...
Further details may vary from block to block and are to be specified for each data block. Hence two types of headers must be used in a data file, viz.:
- A File Headerheader at the top of a data file, and
- Block Headerheader(s) , one above each data block.
...
row: one series per block and the data are time sequential row-wise
col: one series per block and the data are time sequential column-wise
par: multiple series per block and the data are time sequential column-wise
Block Header(s)
The layout of the data block header depends beside on the layout of the data block also and on the contents of the File Header; following distinction is madefile header :
- single series row-wise reading
- single series column-wise reading,
- multiple series column-wise reading,
- data files without headers (#7) require data block(s):
- in fixed format, and
- row-wise time sequential data.
- The data files of non-equidistant time series comprise only one data block, which may contain a number of series, structured column-wise reading(#8).
Note
- The HYMOS special ASCII file formats were developed under HYMOS 3. In this version of HYMOS the number of characters entered for a station ID were limited to 10, the number of characters entered for a parameter ID were limited to 2. It is not possible to use ID's with more characters than these limits. When your ID's have 12 or 3 character lengths for station and/or parameter use the SWDES file format instead.
- The files that are imported via special ASCII format should not end with missing values. It is better to leave out missing values at the end of the file. This is a known flaw in the import routine.
- If only one block of equidistant data exists in the file, the number of rows (see below) does not have to be entered. The value can be skipped in the input by entering ,, instead of e.g.,12,.
...