[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3. Usage

This chapter describes the usage of EUDC. Most functions and customization options are available through the `Directory Search' submenu of the `Tools' submenu.

3.1 Querying Servers  How queries are performed and handled
3.2 Query Form  How to use and customize the query form
3.3 Display of Query Results  Controlling how query results are presented
3.4 Inline Query Expansion  How to use and customize inline queries
3.5 The Server Hotlist  How to use and manage the server hotlist
3.6 Multi-server Queries  How to query multiple servers successively
3.7 Creating BBDB Records  How to insert query results into your BBDB
3.8 Server/Protocol Locals  Customizing on a per server/protocol basis

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.1 Querying Servers

EUDC's basic functionality is to let you query a directory server and return the results back to you. There are several things you may want to customize in this process.

3.1.1 Selecting a Server  The first thing to do
3.1.2 Return Attributes  Configuring what the server should return
3.1.3 Duplicate Attributes  What to do when records have duplicate attributes

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.1.1 Selecting a Server

Before doing any query you will need to set the directory server. You need to specify the name of the host machine running the server software and the protocol to use. If you do not set the server in any fashion, EUDC will ask you for one when you make your first query.

You can set the server by selecting one from your hotlist of servers (see section 3.5 The Server Hotlist) available in the `Server' submenu or by selecting `New Server' in that same menu.

LDAP servers generally require some configuration before you can perform queries on them. In particular, the search base must be configured. If the server you select has no configured search base then EUDC will propose you to configure it at this point. A customization buffer will be displayed where you can edit the search base and other parameters for the server.

Variable: eudc-server
The name or IP address of the remote directory server. A TCP port number may be specified by appending a colon and a number to the name of the server. You will not need this unless your server runs on a port other than the default (which depends on the protocol). If the directory server resides on your own computer (which is the case if you use the BBDB backend) then `localhost' is a reasonable value but it will be ignored anyway.

Variable: eudc-protocol
The directory protocol to use to query the server. Currently supported protocols in this version of EUDC are ph, ldap and bbdb.

Command: eudc-set-server
This command accessible from `Server' submenu lets you specify a new directory server and protocol.

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.1.2 Return Attributes

Directory servers may be configured to return a default set of attributes for each record matching a query if the query specifies none. The variable eudc-default-return-attributes controls the return attributes you want to see, if different from the server defaults.

Variable: eudc-default-return-attributes
A list of the default attributes to extract from directory entries. If set to the symbol all then all available attributes are returned. A value of nil, the default, means to return the default attributes as configured in the server.

The server may return several matching records to a query. Some of the records may however not contain all the attributes you requested. You can discard those records.

User Option: eudc-strict-return-matches
If non-nil, entries that do not contain all the requested return attributes are ignored. Default is t.

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.1.3 Duplicate Attributes

Directory standards may authorize different instances of the same attribute in a record. For instance the record of a person may contain several email fields containing different email addresses. When using a QI directory server this is difficult to distinguish from attributes having multi-line values such as the postal address that may contain a line for the street and another one for the zip code and city name. In both cases, EUDC will consider the attribute duplicated.

EUDC has several methods to deal with duplicated attributes. The available methods are:

Makes a list with the different values of the duplicate attribute. The record is returned with only one instance of the attribute with a list of all the different values as a value. This is the default method that is used to handle duplicate fields for which no other method has been specified.
Discards all the duplicate values of the field keeping only the first one.
Concatenates the different values using a newline as a separator. The record keeps only one instance of the field the value of which is a single multi-line string.
Duplicates the whole record into as many instances as there are different values for the field. This is the default for the email field. Thus a record containing 3 different email addresses is duplicated into three different records each having a single email address. This is particularly useful in combination with select as the method to handle multiple matches in inline expansion queries (see section 3.4 Inline Query Expansion) because you are presented with the 3 addresses in a selection buffer

Because a method may not be applicable to all fields, the variable eudc-duplicate-attribute-handling-method lets you specify either a default method for all fields or a method for each individual field.

Variable: eudc-duplicate-attribute-handling-method
A method to handle entries containing duplicate attributes. This is either an alist (attr . method) or a symbol method. The alist form of the variable associates a method to an individual attribute name, the second form specifies a method applicable to all attribute names. Available methods are: list, first, concat, duplicate (see above). Defaults to list.

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.2 Query Form

The simplest way to query your directory server is to use the query form. You display the query form with the `Query with Form' menu item or by invoking the command M-x eudc-query-form. The attribute names presented in this form are defined by the eudc-query-form-attributes variable (unless a non-nil argument is supplied to eudc-query-form).

Since the different directory protocols to which EUDC interfaces may use different names for equivalent attributes, EUDC defines its own set of attribute names and a mapping between these names and their protocol-specific equivalent through the variable eudc-protocol-attributes-translation-alist. Names currently defined by EUDC are name, firstname, email and phone.

Variable: eudc-query-form-attributes
A list of attributes presented in the query form. Attribute names in this list should be either EUDC attribute names or valid attribute names. You can get a list of valid attribute names for the current protocol with the `List Valid Attribute Names' menu item or the M-x eudc-get-attribute-list command. Defaults to name, email and phone.

Command: eudc-query-form get-fields-from-server
Display a form to query the directory server. If given a non-nil argument the function first queries the server for the existing fields and displays a corresponding form. Not all protocols may support a non-nil argument here.

Since the names of the fields may not be explicit enough or adapted to be directly displayed as prompt strings in the form, the variable eudc-user-attribute-names-alist lets you define more explicit names for directory attribute names. This variable is ignored if eudc-use-raw-directory-names is non-nil.

Variable: eudc-user-attribute-names-alist
This is an alist of user-defined names for the directory attributes used in query/response forms. Prompt strings for attributes that are not in this alist are derived by splitting the attribute name at underscores and capitalizing the individual words.

Variable: eudc-use-raw-directory-names
If non-nil, use attributes names as defined in the directory. Otherwise, directory query/response forms display the user attribute names defined in eudc-user-attribute-names-alist.

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.3 Display of Query Results

Upon successful completion of a form query, EUDC will display a buffer containing the results of the query.

The fields that are returned for each record are controlled by eudc-default-return-attributes (see section 3.1.2 Return Attributes).

The display of each individual field can be performed by an arbitrary function which allows specific processing for binary values like images or audio samples as well as values with computer semantics like URLs.

Variable: eudc-attribute-display-method-alist
An alist specifying methods to display attribute values. Each member of the list is of the form (name . func) where name is a lowercased string naming a directory attribute (translated according to eudc-user-attribute-names-alist if eudc-use-raw-directory-names is non-nil) and func a function that will be passed the corresponding attribute values for display.

This variable has protocol-local definitions (see see section 3.8 Server/Protocol Locals). For instance, it is defined as follows for LDAP:

(eudc-protocol-set 'eudc-attribute-display-method-alist 
                   '(("jpegphoto" . eudc-display-jpeg-inline)
                     ("labeledurl" . eudc-display-url)
                     ("audio" . eudc-display-sound)
                     ("labeledurl" . eudc-display-url)
                     ("url" . eudc-display-url)) 

EUDC provides a set of built-in functions to display binary value types:

Function: eudc-display-generic-binary data
Display a button for unidentified binary data.

Function: eudc-display-url url
Display URL and make it clickable.

Function: eudc-display-sound data
Display a button to play the sound data.

Function: eudc-display-jpeg-inline data
Display the JPEG data inline at point if possible.

Function: eudc-display-jpeg-as-button data
Display a button for the JPEG data.

Right-clicking on a binary value button pops up a contextual menu with options to process the value. Among these are saving the attribute value to a file or sending it to an external viewer command. External viewers should expect the value on their standard input and should display it or perform arbitrary processing on it. Messages sent to standard output are discarded. External viewers are listed in the variable eudc-external-viewers which you can customize.

Variable: eudc-external-viewers
This is a list of viewer program specifications. Each specification is a list whose first element is a string naming the viewer for unique identification, the second element is the executable program which should be invoked and the following elements are arguments that should be passed to the program.

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.4 Inline Query Expansion

Inline query expansion is a powerful method to get completion from your directory server. The most common usage is for expanding names to email addresses in mail message buffers. The expansion is performed by the command M-x eudc-expand-inline which is available from the `Directory Search' menu but can also be conveniently bound to a key shortcut (see section 2. Installation) The operation is controlled by the variables eudc-inline-expansion-format, eudc-inline-query-format, eudc-expanding-overwrites-query and eudc-multiple-match-handling-method.

If the query fails for a server, other servers may be tried successively until one of them finds a match (see section 3.6 Multi-server Queries).

Command: eudc-expand-inline replace-p
Query the server and expand the query string before point. The query string consists of the buffer substring from the point back to the preceding comma, colon or beginning of line. eudc-inline-query-format controls how individual words are mapped onto directory attribute names. After querying the server for the given string, the expansion specified by eudc-inline-expansion-format is inserted in the buffer at point. If replace-p is t then this expansion replaces the query string in the buffer. If eudc-expanding-overwrites-query is non-nil then the meaning of replace-p is negated.

Variable: eudc-inline-query-format
Format of an inline expansion query. This is actually a list of formats. A format is a list of one or more EUDC attribute names. A format applies if it contains as many attributes as individual words in the inline query string. If several formats apply then they are tried in order until a match is found. If nil all the words will be mapped onto the default server/protocol attribute name (generally name).

For instance, use the following

(setq eudc-inline-query-format '((name)
                                 (firstname name)))
to indicate that single word expansion queries are to be considered as surnames and if no match is found then they should be tried as first names. Inline queries consisting of two words are considered as consisting of a first name followed by a surname. If the query consists of more than two words, then the first one is considered as the first name and the remaining words are all considered as surname constituents.

formats are in fact not limited to EUDC attribute names, you can use server or protocol specific names in them. If you do so it is highly recommended to set the variable eudc-inline-query-format in a protocol or server local fashion (see see section 3.8 Server/Protocol Locals).

For instance you could use the following to match up to three words against the cn attribute of LDAP servers:

(eudc-protocol-set 'eudc-inline-query-format
                     (cn cn)
                     (cn cn cn))

Variable: eudc-inline-expansion-format
This variable lets you control exactly what is inserted into the buffer upon an inline expansion request. It is a list whose first element is a string passed to format. Remaining elements are symbols corresponding to directory attribute names. The corresponding attribute values are passed as additional arguments to format. Default is ("%s" email) but you may want to consider a value like ("%s <%s>" name email)

Variable: eudc-multiple-match-handling-method
This variable controls what to do when multiple entries match a query for an inline expansion. Possible values are:
The first match is considered as being the only one, the others are discarded.
A selection buffer pops up where you can choose a particular match. This is the default value of the variable.
The expansion uses all records successively
An error is signaled. The expansion aborts.

Defaults to select

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.5 The Server Hotlist

EUDC lets you maintain a list of frequently used servers so that you can easily switch from one to another. This hotlist appears in the `Server' submenu. You select a server in this list by clicking on its name. You can add the current server to the list with the command M-x eudc-bookmark-current-server. The list is contained in the variable eudc-server-hotlist which is stored in and retrieved from the file designated by eudc-options-file. EUDC also provides a facility to edit the hotlist interactively (see section 3.5.1 The Hotlist Edit Buffer).

The hotlist is also used to make queries on multiple servers successively (see section 3.6 Multi-server Queries). The order in which the servers are tried is the order they appear in the hotlist, therefore it is important to sort the hotlist appropriately.

Command: eudc-bookmark-server server
Add server to the hotlist of servers

Command: eudc-bookmark-current-server
Add the current server to the hotlist of servers

Variable: eudc-options-file
The name of a file where EUDC stores its internal variables (the hotlist and the current server). EUDC will try to load that file upon initialization so, if you choose a file name different from the defaults `~/.eudc-options', be sure to set this variable to the appropriate value before EUDC is itself loaded.

3.5.1 The Hotlist Edit Buffer  An interactive hotlist editing facility

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.5.1 The Hotlist Edit Buffer

The hotlist edit buffer offers a means to manage a list of frequently used servers. Commands are available in the context pop-up menu generally bound to the right mouse button. Those commands also have equivalent keybindings.

Command: eudc-hotlist-add-server
Bound to a. Add a new server to the hotlist on the line after point

Command: eudc-hotlist-delete-server
Bound to d. Delete the server on the line point is on

Command: eudc-hotlist-select-server
Bound to s. Select the server the point is on as the current directory server for the next queries

Command: eudc-hotlist-transpose-servers
Bound to t. Bubble up the server the point is on to the top of the list

Command: eudc-hotlist-quit-edit
Bound to q. Save the changes and quit the hotlist edit buffer. Use x or M-x kill-buffer to exit without saving.

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.6 Multi-server Queries

When performing inline or form queries, EUDC can try to contact a sequence of directory servers. When performing form queries EUDC tries all the servers specified by the variable eudc-multi-query-policy and displays all the matching records. When performing inline expansion queries however, EUDC stops trying different servers as soon as a match has been found on one of them.

Variable: eudc-multi-query-policy
This variable the policy for querying multiple servers. Possible values are:
Only the current directory server is tried
Up to eudc-max-servers-to-query servers in the hotlist are tried in order
The current server then Up to eudc-max-servers-to-query-1 servers in the hotlist are tried. This is the default.

Variable: eudc-max-servers-to-query
This variable indicates the maximum number of servers to query when performing a multi-server query. The default, nil, indicates that all available servers should be tried.

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.7 Creating BBDB Records

With EUDC, you can automatically create BBDB records (see section `BBDB' in BBDB Manual) from records you get from a directory server. You do this by moving point to the appropriate record in a query result display buffer and invoking the command M-x eudc-insert-record-at-point-into-bbdb with the keyboard binding b (1), or with the menu. EUDC cannot update an existing BBDB record and will signal an error if you try to insert a record matching an existing one.

It is also possible to export to BBDB the whole batch of records contained in the directory query result with the command M-x eudc-batch-export-records-to-bbdb.

Because directory systems may not enforce a strict record format, local server installations may use different attribute names and have different ways to organize the information. Furthermore BBDB has its own record structure. For these reasons converting a record from its external directory format to the BBDB format is a highly customizable process.

Variable: eudc-bbdb-conversion-alist
The value of this variable should be a symbol naming an alist defining a mapping between BBDB field names onto directory attribute names records. This is a protocol-local variable and is initialized upon protocol switch (see section 3.8 Server/Protocol Locals) The alist is made of cells of the form (bbdb-field . spec-or-list). bbdb-field is the name of a field that must be defined in your BBDB environment (standard field names are name, company, net, phone, address and notes). spec-or-list is either a single mapping specification or a list of mapping specifications. Lists of mapping specifications are valid for the phone and address BBDB fields only. specs are actually s-expressions which are evaluated as follows:

a string
evaluates to itself
a symbol
evaluates to the symbol value. Symbols corresponding to directory attribute names present in the record evaluate to the value of the field in the record
a form
is evaluated as a function. The argument list may contain attribute names which evaluate to the corresponding values in the record. The form evaluation should return something appropriate for the particular bbdb-field (see bbdb-create-internal). eudc-bbdbify-phone and eudc-bbdbify-address are provided as convenience functions to parse phones and addresses.

The default value of the PH-specific value of that variable is eudc-ph-bbdb-conversion-alist:

((name . name)
 (net . email)
 (address . (eudc-bbdbify-address address "Address"))
 (phone . ((eudc-bbdbify-phone phone "Phone")
           (eudc-bbdbify-phone office_phone "Office Phone"))))

This means that:

Function: eudc-bbdbify-phone phone location
This is a convenience function provided for use in eudc-bbdb-conversion-alist. It parses phone into a vector compatible with bbdb-create-internal. phone is either a string supposedly containing a phone number or a list of such strings which are concatenated. location is used as the phone location for BBDB.

Function: eudc-bbdbify-address addr location
This is a convenience function provided for use in eudc-bbdb-conversion-alist. It parses addr into a vector compatible with bbdb-create-internal. addr should be an address string of no more than four lines or a list of lines. The last line is searched for the zip code, city and state name. location is used as the phone location for BBDB.

Note that only a subset of the attributes you selected with eudc-default-return-attributes and that are actually displayed may actually be inserted as part of the newly created BBDB record.

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.8 Server/Protocol Locals

EUDC can be customized independently for each server or directory protocol. All variables can be given local bindings that are activated when a particular server and/or protocol becomes active. This is much like buffer-local bindings but on a per server or per protocol basis.

3.8.1 Manipulating local bindings  Functions to set and query local bindings

[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.8.1 Manipulating local bindings

EUDC offers functions that let you set and query variables on a per server or per protocol basis.

The following predicates allow you to test the existence of server/protocol local bindings for a particular variable.

Function: eudc-server-local-variable-p var
Return non-nil if var has server-local bindings

Function: eudc-protocol-local-variable-p var
Return non-nil if var has protocol-local bindings

The following functions allow you to set the value of a variable with various degrees of localness.

Function: eudc-default-set var val
Set the EUDC default value of var to val. The current binding of var (if local to the current server or protocol) is not changed.

Function: eudc-protocol-set var val &optional protocol
Set the binding of var local to protocol to val. If omitted, protocol defaults to the current value of eudc-protocol. The current binding of var is changed only if protocol is omitted.

Function: eudc-server-set var val &optional server
Set the binding of var local to server to val. If omitted, server defaults to the current value of eudc-server. The current binding of var is changed only if server is omitted.

Function: eudc-set var val
Set the most local (server, protocol or default) binding of var to val. The current binding of var is also set to val.

The following variables allow you to query the various bindings of a variable (local or non-local).

Function: eudc-variable-default-value var
Return the default binding of var (outside of a particular server or protocol local binding). Return unbound if var has no EUDC default value.

Function: eudc-variable-protocol-value var &optional protocol
Return the value of var local to protocol. Return unbound if var has no value local to protocol. protocol defaults to eudc-protocol.

Function: eudc-variable-server-value var &optional server
Return the value of var local to server. Return unbound if var has no value local to server. server defaults to eudc-server.

Changing a protocol-local or server-local value of a variable has no effect on its current value. The following command is used to synchronize the current values of variables with their local values given the current eudc-server and eudc-protocol:

Function: eudc-update-local-variables
Update all EUDC variables according to their local settings.

[ << ] [ >> ]           [Top] [Contents] [Index] [ ? ]

This document was generated by XEmacs Webmaster on October, 2 2007 using texi2html