In the beginning, programmers created the databases. Now the databases were formless and empty, and they had no data.
And so Salesforce.org said “Let there be data,” and there was Snowfakery. And it was good.
Drop support for Python versions 3.8, 3.9, and 3.10 (#1091)
Update dependencies.
Improved temporary directory handling on Windows (#1029)
Update dependencies.
Added educational and non-profit fakers (#897)
Better error messages for missing database drivers. (#873)
Improve error message for random_record query into Salesforce org. (#844)
Simpler syntax for choice feature. (#441)
Snowfakery's Debug and CSV outputs now use the same format that Salesforce prefers. (#778)
Update dependencies.
Snowfakery can now generate Event or Meeting Schedules similar to Calendar Apps (#739)
Snowfakery now has a Salesforce.ContentFile feature for generating Content Versions (#446)
Snowfakery has a datetime_between function (#779)
Date-time values can now be coerced from strings and dates (#779)
Fixes to documentation (thank you @BrettMN) (#727)
find_record is now cached so that it only calls into Salesforce once. (#726)
Snowfakery can now do random_reference to nicknames. (#639)
Removed some limitations on what Snowfakery can do with objects referenced through random_reference, especially using a.b.c.d syntax. (#681 and #639)
Add a unique feature for random_reference (#684)
When embedded in CumulusCI, Snowfakery can now do SObject Upserts. (#644)
Add a Decimal datatype (#631)
Output Boolean values using the JSON boolean type in JSON outputs. (#663)
When embedded in CumulusCI, Snowfakery can now do SObject Updates. (#586)
Snowfakery can be used to do transformations of CSV files. (#586)
Snowfakery has a now function which returns a UTC datetime representing the present moment. (#485)
Snowfakery has a mode (snowfakery_version: 3) that lets formulas output types other than just strings. For example, they can generate date objects, which are amenable to date manipulations. This is an opt-in feature for the time being, but will become the default in June 2022. (#617)
Snowfakery has a debug function for inspecting expressions and values. (#613 and #615)
Snowfakery outputs "datetimes" in the UTC time zone, for better Salesforce compatibility. (#601, #694)
Snowfakery now ensures that faked firstnames and lastnames are ASCII alphanumeric before attemptiing to incorporate them into a fake email (#597)
Snowfakery publishes a JSON Schema for IDE usage (#556)
Snowfakery gives a better error message if a CSV datasource has rows that have more columns than the header row. (#594)
The Salesforce load order now follows the order of object template declarations in the recipe. The other considerations are described in the document (#563 and #566)
Improved syntax error messaging (#565)
Remove support for Python 3.6 and Python 3.7. (#554)
Allow zero-prefixed strings of digits, including zero-prefixed postal codes. (#551)
Limit fake Usernames to 80 characters (#552)
Removed an unnecessary dependency which could cause installation problems on some systems. (#534)
The unique_id and unique_alpha_code variables return a unique number that can be used to distinguish any record from others. For example, we can incorporate a unique ID into an email address or an employee ID. (#525)
Numeric and date counters are now available with the snowfakery.standard_plugins.Counters plugin. (#525)
Snowfakery will now accept UTF-8 CSV files generated from Excel with a Byte Order Mark (#516)
Snowfakery gives more detailed information in the snowfakery --version message (#469)
Booleans and dates can be the keys for random_choice lists. (#500)
Fix an issue with datasets not being closed in an orderly fashion (PR #492)
Emails and usernames will now be generated to match firstnames and lastnames (PR #420)
Documented Salesforce.ProfileId function which looks up a Profile in Salesforce by name and substitutes the ID automatically. (#438)
Fix problems installing through pipx.
Recipes can now merge in data from Salesforce orgs as records, IDs or datasets. (#395 and #401)
Output streams can now be specified as 3rd party libraries. Any Python class name can be passed to --output-format as long as it adheres to the OutputStream protocol as shown in examples/YamlOutputStream.py (#351)
Snowfakery is thread-safe/multi-processor-ready and available for parallel usage in CumulusCI. (CumulusCI Repo)
Snowfakery now supports a --reps option as an easier way to repeat a recipe (#416)
Snowfakery now accepts LeadingCaps syntax for all fakers, as well as underscore_separated and everythinglowercase. (#403)
Salesforce.ContentVersion and Salesforce.ProfileId convenience functions were added (#421)
Snowfakery now has voluminous documentation about Fakes in many languages and locales (#409)
Fix a regression: In some contexts it was impossible to call Faker with either positional arguments or keyword arguments. It is now possible to call with keyword arguments again. Positional arguments are still disallowed and older recipes may need to be updated to use keyword arguments. (#388)
e.g. ${{fake.sentence(nb_words=4)}} instead of ${{fake.sentence(4)}}
Snowfakery now informs you that it is not possible to use just_once in
a nested context. Previously, the results of doing so were unpredictable.
(#333)
Add a simple benchmarking tool available with the command snowbench. #346
Update the namespace used for inline fakes to support the same new names as block-based fakes. (i.e. ${{fake.FirstName}} is the same as fake: FirstName) #357
Fix a bug which caused problems outputting images. #351
Snowfakery now has some of its own built-in "Fake data types". For example "FirstName" (instead of first_name) and "Username" (which is like a Salesforce Username). These new names should be more intuitive for Salesforce users. PR #343
Fake "emails" now use domains based on "example.com" and similar domains specified in RFC 2606. This ensure that they are never real email addresses. Broader domain email addressses are still available as described in th documentation. PR #343
Fake names are now case agnostic. FIRSTname, firstNAME, firstname, FIRSTNAME, etc. are all identical. PR #343
Snowfakery's APIs have expanded. The changes should be backwards-compatible. #315
Add support for embedding external files (text or base64 binary). #322
Add support specifically for Salesforce ContentVersion. #322
Fixed packaging issue.
Fix regression when an SObject declares a RecordType on some records but not others. #301
Add a new feature for integrating with CumulusCI's Load architecture. This feature will obsolete most (hopefully all) uses of mapping.yml with Snowfakery. Documentation for it will be in CumulusCI. #290
Adds support for Salesforce Person Accounts to Snowfakery. See the documentation for details. (#283)
Fix regression: Can set RecordType on objects with names that are SQL keywords (like Case). (#277)
Record types can now be specified with random_choice. #273
Plugins are now searched for in relative directories called "plugins" as well as a directory under the user's home directory. See the docs for more information. #229
Snowfakery can now generate SQL text as an output format. The specific schema is compatible with CCI's data loader, but can be used in other contexts. #162
Add a snowfakery_filename variable to allow injecting the
recipe filename into recipe output. #249
Add NULL to Snowfakery formulas #239
Allow references to be indirect and include dots in them. Consequently, dots and other punctuation are now disallowed in nicknames and discouraged in object/table names. #186
Snowfakery's default branch is now main, so references to master should be updated.
Adds a "random_reference" feature to allow randomized connections between objects as described in the documentation. (PR #184)
Adds a notion of "variables" as described in the documentation. These allow computed values to be easily shared between object templates. (PR #231)
Adds a "snowfakery_locale" variable which determines what locale drives fake data creation (person nammes, state/province name etc.) (PR #231)
Does not depend upon pygraphviz library which can be difficult to install. One will still generally want to install graphviz, but not both graphviz AND pygraphviz. (PR #171)
Calls to "fake" are optimized. Recipes that are heavily reliant on faked fields may be 10-15% faster. One tradeoff is that e.g. person names or state names do not follow a real-world statistical distribution. Rare names (in en_US locale) like "Mathias" are now as common as real-world common names like "Matthew". (PR #214)
Adds Salesforce Example Files.
Adds an official API for embedding in Python.
Updates dependencies including Faker, which has Salesforce-contributed optimizations.
Adds min, max, round to the math plugin.
random_number can now accept a "step" argument as described in the docs.
Allow a short-form random_choice syntax that matches the docs.
date now casts datetimes to dates instead of returning them unchanged.
The --target-number value now allows its two sub-arguments to be specified in either order: Tablename then number or number then tablename.
The command line interface gives better error messages about unknown file extensions.
The continuation file format changed. Do not upgrade Snowfakery while running a long-running Snowfakery process. The new configuration file should be less prone to issues relating to failures to serialize results.
Improvements to plugin API: add evaluate_raw and simplify
Dependency updates.
Minor improvements in error reporting.
Update to Faker 4.1.3
Fix a deployment problem relating to version.txt
Support for External Datasets. (PR #102)
Macros within Macros now work correctly. (PR #120)
It is now possible to have forward references to Nicknames. (PR #113)
Closed Issue #42: relating to relative month parsing
Closed Issue #131: Parse errors would result in an invalid rather than empty JSON file.
Change the architecture to always initialize the database based on the recipe and never do it from the CumulusCI Mapping file.
Update URLs for PyPI. No changes to the Snowfakery code itself.
Add child_index feature for calculating the index number of each child object (see documentation for more information).
Internal changes to support upcoming features in CumulusCI.
Documentation for the Math plugin.
Dates can now be parsed from strings with the date() function. (see docs)
Added relativedelta() function. (see docs)
Official formula syntax is now ${{foo}} instead of <<foo>>.
The old syntax will be supported for at least the remainder of 2020.
Docs are checked into Git in Markdown format.
Official terminology changed so that Snowfakery files are now called "recipes".
Stabilize the output order for CCI Mapping files generated by Snowfakery.
New syntax for Internalized fake data. (see docs)
Internally use the plugin architecture to support even the builtin functions.
Snowfakery includes support for Salesforce RecordTypes.
Snowfakery can output SQL/JSON NULL using YAML blank fields or the YAML literal 'null'
Fields starting with __ are now properly suppressed as per the documentation.
Various performance and reliability improvements:
- parsed dates are now cached
- Jinja values are always coerced to a string where appropriate
- internal attributes were renamed for clarity
- lookups are only generated in CCI mappings if they are actually needed
Snowfakery now specifies its install requirements with minimum versions rather than specific versions, for improved compatibility with CumulusCI.
Windows compatibility improved, especially in test cases.
Changed the syntax for CSV handling.
Added --output-folder command.
Bugfixes (not user visible) and code cleanup.
Snowfakery 0.6 is open source!
The first version of Snowfakery was Salesforce.org internal-only.