Skip to content

Latest commit

 

History

History
363 lines (315 loc) · 16.1 KB

sfdx_notes.md

File metadata and controls

363 lines (315 loc) · 16.1 KB
Raw

sfdx

– Notes on https://www.udemy.com/course/salesforcedx
For my sfdx related aliases and functions see the sfdx.conf file in my linux repo.

installation process (x64 linux)

# create sfdx directory in your home directory
cd sfdx

# get installer
wget https://developer.salesforce.com/media/salesforce-cli/sfdx-cli/channels/stable/sfdx-cli-linux-x64.tar.xz

# unpack file into sfdx directory
tar xJf sfdx-cli-linux-x64.tar.xz -C sfdx --strip-components 1

# run installer
./sfdx/install

The installer will istall the Salesforce CLI in /usr/local/bin/sfdx.
Update to newest version with sfdx update.
script: sfdx/sfdx_scripts/sfdx_linux_installer automates this process on linux systems

oclif

– the open cli framework

api versin / runtime config

CLI Runtime Configuration Values
sfdx force:config:set apiVersion=42.0 --global will set the used api version to 42.
sfdx force:config:set apiVersion= --global will set it back to current version.
sfdx force:config:list will show a list of overridden configs. Default config doesn't show.

environment variables

Environment Variables

creating a new project

sfdx force:project:create -n Demo1 will create new project Demo1 as a sub-directory of the working directory.
Open project folder in VS Code by entering code Demo1/ in the terminal …
… or by pressing CTRL + o in VS Code and selecting Demo1 folder manually.
The sfdx-project.json was created with the global set api version. The login url can be edited here.

org authentication

sfdx force:auth:web:login -a Demo1 will open production org web-login to authenticate the cli.
To authenticate a Sandbox or Custom Domain use sfdx force:auth:web:login -a Demo1 --instanceurl https://test.salesforce.com
If the authentication via web login doesn't work, try sfdx force:auth:device:login -a Demo1

org logout

sfdx force:auth:logout -u Demo1 will disconnect the prior authenticated Demo1 org from sfdx.

open org

To open an authenticated org type sfdx force:org:open -u Demo1 or sfdx force:org:open -u 0221@myforce.net
Demo1 is the org alias, 0221@myforce.net is the user name – both work fine.
So it makes absolutely sense to use -a Demo1 option with :auth to give your orgs short and descriptive aliases.

org overview

List all your authenticated orgs with sfdx force:auth:list 

=== authenticated orgs
ALIAS  USERNAME          ORG ID              INSTANCE URL                                 OAUTH METHOD
─────  ────────────────  ──────────────────  ───────────────────────────────────────────  ────────────
Demo1  0221@myforce.net  00D09000007FkECEA0  https://myforcenet-dev-ed.my.salesforce.com  web

retreive change set from org

sfdx force:mdapi:retrieve -r ./mdapi -u 0221@myforce.net -p SFDX_Test will receive SFDX_Test change set from org associated with user 0221@myforce.net.
./mdapi is the target directory for the extract. I've created it within the project directory.
A file unpackage.zip will be created. We can use unzip mdapi/unpackaged.zip to extract all files. 

Archive:  mdapi/unpackaged.zip
  inflating: SFDX_Test/objects/Test_Object__c.object  
  inflating: SFDX_Test/layouts/Test_Object__c-Test Object Layout.layout  
  inflating: SFDX_Test/package.xml

The directory SFDX_Test will contain all these extract elements.

package.xml to sfdx project

sfdx force:mdapi:retrieve -r ./mdapi -u Demo1 -k mdapi/package.xml if a package.xml exists.
-r ./mdapi = target location
-u Demo1 = org alias or username
-k mdapi/package.xml = package.xml location
This command will again result in an unpackaged.zip file in the targer directory.

Another way to retrieve metadata via package.xml is sfdx force:source:retrieve -u Demo1 -x ./package.xml.
It is described in the Trailhead Package.xml Metadata Management module.
This method doesn't require to unzip the retrieved metadata, it will be placed in the force-app/main/default directory of the project.

convert metadata to source format

metadata structure – everything is in the same place.
source format – own xml for each field, object, etc.
sfdx force:mdapi:convert -r mdapi -d force-app to convert with metadata api.
I didn't manage to get this method to work with my sfdx version on linux

metadata to source without conversion step

sfdx force:source:retrieve --manifest mdapi/package.xml -u Demo1 uses an existing package.xml (manifest) to retrieve the specified content directly in source format.

Retrieving data without package.xml

sfdx force:source:retrieve -m ApexClass -u Demo1 would retrieve all Apex classes from the Demo1 org.
The Metadata Coverage Report shows what data can be queried via the metadata api.

create and deploy apex class via cli

sfdx force:apex:class:create --classname MyApexClass --template DefaultApexClass --outputdir force-app/main/default/classes/
This command creates apex class MyApexClass in the specified directory.
The apex class can be edited with your editor of chice and deploeyed as following: 

$ sfdx force:source:deploy --sourcepath force-app/main/default/classes/MyApexClass.cls -u Demo1
WARNING: apiVersion configuration overridden at "47.0"
Job ID | 0Ah7R9800NSjLHFGDA3
SOURCE PROGRESS | ████████████████████████████████████████ | 1/1 Components
=== Deployed Source
FULL NAME    TYPE       PROJECT PATH
───────────  ─────────  ───────────────────────────────────────────────────────
MyApexClass  ApexClass  force-app/main/default/classes/MyApexClass.cls
MyApexClass  ApexClass  force-app/main/default/classes/MyApexClass.cls-meta.xml

retrieve elswhere updated class

update already retrieved elements

sfdx force:source:retrieve --sourcepath force-app/main/default/classes -u Demo1 sourcepath can be used to specify a directory from the source org and retrieve all updates from there.
NOTE: This will only retrieve classes already existend in the local project. New classes won't be synced down automatically.

retrieve all or specific elements

`sfdx force:source:retrieve --manifest mdapi/classes.xml -u Demo1 manifest to receive all or specific apex classes.
The following manifest will retrieve all clases due to the star wildcard: 

<?xml version="1.0" encoding="UTF-8"?>
<Package xmlns="http://soap.sforce.com/2006/04/metadata">
    <fullName>Apex_Classes</fullName>
    <description>Let's see if I can retrieve classes only.</description>
    <types>
        <members>*</members>
        <name>ApexClass</name>
    </types>
    <version>50.0</version>
</Package>

To retrive only specific classes those individual members have to be specified: 

    <types>
        <members>MyApexClass</members>
        <members>MyApexClass2</members>
        <name>ApexClass</name>
    </types>

The manifest dosn't have to be named package.xml. I've named the above file classes.xml.

the help command

As it it is impossible to remember all terminal commands, most terminal applications come with am manual. 

$ man sfdx
No manual entry for sfdx

sfdx apparently has no manual, but it makes very good use of the --help command. 

$ sfdx --help
Salesforce CLI

VERSION
  sfdx-cli/7.85.1-2fb9e41053 linux-x64 node-v12.18.3

USAGE
  $ sfdx [COMMAND]

TOPICS
  alias    manage username aliases
  auth     authorize an org for use with the Salesforce CLI
  config   configure the Salesforce CLI
  force    tools for the Salesforce developer
  plugins  add/remove/create CLI plug-ins
  schema   List metadata types in your Salesforce org using the CLI

COMMANDS
  autocomplete  display autocomplete installation instructions
  commands      list all the commands
  help          display help for sfdx
  plugins       list installed plugins
  update        update the sfdx CLI
  which         show which plugin a command is in

To see the help page of source:retrieve and all its available parameters the command would be:
sfdx force:source:retrieve --help there are also their short forms specified: 

  -p, --sourcepath=sourcepath
      comma-separated list of source file paths to retrieve

  -x, --manifest=manifest
      file path for manifest (package.xml) of components to retrieve

The prior used --sourcepath is short -p, the --manifes is -x.
The short form of --help is -h.

running test classes via sfdx

sfdx force:apex:test:run -h will provide all options for an apex test run.

working with data

Create a record sfdx force:data:record:create -h
sfdx force:data:record:create -s Account -u Demo1 -v "name='CFC – created from cli'"
will create an account with the name field filled. Enter multiple fields separated by space:
sfdx force:data:record:create -s Account -u Demo1 -v "name='CFC2' Industry='Energgy'"
The command creates an account named CFC2 and industry Energgy.

Update a record to correct that spelling mistake we can use the following command:
sfdx force:data:record:update -s Account -u Demo1 -v "Industry='Energy'" -i 0017R00002TF7zQQAT

Retrive information with get – in the case of the following command all fields and values from the CFC2 account:
sfdx force:data:record:get -s Account -u Demo1 -w "Name='CFC2'" to receive the outcome in json format use:
sfdx force:data:record:get -s Account -u isv -w "Name='CFC2'" --json

Delete a record with command: sfdx force:data:record:delete -s Account -u Demo1 -w "Name='CFC2'"
Like the others delete would accept -i, --id or other options sfdx force:data:record:delete -h for more.

soql in sfdx example: sfdx force:data:soql:query -u Demo1 -q "SELECT Id, Name, Email FROM User"

Export relational data: To receive relational data, like contacts of accounts use such an command:
sfdx force:data:tree:export -q "SELECT Id, Name, (SELECT FirstName, LastName FROM Contacts) FROM Account" -d /home/heiko/test -p -u Demo1.
-d specifies the directory where the results should be saved (in json format).
-p generates mulitple sobject tree files and a plan definition file for aggregated import.
The output of this command looks like this: 

Wrote 13 records to /home/heiko/test/Accounts.json
Wrote 20 records to /home/heiko/test/Contacts.json
Wrote 0 records to /home/heiko/test/Account-Contact-plan.json

Accounts.json (two example records): 

{
    "records": [
        {
            "attributes": {
                "type": "Account",
                "referenceId": "AccountRef1"
            },
            "Name": "Edge Communications"
        },
        {
            "attributes": {
                "type": "Account",
                "referenceId": "AccountRef2"
            },
            "Name": "Burlington Textiles Corp of America"
        }
    ]
}

Contacts.json (three example records from two accounts): 

{
    "records": [
        {
            "attributes": {
                "type": "Contact",
                "referenceId": "ContactRef1"
            },
            "FirstName": "Rose",
            "LastName": "Gonzalez",
            "AccountId": "@AccountRef1"
        },
        {
            "attributes": {
                "type": "Contact",
                "referenceId": "ContactRef2"
            },
            "FirstName": "Sean",
        
        {
            "attributes": {
                "type": "Contact",
                "referenceId": "ContactRef3"
            },
            "FirstName": "Jack",
            "LastName": "Rogers",
            "AccountId": "@AccountRef2"
        }
    ]
}

Account-Contact-plan.json (complete): 

[
    {
        "sobject": "Account",
        "saveRefs": true,
        "resolveRefs": false,
        "files": [
            "Accounts.json"
        ]
    },
    {
        "sobject": "Contact",
        "saveRefs": false,
        "resolveRefs": true,
        "files": [
            "Contacts.json"
        ]
    }
]

The entry "referenceId": "AccountRef1" on the account object corresponds to the "AccountId": "@AccountRef1 entry on the contact.
This referential mapping together with Account-Contact-plan.json is required to import those records with their relations into an other org.
The Salesforce 18-digit id can't be used in this case as those ids would conflict with the target org.

To import those records into an other org use the plan file: 

$ sfdx force:data:tree:import -p Account-Contact-plan.json -u Demo2
=== Import Results
Reference ID  Type     ID
────────────  ───────  ──────────────────
AccountRef1   Account  0017R00002TFgEgQAL
AccountRef2   Account  0017R00002TFgEhQAL
ContactRef1   Contact  0037R00002pcMTJQA2
ContactRef2   Contact  0037R00002pcMTKQA2
ContactRef3   Contact  0037R00002pcMTLQA2

Use simplified commands to export/import records individual without keeping their relations: 

$ sfdx force:data:tree:export -q "SELECT Id, Name FROM Account" -u Demo1
$ sfdx force:data:tree:export -q "SELECT Id, LastName, FirstName FROM Contact" -u Demo1
# Then run the import command:
$ sfdx force:data:tree:import -f Contact.json,Account.json -u Demo2

sObject tree import requests can contain

  • Up to a total of 200 records across all trees
  • Up to five records of different types
  • sObject trees up to five levels deep

Check out the Salesforce documentation for more information.

Scratch orgs

sfdx force:org:create -f config/project-scratch-def.json executed in the main sfdx project folder will create a scratch org as defined in project-scratch-def.json.
If yo forget to set the alias with -a you can set it anytime later with sfdx alias:set scratch=test-wkuerukfcxub@example.com.

Limits

You can display an ors limits (remaining, mx) with sfdx force:limits:api:display -u Demo1. 

$ sfdx force:limits:api:display -u Demo1
Name                                         Remaining  Max
───────────────────────────────────────────  ─────────  ─────────
ActiveScratchOrgs                            4          5
AnalyticsExternalDataSizeMB                  40960      40960
BOZosCalloutHourlyLimit                      20000      20000
ConcurrentAsyncGetReportInstances            200        200
ConcurrentEinsteinDataInsightsStoryCreation  5          5

Procedure to export and mass-delete debug logs

sfdx force:data:soql:query -u prod -t -q "SELECT Id FROM ApexLog WHERE LogUser.Name = 'B2BMA Integration' AND Status = 'Success'" -r "csv" > out.csv
sfdx force:data:bulk:delete -u prod -s ApexLog -f ./out.csv