rlam - laminate records from multiple files
rlam [ -tS ][ -u ][ -iaN | -ifN | -idN | -iiN | -iwN | -ibN ][ -in M ] input1 [options] input2 ..
Rlam simply joins records (or lines) from multiple inputs, separating them with the given string (TAB by default). Different separators may be given for different files by specifying additional -t options in between each file name. Note that there is no space between this option and its argument. If no argument is given, then a newline is assumed. If none of the input files uses an ASCII separator, then no end-of-line character will be printed, either.
An input is either a file or a command. Commands are given in quotes, and begin with an exclamantion point (’!’). If the inputs do not have the same number of records, then shorter files will stop contributing to the output as they run out.
The -ia option may be used to specify ASCII input (the default), or the -if option may be used to indicated binary IEEE 32-bit floats on input. Similarly, the -id and -ii options may be used to indicate binary 64-bit doubles or integer words, respectively. The -iw option specifies 2-byte short words, and the -ib option specifies bytes. If a number immediately follows any of these options, then it indicates that multiple such values are expected for each record. For example, -if3 indicates three floats per input record for the next named input. In the case of the -ia option, no number reads one line per record, and numbers greater than zero take multiple lines at a time from the same stream. For binary input formts, no number implies one value per record. For anything other than EOL-separated input, the default tab separator is reset to the empty string.
These options may be interspersed with input specifications, indicating different data to draw from the different files.
If a -in option is given with a positive integer argument, this will be the maximum number of records that will be read or produced by rlam.
A hyphen (’-’) by itself can be used to indicate the standard input, and may appear multiple times. The -u option forces output after each record (i.e., one run through inputs).
To join files output1 and output2, separated by a comma:
rlam -t, output1 output2
To join a file with line numbers (starting at 0) and its reverse:
cnt ‘wc -l < lam.c‘ | rlam - -t: lam.c -t ’!tail -r lam.c’
To join four data files, each having three doubles per record:
rlam -id3 file1.dbl file2.dbl file3.dbl file4.dbl > combined.dbl
To join two data files, the first having two words per record and the second having one:
rlam -iw2 file1.hi2 -iw file2.hi1 > combined.hi3
cnt(1), histo(1), neaten(1), rcalc(1), rsplit(1), tabfunc(1), total(1)