Administrace komentářů

Ono je dobre vedet, ze less neni jen less. Napr. podle dokumentace

INPUT PREPROCESSOR
       You may define an "input preprocessor" for less.  Before less  opens  a
       file, it first gives your input preprocessor a chance to modify the way
       the contents of the file are displayed.  An input preprocessor is  sim‐
       ply  an executable program (or shell script), which writes the contents
       of the file to a different file, called the replacement file.  The con‐
       tents  of  the replacement file are then displayed in place of the con‐
       tents of the original file.  However, it will appear to the user as  if
       the  original  file  is opened; that is, less will display the original
       filename as the name of the current file.

A skutecne

$ set |grep LESS
LESS='-R -M --shift 5'
LESSCHARSET=utf-8
LESSOPEN='|lesspipe.sh %s'

Vtip je v tom, ze ze less ctouci data ze stdin zadne filtry neuplatnuje.

Jinak ten vas pripad mi prijde trochu divny. Ja kdyz mam manualovou stranku v UTF-8, tak mi ji zmrsi nroff, protoze zadny vstupni preprocessing type -mutf8 neumi (defaultni kodovani je totiz latin1). Naopak less se chova zcela korektne.

Kdyz se podivam do manualove strankym tak mi file rika, ze to neni v utf8

$ zcat /usr/share/man/cs/man1/dd.1.gz |file  -
/dev/stdin: ISO-8859 troff or preprocessor input text

A skutecne

$ zcat /usr/share/man/cs/man1/dd.1.gz | hexdump -C |head
00000000  2e 54 48 20 44 44 20 31  20 22 47 4e 55 20 46 69  |.TH DD 1 "GNU Fi|
00000010  6c 65 20 55 74 69 6c 69  74 69 65 73 22 20 22 46  |le Utilities" "F|
00000020  53 46 22 20 5c 22 20 2d  2a 2d 20 6e 72 6f 66 66  |SF" \" -*- nroff|
00000030  20 2d 2a 2d 0a 2e 64 6f  20 68 6c 61 20 63 73 0a  | -*-..do hla cs.|
00000040  2e 64 6f 20 68 70 66 20  68 79 70 68 65 6e 2e 63  |.do hpf hyphen.c|
00000050  73 0a 2e 53 48 20 4a 4d  c9 4e 4f 0a 64 64 20 5c  |s..SH JM.NO.dd \|
00000060  2d 20 6b 6f 6e 76 65 72  74 75 6a 65 20 73 6f 75  |- konvertuje sou|
00000070  62 6f 72 20 70 f8 69 20  6b 6f 70 ed 72 6f 76 e1  |bor p.i kop.rov.|
00000080  6e ed 0a 2e 53 48 20 50  4f 55 ae 49 54 cd 0a 2e  |n...SH POU.IT...|
00000090  42 20 64 64 0a 5b 5c 2d  5c 2d 68 65 6c 70 5d 20  |B dd.[\-\-help] |

Zamerte se na ř ve souslovi při kopírování.

Dobre je take prozkoumat manual k nroffu

       The nroff script emulates the nroff command using groff.   Only  ascii,
       latin1, utf8, and cp1047 are valid arguments for the -T option, select‐
       ing the output encoding emitted by grotty, groff’s TTY  output  device.
       If an invalid or no -T option is given, nroff checks the current locale
       to select a default output device.  It first tries the locale  program,
       then  the environment variables LC_ALL, LC_CTYPE, and LANG, and finally
       the LESSCHARSET environment variable.

Bohuzel o vstupnim kodovani se nikde nedoctete. Podle mych zkusenosti nepatchovany nroff predpoklada iso-8859-1, pokud nepouziji makro (parametr -m) pre samotnym formatovanim.