Moldova is a lightweight template interpreter, used to generate random values that plug into the template, as defined by a series of custom tokens.
It understands the tokens as defined further down in the document.
Moldova comes both as a library you can import, as well as a binary executable that you can run, which will output results to STDOUT.
The intended use is via the binary executable.
Moldova was originally designed as input to a database load testing tool, called the Slammer. It's primary purpose was to generate INSERT and SELECT statements, which were then loaded into the Slammer. If you're looking for a good tool to test the throughput and latency of database operations at scale, checkout the Slammer.
And if you're interested in using them together, checkout the Moldovan Slammer, a quick helper repo to demonstrate using them together
To use the command, first install
go install github.com/StabbyCutyou/moldova/cmd/moldovaThe command accepts 2 arguments:
- n - How many templates to render to STDOUT. The default is 1, and it cannot be less than 1.
- t - The template to render
moldova -t "INSERT INTO floof VALUES ('{guid}','{guid:ordinal:0}','{country}',{int:min:-2000|max:0},{int:min:100|max:1000},{float:min:-1000.0|max:-540.0},{int:min:1|max:40},'{now}','{now:ordinal:0}','{country:case:up}',NULL,-3)" -n 100This would provide sample output like the following:
...
INSERT INTO floof VALUES ('791add99-43df-44c8-8251-6f7af7a014df','791add99-43df-44c8-8251-6f7af7a014df','MU',-1540,392,-624.529332,39,'2016-01-24 23:42:49','2016-01-24 23:42:49','UN',NULL,-3)
INSERT INTO floof VALUES ('0ab4cc33-6689-404f-a801-4fd431ca3f30','0ab4cc33-6689-404f-a801-4fd431ca3f30','PL',-1707,112,-550.333145,1,'2016-01-24 23:42:49','2016-01-24 23:42:49','SS',NULL,-3)
INSERT INTO floof VALUES ('a3f4151a-a304-4190-a3df-7fd97ce58588','a3f4151a-a304-4190-a3df-7fd97ce58588','CM',-1755,569,-961.122173,25,'2016-01-24 23:42:49','2016-01-24 23:42:49','NE',NULL,-3)Tokens are represented by special values placed inside of { } characters.
Each token follows this pattern:
{command:argument_name:value|argument_name:value|...|...}
Each token defines it's own set of arguments, which are outlined below. If you're provided multiple arguments, the key-value pairs of argument names to values are delimited by a | pipe character. These can be provided in any order, so long as they are valid arguments.
It is only necessary to provide a | when you have further arguments to provide. For example:
- {int}
- {int:min:10}
- {int:min:10|max:50}
- {int:min:10|max:50|ordinal:0}
- ordinal : integer >= 0
Moldova will replace any instance of {guid} with a GUID/UUID
If you provide the ordinal: option, for the current line of text being generated, you can have Moldova insert an existing value, rather than a new one. For example:
"{guid} - {guid:ordinal:0}"
In this example, both guids will be replaced with the same value. This is a way to back-reference existing generated values, for when you need something repeated.
- format : string, either "simple", "simpletz", or a golang date format string
- ordinal : integer >= 0
- zone: A timezone to represent the time in. You can use any value accepted here
Moldova will replace any instance of {now} with a string representation of Golangs time.Now() function, formatted per the provided date format. There are 2 built in formats, for convenience, that are compatible with many databases.
- simple - "2006-01-02 15:04:05"
- simpletz - "2006-01-02 15:04:05 -0700"
Additionally, you can provide your own format string.
{now} also supports the ordinal: option
- min : integer < max, unix epoch value
- max : integer > min, unix epoch value
- format : string, either "simple", "simpletz", or a golang date format string
- ordinal : integer >= 0
- zone: A timezone to represent the time in. You can use any value accepted here
Moldova will replace any instance of {time} with a string representation of a random time, between min and max in terms of Unix Epoch values. The defaults are between 0, and roughly Now (determined at runtime) There are 2 built in formats, for convenience, that are compatible with many databases.
- simple - "2006-01-02 15:04:05"
- simpletz - "2006-01-02 15:04:05 -0700"
Additionally, you can provide your own format string.
{time} also supports the ordinal: option
- min : integer < max
- max : integer > min
- ordinal : integer >= 0
Moldova will replace any instance of {int} with a random int value, optionally between the range provided. The defaults, if not provided, are 0 to 100.
{int} also supports ordinal: option
- min : integer < max
- max : integer > min
- ordinal : integer >= 0
Moldova will replace any instance of {float} with a random Float64, optionally between the range provided. The defaults, if not provided, are 0.0 to 100.0
{float} also supports ordinal: option
- length : integer >= 1
- case : "up" or "down"
- ordinal : integer >= 0
Moldova will replace any instance of {unicode} with a randomly generated set of unicode characters, of a length specified by :number. The default value is 2.
{unicode} also takes the :case argument, which is either 'up' or 'down', like so
{unicode:case:up} {unicode:case:down}
{unicode} also supports ordinal: option
Only a certain subset of unicode character ranges are supported by default, as defined in the moldova/data/unicode.go file.
- length : integer >= 1
- case : "up" or "down"
- ordinal : integer >= 0
Moldova will replace any instance of {ascii} with a randomly generated set of ASCII characters, of a length specified by :number. The default value is 2.
{ascii} also takes the :case argument, which is either 'up' or 'down', like so
{ascii:case:up} {ascii:case:down}
{ascii} also supports ordinal: option
- language : any string value
- case : "up" or "down"
- ordinal : integer >= 0
Moldova will replace any instance of {firstname} with a randomly generated first name, from the list defined in data/names.go
{firstname} takes a :langage argument, which can be any of the supported language strings in data/names.go. If the chosen name does not have an entry for the selected language, the configured default will be used in it's place. For exmaple:
{firstname:language:romanian}
{firstname} also takes the :case argument, which is either 'up' or 'down', like so
{firstname:case:up} {firstname:case:down}
{firstname} also supports :ordinal option
Only a certain subset of unicode character ranges are supported by default, as defined in the moldova/data/unicode.go file.
- language : any string value
- case : "up" or "down"
- ordinal : integer >= 0
Moldova will replace any instance of {firstname} with a randomly generated first name, from the list defined in data/names.go
{lastname} takes a :langage argument, which can be any of the supported language strings in data/names.go. If the chosen name does not have an entry for the selected language, the configured default will be used in it's place. For example:
{lastname:language:romanian}
{lastname} also takes the :case argument, which is either 'up' or 'down', like so
{lastname:case:up} {lastname:case:down}
{lastname} also supports :ordinal option
Only a certain subset of unicode character ranges are supported by default, as defined in the moldova/data/unicode.go file.
- case : "up" or "down"
- ordinal : integer >= 0
Moldova will replace any instance of {country} with an ISO 3166-1 alpha-2 country code.
{country} supports the same case: argument as {unicode}. The default value is "up"
{country} also supports the ordinal: argument.
I'll continue to add support for more random value categories. There are also hooks to support ascii-only string generation, but as of yet it is not implemented.
Add proper support for escaping control characters, so they are not interpreted as part of a token. These characters are {, }, :, and |.
Apache v2 - See LICENSE