Peter Dennis wrote: > The second is the "o" option which will allow the user to > skip fields in the order right to left. On the omission of > fields in random order, the alignment of the header is not > guaranteed. At all times, all fields may not be of interest > to the user. Under such scenarios, selective field data > can be obtained. The fields that are allowed for the "o" > option are: > kthr, memory, page, faults, cpu > > The same holds for the "-po" flags. > > The order of the displayed fields is determined by the > order they appear on the command line after the "-o".
I know it doesn't usually take much, but I'm confused :-) In particular, the phrase ... > .... skip fields in the order right to left. On the omission of > fields in random order, the alignment of the header is not > guaranteed. ... sounds like you are defining a "skip list", but later it seems you aren't. It also seems to say that there is no way to change the order of displayed fields. The comment about not guaranteeing header alignment also bothers me - it sounds more like a bug, and not an architectural requirement. I would have expected -o to follow the ps convention of defining a list of fields to display, and then displaying them (and their headers) in that order: > -o format Prints information according to the format > specification given in format. This is fully > described in DISPLAY FORMATS. Multiple -o > options can be specified; the format specifica- > tion is interpreted as the space-character- > separated concatenation of all the format > option-arguments. -John
