Page Comparison

CleverMaps Shell is controlled solely by a set of predefined commands. These commands can be divided in 4 categories, or workflow states. All commands and parameters are case sensitive.

Each command can be further specified by parameters, which can have default values. Each parameter also has a "--" prefix, which is a technicality, and is not mentioned in the tables below, for the sake of readability.

Parameters with string type take string input as a value, if they are mentioned. Parameters with enum type have predefined enumeration of strings, which can be passed as a value. Parameters with no type (-) are passed no values.

Workflow states

• Started - you have started the tool
• Connected to server - you have successfully logged in to your account on a specific server
• Opened project - you have opened a project you have access to
• Opened dump - you have created a new dump, or opened an existing one

_

...

_

Started state

login

Log in as a user with correct credentials.

...

email

...

...

...

...

server to connect to

default = https://secure.clevermaps.io

...

...

proxy server hostname

...

...

setup

Store your config and credentials in a file so you don't have to specify them each time you log in.

...

email

...

...

...

...

server to connect to

...

...

...

proxy server hostname

...

...

Connected to server

openProject

Open a specific project.

...

project

...

...

listProjects

List all your projects on the server.

...

...

...

...

createProject

Create a new project.

...

...

...

...

editProject

Edit project properties.

...

...

...

...

...

...

deleteProject

Delete an existing project.

...

...

Opened project

importProject

Allows you to import either:

• project from a server (e.g. a data dimension)
• create a complete project clone (See the tutorial)
• a local dump of a project (See the tutorial)

You can also import a part of a project with one of the parameters (dashboards, datasets, indicators, indicatorDrills, markers, markerSelectors, metrics, views). If you specify none of these parameters, the whole project will be imported. Everytime you specify a datasets parameter, corresponding data will be imported.

Before each import, validate command is called in the background. If there are any model validation violations in the source project, the import is stopped, unless you also provide the --force parameter.

During the import, the origin key is set to all metadata objects. This key indicates the original location of the object (server and the project). This has a special use in case of datasets & data import. import first takes a look at what datasets currently are in the project and compares them with datasets that are about to be imported. Datasets that are not present in the destination project will be imported automatically. In case of datasets that are present in the destination project, 3 cases might occur:

• if they have the same name and origin, the dataset will not be imported
• if they have the same name but different origin, a warning is shown and the dataset will not be imported
• if a prefix is specified, all source datasets will be imported

...

...

specifies if the output should be more verbose

default = https://secure.clevermaps.io

...

...

id of the project to import

either project or dump must be specified

...

...

id of the dump to import

either project or dump must be specified

...

...

...

load and dump request execution type

default = async

...

...

...

...

...

indicatorDrills

...

...

...

...

...

...

...

...

...

ignore source project validate errors and proceed with import anyway

skip failed dataset dumps (for projects with incomplete data)

...

...

importDatabase

Allows you to create datasets and import data from an external database.

This command reads the database metadata from which datasets are created, then imports the data and saves them as CSV files. You can choose to import either of which with --skipMetadata and --skipData parameters. Please note that this command does not create any other metadata objects than datasets. It's also possible to import only specific tables using the --tables parameter.

The database must be located on a running database server which is accessible under an URL. This can be on localhost, or anywhere on the internet. Valid credentials to the database are of course necessary.

So far, the command supports these database engines:

• PostgreSQL (https://www.postgresql.org/)

...

...

name of the database engine

...

...

database server hostname

for local databases, use localhost

...

...

...

name of the database schema

leave out if your engine does not support schemas, or the schema is public

...

...

...

...

...

list of tables to import

leave out if you want to import all tables from the database

example = "orders,clients,stores"

...

...

...

Usage examples:

importDatabase --engine postgresql --host localhost --port 5432 --database my_db --user postgres --password test
importDatabase --engine postgresql --host 172.16.254.1 --port 6543 --schema my_schema --database my_db --user postgres --password test --tables orders,clients,stores

loadCsv

Load data from a CSV file into a specified dataset.

This comand offers the execution parameter, which may be synchronous (sync) or asynchronous (async). Synchronous execution mode loads the CSV content through a single, constantly open connection. In most cases, synchronous execution is sufficient. To upload larger CSV files, it is better to use asynchronous execution. In case of asynchronous execution, after the upload of the content to an external storage (AWS S3), the load to the project itself is processed asynchronously as a job (AWS Step Functions). Shell then polls on this job and waits for a result.

loadCsv also offers various CSV input settings. Your CSV file may contain specific features, like custom quote or separator characters. The parameters with the csv prefix allow you to configure the data load to fit these features, instead of transforming the CSV file to one specific format. Special cases include the csvNull and csvForceNull parameters.

• csvNull allows you to specify a value, which will be interpreted as a null value
  • e.g. "false" or "_"
• csvForceNull then specifies on which columns the custom null replacement should be enforced
  • e.g. "name,title,description"

...

...

...

incremental mode appends the data to the end of the table

full mode truncates the table and loads the table anew

...

...

load request execution type

default = async

...

[sync, async]

...

csvHeader

...

...

specifies if the CSV file to upload has a header

default = true

...

[true, false]

...

...

specifies the CSV column separator character

default = ,

...

...

specifies the CSV quote character

default = "

...

...

specifies the CSV escape character

default = \

...

...

...

...

enables more verbose output

default = false

...

...

enables multipart file upload (recommended for files larger than 2 GB)

default = true

...

...

enables gzip compression

default = true

...

dumpCsv

Dump data from a specified dataset into a CSV file.

This command also offers the synchronous/asynchronous execution type, analogously to the loadCsv command, as described above.

...

...

...

...

dump request execution type

default = async

...

dumpProject

Dump project data & metadata to a directory. If the dump if successfull, the current dump is opened.

...

...

stored in config file

...

...

...

...

dump request execution type

default = async

...

...

listDumps

List all local project dumps.

...

...

stored in config file

openDump

Open a specific dump.

...

...

stored in config file

...

...

open a specific dump

if not specified, latest dump is opened

truncateProject

Deletes all metadata and data from the project.

This command has no parameters.

validate

Validate the project's data model and data.

...

...

...

...

skip validations of the data itself

...

...

validation request execution type

default = async

...

Opened dump

addMetadata

Add a metadata object to the project. The file must be located in a currently opened dump, and in the correct directory.

If the --objectName parameter is not specified, addMetadata will add all new objects in the current dump.

...

...

createMetadata

Create a new metadata object.

 At this moment, only dataset type is supported. Datasets are generated from a provided CSV file.

...

...

[dataset]

...

...

...

name of the object copy (with or without .json extension)

required only for dataset type

...

...

path to the CSV file (located either in dump, or anywhere in the file system)

required only for dataset type

...

...

name of the CSV column that will be marked as primary key

if not specified, first CSV column is selected

...

...

name of the geometry key

required only for geometryPolygon subtype

...

...

specifies custom CSV column separator character

default = ,

...

...

specifies custom CSV column quote character

default = "

...

...

specifies custom CSV column escape character

default = \

Usage examples:

createMetadata --type dataset --subtype basic --objectName "baskets" --file "baskets.csv" --primaryKey "basket_id"
createMetadata --type dataset --subtype geometryPoint --objectName "shops" --file "shops.csv" --primaryKey "shop_id"
createMetadata --type dataset --subtype geometryPolygon --objectName "district" --geometry "districtgeojson" --file "district.csv" --primaryKey "district_code"

removeMetadata

Remove a metadata object from the project and from the dump. The file must be located in a currently opened dump, and must not be new.

...

...

renameMetadata

Rename an object in a local dump and on the server. If the object is referenced in some other objects by URI (/rest/projects/$projectId/md/{objectType}?name=), the references will be renamed as well.

...

...

...

copyMetadata

Create a copy of an object existing in a currently opened dump.

This command unwraps the object from a wrapper, renames it and removes generated common syntax keys. If the objectName and newName arguments are the same, the object is unwrapped only.

...

...

...

pushProject

Upload all modified files (data & metadata) to the project.

This command basically wraps the functionality of loadCsv. It collects all modified metadata to upload, and performs asynchronous full load of CSV files.

...

...

...

...

...

load request execution type

default = async

...

...

...

enables more verbose output

default = false

...

...

enables multipart file upload (recommended for files larger than 2 GB)

default = true

...

...

enables gzip compression

default = true

...

[true, false]

status

Check the status of a currently opened dump against the project on the server.

This command detects files which have been locally or remotely changed, and also detects files which have a syntax error or constraint violations.

...

...

applyDiff

Create and apply metadata diff between two live projects.

This command compares all metadata objects of the currently opened project and project specified in the --sourceProject command, and applies changes to the currently opened dump. Metadata objects in dump can be either:

• added (completely new objects that aren't present in currently opened project)
• modified
• deleted (deleted objects not present in the sourceProject)

When the command finishes, you can review the changes applied to the dump using either status or diff comands. The command then tells you to perform specific subsequent steps. This can be one of (or all) these commands:

• addMetadata (to add the new objects)
• pushProject (to push the changes in modified objects)
• removeMetadata (to remove the deleted objects - a command list which must be copypasted to Shell is generated )

...

...

list of object types to be compared

example = "views,indicators,metrics"

diff

Compare local metadata objects with those in project line by line.

If the --objectName parameter is not specified, all wrapped modifed objects are compared.

...

...

The command outputs sets of changes - "deltas" done in each object. Each object can have multiple deltas, which are followed by header with syntax:

/{objectType}/{objectName}.json
[ A1 A2 | B1 B2 ]
...

Where:
A1 = start delta line number in the dump file
A2 = end delta line number in the dump file
B1 = start delta line number in the remote file
B2 = end delta line number in the remote file

Specific output example:

Image Removed

...