mkload

NAME

USAGE

DESCRIPTION

Wild card patterns can be used in the list of filenames. The output of the task is a load file for each file name on the command line. The load file has the same root file name as the original file, but has the extension .lod.

Each load file describes a single reference file. A load file is an ASCII file with no upper limit on the length of its fields. Field widths are variable, so no space is wasted. Because the load file is ASCII, it can be viewed and editing with any text editor and transferred between computer systems without conversion.

The loadfile is divided into two parts: a header and body. The body is further subdivided into rows. Each division has a terminator. The terminator for the heading is as ENDHEADER line. For the body it is an ENDFILE line. For each row it is an ENDROW line. A terminator is a single word, in either case, starting in the first column, as the sole word on the line. Comments can be placed anywhere in the loadfile before the ENDFILE line. Comments are any line whose first non-white character is a sharp #. Blank lines are treated as comments, except that they can also follow the ENDFILE line.

This task copies the information that will be stored in the database from the reference file to the load file. It reads the CDBS database to get the comparison file name sets the COMPARISON_FILE to this value. If no comparison file is found in the database, it sets the field to "(INITIAL)". If the useafter date is not found in the reference file header, it sets the USEAFTER_DATE field to the useafter date of the comparison file on the assumption that the useafter date of the reference file is the same as that of the comparison file.

Some other fields are left blank that must be filled in by the user. These are the OPUS_FLAG, CHANGE_LEVEL, and COMMENT in the header and PEDIGREE, OBSERVATION_BEGIN_DATE, OBSERVATION_END_DATE, and COMMENT in the body of the load file.

PARAMETERS

• file [string] The reference file used to create the load file.

DATA FILES

ALGORITHM

Mkload determines the type of the reference file from the catalog file, creates an empty table, and calls mkload_header and mkload_body to read the information from the reference file into the table. It then calls mkload_name to create the name of the load file, writes the table to the load file, and frees the table. Mkload_header writes the load file header keywords. Some of these keywords are read from the database, so, for the present, they are left blank. Mkload_body writes the observation mode keywords and the pedigree information to the body of the table. If the reference file is an image, it writes one row to the body of the table from the information in the image header. If the reference file is a table, it writes one row for each unique row in the table. The observation mode keywords are read from an array in obsmode.c which contains all the observation mode keywords indexed by instrument. The function searches the reference file for the keyword and if it finds it, it is written to the table. If not, that particular keyword is not written. Mkload_compfile sets the comparison file name and useafter dates in the header to values read from the CDBS database. Mkload_name creates the name of the load file by changing the extension of the reference file to .lod.

The loadfile library handles reading and writing of load files. The load file is read into an in memory structure called a table, which is then accessed by the program. The functions to read and write a load file are:

Write_loadfile does not free the table structure or modify it in any way. To free the table, the function free_table must be called.

All characters are converted to upper case when read into the table and superfluous blanks are deleted. Superfluous blanks are leading and trailing blanks and more than one blank between words. So when retrieving information from the table, specify keyword names and column names in upper case with only one blank between words.

The table functions implement an in-memory ASCII table. The table has header keywords accessed by name and cells accessed by column name and row number. The values returned by the get functions are pointers to the contents of the table and should not be modified by the calling function. Also, doing a put to the table may invalidate the pointers returned if memory is shuffled. Copy the results of a get to another string before doing a put. Table access is oriented towards random access, although sequential access is also supported. The get functions return a value of NULL if the get does not succeed.

For random access, function get_tab_key reads a header keyword and get_tab_cell reads a table cell. For sequential access, get_tab_keyname_bynum retrieves a header keyword name, get_tab_keyval_bynum retrieves a header keyword value, and get_tab_cell_bynum reads a table cell. The macros nkey_table, nrow_table, and ncol_table get the dimensions of a table.

Most often a table is created by reading the contents of a load file into it. However, it is possible to create a table in memory or modify the contents of a table read from a load file. The use of the add column routine is optional, as columns are implicitly added when the first row of the table is written. The add column routine can be used when the table column names are known before the table is created. The add column routine returns the number of the column created or -1 if the column cannot be created because the table is already written to.

The put functions return a value of -1 if the put does not succeed and 0 if it does. In addition to the error codes returned by the get and put functions, an error message is stored in the global variable lod_error.

The functions put_tab_key and put_tab_cell put a new value to the table header and body respective. The functions cat_tab_key and cat_tab_cell concatenate a value to an existing value in a table.

INSTALLATION DEPENDENCIES

In addition to these environment variables that control cdbs access, the stdb library that this task is linked with requires the two environment variables SYBASE and ACAREA be set. These variables define the location of the Sybase and stdb executables and libraries.

EXAMPLES