home *** CD-ROM | disk | FTP | other *** search
- READ File Format
-
- The READ function processes ASCII text records containing a mixture of
- time tag fields (i.e. the independent variable's value), one or more
- data fields, and optional comments. The carriage return character is
- used as a record separator and spaces, commas, tabs, or comments are
- used as field separators within records.
-
- Comments are any string of characters (except '}') enclosed between a
- '{' and a '}' or between pairs of double quote characters (").
- Comments may appear anywhere within the file except between the
- characters of a time tag or data field; i.e. if a data field is e.g.
- '12' then a comment should not appear between the '1' and the '2'.
- Comments may be continued across records but should not, in this case,
- appear between time tag or data fields.
- +
- An example comment is:
- {Entry High Low Close Date Time}
-
- The first field in a record (i.e. the first field on the line
- following a carriage return) is considered to be the time tag for all
- subsequent values in the record. Time tag fields should be in
- increasing order. All fields following the time tag are data fields
- until another carriage return is encountered. For example,
-
- 1.2 4E1 2 3.625 7/7/86 13:45:22.5<CR><LF>
- where:
- 1.2 is the time tag.
- 4E1, 2 and 3.625 are numeric data fields.
- 7/7/86 is a date data field.
- 13:45:22.5 is a time data field.
-
- Time tag and data fields may have any of the formats that constants
- may have, except the leading '#' character must not be used (see
- Constants topic).
- +
- When processing a file whose records contain multiple data fields on
- the same line, multiple READ function calls must be used; one READ call
- for each data field. These READ calls may be grouped via the '&'
- statement terminator, occur multiple times in an expression (e.g. READ+
- READ), or a combination of both. Each occurrence of a READ causes one
- data value (time tag values do not count) to be read from the ASCII
- file. For example, to process an ASCII file containing records as in
- the above example, the following statements might be employed:
-
- high = READ & low = READ & close = READ &
- quote_date = READ & quote_time = READ;
-
- The DAN algorithm could then go on to reference the files 'high',
- 'low', 'close', 'quote_date', and 'quote_time' in subsequent
- statements. See the PLOTSTOK and PLOTSPOT examples and use the text
- editor to examine the data files used in these algorithms.
- +
- Since DAN performs a linear interpolation to determine the value of a
- function at step times other than those explicitly given, it is not
- necessary for time tag values to be evenly spaced between records (i.e.
- the difference between time tags for successive records does not have
- to be a constant value). Also, there is no requirement on the number
- of records.
-
- If the algorithm does not specify START or DUR values, the READ
- function will automatically set these control values from the first
- time tag in the file and the time span processed during input. That
- is, START is set to the first time tag, and DUR is set to the
- difference between the last time tag found and the first time tag.
- These become the default values for subsequent processing.
- +
- Accepting OPER Input From A Text File
- -------------------------------------
-
- A less flexible, but still usable, way to enter data from a text file
- is to use the I/O redirection facilities of DOS and the OPER predefined
- variable.
-
- In this method the user creates a text file containing just data values
- (no time tags). There must be one data value for each step time in the
- processing algorithm and the data values must correspond to the even
- spacing specified by the STEP value used in the algorithm. Comments
- are not allowed in this text file.
-
- An algorithm is then created that references the OPER variable when
- successive values from the text file are desired.
- +
- The user then invokes DAN with the following command line:
-
- C>DAN algor_name <text_file
-
- where `algor_name' is the name given to the algorithm file and
- `text_file' is the file name of the data file. Note that only one data
- file can be input this way since I/O redirection only applies to one
- file. The `Suppress OPER Prompt' Execution Option is used to suppress
- the message that is normally output when operator input is expected.
-
- If the number of data points changes then the STEP and/or DUR values
- must be changed in the algorithm.
-