Inheritance diagram for pyfortool.variables.Variables:

[legend]

Public Member Functions
	__init__ (self, **kwargs)

	varList (self)

	attachArraySpecToEntity (self)

	checkImplicitNone (self, mustRaise=False)

	checkIntent (self, mustRaise=False)

	removeVar (self, varList, simplify=False)

	addVar (self, varList)

	addModuleVar (self, moduleVarList)

	showUnusedVar (self)

	removeUnusedLocalVar (self, excludeList=None, simplify=False)

	addExplicitArrayBounds (self, node=None)

	addArrayParentheses (self)

	addArrayParenthesesInNode (self, node)

	modifyAutomaticArrays (self, declTemplate=None, startTemplate=None, endTemplate=None)

	findIndexArrayBounds (self, arr, index, loopVar)

	arrayR2parensR (self, namedE, table)

	findArrayBounds (self, arr, loopVar, extraVarList=None)

	renameVar (self, oldName, newName)

	removeVarIfUnused (self, varList, excludeDummy=False, excludeModule=False, simplify=False)

	isVarUsed (self, varList, exactScope=False, dummyAreAlwaysUsed=False)

	addArgInTree (self, varName, declStmt, pos, stopScopes, moduleVarList=None, otherNames=None, parser=None, parserOptions=None, wrapH=False)

Static Public Member Functions
	varSpec2stmt (varSpec)

Public Attributes
	path

Protected Member Functions
	_normalizeScopeVar (self, scopeVarList)

	_normalizeUniqVar (self, scopeVarList)

Protected Attributes
	_varList

	_excludeContains

Detailed Description

Methos to deal with variables

Definition at line 250 of file variables.py.

Constructor & Destructor Documentation

◆ init()

pyfortool.variables.Variables.__init__	(		self,
		**	kwargs
	)

**kwargs is used to enable the use of super().__init__

Reimplemented in pyfortool.pyfortool.PYFT, and pyfortool.scope.PYFTscope.

Definition at line 254 of file variables.py.

Member Function Documentation

◆ _normalizeScopeVar()

pyfortool.variables.Variables._normalizeScopeVar	(	self,
		scopeVarList
	)

protected

Internal method to normalize scopeVarList
(list of tuples made of scope path, variable name, and optional other values)

Definition at line 274 of file variables.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ _normalizeUniqVar()

pyfortool.variables.Variables._normalizeUniqVar	(	self,
		scopeVarList
	)

protected

Internal method to suppress duplicates in scopeVarList
(list of tuples made of scope path, variable name, and optional other values)

Definition at line 283 of file variables.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ addArgInTree()

pyfortool.variables.Variables.addArgInTree	(	self,
		varName,
		declStmt,
		pos,
		stopScopes,
		moduleVarList = `None`,
		otherNames = `None`,
		parser = `None`,
		parserOptions = `None`,
		wrapH = `False`
	)

Adds an argument to the routine and propagates it upward until we encounter a scope
where the variable exists or a scope in stopScopes
:param varName: variable name
:param declStmt: declarative statment (will be used by addVar)
:param pos: position of the variable in the list of dummy argument
:param stopScopes: list of scopes to reach
:param moduleVarList: list of module variable specification to insert in the xml code
                      a module variable specification is a list of two elements:
                      - module name
                      - variable name or or list of variable names
                        or None to add a USE statement without the ONLY attribute
                      use moduleVarList to not add module variables
:param otherNames: None or list of other variable names that can be used
                   These variables are used first
:param parser, parserOptions, wrapH: see the PYFT class

Argument is inserted only on paths leading to one of scopes listed in stopScopes

Definition at line 1358 of file variables.py.

Here is the call graph for this function:

◆ addArrayParentheses()

pyfortool.variables.Variables.addArrayParentheses ( self )

Look for arrays and add parenthesis. A => A(:)

Definition at line 749 of file variables.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ addArrayParenthesesInNode()

pyfortool.variables.Variables.addArrayParenthesesInNode	(	self,
		node
	)

Look for arrays and add parenthesis. A => A(:)
:param node: xml node in which ':' must be added

Definition at line 793 of file variables.py.

Here is the call graph for this function:

◆ addExplicitArrayBounds()

pyfortool.variables.Variables.addExplicitArrayBounds	(	self,
		node = `None`
	)

Replace ':' by explicit arrays bounds.
:param node: xml node in which ':' must be replaced (None to replace everywhere)

Definition at line 693 of file variables.py.

Here is the call graph for this function:

◆ addModuleVar()

pyfortool.variables.Variables.addModuleVar	(	self,
		moduleVarList
	)

:param moduleVarList: list of module variable specification to insert in the xml code
                      a module variable specification is a list of three elements:
                      - scope path (path to module, subroutine, function or type
                        declaration)
                      - module name
                      - variable name or or list of variable names
                        or None to add a USE statement without the ONLY attribute
For example addModuleVar('sub:FOO', 'MODD_XX', 'Y') will add the following line in
subroutine FOO:
USE MODD_XX, ONLY: Y

Definition at line 594 of file variables.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ addVar()

pyfortool.variables.Variables.addVar	(	self,
		varList
	)

:param varList: list of variable specification to insert in the xml code
                a variable specification is a list of four elements:
                - variable scope path (path to module, subroutine, function or type
                  declaration)
                - variable name
                - declarative statment
                - position of the variable in the list of dummy argument,
                  None for a local variable

Definition at line 504 of file variables.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ arrayR2parensR()

pyfortool.variables.Variables.arrayR2parensR	(	self,
		namedE,
		table
	)

Transform a array-R into a parens-R node by replacing slices by variables
In 'A(:)', the ':' is in a array-R node whereas in 'A(JL)', 'JL' is in a parens-R node.
Both the array-R and the parens-R nodes are inside a R-LT node
:param namedE: a named-E node
:param table: dictionnary returned by the decode function
:param varList: None or a VarList object in which varaibles are searched for

Definition at line 1043 of file variables.py.

Here is the call graph for this function:

◆ attachArraySpecToEntity()

pyfortool.variables.Variables.attachArraySpecToEntity ( self )

Find all T-decl-stmt elements that have a child element 'attribute' with
attribute-N="DIMENSION" and move the attribute into EN-N elements
E.g., before :
REAL, DIMENSION(D%NIJT,D%NKT) :: ZTLK, ZRT
INTEGER, PARAMETER, DIMENSION(1,1) :: IBUEXTRAIND=(/18, 30/)
after :
REAL :: ZTLK(D%NIJT,D%NKT), ZRT(D%NIJT,D%NKT)
INTEGER, PARAMETER  :: IBUEXTRAIND(1,1)=(/18, 30/)
Limitations : "DIMENSION" must be in upper case in attribute.text

Definition at line 297 of file variables.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ checkImplicitNone()

pyfortool.variables.Variables.checkImplicitNone	(	self,
		mustRaise = `False`
	)

:param mustRaise: True to raise
Issue a logging.warning if the "IMPLICIT NONE" statment is missing
If mustRaise is True, issue a logging.error instead and raise an error

Definition at line 327 of file variables.py.

Here is the call graph for this function:

◆ checkIntent()

pyfortool.variables.Variables.checkIntent	(	self,
		mustRaise = `False`
	)

:param mustRaise: True to raise
Issue a logging.warning if some "INTENT" attributes are missing
If mustRaise is True, issue a logging.error instead and raise an error

Definition at line 349 of file variables.py.

Here is the call graph for this function:

◆ findArrayBounds()

pyfortool.variables.Variables.findArrayBounds	(	self,
		arr,
		loopVar,
		extraVarList = `None`
	)

Find bounds and loop variable given an array
:param arr: array node (named-E node with a array-R child)
:param loopVar: None to create new variable for each added DO loop
                or a function that return the name of the variable to use for the loop
                    control.
                This function returns a string (name of the variable), or True to create
                a new variable, or False to not transform this statement
                The functions takes as arguments:
                  - lower and upper bounds as defined in the declaration statement
                  - lower and upper bounds as given in the statement
                  - name of the array
                  - index of the rank
:param extraVarList: None or list of variables (such as those contained in a VarList object)
                     defined but not yet available in the self.varList object.
:return: the tuple (table, newVar) where:
            table is a dictionnary: keys are loop variable names
                                    values are tuples with lower and upper bounds
            newVar is a list of loop variables not found in varList. This list has the same
                   format as the varList list.

In case the loop variable cannot be defined, the function returns (None, [])

Definition at line 1137 of file variables.py.

Here is the call graph for this function:

◆ findIndexArrayBounds()

pyfortool.variables.Variables.findIndexArrayBounds	(	self,
		arr,
		index,
		loopVar
	)

Find bounds and loop variable for a given array index
:param arr: array node (named-E node with a array-R child)
:param index: index of the rank of the array
:param loopVar: None to create new variable for each added DO loop
                or a function that return the name of the variable to use for the
                    loop control.
                This function returns a string (name of the variable), or True to create
                a new variable, or False to not transform this statement
                The functions takes as arguments:
                  - lower and upper bounds as defined in the declaration statement
                  - lower and upper bounds as given in the statement
                  - name of the array
                  - index of the rank
:return: the tuple (loopName, lowerBound, upperBound) where:
            loopName is the name of the variable to use for the loop
            lower and upper bounds are the bounds to use for the DO loop
loopName can be:
    - a string
    - False to discard this index
    - True to create a new variable to loop with

Definition at line 989 of file variables.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ isVarUsed()

pyfortool.variables.Variables.isVarUsed	(	self,
		varList,
		exactScope = `False`,
		dummyAreAlwaysUsed = `False`
	)

:param varList: list of variables to test. Each item is a list or tuple of two elements.
                The first one describes where the variable is declared, the second one is
                the name of the variable. The first element is a '/'-separated path with
                each element having the form 'module:<name of the module>',
                'sub:<name of the subroutine>' or 'func:<name of the function>'
:param exactScope: True to search strictly in scope
:param dummyAreAlwaysUsed: Returns True if variable is a dummy argument
:return: a dict whose keys are the elements of varList, and values are True when the
         variable is used, False otherwise

If exactScope is True, the function will search for variable usage
only in this scope. But this feature has a limited interest.

If exactScope is False:
  - if scopePath is a subroutine/function in a contains section,
    and if the variable is not declared in this scope, usages are
    searched in the module/subroutine/function upper that declared
    the variable and in all subroutines/functions in the contains section
  - if scopePath is a module/subroutine/function that has a
    contains sections, usages are searched in all subroutines/functions
    in the contains section

To know if a variable can be removed, you must use exactScope=False

Definition at line 1261 of file variables.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ modifyAutomaticArrays()

pyfortool.variables.Variables.modifyAutomaticArrays	(	self,
		declTemplate = `None`,
		startTemplate = `None`,
		endTemplate = `None`
	)

:param declTemplate: declaration template
:param startTemplate: template for the first executable statement
:param: endTemplate: template for the last executable statement
:return: number of arrays modified
Modifies all automatic arrays declaration in subroutine and functions. The declaration is
replaced by the declaration template, the start template is inserted as first executable
statement and the end template as last executable statement. Each template can use the
following place holders:
"{doubledotshape}", "{shape}", "{lowUpList}", "{name}" and "{type}" wich are, respectively
modified into ":, :, :", "I, I:J, 0:I", "1, I, I, J, 0, I", "A", "REAL" if the original
declaration statement was "A(I, I:J, 0:I)". The template
"{type}, DIMENSION({doubledotshape}), ALLOCATABLE :: {name}#
 ALLOCATE({name}({shape}))#DEALLOCATE({name})"
replaces automatic arrays by allocatables

Definition at line 826 of file variables.py.

Here is the call graph for this function:

◆ removeUnusedLocalVar()

pyfortool.variables.Variables.removeUnusedLocalVar	(	self,
		excludeList = `None`,
		simplify = `False`
	)

Remove unused local variables (dummy and module variables are not suppressed)
:param excludeList: list of variable names to exclude from removal (even if unused)
:param simplify: try to simplify code (if we delete a declaration statement that used a
                 variable as kind selector, and if this variable is not used else where,
                 we also delete it)

Definition at line 673 of file variables.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ removeVar()

pyfortool.variables.Variables.removeVar	(	self,
		varList,
		simplify = `False`
	)

:param varList: list of variables to remove. Each item is a list or tuple of two elements.
                The first one describes where the variable is used, the second one is
                the name of the variable. The first element is a '/'-separated path with
                each element having the form 'module:<name of the module>',
                'sub:<name of the subroutine>', 'func:<name of the function>' or
                'type:<name of the type>'
:param simplify: try to simplify code (if we delete a declaration statement that used a
                 variable as kind selector, and if this variable is not used else where,
                 we also delete it)
Remove the variable from declaration, and from the argument list if needed

Definition at line 370 of file variables.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ removeVarIfUnused()

pyfortool.variables.Variables.removeVarIfUnused	(	self,
		varList,
		excludeDummy = `False`,
		excludeModule = `False`,
		simplify = `False`
	)

:param varList: list of variables to remove if unused. Each item is a list or tuple of two
                elements.
                The first one describes where the variable is used, the second one is
                the name of the variable. The first element is a '/'-separated path with
                each element having the form 'module:<name of the module>',
                'sub:<name of the subroutine>' or 'func:<name of the function>'
:param excludeDummy: if True, dummy arguments are always kept untouched
:param excludeModule: if True, module variables are always kept untouched
:param simplify: try to simplify code (if we delete a declaration statement that used a
                 variable as kind selector, and if this variable is not used else where,
                 we also delete it)
:return: the varList without the unremovable variables
If possible, remove the variable from declaration, and from the argument list if needed

Definition at line 1230 of file variables.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ renameVar()

pyfortool.variables.Variables.renameVar	(	self,
		oldName,
		newName
	)

:param oldName: old name of the variable
:param newName: new name of the variable

Definition at line 1216 of file variables.py.

Here is the call graph for this function:

◆ showUnusedVar()

pyfortool.variables.Variables.showUnusedVar ( self )

Displays on stdout a list of unued variables

Definition at line 657 of file variables.py.

Here is the call graph for this function:

◆ varList()

pyfortool.variables.Variables.varList ( self )

Returns the varList object corresponding to the node

Definition at line 261 of file variables.py.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ varSpec2stmt()

pyfortool.variables.Variables.varSpec2stmt ( varSpec )

static

:param varSpec: a variable description, same form as the items return by self.varList
:return: the associated declarative statement

Definition at line 951 of file variables.py.

Here is the caller graph for this function:

Member Data Documentation

◆ _excludeContains

pyfortool.variables.Variables._excludeContains

protected

Definition at line 271 of file variables.py.

◆ _varList

pyfortool.variables.Variables._varList

protected

Definition at line 258 of file variables.py.

◆ path

pyfortool.variables.Variables.path

Reimplemented in pyfortool.scope.PYFTscope.

Definition at line 271 of file variables.py.

The documentation for this class was generated from the following file:

/home/runner/work/pyfortool/pyfortool/src/pyfortool/variables.py

Public Member Functions

Static Public Member Functions

Public Attributes

Protected Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ __init__()

Member Function Documentation

◆ _normalizeScopeVar()

◆ _normalizeUniqVar()

◆ addArgInTree()

◆ addArrayParentheses()

◆ addArrayParenthesesInNode()

◆ addExplicitArrayBounds()

◆ addModuleVar()

◆ addVar()

◆ arrayR2parensR()

◆ attachArraySpecToEntity()

◆ checkImplicitNone()

◆ checkIntent()

◆ findArrayBounds()

◆ findIndexArrayBounds()

◆ isVarUsed()

◆ modifyAutomaticArrays()

◆ removeUnusedLocalVar()

◆ removeVar()

◆ removeVarIfUnused()

◆ renameVar()

◆ showUnusedVar()

◆ varList()

◆ varSpec2stmt()

Member Data Documentation

◆ _excludeContains

◆ _varList

◆ path

◆ init()