Ubuntu Manpage: yaz-client - Z39.50/SRU client for implementors

NAME

       yaz-client - Z39.50/SRU client for implementors

SYNOPSIS

       yaz-client [-a apdulog] [-b berdump] [-c cclfile] [-d dump] [-f cmdfile] [-k size]
                  [-m marclog] [-p proxy-addr] [-q cqlfile] [-t dispcharset] [-u auth]
                  [-v loglevel] [-V] [-x] [server-addr]

DESCRIPTION

       yaz-client is a Z39.50[1]/SRU[2] client (origin) with a simple command line interface that
       allows you to test behavior and performance of Z39.50 targets and SRU servers.

       From YAZ version 4.1.0 yaz-client may also operate as a Solr[3] Web Service client.

       If the server-addr is specified, the client creates a connection to the Z39.50/SRU target
       at the address given.

       When yaz-client is started it tries to read commands from one of the following files:

       •   Command file if it is given by option -f.

       •   .yazclientrc in current working directory.

       •   .yazclientrc in the user's home directory. The value of the $HOME environment variable
           is used to determine the home directory. Normally, $HOME is only set on POSIX systems
           such as Linux, FreeBSD, Solaris.

OPTIONS

-a filename
If specified, logging of protocol packages will be appended to the file given. If
filename is specified as -, the output is written to stdout.

-b filename
If specified, YAZ will dump BER data in readable notation to the file specified. If
filename is specified as - the output is written to stdout.

-c filename
If specified, CCL configuration will be read from the file given.

-d dump
If specified, YAZ will dump BER data for all PDUs sent and received to individual
files, named dump.DDD.raw, where DDD is 001, 002, 003, ...

-f cmdfile
Reads commands from cmdfile. When this option is used, YAZ client does not read
.yazclientrc from current directory or home directory.

-k size
Sets preferred messages and maximum record size for Initialize Request in kilobytes.
Default value is 65536 (64 MB).

-m filename
If specified, retrieved records will be appended to the file given.

-p proxy-addr
If specified, the client will use the proxy at the address given. YAZ client will
connect to a proxy on the address and port given. The actual target will be specified
as part of the InitRequest to inform the proxy about the actual target.

-q filename
If specified, CQL configuration will be read from the file given.

-t displaycharset
If displaycharset is given, it specifies name of the character set of the output (on
the terminal on which YAZ client is running).

-u auth
If specified, the auth string will be used for authentication.

-v level
Sets the LOG level to level. Level is a sequence of tokens separated by comma. Each
token is a integer or a named LOG item - one of fatal, debug, warn, log, malloc, all,
none.

-V
Prints YAZ version.

-x
Makes the YAZ client print hex dumps of packages sent and received on standard output.

COMMANDS

The YAZ client accepts the following commands.

open zurl
Opens a connection to a server. The syntax for zurl is the same as described above for
connecting from the command line.

Syntax:

[(tcp|ssl|unix|http)':']host [:port][/base]

quit
Quits YAZ client

find query
Sends a Search Request using the query given. By default the query is assumed to be
PQF. See command querytype for more information.

delete setname
Deletes result set with name setname on the server.

base base1 base2 ...
Sets the name(s) of the database(s) to search. One or more databases may be specified,
separated by blanks. This command overrides the database given in zurl.

show [start[+number [+resultset]]]
Fetches records by sending a Present Request from the start position given by start
and a number of records given by number, from the result set resultset. If start is
not given, then the client will fetch from the position of the last retrieved record
plus 1. If number is not given, then one record will be fetched at a time. If
resultset is not given, the most recently retrieved result set is used.

scan term
Scans database index for a term. The syntax resembles the syntax for find. If you want
to scan for the word water you could write

scan water

but if you want to scan only in, say the title field, you would write

scan @attr 1=4 water

setscan set term
Scans database index for a term within a result set. This is similar to the scan
command but has a result set as its first argument.

scanpos pos
Sets preferred position for scan. This value is used in the next scan. By default,
position is 1.

scansize size
Sets number of entries to be returned by scan. Default number of entries is 20.

scanstep step
Set step-size for scan. This value is used in the next scan sent to the target. By
default step-size is 0.

sort sortspecs
Sorts a result set. The sort command takes a sequence of space-separated sort
specifications, with each sort specification consisting of two space-separated words
(so that the whole specification list is made up of an even number of words). The
first word of each specification holds a field (sort criterion) and the second holds
flags. If the sort criterion includes = it is assumed that the SortKey is of type
sortAttributes using Bib-1: in this case the integer before = is the attribute type
and the integer following = is the attribute value. If no = character is in the
criterion, it is treated as a sortfield of type InternationalString. The flags word of
each sort specification must consist of s for case sensitive or i for case
insensitive, and < for ascending order or > for descending order.

sort+
Same as sort but stores the sorted result set in a new result set.

authentication openauth
Sets up an authentication string if a server requires authentication (v2 OpenStyle).
The authentication string is first sent to the server when the open command is issued
and the Z39.50 Initialize Request is sent, so this command must be used before open in
order to be effective. A common convention for the authopen string is that the
username - and password is separated by a slash, e.g. myusername/mysecret.

sru method version
Selects Web Service method and version. Must be one of post, get, soap (default) or
solr. Version should be either 1.1, 1.2 or 2.0 for SRU. Other versions are allowed -
for testing purposes (version negotiation with SRU server). The version is currently
not used for Solr Web Services

list_all
This command displays status and values for many settings.

lslb n
Sets the limit for when no records should be returned together with the search result.
See the Z39.50 standard on set bounds[4] for more details.

ssub n
Sets the limit for when all records should be returned with the search result. See the
Z39.50 standard on set bounds[4] for more details.

mspn n
Sets the number of records that should be returned if the number of records in the
result set is between the values of lslb and ssub. See the Z39.50 standard on set
bounds[4] for more details.

status
Displays the values of lslb, ssub and mspn.

setname
Switches named result sets on and off. Default is on.

cancel
Sends a Trigger Resource Control Request to the target.

facets spec
Specifies requested facets to be used in search. The notation is specified in ???.

format oid
Sets the preferred transfer syntax for retrieved records. yaz-client supports all the
record syntaxes that currently are registered. See Z39.50 Record Syntax Identifiers[5]
for more details. Commonly used records syntaxes include usmarc, sutrs and xml.

elements e
Sets the element set name for the records. Many targets support element sets B (for
brief) and F (for full).

close
Sends a Z39.50 Close APDU and closes connection with the peer

querytype type
Sets the query type as used by command find. The following is supported: prefix for
Prefix Query Notation (Type-1 Query); ccl for CCL search (Type-2 Query), cql for CQL
(Type-104 search with CQL OID), ccl2rpn for CCL to RPN conversion (Type-1 Query),
cql2rpn for CQL to RPN conversion (Type-1 Query).

attributeset set
Sets attribute set OID for prefix queries (RPN, Type-1).

refid id
Sets reference ID for Z39.50 Request(s).

itemorder type no
Sends an Item Order Request using the ILL External. type is either 1 or 2 which
corresponds to ILL-Profile 1 and 2 respectively. The no is the Result Set position of
the record to be ordered.

update action recid doc
Sends Item Update Request. The action argument must be the action type: one of insert,
replace, delete and update. The second argument, recid, is the record identifier (any
string). Third argument which is optional is the record document for the request. If
doc is preceded with "<", then the following characters are treated as a filename with
the records to be updated. Otherwise doc is treated as a document itself. The doc may
also be quoted in double quotes. If doc is omitted, the last received record (as part
of present response or piggybacked search response) is used for the update.

source filename
Executes list of commands from file filename, just like 'source' on most UNIX shells.
A single dot (.) can be used as an alternative.

! args
Executes command args in subshell using the system call.

push_command command
The push_command takes another command as its argument. That command is then added to
the history information (so you can retrieve it later). The command itself is not
executed. This command only works if you have GNU readline/history enabled.

set_apdufile filename
Sets that APDU should be logged to file filename. Another way to achieve APDU log is
by using command-line option -a.

set_auto_reconnect flag
Specifies whether YAZ client automatically reconnects if the target closes connection
(Z39.50 only).

flag must be either on or off.

set_auto_wait flag
Specifies whether YAZ client should wait for response protocol packages after a
request. By default YAZ client waits (on) for response packages immediately after a
command (find, show) has been issued. If off is used, YAZ client does not attempt to
receive packages automatically. These will have to be manually received when command
wait_response is used.

flag must be either on or off.

set_marcdump filename
Specifies that all retrieved records should be appended to file filename. This command
does the same thing as option -m.

schema schemaid
Specifies schema for retrieval. Schema may be specified as an OID for Z39.50. For SRU,
schema is a simple string URI.

charset negotiationcharset [displaycharset] [[marccharset]]
Specifies character set (encoding) for Z39.50 negotiation / SRU encoding and/or
character set for output (terminal).

negotiationcharset is the name of the character set to be negotiated by the server.
The special name - for negotiationcharset specifies no character set to be negotiated.

If displaycharset is given, it specifies name of the character set of the output (on
the terminal on which YAZ client is running). To disable conversion of characters to
the output encoding, the special name - (dash) can be used. If the special name auto
is given, YAZ client will convert strings to the encoding of the terminal as returned
by nl_langinfo call.

If marccharset is given, it specifies name of the character set of retrieved MARC
records from server. See also marccharset command.

Note
Since character set negotiation takes effect in the Z39.50 Initialize Request you
should issue this command before command open is used.

Note
MARC records are not covered by Z39.50 character set negotiation, so that's why
there is a separate character that must be known in order to do meaningful
conversion(s).

negcharset charset
Specifies character set for negotiation (Z39.50). The argument is the same as second
argument for command charset.

displaycharset charset
Specifies character set for output (display). The argument is the same as second
argument for command charset.

marccharset charset
Specifies character set for retrieved MARC records so that YAZ client can display them
in a character suitable for your display. See charset command. If auto is given, YAZ
will assume that MARC21/USMARC is using MARC8/UTF8 and ISO-8859-1 for all other MARC
variants. The charset argument is the same as third argument for command charset.

querycharset charset
Specifies character set for query terms for Z39.50 RPN queries and Z39.50 Scan
Requests (termListAndStartPoint). This is a pure client-side conversion which converts
from displayCharset to queryCharset.

set_cclfile filename
Specifies that CCL fields should be read from file file filename. This command does
the same thing as option -c.

set_cqlfile filename
Specifies that CQL fields should be read from file file filename. This command does
the same thing as option -q.

register_oid name class OID
This command allows you to register your own object identifier - so that instead of
entering a long dot-notation you can use a short name instead. The name is your name
for the OID, class is the class, and OID is the raw OID in dot notation. Class is one
of: appctx, absyn, attet, transyn, diagset, recsyn, resform, accform, extserv,
userinfo, elemspec, varset, schema, tagset, general. If you're in doubt use the
general class.

register_tab command string
This command registers a TAB completion string for the command given.

sleep seconds
This command makes YAZ client sleep (be idle) for the number of seconds given.

wait_response [ number]
This command makes YAZ client wait for a number of response packages from target. If
number is omitted, 1 is assumed.

This command is rarely used and is only useful if command set_auto_wait is set to off.

xmles OID doc
Sends XML Extended Services request using the OID and doc given.

zversion ver
This command sets Z39.50 version for negotiation. Should be used before open. By
default 3 (version 3) is used.

options op1 op2..
This command sets Z39.50 options for negotiation. Should be used before open.

The following options are supported: search, present, delSet, resourceReport,
triggerResourceCtrl, resourceCtrl, accessCtrl, scan, sort, extendedServices,
level_1Segmentation, level_2Segmentation, concurrentOperations, namedResultSets,
encapsulation, resultCount, negotiationModel, duplicationDetection, queryType104,
pQESCorrection, stringSchema.

EXAMPLE

       The simplest example of a Prefix Query would be something like

               f knuth

       or

               f "donald knuth"

       In those queries, no attributes were specified. This leaves it up to the server what
       fields to search but most servers will search in all fields. Some servers do not support
       this feature though, and require that some attributes are defined. To add one attribute
       you could do:

               f @attr 1=4 computer

       where we search in the title field, since the use(1) is title(4). If we want to search in
       the author field and in the title field, and in the title field using right truncation it
       could look something like this:

               f @and @attr 1=1003 knuth @attr 1=4 @attr 5=1 computer

       Finally using a mix of Bib-1 and GILS attributes could look something like this:

               f @attrset Bib-1 @and @attr GILS 1=2008 Washington @attr 1=21 weather

FILES

       yaz-<version>/client/client.c

       $HOME/.yazclientrc

       $HOME/.yazclient.history

AUTHORS

       Index Data

NOTES

        1. Z39.50
           https://www.loc.gov/z3950/agency/

        2. SRU
           https://www.loc.gov/standards/sru/

        3. Solr
           https://lucene.apache.org/solr/

        4. Z39.50 standard on set bounds
           http://www.loc.gov/z3950/agency/markup/04.html#3.2.2.1.6

        5. Z39.50 Record Syntax Identifiers
           http://www.loc.gov/z3950/agency/defns/oids.html#5