Date: Fri, 29 Mar 2013 21:00:01 GMT From: "Ronald F. Guilmette" <rfg@tristatelogic.com> To: freebsd-doc@FreeBSD.org Subject: Re: docs/177429: dd(1) man page is unclear about semantics of conv=sync Message-ID: <201303292100.r2TL01ZJ038962@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
The following reply was made to PR docs/177429; it has been noted by GNATS. From: "Ronald F. Guilmette" <rfg@tristatelogic.com> To: Peter Pentchev <roam@ringlet.net> Cc: bug-followup@FreeBSD.org Subject: Re: docs/177429: dd(1) man page is unclear about semantics of conv=sync Date: Fri, 29 Mar 2013 13:56:51 -0700 In message <20130329131835.GA15952@straylight.m.ringlet.net>, you wrote: >On Wed, Mar 27, 2013 at 08:21:11PM -0700, Ronald F.Guilmette wrote: >>=20 >> >Number: 177429 >> >Category: docs >> >Synopsis: dd(1) man page is unclear about semantics of conv=3Dsync >[snip] >> >Description: >>=20 >> The man page for dd(1) describes the "sync" parameter of the "conv=3D" op= >tion >> thusly: >>=20 >> sync Pad every input block to the input buffer size. S= >paces >> are used for pad bytes if a block oriented convers= >ion >> value is specified, otherwise NUL bytes are used. >>=20 >> This verbage is entirely unclear. What is a "block oriented conversion va= >lue" >> and how would a user know whether or not he had specified one? >>=20 >> The man page should make this more clear & explicit. > >Well, the "conversion value" is the set of all parameters that you have >supplied to "conv" - you can supply more than one using commas. And I >think that "block-oriented" should mean that some of the things that you >have specified say that the output should be in blocks - either that it >is in blocks by default (and you have *not* requested unblocking it by >using "ascii", "oldascii" or "unblock"), or that you've requested that >dd(1) make it into blocks using one of the "block", "ebcdic", "ibm", >"oldebcdic" and "oldibm" conversion specifiers. > >Of course, I could be wrong, but that's how I myself understand the >manual page. Thank you, but I think that the important point here is that the man page should be clear about all this, so that people can use the tool without guessing as to the semantics, and without having to rely on additional sources of information about the semantics (e.g. the source code). Regards, rfg
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201303292100.r2TL01ZJ038962>