NAME
ce_db_merge - merges new information into a Classing Engine database from an
ASCII description of the new information.
SYNOPSIS
ce_db_merge user | system | network -from_ascii file-name \
[-db_file db-file-name]
DESCRIPTION
ce_db_merge merges new information into a Classing Engine database from an
ASCII description file.
The Classing Engine implements a simple hierarchy of databases. Each database
contains tables of namespaces. Each namespace table allows for namespace
specific mappings of names to attributes of names. For example, the Files
namespace allows for mappings from file names to file attributes e.g. file
type.
The ASCII description for merger has the same syntax as that used for
building a database from scratch (see ce_db_build (1)). An example
ASCII file is:
{
NS_NAME=Types
NS_ATTR= ((NS_MANAGER,string, <$CEPATH/tns_mgr.so>))
NS_ENTRIES= (
(
(TYPE_NAME,type-id,<binder-prog>)
(TYPE_ICON,icon-file,<$OPENWINHOME/include/images/bind.icon>)
(TYPE_ICON_MASK,icon-file,<$OPENWINHOME/include/images/bind.mask.icon>)
(TYPE_FGCOLOR,color,<91 229 229>)
(TYPE_BGCOLOR,color,<91 126 229>)
(TYPE_PRINT,string,<lpr -Plp>)
)(
(TYPE_NAME,type-id,<calctool-prog>)
(TYPE_ICON,icon-file,<$OPENWINHOME/include/images/calctool.icon>)
(TYPE_ICON_MASK,icon-file,<$OPENWINHOME/include/images/calctool.mask.icon>)
(TYPE_FGCOLOR,color,<255 255 255>)
(TYPE_BGCOLOR,color,<229 45 183>)
(TYPE_PRINT,string,<lpr -Plp>)
)
)
)
}
{
NS_NAME=Files
NS_ATTR=((NS_MANAGER,junk,<$CEPATH/fns_mgr.so>))
NS_ENTRIES=(
(
(FNS_TYPE,refto-Types,<binder-prog>)
(FNS_FILENAME,str,<binder>)
)(
(FNS_TYPE,refto-Types,<default-app>)
(FNS_FILENAME,str,<default|app>)
)
)
}
In the above example, the ASCII description file defines two namespaces
called Types and Files.
The keyword NS_NAME precedes the name of a namespace.
The keyword NS_ATTR precedes a list of the attributes of the namespace. An
important namespace attribute is called NS_MANAGER, which identifies the
name of the shared library (see dlopen (1)) that implements the namespace
manager for the namespace.
NS_ENTRIES precedes a list of entries in the namespace. Each entry consists
of a list of attributes.
An attribute has a name, a type and a value.
Attribute names in the example are in upper case e.g. TYPE_NAME, FNS_TYPE.
Attribute types are just strings that make sense to the CE client. For
example, FNS_TYPE has a type of "refto-Types" - an indication to the CE
client that FNS_TYPE values are really TYPE_NAMEs that can be looked up in
Types namespace.
A BNF for the ASCII description file is:
database : name_space | database name_space
name_space : { name namespace_attributes entries }
name : NS_NAME = variable
ns_attrs : NS_ATTR = ( attribute_list )
manager_attribute : ( NS_MANAGER , attribute_type , attribute_value )
attribute_list : attribute
| attribute_list attribute
attribute : ( attribute_name , attribute_type , attribute_value )
entries : NS_ENTRIES = ( list_of_entries )
list_of_entries : an_entry
| list_of_entries an_entry
an_entry : ( attribute_list )
attribute_name : variable
attribute_type : variable
variable : Identifier
attribute_value : attribute_token
The terminals are NS_NAME, NS_ATTR, NS_MANAGER, NS_ENTRIES,
"{", "}", "(", ")", "," "=", Identifier and attribute_token.
Identifier can have a-z, A-Z, 0-9, _, -.
attribute_token can come in 2 flavors:
- it can begin with a "<" and end with a ">" and can have any ASCII
character (except a ">") within it.
- it can begin with one or more digits (which represent a number n)
followed by zero or more spaces followed by a "<" followed by any n
characters closed off by a ">". This is the escape mechanism to
allow for arbitrary byte string attributes that could have ">"
characters within them.
If the database already had a namespace defined e.g. Types, the
new entries would just be appended to the existing namespace with no
checking for duplicate entries.
If the namespace description has some namespace attributes defined e.g.
NS_MANAGER that already exist in the database, the new namespace attributes
overwrite the existing attributes.
OPTIONS
user | system | network
indicates whether the user wants to access the user, system or
network CE database
-from_ascii file-name
indicates that the user wishes to write the stated CE database
from the ASCII file file-name. The entire CE database will be
re-written i.e. this is an all or nothing update of the CE database
-db_file db-file-name
should be used in the case that a particular database is to be read
from/written to using db-file-name as the pathname of the CE
database. The CE has the following default CE database file names:
if the database is:
"user" - ~/.cetables/cetables
"system" - /etc/cetables/cetables
"network" - $OPENWINHOME/lib/cetables/cetables
ENVIRONMENT
CEPATH
This is a colon separated list of pathnames that the CE uses when
looking for CE database when it is reading them. It also uses CEPATH
as a list of places to look if a particular NS_MANAGER filename has
CEPATH prepended to it.
FILES
cetables
This is the Classing Engine database file produced in the appropriate
directory after ce_db_merge is complete. A CE database file can
be identified by its first characters which state:
"Classing Engine Data File Version 1.0aaa"
SEE ALSO
ce_db_merge (1)
NOTES
None
BUGS
1. If the CE database file you are mergeing is mounted from a
pre-SunOS 4.1 machine, the locking protocol used by ce_db_merge
does not work i.e. the database does not get locked for writing.
2. Running ce_db_merge on an empty ASCII file causes it to hang
indefinitely as though it were in an infinite loop.