ENVIRON(5) SysV ENVIRON(5)
NAME
environ - user environment
DESCRIPTION
An array of strings called the "environment" is made available by exec(2)
when a process begins. By convention, these strings have the form
"name=value." The following names are used by various commands:
APOLLO_STATUS When set to "true," extends system error reporting. Use
perror(3C) to print the more verbose, Domain/OS-specific
error messages.
HOME Name of the user's login directory, set by login(1) from
the password file (see passwd(4)).
ISP Instruction set processor type. The value of this
variable determines the resolution of variant links in
/usr/lib having names of the general form
linkname -> isp_$(ISP)/filename
The linknames are those of the various bindable libraries
and C initialization routines used by ld and f77:
crt0.o, libF77.a, libld.a, and so on. When ISP=m68k,
these variant links resolve to files of the corresponding
names in /usr/lib/isp_m68k, a directory containing
bindable libraries and C initialization routines for
M680X0-based workstations. When ISP=a88k, these variant
links resolve to files of the corresponding names in
/usr/lib/isp_a88k, a directory containing bindable
libraries and C initialization routines for the Series
10000.
ld and f77 set ISP according to the option you specify to
the -cpu switch, allowing you to link and compile either
for M680X0-based workstations (-cpu m68k) or for the
Series 10000 (-cpu a88k).
LANG A value that specifies requirements for a language-
specific environment; enables users to select desired
native language. LANG defines language, territory, and
codeset. A unique value of LANG is defined for each
supported language/territory/codeset combination. Each
LANG setting includes instances of collating sequence,
character conversion, character classification and
langinfo tables and message catalogs.
LANG uses the Domain/OS locale naming convention and
specifies the language, territory and codeset as follows:
language[_territory[.codeset]]
where the length of the entire string should not exceed
[NL_LANGMAX] characters.
Either an individual user or a system administrator can
set LANG. Typically, system administrators set up the
default locale for an entire site. Users can choose to
override the default.
Domain/OS provides the following Latin-1, iso88591-based
locales:
___________________________________
|Locale | Language |
___________________________________|
_da_DK.iso88591___danish___________|
|de_DE.iso88591 | german |
___________________________________|
_en_GB.iso88591___british_english__|
|en_US.iso88591 | american english |
___________________________________|
_es_ES.iso88591___spanish__________|
|fi_FI.iso88591 | finnish |
___________________________________|
_fr_FR.iso88591___french___________|
|fr_CA.iso88591 | canadian french |
___________________________________|
_is_IS.iso88591___icelandic________|
|it_IT.iso88591 | italian |
___________________________________|
_nl_NL.iso88591___dutch____________|
|no_NO.iso88591 | norwegian |
___________________________________|
_pt_PT.iso88591___portuguese_______|
|sv_SE.iso88591 | swedish |
___________________________________|
LC Variables Although assigning a value to LANG is the most common way
to set a locale, there may be times when you want to
assign a particular value to a smaller piece of it. The
following variables define these more specific categories
of native language support:
+ LC_COLLATE - Value that controls collation
+ LC_CTYPE - Value that controls character classification
(ctype functions)
+ LC_NUMERIC - Value that controls numeric formatting
+ LC_MONETARY - Value that controls monetary formatting
+ LC_TIME - Value that controls date and time
As with LANG, all of these categories are environment
variables to which you can assign locale names. However,
you can add an additional field (@modifier) to names for
these categories. This allows you to select a specific
version of locale-specific data.
For example, a locale might sort data two ways: in
dictionary order and in telephone-book order. To set up a
locale that provides standard dictionary order and
permits telephone-book order, one way to set the
environment variables is:
LANG=fr_FR.iso88591
LC_COLLATE=fr_FR.iso88591@phone
The explicit setting of LC_COLLATE overrides the implicit
setting of LANG for that portion of the locale.
NLSPATH Value that specifies the location of locale-specific
message catalogs. Message catalogs are native language
message strings that programs use. These catalogs are
located in /usr/nlslib/<locale>/name.cat
Domain/OS sets its default value for NLSPATH as follows:
NLSPATH=/usr/nlslib/%L/%N.cat
where %L gets filled in with the current locale name, and
%N gets filled in with the value of <name>.
It is recommended that both LANGand NLSPATH be defined in
a startup script along with the other user specified
environment variables. (For example, the user might put
them in /etc/profile and export the environment variables
after specifying them.)
To set the locale for the English/ISO 88591 - Latin 1
code set (default), add the following lines:
LANG=en_US.iso88591
NLSPATH=/usr/nlslib/%L/%N.cat:/usr/nlslib/%N/%L
NLSPATH, in the case above, is set for catopen(3c) calls
to look for the named catalog and referenced by %N, in:
/usr/nlslib/<LANG env var>/<named catalog>.cat
and if not found, then in:
/usr/nlslib/<named catalog>/<LANG env var>
NODEID Hexadecimal ID of current network node (4DD0, for
example).
NODETYPE Model of the Apollo workstation on which the process is
running (DN4500, for example).
ORGANIZATION Organization associated with log-in name.
PATH The sequence of directory prefixes that sh(1), time(1),
nice(1), nohup(1), etc., apply in searching for a file
known by an incomplete pathname. The prefixes are
separated by colons (:). login(1) sets
PATH=:/bin:/usr/bin:/usr/apollo/bin.
PROJECT Project associated with log-in name.
SYSTYPE Environment as set by "systype".
TERM The kind of terminal for which output is to be prepared.
This information is used by commands such as tplot(1G),
which may exploit special capabilities of that terminal.
TZ Time zone information. The simplest format is xxxnzzz
where xxx is standard local time zone abbreviation, n is
the difference in hours from GMT (Greenwich Mean Time),
and zzz is the abbreviation for an alternate time zone
(usually the daylight-saving local time zone), if any;
for example,
TZ="EST5EDT"
The most complex format allows you to specify the
difference in hours of the alternate time zone from GMT
and the starting day and time and ending day and time for
using this alternate time zone. For example, in 1985 the
complex format corresponding to the above example is:
TZ="EST5:00:00EDT4:00:00;118/2:00:00,300/2:00:00"
When the complex format is used, it must be surrounded by
double quotation marks. See ctime(3C) and timezone(4).
Variables Used by Obsolete Calls
The CFTIME, CHRCLASS, and LANGUAGE environment variables are used by
obsolete routines to specify format strings and language-specific
character classification and conversion data. The cftime(3C),
ascftime(3C), routines are obsoleted by standard compliant strftime(3C)
call. And the and setchrclass(3C) routine is obsoleted by standard
compliant setlocale(3C) call.
We strongly recommend:
+ that you use strftime(3C) in place of cftime(3C) and ascftime(3C).
+ that you use setlocale(3C) in place of setchrclass(3C).
+ that you use LANG, LC Variables, and NLSPATH environment variables (see
above) to specify requirements for native language support.
CFTIME
The default format string to be used by the date(1) command and the
ascftime and cftime routines (see ctime(3C)). If CFTIME is not set
or is null, the default format string specified in the
/usr/lib/cftime/language file is used in its place (if it exists),
where language is the value to which the environment variable
LANGUAGE is set. See cftime(4).
CHRCLASS
A value corresponding to the name of a file in /usr/lib/chrclass
containing language-specific character classification and conversion
data. Programs can initialize the tables used by the ctype(3C)
macros to the data contained in the file corresponding to the
CHRCLASS value by calling setchrclass with its argument set to (char
*)0. If CHRCLASS is non-existent, null, set to a value for which no
file exists in /usr/lib/chrclass, or an error occurs while reading
the file, the ascii character set is used. Programs that check the
value of CHRCLASS include cat(1), ed(1), egrep(1), grep(1), sed(1),
and sort(1).
Programs can initialize character classification and conversion data
to a specific language by calling setchrclass with the appropriate
argument. See ctype(3C). See chrtbl(1M) for information on
creating language-specific character classification and conversion
tables.
LANGUAGE
A language for which a format file by that name exists in
/usr/lib/cftime. This information is used by commands such as
date(1), ls(1), sort(1), etc. to print date and time information in
the language specified.
If LANGUAGE is not set, null, or set to a value for which no file
exists in /usr/lib/cftime, /or an error occurs while reading the
file, the last language requested will be used. If no language has
been requested, usa_english is the default. See cftime(4).
Additional Names
Further names may be placed in the environment by the export command and
"name=value" arguments in sh(1), or by exec(2). It is unwise to conflict
with certain shell variables that are frequently exported by .profile
files: MAIL, PS1, PS2, IFS (see profile(4)).
SEE ALSO
ctime(3C), buildlang(1m), setlocale(3C), cftime(4), passwd(4),
profile(4), timezone(4), exec(2).
env(1), f77(1), ld(1), login(1), sh(1), nice(1), nohup(1), time(1),
tplot(1G) buildlang(1m) in the SysV Command Reference.
chrtbl(1M) in Managing SysV System Software.
For more information on Internationalization environment variables refer
to Domain System Software Release Notes, Software Release 10.4.
NOTES
Setting the time to be within the interval between the end of the main
time zone and the beginning of the alternate time zone, or vice versa, is
not recommended. (For example, on the morning Daylight Savings Time is
scheduled to begin, don't set the time to be within the hour "lost").