*\ To avoid line-feeding you need to write *\, so that the line currently being used continues on the following line of the file filename.bas.
- *# If this is placed at the beginning of the line, it is considered as a comment and therefore is not written.
- ** In order for an asterisk symbol to appear in the text, two asterisks ** must be written.
- *Include. The include command allows you to include the contents of a slave file inside a master .bas file, setting a relative path from the Problem Type directory to this secondary file.
Example:
*include includes\execntrlmi.h
Note: The *.bas extension cannot be used for the slave file to avoid multiple output files.
- *MessageBox. This command stops the execution of the .bas file and prints a message in a window; this command should only be used when a fatal error occurs.
Example:
*MessageBox error: Quadrilateral elements are not permitted.
- *WarningBox. This is the same as MessageBox, but the execution is not stopped.
Example:
WarningBox Warning: Bad elements. A STL file is a collection of triangles bounding a volume.
The following commands must be written at the beginning of a line and the rest of the line will serve as their modifiers. No additional text should be written.
- *loop, *end, *break. These are declared for the use of loops. A loop begins with a line that starts with loop (none of these commands is case-sensitive) and contains another word to express the variable of the loop. There are some lines in the middle that will be repeated depending on the values of the variable, and whose parameters will keep on changing throughout the iterations if necessary. Finally, a loop will end with a line that finishes with *end. After *end, you may write any kind of comments in the same line. The command **break inside a *loop or *for block, will finish the execution of the loop and will continue after the *end line.
The variables that are available for *loop are the following:
- elems, nodes, faces, materials, conditions, layers, groups, intervals, localaxes. These commands mean, respectively, that the loop will iterate over the elements, nodes, faces of a group, materials, conditions, layers, groups, intervals or local axes systems. The loops can be nested among them. The loop over the materials will iterate only over the effectively assigned materials to an entity, in spite of the fact that more materials have been defined. The number of the materials will begin with the number 1. If a command that depends on the loop is located outside it, the number will also take by default the value 1.
After the command *loop:
- If the variable is nodes, elems or faces, you can include one of the modifiers: *all, *OnlyInCond,*OnlyInLayer or *OnlyInGroup. The first one signifies that the iteration is going to be performed over all the entities.
The *OnlyInCond modifier implies that the iteration will only take place over the entities that satisfy the relevant condition. This condition must have been previously defined with *set cond.
*OnlyInLayer implies that the iteration will only take place over the entities that are in the specified layer; layers must be specified with the command *set Layer.
*OnlyInGroup implies that the iteration will only take place over the entities that are in the specified group;
group must be specified inside a loop groups with the command *set Group *GroupName *nodes|elems|faces, or *set Group <name> , with <name> the full name of the group.
By default, it is assumed that the iteration will affect all the entities. - If the variable is material you can include the modifier *NotUsed to make a loop over those materials that are defined but not used.
- If the variable is conditions you must include one of the modifiers: *Nodes, *BodyElements, *FaceElements, *Layers or *Groups, to do the loop on the conditions defined over this kind of mesh entity, or only the conditions declared 'over layers' or only the ones declared 'over groups'.
- If the variable is layers you can include modifiers: OnlyInCond if before was set a condition defined 'over layers'
- If the variable is groups you can include modifiers: OnlyInCond if before was set a condition defined 'over groups' (e.g. inside a *loop conditions *groups)
Example 1:
*loop nodes *format "%5i%14.5e%14.5e" *NodesNum *NodesCoord(1,real) *NodesCoord(2,real) *end nodes
This command carries out a rundown of all the nodes of the mesh, listing their identifiers and coordinates (x and y coordinates).
Example 2:
*Set Cond Point-Weight *nodes *loop nodes OnlyInCond *NodesNum *cond(1) *end
This carries out a rundown of all the nodes assigned the condition "Point-Weight" and provides a list of their identifiers and the first "weight" field of the condition in each case.
Example 3:
*Loop Elems *ElemsNum *ElemsLayerNum *End Elems
This carries out a rundown of all the elements and provides a list of their identifier and the identifier of the layer to which they belong.
Example 4:
*Loop Layers *LayerNum *LayerName *LayerColorRGB *End Layers
This carries out a rundown of all the layers and for each layer it lists its identifier and name.
Example 5:
*Loop Conditions OverFaceElements *CondName *Loop Elems OnlyInCond *elemsnum *condelemface *cond *End Elems *End Conditions
This carries out a rundown of all conditions defined to be applied on the mesh 'over face elements', and for each condition it lists its name and for each element where this condition is applied are printed the element number, the marked face and the condition field values.
Example 6:
*loop intervals interval=*loopvar *loop conditions *groups *if(condnumentities) condition name=*condname *loop groups *onlyincond *groupnum *groupname *cond *end groups *end if *end conditions *end intervals
This do a loop for each interval, and for each condition defined 'over groups' list the groups where the condition was applied and its values.
- *if, *else, *elseif, *endif. These commands create the conditionals. The format is a line which begins with *if followed by an expression between parenthesis. This expression will be written in C-language syntax, value return commands, will not begin with *, and its variables must be defined as integers or real numbers (see *format, *intformat, *realformat), with the exception of strcmp and strcasecmp. It can include relational as well as arithmetic operators inside the expressions.
The following are valid examples of the use of the conditionals:
*if((fabs(loopvar)/4)<1.e+2) *if((p3<p2)||p4) *if((strcasecmp(cond(1),"XLoad")==0)&&(cond(2)!=0))
The first example is a numerical example where the condition is satisfied for the values of the loop under 400, while the other two are logical operators; in the first of these two, the condition is satisfied when p3<p2 or p4 is different from 0, and in the second, when the first field of the condition is called XLoad (with this particular writing) and the second is not null.
If the checked condition is true, GiD will write all the lines until it finds the corresponding *else, *elseif or *endif (*end is equivalent to *endif after *if). *else or *elseif are optional and require the writing of all the lines until the corresponding *endif, but only when the condition given by *if is false. If either *else or *elseif is present, it must be written between *if and *endif. The conditionals can be nested among them.
The behaviour of *elseif is identical to the behaviour of *else with the addition of a new condition:
*if(GenData(31,int)==1) ...(1) *elseif(GenData(31,int)==2) ...(2) *else ...(3) *endif
In the previous example, the body of the first condition (written as 1) will be written to the data file if GenData(31,int) is 1, the body of the second condition (written as 2) will be written to the data file if GenData(31,int) is 2, and if neither of these is true, the body of the third condition (written as 3) will be written to the data file.
Note: A conditional can also be written in the middle of a line. To do this, begin another line and write the conditional by means of the command *\.
- *for, *end, *break. The syntax of this command is equivalent to *for in C-language.
*for(varname=expr.1;varname<=expr.2;varname=varname+1)
*end for
The meaning of this statement is the execution of a controlled loop, since varname is equal to expr.1 until it is equal to expr.2, with the value increasing by 1 for each step. varname is any name and expr.1 and expr.2 are arithmetical expressions or numbers whose only restrictions are to express the range of the loop.
The command *break inside a *loop or *for block, will finish the execution of the loop and will continue after the *end line.
Example:
*for(i=1;i<=5;i=i+1) variable i=*i *end for
- *set. This command has the following purposes:
- *set cond: To set a condition.
- *set Layer "layer name" *nodes|elems: To set a layer.
- *set Group "group name" *nodes|elems|faces: To set a group. (inside a *loop groups can use *GroupName as "group name" ,to get the name of the group of the current loop)
- *set elems: To indicate the elements.
- *set var: To indicate the variables to use.
It is not necessary to write these commands in lowercase, so *Set will also be valid in all the examples.
*set cond.: In the case of the conditions, GiD allows the combination of a group of them via the use of *add cond. When a specific condition is about to be used, it must first be defined, and then this definition will be used until another is defined. If this feature is performed inside a loop over intervals, the corresponding entities will be chosen. Otherwise, the entities will be those referred to in the first interval.
It is done in this way because when you indicate to the program that a condition is going to be used, GiD creates a table that lets you know the number of entities over which this condition has been applied. It is necessary to specify whether the condition takes place over the *nodes, over the *elems or over *layers to create the table.
So, a first example to check the nodes where displacement constraints exist could be:
*Set Cond Volu-Cstrt *nodes *Add Cond Surf-Cstrt *nodes *Add Cond Line-Cstrt *nodes *Add Cond Poin-Cstrt *nodes
These let you apply the conditions directly over any geometric entity.
*Set Layer "layer name" *elems|nodes
*Add Layer "layer name"
*Remove Layer "layer name"
This command sets a group of nodes. In the following loops over nodes/elements with the modifier *OnlyInLayer, the iterations will only take place over the nodes/elements of that group.
Example 1:
*set Layer example_layer_1 *elems *loop elems *OnlyInLayer Nº:*ElemsNum Name of Layer:*ElemsLayerName Nº of Layer :*ElemsLayerNum *end elems
Example 2:
*loop layers *set Layer *LayerName *elems *loop elems *OnlyInLayer Nº:*ElemsNum Name of Layer:*ElemsLayerName Nº of Layer :*ElemsLayerNum *end elems *end layers
In this example the command *LayerName is used to get the layer name.
There are some modifiers available to point out particular specifications of the conditions.
If the command *CanRepeat is added after *nodes or *elems in *Set cond, one entity can appear several times in the entities list. If the command *NoCanRepeat is used, entities will appear only once in the list. By default, *CanRepeat is off except where one condition has the *CanRepeat flag already set.
A typical case where you would not use *CanRepeat might be:
*Set Cond Line-Constraints *nodes
In this case, when two lines share one endpoint, instead of two nodes in the list, only one is written.
A typical situation where you would use *CanRepeat might be:
*Set Cond Line-Pressure *elems *CanRepeat
In this case, if one triangle of a quadrilateral has more than one face in the marked boundary then we want this element to appear several times in the elements list, once for each face.
Other modifiers are used to inform the program that there are nodes or elements that can satisfy a condition more than once (for instance, a node that belongs to a certain number of lines with different prescribed movements) and that have to appear unrepeated in the data input file, or, in the opposite case, that have to appear only if they satisfy more than one condition. These requirements are achieved with the commands *or(i,type) and *and(i,type), respectively, after the input of the condition, where i is the number of the condition to be considered and type is the type of the variable (integer or real).
For the previous example there can be nodes or elements in the intersection of two lines or maybe belonging to different entities where the same condition had been applied. To avoid the repetition of these nodes or elements, GiD has the modifier *or, and in the case where two or more different values were applied over a node or element, GiD only would consider one, this value being different from zero. The reason for this can be easily understood by looking at the following example. Considering the previous commands transformed as:
*Set Cond Volu-Cstrt *nodes *or(1,int) *or(2,int) *Add Cond Surf-Cstrt *nodes *or(1,int) *or(2,int) *Add Cond Line-Cstrt *nodes *or(1,int) *or(2,int) *Add Cond Poin-Cstrt *nodes *or(1,int) *or(2,int)
where *or(1,int) means the assignment of that node to the considered ones satisfying the condition if the integer value of the first condition's field is different from zero, and (*or(2,int) means the same assignment if the integer value of the second condition's field is different from zero). Let us imagine that a zero in the first field implies a restricted movement in the direction of the X-axis and a zero in the second field implies a restricted movement in the direction of the Y-axis. If a point belongs to an entity whose movement in the direction of the X-axis is constrained, but whose movement in the direction of the Y-axis is released and at the same time to an entity whose movement in the direction of the Y-axis is constrained, but whose movement in the direction of the X-axis is released, GiD will join both conditions at that point, appearing as a fixed point in both directions and as a node satisfying the four expressed conditions that would be counted only once.
The same considerations explained for adding conditions through the use of *add cond apply to elements, the only difference being that the command is *add elems. Moreover, it can sometimes be useful to remove sets of elements from the ones assigned to the specific conditions. This can be done with the command *remove elems. So, for instance, GiD allows combinations of the type:
*Set Cond Dummy *elems *Set elems(All) *Remove elems(Linear)
To indicate that all dummy elements apart from the linear ones will be considered, as well as:
*Set Cond Dummy *elems *Set elems(Hexahedra) *Add elems(Tetrahedra) *Add elems(Quadrilateral) *Add elems(Triangle)
The format for *set var differs from the syntax for the other two *set commands. Its syntax is as follows:
*Set var varname = expression
where varname is any name and expression is any arithmetical expression, number or command, where the latter must be written without *** and must be defined as Int or Real.
A Tcl procedure can also be called, but it must return a numerical result.The following are valid examples for these assignments:
*Set var ko1=cond(1,real)
*Set var ko2=2
*Set var S1=CondNumEntities
*Set var p1=elemsnum()
*Set var b=operation(p1*2)
*tcl(proc MultiplyByTwo { x } { return [expr {$x*2}] })*\
*Set var a=tcl(MultiplyByTwo *p1)
- *intformat, *realformat, *format. These commands explain how the output of different mathematical expressions will be written to the analysis file. The use of this command consists of a line which begins with the corresponding version, *intformat, *realformat or *format (again, these are not case-sensitive), and continues with the desired writing format, expressed in C-language syntax argument, between double quotes (").
The integer definition of *intformat and the real number definition of *realformat remain unchanged until another definition is provided via *intformat and *realformat, respectively. The argument of these two commands is composed of a unique field. This is the reason why the *intformat and *realformat commands are usually invoked in the initial stages of the .bas file, to set the format configuration of the integer or real numbers to be output during the rest of the process.
The *format command can include several field definitions in its argument, mixing integer and real definitions, but it will only affect the line that follows the command's instance one. Hence, the *format command is typically used when outputting a listing, to set a temporary configuration.
In the paragraphs that follow, there is an explanation of the C format specification, which refers to the field specifications to be included in the arguments of these commands. Keep in mind that the type of argument that the *format command expects may be composed of several fields, and the *intformat and *realformat commands' arguments are composed of an unique field, declared as integer and real, respectively, all inside double quotes:
A format specification, which consists of optional and required fields, has the following form: %[flags][width][.precision]typeThe start of a field is signaled by the percentage symbol (%). Each field specification is composed of: some flags, the minimum width, a separator point, the level of precision of the field, and a letter which specifies the type of the data to be represented. The field type is the only one required.
The most common flags are:
- To left align the result
+ To prefix the numerical output with a sign (+ or -)
# To force the real output value to contain a decimal point.
The most usual representations are integers and floats. For integers the letters d and i are available, which force the data to be read as signed decimal integers, and u for unsigned decimal integers.
For floating point representation, there are the letters e, f and g, these being followed by a decimal point to separate the minimum width of the number from the figure giving the level of precision.The number of digits after the decimal point depends on the requested level of precision.
Note: The standard width specification never causes a value to be truncated. A special command exists in GiD: *SetFormatForceWidth, which enables this truncation to a prescribed number of digits.
For string representations, the letter s must be used. Characters are printed until the precision value is reached.
The following are valid examples of the use of format:
*Intformat "%5i"
With this sentence, usually located at the start of the file, the output of an integer quantity is forced to be right aligned on the fifth column of the text format on the right side. If the number of digits exceeds five, the representation of the number is not truncated.
*Realformat "%10.3e"
This sentence, which is also frequently located in the first lines of the template file, sets the output format for the real numbers as exponential with a minimum of ten digits, and three digits after the decimal point.
*format "%10i%10.3e%10i%15.6e"
This complex command will specify a multiple assignment of formats to some output columns. These columns are generated with the line command that will follow the format line. The subsequent lines will not use this format, and will follow the general settings of the template file or the general formats: *IntFormat, *RealFormat.
- *SetFormatForceWidth, *SetFormatStandard The default width specification of a "C/C+" format, never causes a value to be truncated.
*SetFormatForceWidth is a special command that allows a figure to be truncated if the number of characters to print exceeds the specified width.
*SetFormatStandard changes to the default state, with truncation disabled.
For example:
*SetFormatForceWidth *set var num=-31415.16789 *format "%8.3f" *num *SetFormatStandard *format "%8.3f" *num
Output:
-31415.1
-31415.168
The first number is truncated to 8 digits, but the second number, printed with "C" standard, has 3 numbers after the decimal point, but more than 8 digits.
- *Tcl This command allows information to be printed using the Tcl extension language. The argument of this command must be a valid Tcl command or expression which must return the string that will be printed. Typically, the Tcl command is defined in the Tcl file (.tcl , see TCL AND TK EXTENSION for details).
Example: In this example the *Tcl command is used to call a Tcl function defined in the problem type .tcl file. That function can receive a variable value as its argument with *variable. It is also possible to assign the returned value to a variable, but the Tcl procedure must return a numerical value.
In the .bas file:
*set var num=1 *tcl(WriteSurfaceInfo *num) *set var num2=tcl(MultiplyByTwo *num)
In the .tcl file:
proc WriteSurfaceInfo { num } {
return [GiD_Info list_entities surfaces $num]
}
proc MultiplyByTwo { x } {
return [expr {$x*2}]
}