diff options
author | Mattias Andrée <maandree@kth.se> | 2021-02-23 01:11:51 +0100 |
---|---|---|
committer | Mattias Andrée <maandree@kth.se> | 2021-02-23 01:12:15 +0100 |
commit | 33d0ee66d80d664aec3511b1827049e538642658 (patch) | |
tree | fa3e0ae6a88ff139bb37923d348562b9d851420c /README | |
parent | update dist (diff) | |
download | median-a41833d227c13d41d6ca850bae876e9b1d242043.tar.gz median-a41833d227c13d41d6ca850bae876e9b1d242043.tar.bz2 median-a41833d227c13d41d6ca850bae876e9b1d242043.tar.xz |
Change license, rewrite in C, remove info manual, remove dist2.0
Signed-off-by: Mattias Andrée <maandree@kth.se>
Diffstat (limited to 'README')
-rw-r--r-- | README | 21 |
1 files changed, 15 insertions, 6 deletions
@@ -2,13 +2,25 @@ NAME median - Calculate the median values for a set of groups SYNOPSIS - median < INPUT + median DESCRIPTION median takes the first blank space-separated column values and the remainder as keys. It will then print the median for - each key. As a side effect median will print the result - sorted by the keys. + each key. Lines with the same key form a group. + + median outputs the median value for each key, no order of + the output lines are prescribed. No order is prescribed for + the input lines. + + Lines without a blank space are parsed as having the empty + string as the key, for lines with a blank space, the first + blank space is parsed as part of the key. + + For groups with an even number of elements, if the mean of + the middle two values are used as the median if all values + in the group are numerical, otherwise the lower value is + used as the median. EXAMPLES $ cat <<EOF | median @@ -38,6 +50,3 @@ RATIONALE SEE ALSO sort(1), expr(1), sets(1) - - Full documentation available locally via: info '(median)' - |