dennis@rlgvax.UUCP (Dennis Bednar) (07/04/86)
The -inum option of find(1) which finds files based on i-node number is undocumented in the man pages. By the way, what does the -depth option of find do? The man page for cpio attempts to explain what the -depth option of find does, but I'm confused. Also the man page for find doesn't describe -depth either. -- -Dennis Bednar {decvax,ihnp4,harpo,allegra}!seismo!rlgvax!dennis UUCP
gwyn@brl-smoke.ARPA (Doug Gwyn ) (07/05/86)
In article <1036@rlgvax.UUCP> dennis@rlgvax.UUCP (Dennis Bednar) writes: >By the way, what does the -depth option of find do? Forces a depth-first traversal of the directory hierarchy. The main value of this is that a later restore using cpio can handle read-only diretories properly, since they'll not be made read-only until all their files are in place. >Also the man page for find doesn't describe -depth either. Whose manual? I think it's in the current one from AT&T.
geoff@desint.UUCP (Geoff Kuenning) (07/08/86)
In article <1036@rlgvax.UUCP> dennis@rlgvax.UUCP (Dennis Bednar) writes: > The -inum option of find(1) which finds files based on > i-node number is undocumented in the man pages. This is true. However, the BUGS section should mention that, for almost all applications, ncheck(1m) is a much better way to locate a file based on its i-node number. > By the way, what does the -depth option of find do? > The man page for cpio attempts to explain what the -depth > option of find does, but I'm confused. > > Also the man page for find doesn't describe -depth either. The -depth option changes the behavior of "find / -print" so that the name of a directory is printed *after* the contents, instead of before. In data-structure terminology, it changes the directory tree search from preorder to postorder. It is useful with cpio because, with the -m switch, it causes the directory's modification time to be updated *after* all files have been placed in it--otherwise, placing the next file would wipe out that carefully-preserved mod time. -- Geoff Kuenning {hplabs,ihnp4}!trwrb!desint!geoff
mikel@codas.UUCP (07/09/86)
> By the way, what does the -depth option of find do? > The man page for cpio attempts to explain what the -depth > option of find does, but I'm confused. > > Also the man page for find doesn't describe -depth either. > > -Dennis Bednar > {decvax,ihnp4,harpo,allegra}!seismo!rlgvax!dennis UUCP From my man page from "AT&T 3B2 Computer Unix System V User Reference Manual" " -depth Always true; causes descent of the directory hierarchy to be done so that all entries in a directory are acted on before the directory itself. This can be helpful when _f_i_n_d is used with _c_p_i_o_(_1_) to transfer files that are contained in directories without write permission." -- ___ / \ Mikel Manitius @ AT&T-IS Altamonte Springs, FL | RPI | ...{seismo!akgua|ihnp4|cbosgd|mcnc}!codas!mikel | . | \\-------//
guy@sun.UUCP (07/10/86)
> This is true. However, the BUGS section should mention that, for almost > all applications, ncheck(1m) is a much better way to locate a file based > on its i-node number. Only if you have read permission on the partition containing the file system you're searching - which normal users will not have if the system administrator is sane. In most cases, it is the administrator who's searching for that file, so this is not a problem. (Then again, I can't remember the last time I searched for a file by i-number anyway.) -- Guy Harris {ihnp4, decvax, seismo, decwrl, ...}!sun!guy guy@sun.com (or guy@sun.arpa)
glen@proexam.UUCP (07/14/86)
In article <625@codas.ATT.UUCP> mikel@codas.UUCP writes: > >" -depth Always true; causes descent of the directory hierarchy to > be done so that all entries in a directory are acted on > before the directory itself. This can be helpful when _f_i_n_d > is used with _c_p_i_o_(_1_) to transfer files that are contained > in directories without write permission." >-- Another use for this option is to clean out your disk directory hierarchy of directories, empty except for other directories. find / -depth -type d -exec rmdir >& /dev/null or better: find / -depth -type d -print | xargs rmdir >& /dev/null
davidsen@steinmetz.UUCP (Davidsen) (08/04/86)
In article <232@desint.UUCP> geoff@desint.UUCP (Geoff Kuenning) writes: >In article <1036@rlgvax.UUCP> dennis@rlgvax.UUCP (Dennis Bednar) writes: > >> By the way, what does the -depth option of find do? >> The man page for cpio attempts to explain what the -depth >> option of find does, but I'm confused. >> >> Also the man page for find doesn't describe -depth either. > >The -depth option changes the behavior of "find / -print" so that the >name of a directory is printed *after* the contents, instead of before. >In data-structure terminology, it changes the directory tree search from >preorder to postorder. It is useful with cpio because, with the -m switch, >it causes the directory's modification time to be updated *after* all files >have been placed in it--otherwise, placing the next file would wipe out that >carefully-preserved mod time. I learn something every day! I admit to not using the mod time on directories much, but I never thought of this. I always thought it was to allow restoring a directory which is read-only, by creating it by default (-d) and then fixing the permissions later. I have a few things in directories which are mode 555 and don't create well on a new system without -depth. Hope someone else learned something, too. -- -bill davidsen ihnp4!seismo!rochester!steinmetz!--\ \ unirot ------------->---> crdos1!davidsen chinet ------/ sixhub ---------------------/ (davidsen@ge-crd.ARPA) "Stupidity, like virtue, is its own reward"
simon@einode.UUCP (Simon Kenyon) (08/11/86)
> By the way, what does the -depth option of find do? > The man page for cpio attempts to explain what the -depth > option of find does, but I'm confused. > > Also the man page for find doesn't describe -depth either. they have been missing from the manual for an awfully long time. (back when system V was just plain ole system V - no releases, no versions, no technologies) surprised this hasen't cropped up before :-) -- Simon Kenyon The National Software Centre, Dublin, IRELAND simon@einode.UUCP +353-1-716255 EEEK /dev/mouse escaped (Glad to see my competition went down well at USENIX)