When dealing with various data formats in Sci2, column header names sometimes need to be standardized to a generic Sci2 format before algorithms can be run on the data. This process is called header standardization, and can be achieved using the Convert to Generic Publication Algorithm. This algorithm uses a specially formatted file, ending with the .hmap extension, to replace preexisting headers with the generic Sci2 equivalents.
The syntax of a .hmap file is identical to that of a Java .properties file. Line comments can be included, and should start with a #. The purpose of the file is to map out preexisting headers with their replacement headers. This is done using a key-value format, where the old headers are the keys, and the new generic headers are the values. Each line should contain a key, followed by a =, then a value. It is important to note that any spaces in a key (old header) should be represented by the Unicode space value - \u0020. Normal spaces can be used in the value (replacement headers), however. The reason for this is that, when the .hmap file is read in by Java (using .properties syntax interpretation), it will mistake any spaces before the equal sign as a key-value separator. Therefore, Unicode space values must be used to escape any spaces before the equal sign.
It may be hard to visualize what a .hmap file should look like without an example, so one is provided below. This is a sample .hmap file mapping SCOPUS format headers to their generic Sci2 equivalents: