db

package
v1.17.6 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Dec 26, 2024 License: MIT Imports: 19 Imported by: 0

Documentation

Overview

Package db to support openM++ model database operations.

Database may contain multiple openM++ models and consists of two parts: (a) metadata to describe models: data types, input parameters, output tables, etc. (b) actual modeling data itself: database tables for input parameters and output tables.

To run the model we need to have a set of model input parameters, called "input working set" or "workset". User can "open" workset in read-write mode and modify model input parameters. To use that set as model run input user must "close" it by marking workset as read-only. After model run user can again open workset as read-write and continue input editing. Each workset has a name (unique inside of the model) and set id (database unique positive int).

Result of model run stored in output tables and also include copy of all input parameters used to run the model. That pair of input and output data called "run" and identified by run id (database unique positive int).

Index

Constants

View Source 
const (
	SQLiteDbDriver  = "SQLite"  // default db driver name
	SQLiteTimeout   = 86400     // default SQLite busy timeout
	Sqlite3DbDriver = "sqlite3" // SQLite db driver name
	OdbcDbDriver    = "odbc"    // ODBC db driver name
)

Database connection values

View Source 
const (
	InitRunStatus     = "i" // i = initial status
	ProgressRunStatus = "p" // p = run in progress
	WaitRunStatus     = "w" // w = wait: run in progress, under external supervision
	DoneRunStatus     = "s" // s = completed successfully
	ExitRunStatus     = "x" // x = exit and not completed
	ErrorRunStatus    = "e" // e = error failure
	DeleteRunStatus   = "d" // d = deleted model run
)

Model run status (run_lst table) and modeling task run status (task_run_lst table): 

if task status = w (wait) then
   model wait and NOT completed until other process set status to one of finals: s,x,e
   model check if any new sets inserted into task_set and run it as they arrive
View Source 
const CALCULATED_ID_OFFSET = 12000 // calculated exprssion id offset, for example for Expr1 calculated expression id is 12001
View Source 
const IsOdbcSupported = false

IsOdbcSupported indicate support of ODBC connections built-in

View Source 
const MaxSchemaVersion = 105

MaxSchemaVersion is a maximum compatible db schema version

View Source 
const MinSchemaVersion = 105

MinSchemaVersion is a minimal compatible db schema version

View Source 
const TotalEnumCode = "all"

TotalEnumCode is predefined enum code for "total" enum

Variables

This section is empty.

Functions

func CheckOpenmppSchemaVersion 

func CheckOpenmppSchemaVersion(dbConn *sql.DB) error

CheckOpenmppSchemaVersion return error if it is not openM++ db or schema version incompatible

func CopyParameterFromRun 

func CopyParameterFromRun(dbConn *sql.DB, modelDef *ModelMeta, ws *WorksetRow, paramName string, isReplace bool, rs *RunRow) error

CopyParameterFromRun copy parameter metadata and parameter values into workset from model run.

If isReplace is true and parameter already exist in destination workset then error returned. If isReplace is false then delete existing metadata and new insert new from model run. Destination workset must be in read-write state. Source model run must be completed, run status one of: s=success, x=exit, e=error.

func CopyParameterFromWorkset 

func CopyParameterFromWorkset(dbConn *sql.DB, modelDef *ModelMeta, dstWs *WorksetRow, paramName string, isReplace bool, srcWs *WorksetRow) error

CopyParameterFromWorkset copy parameter metadata and parameter values from one workset to another.

If isReplace is true and parameter already exist in destination workset then error returned. If isReplace is false then delete existing metadata and new insert new from source workset. Destination workset must be in read-write state, source workset must be read-only.

func DeleteModel 

func DeleteModel(dbConn *sql.DB, modelId int) error

Delete existing model metadata and drop model data tables from database.

func DeleteProfile 

func DeleteProfile(dbConn *sql.DB, name string) error

DeleteProfile delete rows from profile in profile_lst and profile_option tables.

func DeleteProfileOption 

func DeleteProfileOption(dbConn *sql.DB, name, key string) error

DeleteProfileOption delete row from profile_option table.

func DeleteRun 

func DeleteRun(dbConn *sql.DB, runId int) error

Delete model run: delete metadata, parameters run values and output tables run values from database.

func DeleteTask 

func DeleteTask(dbConn *sql.DB, taskId int) error

Delete modeling task and task run history from database. Only task itself is deleted, model runs and worksets are not affected.

func DeleteWorkset 

func DeleteWorkset(dbConn *sql.DB, setId int) error

DeleteWorkset delete workset metadata and workset parameter values from database.

func DeleteWorksetAllParameters added in v1.6.3 

func DeleteWorksetAllParameters(dbConn *sql.DB, setId int) error

DeleteWorksetAllParameters delete all parameters metadata and values from workset.

Workset must be read-write in order to delete all parameters. If workset does not exist then nothing deleted and no errors returned, it is empty operation.

func DeleteWorksetParameter 

func DeleteWorksetParameter(dbConn *sql.DB, modelId int, setName, paramName string) (int, error)

DeleteWorksetParameter do delete parameter metadata and values from workset.

If parameter not exist in workset then nothing deleted. Workset must be read-write in order to delete parameter. It is return parameter Hid = 0 if nothing deleted.

func GetModelId 

func GetModelId(dbConn *sql.DB, name, digest string) (bool, int, error)

GetModelId return model id if exists.

Model selected by name and/or digest, i.e.: ("modelOne", "abcd20120817160459148") if digest is empty then first model with min(model_id) is used

func GetModelRunOptions 

func GetModelRunOptions(dbConn *sql.DB, modelId int) (map[int]map[string]string, error)

GetModelOptions return model run_option table rows as map of maps: map(run_id, map(key, value)).

func GetProfileList 

func GetProfileList(dbConn *sql.DB) ([]string, error)

GetProfileList return profile names: profile_lst table rows.

Profile is a named group of (key, value) options, similar to ini-file. Default model options has profile_name = model_name.

func GetRunListText 

func GetRunListText(dbConn *sql.DB, modelId int, langCode string) ([]RunRow, []RunTxtRow, error)

GetRunListText return list of model runs with description and notes: run_lst and run_txt rows.

If langCode not empty then only specified language selected else all languages

func GetRunOptions 

func GetRunOptions(dbConn *sql.DB, runId int) (map[string]string, error)

GetRunOptions return run_option table rows as (key, value) map.

func GetTaskListText 

func GetTaskListText(dbConn *sql.DB, modelId int, langCode string) ([]TaskRow, []TaskTxtRow, error)

GetTaskListText return list of modeling tasks with description and notes: task_lst and task_txt rows.

If langCode not empty then only specified language selected else all languages

func GetTaskSetIds 

func GetTaskSetIds(dbConn *sql.DB, taskId int) ([]int, error)

GetTaskSetIds return modeling task set id's by task id from task_set db table

func GetWorksetListText 

func GetWorksetListText(dbConn *sql.DB, modelId int, langCode string) ([]WorksetRow, []WorksetTxtRow, error)

GetWorksetListText return list of model worksets, description and notes: workset_lst and workset_txt rows.

If langCode not empty then only specified language selected else all languages

func GetWorksetParam added in v1.6.3 

func GetWorksetParam(dbConn *sql.DB, setId int, paramHid int) (int, int, error)

GetWorksetParam return sub-values count and default sub-value id for workset parameter: SELECT sub_count, default_sub_id FROM workset_parameter

func GetWorksetParamList 

func GetWorksetParamList(dbConn *sql.DB, setId int) ([]int, []int, []int, error)

GetWorksetParamList return Hid, sub-values count and default sub-value id for all parameters included in that workset: SELECT parameter_hid, sub_count, default_sub_id FROM workset_parameter

func GetWorksetRunIds 

func GetWorksetRunIds(dbConn *sql.DB, setId int) ([]int, error)

GetWorksetRunIds return ids of model run results (run_id) where input parameters are from specified working set.

Only successfully completed run ids returned, not failed, not "in progress". This method is "local" to database and if data transfered between databases it very likely return wrong results. This method is not recommended, use modeling task to establish relationship between input set and model run.

func IfEmptyMakeDefault 

func IfEmptyMakeDefault(modelName, sqlitePath, dbConnStr, dbDriver string) (string, string)

return SQLite connection string and driver name based on model name: 

Database=modelName.sqlite; Timeout=86400; OpenMode=ReadWrite;

func IfEmptyMakeDefaultReadOnly added in v1.9.0 

func IfEmptyMakeDefaultReadOnly(modelName, sqlitePath, dbConnStr, dbDriver string) (string, string)

return read-only SQLite connection string and driver name based on model name: 

Database=modelName.sqlite; Timeout=86400; OpenMode=ReadWrite;

func IsRunCompleted 

func IsRunCompleted(status string) bool

IsRunCompleted return true if run status one of: s=success, x=exit, e=error

func IsUnsafeQuote added in v1.9.0 

func IsUnsafeQuote(c rune) bool

return true if character is can be interpreted as sql ' quote MSSQL silently replace following utf-16 chars with 'single' quote: 

ʹ    697  Modifier Letter Prime
ʼ    700  Modifier Letter Apostrophe
ˈ    712  Modifier Letter Vertical Line
′  8242  Prime
＇ 65287  Fullwidth Apostrophe

func MakeSqliteDefault 

func MakeSqliteDefault(sqlitePath string) string

return default SQLite connection string based on model.sqlite file path: 

Database=model.sqlite; Timeout=86400; OpenMode=ReadWrite;

func MakeSqliteDefaultReadOnly added in v1.9.0 

func MakeSqliteDefaultReadOnly(sqlitePath string) string

return default read-only SQLite connection string based on model.sqlite file path: 

Database=model.sqlite; Timeout=86400; OpenMode=ReadOnly;

func NameOfRunStatus added in v1.10.0 

func NameOfRunStatus(status string) string

NameOfRunStatus return short name by run run status code: s=success, x=exit, e=error

func OpenmppSchemaVersion 

func OpenmppSchemaVersion(dbConn *sql.DB) (int, error)

OpenmppSchemaVersion return db schema: select id_value from id_lst where id_key = 'openmpp'

func RenameRun added in v1.6.3 

func RenameRun(dbConn *sql.DB, runId int, newRunName string) (bool, error)

RenameRun do rename model run if new name is not empty "" string.

func RenameTask added in v1.6.3 

func RenameTask(dbConn *sql.DB, taskId int, newTaskName string) (bool, error)

RenameTask do rename task if new name is not empty "" string

func RenameWorkset added in v1.6.3 

func RenameWorkset(dbConn *sql.DB, setId int, newSetName string, isAnyReadonly bool) (bool, error)

RenameWorkset do rename workset if new name is not empty "" string.

If isAnyReadonly parameter is true then do rename regarless of workset read-only status else return error if workset is not read-write.

func SelectFirst 

func SelectFirst(dbConn *sql.DB, query string, cvt func(row *sql.Row) error) error

SelectFirst select first db row and pass it to cvt() for row.Scan()

func SelectRows 

func SelectRows(dbConn *sql.DB, query string, cvt func(rows *sql.Rows) error) error

SelectRows select db rows and pass each to cvt() for rows.Scan()

func SelectRowsTo added in v1.9.7 

func SelectRowsTo(dbConn *sql.DB, query string, cvt func(rows *sql.Rows) (bool, error)) error

SelectRowsTo select db rows and pass each row to cvt(). cvt() return true to continue or false to stop rows processing.

func ToQuoted added in v1.9.0 

func ToQuoted(src string) string

make sql quoted string, ie: 'O”Brien'.

func TrxSelectFirst 

func TrxSelectFirst(dbTrx *sql.Tx, query string, cvt func(row *sql.Row) error) error

TrxSelectFirst select first db row in transaction scope and pass it to cvt() for row.Scan()

func TrxSelectRows 

func TrxSelectRows(dbTrx *sql.Tx, query string, cvt func(rows *sql.Rows) error) error

TrxSelectRows select db rows in transaction scope and pass each to cvt() for rows.Scan()

func TrxUpdate 

func TrxUpdate(dbTrx *sql.Tx, query string) error

TrxUpdate execute sql query in transaction scope

func TrxUpdateStatement 

func TrxUpdateStatement(dbTrx *sql.Tx, query string, put func() (bool, []interface{}, error)) error

TrxUpdateStatement execute sql statement in transaction scope until put() return true

func Update 

func Update(dbConn *sql.DB, query string) error

Update execute sql query outside of transaction scope (on different connection)

func UpdateLanguage 

func UpdateLanguage(dbConn *sql.DB, langDef *LangMeta) error

UpdateLanguage insert new or update existing language and words in lang_lst and lang_word tables. Language ids updated with actual id's from database

func UpdateModel 

func UpdateModel(dbConn *sql.DB, dbFacet Facet, modelDef *ModelMeta) (bool, error)

UpdateModel insert new model metadata in database, return true if model inserted or false if already exist.

If model with same digest already exist then call simply return existing model metadata (no changes made in database). Parameters and output tables Hid's and db table names updated with actual database values If new model inserted then modelDef updated with actual id's (model id, parameter Hid...) If parameter (output table) not exist then create db tables for parameter values (output table values) If db table names is "" empty then make db table names for parameter values (output table values)

func UpdateModelText 

func UpdateModelText(dbConn *sql.DB, modelDef *ModelMeta, langDef *LangMeta, modelTxt *ModelTxtMeta) error

delete existing and insert new model text (description and notes) in database.

Model id, type Hid, parameter Hid, table Hid, language id updated with actual database id's

func UpdateModelWord 

func UpdateModelWord(dbConn *sql.DB, modelDef *ModelMeta, langDef *LangMeta, mwDef *ModelWordMeta) error

UpdateModelWord insert new or update existing model language-specific strings in model_word table.

func UpdateProfile 

func UpdateProfile(dbConn *sql.DB, profile *ProfileMeta) error

UpdateProfile insert new or replace existing profile in profile_lst and profile_option tables.

It always replacing all existing rows by delete from profile_lst and profile_option where profile_name = profile.Name and insert new rows into profile_lst and profile_option tables.

func UpdateProfileOption 

func UpdateProfileOption(dbConn *sql.DB, name, key, val string) error

UpdateProfileOption insert new or replace existing profile option in profile_option table. It is also will insert profile name into profile_lst table if such name does not already exist.

func UpdateRunParameterText added in v1.9.3 

func UpdateRunParameterText(dbConn *sql.DB, modelDef *ModelMeta, runId int, paramTxtPub []ParamRunSetTxtPub, langDef *LangMeta) error

UpdateRunParameterText merge (insert or update) run parameter value notes into run_parameter_txt table.

New parameter notes merged with existing db rows by update or insert if rows not exist db tables. If run does not exist then function does nothing. Run status must be completed (success, exit or error) otherwise error returned. If input array of ParamRunSetTxtPub is empty then it is empty operation and return is success.

func UpdateRunStatus added in v1.10.0 

func UpdateRunStatus(dbConn *sql.DB, runId int, status string) error

UpdateRunStatus updates run status to the one of run completed: s=success, x=exit, e=error.

func UpdateRunValueDigest 

func UpdateRunValueDigest(dbConn *sql.DB, runId int) (string, error)

UpdateRunValueDigest does recalculate and update run_lst table with new run value digest and return it. If run not exist or status is not success or exit then digest is "" empty (not updated).

func UpdateWorksetParameterText added in v1.9.3 

func UpdateWorksetParameterText(dbConn *sql.DB, modelDef *ModelMeta, setName string, paramTxtPub []ParamRunSetTxtPub, langDef *LangMeta) error

UpdateWorksetParameterText merge parameter value notes into workset_parameter_txt table.

Set name is used to find workset and set id updated with actual database value. Workset must exist and must be read-write for replace or merge.

Parameter must exist exist in the model otherwise it is an error. If parameter not exist in workset then function does nothing (it is empty operation). If input array of ParamRunSetTxtPub is empty then it is empty operation and return is success.

func UpdateWorksetReadonly 

func UpdateWorksetReadonly(dbConn *sql.DB, setId int, isReadonly bool) error

UpdateWorksetReadonly update workset readonly status.

func UpdateWorksetReadonlyByName 

func UpdateWorksetReadonlyByName(dbConn *sql.DB, modelId int, name string, isReadonly bool) error

UpdateWorksetReadonlyByName update workset readonly status by workset name.

func WriteMicrodataFrom added in v1.13.0 

func WriteMicrodataFrom(
	dbConn *sql.DB, dbFacet Facet, modelDef *ModelMeta, runMeta *RunMeta, layout *WriteMicroLayout, from func() (interface{}, error),
) error

WriteMicrodataFrom insert microdata values into model run until from() return not nil CellMicro value.

Model run should not already contain microdata values: microdata can be inserted only once in model run and cannot be updated after. Entity generation digest is used to find existing entity generation, it cannot be empty "" string otherwise error returned. Entity generation metadata updated with actual database values of generation Hid

Double format is used for float model types digest calculation, if non-empty format supplied.

Return run entity metadata rows.

func WriteOutputTableFrom added in v1.9.7 

func WriteOutputTableFrom(
	dbConn *sql.DB, modelDef *ModelMeta, layout *WriteTableLayout, accFrom func() (interface{}, error), exprFrom func() (interface{}, error),
) error

WriteOutputTableFrom insert output table values (accumulators or expressions) into model run from accFrom() and exprFrom() readers.

Model run must exist and be in completed state (i.e. success or error state). Model run should not already contain output table values: it can be inserted only once in model run and cannot be updated after. Accumulators and expressions values must come in the order of primary key otherwise digest calculated incorrectly. Double format is used for float model types digest calculation, if non-empty format supplied.

func WriteParameterFrom added in v1.9.7 

func WriteParameterFrom(dbConn *sql.DB, modelDef *ModelMeta, layout *WriteParamLayout, from func() (interface{}, error)) error

WriteParameterFrom insert or update parameter values in workset or insert parameter values into model run until from() return not nil CellParam value.

If this is model run update (layout.IsToRun is true) then model run must exist and be in completed state (i.e. success or error state). Model run should not already contain parameter values: parameter can be inserted only once in model run and cannot be updated after.

If workset already contain parameter values then values updated else inserted. If only "page" of workset parameter rows supplied (layout.IsPage is true) then each row deleted by primary key before insert else all rows deleted by one delete by set id.

Double format is used for float model types digest calculation, if non-empty format supplied.

Types

type CalcMaps added in v1.16.0 

type CalcMaps struct {
	IdToDigest   map[int]string // map of run id's to run digests
	DigestToId   map[string]int // map of run digests to run id's
	CalcIdToName map[int]string // map of calculation id to name
	CalcNameToId map[string]int // map of calculation name to id
}

calculation maps between runs digest and id and calculations name and id

func EmptyCalcMaps added in v1.16.0 

func EmptyCalcMaps() CalcMaps

return empty value of calculation maps

type CalculateLayout added in v1.15.4 

type CalculateLayout struct {
	Calculate string // expression to calculate, ex.: Expr0[base] - Expr0[variant]
	CalcId    int    // calculated expression id, calc_id column in csv,     ex.: 0, 12001, 24001
	Name      string // calculated expression name, calc_name column in csv, ex.: Expr0, AVG_Expr0, RATIO_Expro0
}

CalculateLayout describes calculation expression for parameters, output table values or microdata entity. It can be comparison calculation for multiple model runs, ex.: Expr0[base] - Expr0[variant].

type CalculateMicroLayout added in v1.16.0 

type CalculateMicroLayout struct {
	Calculation []CalculateLayout // aggregation measures, ex.: OM_MIN(Salary), OM_AVG(Income[base] - Income[variant])
	GroupBy     []string          // attributes to group by
}

CalculateMicroLayout describes aggregations of microdata.

It can be comparison aggregations and/or calculation aggregations. Comparison aggregation must contain [base] and [variant] attribute(s), ex.: OM_AVG(Income[base] - Income[variant]). Calculation aggregation is attribute(s) aggregation expression, ex.: OM_MAX(Income) / OM_MIN(Salary).

type CalculateTableLayout added in v1.15.4 

type CalculateTableLayout struct {
	CalculateLayout      // expression to calculate and layout
	IsAggr          bool // if true then select output table accumulator else expression
}

CalculateLayout describes calculation of output table values. It can be comparison calculation for multiple model runs, ex.: Expr0[base] - Expr0[variant].

type CellAcc 

type CellAcc struct {
	AccId int // output table accumulator id
	SubId int // output table subvalue id
	// contains filtered or unexported fields
}

CellAcc is value of output table accumulator.

type CellAccConverter added in v1.9.2 

type CellAccConverter struct {
	CellTableConverter // model metadata and output table name
}

CellAccConverter is a converter for output table accumulator to implement CsvConverter interface.

func (*CellAccConverter) CsvFileName added in v1.9.2 

func (cellCvt *CellAccConverter) CsvFileName() (string, error)

CsvFileName return file name of csv file to store output table accumulator rows

func (*CellAccConverter) CsvHeader added in v1.9.2 

func (cellCvt *CellAccConverter) CsvHeader() ([]string, error)

CsvHeader return first line for csv file: column names. For example: acc_name,sub_id,dim0,dim1,acc_value or if IsIdCsv is true then: acc_id,sub_id,dim0,dim1,acc_value

func (*CellAccConverter) IdToCodeCell added in v1.9.2 

func (cellCvt *CellAccConverter) IdToCodeCell(modelDef *ModelMeta, name string) (func(interface{}) (interface{}, error), error)

IdToCodeCell return converter from output table cell of ids: (acc_id, sub_id, dimensions, value) to cell of codes: (acc_id, sub_id, dimensions as enum code, value)

If dimension type is enum based then dimensions enum ids can be converted to enum code. If dimension type is simple (bool or int) then dimension value converted to string.

func (*CellAccConverter) IsUseEnumId added in v1.13.0 

func (cellCvt *CellAccConverter) IsUseEnumId() bool

return true if csv converter is using enum id's for dimensions

func (*CellAccConverter) KeyIds added in v1.9.7 

func (cellCvt *CellAccConverter) KeyIds(name string) (func(interface{}, []int) error, error)

KeyIds return converter to copy primary key: (acc_id, sub_id, dimension ids) into key []int.

Converter will return error if len(key) not equal to row key size.

func (*CellAccConverter) ToCell added in v1.17.3 

func (cellCvt *CellAccConverter) ToCell() (func(row []string) (interface{}, error), error)

CsvToCell return closure to convert csv row []string to output table accumulator cell (dimensions and value).

It does return error if len(row) not equal to number of fields in cell db-record. If dimension type is enum based then csv row is enum code and it is converted into cell.DimIds (into dimension type type enum ids).

func (*CellAccConverter) ToCsvIdRow added in v1.13.0 

func (cellCvt *CellAccConverter) ToCsvIdRow() (func(interface{}, []string) (bool, error), error)

ToCsvIdRow return converter from output table cell (acc_id, sub_id, dimensions, value) to csv id's row []string.

Converter return isNotEmpty flag, it return false if IsNoZero or IsNoNull is set and cell value is empty or zero. Converter simply does Sprint() for each dimension item id, accumulator id, subvalue number and value. Converter will return error if len(row) not equal to number of fields in csv record.

func (*CellAccConverter) ToCsvRow added in v1.13.0 

func (cellCvt *CellAccConverter) ToCsvRow() (func(interface{}, []string) (bool, error), error)

ToCsvRow return converter from output table cell (acc_id, sub_id, dimensions, value) to csv row []string (acc_name, sub_id, dimensions, value).

Converter return isNotEmpty flag, it return false if IsNoZero or IsNoNull is set and cell value is empty or zero. Converter return error if len(row) not equal to number of fields in csv record. If dimension type is enum based then csv row is enum code.

type CellAccLocaleConverter added in v1.17.6 

type CellAccLocaleConverter struct {
	CellAccConverter
	Lang    string            // language code, expected to compatible with BCP 47 language tag
	LangDef *LangMeta         // language metadata to find translations
	DimsTxt []TableDimsTxtRow // output table dimension text rows: table_dims_txt join to model_table_dic
	EnumTxt []TypeEnumTxtRow  // type enum text rows: type_enum_txt join to model_type_dic
	AccTxt  []TableAccTxtRow  // output table accumulator text rows: table_acc_txt join to model_table_dic
}

Converter for output table accumulators to implement CsvLocaleConverter interface.

func (*CellAccLocaleConverter) CsvHeader added in v1.17.6 

func (cellCvt *CellAccLocaleConverter) CsvHeader() ([]string, error)

CsvHeader return first line for csv file: column names. For example: acc_name,sub_id,Age,Sex,acc_value

func (*CellAccLocaleConverter) ToCsvRow added in v1.17.6 

func (cellCvt *CellAccLocaleConverter) ToCsvRow() (func(interface{}, []string) (bool, error), error)

Return converter from output table cell (acc_id, sub_id, dimensions, value) to language-specific csv []string row of accumulator description, sub_id, dimension enum labels and value.

Converter return isNotEmpty flag, it return false if IsNoZero or IsNoNull is set and cell value is empty or zero. Converter return error if len(row) not equal to number of fields in csv record. If dimension type is enum based then csv row is enum label. Value and dimesions of built-in types converted to locale-specific strings, e.g.: 1234.56 => 1 234,56

type CellAllAcc 

type CellAllAcc struct {
	DimIds []int     // dimensions item as enum ids or int values if dimension type simple
	SubId  int       // output table subvalue id
	IsNull []bool    // if true then value is NULL
	Value  []float64 // accumulator value(s)
}

CellAllAcc is value of multiple output table accumulators.

type CellAllAccConverter added in v1.9.2 

type CellAllAccConverter struct {
	CellTableConverter        // model metadata and output table name
	ValueName          string // If ValueName is "" empty then all accumulators use for csv else one
}

CellAllAccConverter is a converter for multiple output table accumulators to implement CsvConverter interface.

func (*CellAllAccConverter) CsvFileName added in v1.9.2 

func (cellCvt *CellAllAccConverter) CsvFileName() (string, error)

CsvFileName return file name of csv file to store all accumulators rows

func (*CellAllAccConverter) CsvHeader added in v1.9.2 

func (cellCvt *CellAllAccConverter) CsvHeader() ([]string, error)

CsvHeader return first line for csv file: column names, for example: sub_id,dim0,dim1,acc0,acc1 If ValueName is "" empty then use all accumulators for csv else only one where accumulator name is ValueName

func (*CellAllAccConverter) IdToCodeCell added in v1.9.2 

func (cellCvt *CellAllAccConverter) IdToCodeCell(modelDef *ModelMeta, name string) (func(interface{}) (interface{}, error), error)

IdToCodeCell return converter from output table cell of ids: (sub_id, dimensions, acc0, acc1, acc2) to cell of codes: (sub_id, dimensions as enum code, acc0, acc1, acc2)

If dimension type is enum based then dimensions enum ids can be converted to enum code. If dimension type is simple (bool or int) then dimension value converted to string.

func (*CellAllAccConverter) IsUseEnumId added in v1.13.0 

func (cellCvt *CellAllAccConverter) IsUseEnumId() bool

return true if csv converter is using enum id's for dimensions

func (*CellAllAccConverter) KeyIds added in v1.9.7 

func (cellCvt *CellAllAccConverter) KeyIds(name string) (func(interface{}, []int) error, error)

KeyIds return converter to copy primary key: (sub id, dimension ids) into key []int.

Converter will return error if len(key) not equal to row key size.

func (*CellAllAccConverter) ToCell added in v1.17.3 

func (cellCvt *CellAllAccConverter) ToCell() (func(row []string) (interface{}, error), error)

CsvToCell return closure to convert csv row []string to output table accumulator cell (dimensions and value).

It does return error if len(row) not equal to number of fields in cell db-record. If dimension type is enum based then csv row is enum code and it is converted into cell.DimIds (into dimension type type enum ids).

func (*CellAllAccConverter) ToCsvIdRow added in v1.13.0 

func (cellCvt *CellAllAccConverter) ToCsvIdRow() (func(interface{}, []string) (bool, error), error)

ToCsvIdRow return converter from output table cell (sub_id, dimensions, acc0, acc1, acc2) to csv id's row []string.

Converter return isNotEmpty flag, it return false if IsNoZero or IsNoNull is set and all values are empty or zero. Converter simply does Sprint() for each dimension item id, subvalue number and value(s). Converter will return error if len(row) not equal to number of fields in csv record. If ValueName is "" empty then all accumulators converted else one

func (*CellAllAccConverter) ToCsvRow added in v1.13.0 

func (cellCvt *CellAllAccConverter) ToCsvRow() (func(interface{}, []string) (bool, error), error)

ToCsvRow return converter from output table cell (sub_id, dimensions, acc0, acc1, acc2) to csv row []string (sub_id, dimensions, acc0 value, acc1 value,....).

Converter return isNotEmpty flag, it return false if IsNoZero or IsNoNull is set and all values are empty or zero. Converter return error if len(row) not equal to number of fields in csv record. If dimension type is enum based then csv row is enum code. If ValueName is "" empty then all accumulators converted else one

type CellAllAccLocaleConverter added in v1.17.6 

type CellAllAccLocaleConverter struct {
	CellAllAccConverter
	Lang    string            // language code, expected to compatible with BCP 47 language tag
	LangDef *LangMeta         // language metadata to find translations
	DimsTxt []TableDimsTxtRow // output table dimension text rows: table_dims_txt join to model_table_dic
	EnumTxt []TypeEnumTxtRow  // type enum text rows: type_enum_txt join to model_type_dic
	AccTxt  []TableAccTxtRow  // output table accumulator text rows: table_acc_txt join to model_table_dic
}

Converter for output table accumulators to implement CsvLocaleConverter interface.

func (*CellAllAccLocaleConverter) CsvHeader added in v1.17.6 

func (cellCvt *CellAllAccLocaleConverter) CsvHeader() ([]string, error)

CsvHeader return first line for csv file: column names, for example: sub_id,Age,Sex,AVG Income,SE Income For example: acc_name,sub_id,Age,Sex,acc_value

func (*CellAllAccLocaleConverter) ToCsvRow added in v1.17.6 

func (cellCvt *CellAllAccLocaleConverter) ToCsvRow() (func(interface{}, []string) (bool, error), error)

ToCsvRow return converter from output table cell (sub_id, dimensions, acc0, acc1, acc2) to language-specific csv []string row of (sub_id, dimension enum labels, acc0 value, acc1 value,....).

Converter return isNotEmpty flag, it return false if IsNoZero or IsNoNull is set and all values are empty or zero. Converter return error if len(row) not equal to number of fields in csv record. If dimension type is enum based then csv row is enum label. If ValueName is "" empty then all accumulators converted else one Values and dimesions of built-in types converted to locale-specific strings, e.g.: 1234.56 => 1 234,56

type CellCodeAcc 

type CellCodeAcc struct {
	AccId int // output table accumulator id
	SubId int // output table subvalue id
	// contains filtered or unexported fields
}

CellCodeAcc is value of output table accumulator. Dimension(s) items are enum codes, not enum ids.

type CellCodeAllAcc 

type CellCodeAllAcc struct {
	Dims   []string  // dimensions as enum codes or string of item if dimension type simple then
	SubId  int       // output table subvalue id
	IsNull []bool    // if true then value is NULL
	Value  []float64 // accumulator value(s)
}

CellCodeAllAcc is value of multiple output table accumulators. Dimension(s) items are enum codes, not enum ids.

type CellCodeExpr 

type CellCodeExpr struct {
	ExprId int // output table expression id
	// contains filtered or unexported fields
}

CellCodeExpr is value of output table expression. Dimension(s) items are enum codes, not enum ids.

type CellCodeMicro added in v1.13.0 

type CellCodeMicro struct {
	Key  uint64      // microdata key, part of row primary key: entity_key
	Attr []attrValue // attributes value: built-in type value or enum code
}

CellCodeMicro is a row of entity microdata: entity_key and attribute values, if attribute type is enum based value is enum code

type CellCodeMicroCalc added in v1.16.0 

type CellCodeMicroCalc struct {
	Attr      []attrValue // group by attributes and calculated value attribute, using enum codes for enum based attributes
	CalcName  string      // aggregation name
	RunDigest string      // model run digest
}

CellCodeMicroCalc is value of entity microdata calculated expression. if attribute type is enum based value is enum code

type CellCodeParam 

type CellCodeParam struct {
	SubId int // parameter subvalue id
	// contains filtered or unexported fields
}

CellCodeParam is value of input parameter. Dimension(s) items are enum codes, not enum ids.

type CellCodeTableCalc added in v1.15.4 

type CellCodeTableCalc struct {
	CalcName  string // calculated expression name
	RunDigest string // model run digest
	// contains filtered or unexported fields
}

CellCodeTableCalc is value of output table calculated expression. Dimension(s) items are enum codes, not enum ids.

type CellEntityConverter added in v1.16.0 

type CellEntityConverter struct {
	ModelDef    *ModelMeta     // model metadata
	Name        string         // model entity name
	EntityGen   *EntityGenMeta // model run entity generation
	IsIdCsv     bool           // if true then use enum id's else use enum codes
	DoubleFmt   string         // if not empty then format string is used to sprintf if value type is float, double, long double
	IsNoZeroCsv bool           // if true then do not write zero values into csv output
	IsNoNullCsv bool           // if true then do not write NULL values into csv output
	// contains filtered or unexported fields
}

CellMicroConverter is a parent for for entity microdata converters.

type CellExpr 

type CellExpr struct {
	ExprId int // output table expression id
	// contains filtered or unexported fields
}

CellExpr is value of output table expression.

type CellExprConverter added in v1.9.2 

type CellExprConverter struct {
	CellTableConverter // model metadata and output table name
}

CellExprConverter is a converter for output table expression to implement CsvConverter interface.

func (*CellExprConverter) CsvFileName added in v1.9.2 

func (cellCvt *CellExprConverter) CsvFileName() (string, error)

CsvFileName return file name of csv file to store output table expression rows

func (*CellExprConverter) CsvHeader added in v1.9.2 

func (cellCvt *CellExprConverter) CsvHeader() ([]string, error)

CsvHeader return first line for csv file: column names. For example: expr_name,dim0,dim1,expr_value or if IsIdCsv is true: expr_id,dim0,dim1,expr_value

func (*CellExprConverter) IdToCodeCell added in v1.9.2 

func (cellCvt *CellExprConverter) IdToCodeCell(modelDef *ModelMeta, name string) (func(interface{}) (interface{}, error), error)

IdToCodeCell return converter from output table cell of ids: (expr_id, dimensions enum ids, value) to cell of codes: (expr_id, dimensions as enum codes, value).

If dimension type is enum based then dimensions enum ids can be converted to enum code. If dimension type is simple (bool or int) then dimension value converted to string.

func (*CellExprConverter) IsUseEnumId added in v1.13.0 

func (cellCvt *CellExprConverter) IsUseEnumId() bool

return true if csv converter is using enum id's for dimensions

func (*CellExprConverter) KeyIds added in v1.9.7 

func (cellCvt *CellExprConverter) KeyIds(name string) (func(interface{}, []int) error, error)

KeyIds return converter to copy primary key: (expr_id, dimension ids) into key []int.

Converter will return error if len(key) not equal to row key size.

func (*CellExprConverter) ToCell added in v1.17.3 

func (cellCvt *CellExprConverter) ToCell() (func(row []string) (interface{}, error), error)

CsvToCell return closure to convert csv row []string to output table expression cell (dimensions and value).

Converter return error if len(row) not equal to number of fields in cell db-record. If dimension type is enum based then csv row is enum code and it is converted into cell.DimIds (into dimension type type enum ids).

func (*CellExprConverter) ToCsvIdRow added in v1.13.0 

func (cellCvt *CellExprConverter) ToCsvIdRow() (func(interface{}, []string) (bool, error), error)

ToCsvIdRow return converter from output table cell (expr_id, dimensions, value) to csv id's row []string.

Converter return isNotEmpty flag, it return false if IsNoZero or IsNoNull is set and cell value is empty or zero. Converter simply does Sprint() for each dimension item id, expression id and value. Converter will return error if len(row) not equal to number of fields in csv record.

func (*CellExprConverter) ToCsvRow added in v1.13.0 

func (cellCvt *CellExprConverter) ToCsvRow() (func(interface{}, []string) (bool, error), error)

ToCsvRow return converter from output table cell (expr_id, dimensions, value) to csv row []string (expr_name, dimensions, value).

Converter return isNotEmpty flag, it return false if IsNoZero or IsNoNull is set and cell value is empty or zero. Converter return error if len(row) not equal to number of fields in csv record. If dimension type is enum based then csv row is enum code.

type CellExprLocaleConverter added in v1.17.4 

type CellExprLocaleConverter struct {
	CellExprConverter
	Lang    string            // language code, expected to compatible with BCP 47 language tag
	LangDef *LangMeta         // language metadata to find translations
	DimsTxt []TableDimsTxtRow // output table dimension text rows: table_dims_txt join to model_table_dic
	EnumTxt []TypeEnumTxtRow  // type enum text rows: type_enum_txt join to model_type_dic
	ExprTxt []TableExprTxtRow // output table expression text rows: table_expr_txt join to model_table_dic
}

Converter for output table expression to implement CsvLocaleConverter interface.

func (*CellExprLocaleConverter) CsvHeader added in v1.17.6 

func (cellCvt *CellExprLocaleConverter) CsvHeader() ([]string, error)

CsvHeader return first line for csv file: column names. For example: expr_name,Age,Sex,expr_value

func (*CellExprLocaleConverter) ToCsvRow added in v1.17.4 

func (cellCvt *CellExprLocaleConverter) ToCsvRow() (func(interface{}, []string) (bool, error), error)

Return converter from output table cell (expr_id, dimensions, value) to language-specific csv []string row of dimension enum labels and value.

Converter return isNotEmpty flag, it return false if IsNoZero or IsNoNull is set and cell value is empty or zero. Converter return error if len(row) not equal to number of fields in csv record. If dimension type is enum based then csv row is enum label. Value and dimesions of built-in types converted to locale-specific strings, e.g.: 1234.56 => 1 234,56

type CellIntKeys added in v1.13.0 

type CellIntKeys interface {

	// KeyIds return converter to copy row primary key into key []int.
	// Row primary keys are:
	//   parameter: (sub_id, dimension ids)
	//   accumulators: (acc_id, sub_id, dimension ids)
	//   all accumulators: (sub_id, dimension ids)
	//   expressions: (expr_id, dimension ids)
	KeyIds(name string) (func(interface{}, []int) error, error)
}

CellIntKeys provide method to get a copy of cell keys as []int for parameter or output table row, for example, return [parameter row sub id and dimension ids].

type CellMicro added in v1.13.0 

type CellMicro struct {
	Key  uint64      // microdata key, part of row primary key: entity_key
	Attr []attrValue // attributes value: built-in type value or enum id
}

CellMicro is a row of entity microdata: entity_key and attribute values, if attribute type is enum based value is enum id

type CellMicroCalc added in v1.16.0 

type CellMicroCalc struct {
	Attr   []attrValue // group by attributes and calculated value attribute, using enum id's for enum based attributes
	CalcId int         // aggregation id
	RunId  int         // model run id
}

CellMicroCalc is value of entity microdata calculated expression. if attribute type is enum based value is enum id

type CellMicroCalcConverter added in v1.16.0 

type CellMicroCalcConverter struct {
	CellEntityConverter          // model metadata, entity generation and and attributes
	CalcMaps                     // map between runs digest and id and calculations name and id
	GroupBy             []string // attributes to group by
	// contains filtered or unexported fields
}

CellMicroCalcConverter is a converter for calculated microdata row to implement CsvConverter interface.

func (*CellMicroCalcConverter) CsvFileName added in v1.16.0 

func (cellCvt *CellMicroCalcConverter) CsvFileName() (string, error)

CsvFileName return file name of csv file to store calculated microdata rows

func (*CellMicroCalcConverter) CsvHeader added in v1.16.0 

func (cellCvt *CellMicroCalcConverter) CsvHeader() ([]string, error)

CsvHeader return first line for csv file column names: names of group by attributes. For example: run_digest,calc_name,AgeGroup,Income,calc_value Or: run_id,calc_id,AgeGroup,Income,calc_value

func (*CellMicroCalcConverter) IdToCodeCell added in v1.16.0 

func (cellCvt *CellMicroCalcConverter) IdToCodeCell(modelDef *ModelMeta, _ string) (func(interface{}) (interface{}, error), error)

IdToCodeCell return converter from calculated microdata cell of ids: (run id, calc_id, group by attributes as built-in values or enum id's, calculated value attribute) into cell of codes: (run_digest, CalcName, group by attributes as enum codes or built-in values, calc_value).

If attribute type is enum based then attribute enum id converted to enum code. If attribute type is built-in (bool, int, float) then return attribute value as is, no conversion.

func (*CellMicroCalcConverter) IsUseEnumId added in v1.16.0 

func (cellCvt *CellMicroCalcConverter) IsUseEnumId() bool

return true if csv converter is using enum id's for dimensions

func (*CellMicroCalcConverter) SetCalcIdNameMap added in v1.16.0 

func (cellCvt *CellMicroCalcConverter) SetCalcIdNameMap(calcLt []CalculateLayout) error

Set calculation name to Id and Id to name maps

func (*CellMicroCalcConverter) ToCsvIdRow added in v1.16.0 

func (cellCvt *CellMicroCalcConverter) ToCsvIdRow() (func(interface{}, []string) (bool, error), error)

ToCsvIdRow return converter from microdata cell: (microdata key, attributes as enum id or built-in type value) to csv id's row []string.

Converter return isNotEmpty flag, it is always true if there were no error during conversion. Converter simply does Sprint() for key and each attribute value, if value is NULL then empty "" string used. Converter will return error if len(row) not equal to number of fields in csv record.

func (*CellMicroCalcConverter) ToCsvRow added in v1.16.0 

func (cellCvt *CellMicroCalcConverter) ToCsvRow() (func(interface{}, []string) (bool, error), error)

ToCsvRow return converter from microdata cell: (microdata key, attributes as enum code or built-in type value) to csv row []string.

Converter return isNotEmpty flag, it is always true if there were no error during conversion. Converter simply does Sprint() for key and each attribute value, if value is NULL then empty "" string used. If attribute type is float and double format is not empty "" string then converter does Sprintf(using double format). If attribute type is enum based then converter return enum code for attribute enum id. Converter will return error if len(row) not equal to number of fields in csv record.

type CellMicroConverter added in v1.13.0 

type CellMicroConverter struct {
	CellEntityConverter // model metadata, entity generation and and attributes
}

CellMicroConverter is a converter for entity microdata row to implement CsvConverter interface.

func (*CellMicroConverter) CodeToIdCell added in v1.13.0 

func (cellCvt *CellMicroConverter) CodeToIdCell(modelDef *ModelMeta, _ string) (func(interface{}) (interface{}, error), error)

CodeToIdCell return converter from parameter cell of codes: (entity key, attributes as built-in values or enum codes) to cell of ids: (entity key, attributes as built-in values or enum id's)

If attribute type is enum based then attribute enum codes converted to enum ids. If attribute type is built-in (bool, int, float) then attribute value converted from string to attribute type.

func (*CellMicroConverter) CsvFileName added in v1.13.0 

func (cellCvt *CellMicroConverter) CsvFileName() (string, error)

CsvFileName return file name of csv file to store entity microdata rows

func (*CellMicroConverter) CsvHeader added in v1.13.0 

func (cellCvt *CellMicroConverter) CsvHeader() ([]string, error)

CsvHeader return first line for csv file: column names. For example: key,AgeGroup,Income.

func (*CellMicroConverter) IdToCodeCell added in v1.13.0 

func (cellCvt *CellMicroConverter) IdToCodeCell(modelDef *ModelMeta, _ string) (func(interface{}) (interface{}, error), error)

IdToCodeCell return converter from microdata cell of ids: (entity key, attributes as built-in values or enum id's) into cell of codes: (entity key, attributes as built-in values or enum codes)

If attribute type is enum based then attribute enum id converted to enum code. If attribute type is built-in (bool, int, float) then return attribute value as is, no conversion.

func (*CellMicroConverter) IsUseEnumId added in v1.13.0 

func (cellCvt *CellMicroConverter) IsUseEnumId() bool

return true if csv converter is using enum id's for attributes

func (*CellMicroConverter) ToCell added in v1.17.3 

func (cellCvt *CellMicroConverter) ToCell() (func(row []string) (interface{}, error), error)

CsvToCell return closure to convert csv row []string to microdata cell (key, attributes value).

If attribute type is enum based then csv row contains enum code and it is converted into cell attribute enum id. It does return error if len(row) not equal to number of fields in cell db-record. Converter will return error if len(row) not equal to number of fields in csv record.

func (*CellMicroConverter) ToCsvIdRow added in v1.13.0 

func (cellCvt *CellMicroConverter) ToCsvIdRow() (func(interface{}, []string) (bool, error), error)

ToCsvIdRow return converter from microdata cell: (microdata key, attributes as enum id or built-in type value) to csv id's row []string.

Converter return isNotEmpty flag, it return false if IsNoZero or IsNoNull is set and all attributes values are empty or zero, only attributes of type float or integer or string are considered as "value" attributes. Converter simply does Sprint() for key and each attribute value, if value is NULL then empty "" string used. Converter will return error if len(row) not equal to number of fields in csv record.

func (*CellMicroConverter) ToCsvRow added in v1.13.0 

func (cellCvt *CellMicroConverter) ToCsvRow() (func(interface{}, []string) (bool, error), error)

ToCsvRow return converter from microdata cell: (microdata key, attributes as enum id or built-in type value) to csv row []string.

Converter return isNotEmpty flag, it return false if IsNoZero or IsNoNull is set and all values of float or integer or string type are empty or zero. Converter simply does Sprint() for key and each attribute value, if value is NULL then empty "" string used. If attribute type is float and double format is not empty "" string then converter does Sprintf(using double format). If attribute type is enum based then converter return enum code for attribute enum id. Converter will return error if len(row) not equal to number of fields in csv record.

type CellMicroLocaleConverter added in v1.17.6 

type CellMicroLocaleConverter struct {
	CellMicroConverter
	Lang    string             // language code, expected to compatible with BCP 47 language tag
	LangDef *LangMeta          // language metadata to find translations
	EnumTxt []TypeEnumTxtRow   // type enum text rows: type_enum_txt join to model_type_dic
	AttrTxt []EntityAttrTxtRow // entity attributes text rows: entity_attr_txt join to model_entity_dic table
	// contains filtered or unexported fields
}

Converter for entity microdata to implement CsvLocaleConverter interface.

func (*CellMicroLocaleConverter) CsvHeader added in v1.17.6 

func (cellCvt *CellMicroLocaleConverter) CsvHeader() ([]string, error)

CsvHeader return first line for csv file: column names. For example: key,Age Group,Taxable Income.

func (*CellMicroLocaleConverter) ToCsvRow added in v1.17.6 

func (cellCvt *CellMicroLocaleConverter) ToCsvRow() (func(interface{}, []string) (bool, error), error)

Return converter from microdata cell: (microdata key, attributes as enum id or built-in type value) to language-specific csv []string row of dimension enum labels and value.

Converter return isNotEmpty flag, it return false if IsNoZero or IsNoNull is set and all values of float or integer or string type are empty or zero. Converter will return error if len(row) not equal to number of fields in csv record. Microdata row key and attribute values of built-in type converted to locale-specific strings, e.g.: 1234.56 => 1 234,56. If value is NULL then empty "" string used. If attribute type is enum based then csv vslur is enum label.

type CellParam 

type CellParam struct {
	SubId int // parameter subvalue id
	// contains filtered or unexported fields
}

CellParam is value of input parameter.

type CellParamConverter added in v1.9.2 

type CellParamConverter struct {
	ModelDef  *ModelMeta // model metadata
	Name      string     // parameter name
	IsIdCsv   bool       // if true then use enum id's else use enum codes
	DoubleFmt string     // if not empty then format string is used to sprintf if value type is float, double, long double
	// contains filtered or unexported fields
}

CellParamConverter is a converter for input parameter to implement CsvConverter interface.

func (*CellParamConverter) CodeToIdCell added in v1.9.2 

func (cellCvt *CellParamConverter) CodeToIdCell(modelDef *ModelMeta, name string) (func(interface{}) (interface{}, error), error)

CodeToIdCell return converter from parameter cell of codes: (sub id, dimensions as enum code, value) to cell of ids: (sub id, dimensions, value)

If dimension type is enum based then dimensions enum codes converted to enum ids. If dimension type is simple (bool or int) then dimension code converted from string to dimension type. If parameter type is enum based then cell value enum code converted to enum id.

func (*CellParamConverter) CsvFileName added in v1.9.2 

func (cellCvt *CellParamConverter) CsvFileName() (string, error)

CsvFileName return file name of csv file to store parameter rows

func (*CellParamConverter) CsvHeader added in v1.9.2 

func (cellCvt *CellParamConverter) CsvHeader() ([]string, error)

CsvHeader return first line for csv file: column names. For example: sub_id,dim0,dim1,param_value.

func (*CellParamConverter) IdToCodeCell added in v1.9.2 

func (cellCvt *CellParamConverter) IdToCodeCell(modelDef *ModelMeta, name string) (func(interface{}) (interface{}, error), error)

IdToCodeCell return converter from parameter cell of ids: (sub id, dimensions, value) to cell of codes: (sub id, dimensions as enum code, value)

If dimension type is enum based then dimensions enum ids can be converted to enum code. If dimension type is simple (bool or int) then dimension value converted to string. If parameter type is enum based then cell value enum id converted to enum code.

func (*CellParamConverter) IsUseEnumId added in v1.13.0 

func (cellCvt *CellParamConverter) IsUseEnumId() bool

return true if csv converter is using enum id's for dimensions

func (*CellParamConverter) KeyIds added in v1.9.7 

func (cellCvt *CellParamConverter) KeyIds(name string) (func(interface{}, []int) error, error)

KeyIds return converter to copy primary key: (sub id, dimension ids) into key []int.

Converter will return error if len(key) not equal to row key size.

func (*CellParamConverter) ToCell added in v1.17.3 

func (cellCvt *CellParamConverter) ToCell() (func(row []string) (interface{}, error), error)

CsvToCell return closure to convert csv row []string to parameter cell (sub id, dimensions, value).

It does return error if len(row) not equal to number of fields in cell db-record. If dimension type is enum based then csv row is enum code and it is converted into cell.DimIds (into dimension type type enum ids). If parameter type is enum based then csv row value is enum code and it is converted into value enum id.

func (*CellParamConverter) ToCsvIdRow added in v1.13.0 

func (cellCvt *CellParamConverter) ToCsvIdRow() (func(interface{}, []string) (bool, error), error)

Return converter from parameter cell (sub id, dimensions, value) to csv id's row []string.

Converter return isNotEmpty flag, it is always true if there were no error during conversion. Converter simply does Sprint() for each sub-value id, dimension item id and value. Converter will return error if len(row) not equal to number of fields in csv record.

func (*CellParamConverter) ToCsvRow added in v1.13.0 

func (cellCvt *CellParamConverter) ToCsvRow() (func(interface{}, []string) (bool, error), error)

Return converter from parameter cell (sub id, dimensions, value) to csv row []string.

Converter return isNotEmpty flag, it is always true if there were no error during conversion. Converter will return error if len(row) not equal to number of fields in csv record. If dimension type is enum based then csv row is enum code. If parameter type is enum based then csv row value is enum code.

type CellParamLocaleConverter added in v1.17.3 

type CellParamLocaleConverter struct {
	CellParamConverter
	Lang    string            // language code, expected to compatible with BCP 47 language tag
	DimsTxt []ParamDimsTxtRow // parameter dimension text rows: parameter_dims_txt join to model_parameter_dic
	EnumTxt []TypeEnumTxtRow  // type enum text rows: type_enum_txt join to model_type_dic
}

Converter for input parameter to implement CsvLocaleConverter interface.

func (*CellParamLocaleConverter) CsvHeader added in v1.17.6 

func (cellCvt *CellParamLocaleConverter) CsvHeader() ([]string, error)

CsvHeader return first line for csv file: column names. For example: sub_id,Age,Sex,param_value.

func (*CellParamLocaleConverter) ToCsvRow added in v1.17.3 

func (cellCvt *CellParamLocaleConverter) ToCsvRow() (func(interface{}, []string) (bool, error), error)

Return converter from parameter cell (sub id, dimensions, value) to language-specific csv []string row of enum labels and value.

Converter return isNotEmpty flag, it is always true if there were no error during conversion. Converter will return error if len(row) not equal to number of fields in csv record. If dimension type is enum based then csv row is enum label. If parameter type is enum based then csv row value is enum label. Value and dimesions of built-in types converted to locale-specific strings, e.g.: 1234.56 => 1 234,56

type CellTableCalc added in v1.15.4 

type CellTableCalc struct {
	CalcId int // calculated expression id
	RunId  int // model run id
	// contains filtered or unexported fields
}

CellTableCalc is value of output table calculated expression.

type CellTableCalcConverter added in v1.15.4 

type CellTableCalcConverter struct {
	CellTableConverter // model metadata and output table name
	CalcMaps           // map between runs digest and id and calculations name and id
}

CellTableCalcConverter is a converter for output table calculated cell to implement CsvConverter interface.

func (*CellTableCalcConverter) CsvFileName added in v1.15.4 

func (cellCvt *CellTableCalcConverter) CsvFileName() (string, error)

CsvFileName return file name of csv file to store output table calculated rows

func (*CellTableCalcConverter) CsvHeader added in v1.15.4 

func (cellCvt *CellTableCalcConverter) CsvHeader() ([]string, error)

CsvHeader return first line for csv file: column names, it's look like: run_digest,calc_id,dim0,dim1,calc_value

func (*CellTableCalcConverter) IdToCodeCell added in v1.15.4 

func (cellCvt *CellTableCalcConverter) IdToCodeCell(modelDef *ModelMeta, name string) (func(interface{}) (interface{}, error), error)

IdToCodeCell return converter from output table calculated cell of ids: (run_id, calc_id, dimensions enum ids, calc_value) to cell of codes: (run_digest, CalcName, dimensions as enum codes, calc_value).

If dimension type is enum based then dimensions enum ids can be converted to enum code. If dimension type is simple (bool or int) then dimension value converted to string.

func (*CellTableCalcConverter) IsUseEnumId added in v1.15.4 

func (cellCvt *CellTableCalcConverter) IsUseEnumId() bool

return true if csv converter is using enum id's for dimensions

func (*CellTableCalcConverter) KeyIds added in v1.15.4 

func (cellCvt *CellTableCalcConverter) KeyIds(name string) (func(interface{}, []int) error, error)

KeyIds return converter to copy primary key: (run_id, calc_id, dimension ids) into key []int.

Converter will return error if len(key) not equal to row key size.

func (*CellTableCalcConverter) SetCalcIdNameMap added in v1.15.6 

func (cellCvt *CellTableCalcConverter) SetCalcIdNameMap(calcLt []CalculateTableLayout) error

Set calculation name to Id and Id to name maps

func (*CellTableCalcConverter) ToCsvIdRow added in v1.15.4 

func (cellCvt *CellTableCalcConverter) ToCsvIdRow() (func(interface{}, []string) (bool, error), error)

ToCsvIdRow return converter from output table calculated cell (run_id, calc_id, dimensions, calc_value) to csv id's row []string.

Converter return isNotEmpty flag, it is always true if there were no error during conversion. Converter simply does Sprint() for each dimension item id, run id and value. Converter will return error if len(row) not equal to number of fields in csv record.

func (*CellTableCalcConverter) ToCsvRow added in v1.15.4 

func (cellCvt *CellTableCalcConverter) ToCsvRow() (func(interface{}, []string) (bool, error), error)

ToCsvRow return converter from output table calculated cell (run_id, calc_id, dimensions, calc_value) to csv row []string (run digest, calc_name, dimensions, calc_value).

Converter return isNotEmpty flag, it is always true if there were no error during conversion. Converter will return error if len(row) not equal to number of fields in csv record. Converter will return error if run_id not exist in the list of model runs (in run_lst table). Double format string is used if parameter type is float, double, long double. If dimension type is enum based then csv row is enum code.

type CellTableConverter added in v1.13.0 

type CellTableConverter struct {
	ModelDef *ModelMeta // model metadata
	Name     string     // output table name

	IsIdCsv     bool   // if true then use enum id's else use enum codes
	DoubleFmt   string // if not empty then format string is used to sprintf if value type is float, double, long double
	IsNoZeroCsv bool   // if true then do not write zero values into csv output
	IsNoNullCsv bool   // if true then do not write NULL values into csv output
	// contains filtered or unexported fields
}

CellTableConverter is a parent for for output table converters.

type CellToCodeConverter 

type CellToCodeConverter interface {

	// IdToCodeCell return converter from id cell to code cell.
	// Cell is dimensions and value of parameter or output table.
	// It does convert from enum id to code for all dimensions and enum-based parameter value.
	IdToCodeCell(modelDef *ModelMeta, name string) (func(interface{}) (interface{}, error), error)
}

CellToCodeConverter provide methods to convert parameters or output table row from enum id to enum code. If dimension type is enum based then dimensions enum ids can be converted to enum code. If dimension type is simple (bool or int) then dimension value converted to string. If parameter type is enum based then cell value enum id converted to enum code.

type CellToIdConverter 

type CellToIdConverter interface {

	// CodeToIdCell return converter from code cell to id cell.
	// Cell is dimensions and value of parameter or output table.
	// It does convert from enum code to id for all dimensions and enum-based parameter value.
	CodeToIdCell(modelDef *ModelMeta, name string) (func(interface{}) (interface{}, error), error)
}

CellToIdConverter provide methods to convert parameters or output table row from enum code to enum id. If dimension type is enum based then dimensions enum codes converted to enum ids. If dimension type is simple (bool or int) then dimension code converted from string to dimension type. If parameter type is enum based then cell value enum code converted to enum id.

type CsvConverter 

type CsvConverter interface {
	// return file name of csv file to store parameter or output table rows
	CsvFileName() (string, error)

	// return true if csv converter is using enum id's for dimensions or attributes
	IsUseEnumId() bool

	// return first line of csv file with column names: expr_name,dim0,dim1,expr_value.
	// if IsIdCsv is true: expr_id,dim0,dim1,expr_value
	// if isAllAcc is true: sub_id,dim0,dim1,acc0,acc1,acc2
	CsvHeader() ([]string, error)

	// return converter from cell of parameter, output table or microdata to csv id's row []string.
	// converter simply sprint() dimension id's and value into []string buffer.
	// converter return isNotEmpty flag if cell value is not empty.
	ToCsvIdRow() (func(interface{}, []string) (bool, error), error)

	// return converter from cell of parameter, output table or microdata to csv row []string.
	// it does convert from enum id to code for all dimensions into []string buffer.
	// if this is a enum-based parameter value then it is also converted to enum code.
	// converter return isNotEmpty flag if cell value is not empty.
	ToCsvRow() (func(interface{}, []string) (bool, error), error)
}

CsvConverter provide methods to convert parameter row, output table row or microdata row from or to row []string for csv file. Double format string is used for values or if value type is float, double, long double. If dimension type is enum based then csv row is enum code and cell.DimIds is enum id. If parameter type is enum based then cell value is enum id and csv row value is enum code.

type CsvIntKeysConverter added in v1.13.0 

type CsvIntKeysConverter interface {
	CsvConverter // convert parameter row or output table row to row []string for csv file
	CellIntKeys  // provide a method to get row keys as []int, for example, get sub id and dimension ids
}

CsvConverter provide methods to convert parameters or output table data to row []string for csv file.

type CsvLocaleConverter added in v1.17.4 

type CsvLocaleConverter interface {
	// return converter from cell of parameter, output table or microdata to csv row []string.
	// If dimension type is enum based then csv row is enum label.
	// If parameter type is enum based then csv row value is enum label.
	// Value and dimesions of built-in types converted to locale-specific strings, e.g.: 1234.56 => 1 234,56
	ToCsvRow() (func(interface{}, []string) (bool, error), error)

	// return first line of csv file with column names, for example: expr_name,Age,City,expr_value.
	CsvHeader() ([]string, error)
}

Provide locale-specific methods to convert parameter row, output table row or microdata row to []string row for csv file. For example convert dimension enum id to enum label in specific language or use locale format for numeric values: 1234.56 => 1 234,56

type CsvToCellConverter added in v1.17.3 

type CsvToCellConverter interface {
	// return converter from csv row []string to parameter, output table or microdata cell (dimensions and value or microdata key and attributes value)
	ToCell() (func(row []string) (interface{}, error), error)
}

Convert from csv row []string to parameter, output table or microdata cell (dimensions and value or microdata key and attributes value).

type DescrNote 

type DescrNote struct {
	LangCode string // lang_code VARCHAR(32)  NOT NULL
	Descr    string // descr     VARCHAR(255) NOT NULL
	Note     string // note      VARCHAR(32000)
}

DescrNote is a holder for language code, descripriton and notes

type EntityAttrRow added in v1.13.0 

type EntityAttrRow struct {
	ModelId    int    // model_id        INT          NOT NULL
	EntityId   int    // model_entity_id INT          NOT NULL
	AttrId     int    // attr_id         INT          NOT NULL
	Name       string // attr_name       VARCHAR(255) NOT NULL
	TypeId     int    // model_type_id   INT          NOT NULL
	IsInternal bool   // is_internal     SMALLINT     NOT NULL
	// contains filtered or unexported fields
}

EntityAttrRow is db row of entity_attr join to model_entity_dic table

type EntityAttrTxtRow added in v1.13.0 

type EntityAttrTxtRow struct {
	ModelId  int    // model_id        INT          NOT NULL
	EntityId int    // model_entity_id INT          NOT NULL
	AttrId   int    // attr_id         INT          NOT NULL
	LangCode string // lang_code       VARCHAR(32)  NOT NULL
	Descr    string // descr           VARCHAR(255) NOT NULL
	Note     string // note            VARCHAR(32000)
}

EntityAttrTxtRow is db row of entity_attr_txt join to model_entity_dic table

type EntityDicRow added in v1.13.0 

type EntityDicRow struct {
	ModelId   int    // model_id         INT          NOT NULL
	EntityId  int    // model_entity_id  INT          NOT NULL
	EntityHid int    // entity_hid       INT          NOT NULL, -- unique entity id
	Name      string // entity_name      VARCHAR(255) NOT NULL
	Digest    string // entity_digest    VARCHAR(32)  NOT NULL
}

EntityDicRow is db row of entity_dic join to model_entity_dic table.

EntityHid (entity_dic.entity_hid) is db-unique id of the entity, use digest to find same table in other db. EntityId (model_entity_dic.model_entity_id) is model-unique entity id, assigned by model compiler.

type EntityGenMeta added in v1.13.0 

type EntityGenMeta struct {
	GenAttr []entityGenAttrRow // entity generation attributes: entity_gen_attr join to entity_gen
	// contains filtered or unexported fields
}

EntityGenMeta is a holder of entity generation db rows from entity_gen, model_entity_dic, entity_gen_attr and run_entity tables Entity generation is a model entity with set of attributes included in particular model run(s). Model run typically include less attributes in microdata output than entity has in model metadata.

func GetEntityGenList added in v1.13.0 

func GetEntityGenList(dbConn *sql.DB, runId int) ([]EntityGenMeta, error)

GetEntityGenList return entity generation rows by run id: entity_gen and entity_gen_attr table rows.

type EntityGroupLstRow added in v1.17.6 

type EntityGroupLstRow struct {
	ModelId  int    // model_id        INT          NOT NULL
	EntityId int    // model_entity_id INT          NOT NULL
	GroupId  int    // group_id        INT          NOT NULL
	Name     string // group_name      VARCHAR(255) NOT NULL
	IsHidden bool   // is_hidden       SMALLINT     NOT NULL
}

EntityGroupLstRow is db row of entity_group_lst table

type EntityGroupMeta added in v1.17.6 

type EntityGroupMeta struct {
	EntityGroupLstRow                    // entity attribute group rows: entity_group_lst
	GroupPc           []EntityGroupPcRow // group parent-child relationship rows: entity_group_pc
}

EntityGroupMeta is db rows to describe parent-child group of entity attributes, it is join of entity_group_lst to entity_group_pc

type EntityGroupPcRow added in v1.17.6 

type EntityGroupPcRow struct {
	ModelId      int // model_id        INT NOT NULL
	EntityId     int // model_entity_id INT NOT NULL
	GroupId      int // group_id        INT NOT NULL
	ChildPos     int // child_pos       INT NOT NULL
	ChildGroupId int // child_group_id  INT, -- if not IS NULL then child group
	AttrId       int // attr_id         INT, -- if not IS NULL then entity attribute index
}

EntityGroupPcRow is db row of entity_group_pc table

type EntityGroupTxtRow added in v1.17.6 

type EntityGroupTxtRow struct {
	ModelId   int // model_id        INT NOT NULL
	EntityId  int // model_entity_id INT NOT NULL
	GroupId   int // group_id        INT NOT NULL
	DescrNote     // language, description, notes
}

EntityGroupTxtRow is db row of entity_group_txt table

type EntityMeta added in v1.13.0 

type EntityMeta struct {
	EntityDicRow                 // model entity row: entity_dic join to model_entity_dic table
	Attr         []EntityAttrRow // entity attribute rows: entity_attr join to model_entity_dic table
}

EntityMeta is entity metadata: entity name, digest, attributes

func (*EntityMeta) AttrByKey added in v1.13.0 

func (entity *EntityMeta) AttrByKey(attrId int) (int, bool)

AttrByKey return index of entity attribute by key: attrId

func (*EntityMeta) AttrByName added in v1.13.0 

func (entity *EntityMeta) AttrByName(name string) (int, bool)

AttrByName return index of entity attribute by name

type EntityRunPub added in v1.13.0 

type EntityRunPub struct {
	Name        string   // entity name
	GenDigest   string   // digest of entity generation, not empty only as result of select from run_entity; input from "public" digest is ignored
	ValueDigest string   // value digest, not empty only as result of select from run_entity; input from "public" value digest is ignored
	RowCount    int      // if not zero then entity microdata row count
	Attr        []string // names of entity generation attributes
}

EntityRunPub is "public" metadata of entity run generation and attributes for json import-export

type EntityTxtRow added in v1.13.0 

type EntityTxtRow struct {
	ModelId  int    // model_id         INT          NOT NULL
	EntityId int    // model_entity_id  INT          NOT NULL
	LangCode string // lang_code        VARCHAR(32)  NOT NULL
	Descr    string // descr            VARCHAR(255) NOT NULL
	Note     string // note             VARCHAR(32000)
}

EntityTxtRow is db row of entity_dic_txt join to model_entity_dic table

type Facet 

type Facet uint8

Facet is type to define database provider and driver facets, ie: name of bigint type 

const (
	DefaultFacet Facet = iota // common default db facet
	SqliteFacet               // SQLite db facet
	PgSqlFacet                // PostgreSQL db facet
	MySqlFacet                // MySQL and MariaDB facet
	MsSqlFacet                // MS SQL db facet
	OracleFacet               // Oracle db facet
	Db2Facet                  // DB2 db facet
)

func Open 

func Open(dbConnStr, dbDriver string, isFacetRequired bool) (*sql.DB, Facet, error)

Open database connection.

Default driver name: "SQLite" and connection string is compatible with model connection, ie: 

Database=modelName.sqlite; Timeout=86400; OpenMode=ReadWrite;

Otherwise it is expected to be driver-specific connection string, ie: 

DSN=ms2014; UID=sa; PWD=secret;
file:m1.sqlite?mode=rw&_busy_timeout=86400000

If isFacetRequired is true then database facet determined

func (Facet) String 

func (facet Facet) String() string

String is default printable value of db facet, Stringer implementation

type FilterColumn 

type FilterColumn struct {
	Name   string   // dimension or attribute name
	Op     FilterOp // filter operator: equal, IN, BETWEEN
	Values []string // enum code(s): one, two or many values depending on filter condition
}

FilterColumn define dimension or attribute column and condition to filter enum codes to build select where

type FilterIdColumn 

type FilterIdColumn struct {
	Name    string   // dimension or attribute name
	Op      FilterOp // filter operator: equal, IN, BETWEEN
	EnumIds []int    // enum id(s): one, two or many ids depending on filter condition
}

FilterIdColumn define dimension or attribute column and condition to filter enum ids to build select where

type FilterOp 

type FilterOp string

FilterOp is enum type for filter operators in select where conditions 

const (
	InAutoOpFilter  FilterOp = "IN_AUTO" // auto convert IN list filter into equal or BETWEEN if possible
	InOpFilter      FilterOp = "IN"      // dimension enum ids in: dim2 IN (11, 22, 33)
	EqOpFilter      FilterOp = "="       // dimension equal: dim1 = 12
	NeOpFilter      FilterOp = "!="      // dimension equal: dim1 <> 12
	GtOpFilter      FilterOp = ">"       // value greater than: attr1 > 12
	GeOpFilter      FilterOp = ">="      // value greater or equal: attr1 >= 12
	LtOpFilter      FilterOp = "<"       // value less than: attr1 < 12
	LeOpFilter      FilterOp = "<="      // value less or equal: attr1 <= 12
	BetweenOpFilter FilterOp = "BETWEEN" // dimension enum ids between: dim3 BETWEEN 44 AND 88
)

Select filter operators for dimension enum values or attribute values.

type GroupLstRow 

type GroupLstRow struct {
	ModelId  int    // model_id     INT          NOT NULL
	GroupId  int    // group_id     INT          NOT NULL
	IsParam  bool   // is_parameter SMALLINT     NOT NULL, -- if <> 0 then parameter group else output table group
	Name     string // group_name   VARCHAR(255) NOT NULL
	IsHidden bool   // is_hidden    SMALLINT     NOT NULL
}

GroupLstRow is db row of group_lst table

type GroupMeta 

type GroupMeta struct {
	GroupLstRow              // parameters or output tables group rows: group_lst
	GroupPc     []GroupPcRow // group parent-child relationship rows: group_pc
}

GroupMeta is db rows to describe parent-child group of parameters or output tables, it is join of group_lst to group_pc

type GroupPcRow 

type GroupPcRow struct {
	ModelId      int // model_id       INT NOT NULL
	GroupId      int // group_id       INT NOT NULL
	ChildPos     int // child_pos      INT NOT NULL
	ChildGroupId int // child_group_id INT NULL, -- if not NULL then id of child group
	ChildLeafId  int // leaf_id        INT NULL, -- if not NULL then id of parameter or output table
}

GroupPcRow is db row of group_pc table

type GroupTxtRow 

type GroupTxtRow struct {
	ModelId   int // model_id  INT          NOT NULL
	GroupId   int // group_id  INT          NOT NULL
	DescrNote     // language, description, notes
}

GroupTxtRow is db row of group_txt table

type LangLstRow 

type LangLstRow struct {
	LangCode string // lang_code VARCHAR(32)  NOT NULL
	Name     string // lang_name VARCHAR(255) NOT NULL
	// contains filtered or unexported fields
}

LangLstRow is db row of lang_lst table.

langId is db-unique id of the language. LangCode is unique language code: EN, FR.

type LangMeta 

type LangMeta struct {
	Lang []langWord // language lang_lst row and lang_word rows in that language
	// contains filtered or unexported fields
}

LangMeta is languages and words for each language

func GetLanguages 

func GetLanguages(dbConn *sql.DB) (*LangMeta, error)

GetLanguages return language rows from lang_lst join to lang_word tables and map from lang_code to lang_id.

func (*LangMeta) Clone 

func (src *LangMeta) Clone() (*LangMeta, error)

Clone return deep copy of source language metadata

func (*LangMeta) CodeById 

func (langDef *LangMeta) CodeById(langId int) (string, bool)

CodeIdId return language code by language id or first language if id not found

func (*LangMeta) FromJson 

func (dst *LangMeta) FromJson(srcJson []byte) (bool, error)

FromJson restore language list from json string bytes

func (*LangMeta) IdByCode 

func (langDef *LangMeta) IdByCode(langCode string) (int, bool)

IdByCode return language id by language code or first language if code not found

type LangNote 

type LangNote struct {
	LangCode string // lang_code VARCHAR(32)  NOT NULL
	Note     string // note      VARCHAR(32000)
}

LangNote is a holder for language code and notes

type ModelDicDescrNote added in v1.17.6 

type ModelDicDescrNote struct {
	Model     ModelDicRow // model_dic db row
	DescrNote DescrNote   // from model_dic_txt
}

ModelDicDescrNote is join of model_dic db row and model_dic_txt row

type ModelDicRow 

type ModelDicRow struct {
	ModelId         int    // model_id         INT          NOT NULL
	Name            string // model_name       VARCHAR(255) NOT NULL
	Digest          string // model_digest     VARCHAR(32)  NOT NULL
	Type            int    // model_type       INT          NOT NULL
	Version         string // model_ver        VARCHAR(32)  NOT NULL
	CreateDateTime  string // create_dt        VARCHAR(32)  NOT NULL
	DefaultLangCode string // model default language code
}

ModelDicRow is db row of model_dic table.

ModelId (model_dic.model_id) is db-unique id of the model, use digest to find same model in other db.

func GetModelList 

func GetModelList(dbConn *sql.DB) ([]ModelDicRow, error)

GetModelList return list of the models: model_dic table rows.

func GetModelRow added in v1.6.3 

func GetModelRow(dbConn *sql.DB, modelId int) (*ModelDicRow, error)

GetModelRow return model_dic table row by model id.

type ModelMeta 

type ModelMeta struct {
	Model       ModelDicRow       // model_dic table row
	Type        []TypeMeta        // types metadata: type name and enums
	Param       []ParamMeta       // parameters metadata: parameter name, type, dimensions
	Table       []TableMeta       // output tables metadata: table name, dimensions, accumulators, expressions
	Entity      []EntityMeta      // model entities and attributes
	Group       []GroupMeta       // groups of parameters or output tables
	EntityGroup []EntityGroupMeta // entity groups of attributes
}

ModelMeta is model metadata db rows, language-neutral portion of it.

Types, parameters and output tables can be shared between different models and even between different databases. Use digest hash to find same type (parameter, table or model) in other database. As it is today language-specific part of model metadata (labels, description, notes, etc.) does not considered for "equality" comparison and not included in digest.

For example, logical type consists of 2 enum (code, value) pairs: [(0, "false") (1, "true")] and even it has different labels in different databases, i.e. (1, "Truth") vs (1, "OK") such type(s) considered the same and should have identical digest(s).

Inside of database *_hid (type_hid, parameter_hid, table_hid) is a unique id of corresponding object (primary key). Those _hid's are database-unique and should be used to find same type (parameter, output table) in other database. Also each type, parameter, output table have model-unique *_id (type_id, parameter_id, table_id) assigned by compiler and it is possible to find type, parameter or table by combination of (model_id, type_id).

Unless otherwise specified each array is ordered by model-specific id's and binary search can be used. For example type array is ordered by (model_id, type_id) and type enum array by (model_id, type_id, enum_id).

func GetModel 

func GetModel(dbConn *sql.DB, name, digest string) (*ModelMeta, error)

GetModel return model metadata: parameters and output tables definition.

Model selected by name and/or digest, i.e.: ("modelOne", "abcd201208171604590148") if digest is empty then first model with min(model_id) is used

func GetModelById 

func GetModelById(dbConn *sql.DB, modelId int) (*ModelMeta, error)

GetModelById return model metadata: parameters and output tables definition.

Model selected by model id, which expected to be positive.

func (*ModelMeta) Clone 

func (src *ModelMeta) Clone() (*ModelMeta, error)

Clone return deep copy of source model metadata

func (*ModelMeta) EntityByKey added in v1.13.0 

func (modelDef *ModelMeta) EntityByKey(entityId int) (int, bool)

EntityByKey return index of entity by key: entityId

func (*ModelMeta) EntityByName added in v1.13.0 

func (modelDef *ModelMeta) EntityByName(name string) (int, bool)

EntityByName return index of entity by name

func (*ModelMeta) EntityHidById added in v1.13.0 

func (modelDef *ModelMeta) EntityHidById(entityId int) int

EntityHidById return entity Hid by id or -1 if not found

func (*ModelMeta) FromJson 

func (dst *ModelMeta) FromJson(srcJson []byte) (bool, error)

FromJson restore model metadata list from json string bytes

func (*ModelMeta) OutTableByHid 

func (modelDef *ModelMeta) OutTableByHid(tableHid int) (int, bool)

OutTableByHid return index of output table by table Hid

func (*ModelMeta) OutTableByKey 

func (modelDef *ModelMeta) OutTableByKey(tableId int) (int, bool)

OutTableByKey return index of output table by key: tableId

func (*ModelMeta) OutTableByName 

func (modelDef *ModelMeta) OutTableByName(name string) (int, bool)

OutTableByName return index of output table by name

func (*ModelMeta) OutTableHidById 

func (modelDef *ModelMeta) OutTableHidById(tableId int) int

OutTableHidById return output table Hid by id or -1 if not found

func (*ModelMeta) ParamByHid 

func (modelDef *ModelMeta) ParamByHid(paramHid int) (int, bool)

ParamByHid return index of parameter by parameter Hid

func (*ModelMeta) ParamByKey 

func (modelDef *ModelMeta) ParamByKey(paramId int) (int, bool)

ParamByKey return index of parameter by key: paramId

func (*ModelMeta) ParamByName 

func (modelDef *ModelMeta) ParamByName(name string) (int, bool)

ParamByName return index of parameter by name

func (*ModelMeta) ParamHidById 

func (modelDef *ModelMeta) ParamHidById(paramId int) int

ParamHidById return parameter Hid by id or -1 if not found

func (*ModelMeta) TypeByKey 

func (modelDef *ModelMeta) TypeByKey(typeId int) (int, bool)

TypeByKey return index of type by key: typeId

func (*ModelMeta) TypeOfDouble added in v1.17.3 

func (modelDef *ModelMeta) TypeOfDouble() (int, bool)

return double type index, it sia type of output value table and calculated value

type ModelTxtMeta 

type ModelTxtMeta struct {
	ModelName      string              // model name for text metadata
	ModelDigest    string              // model digest for text metadata
	ModelTxt       []ModelTxtRow       // model text rows: model_dic_txt
	TypeTxt        []TypeTxtRow        // model type text rows: type_dic_txt join to model_type_dic
	TypeEnumTxt    []TypeEnumTxtRow    // type enum text rows: type_enum_txt join to model_type_dic
	ParamTxt       []ParamTxtRow       // model parameter text rows: parameter_dic_txt join to model_parameter_dic
	ParamDimsTxt   []ParamDimsTxtRow   // parameter dimension text rows: parameter_dims_txt join to model_parameter_dic
	TableTxt       []TableTxtRow       // model output table text rows: table_dic_txt join to model_table_dic
	TableDimsTxt   []TableDimsTxtRow   // output table dimension text rows: table_dims_txt join to model_table_dic
	TableAccTxt    []TableAccTxtRow    // output table accumulator text rows: table_acc_txt join to model_table_dic
	TableExprTxt   []TableExprTxtRow   // output table expression text rows: table_expr_txt join to model_table_dic
	EntityTxt      []EntityTxtRow      // model entities text rows: entity_dic_txt join to model_entity_dic table
	EntityAttrTxt  []EntityAttrTxtRow  // entity attributes: entity_attr_txt join to model_entity_dic table
	GroupTxt       []GroupTxtRow       // group text rows: group_txt
	EntityGroupTxt []EntityGroupTxtRow // entity group text rows: entity_group_txt
}

ModelTxtMeta is language-specific portion of model metadata db rows.

func GetModelText 

func GetModelText(dbConn *sql.DB, modelId int, langCode string, isPack bool) (*ModelTxtMeta, error)

GetModelText return model text metadata: description and notes. If langCode not empty then only specified language selected else all languages. If isPack is true then return empty text metadata for range types

func (*ModelTxtMeta) Clone added in v1.15.3 

func (src *ModelTxtMeta) Clone() (*ModelTxtMeta, error)

Clone return deep copy of source model text metadata

type ModelTxtRow 

type ModelTxtRow struct {
	ModelId  int    // model_id     INT          NOT NULL
	LangCode string // lang_code    VARCHAR(32)  NOT NULL
	Descr    string // descr        VARCHAR(255) NOT NULL
	Note     string // note         VARCHAR(32000)
}

ModelTxtRow is db row of model_dic_txt join to model_dic

func GetModelTextRowById added in v1.13.0 

func GetModelTextRowById(dbConn *sql.DB, modelId int, langCode string) ([]ModelTxtRow, error)

GetModelTextById return model_dic_txt table rows by model id.

If langCode not empty then only specified language selected else all languages.

type ModelWordMeta 

type ModelWordMeta struct {
	ModelName   string          // model name for text metadata
	ModelDigest string          // model digest for text metadata
	ModelWord   []modelLangWord // language and db rows of model_word in that language
}

ModelWordMeta is language-specific model_word db rows.

func GetModelWord 

func GetModelWord(dbConn *sql.DB, modelId int, langCode string) (*ModelWordMeta, error)

GetModelWord return model "words": language-specific stirngs. If langCode not empty then only specified language selected else all languages.

func (*ModelWordMeta) Clone added in v1.15.3 

func (src *ModelWordMeta) Clone() (*ModelWordMeta, error)

Clone return deep copy of source model words

type OrderByColumn 

type OrderByColumn struct {
	IndexOne int  // one-based column index
	IsDesc   bool // if true then descending order
}

OrderByColumn define column to order by rows selected from parameter or output table.

type ParamDicRow 

type ParamDicRow struct {
	ModelId      int    // model_id           INT          NOT NULL
	ParamId      int    // model_parameter_id INT          NOT NULL
	ParamHid     int    // parameter_hid      INT          NOT NULL, -- unique parameter id
	Name         string // parameter_name     VARCHAR(255) NOT NULL
	Digest       string // parameter_digest   VARCHAR(32)  NOT NULL
	Rank         int    // parameter_rank     INT          NOT NULL
	TypeId       int    // model_type_id      INT          NOT NULL
	IsExtendable bool   // is_extendable      SMALLINT     NOT NULL
	IsHidden     bool   // is_hidden          SMALLINT     NOT NULL
	NumCumulated int    // num_cumulated      INT          NOT NULL
	DbRunTable   string // db_run_table       VARCHAR(64)  NOT NULL
	DbSetTable   string // db_set_table       VARCHAR(64)  NOT NULL
	ImportDigest string // import_digest      VARCHAR(32)  NOT NULL
}

ParamDicRow is db row of parameter_dic join to model_parameter_dic table

ParamHid (parameter_dic.parameter_hid) is db-unique id of the parameter, use digest to find same parameter in other db. ParamId (model_parameter_dic.model_parameter_id) is model-unique parameter id, assigned by model compiler.

type ParamDimsRow 

type ParamDimsRow struct {
	ModelId int    // model_id           INT          NOT NULL
	ParamId int    // model_parameter_id INT          NOT NULL
	DimId   int    // dim_id             INT          NOT NULL
	Name    string // dim_name           VARCHAR(255) NOT NULL
	TypeId  int    // model_type_id      INT          NOT NULL
	// contains filtered or unexported fields
}

ParamDimsRow is db row of parameter_dims join to model_parameter_dic table

type ParamDimsTxtRow 

type ParamDimsTxtRow struct {
	ModelId  int    // model_id           INT          NOT NULL
	ParamId  int    // model_parameter_id INT          NOT NULL
	DimId    int    // dim_id             INT          NOT NULL
	LangCode string // lang_code          VARCHAR(32)  NOT NULL
	Descr    string // descr              VARCHAR(255) NOT NULL
	Note     string // note               VARCHAR(32000)
}

ParamDimsTxtRow is db row of parameter_dims_txt join to model_parameter_dic table

type ParamImportRow 

type ParamImportRow struct {
	ModelId     int    // model_id           INT          NOT NULL
	ParamId     int    // model_parameter_id INT          NOT NULL
	FromName    string // from_name          VARCHAR(255) NOT NULL
	FromModel   string // from_model_name    VARCHAR(255) NOT NULL
	IsSampleDim bool   // is_sample_dim      SMALLINT     NOT NULL
}

ParamImportRow is db row of model_parameter_import table

type ParamMeta 

type ParamMeta struct {
	ParamDicRow                  // model parameter row: parameter_dic join to model_parameter_dic table
	Dim         []ParamDimsRow   // parameter dimension rows: parameter_dims join to model_parameter_dic table
	Import      []ParamImportRow // parameter import from upstream model
	// contains filtered or unexported fields
}

ParamMeta is parameter metadata: parameter name, type, dimensions

func (*ParamMeta) DimByKey 

func (param *ParamMeta) DimByKey(dimId int) (int, bool)

DimByKey return index of parameter dimension by key: dimId

type ParamRunSetPub 

type ParamRunSetPub struct {
	ParamRunSetTxtPub
	SubCount     int    // number of parameter sub-values
	DefaultSubId int    // default sub-value id for that parameter workset
	ValueDigest  string // value digest, not empty only as result of select from run_parameter; input from "public" value digest is ignored
}

ParamRunSetPub is "public" run or workset parameter value metadata for json import-export

type ParamRunSetTxtPub added in v1.9.3 

type ParamRunSetTxtPub struct {
	Name string     // parameter name
	Txt  []LangNote // parameter value notes by language
}

ParamRunSetTxtPub is "public" run or workset parameter metadata for json import-export: name, description and notes

type ParamTxtRow 

type ParamTxtRow struct {
	ModelId  int    // model_id           INT          NOT NULL
	ParamId  int    // model_parameter_id INT          NOT NULL
	LangCode string // lang_code          VARCHAR(32)  NOT NULL
	Descr    string // descr              VARCHAR(255) NOT NULL
	Note     string // note               VARCHAR(32000)
}

ParamTxtRow is db row of parameter_dic_txt join to model_parameter_dic table

type ParamValuePub 

type ParamValuePub struct {
	ParamRunSetPub                 // parameter metadata
	Kind           string          // where to get parameter from: "value", "run", "set" or empty "" which is "value" by default
	From           string          // run digest (or name or stamp) or workset name to copy parameter from
	Value          []CellCodeParam // parameter value(s)
}

ParamValuePub is "public" run or workset parameter metadata and values for json import-export.

type ProfileMeta 

type ProfileMeta struct {
	Name string            // profile name
	Opts map[string]string // profile (key, value) options
}

ProfileMeta is rows from profile_option table.

Profile is a named group of (key, value) options, similar to ini-file. Default model options has profile_name = model_name.

func GetProfile 

func GetProfile(dbConn *sql.DB, name string) (*ProfileMeta, error)

GetProfile return profile_option table rows as (key, value) map.

Profile is a named group of (key, value) options, similar to ini-file. Default model options has profile_name = model_name.

type ReadCalculteMicroLayout added in v1.16.0 

type ReadCalculteMicroLayout struct {
	ReadLayout           // entity name, run id, page size, where filters and order by
	CalculateMicroLayout // microdata aggregations
}

ReadCalculteMicroLayout describe microdata generation read layout, aggregation measures and group by attributes.

type ReadCalculteTableLayout added in v1.15.4 

type ReadCalculteTableLayout struct {
	ReadLayout                         // output table name, run id, page size, where filters and order by
	Calculation []CalculateTableLayout // additional measures to calculate
}

ReadCalculteTableLayout describe table read layout and additional measures to calculte.

type ReadCompareMicroLayout added in v1.16.0 

type ReadCompareMicroLayout struct {
	ReadCalculteMicroLayout          // aggregation measures and group by attributes
	Runs                    []string // runs to compare: list of digest, stamp or name
}

ReadCompareMicroLayout to compare microdata runs with base run using multiple comparison aggregations and/or calculation aggregations.

Comparison aggregation must contain [base] and [variant] attribute(s), ex.: OM_AVG(Income[base] - Income[variant]). Calculation aggregation is attribute(s) aggregation expression, ex.: OM_MAX(Income) / OM_MIN(Salary).

type ReadCompareTableLayout added in v1.15.5 

type ReadCompareTableLayout struct {
	ReadCalculteTableLayout          // output table, base run and comparison expressions or calculations
	Runs                    []string // runs to compare: list of digest, stamp or name
}

ReadCompareTableLayout to compare output table runs with base run using multiple comparison expressions and/or calculation measures.

Comparison expression(s) must contain [base] and [variant] expression(s), ex.: Expr0[base] - Expr0[variant]. Calculation measure(s) can include table exprissions, ex.: Expr0 + Expr1 or aggregation of table accumulators, ex.: OM_SUM(acc0) / OM_COUNT(acc0)

type ReadLayout 

type ReadLayout struct {
	Name           string           // parameter name, output table name or entity microdata name
	FromId         int              // run id or set id to select input parameter, output table values or microdata from
	ReadPageLayout                  // read page first row offset, size and last page flag
	Filter         []FilterColumn   // dimension or attribute or value filters, final WHERE does join all filters by AND
	FilterById     []FilterIdColumn // dimension or attribute filters by enum ids, final WHERE does join filters by AND
	OrderBy        []OrderByColumn  // order by columnns, if empty then dimension id ascending order is used
}

ReadLayout describes source and size of data page to read input parameter, output table values or microdata.

Row filters combined by AND and allow to select dimension or attribute items, it can be enum codes or enum id's, ex.: dim0 = 'CA' AND dim1 IN (2010, 2011, 2012)

Order by applied to output columns. Because dimension or attribute columns always contain enum id's, therefore result ordered by id's and not by enum codes. Columns list depending on output table or parameter query:

parameter values: 

SELECT sub_id, dim0, dim1, param_value FROM parameterTable ORDER BY...

output table expressions: 

SELECT expr_id, dim0, dim1, expr_value FROM outputTable ORDER BY...

output table accumulators: 

SELECT acc_id, sub_id, dim0, dim1, acc_value FROM outputTable ORDER BY...

all-accumulators view: 

SELECT sub_id, dim0, dim1, acc0_value, acc1_value... FROM outputTable ORDER BY...

entity microdata table: 

SELECT entity_key, attr0, attr1,... FROM microdataTable ORDER BY...

type ReadMicroLayout added in v1.13.0 

type ReadMicroLayout struct {
	ReadLayout        // entity name, run id, page size, where filters and order by
	GenDigest  string // entity generation digest
}

ReadMicroLayout describes source and size of data page to read entity microdata.

Only one entity generation digest expected for each run id + entity name, but there is no such constarint in db schema.

type ReadPageLayout 

type ReadPageLayout struct {
	Offset     int64 // first row to return from select, zero-based ofsset
	Size       int64 // max row count to select, if <= 0 then all rows
	IsLastPage bool  // output last page flag: return true if it was a last page of rows
	IsFullPage bool  // input last page flag: if true then adjust offset to return full last page
}

ReadPageLayout describes first row offset and size of data page to read input parameter or output table values.

func CalculateMicrodata added in v1.16.0 

func CalculateMicrodata(dbConn *sql.DB, modelDef *ModelMeta, microLt *ReadCalculteMicroLayout, runIds []int) (*list.List, *ReadPageLayout, error)

CalculateMicrodata aggregates microdata using group by attributes as dimensions and calculate aggregated measure(s).

func CalculateOutputTable added in v1.15.4 

func CalculateOutputTable(dbConn *sql.DB, modelDef *ModelMeta, tableLt *ReadCalculteTableLayout, runIds []int) (*list.List, *ReadPageLayout, error)

CalculateOutputTable read output table page (dimensions and values) and calculate extra measure(s).

If calcLt.IsAggr true then do accumulator(s) aggregation else calculate expression value(s), ex: Expr1[variant] - Expr1[base].

func ReadMicrodataCalculateTo added in v1.16.0 

func ReadMicrodataCalculateTo(
	dbConn *sql.DB, modelDef *ModelMeta, layout *ReadMicroLayout, calcLt *CalculateMicroLayout, runIds []int, cvtTo func(src interface{}) (bool, error),
) (*ReadPageLayout, error)

ReadMicrodataCalculateTo read aggregated microdata rows (run_id, calc_id, group by attributes, calc_value) and process each row by cvtTo().

It can calculate multiple measure values using simple aggregation of attributes, ex.: OM_AVG(Income) or on aggregate for run comparison, ex.: OM_AVG(Income[variant] - Income[base]). Optional list of run id's can be supplied to read more than one run from microdata table.

func ReadMicrodataTo added in v1.13.0 

func ReadMicrodataTo(dbConn *sql.DB, modelDef *ModelMeta, layout *ReadMicroLayout, cvtTo func(src interface{}) (bool, error)) (*ReadPageLayout, error)

ReadMicrodataTo read entity microdata rows (microdata key, attributes) from model run results and process each row by cvtTo().

func ReadOutputTableCalculteTo added in v1.15.4 

func ReadOutputTableCalculteTo(
	dbConn *sql.DB, modelDef *ModelMeta, layout *ReadTableLayout, calcLt []CalculateTableLayout, runIds []int, cvtTo func(src interface{}) (bool, error),
) (*ReadPageLayout, error)

ReadOutputTableCalculteTo read calculated output table page (run id, expression(s) id, dimensions and values) and process each row by cvtTo().

It can calculate multiple values based on expressions and/or accumulators aggregation. Optional runIds list of run id's can be supplied to read more than one run from output table.

func ReadOutputTableTo added in v1.9.7 

func ReadOutputTableTo(dbConn *sql.DB, modelDef *ModelMeta, layout *ReadTableLayout, cvtTo func(src interface{}) (bool, error)) (*ReadPageLayout, error)

ReadOutputTableTo read output table page (dimensions and values) from model run results and process each row by cvtTo().

If layout.IsAccum true then select accumulator(s) else output expression value(s) If layout.ValueName not empty then select only that expression (accumulator) else all expressions (accumulators)

func ReadParameterTo added in v1.9.7 

func ReadParameterTo(dbConn *sql.DB, modelDef *ModelMeta, layout *ReadParamLayout, cvtTo func(src interface{}) (bool, error)) (*ReadPageLayout, error)

ReadParameterTo read input parameter rows (sub id, dimensions, value) from workset or model run results and process each row by cvtTo().

func SelectToList 

func SelectToList(
	dbConn *sql.DB, query string, layout ReadPageLayout, cvt func(rows *sql.Rows) (interface{}, error)) (*list.List, *ReadPageLayout, error)

SelectToList select db rows into list using cvt to convert (scan) each db row into struct.

It selects "page size" number of rows starting from row number == offset (zero based). If page size <= 0 then all rows returned. If IsFullPage is true then adjust offset to return full last page

type ReadParamLayout 

type ReadParamLayout struct {
	ReadLayout           // parameter name, run id or set id page size, where filters and order by
	IsFromSet       bool // if true then select from workset else from model run
	IsEditSet       bool // if true then workset must be editable (readonly = false)
	ReadSubIdLayout      // sub-value id filter: select rows with only one sub-value id
}

ReadParamLayout describes source and size of data page to read input parameter values.

It can read parameter values from model run results or from input working set (workset). If this is read from workset then it can be read-only or read-write (editable) workset.

type ReadSubIdLayout added in v1.15.3 

type ReadSubIdLayout struct {
	IsSubId bool // if true then select only single sub-value id
	SubId   int  // sub-value id to select rows from output table or parameter
}

ReadSubIdLayout supply sub-value id filter to select rows with only single sub_id from output table or input parameter values.

type ReadTableLayout 

type ReadTableLayout struct {
	ReadLayout             // output table name, run id, page size, where filters and order by
	ValueName       string // if not empty then expression or accumulator name to select
	IsAccum         bool   // if true then select output table accumulator else expression
	IsAllAccum      bool   // if true then select from all accumulators view else from accumulators table
	ReadSubIdLayout        // sub-value id filter: select rows with only one sub-value id
}

ReadTableLayout describes source and size of data page to read output table values.

If ValueName is not empty then only accumulator or output expression with that name selected (i.e: "acc1" or "expr4") else all output table accumulators (expressions) selected.

type RunEntityRow added in v1.14.0 

type RunEntityRow struct {
	RunId       int    // run_id         INT         NOT NULL
	GenHid      int    // entity_gen_hid INT NOT NULL
	RowCount    int    // row_count    INT NOT NULL, -- if not zero then entity microdata row count
	ValueDigest string // value_digest VARCHAR(32),  -- if not NULL then digest of table value for the run
}

RunEntityRow is db row of run_entity join to entity_gen table

func GetRunEntityGenByModel added in v1.16.0 

func GetRunEntityGenByModel(dbConn *sql.DB, modelId int) ([]RunEntityRow, error)

GetRunEntityGenList return run entity rows by model id: run_entity table rows.

type RunMeta 

type RunMeta struct {
	Run       RunRow            // model run master row: run_lst
	Txt       []RunTxtRow       // run text rows: run_txt
	Opts      map[string]string // options used to run the model: run_option
	Param     []runParam        // run parameters: parameter_hid, sub-value count, run_parameter_txt table rows
	Table     []runTable        // run tables: table_hid fom run_table rows
	EntityGen []EntityGenMeta   // run entity generation: entity_gen, entity_gen_attr table rows
	RunEntity []RunEntityRow    // run microdata entities: run_entity table rows
	Progress  []RunProgress     // run progress by sub-values: run_progress table rows
}

RunMeta is metadata of model run: name, status, run options, description, notes.

func GetRunFull 

func GetRunFull(dbConn *sql.DB, runRow *RunRow) (*RunMeta, error)

GetRunFull return full metadata for model run: run_lst, run_option, run_parameter, run_table, run_progress, and entity generation rows.

func GetRunFullText 

func GetRunFullText(dbConn *sql.DB, runRow *RunRow, isSuccess bool, langCode string) (*RunMeta, error)

GetRunFullText return full metadata, including text, for completed model run: run_lst, run_txt, run_option, run_parameter, run_parameter_txt, run_table, run_progress, and entity generation rows.

If isSuccess true then return only successfully completed runs else return all runs: success, error, exit, progress. If langCode not empty then only specified language selected else all languages

func GetRunFullTextList 

func GetRunFullTextList(dbConn *sql.DB, modelId int, isSuccess bool, langCode string) ([]RunMeta, error)

GetRunFullTextList return list of full metadata, including text, for model runs: run_lst, run_txt, run_option, run_parameter, run_parameter_txt, run_table, run_progress and entity generation rows.

If isSuccess true then return only successfully completed runs else return all runs: success, error, exit, progress. If langCode not empty then only specified language selected else all languages

func (*RunMeta) EntityGenByDigest added in v1.13.0 

func (run *RunMeta) EntityGenByDigest(digest string) (int, bool)

EntityGenByDigest return index of entity generation by generation digest

func (*RunMeta) EntityGenByEntityId added in v1.13.0 

func (run *RunMeta) EntityGenByEntityId(entityId int) (int, bool)

EntityGenByEntityId return index of entity generation by model entity id. As it is today model do not insert more than one generation for each entity into model run, but there is no such costraint in db schema.

func (*RunMeta) ToPublic 

func (meta *RunMeta) ToPublic(dbConn *sql.DB, modelDef *ModelMeta) (*RunPub, error)

ToPublic convert model run db rows into "public" model run format for json import-export.

func (*RunMeta) UpdateRun 

func (meta *RunMeta) UpdateRun(dbConn *sql.DB, modelDef *ModelMeta, langDef *LangMeta, doubleFmt string) (bool, error)

UpdateRun insert new or return existing model run metadata in database.

Run status must be completed (success, exit or error) otherwise error returned. If this run already exist then nothing is updated in database, only metadata updated with actual run id. Run digest is used to find existing model run, it cannot be empty "" string otherwise error returned.

Double format is used for progress value float conversion, if non-empty format supplied.

It return "is found" flag and update metadata with actual run id in database.

func (*RunMeta) UpdateRunText 

func (meta *RunMeta) UpdateRunText(dbConn *sql.DB, modelDef *ModelMeta, runId int, langDef *LangMeta) error

UpdateRunText merge run text (run description and notes) and run parameter notes into run_txt and run_parameter_txt tables.

New run text merged with existing db rows by update or insert if rows not exist db tables. If run not exist then function does nothing. Run status must be completed (success, exit or error) otherwise error returned. Run id of the input run metadata updated with runId value.

type RunParamTxtRow 

type RunParamTxtRow struct {
	RunId    int    // run_id        INT         NOT NULL
	ParamHid int    // parameter_hid INT         NOT NULL
	LangCode string // lang_code     VARCHAR(32) NOT NULL
	Note     string // note          VARCHAR(32000)
}

RunParamTxtRow is db row of run_parameter_txt

func GetRunAllParamText 

func GetRunAllParamText(dbConn *sql.DB, runId int, langCode string) ([]RunParamTxtRow, error)

GetRunAllParamText return all run parameters value notes: run_parameter_txt table rows.

If langCode not empty then only specified language selected else all languages

func GetRunParamText 

func GetRunParamText(dbConn *sql.DB, runId int, paramHid int, langCode string) ([]RunParamTxtRow, error)

GetRunParamText return run parameter value notes: run_parameter_txt table rows.

If langCode not empty then only specified language selected else all languages

type RunProgress 

type RunProgress struct {
	SubId          int     // sub_id         INT         NOT NULL, -- sub-value id (zero based index)
	CreateDateTime string  // create_dt      VARCHAR(32) NOT NULL, -- start date-time
	Status         string  // status         VARCHAR(1)  NOT NULL, -- run status: i=init p=progress s=success x=exit e=error(failed)
	UpdateDateTime string  // update_dt      VARCHAR(32) NOT NULL, -- last update date-time
	Count          int     // progress_count INT         NOT NULL, -- progress count: percent completed
	Value          float64 // progress_value FLOAT       NOT NULL, -- progress value: number of cases (case based) or time (time based)
}

RunProgress is a "public" sub-value run_progress db row

func GetRunProgress 

func GetRunProgress(dbConn *sql.DB, runId int) ([]RunProgress, error)

GetRunProgress return sub-values run progress for specified run id: run_progress table rows.

type RunPub 

type RunPub struct {
	ModelName           string            // model name for that run
	ModelDigest         string            // model digest for that run
	ModelVersion        string            // model_ver     VARCHAR(32)  NOT NULL
	ModelCreateDateTime string            // create_dt     VARCHAR(32)  NOT NULL
	Name                string            // run_name      VARCHAR(255) NOT NULL
	SubCount            int               // sub_count     INT          NOT NULL, -- subvalue count
	SubStarted          int               // sub_started   INT          NOT NULL, -- number of subvalues started
	SubCompleted        int               // sub_completed INT          NOT NULL, -- number of subvalues completed
	CreateDateTime      string            // create_dt     VARCHAR(32)  NOT NULL, -- start date-time
	Status              string            // status        VARCHAR(1)   NOT NULL, -- run status: i=init p=progress s=success x=exit e=error(failed)
	UpdateDateTime      string            // update_dt     VARCHAR(32)  NOT NULL, -- last update date-time
	RunId               int               // used only as output value, input value ignored
	RunDigest           string            // run_digest    VARCHAR(32)  NULL,     -- digest of the run metadata: model digest, run name, sub count, created date-time, run stamp
	ValueDigest         string            // value_digest  VARCHAR(32),           -- if not NULL then digest of the run values: all parameters and output tables
	RunStamp            string            // run_stamp     VARCHAR(32)  NOT NULL, -- process run stamp, by default is log time stamp
	Txt                 []DescrNote       // run text: description and notes by language
	Opts                map[string]string // options used to run the model: run_option
	Param               []ParamRunSetPub  // run parameters: name, sub-value count and value notes by language
	Table               []TableRunPub     // run tables: name for tables included in run_table
	Entity              []EntityRunPub    // run entities: entity generation and attributes
	Progress            []RunProgress     // run progress by sub-values: run_progress table rows
}

RunPub is "public" model run metadata for json import-export

func (*RunPub) FromPublic 

func (pub *RunPub) FromPublic(dbConn *sql.DB, modelDef *ModelMeta) (*RunMeta, error)

FromPublic convert model run metadata from "public" format (coming from json import-export) into db rows.

type RunRow 

type RunRow struct {
	RunId          int    // run_id        INT          NOT NULL, -- unique run id
	ModelId        int    // model_id      INT          NOT NULL
	Name           string // run_name      VARCHAR(255) NOT NULL, -- model run name
	SubCount       int    // sub_count     INT          NOT NULL, -- subvalue count
	SubStarted     int    // sub_started   INT          NOT NULL, -- number of subvalues started
	SubCompleted   int    // sub_completed INT          NOT NULL, -- number of subvalues completed
	CreateDateTime string // create_dt     VARCHAR(32)  NOT NULL, -- start date-time
	Status         string // status        VARCHAR(1)   NOT NULL, -- run status: i=init p=progress s=success x=exit e=error(failed)
	UpdateDateTime string // update_dt     VARCHAR(32)  NOT NULL, -- last update date-time
	RunDigest      string // run_digest    VARCHAR(32)  NULL,     -- digest of the run metadata: model digest, run name, sub count, created date-time, run stamp
	ValueDigest    string // value_digest  VARCHAR(32),           -- if not NULL then digest of the run values: all parameters and output tables
	RunStamp       string // run_stamp     VARCHAR(32)  NOT NULL, -- process run stamp, by default is log time stamp
}

RunRow is model run row: run_lst table row.

Run status: i=init p=progress s=success x=exit e=error(failed). Run id must be different from working set id (use id_lst to get it)

func GetFirstRun 

func GetFirstRun(dbConn *sql.DB, modelId int) (*RunRow, error)

GetFirstRun return first run of the model: run_lst table row.

func GetLastCompletedRun 

func GetLastCompletedRun(dbConn *sql.DB, modelId int) (*RunRow, error)

GetLastCompletedRun return last completed run of the model: run_lst table row.

Run completed if run status one of: s=success, x=exit, e=error

func GetLastRun 

func GetLastRun(dbConn *sql.DB, modelId int) (*RunRow, error)

GetLastRun return last run of the model: run_lst table row.

func GetLastRunByName added in v1.6.3 

func GetLastRunByName(dbConn *sql.DB, modelId int, name string) (*RunRow, error)

GetLastRunByName return model run row by run name: run_lst table row.

If there is multiple runs with this name then run with max(run_id) returned

func GetRun 

func GetRun(dbConn *sql.DB, runId int) (*RunRow, error)

GetRun return model run row by id: run_lst table row.

func GetRunByDigest 

func GetRunByDigest(dbConn *sql.DB, digest string) (*RunRow, error)

GetRunByDigest return model run row by digest: run_lst table row.

func GetRunByDigestStampName added in v1.16.0 

func GetRunByDigestStampName(dbConn *sql.DB, modelId int, rdsn string) (*RunRow, error)

GetRunByDigestStampName return model run row by run digest or run stamp or run name: run_lst table row.

It does select run row by digest, if not found then by model id and stamp, if not found by model id and run name. If there is multiple runs with this stamp or name then run with min(run_id) returned

func GetRunByName 

func GetRunByName(dbConn *sql.DB, modelId int, name string) (*RunRow, error)

GetRunByName return model run row by run name: run_lst table row.

If there is multiple runs with this name then run with min(run_id) returned

func GetRunByStamp 

func GetRunByStamp(dbConn *sql.DB, modelId int, stamp string) (*RunRow, error)

GetRunByStamp return model run row by run stamp: run_lst table row.

If there is multiple runs with this stamp then run with min(run_id) returned

func GetRunList 

func GetRunList(dbConn *sql.DB, modelId int) ([]RunRow, error)

GetRunList return list of model runs by model_id: run_lst rows.

func GetRunListByDigestStampName added in v1.16.0 

func GetRunListByDigestStampName(dbConn *sql.DB, modelId int, rdsn string) ([]RunRow, error)

GetRunListByDigestStampName return list of model run rows by run digest or run stamp or run name: run_lst table rows.

It does select run row by digest, if not found then by model id and stamp, if not found by model id and run name. If there is multiple runs with this stamp or name then multiple rows returned

type RunTxtRow 

type RunTxtRow struct {
	RunId    int    // run_id    INT          NOT NULL
	LangCode string // lang_code VARCHAR(32)  NOT NULL
	Descr    string // descr     VARCHAR(255) NOT NULL
	Note     string // note      VARCHAR(32000)
}

RunTxtRow is db row of run_txt

func GetRunText 

func GetRunText(dbConn *sql.DB, runId int, langCode string) ([]RunTxtRow, error)

GetRunText return model run description and notes: run_txt table rows.

If langCode not empty then only specified language selected else all languages

type TableAccRow 

type TableAccRow struct {
	ModelId   int    // model_id       INT          NOT NULL
	TableId   int    // model_table_id INT           NOT NULL
	AccId     int    // acc_id         INT           NOT NULL
	Name      string // acc_name       VARCHAR(255)  NOT NULL
	IsDerived bool   // is_derived     SMALLINT      NOT NULL
	SrcAcc    string // acc_src        VARCHAR(255)  NOT NULL
	AccSql    string // acc_sql        VARCHAR(2048) NOT NULL
	// contains filtered or unexported fields
}

TableAccRow is db row of table_acc join to model_table_dic table

type TableAccTxtRow 

type TableAccTxtRow struct {
	ModelId  int    // model_id       INT          NOT NULL
	TableId  int    // model_table_id INT          NOT NULL
	AccId    int    // acc_id         INT          NOT NULL
	LangCode string // lang_code      VARCHAR(32)  NOT NULL
	Descr    string // descr          VARCHAR(255) NOT NULL
	Note     string // note           VARCHAR(32000)
}

TableAccTxtRow is db row of table_acc_txt join to model_table_dic table

type TableDicRow 

type TableDicRow struct {
	ModelId      int    // model_id        INT          NOT NULL
	TableId      int    // model_table_id  INT          NOT NULL
	TableHid     int    // table_hid       INT          NOT NULL, -- unique table id
	Name         string // table_name      VARCHAR(255) NOT NULL
	Digest       string // table_digest    VARCHAR(32)  NOT NULL
	IsUser       bool   // is_user         SMALLINT     NOT NULL
	Rank         int    // table_rank      INT          NOT NULL
	IsSparse     bool   // is_sparse       SMALLINT     NOT NULL
	DbExprTable  string // db_expr_table   VARCHAR(64)  NOT NULL
	DbAccTable   string // db_acc_table    VARCHAR(64)  NOT NULL
	DbAccAllView string // db_acc_all_view VARCHAR(64)  NOT NULL
	ExprPos      int    // expr_dim_pos    INT          NOT NULL
	IsHidden     bool   // is_hidden       SMALLINT     NOT NULL
	ImportDigest string // import_digest   VARCHAR(32)  NOT NULL
}

TableDicRow is db row of table_dic join to model_table_dic table.

TableHid (table_dic.table_hid) is db-unique id of the output table, use digest to find same table in other db. TableId (model_table_dic.model_table_id) is model-unique output table id, assigned by model compiler.

type TableDimsRow 

type TableDimsRow struct {
	ModelId int    // model_id       INT          NOT NULL
	TableId int    // model_table_id INT          NOT NULL
	DimId   int    // dim_id         INT          NOT NULL
	Name    string // dim_name       VARCHAR(255) NOT NULL
	TypeId  int    // model_type_id  INT          NOT NULL
	IsTotal bool   // is_total       SMALLINT     NOT NULL
	DimSize int    // dim_size       INT          NOT NULL
	// contains filtered or unexported fields
}

TableDimsRow is db row of table_dims join to model_table_dic table

type TableDimsTxtRow 

type TableDimsTxtRow struct {
	ModelId  int    // model_id       INT          NOT NULL
	TableId  int    // model_table_id INT          NOT NULL
	DimId    int    // dim_id         INT          NOT NULL
	LangCode string // lang_code      VARCHAR(32)  NOT NULL
	Descr    string // descr          VARCHAR(255) NOT NULL
	Note     string // note           VARCHAR(32000)
}

TableDimsTxtRow is db row of table_dims_txt join to model_table_dic table

type TableExprRow 

type TableExprRow struct {
	ModelId  int    // model_id       INT           NOT NULL
	TableId  int    // model_table_id INT           NOT NULL
	ExprId   int    // expr_id        INT           NOT NULL
	Name     string // expr_name      VARCHAR(255)  NOT NULL
	Decimals int    // expr_decimals  INT           NOT NULL
	SrcExpr  string // expr_src       VARCHAR(255)  NOT NULL
	ExprSql  string // expr_sql       VARCHAR(2048) NOT NULL
	// contains filtered or unexported fields
}

TableExprRow is db row of table_expr join to model_table_dic table

type TableExprTxtRow 

type TableExprTxtRow struct {
	ModelId  int    // model_id       INT          NOT NULL
	TableId  int    // model_table_id INT          NOT NULL
	ExprId   int    // expr_id        INT           NOT NULL
	LangCode string // lang_code      VARCHAR(32)  NOT NULL
	Descr    string // descr          VARCHAR(255) NOT NULL
	Note     string // note           VARCHAR(32000)
}

TableExprTxtRow is db row of table_expr_txt join to model_table_dic table

type TableMeta 

type TableMeta struct {
	TableDicRow                // model output table row: table_dic join to model_table_dic
	Dim         []TableDimsRow // output table dimension rows: table_dims join to model_table_dic
	Acc         []TableAccRow  // output table accumulator rows: table_acc join to model_table_dic
	Expr        []TableExprRow // output table expression rows: table_expr join to model_table_dic
	// contains filtered or unexported fields
}

TableMeta is output table metadata: table name, dimensions, accumulators, expressions

func (*TableMeta) DimByKey 

func (table *TableMeta) DimByKey(dimId int) (int, bool)

DimByKey return index of output table dimension by key: dimId

type TableRunPub added in v1.8.4 

type TableRunPub struct {
	Name        string // parameter name
	ValueDigest string // value digest, not empty only as result of select from table_parameter; input from "public" value digest is ignored
}

TableRunPub is "public" metadata for output tables included in model run results

type TableTxtRow 

type TableTxtRow struct {
	ModelId   int    // model_id       INT          NOT NULL
	TableId   int    // model_table_id INT          NOT NULL
	LangCode  string // lang_code      VARCHAR(32)  NOT NULL
	Descr     string // descr          VARCHAR(255) NOT NULL
	Note      string // note           VARCHAR(32000)
	ExprDescr string // expr_descr     VARCHAR(255) NOT NULL
	ExprNote  string // expr_note      VARCHAR(32000)
}

TableTxtRow is db row of table_dic_txt join to model_table_dic table

type TaskDef 

type TaskDef struct {
	Task TaskRow      // modeling task row: task_lst
	Txt  []TaskTxtRow // task text rows: task_txt
	Set  []int        // task body (current list of workset id's): task_set
}

TaskDef is modeling task definition: metadata and input worksets

type TaskDefPub 

type TaskDefPub struct {
	ModelName           string      // model name for that task list
	ModelDigest         string      // model digest for that task list
	ModelVersion        string      // model_ver     VARCHAR(32)  NOT NULL
	ModelCreateDateTime string      // create_dt     VARCHAR(32)  NOT NULL
	Name                string      // task_name    VARCHAR(255) NOT NULL
	Txt                 []DescrNote // task text: description and notes by language
	Set                 []string    // task body: list of workset names
}

TaskDefPub is "public" modeling task metadata and task input worksets for json import-export

type TaskMeta 

type TaskMeta struct {
	TaskDef               // task definition: metadata and input worksets
	TaskRun []taskRunItem // task run history: task_run_lst and task_run_set rows
}

TaskMeta is metadata for modeling task: name, status, description, notes, task run history.

Modeling task is a named set of input model inputs (of workset ids) to run the model. Typical use case: create multiple input sets by varying some model parameters, combine it under named "task" and run the model with that task name. As result multiple model "runs" created ("run" is input and output data of model run). Such run of model called "task run" and allow to study dependencies between model input and output.

Task can be edited by user: new input workset ids added or some workset id(s) excluded. As result current task body (workset ids of the task) may be different from older version of it: task_set set_id's may not be same as task_run_set set_id's. TaskRun and TaskRunSet is a history and result of that task run, but there is no guarantee of any workset in task history still exist or contain same input parameter values as it was at the time of task run. To find actual input for any particular model run and/or task run we must use run_id.

func GetTaskFull 

func GetTaskFull(dbConn *sql.DB, taskRow *TaskRow, isSuccess bool, langCode string) (*TaskMeta, error)

GetTaskFull return modeling task metadata, description, notes and run history from db-tables: task_lst, task_txt, task_set, task_run_lst, task_run_set.

If isSuccess true then return only successfully completed runs else return all runs: success, error, exit, progress. If langCode not empty then only specified language selected else all languages

func GetTaskFullList 

func GetTaskFullList(dbConn *sql.DB, modelId int, isSuccess bool, langCode string) ([]TaskMeta, error)

GetTaskFullList return list of modeling tasks metadata, description, notes and run history from db-tables: task_lst, task_txt, task_set, task_run_lst, task_run_set.

If isSuccess true then return only successfully completed runs else return all runs: success, error, exit, progress. If langCode not empty then only specified language selected else all languages

func GetTaskRunList 

func GetTaskRunList(dbConn *sql.DB, taskRow *TaskRow) (*TaskMeta, error)

GetTaskRunList return model run history: master row from task_lst and all rows from task_run_lst, task_run_set.

func (*TaskMeta) MergeTaskDef 

func (meta *TaskMeta) MergeTaskDef(dbConn *sql.DB, modelDef *ModelMeta, langDef *LangMeta) error

MergeTaskDef update existing and insert new modeling task definition: task metadata and task input worksets.

It does replace: task_txt and task_set db rows. It does not change anything in task run history: task_run_lst and task_run_set tables. If task not exist then new task created. Task input worksets (new content of task_set table) must exist in workset_lst.set_id. Model id and task id updated with actual database id's.

func (*TaskMeta) ReplaceTaskDef 

func (meta *TaskMeta) ReplaceTaskDef(dbConn *sql.DB, modelDef *ModelMeta, langDef *LangMeta) error

ReplaceTaskDef delete existing and insert new modeling task definition: task metadata and task input worksets.

It does replace: task_txt and task_set db rows. It does not change anything in task run history: task_run_lst and task_run_set tables. If task not exist then new task created. Task input worksets (new content of task_set table) must exist in workset_lst.set_id. Model id and task id updated with actual database id's.

func (*TaskMeta) ToPublic 

func (meta *TaskMeta) ToPublic(dbConn *sql.DB, modelDef *ModelMeta) (*TaskPub, error)

ToPublic convert modeling task metadata db rows into "public" modeling task format for json import-export.

func (*TaskMeta) UpdateTaskFull 

func (meta *TaskMeta) UpdateTaskFull(dbConn *sql.DB, modelDef *ModelMeta, langDef *LangMeta) error

UpdateTaskFull delete existing and insert new modeling task and task run history in database.

Model id, task id, run id, set id updated with actual database id's. Task input worksets (new content of task_set table) must exist in workset_lst table.

type TaskPub 

type TaskPub struct {
	TaskDefPub              // task definition: metadata and input worksets
	TaskRun    []taskRunPub // task run history: task_run_lst
}

TaskPub is "public" modeling task metadata, task input worksets and task run history for json import-export

func (*TaskPub) FromPublic 

func (pub *TaskPub) FromPublic(dbConn *sql.DB, modelDef *ModelMeta, isCompleted bool) (*TaskMeta, bool, bool, error)

FromPublic convert modeling task metadata from "public" format (coming from json import-export) into db rows.

It return task metadata and two boolean flags: (1) isSetNotFound = true if some of task workset names not found in current database (2) isTaskRunNotFound = true if some of task run (pairs of set, model run) set or model run not found in current database.

Worksets are searched by set name, which is unique inside of the model. Model run searched by run digest. If isCompleted true then return only completed runs where status is success, error or exit else return all runs: success, error, exit, progress.

type TaskRow 

type TaskRow struct {
	TaskId  int    // task_id      INT          NOT NULL, -- unique task id
	ModelId int    // model_id     INT          NOT NULL
	Name    string // task_name    VARCHAR(255) NOT NULL
}

TaskRow is db row of task_lst.

func GetTask 

func GetTask(dbConn *sql.DB, taskId int) (*TaskRow, error)

GetTask return modeling task rows by id: task_lst table row and set ids from task_set

func GetTaskByName 

func GetTaskByName(dbConn *sql.DB, modelId int, name string) (*TaskRow, error)

GetTaskByName return modeling task rows by id: task_lst table row and set ids from task_set

func GetTaskList 

func GetTaskList(dbConn *sql.DB, modelId int) ([]TaskRow, error)

GetTaskList return list of model tasks: task_lst rows.

type TaskRunRow 

type TaskRunRow struct {
	TaskRunId      int    // task_run_id INT          NOT NULL, -- unique task run id
	TaskId         int    // task_id     INT          NOT NULL
	Name           string // run_name    VARCHAR(255) NOT NULL, -- task run name
	SubCount       int    // sub_count   INT          NOT NULL, -- subvalue count of task run
	CreateDateTime string // create_dt   VARCHAR(32)  NOT NULL, -- start date-time
	Status         string // status      VARCHAR(1)   NOT NULL, -- task status: i=init p=progress w=wait s=success x=exit e=error(failed)
	UpdateDateTime string // update_dt   VARCHAR(32)  NOT NULL, -- last update date-time
	RunStamp       string // run_stamp   VARCHAR(32)  NOT NULL, -- process run stamp, by default is log time stamp
}

TaskRunRow is db row of task_run_lst. This table contains task run history and status.

Task status: i=init p=progress w=wait s=success x=exit e=error(failed) 

if task status = w (wait) then
   model wait and NOT completed until other process set status to one of finals: s,x,e
   model check if any new sets inserted into task_set and run it as they arrive

func GetTaskFirstRun 

func GetTaskFirstRun(dbConn *sql.DB, taskId int) (*TaskRunRow, error)

GetTaskFirstRun return first run of the modeling task: task_run_lst table row.

func GetTaskLastCompletedRun 

func GetTaskLastCompletedRun(dbConn *sql.DB, taskId int) (*TaskRunRow, error)

GetTaskLastCompletedRun return last completed run of the modeling task: task_run_lst table row.

Task run completed if run status one of: s=success, x=exit, e=error

func GetTaskLastRun 

func GetTaskLastRun(dbConn *sql.DB, taskId int) (*TaskRunRow, error)

GetTaskLastRun return last run of the modeling task: task_run_lst table row.

func GetTaskRun 

func GetTaskRun(dbConn *sql.DB, taskRunId int) (*TaskRunRow, error)

GetTaskRun return modeling task run status: task_run_lst table row.

func GetTaskRunByName 

func GetTaskRunByName(dbConn *sql.DB, taskId int, name string) (*TaskRunRow, error)

GetTaskRunByName return modeling task run status by task id and task run name: task_run_lst table row. If there is multiple task runs with this name then run with min(task_run_id) returned

func GetTaskRunByStamp 

func GetTaskRunByStamp(dbConn *sql.DB, taskId int, stamp string) (*TaskRunRow, error)

GetTaskRunByStamp return modeling task run status by task id and task run stamp: task_run_lst table row. If there is multiple task runs with this stamp then run with min(task_run_id) returned

func GetTaskRunByStampOrName 

func GetTaskRunByStampOrName(dbConn *sql.DB, taskId int, trsn string) (*TaskRunRow, error)

GetTaskRunByStampOrName return modeling task run status by task id and task run stamp or task run name: task_run_lst table row.

It does select task run row by task id and stamp, if not found then by task id and run name. If there is multiple task runs with this stamp or name then run with min(task_run_id) returned.

func GetTaskRunListByStampOrName 

func GetTaskRunListByStampOrName(dbConn *sql.DB, taskId int, trsn string) ([]TaskRunRow, error)

GetTaskRunListByStampOrName return modeling task run rows from task_run_lst table by task id and task run stamp or task run name.

It does select task run rows by task id and stamp, if not found then by task id and run name. If there is multiple task runs with this stamp or name then multiple rows returned.

type TaskRunSetRow 

type TaskRunSetRow struct {
	TaskRunId int // task_run_id INT NOT NULL
	RunId     int // run_id      INT NOT NULL, -- if > 0 then result run id
	SetId     int // set_id      INT NOT NULL, -- if > 0 then input working set id
	TaskId    int // task_id     INT NOT NULL
}

TaskRunSetRow is db row of task_run_set. This table contains task run input (working set id) and output (model run id)

func GetTaskRunSetRows added in v1.8.4 

func GetTaskRunSetRows(dbConn *sql.DB, taskRunId int) ([]TaskRunSetRow, error)

GetTaskRunSetRows return task run body (pairs of run id and set id) by task run id: task_run_set table rows.

type TaskRunSetTxt 

type TaskRunSetTxt struct {
	SetTxt map[string][]DescrNote // map workset name to description and notes by language
	RunTxt map[string][]DescrNote // map run digest-or-name to description and notes by language
}

TaskRunSetTxt is additional task text: description and notes by language for all input worksets and model runs of the task. Run identified by run digest or, if digest is null then by run name

func GetTaskRunSetText 

func GetTaskRunSetText(dbConn *sql.DB, taskId int, isSuccess bool, langCode string) (*TaskRunSetTxt, error)

GetTaskRunSetText return additinal modeling task text description and notes for all task worksets and model runs: workset_txt, run_txt table rows.

It includes all input worksets text and model runs text. If isSuccess true then return only successfully completed runs else return all runs: success, error, exit, progress. If langCode not empty then only specified language selected else all languages.

type TaskTxtRow 

type TaskTxtRow struct {
	TaskId   int    // task_id  INT           NOT NULL
	LangCode string // lang_code VARCHAR(32)  NOT NULL
	Descr    string // descr     VARCHAR(255) NOT NULL
	Note     string // note      VARCHAR(32000)
}

TaskTxtRow is db row of task_txt

type TypeDicRow 

type TypeDicRow struct {
	ModelId     int    // model_id      INT          NOT NULL
	TypeId      int    // model_type_id INT          NOT NULL
	TypeHid     int    // type_hid      INT          NOT NULL, -- unique type id
	Name        string // type_name     VARCHAR(255) NOT NULL, -- type name: int, double, etc.
	Digest      string // type_digest   VARCHAR(32)  NOT NULL
	DicId       int    // dic_id        INT NOT NULL, -- dictionary id: 0=simple 1=logical 2=classification 3=range 4=partition 5=link
	TotalEnumId int    // total_enum_id INT NOT NULL, -- if total enabled this is enum_value of total item =max+1
	IsRange     bool   // if true then it is range type and enums calculated
	MinEnumId   int    // min enum id
	MaxEnumId   int    // max enum id
	// contains filtered or unexported fields
}

TypeDicRow is db row of type_dic join to model_type_dic table and min, max, count of enum id's.

TypeHid (type_dic.type_hid) is db-unique id of the type, use digest to find same type in other db. TypeId (model_type_dic.model_type_id) is model-unique type id, assigned by model compiler.

func (*TypeDicRow) IsBool 

func (typeRow *TypeDicRow) IsBool() bool

IsBool return true if model type is boolean.

func (*TypeDicRow) IsBuiltIn 

func (typeRow *TypeDicRow) IsBuiltIn() bool

IsBuiltIn return true if model type is built-in, ie: int, double, logical.

func (*TypeDicRow) IsFloat 

func (typeRow *TypeDicRow) IsFloat() bool

IsFloat return true if model type is any of float family.

func (*TypeDicRow) IsFloat32 added in v1.13.0 

func (typeRow *TypeDicRow) IsFloat32() bool

IsFloat32 return true if model type is float 32 bit.

func (*TypeDicRow) IsInt 

func (typeRow *TypeDicRow) IsInt() bool

IsInt return true if model type is integer (not float, string or boolean). If type is not a built-in then it must be integer enums.

func (*TypeDicRow) IsString 

func (typeRow *TypeDicRow) IsString() bool

IsString return true if model type is string.

type TypeEnumRow 

type TypeEnumRow struct {
	ModelId int    // model_id      INT NOT NULL
	TypeId  int    // model_type_id INT NOT NULL
	EnumId  int    // enum_id       INT NOT NULL
	Name    string // enum_name     VARCHAR(255) NOT NULL
}

TypeEnumRow is db row of type_enum_lst join to model_type_dic table

type TypeEnumTxtRow 

type TypeEnumTxtRow struct {
	ModelId  int    // model_id      INT          NOT NULL
	TypeId   int    // model_type_id INT          NOT NULL
	EnumId   int    // enum_id       INT          NOT NULL
	LangCode string // lang_code     VARCHAR(32)  NOT NULL
	Descr    string // descr         VARCHAR(255) NOT NULL
	Note     string // note          VARCHAR(32000)
}

TypeEnumTxtRow is db row of type_enum_txt join to model_type_dic table

type TypeMeta 

type TypeMeta struct {
	TypeDicRow               // model type rows: type_dic join to model_type_dic
	Enum       []TypeEnumRow // type enum rows: type_enum_lst join to model_type_dic
}

TypeMeta is type metadata: type name and enums

type TypeTxtRow 

type TypeTxtRow struct {
	ModelId  int    // model_id      INT          NOT NULL
	TypeId   int    // model_type_id INT          NOT NULL
	LangCode string // lang_code     VARCHAR(32)  NOT NULL
	Descr    string // descr         VARCHAR(255) NOT NULL
	Note     string // note          VARCHAR(32000)
}

TypeTxtRow is db row of type_dic_txt join to model_type_dic table

type WorksetCreatePub added in v1.9.2 

type WorksetCreatePub struct {
	WorksetHdrPub
	Param []ParamValuePub // workset parameters: name, text (value notes by language) and value
}

WorksetCreatePub is "public" workset metadata and parameters list for json import-export. Each parameter must have metadata: name, subvalues count, optinal notes and value(s). Values can be either literal cell values or copy direction, for example run digest to copy from.

type WorksetHdrPub 

type WorksetHdrPub struct {
	ModelName           string      // model name for that workset
	ModelDigest         string      // model digest for that workset
	ModelVersion        string      // model_ver     VARCHAR(32)  NOT NULL
	ModelCreateDateTime string      // create_dt     VARCHAR(32)  NOT NULL
	Name                string      // workset name: set_name VARCHAR(255) NOT NULL
	BaseRunDigest       string      // if not empty then digest of the base run
	IsReadonly          bool        // readonly flag
	UpdateDateTime      string      // last update date-time
	IsCleanBaseRun      bool        // if true then update set base run digest to NULL
	Txt                 []DescrNote // workset text: description and notes by language
}

WorksetHdrPub is "public" workset metadata for json import-export

type WorksetMeta 

type WorksetMeta struct {
	Set   WorksetRow      // workset master row: workset_lst
	Txt   []WorksetTxtRow // workset text rows: workset_txt
	Param []worksetParam  // workset parameter: parameter_hid, sub-value count and workset_parameter_txt rows
}

WorksetMeta is a model workset metadata: name, parameters, decription, notes.

Workset (working set of model input parameters): it can be a full set, which include all model parameters or subset and include only some parameters.

Each model must have "default" workset. Default workset must include ALL model parameters (it is a full set). Default workset is a first workset of the model: set_id = min(set_id). If workset is a subset (does not include all model parameters) then it can be based on model run results, specified by run_id (not NULL).

Workset can be editable or read-only. If workset is editable then you can modify input parameters or workset description, notes, etc. If workset is read-only then you can run the model using that workset as input.

Important: working set_id must be different from run_id (use id_lst to get it) Important: always update parameter values inside of transaction scope Important: before parameter update do is_readonly = is_readonly + 1 to "lock" workset

WorksetMeta is workset metadata db rows: workset_lst, workset_txt, workset_parameter, workset_parameter_txt

func GetWorksetFull 

func GetWorksetFull(dbConn *sql.DB, setRow *WorksetRow, langCode string) (*WorksetMeta, error)

GetWorksetFull return full workset metadata: workset_lst, workset_txt, workset_parameter, workset_parameter_txt table rows.

If langCode not empty then only specified language selected else all languages

func GetWorksetFullList 

func GetWorksetFullList(dbConn *sql.DB, modelId int, isReadonly bool, langCode string) ([]WorksetMeta, error)

GetWorksetFullList return list of full workset metadata: workset_lst, workset_txt, workset_parameter, workset_parameter_txt table rows.

If isReadonly true then return only readonly worksets else all worksets. If langCode not empty then only specified language selected else all languages

func (*WorksetMeta) ToPublic 

func (meta *WorksetMeta) ToPublic(dbConn *sql.DB, modelDef *ModelMeta) (*WorksetPub, error)

ToPublic convert workset db rows into "public" workset format for json import-export

func (*WorksetMeta) UpdateWorkset 

func (meta *WorksetMeta) UpdateWorkset(dbConn *sql.DB, modelDef *ModelMeta, isReplace bool, langDef *LangMeta) error

UpdateWorkset create new workset metadata, replace or merge existing workset metadata in database.

Set name is used to find workset and set id updated with actual database value. Workset must be read-write for replace or merge.

Only metadata updated: list of workset parameters not changed. To add or remove parameters form workset use UpdateWorksetParameterFrom() method. It is an error if incoming list parameters include any parameter which are not already in workset_parameter table.

If workset does not exist then empty workset created, without parameters. Parameter list in workset metadata must be empty if workset does not exist.

Replace is replace of existing metadata or create empty new workset. If workset exist then workset metadata replaced and parameters text metadata replaced.

Merge does merge of text metadata with existing workset or create empty new workset. If workset exist then text is updated if such language already exist or inserted if no text in that language.

func (*WorksetMeta) UpdateWorksetParameterFrom added in v1.9.7 

func (meta *WorksetMeta) UpdateWorksetParameterFrom(
	dbConn *sql.DB, modelDef *ModelMeta, isReplaceMeta bool, param *ParamRunSetPub, langDef *LangMeta, from func() (interface{}, error),
) (int, error)

UpdateWorksetParameterFrom add new or replace existing workset parameter.

If parameter already exist in workset then parameter metadata either merged or replaced. If parameter not exist in workset then parameter values must be supplied, it cannot be empty. If parameter not exist in workset then new parameter metadata and values inserted. If from() not nil then parameter values completely replaced, on each call from() must return CellParam for all parameter values.

Set name is used to find workset and set id updated with actual database value. Workset must be read-write for replace or merge.

type WorksetParamTxtRow 

type WorksetParamTxtRow struct {
	SetId    int    // set_id        INT NOT NULL
	ParamHid int    // parameter_hid INT NOT NULL
	LangCode string // lang_code VARCHAR(32)  NOT NULL
	Note     string // note          VARCHAR(32000), -- parameter value note
}

WorksetParamTxtRow is workset_parameter_txt table row.

func GetWorksetAllParamText 

func GetWorksetAllParamText(dbConn *sql.DB, setId int, langCode string) ([]WorksetParamTxtRow, error)

GetWorksetAllParamText return all workset parameters value notes: workset_parameter_txt table rows.

If langCode not empty then only specified language selected else all languages

func GetWorksetParamText 

func GetWorksetParamText(dbConn *sql.DB, setId int, paramHid int, langCode string) ([]WorksetParamTxtRow, error)

GetWorksetParamText return parameter value notes: workset_parameter_txt table rows.

If langCode not empty then only specified language selected else all languages

type WorksetPub 

type WorksetPub struct {
	WorksetHdrPub
	Param []ParamRunSetPub // workset parameters: name and text (value notes by language)
}

WorksetPub is "public" workset metadata and parameter metadata for json import-export

func (*WorksetPub) FromPublic 

func (pub *WorksetPub) FromPublic(dbConn *sql.DB, modelDef *ModelMeta) (*WorksetMeta, error)

FromPublic convert workset metadata from "public" format (coming from json import-export) into db rows.

type WorksetRow 

type WorksetRow struct {
	SetId          int    // unique working set id
	BaseRunId      int    // if not NULL and positive then base run id (source run id)
	ModelId        int    // model_id     INT          NOT NULL
	Name           string // set_name     VARCHAR(255) NOT NULL
	IsReadonly     bool   // is_readonly  SMALLINT     NOT NULL
	UpdateDateTime string // update_dt    VARCHAR(32)  NOT NULL, -- last update date-time
	// contains filtered or unexported fields
}

WorksetRow is workset_lst table row.

func GetDefaultWorkset 

func GetDefaultWorkset(dbConn *sql.DB, modelId int) (*WorksetRow, error)

GetDefaultWorkset return default working set for the model: workset_lst table row.

Default workset is a first workset for the model, each model must have default workset.

func GetWorkset 

func GetWorkset(dbConn *sql.DB, setId int) (*WorksetRow, error)

GetWorkset return working set by id: workset_lst table row.

func GetWorksetByName 

func GetWorksetByName(dbConn *sql.DB, modelId int, name string) (*WorksetRow, error)

GetWorksetByName return working set by name.

If model has multiple worksets with that name then return first set.

func GetWorksetList 

func GetWorksetList(dbConn *sql.DB, modelId int) ([]WorksetRow, error)

GetWorksetList return list of model worksets: workset_lst rows.

type WorksetTxtRow 

type WorksetTxtRow struct {
	SetId    int    // set_id    INT          NOT NULL
	LangCode string // lang_code VARCHAR(32)  NOT NULL
	Descr    string // descr     VARCHAR(255) NOT NULL
	Note     string // note      VARCHAR(32000)
}

WorksetTxtRow is db row of workset_txt

func GetWorksetText 

func GetWorksetText(dbConn *sql.DB, setId int, langCode string) ([]WorksetTxtRow, error)

GetWorksetText return workset description and notes: workset_txt table rows.

If langCode not empty then only specified language selected else all languages

type WriteLayout 

type WriteLayout struct {
	Name string // parameter name or output table name
	ToId int    // run id or set id to write parameter or output table values
}

WriteLayout describes parameters or output tables values for insert or update.

Name is a parameter or output table name to read.

type WriteMicroLayout added in v1.13.0 

type WriteMicroLayout struct {
	WriteLayout        // common write layout: entity name and run id
	DoubleFmt   string // used for float model types digest calculation
}

WriteMicroLayout describes source and size of data page to read entity microdata.

Double format string is used for digest calculation if value type if float or double.

type WriteParamLayout 

type WriteParamLayout struct {
	WriteLayout        // common write layout: parameter name, run or set id
	SubCount    int    // sub-values count
	IsToRun     bool   // if true then write into into model run else into workset
	IsPage      bool   // if true then write only page of data else all parameter values
	DoubleFmt   string // used for float model types digest calculation
}

WriteParamLayout describes parameter values for insert or update.

Double format string is used for digest calculation if value type if float or double.

type WriteTableLayout 

type WriteTableLayout struct {
	WriteLayout        // common write layout: output table name and run id
	SubCount    int    // sub-values count
	DoubleFmt   string // used for float model types digest calculation
}

WriteTableLayout describes output table values for insert or update.

Double format string is used for digest calculation if value type if float or double.

Source Files

View all Source files

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.