PINFO
Section: User Commands (1)
Updated: 01 Dec 2001
Index
Return to Main Contents
NAME
pinfo
- curses based lynx-style info browser
SYNTAX
pinfo
[options]
[infopage]
DESCRIPTION
This is a program for viewing info files. You specify which page you want to
read by passing it an
infopage
argument. This argument contains the name of an info page (i.e. 'bash').
The program will then (by default) search for it in the current directory,
/usr/share/info,
/usr/info,
/usr/local/share/info,
/usr/local/info.
and
/opt/info.
The search path can be adjusted by INFOPATH environment variable or in the configuration
file. Pinfo will also automatically add the suffix '-info', '-info.Z', '-info.gz', or '-info.bz2'.
At present other suffixes are not recognized, but you can easily add them to
the function
openinfo()
in
filehandling_functions.c.
When the search for info pages fails, man is called with the infopage
argument, and it's output is parsed by pinfo. This means that when you don't
have the appropriate info page, but have a man page instead; the man page
will be viewed.
When no infopage is specified, the default `dir' page is shown.
Supported options are
-h,
--help
- print help information and exit.
-v,
--version
- print version information and exit.
-m,
--manual
- uses manual page instead of info by default. (pinfo -m could be used as a
manual pager). Warning: Everything what follows this option is passed
to the `man' program. Don't be confused if pinfo options, which
followed `-m' don't work. When using this option, pinfo does not parse
the info options as usual! It invokes the man part of program.
You can also call the man function of pinfo in another way. When pinfo is
called with an argv[0] (the program file name), which contains the word 'man'
in it's name, the man functions are enabled automatically.
Previously there was a symlink to pinfo, called pman, but I had to remove it
from the distribution, since it's name was in conflict with some other
utility. Anyway, you can feel free to create such a link if you wish.
-r,
--raw-filename
- uses a raw filename first (i.e. the name which you specified as
infopage is considered to be a real file in the specified location).
-f,
--file
synonym for -r.
-a,
--apropos
- if this is set, apropos is called when no man or info page could be found.
-c,
--cut-man-headers
- if this is set, man parsing code will try to cut out the repeated man
headers. Use with care. ;)
-s,
--squeeze-lines-
cut empty lines from manual pages. This option enables auto cutting of every
repeated newline in a manual page.
-t,
--force-manual-tag-table-
forces manual detection of tag table. This allows you to view info pages,
which may be corrupted. (as i.e. version of jed's pages, shipped with
RH5.0). The tag table corruption usually appears in that the info links,
which you follow, move you to quite unexpected nodes.
--node=nodename,
--node nodename-
Go to the node `nodename' of info file. Since 0.6.7 it is also
possible to specify nodes as in standalone info via file names, like
`(gcc)Introduction'.
--rcfile=filename,
--rcfile filename-
Use alternate configuration file.
--long-manual-links,
-l-
Use long link names in manuals. On some systems the manual hierarchy is
divided into subsections like `3ncurses', etc, while on other systems all
belongs to section `3'. If this option is what your system is like, feel
free to use it.
--clear-at-exit,
-x-
Clear screen at exit.
The options are handled by GNU getopt, so you can here (as in other
programs) abbreviate the option names to the minimal number of
characters by which the options differ.
Warning! If you do not have getopt, these options will not work!
DEFAULT KEYS WHEN BROWSING INFO FILE
Just take a look at the example configuration file (below), and at the key
descriptions. Keys available in manual viewer differ a bit from the keys
available in info viewer.
ENVIRONMENT
There is a variable $INFOPATH, which can specify the paths to be searched
for info files. It's format is similar to that of the $PATH variable. An
example setting could look like:
/usr/info:/usr/somewhere/info:/not/even/in/usr/info
etc. Directories are separated by colons.
COLOR AND KEY DEFINITIONS
There are configuration files called ~/.pinforc and
[prefix]/etc/pinforc, for local and global configuration (where prefix is the
prefix of the directory, where pinfo is installed, i.e. /usr/local, or /).
Here's an example of such a file; we'll discuss the contents below:
# Here are some color setting.
# Whitespace between the entries is optional.
COL_NORMAL = COLOR_WHITE, COLOR_BLACK, NO_BOLD, NO_BLINK
COL_MENUSELECTED = COLOR_RED, COLOR_BLACK, BOLD, NO_BLINK
COL_MENU=COLOR_BLUE,COLOR_BLACK,BOLD, NO_BLINK
COL_NOTESELECTED=COLOR_RED,COLOR_BLACK,BOLD, NO_BLINK
COL_NOTE=COLOR_GREEN,COLOR_BLACK,BOLD, NO_BLINK
COL_TOPLINE=COLOR_YELLOW,COLOR_BLUE,BOLD, NO_BLINK
COL_BOTTOMLINE=COLOR_YELLOW,COLOR_BLUE,BOLD, NO_BLINK
COL_MANUALBOLD=COLOR_WHITE,COLOR_BLACK,BOLD, NO_BLINK
COL_MANUALITALIC=COLOR_WHITE,COLOR_BLACK,BOLD, NO_BLINK
COL_URL=COLOR_MAGENTA,COLOR_BLACK,BOLD, NO_BLINK
COL_URLSELECTED=COLOR_RED,COLOR_BLACK,NO_BOLD, NO_BLINK
COL_INFOHIGHLIGHT=COLOR_WHITE,COLOR_BLACK,BOLD, NO_BLINK
#
# Here are some keybindings as well...
#
KEY_TOTALSEARCH_1 = 's'
KEY_TOTALSEARCH_2 = 'S'
KEY_SEARCH_1 = '/'
KEY_SEARCH_2 = '.'
KEY_GOTO_1='g'
KEY_GOTO_2='m'
KEY_HOME_1='h'
KEY_HOME_2='H'
KEY_PREVNODE_1='p'
KEY_PREVNODE_2='P'
KEY_NEXTNODE_1='n'
KEY_NEXTNODE_2='N'
KEY_UP_1=KEY_UP
KEY_UP_2='u'
KEY_END_1=
KEY_END_2='e'
KEY_PGDN_1=KEY_NPAGE
KEY_PGDN_2=' '
KEY_PGDN_AUTO_1=0
KEY_PGDN_AUTO_2=' '
KEY_PGUP_1=KEY_PPAGE
KEY_PGUP_2='b'
KEY_PGUP_AUTO_1=0
KEY_PGUP_AUTO_2='b'
KEY_DOWN_1=KEY_DOWN
KEY_DOWN_2='d'
KEY_TOP_1=KEY_HOME
KEY_TOP_2='t'
KEY_BACK_1=KEY_LEFT
KEY_BACK_2='l'
KEY_FOLLOWLINK_1=KEY_RIGHT
KEY_FOLLOWLINK_2='
# 12 is a code for ctrl+l
KEY_REFRESH_1=12
KEY_REFRESH_2='~'
KEY_SHELLFEED_1='!'
KEY_SHELLFEED_2='1'
KEY_QUIT_1='q'
KEY_QUIT_2='Q'
KEY_DIRPAGE_1='d'
KEY_DIRPAGE_2='D'
KEY_GOLINE_1='l'
KEY_GOLINE_2=0
KEY_PRINT_1=']'
KEY_PRINT_2=0
#
# Some options, explained in the man page
#
MANUAL=false
CUT-MAN-HEADERS=true
CUT-EMPTY-MAN-LINES=true
RAW-FILENAME=false
APROPOS=false
DONT-HANDLE-WITHOUT-TAG-TABLE=false
LONG-MANUAL-LINKS=false
FILTER-0xB7=true
QUIT-CONFIRMATION=false
QUIT-CONFIRM-DEFAULT=no
CLEAR-SCREEN-AT-EXIT=true
STDERR-REDIRECTION="2> /dev/null"
HTTPVIEWER=lynx
FTPVIEWER=lynx
MAILEDITOR=pine
MANLINKS=1:8:2:3:4:5:6:7:9:n:p:o:3X11:3Xt
INFOPATH=/usr/info:/usr/share/info:/usr/local/info
HIGHLIGHTREGEXP=Bash.*has
SAFE-USER=nobody
SAFE-GROUP=nobody
As you can see, the format is simple. First I'll explain the color
definitions. First you must enter a color name (all available color
names are present in the example, and they're self explanatory, I
think. There is also a special color COLOR_DEFAULT, which stands for
transparency). Then you enter the foreground color, and the background
color. The BOLD attribute means that we want the foreground color to
be highlighted. (i.e. light blue, light green). BLINK attribute is the
blinking attribute, or highlighted background in some other configurations.
Now let's move to the key definitions. Here we first put a key name (again
all keys are present in the example); then we enter it's value -- either
surrounded
by apostrophes, or a keycode number (like in KEY_REFRESH_1), or its mnemonic
code name if it's a special key (like i.e. in KEY_FOLLOWLINK_1).
If you wish to specify key by code value, use the supplied program 'testkey'
to obtain the needed value. It mainly is a feature, when you want to add
some CTRL+letter keybindings, and similar.
For each function you can bind two keys, i.e. you could bind both
Enter and Cursor Right to the FollowLink-function. As you can see in
the example above, the two key names are KEY_FOLLOWLINK_1 and
KEY_FOLLOWLINK_2.
Here's an explanation of the key names:
-
- KEY_TOTALSEARCH_1
-
Key for searching through all nodes of info file.
- KEY_TOTALSEARCH_2
-
Alternate key for searching through all nodes of info file.
- KEY_SEARCH_1
-
Key for searching through current node (or manual).
- KEY_SEARCH_2
-
Alternate key for searching through current node (or manual).
- KEY_SEARCH_AGAIN_1
-
Key for repeating the last search.
- KEY_SEARCH_AGAIN_2
-
Alternate key for repeating the last search.
- KEY_GOTO_1
-
Key for explicitly going to a node (by specifying it's name).
- KEY_GOTO_2
-
Alternate key for explicitly going to a node (by specifying it's name).
- KEY_PREVNODE_1
-
Key for going to a node marked as 'Prev' in the header. In man page viewer
this goes to the previous man section.
- KEY_PREVNODE_2
-
Alternate key for going to a node marked as 'Prev' in the header. In man page
viewer this goes to the previous man section.
- KEY_NEXTNODE_1
-
Key for going to a node marked as 'Next' in the header. In man page viewer
this goes to the next man section.
- KEY_NEXTNODE_2
-
Alternate key for going to a node marked as 'Next' in the header. In man page
viewer this goes to the next man section.
- KEY_UP_1
-
Key for scrolling text one line up.
- KEY_UP_2
-
Alternate key for scrolling text one line up.
- KEY_END_1
-
Key for going to the end of the node.
- KEY_END_2
-
Alternate key for going to the end of the node.
- KEY_PGDN_1
-
Key for going one page down in the viewed node.
- KEY_PGDN_2
-
Alternate key for going one page down in the viewed node.
- KEY_PGDN_AUTO_1
-
Key for going to the next node when you're at the end of node (default
is zero -- turned off).
- KEY_PGDN_AUTO_2
-
Alternate key for going to the next node when you're at the end of node
(default is space, as for pgdn_2).
- KEY_HOME_1
-
Key for going to the beginning of the node.
- KEY_HOME_2
-
Alternate key for going to the beginning of the node.
- KEY_PGUP_1
-
Key for going one page up in the viewed node.
- KEY_PGUP_2
-
Alternate key for going one page up in the viewed node.
- KEY_PGUP_AUTO_1
-
Key for going to the `up' node, when being at the top of node. (Default value
is zero -- turned off).
- KEY_PGUP_AUTO_2
-
Alternate key for going to the `up' node, when being at the top of node.
(Default value is `-', as for pgup_2).
- KEY_DOWN_1
-
Key for scrolling the text down one line.
- KEY_DOWN_2
-
Alternate key for scrolling the text down one line.
- KEY_TOP_1
-
Key for going to the top (first) node.
- KEY_TOP_2
-
Alternate key for going to the top (first) node.
- KEY_BACK_1
-
Key for going back (in the history of viewed nodes).
- KEY_BACK_2
-
Alternate key for going back (in the history of viewed nodes).
- KEY_FOLLOWLINK_1
-
Key for following a hypertext link.
- KEY_FOLLOWLINK_2
-
Alternate key for following a hypertext link.
- KEY_REFRESH_1
-
Key for refreshing the screen (hard coded is the ^L value).
- KEY_REFRESH_2
-
Alternate key for refreshing the screen.
- KEY_SHELLFEED_1
-
Key for calling a shell command, and passing the viewed node to the stdin of
that command.
- KEY_SHELLFEED_2
-
Alternate key for calling a shell command, and passing the viewed node to the
stdin of that command.
- KEY_QUIT_1
-
Key for exiting the program.
- KEY_QUIT_2
-
Alternate key for exiting the program.
- KEY_GOLINE_1
-
Key for going to a specified line in file.
- KEY_GOLINE_2
-
Alternate key for going to a specified line in file.
- KEY_PRINT_1
-
Key for printing viewed node or man page.
- KEY_PRINT_2
-
Alternate key for printing viewed node or man page.
The special mnemonics for keys (which are defined at present) are:
-
- KEY_BREAK
-
- KEY_DOWN
-
- KEY_UP
-
- KEY_LEFT
-
- KEY_RIGHT
-
- KEY_DOWN
-
- KEY_HOME
-
- KEY_BACKSPACE
-
- KEY_NPAGE
-
- KEY_PPAGE
-
- KEY_END [Note: this works probably ONLY with Linux ncurses]
-
- KEY_F(x)
-
- KEY_CTRL('c')
-
- this assigns the key value to a ctrl+c combination. c may be
any letter you wish.
- KEY_ALT('c')
-
- this assigns the key value to a alt+c combination. c may be
any letter you wish. If alt key won't work, you may use ESC+key
combination.
- 'c'
-
- this means a printable character c. The syntax is just like in
C/C++ ;).
- [number]
-
- you can also specify key as it's code number. It is useful e.g. when
specifying control keys, and some nonstandard keys.
A numerical value of zero turns given key binding off.
See manual page for curs_getch (3x) for description of their meaning.
Warning! Try not to create some serious key binding conflicts!
The options in the last part of the example configuration file should
be fairly self-explanatory. The variables that can be set to true or
false do the same things as the command line arguments with the same
names.
-
- MANUAL
-
If this is set to
true
the default is to first check for a man page, instead of a texinfo
file.
- CUT-MAN-HEADERS
-
If set to true, then pinfo tries to cut off the repeated headers
throughout man pages.
- CUT-EMPTY-MAN-LINES
-
If set to true, then pinfo tries to cut off the repeated newlines (i.e. it
will shorten each set of consecutive newlines to one newline).
- RAW-FILENAME
-
If set to true, the file argument is taken to be the name of a file in
the current working directory, i.e. the directories in
INFOPATH
will only be searched if a file with this name is not in the working
directory.
- APROPOS
-
If set to true,
apropos
is called if no info or man page is found.
- DONT-HANDLE-WITHOUT-TAG-TABLE
-
If set to
true
, pinfo will not attempt to display texinfo pages
without tag tables.
- HTTPVIEWER
-
Set this to the program you want to use to follow http links in
documents.
- FTPVIEWER
-
Set this to the program you want to use to follow ftp links in
documents.
- MAILEDITOR
-
Set this to your favorite email program, and it will be started if
you follow an email link in a document.
- PRINTUTILITY
-
Utility, which you use for printing. I.e. `lpr'. If you don't use any, you
may also try something like `cat >/dev/lp1', or sth. ;)
- MANLINKS
-
This specifies the section names, which may be referenced in your man pages
(i.e. Xtoolkit man pages match the section 3Xt (see for example
XtVaCreateWidget)
manpage), Xlib function pages match section 3X11, etc. Such extensions may
not be recognized by default, so it is a good idea to add them).
- INFOPATH
-
This allows you to override the default search path for info pages. The paths
should be separated by colons.
- MAN-OPTIONS
-
This specifies the options, which should be passed to the `man' program.
(see man(1) for description of what they're like).
- STDERR-REDIRECTION
-
Pinfo allows you to redirect the stderr output of called programs. For
example if you don't want to see man's error messages about manual page
formatting, you can use STDER-REDIRECTION="2> /dev/null". This is the
default.
LONG-MANUAL-LINKS
-
This is another true/false option, which decides whether your system
supports long manual section names, or not. (i.e. "3ncurses" instead of "3").
- FILTER-0xB7
-
This decides, whether you want to convert 0xb7 chars to `o', or not. For
example for iso-8859-2 fonts this makes man's list marks a bit nicer ;)
(look for example at perl's man page, to see how those marks look like).
- QUIT-CONFIRMATION
-
This decides whether you want to use quit confirmation on exit, or not.
- QUIT-CONFIRM-DEFAULT
-
This yes/no option determines the default answer to the QUIT-CONFIRMATION
dialog. (default answer is when you press a key, that does not match the
asked question).
- CLEAR-SCREEN-AT-EXIT
-
This true/false option determines if you want to have your screen cleared at
exit, or no.
- CALL-READLINE-HISTORY
-
This true/false option determines if you want to have a prompt of last
history entry whenever calling readline wrapper, eg. in subsequent searches.
- HIGHLIGHTREGEXP
-
This is an option, through which you may pass to pinfo regexps, which should
be highlighted when working with document. Warning! This may turn very slow
if you use it without care!
- SAFE-USER
-
This option is used to pass the name of user, to which suid when pinfo is
run with root privileges.
- SAFE-GROUP
-
This option is used to pass the name of group, to which suid when pinfo is
run with root privileges.
INTERNATIONALIZATION SUPPORT
Pinfo implements general features of gnu gettext library (the thing, which
you need to see national messages ;). But it is not the end. Pinfo allows
you to use national info pages! You only need to put them to your info
directory, into a subdirectory, which is called `$LANG'.
LICENSE
This program is distributed under the terms of GPL.
BUGS
Please send bug reports to the author.
AUTHOR
Przemek Borys <pborys@dione.ids.pl>
If that E-mail address wont work (since the machine where it is being
handled is a bit damaged lately), you can try
pborys@zeus.polsl.gliwice.pl, or
pborys@p-soft.silesia.linux.org.pl.
There was also a lot of other people, who contributed to this code. See the
AUTHORS file.
COMMENTS
The author would like to read some comments and suggestions from you, if any.