3. Operands#
3.1. Operand UNITE#
♦ UNITE = u
Integer, logical number of the reading unit, defining the file from which the data is read.
3.2. Operand FORMAT#
◊ FORMAT = 'LIBRE' or 'NUMPY',
Indicates the format of the file to be read. Format NUMPY indicates that the file was produced by numpy.save; in LIBRE, it is a simple text file.
3.3. Operand SEPARATEUR#
◊ SEPARATEUR = 'None' or Sep
Character separator of the columns of numbers in the file to be read.
The default value “None” means one or more blank characters. For other separators, white characters are allowed between the numbers and the separator Sep itself.
Not used in the NUMPY format.
3.4. Operands PROL_DROITE/PROL_GAUCHE/NOM_PARA/NOM_RESU//INTERPOL//VERIF/TITRE/INFO#
The numeric values are read from the specified file, then the function is actually created by read_function by calling DEFI_FONCTION or DEFI_NAPPE.
Thus, these keywords are used identically to DEFI_FONCTION or DEFI_NAPPE and make it possible to specify the characteristics of the function produced: interpolation, extensions, names of the access parameters.
3.5. Operand INDIC_PARA#
◊ INDIC_PARA =/[:ref:`indice1,indice2 <indice1,indice2>`]
This keyword is identical in principle to the VALE_PARA keyword in the DEFI_FONCTION command. It allows you to choose the column of numbers in the file read that will define the x-axis of the function, i.e. the values of the parameter.
We necessarily expect a list of two integers: the first designating the number of “block” in the file (see paragraph [§4] Examples), the second designating the number of the column. To identify the blocks in the file, a block separator is considered to be any row that does not contain exclusively numbers and column separators. Columns are defined by the presence of column separators delimiting them (keyword SEPARATEUR).
The default value is [1,1]: we read the first column of the first block.
In the NUMPY format, there is only one « block » so the first clue is always equal to 1.
3.6. Operand TYPE#
◊ TYPE =/'FONCTION', [DEFAUT]
/”FONCTION_C”, /”NAPPE”,
Specify the type of function read and created: “FONCTION”, “FONCTION_C” (call in the macro command to DEFI_FONCTION) or “NAPPE” (call to DEFI_NAPPE).
3.7. Case of a real function: TYPE =” FONCTION “#
◊ INDIC_RESU =/[:ref:`indice1,indice2 <indice1,indice2>`]
Operating identical to INDIC_PARA, this keyword allows you to choose the number column in the file read defining the ordinates of the function, i.e. the values of the result.
The default value is [1,2]: we read the second column of the first block.
In NUMPY format, there is only one « block ».
3.8. Case of a complex function: TYPE =” FONCTION_C “#
◊ FORMAT_C = 'REEL_IMAG'
♦ INDIC_REEL =/[:ref:`indice1,indice2 <indice1,indice2>`]
♦ INDIC_IMAG =/[:ref:`indice1,indice2 <indice1,indice2>`]
◊ FORMAT_C = 'MODULE_PHASE'
♦ INDIC_MODU =/[:ref:`indice1,indice2 <indice1,indice2>`]
♦ INDIC_PHAS =/[:ref:`indice1,indice2 <indice1,indice2>`]
In the case of a complex function, two columns of numbers (real values) must be identified in the file read to define the list of complex values of the result. It is thus possible to read either real part/imaginary part or module/phase. The phase is expressed in radians.
The values by default are respectively [1,2] and [1,3]: we read the second and third columns of the first block. The first column read by default was the value of the parameter (INDIC_PARA).
3.9. Case of a tablecloth: TYPE =” NAPPE “#
♦ NOM_PARA_FONC = cf keyword NOM_PARA
♦ INDIC_ABSCISSE = [:ref:`indice1,indice2 <indice1,indice2>`],
◊ INTERPOL_FONC = cf keyword INTERPOL
◊ PROL_DROITE_FONC = cf keyword PROL_DROITE
◊ PROL_GAUCHE_FONC = cf keyword PROL_GAUCHE
◊ DEFI_FONCTION = _F (
♦ INDIC_RESU = [:ref:`indice1,indice2 <indice1,indice2>`],
The principle is the same as that of DEFI_NAPPE [U4.31.03]: the list of possible values of the parameter (second variable) having been defined by INDIC_PARA, it remains to identify the functions in the file read by the keyword factor DEFI_FONCTION (analogous to the keyword factor FONCTION of DEFI_NAPPE). Attention, the number of functions identified (occurrences of the factor keyword) must be identical to the cardinal of the values read by INDIC_PARA.
Unlike tablecloths in general, functions are required to share the same list of INDIC_ABSCISSE x-axis values, which is why only INDIC_RESU is present under the key word factor DEFI_FONCTION.
The keywords “_ FONC “apply to the functions of the table (see DEFI_NAPPE, keyword factor DEFI_FONCTION [U4.31.03]).