Update build boundaries #418

lacardonap · 2021-02-19T16:08:24Z

Actualización del algoritmo que permite construir los linderos

Implementar modelo de procesamiento
Actualizar tests
Soporte para linderos seleccionados
Remover dependencia de los linderos con las tablas de topolofía almacenada (puntosccl, masccl y menosccl)
Actualizar capa de linderos
Resolver inquietudes.
Agregar tests para comprobar llenado de tablas topológicas (con y sin selección)
Corregir ETL model_build_boundaries.model3
Agregar barra de progreso para la toolbar (construir linderos y llenar tablas de topología) Resuelve issue Comunicar estado en la herramienta construir linderos #333

lacardonap · 2021-02-19T17:23:17Z

Se esta trabajando en la implementación del nuevo algoritmo de construcción de linderos, en un principio se desea conservar el comportamiento que tenia el algoritmo existente en el cual se le permitia al usuario realizar una selección de linderos antes de proceder a construir los linderos.

Realizar la selección de linderos tiene la ventaja de que si el conjunto de datos es muy grande no es necesario correr todo el algoritmo para toda la base de datos sino unicamente para los linderos seleccionados. Parece una buena estragería, el problema que actualmente existe al adoptarla es que al no seleccionar todos los linderos que estan conectados el resultado obtenido no es el esperado.

Por ejemplo, se puede apreciar en el siguiente gif que al generar los linderos para la zona seleccionada el total de linderos es 58602. Pero si lo genero para toda la base de datos el total de linderos es 58613.

A partir de los linderos seleccionados por el usuario se realiza una selección por localización y se seleccionan los linderos que estan conectados pero puede ocurrir lo siguiente: los linderos seleccionados no incluyen todos los linderos vecinos razón por la cual la construcción no es la correcta.

lacardonap · 2021-02-19T17:24:05Z

Actualmente si doy clic en el botón de construir linderos y la capa de linderos no esta cargada se desplega un mensaje en el cual se le dice a la persona que primero se debe cargar la capa de linderos. Propongo que este mensaje no se muestre al usuario y se cargue automáticamente la capa

gacarrillor

Could we rebase this PR (on lev_cat_1_1) to make it ready to be reviewed/merged?
That would also show us the unit tests status.

lacardonap · 2021-05-25T17:22:50Z

Procedere con la actualización 👍

lacardonap · 2021-05-26T19:52:13Z

This PR was rebase using lev_cat_1_1. Now this one is ready to be reviewed/merged.

gacarrillor

We've never used [feat] in our commits. For your information, inside the square brackets we are get used to add the module name.

asistente_ladm_col/gui/toolbar.py

asistente_ladm_col/lib/geometry.py

asistente_ladm_col/utils/decorators.py

gacarrillor · 2021-06-02T22:07:32Z

I'm getting errors on GNU/Linux, using QGIS v3.16.5 and master:

The result of the Build Boundaries model is a layer which has its field names truncated, like in the image below:

Because of that, the ETL cannot run well, and I get errors like the following in the console:

../src/core/qgsmessagelog.cpp:29 : (logMessage) [4ms] 2021-06-02T16:56:21 Messages[2] Layer Lindero : OGR error creating feature 40: failed to execute insert : NOT NULL constraint failed: lc_lindero.comienzo_vida_util_version
ERROR 1: failed to execute insert : NOT NULL constraint failed: lc_lindero.comienzo_vida_util_version
../src/core/vector/qgsvectordataprovider.cpp:756 : (pushError) [0ms] OGR error creating feature 41: failed to execute insert : NOT NULL constraint failed: lc_lindero.comienzo_vida_util_version

We cannot merge until we know what's happening and fix it.

This should be done on top of the latest commit (7830fec). Testing on your latest commit (392b326) gave me the same results.

gacarrillor · 2021-06-02T22:09:33Z

Apart from that, it would be required to have a test for the toolbar's method called build_boundary(). The only test added by this PR is testing the model, but we also need to test all the logic involved in build_boundary(), that is:

Are topology tables finally updated?
Is the boundary layer updated (number of features should be the same before and after the process)?

Ideally the test should act in two modes: using a selection and using all the features.

lacardonap · 2021-06-03T16:02:38Z

Thank you very much for the review and adjustments made. I will proceed to review the observations. I would like to confirm if the dataset with which the error was generated corresponds to the structured data for the tutorial?

gacarrillor · 2021-06-04T19:41:57Z

I would like to confirm if the dataset with which the error was generated corresponds to the structured data for the tutorial?

Yes, but migrated to Levantamiento Catastral v1.1.

lacardonap · 2021-06-22T18:22:16Z

I'm getting errors on GNU/Linux, using QGIS v3.16.5 and master:

Build_Boundaries model was updated to prevent truncated data.
It can be tested in the commit ea0f7c4

lacardonap · 2021-06-24T16:43:15Z

I have designed the test to validate the entire build_boundaries function. But I am having a problem trying to run the test because the iface object is null.

Do you think it is a good idea to validate the whole build_boundaries function?

https://gist.github.com/lacardonap/157b1aa7f36b0e178758d2f00a3e75e5

I am getting this error when running the test:

======================================================================
ERROR: test_build_boundaries (asistente_ladm_col.tests.test_toolbar.TestToolbar)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/qgis/utils.py", line 359, in startPlugin
    plugins[packageName].initGui()
  File "/usr/share/qgis/python/plugins/processing/ProcessingPlugin.py", line 181, in initGui
    iface.registerOptionsWidgetFactory(self.options_factory)
AttributeError: 'NoneType' object has no attribute 'registerOptionsWidgetFactory'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/src/asistente_ladm_col/tests/test_toolbar.py", line 64, in test_build_boundaries
    self.toolbar.build_boundary(self.db_gpkg)
  File "/usr/src/asistente_ladm_col/gui/toolbar.py", line 121, in build_boundary
    self.app.core.run_etl_model_in_backgroud_mode(db, build_boundaries_layer, db.names.LC_BOUNDARY_T)
  File "/usr/src/asistente_ladm_col/utils/decorators.py", line 209, in decorated_function
    startPlugin('processing')
  File "/usr/lib/python3/dist-packages/qgis/utils.py", line 362, in startPlugin
    _unloadPluginModules(packageName)
  File "/usr/lib/python3/dist-packages/qgis/utils.py", line 448, in _unloadPluginModules
    mods = _plugin_modules[packageName]
KeyError: 'processing'

Any advice is welcome. Thanks

gacarrillor · 2021-06-28T16:32:41Z

Do you think it is a good idea to validate the whole build_boundaries function?

Do you think it's a good idea to delete features from the user layer and do not try to recover those deleted features if any failure occurs while running the model?

Could you commit your test to another branch (created from update_build_boundaries), so that I can inspect it?

lacardonap · 2021-06-29T14:36:04Z

build_boundaries function must be tested, what I meant is if we test the whole function or check it by parts. After thinking about it, it is better to check it all.

I appreciate the support in reviewing the test which was left in the branch https://github.com/SwissTierrasColombia/Asistente-LADM-COL/tree/test_toolbar_build_boundaries, which never ever worked for me locally. I am attentive to what is required.

gacarrillor

The test test_build_boundaries does not fail anymore due to processing plugin. But it's throwing other errors now, that I'll pass for you to adjust.

Additionally, the toolbar.build_boundary() method should be adjusted:

Don't delete features until you know you have obtained the new features from the processing model.
When deleting features, keep them in a variable because if the insertion of new features fails, you should restore the deleted features. The user should never end up with an incomplete data set (which you can see in https://user-images.githubusercontent.com/652785/120557677-e7cddf00-c3c3-11eb-8800-44f61e778a07.png).

lacardonap · 2021-07-08T16:24:40Z

Thank you for your support. I continue with the requested adjustments.

lacardonap · 2021-07-12T15:50:56Z

The requested adjustments are made. Please continue with the review. I stay tuned for what is required

gacarrillor · 2021-07-12T20:08:18Z

asistente_ladm_col/gui/toolbar.py

@@ -118,7 +118,7 @@ def build_boundary(self, db):
                    layers[db.names.LC_BOUNDARY_T].deleteSelectedFeatures()

                # Bring back the features we deleted before, but this time, with the boundaries fixed
-                self.app.core.run_etl_model_in_backgroud_mode(db, build_boundaries_layer, db.names.LC_BOUNDARY_T)
+                self.app.core.run_etl_model_in_backgroud_mode(db, build_boundaries_layer, db.names.LC_BOUNDARY_T, False)


If something has to change, it has to be the call from the tests, not the one from core...

Thank you very much for the feedback. The reply is late but I prefer to respond.
You are right, the suggested adjustment was made and currently in the commit 513f31a

gacarrillor · 2021-07-12T20:36:59Z

asistente_ladm_col/gui/toolbar.py

-                # Bring back the features we deleted before, but this time, with the boundaries fixed
-                self.app.core.run_etl_model_in_backgroud_mode(db, build_boundaries_layer, db.names.LC_BOUNDARY_T, False)
+                # Build boundaries should have generated at least one boundary.
+                if selected_boundaries_count > 0 and build_boundaries_count > 0:


And what if not? Any message for users?

Adjustment made, available in commit 513f31a

gacarrillor · 2021-07-12T20:38:39Z

asistente_ladm_col/core/app_core_interface.py

@@ -1152,7 +1152,7 @@ def copy_csv_to_db(self, csv_layer, db, target_layer_name):
        return True

    @_activate_processing_plugin
-    def run_etl_model_in_backgroud_mode(self, db, input_layer, ladm_col_layer_name):
+    def run_etl_model_in_backgroud_mode(self, db, input_layer, ladm_col_layer_name, force_reprojection=True):


Don't get it. If you need to adjust something here is to set the proper CRS to your layer before calling this method. The method run_etl_model_in_backgroud_mode can remain without changes.

Is there any problem trying to set a proper CRS to your layer?
You should use get_ctm12_crs().

You are right, the suggested adjustment was made and currently in the commit 513f31a

gacarrillor · 2021-07-12T20:40:53Z

asistente_ladm_col/gui/toolbar.py

+                        layers[db.names.LC_BOUNDARY_T].dataProvider().truncate()
+
+                        # Original data is restored because an error occurred while trying to build the boundaries
+                        self.app.core.run_etl_model_in_backgroud_mode(db, copy_boundary_layer, db.names.LC_BOUNDARY_T, False)


Just restore selected features, no need to do it for the whole layer...

wooow thanks for the comment, you are absolutely right, adjustment made in the commit 513f31a

gacarrillor · 2021-07-12T20:41:37Z

asistente_ladm_col/gui/toolbar.py

-                with edit(layers[db.names.LC_BOUNDARY_T]):
-                    # Delete selected features as they will be imported again from a newly created layer after processed
-                    layers[db.names.LC_BOUNDARY_T].deleteSelectedFeatures()
+                copy_boundary_layer = self.app.core.get_layer_copy(layers[db.names.LC_BOUNDARY_T])


I disagree. Imagine a layer of 1000 boundaries and you select 4 of them before clicking on "Build Boundaries"...

Adjustment made, available in commit 513f31a

gacarrillor · 2021-07-12T20:42:44Z

asistente_ladm_col/gui/toolbar.py

+                    # Bring back the features we deleted before, but this time, with the boundaries fixed
+                    self.app.core.run_etl_model_in_backgroud_mode(db, build_boundaries_layer, db.names.LC_BOUNDARY_T, False)
+
+                    # check if features were insert successful


successfully

Adjustment made, available in commit 513f31a

gacarrillor · 2021-07-12T20:43:55Z

asistente_ladm_col/gui/toolbar.py

+
+                    else:
+                        # Clean layer because wrong data could be insert previously
+                        layers[db.names.LC_BOUNDARY_T].dataProvider().truncate()


You're removing all features?

What if the user just selected 1% of them?

the change was made, now only the selected boundaries are deleted and if an error occurs they are restored. Available in commit 513f31a

gacarrillor · 2021-07-12T21:03:46Z

asistente_ladm_col/gui/toolbar.py

+                                                                                  "{} feature(s) was(were) analyzed generating {} boundary(ies)!").format(num_boundaries, build_boundaries_layer.featureCount()))
+
+                    else:
+                        # Clean layer because wrong data could be insert previously


could have been inserted previously

Adjustment made, available in commit 513f31a

gacarrillor · 2021-07-12T21:07:44Z

asistente_ladm_col/tests/test_toolbar.py

+                             'Features count does not match for the layer {}'.format(layer_name))
+
+    def test_build_boundaries_with_empty_geom(self):
+        print('\nINFO: Validating build boundaries with features with null geometry...')


Unclear what you're testing.

What is the result you expect? The NULL geometry is ignored?

the information given in the tests has been updated. This test validates that if there is a boundary without geometry it must be removed. vailable in commit 513f31a

gacarrillor · 2021-07-12T21:08:11Z

asistente_ladm_col/tests/test_toolbar.py

+                         'Null features count does not match for the layer boundary layer')
+
+        boundary_layer.selectAll()
+        self.toolbar.build_boundary(db_gpkg)


The method build_boundaries should be called like this:

build_boundary(db_gpkg, False) # where False is do not use selection

Adjustment made, available in commit 513f31a

gacarrillor · 2021-07-12T21:11:39Z

We are wasting too much time in this review. Please double (and triple) check before submitting to a review.

gacarrillor

The test is not passing.
To keep the commit history clean, squash all commits since 34a5e0f on.

gacarrillor · 2021-07-21T20:42:25Z

asistente_ladm_col/gui/toolbar.py

@@ -45,7 +45,14 @@ def __init__(self, iface):
        self.app = AppInterface()
        self.geometry = GeometryUtils()

-    def build_boundary(self, db):
+    def build_boundary(self, db, use_user_selection=True):


False by default, since the expected and recommended way to run this method is without a selection.

Change use_user_selection by use_selection like in other methods in the same class and like in GeometryUtils().

Change use_user_selection by use_selection like in other methods in the same class and like in GeometryUtils().

The build_boundary method is designed to display a dialog box to the user indicating that no selection has been made and asking if the user wants to use all boundaries. When there are selected boundaries, no dialog box is shown and the selected boundaries are used. The purpose of using the parameter use_user_selection in the function is to omit the user selection and use all the records of the layer.

It is proposed to change use_user_selection to skip_selection where:

skip_selection (Boolean): True if we omit the boundaries selected by the user, False if we validate the user's selection.

gacarrillor · 2021-07-21T20:57:04Z

asistente_ladm_col/gui/toolbar.py

@@ -58,6 +65,9 @@ def build_boundary(self, db):
            }
            self.app.core.get_layers(db, layers, load=True)

+            if not use_user_selection:
+                layers[db.names.LC_BOUNDARY_T].selectAll()


Doesn't seem to make sense.

If use_selection is False, you shouldn't alter selected layer features nor use them in any way. Instead you should use always all layer features (e.g., when creating a copy, when counting features, etc.).

You are right, the suggested adjustment was made and currently in the commit 513f31a

gacarrillor · 2021-07-21T21:00:04Z

asistente_ladm_col/gui/toolbar.py

-                        layers[db.names.LC_BOUNDARY_T].dataProvider().truncate()
+                        if layers[db.names.LC_BOUNDARY_T].featureCount() != boundaries_count - selected_boundaries_count:
+                            # Clean layer because wrong data could have been inserted previously
+                            expr = "{} IN ('{}')".format(db.names.T_ILI_TID_F, "','".join([str(t_ili_tid) for t_ili_tid in boundary_t_ili_tids]))


Not sure if you actually need str(t_ili_tid), since t_ili_tids are already strings.

data type of t_ili_tid is string, no conversion is necessary. Adjustment made, available in commit 513f31a

gacarrillor · 2021-07-21T21:02:09Z

asistente_ladm_col/gui/toolbar.py

+                            # Clean layer because wrong data could have been inserted previously
+                            expr = "{} IN ('{}')".format(db.names.T_ILI_TID_F, "','".join([str(t_ili_tid) for t_ili_tid in boundary_t_ili_tids]))
+                            layers[db.names.LC_BOUNDARY_T].selectByExpression(expr)
+                            layers[db.names.LC_BOUNDARY_T].invertSelection()


If you need to select and then invert the selection, then the expression is wrong.

Perhaps you need expr = "{} NOT IN ('{}')"..., don't you?

Thanks for the observation, the selection is much more optimal. Adjustment made, available in commit 513f31a

gacarrillor · 2021-07-21T21:03:39Z

asistente_ladm_col/gui/toolbar.py


-                        # Original data is restored because an error occurred while trying to build the boundaries
-                        self.app.core.run_etl_model_in_backgroud_mode(db, copy_boundary_layer, db.names.LC_BOUNDARY_T, False)
+                        # the previously deleted boundaries are restored because an error occurred when trying to insert the building boundaries


Any chance to write a unit test for that case?

I am trying to generate a data set that will allow us to test this part of the cosw. It has not been possible, but I would also like to have a test that checks this.

gacarrillor · 2021-07-21T21:06:36Z

asistente_ladm_col/lib/geometry.py

-        return processing.run("model:Build_Boundaries", params)['native:refactorfields_2:built_boundaries']
+        build_boundary_layer = processing.run("model:Build_Boundaries", params)['native:refactorfields_2:built_boundaries']
+
+        # For CTM12 the output layer remains without projection. The projection of the input layer is assigned


Is it necessary?

Wasn't it only an issue for unit tests?

If you really need to set the CRS here, just use layer.setCrs().

gacarrillor · 2021-07-21T21:08:14Z

asistente_ladm_col/tests/test_toolbar.py

@@ -105,8 +105,7 @@ def test_build_boundaries_with_empty_geom(self):
        self.assertEqual(boundary_layer.selectedFeatureCount(), 1,
                         'Null features count does not match for the layer boundary layer')

-        boundary_layer.selectAll()
-        self.toolbar.build_boundary(db_gpkg)
+        self.toolbar.build_boundary(db_gpkg, False)


This call should be simply self.toolbar.build_boundary(db_gpkg).

Adjustment made, available in commit 513f31a

gacarrillor · 2021-07-21T21:09:30Z

asistente_ladm_col/gui/toolbar.py

@@ -45,7 +45,14 @@ def __init__(self, iface):
        self.app = AppInterface()
        self.geometry = GeometryUtils()

-    def build_boundary(self, db):
+    def build_boundary(self, db, use_user_selection=True):


You can rename the method to build_boundaries(), since that's what the method does.

Adjustment made, available in commit 513f31a

gacarrillor · 2021-07-21T21:12:19Z

asistente_ladm_col/tests/test_toolbar.py

@@ -22,6 +22,7 @@ def setUpClass(cls):
        import_qgis_model_baker()
        import_asistente_ladm_col()  # Import plugin
        cls.app = AppInterface()
+        cls.app.core.initialize_ctm12()  # We need to initialize CTM12


What if you initialize CTM12 after this line instead (and remove any other call to such method in unit tests)?

What a good idea, adjustment made. initialize CTRM12 from other classes were removed. Available in commit 513f31a

lacardonap · 2021-07-21T22:28:33Z

@gacarrillor thank you very much for the review, I am checking why if I run only the toolbar tests it works correctly but if I run all the tests it fails.

I will proceed with the requested adjustments

lacardonap · 2021-07-22T20:24:45Z

The problem was solved by moving the test test_build_boundaries from the test_geometry_utils class to test_build_boundaries (Note: the test_toolbar class was renamed).

lacardonap · 2021-07-23T22:16:10Z

I am currently trying to generate test data to extend the test coverage. The last commit will correspond to the squash all commits since 34a5e0f on. Again, thanks for the review.

lacardonap · 2021-07-27T16:37:15Z

In trying to generate a dataset that would allow to extend the coverage of the tests, some improvements were identified and implemented and the corresponding tests were developed. The suggested adjustments are already implemented but to date I have not been able to generate a test data set that fails (the grass algorithm is very good).

You can now proceed with the revision. Thanks

…ng the former inefficient method; add tests

…dd test

… the build_boundaries function and include tests to check build boundaries function with invalid data

…gy tables)

lacardonap added the work in progress label Feb 19, 2021

lacardonap changed the base branch from master to lev_cat_1_1 February 19, 2021 16:12

lacardonap added Question and removed Question work in progress labels Feb 19, 2021

lacardonap requested a review from gacarrillor February 23, 2021 20:24

gacarrillor requested changes May 25, 2021

View reviewed changes

lacardonap self-assigned this May 25, 2021

lacardonap force-pushed the update_build_boundaries branch from e5769c6 to fb6dbeb Compare May 26, 2021 17:44

lacardonap requested a review from gacarrillor May 26, 2021 20:31

gacarrillor added the Under Review Don't add new commits, this is under review. label Jun 2, 2021

gacarrillor reviewed Jun 2, 2021

View reviewed changes

gacarrillor force-pushed the update_build_boundaries branch from fb6dbeb to 392b326 Compare June 2, 2021 19:43

gacarrillor added a commit that referenced this pull request Jun 2, 2021

Several adjustments to PR #418

4b90445

gacarrillor added a commit that referenced this pull request Jun 2, 2021

Several adjustments to PR #418

7830fec

gacarrillor force-pushed the update_build_boundaries branch from 4b90445 to 7830fec Compare June 2, 2021 21:50

gacarrillor added Not ready for a Merge This PR has errors or pending tasks that must be addressed and removed Under Review Don't add new commits, this is under review. labels Jun 2, 2021

gacarrillor added a commit that referenced this pull request Jul 8, 2021

Several adjustments to PR #418

49a6485

gacarrillor force-pushed the update_build_boundaries branch from 53a8209 to 0f722d1 Compare July 8, 2021 16:05

gacarrillor requested changes Jul 8, 2021

View reviewed changes

gacarrillor requested changes Jul 12, 2021

View reviewed changes

gacarrillor requested changes Jul 21, 2021

View reviewed changes

lacardonap added the work in progress label Jul 22, 2021

lacardonap force-pushed the update_build_boundaries branch from 3609a57 to 1adde0c Compare July 26, 2021 22:17

lacardonap removed the work in progress label Jul 27, 2021

lacardonap and others added 6 commits September 2, 2021 09:23

[quality] Add model based on GRASS clean to build boundaries, replaci…

77283e2

…ng the former inefficient method; add tests

Several adjustments to PR #418

a37e1e2

[quality] Build boundaries model updated to prevent truncated data; a…

2c3e87f

…dd test

[tests] Make processing pass decorator checks

775f243

[tests] Minor adjustments to test test_build_boundaries

12c3bfa

[toolbar] Boundaries are restored if an error occurred when executing…

a6ea115

… the build_boundaries function and include tests to check build boundaries function with invalid data

lacardonap force-pushed the update_build_boundaries branch from 1adde0c to a6ea115 Compare September 2, 2021 14:33

lacardonap added the work in progress label Sep 2, 2021

lacardonap removed the work in progress label Sep 10, 2021

lacardonap requested review from gacarrillor and removed request for gacarrillor September 10, 2021 02:35

[toolbar] Add progress bar for toolbar (build boundaries, fill topolo…

6791d59

…gy tables)

lacardonap force-pushed the update_build_boundaries branch from 4755316 to 6791d59 Compare September 10, 2021 15:27

lacardonap requested a review from gacarrillor September 10, 2021 15:44

Base automatically changed from lev_cat_1_1 to master February 27, 2022 01:03

Update build boundaries #418

Are you sure you want to change the base?

Update build boundaries #418

Conversation

lacardonap commented Feb 19, 2021 • edited Loading

lacardonap commented Feb 19, 2021

lacardonap commented Feb 19, 2021

gacarrillor left a comment

Choose a reason for hiding this comment

lacardonap commented May 25, 2021 • edited Loading

lacardonap commented May 26, 2021 • edited Loading

gacarrillor left a comment • edited Loading

Choose a reason for hiding this comment

gacarrillor commented Jun 2, 2021 • edited Loading

gacarrillor commented Jun 2, 2021

lacardonap commented Jun 3, 2021

gacarrillor commented Jun 4, 2021

lacardonap commented Jun 22, 2021

lacardonap commented Jun 24, 2021

gacarrillor commented Jun 28, 2021

lacardonap commented Jun 29, 2021

gacarrillor left a comment

Choose a reason for hiding this comment

lacardonap commented Jul 8, 2021

lacardonap commented Jul 12, 2021 • edited by gacarrillor Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

gacarrillor Jul 12, 2021 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

gacarrillor commented Jul 12, 2021

gacarrillor left a comment • edited by lacardonap Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

lacardonap commented Jul 21, 2021

lacardonap commented Jul 22, 2021

lacardonap commented Jul 23, 2021

lacardonap commented Jul 27, 2021 • edited Loading

lacardonap commented Feb 19, 2021 •

edited

Loading

lacardonap commented May 25, 2021 •

edited

Loading

lacardonap commented May 26, 2021 •

edited

Loading

gacarrillor left a comment •

edited

Loading

gacarrillor commented Jun 2, 2021 •

edited

Loading

lacardonap commented Jul 12, 2021 •

edited by gacarrillor

Loading

gacarrillor Jul 12, 2021 •

edited

Loading

gacarrillor left a comment •

edited by lacardonap

Loading

lacardonap commented Jul 27, 2021 •

edited

Loading