be created, but if there is no occurrence in the processed S57 file, the table will have no record.<\/p>\n\n\n\nUsing Python scripts for database management<\/h2>\n\n\n\nWe will use the QGis Python console for all database management operations.<\/p>\n\n\n\n To open it, go to Extensions -> Python console. The Python console panels open:<\/p>\n\n\n\n<\/a><\/figure>\n\n\n\nTo execute a Python script, load the code into the Editor panel. You can either copy and paste, or use the Directory icon to point to the .py file. As indentation is crucial for Python scripts, you should use the second option.<\/p>\n\n\n\nRemove empty tables and add map name and scale<\/h2>\n\n\n\nAs we saw earlier, when executing ogr2ogr commands, empty tables are created for layers where the geometry requested in the command line is not present. We will therefore remove these empty tables and take advantage of the Python script to add two attributes to all tables: the name of the ENC map and its scale.<\/p>\n\n\n\n As far as the name is concerned, it may be useful in future database maintenance to be able to select the entities corresponding to an S57 source file. This information is not contained in any attribute of the S57 file. We therefore have to enter it manually.<\/p>\n\n\n\n The same applies to the map scale. Although an attribute is provided in S57 to enter the scale of the \u00ab\u00a0paper\u00a0\u00bb map, the CSCALE attribute in the M_CSCL table, it is very rare for this table and its attribute to be present in S57 files.<\/p>\n\n\n\n While the name of the map is of relative use, the same cannot be said of the scale, as this is an essential criterion in the search for duplicates. Indeed, when mixing entities from several S57 files, data from different scales will coexist more or less happily.<\/p>\n\n\n\n The following Python code deletes empty tables from an import schema and adds two attributes (enc_chart and scale) to all tables:<\/p>\n\n\n\n<\/div>tables vides .py<\/div><\/div><\/div> <\/p>\n\n\n\nfrom osgeo import ogr<\/code><\/pre>\n\n\n\ndef delete_empty_tables_in_geopackages(geopackage_paths, valeur_par_defaut_enc, valeur_par_defaut_scale): for geopackage_path in geopackage_paths: delete_empty_tables(geopackage_path, valeur_par_defaut_enc, valeur_par_defaut_scale) print(f\"Fin du traitement\") def delete_empty_tables(geopackage_path, valeur_par_defaut_enc, valeur_par_defaut_scale): # Ouvrir le GeoPackage en mode \u00e9dition geopackage = ogr.Open(geopackage_path, 1)<\/code><\/pre>\n\n\n\nif geopackage is not None:\n # R\u00e9cup\u00e9rer le nombre de tables dans le GeoPackage\n num_tables = geopackage.GetLayerCount()\n\n # Liste des tables \u00e0 supprimer\n tables_to_delete = []\n tables_to_update = []\n\n # Identifier les tables \u00e0 supprimer\n for i in range(num_tables):\n table = geopackage.GetLayerByIndex(i)\n\n # V\u00e9rifier si la table est vide (aucun enregistrement)\n if table.GetFeatureCount() == 0:\n tables_to_delete.append(table.GetName())\n else:\n tables_to_update.append(table.GetName())\n\n # Supprimer les tables\n for table_name in tables_to_delete:\n geopackage.DeleteLayer(table_name)\n print(f\"Table supprim\u00e9e dans {geopackage_path}: {table_name}\")\n\n # Mettre \u00e0 jour les tables restantes\n for table_name in tables_to_update:\n table = geopackage.GetLayerByName(table_name)\n\n # Ajouter le premier champ texte\n champ1 = ogr.FieldDefn('enc_chart', ogr.OFTString)\n champ1.SetWidth(50)\n champ1.SetNullable(True)\n champ1.SetDefault(valeur_par_defaut_enc)\n table.CreateField(champ1)\n\n # Ajouter le deuxi\u00e8me champ texte\n champ2 = ogr.FieldDefn('scale', ogr.OFTReal)\n champ2.SetWidth(10)\n champ2.SetPrecision(2)\n champ2.SetDefault(valeur_par_defaut_scale)\n table.CreateField(champ2)\n\n # Mettre \u00e0 jour les valeurs dans les champs apr\u00e8s leur cr\u00e9ation\n for feature in table:\n feature.SetField('enc_chart', valeur_par_defaut_enc)\n feature.SetField('scale', valeur_par_defaut_scale)\n table.SetFeature(feature)\n print(f\"ENC & scale ajout\u00e9s \u00e0 {geopackage_path}: {table_name}\")\n\n # Fermer le GeoPackage\n geopackage = None\n\nelse:\n print(f\"Impossible d'ouvrir le GeoPackage {geopackage_path} en mode \u00e9dition.\")<\/code><\/pre>\n\n\n\n#Lignes \u00e0 \u00e9diter avant l'ex\u00e9cution du script<\/code><\/pre>\n\n\n\ngeopackage_paths = [\"c:\/testgpkg\/pointsENC.gpkg\", \"c:\/testgpkg\/LinesENC.gpkg\", \"c:\/testgpkg\/PolysENC.gpkg\"] valeur_par_defaut_enc = 'US1GC09M' valeur_par_defaut_scale = '2160000' delete_empty_tables_in_geopackages(geopackage_paths, valeur_par_defaut_enc, valeur_par_defaut_scale)<\/code><\/pre>\n\n\n\n <\/div><\/div>\n\n\n\n Be careful to respect the indentation of the Python code. To simplify the task, download the .py file directly from this link: delete_empty_tables_in_geopackages.py<\/a>.<\/p>\n\n\n\n For execution, be sure to modify the script’s input parameters:<\/p>\n\n\n\n <\/a><\/figure>\n\n\n\nYou will only need to enter the geopackage_paths the first time you run the script. On the other hand, the ENC map name and scale must be modified when importing each new S57 file. At the end of this run, you’ll have the new attributes in each table<\/p>\n\n\n\n<\/a><\/figure>\n\n\n\nWorkflow for loading several S57 files at once<\/h1>\n\n\n\nThe command lines in the previous paragraphs assume that you have a single S57 file to load. But you can load several files simultaneously. By using the -append -update options, you can loop through all the .000 files in a directory. You’ll then have all the files loaded in the import geopackages.<\/p>\n\n\n\n Unlike loading a single file, when loading a batch of files, the addition of the file name and scale must be integrated into the S57->geopackage transcoding loop. To do this, we’ll use a python script that adds the enc_chart and scale attributes to the tables created by the ogr2ogr command and populates them with the corresponding file name and a scale defined by the producer. The script also deletes any empty tables present in the import geopackage.<\/p>\n\n\n\n Thanks to Janez AVSEC for the idea of using the DSID table to retrieve the data compilation scale.<\/p>\n\n\n\n<\/div>Update available<\/div><\/div><\/div><\/p>\n\n\n\nAn update on this procedure is available in the article here.<\/a><\/p>\n\n\n\n <\/div><\/div>\n\n\n\n The .bat files must be run from an OSGeo4W window to benefit from the QGis Python environment.<\/p>\n\n\n\n The call to each .bat file is of the form:<\/p>\n\n\n\n .\\fichier.bat \u00ab\u00a0directory containing S57 files to be loaded\u00a0\u00bb \u00ab\u00a0name of import geopackage\u00a0\u00bb <\/em><\/strong><\/p>\n\n\n\n For example:<\/p>\n\n\n\n .\\mass_load_s57_polys.bat \u00ab\u00a0c:\/enc_db\/enc000\u00a0\u00bb \u00ab\u00a0c:\/enc_db\/polys.gpkg\u00a0\u00bb <\/em><\/strong><\/p>\n\n\n\n The .bat files for this task are as follows:<\/p>\n\n\n\n To import point type tables:<\/p>\n\n\n\n <\/div>Fichier .bat pour charger les g\u00e9om\u00e9tries point<\/div><\/div><\/div> <\/p>\n\n\n\n@echo off setlocal enabledelayedexpansion<\/code><\/pre>\n\n\n\nREM Assurez-vous qu'au moins 2 arguments ont \u00e9t\u00e9 fournis if \"%~2\"==\"\" ( echo Usage: %0 directory000 output_geopackage scale exit \/b 1 )<\/code><\/pre>\n\n\n\nREM R\u00e9cup\u00e9rer les arguments de la ligne de commande set \"directory=%~1\" set \"output_geopackage=%~2\"<\/code><\/pre>\n\n\n\nREM It\u00e9rez sur tous les fichiers .000 dans le r\u00e9pertoire for \/r \"%directory%\" %%i in (*.000) do ( echo Traitement du fichier: %%~ni set \"file=%%~ni\" ogr2ogr -f GPKG -skipfailures -append -update -where \"OGR_GEOMETRY='POINT' or OGR_GEOMETRY='MULTIPOINT'\" -oo SPLIT_MULTIPOINT=ON -oo RETURN_LINKAGES=ON -oo LNAM_REFS=ON -oo ADD_SOUNDG_DEPTH=ON -mapFieldType StringList=String,IntegerList=String \"%output_geopackage%\" \"%%i\" ogr2ogr -f GPKG -skipfailures -append -update \"%output_geopackage%\" \"%%i\" \"DSID\" ogr2ogr -f GPKG -skipfailures -append -update \"%output_geopackage%\" \"%%i\" \"C_AGGR\" python c:\/testgpkg\/update_geopackage_dsid.py \"%output_geopackage%\" \"!file!\" )<\/code><\/pre>\n\n\n\necho Traitement termin\u00e9. pause<\/code><\/pre>\n\n\n\n <\/div><\/div>\n\n\n\n You’ll notice that, in addition to the ogr2ogr command for importing point layers, we import two tables without geometry: DSID and C_AGGR. The DSID table contains two attributes: the S57 file name and the data compilation scale. This table will then be used by the Python script (update_geopackage_dsid.py) to fill each geopackage table with the original S57 file and compilation scale.<\/p>\n\n\n\n To import line-type tables:<\/p>\n\n\n\n <\/div>Fichier .bat pour charger les g\u00e9om\u00e9tries ligne<\/div><\/div><\/div> <\/p>\n\n\n\n@echo off setlocal enabledelayedexpansion<\/code><\/pre>\n\n\n\nREM Assurez-vous qu'au moins 2 arguments ont \u00e9t\u00e9 fournis\nif \"%~2\"==\"\" (\necho Usage: %0 directory000 output_geopackage scale\nexit \/b 1\n)<\/code><\/pre>\n\n\n\nREM R\u00e9cup\u00e9rer les arguments de la ligne de commande set \"directory=%~1\" set \"output_geopackage=%~2\" set \"scale=%~3\"<\/code><\/pre>\n\n\n\nREM It\u00e9rez sur tous les fichiers .000 dans le r\u00e9pertoire\nfor \/r \"%directory%\" %%i in (*.000) do (\necho Traitement du fichier: %%~ni\nset \"file=%%~ni\"\nogr2ogr -f GPKG -skipfailures -append -update -where \"OGR_GEOMETRY='LINESTRING' or OGR_GEOMETRY='MULTILINESTRING'\" -oo RETURN_LINKAGES=ON -oo LNAM_REFS=ON -mapFieldType StringList=String,IntegerList=String \"%output_geopackage%\" \"%%i\"\nogr2ogr -f GPKG -skipfailures -append -update \"%output_geopackage%\" \"%%i\" \"DSID\"\nogr2ogr -f GPKG -skipfailures -append -update \"%output_geopackage%\" \"%%i\" \"C_AGGR\"\npython c:\/testgpkg\/update_geopackage_dsid.py \"%output_geopackage%\" \"!file!\" \n)<\/code><\/pre>\n\n\n\necho Traitement termin\u00e9. pause<\/code><\/pre>\n\n\n\n<\/div><\/div>\n\n\n\n To import polygon tables:<\/p>\n\n\n\n <\/div>Fichier .bat pour charger les g\u00e9om\u00e9tries polygone<\/div><\/div><\/div> <\/p>\n\n\n\n@echo off setlocal enabledelayedexpansion<\/code><\/pre>\n\n\n\nREM Assurez-vous qu'au moins 2 arguments ont \u00e9t\u00e9 fournis\nif \"%~2\"==\"\" (\necho Usage: %0 directory000 output_geopackage scale\nexit \/b 1\n)<\/code><\/pre>\n\n\n\nREM R\u00e9cup\u00e9rer les arguments de la ligne de commande set \"directory=%~1\" set \"output_geopackage=%~2\" set \"scale=%~3\"<\/code><\/pre>\n\n\n\nREM It\u00e9rez sur tous les fichiers .000 dans le r\u00e9pertoire\nfor \/r \"%directory%\" %%i in (*.000) do (\necho Traitement du fichier: %%~ni\nset \"file=%%~ni\"\nogr2ogr -f GPKG -skipfailures -append -update -where \"OGR_GEOMETRY='POLYGON' or OGR_GEOMETRY='MULTIPOLYGON'\" -oo RETURN_LINKAGES=ON -oo LNAM_REFS=ON -mapFieldType StringList=String,IntegerList=String \"%output_geopackage%\" \"%%i\"\nogr2ogr -f GPKG -skipfailures -append -update \"%output_geopackage%\" \"%%i\" \"DSID\"\nogr2ogr -f GPKG -skipfailures -append -update \"%output_geopackage%\" \"%%i\" \"C_AGGR\"\npython c:\/testgpkg\/update_geopackage_dsid.py \"%output_geopackage%\" \"!file!\" \n)<\/code><\/pre>\n\n\n\necho Traitement termin\u00e9. pause<\/code><\/pre>\n\n\n\n <\/div><\/div>\n\n\n\n The .bat files contain the ogr2ogr command and a line to execute a Python script, update_geopackage_dsid.py. Be sure to change the path of this script in the .bat file to the one of your choice. This script adds the two attributes enc_chart and scale to all geopackage tables.<\/p>\n\n\n\n The python code used is as follows:<\/p>\n\n\n\n <\/div>update_geopackage_dsid.py<\/div><\/div><\/div> <\/p>\n\n\n\nimport sys from osgeo import ogr<\/p>\n\n\n\n def get_default_scale(geopackage_path, default_enc): # Ouvrir le GeoPackage en mode lecture seule geopackage = ogr.Open(geopackage_path, 0) default_scale = ‘0’ default_enc = default_enc +’.000′<\/p>\n\n\n\n if geopackage is not None:\n # R\u00e9cup\u00e9rer la couche DSID\n dsid_layer = geopackage.GetLayerByName('DSID')\n\n # V\u00e9rifier si la couche DSID existe\n if dsid_layer:\n # Rechercher la valeur de DSID_CSCL correspondant \u00e0 DSID_DSNM\n dsid_layer.SetAttributeFilter(f\"DSID_DSNM = '{default_enc}'\")\n feature = dsid_layer.GetNextFeature()\n if feature:\n default_scale = feature.GetField('DSPM_CSCL')\n\nreturn default_scale<\/code><\/pre>\n\n\n\ndef delete_empty_tables_in_geopackages_dsid(geopackage_path, default_enc): # R\u00e9cup\u00e9rer la valeur de scale \u00e0 partir de la table DSID default_scale = get_default_scale(geopackage_path, default_enc)<\/p>\n\n\n\n # Continuer avec le reste du code\nif default_scale is None:\n print(\"La valeur par d\u00e9faut de l'\u00e9chelle n'a pas \u00e9t\u00e9 trouv\u00e9e dans la table DSID.\")\n return\ndefault_scale_str = str(default_scale)\n# Ouvrir le GeoPackage en mode \u00e9dition\ngeopackage = ogr.Open(geopackage_path, 1)\n\nif geopackage is not None:\n # R\u00e9cup\u00e9rer le nombre de tables dans le GeoPackage\n num_tables = geopackage.GetLayerCount()\n\n # Liste des tables \u00e0 supprimer\n tables_to_delete = []\n tables_to_update = []\n\n # Identifier les tables \u00e0 supprimer\n for i in range(num_tables):\n table = geopackage.GetLayerByIndex(i)\n\n # V\u00e9rifier si la table est vide (aucun enregistrement)\n if table.GetFeatureCount() == 0:\n tables_to_delete.append(table.GetName())\n else:\n tables_to_update.append(table.GetName())\n\n # Supprimer les tables\n for table_name in tables_to_delete:\n geopackage.DeleteLayer(table_name)\n print(f\"Table supprim\u00e9e dans {geopackage_path}: {table_name}\")\n\n # Mettre \u00e0 jour les tables restantes\n for table_name in tables_to_update:\n table = geopackage.GetLayerByName(table_name)\n\n # V\u00e9rifier si le champ 'enc_chart' existe\n enc_chart_index = table.FindFieldIndex('enc_chart', 1)\n if enc_chart_index < 0:\n # Ajouter le champ 'enc_chart' s'il n'existe pas\n champ1 = ogr.FieldDefn('enc_chart', ogr.OFTString)\n champ1.SetWidth(50)\n champ1.SetNullable(True)\n champ1.SetDefault(default_enc)\n table.CreateField(champ1)\n\n # V\u00e9rifier si le champ 'scale' existe\n scale_index = table.FindFieldIndex('scale', 1)\n if scale_index < 0:\n # Ajouter le champ 'scale' s'il n'existe pas\n champ2 = ogr.FieldDefn('scale', ogr.OFTReal)\n champ2.SetWidth(10)\n champ2.SetPrecision(2)\n champ2.SetDefault(default_scale_str) # Convertir en nombre flottant\n table.CreateField(champ2)\n # Mettre \u00e0 jour les valeurs dans les champs seulement si les champs n'ont pas d\u00e9j\u00e0 de valeurs\n for feature in table:\n enc_chart_value = feature.GetField('enc_chart')\n scale_value = feature.GetField('scale')\n\n # V\u00e9rifier si les champs n'ont pas d\u00e9j\u00e0 de valeurs d\u00e9finies\n if enc_chart_value is None or scale_value is None:\n feature.SetField('enc_chart', default_enc)\n feature.SetField('scale', default_scale)\n table.SetFeature(feature)\n\n\n # Fermer le GeoPackage\n geopackage = None\n\nelse:\n print(f\"Impossible d'ouvrir le GeoPackage {geopackage_path} en mode \u00e9dition.\")<\/code><\/pre>\n\n\n\nif name<\/strong> == \u00ab\u00a0main<\/strong>\u00ab\u00a0: if len(sys.argv) != 3: print(\u00ab\u00a0Usage: python script.py geopackage_path default_enc\u00a0\u00bb) sys.exit(1)<\/p>\n\n\n\n geopackage_path = sys.argv[1]\ndefault_enc = sys.argv[2]\n\ndelete_empty_tables_in_geopackages_dsid(geopackage_path, default_enc)<\/code><\/pre>\n\n\n\n<\/div><\/div>\n\n\n\n <\/p>\n\n\n\n You could create a single .bat file for all three treatments, but you’ll need to check the results for each type of geometry.<\/p>\n\n\n\n You can download the .bat files and the Python code by clicking here.<\/a><\/p>\n\n\n\n <\/p>\n\n\n\n Continues regardless of the number of files loaded<\/h1>\n\n\n\nCloning and\/or adding imported tables to the ENC geopackage<\/h2>\n\n\n\nFor the first S57 load, the following Python script creates all the tables present in the three import geopackages. For subsequent loads, if the table already exists in the ENC geopackage, the records of the table in the import geopackage are added to the records already present in ENC. If, on the other hand, the table does not yet exist, it is created and populated with records from the import geopackage. In the ENC geopackage file, tables are prefixed with pt_, li_ and pl_ to indicate their geometry type.<\/p>\n\n\n\n Cloning is performed using the following code:<\/p>\n\n\n\n<\/div>Clonage ou mise \u00e0 jour .py<\/div><\/div><\/div> <\/p>\n\n\n\nfrom osgeo import ogr<\/code><\/pre>\n\n\n\ndef clone_or_append_tables_with_prefix(): input_geopackages = [\"c:\/testgpkg\/pointsENC.gpkg\", \"c:\/testgpkg\/LinesENC.gpkg\", \"c:\/testgpkg\/PolysENC.gpkg\"] prefixes = [\"pt_\", \"li_\", \"pl_\"]<\/code><\/pre>\n\n\n\nfor i in range(len(input_geopackages)):\n # Ouvrir le GeoPackage source en mode lecture\n input_gpkg = ogr.Open(input_geopackages[i], 0)\n\n if input_gpkg is not None:\n # Ouvrir le GeoPackage de destination en mode \u00e9criture\n output_gpkg = ogr.Open(\"c:\/testgpkg\/ENC.gpkg\", 1)\n prefix = prefixes[i]\n\n if output_gpkg is not None:\n # R\u00e9cup\u00e9rer le nombre de tables dans le GeoPackage source\n num_tables = input_gpkg.GetLayerCount()\n\n # Ajouter le contenu des tables avec le pr\u00e9fixe appropri\u00e9\n for j in range(num_tables):\n input_table = input_gpkg.GetLayerByIndex(j)\n output_table_name = f\"{prefix}{input_table.GetName()}\"\n input_table_name = input_table.GetName()\n\n # V\u00e9rifier si la table de destination existe d\u00e9j\u00e0\n output_table = output_gpkg.GetLayerByName(output_table_name)\n if output_table is None:\n # Cr\u00e9er une nouvelle couche de destination si elle n'existe pas\n output_table = output_gpkg.CreateLayer(output_table_name, geom_type=input_table.GetGeomType(), options=[\"OVERWRITE=YES\"])\n\n # Copiez la structure de la table d'origine vers la table temporaire\n input_layer = input_gpkg.GetLayerByName(input_table_name)\n input_layer_defn = input_layer.GetLayerDefn()\n\n for k in range(input_layer_defn.GetFieldCount()):\n field_defn = input_layer_defn.GetFieldDefn(k)\n output_table.CreateField(field_defn)\n\n # Copier les entit\u00e9s de la table source vers la table de destination\n for feature in input_table:\n output_feature = ogr.Feature(output_table.GetLayerDefn())\n output_feature.SetGeometry(feature.GetGeometryRef())\n\n # Copier les valeurs des champs\n for field_index in range(feature.GetFieldCount()):\n output_feature.SetField(field_index, feature.GetField(field_index))\n\n output_table.CreateFeature(output_feature)\n output_feature = None # Lib\u00e9rer la m\u00e9moire\n\n print(f\"Contenu ajout\u00e9 de {input_geopackages[i]}.{input_table.GetName()} vers ENC.gpkg.{output_table_name}\")\n\n # Fermer les GeoPackages\n input_gpkg = None\n output_gpkg = None\n else:\n print(f\"Impossible d'ouvrir le GeoPackage {output_geopackage} en mode \u00e9criture.\")\n else:\n print(f\"Impossible d'ouvrir le GeoPackage {input_geopackages[i]} en mode lecture.\")<\/code><\/pre>\n\n\n\n#Appel de la fonction append_tables_with_prefix<\/code><\/pre>\n\n\n\nclone_or_append_tables_with_prefix()<\/code><\/pre>\n\n\n\n <\/div><\/div>\n\n\n\n <\/p>\n\n\n\n You can download the .py file here: clone_or_append_tables_with_prefix.py.<\/a><\/p>\n\n\n\n Duplicate management<\/h2>\n\n\n\nWhen two ENC maps overlap, the entities present in the overlap zone will be inserted as many times as there are overlaps. To avoid duplicates and more, the database must be cleaned after each import of an S57 file.<\/p>\n\n\n\n But there are two different types of duplicates. On the one hand, overlapping areas between two maps of similar scale will produce \u00ab\u00a0real\u00a0\u00bb duplicates that need to be cleaned up. On the other hand, overlapping maps of very different scales need to be treated differently. The less detailed map (small scale) will have a selection of data from the more detailed map (large scale). In the latter case, it is not advisable to remove this information.<\/p>\n\n\n\n The Python code below will remove the \u00ab\u00a0true\u00a0\u00bb duplicates. For other duplicates, we recommend working with QGis filters. Depending on your needs, you can use the \u00ab\u00a0scale\u00a0\u00bb attribute we added when deleting the empty tables to define which data will or will not be displayed on your map.<\/p>\n\n\n\n<\/a><\/figure>\n\n\n\nIn this example, only entities corresponding to a scale between 12500 and 45000 will be displayed on your map.<\/p>\n\n\n\n This operation corresponds to the filtering of a layer, but\u2026 what do you do when you have 130 layers displayed? To save you having to perform this operation 130 times, here’s a Python code that lets you filter all the layers in your project:<\/p>\n\n\n\n<\/div>filter<br>.py<\/div><\/div><\/div> <\/p>\n\n\n\n#Importer le module QGIS<\/code><\/pre>\n\n\n\nfrom qgis.core import QgsProject<\/code><\/pre>\n\n\n\n#D\u00e9finir la valeur de l'attribut \"scale\" \u00e0 filtrer<\/code><\/pre>\n\n\n\nvaleur_echelle = 50000<\/code><\/pre>\n\n\n\n#Obtenir le projet actif<\/code><\/pre>\n\n\n\nprojet = QgsProject.instance()<\/code><\/pre>\n\n\n\n#Obtenir la liste des couches charg\u00e9es dans le projet<\/code><\/pre>\n\n\n\ncouches = projet.mapLayers().values()<\/code><\/pre>\n\n\n\n#Parcourir toutes les couches<\/code><\/pre>\n\n\n\nfor couche in couches: # V\u00e9rifier si la couche a un champ nomm\u00e9 \"scale\" if couche.fields().indexFromName('scale') != -1: # D\u00e9finir le filtre sur l'attribut \"scale\" \u00e9gal \u00e0 50000 filtre = f'\"scale\" = {valeur_echelle}' couche.setSubsetString(filtre) print(f\"Filre d\u00e9fini pour la couche {couche.name()}\") else: print(f\"La couche {couche.name()} n'a pas d'attribut 'scale'\") <\/code><\/pre>\n\n\n\n<\/div><\/div>\n\n\n\n And another to remove the filter from all displayed layers:<\/p>\n\n\n\n <\/div>unfilter<br>.py<\/div><\/div><\/div> <\/p>\n\n\n\n# Importer le module QGIS\nfrom qgis.core import QgsProject\n\n# Obtenir le projet actif\nprojet = QgsProject.instance()\n\n# Obtenir la liste des couches charg\u00e9es dans le projet\ncouches = projet.mapLayers().values()\n\n# Parcourir toutes les couches\nfor couche in couches:\n # V\u00e9rifier si la couche est une couche vectorielle\n if couche.type() == QgsMapLayer.VectorLayer:\n # Supprimer le filtre de la couche\n couche.setSubsetString('')\n print(f\"Filtre supprim\u00e9 pour la couche {couche.name()}\")\n<\/code><\/pre>\n\n\n\n<\/div><\/div>\n\n\n\n <\/p>\n\n\n\n Download the python codes here.<\/a><\/p>\n\n\n\n The following code creates a temporary table (temp_table) for each table in the ENC geopackage. It then fills this table with the separate LNAM-based records. It then deletes the original table and renames the temporary table with the name of the original table.<\/p>\n\n\n\n This treatment applies to all S57 tables, except:<\/p>\n\n\n\n \nfire (LIGHTS) and bathymetric sounding (SOUNDG) tables, where the LNAM attribute is not useful. In fact, each fire sector has the same LNAM and name_rcid, but corresponds to a different color and geographic sector. Similarly, the LNAM and name_rcid of bathymetric soundings are the same for all sounding points in the original multipoint entity. In the first case, we need to search for duplicate colors and sectors for the same name_rcid, and for probes we need to find probes with the same XY in their geometry.<\/li>\n\n\n\n SEAARE, LNDRGN and ADMARE tables, which cover large areas and are split according to the map footprint, while keeping the same LNAM. In this case, even if the LNAM is the same, the polygons do not have the same shape and area.<\/li>\n\n\n\nDSID and C_AGGR tables, which have no LNAM and, by principle, cannot have duplicates.<\/li>\n<\/ul>\n\n\n\n<\/div>effacer doublons .py<\/div><\/div><\/div> <\/p>\n\n\n\nfrom osgeo import ogr<\/code><\/pre>\n\n\n\ndef delete_doublons(): # Ouvre le GeoPackage en mode mise \u00e0 jour geopackage_path = \"c:\/enc_db\/enc2.gpkg\" geopackage = ogr.Open(geopackage_path, update=True)<\/code><\/pre>\n\n\n\nif geopackage is not None:\n # R\u00e9cup\u00e8re la liste de noms de toutes les tables du GeoPackage\n table_names = [geopackage.GetLayerByIndex(i).GetName() for i in range(geopackage.GetLayerCount())]\n\n # Boucle sur les tables et les traite individuellement\n for table_name in table_names:\n if table_name not in [\"pl_SEAARE\",\"pl_LNDRGN\",\"pl_ADMARE\",\"layer_styles\",\"natsurf\",\"DSID\",\"C_AGGR\"]:\n # Cr\u00e9e une table temporaire en ajoutant un suffixe \"_temp\"\n temp_table_name = table_name + \"_temp\"\n temp_table = geopackage.CreateLayer(temp_table_name)\n # Copie la structure de la table d'origine vers la table temporaire\n input_layer = geopackage.GetLayerByName(table_name)\n input_layer_defn = input_layer.GetLayerDefn()\n\n for i in range(input_layer_defn.GetFieldCount()):\n field_defn = input_layer_defn.GetFieldDefn(i)\n temp_table.CreateField(field_defn)\n if table_name not in [\"pt_LIGHTS\", \"pt_SOUNDG\"]:\n attribut = \"LNAM\"\n geopackage.ExecuteSQL(f\"INSERT INTO {temp_table_name} SELECT * FROM {table_name} GROUP BY {attribut};\")\n elif table_name == \"pt_SOUNDG\":\n attribut = \"name_rcid,depth, ST_X(geom),ST_Y(geom)\"\n geopackage.ExecuteSQL(f\"INSERT INTO {temp_table_name} SELECT * FROM {table_name} GROUP BY {attribut};\")\n elif table_name == \"pt_LIGHTS\":\n attribut = \"name_rcid, colour, sectr1, sectr2\"\n geopackage.ExecuteSQL(f\"INSERT INTO {temp_table_name} SELECT * FROM {table_name} GROUP BY {attribut};\")\n geopackage.ExecuteSQL(f\"DROP TABLE {table_name}\")\n\n # Renomme la table temporaire pour qu'elle ait le m\u00eame nom que la table d'origine\n geopackage.ExecuteSQL(f\"ALTER TABLE {temp_table_name} RENAME TO {table_name}\")\n print(f\"alter table {table_name}\")\n\n # Ferme le GeoPackage\n geopackage = None\nelse:\n print(f\"Impossible d'ouvrir le GeoPackage {geopackage} en mode \u00e9criture.\")<\/code><\/pre>\n\n\n\n#Appel de la fonction delete_doublons<\/code><\/pre>\n\n\n\ndelete_doublons()<\/code><\/pre>\n\n\n\n <\/div><\/div>\n\n\n\n <\/p>\n\n\n\n You can download the .py file from this link: supprime_doublons.py.<\/a> Don’t forget to change the paths and filenames to match your own. Note also the line :<\/p>\n\n\n\n if table_name not in [\u00ab\u00a0pl_SEAARE\u00a0\u00bb, \u00ab\u00a0pl_LNDRGN\u00a0\u00bb, \u00ab\u00a0pl_ADMARE\u00a0\u00bb, \u00ab\u00a0layer_styles\u00a0\u00bb, \u00ab\u00a0natsurf\u00a0\u00bb,\u00a0\u00bbDSID\u00a0\u00bb,\u00a0\u00bbC_AGGR\u00a0\u00bb]:<\/em><\/strong><\/p>\n\n\n\n In principle, you’ll only have the prefixed tables and the layer_styles and natsurf tables. This line prevents pl_SEAARE, pl_LNDRGN and pl_ADMARE from being processed, but more importantly, it prevents the layer_styles and natsurf tables we use for symbologies from being emptied. If you need to create tables other than those from the s57 file in the geopackage, you need to take into account the names of these tables in this script, by adding them to this line of exceptions.<\/p>\n\n\n\n <\/p>\n","protected":false},"excerpt":{"rendered":" The first part of the Collaborative Financing for the Integration of Marine Data in QGIS project has been completed, thanks to contributions from ORANGE Marine, Geoconceptos Uruguay, Janez Avzec and Gaetan Mas. Our warmest thanks to…<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"give_campaign_id":0,"_bbp_topic_count":0,"_bbp_reply_count":0,"_bbp_total_topic_count":0,"_bbp_total_reply_count":0,"_bbp_voice_count":0,"_bbp_anonymous_reply_count":0,"_bbp_topic_count_hidden":0,"_bbp_reply_count_hidden":0,"_bbp_forum_subforum_count":0,"sfsi_plus_gutenberg_text_before_share":"","sfsi_plus_gutenberg_show_text_before_share":"","sfsi_plus_gutenberg_icon_type":"","sfsi_plus_gutenberg_icon_alignemt":"","sfsi_plus_gutenburg_max_per_row":"","_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[1377],"tags":[1995,2795,1449,2721,2793,2715,2036],"class_list":["post-12510","post","type-post","status-publish","format-standard","hentry","category-qgis-en","tag-geopackage-en","tag-ogr2ogr-en","tag-python-en","tag-s57-en","tag-script-en","tag-symbology","tag-tutorial-en"],"aioseo_notices":[],"jetpack_featured_media_url":"","jetpack_shortlink":"https:\/\/wp.me\/p6XU0A-3fM","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/posts\/12510","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/comments?post=12510"}],"version-history":[{"count":0,"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/posts\/12510\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/media?parent=12510"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/categories?post=12510"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.sigterritoires.fr\/index.php\/wp-json\/wp\/v2\/tags?post=12510"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}