GNU
OSX_Resources
SPECS
VCPP
conf
d4_ce
d4_function
gl
tests
unit-tests
.packit.yaml
AlarmHandler.h
Ancillary.cc
Ancillary.h
Array.cc
Array.h
AttrTable.cc
AttrTable.h
BaseType.cc
BaseType.h
BaseTypeFactory.cc
BaseTypeFactory.h
Byte.cc
Byte.h
COPYING
COPYRIGHT_URI
COPYRIGHT_W3C
ChangeLog
Clause.cc
Clause.h
Connect.cc
Connect.h
ConstraintEvaluator.cc
ConstraintEvaluator.h
Constructor.cc
Constructor.h
D4AsyncUtil.cc
D4AsyncUtil.h
D4AttributeType.h
D4Attributes.cc
D4Attributes.h
D4BaseTypeFactory.cc
D4BaseTypeFactory.h
D4Connect.cc
D4Connect.h
D4Dimensions.cc
D4Dimensions.h
D4Enum.cc
D4Enum.h
D4EnumDefs.cc
D4EnumDefs.h
D4FilterClause.cc
D4FilterClause.h
D4Function.h
D4Group.cc
D4Group.h
D4Maps.cc
D4Maps.h
D4Opaque.cc
D4Opaque.h
D4ParserSax2.cc
D4ParserSax2.h
D4RValue.cc
D4RValue.h
D4Sequence.cc
D4Sequence.h
D4StreamMarshaller.cc
D4StreamMarshaller.h
D4StreamUnMarshaller.cc
D4StreamUnMarshaller.h
DAS.cc
DAS.h
DDS.cc
DDS.h
DDXExceptions.h
DDXParserSAX2.cc
DDXParserSAX2.h
DMR.cc
DMR.h
DODSFilter.cc
DODSFilter.h
DapIndent.cc
DapIndent.h
DapObj.h
DapXmlNamespaces.cc
DapXmlNamespaces.h
DataDDS.cc
DataDDS.h
EncodingType.h
Error.cc
Error.h
Error.lex
Error.tab.cc
Error.tab.hh
Error.yy
EventHandler.h
Float32.cc
Float32.h
Float64.cc
Float64.h
Grid.cc
Grid.h
HTTPCache.cc
HTTPCache.h
HTTPCacheDisconnectedMode.h
HTTPCacheInterruptHandler.h
HTTPCacheMacros.h
HTTPCacheResponse.h
HTTPCacheTable.cc
HTTPCacheTable.h
HTTPConnect.cc
HTTPConnect.h
HTTPResponse.h
INSTALL
Int16.cc
Int16.h
Int32.cc
Int32.h
Int64.cc
Int64.h
Int8.cc
Int8.h
InternalErr.cc
InternalErr.h
Keywords2.cc
Keywords2.h
Makefile.am
Makefile.in
Marshaller.h
MarshallerThread.cc
MarshallerThread.h
NEWS
ObjectType.h
Operators.h
PipeResponse.h
RCReader.cc
RCReader.h
README
README.dodsrc
RValue.cc
RValue.h
Resource.h
Response.h
ResponseTooBigErr.cc
ResponseTooBigErr.h
Sequence.cc
Sequence.h
ServerFunction.cc
ServerFunction.h
ServerFunctionsList.cc
ServerFunctionsList.h
SignalHandler.cc
SignalHandler.h
SignalHandlerRegisteredErr.h
StdinResponse.h
Str.cc
Str.h
Structure.cc
Structure.h
Type.h
UInt16.cc
UInt16.h
UInt32.cc
UInt32.h
UInt64.cc
UInt64.h
UnMarshaller.h
Url.cc
Url.h
Vector.cc
Vector.h
XDRFileMarshaller.cc
XDRFileMarshaller.h
XDRFileUnMarshaller.cc
XDRFileUnMarshaller.h
XDRStreamMarshaller.cc
XDRStreamMarshaller.h
XDRStreamUnMarshaller.cc
XDRStreamUnMarshaller.h
XDRUtils.cc
XDRUtils.h
XMLWriter.cc
XMLWriter.h
abi_checker.xml.in
aclocal.m4
ce_expr.lex
ce_expr.tab.cc
ce_expr.tab.hh
ce_expr.yy
ce_parser.h
cgi_util.h
chunked_istream.cc
chunked_istream.h
chunked_ostream.cc
chunked_ostream.h
chunked_stream.h
config.h.in
config_dap.h
configure.ac
crc.h
dap-config-pkgconfig
dap-config.in
dap-config.man1
das.lex
das.tab.cc
das.tab.hh
das.yy
dds.lex
dds.tab.cc
dds.tab.hh
dds.yy
debug.h
dods-datatypes-config.h.in
dods-datatypes-static.h
dods-datatypes.h
dods-limits.h
doxy.conf.in
escaping.cc
escaping.h
expr.h
getdap.cc
getdap.man1
getdap4.cc
getdap4.man1
lex.Error.cc
lex.ce_expr.cc
lex.das.cc
lex.dds.cc
libdap.pc.in
libdap.spec
libdapclient.pc.in
libdapserver.pc.in
main_page.doxygen.in
media_types.h
mime_util.cc
mime_util.h
parser-util.cc
parser-util.h
parser.h
util.cc
util.h
util_mit.cc
util_mit.h
xdr-datatypes-config.h.in
xdr-datatypes-static.h
xdr-datatypes.h
xdrutil_ppc.c
configure
README.dodsrc
$Id$ Current for version 3.7.6 (12 March 2007) The libdap client-side configuration file (.dodsrc) is used to configure how clients cache responses and how they interact with proxy servers. By default the configuration file resides in a users home directory and is called `.dodsrc'. This can be changed by creating the environment variable DODS_CONF and setting it to the full pathname of the configuration file. About the Win32 builds: For Win32 (Windows XP and Vista) a client linked with libdapclient will look for the .dodsrc file in the following locations: C:\Dods; %APPDATA%; %TEMP%; and %TMP%. If no .dodsrc file is found, the library will create one in %APPDATA%. Note that on some Windows XP machines a user's 'Application Data' directory is hidden. To see and edit the .dodsrc file, open a file browser window from the Start menu, go to your home directory and then type in 'Application Data'. The .dodsrc file will be created by the first OPeNDAP client to run if there's not a copy already there. Another note about the Win32 builds and the .dodsrc file: It appears that client-side caching does not work in recent versions of the library, so until further notice, leave USE_CACHE set to zero. New feature added with libdap version 3.7.3 (7 Nov 2006) The configuration file can now be used to control client-side behavior when accessing servers using SSL. By default, certificates must be signed by a certificate authority. The libcurl package recognizes a large set of CAs; you can sign your own certificates as well (but see the OpenSSL documentation for more information). In addition, server hosts must identify themselves using the same name as is used in the certificate. To disable these features, set the configuration parameter VALIDATE_SSL to zero. By default, these features are now enabled following the defaults for libcurl. If a DODS client starts and cannot find the configuration file, then one with default parameters will be created in the user's home directory. By default caching will be enabled with a maximum cache size of 20M and a default expiration time of 24 hours. By default no proxy server is configured and SSL hosts and certificates are validated. Also by default, compression is not activated. A sample configuration file looks like (the line numbers are not part of the file; they've been added to make the description clearer): 0 # Comments start with a `#' in the first column. 1 USE_CACHE=1 2 MAX_CACHE_SIZE=20 3 MAX_CACHED_OBJ=5 4 IGNORE_EXPIRES=0 5 CACHE_ROOT=/home/jimg/.dods_cache/ 6 DEFAULT_EXPIRES=86400 7 DEFLATE=0 8 VALIDATE_SSL=1 9 PROXY_SERVER=http,http://jimg:password@dcz.dods.org/ 10 NO_PROXY_FOR=http,dcz.dods.org 11 AIS_DATABASE=.ais_sst_database COMMENTS Starting a line with a `#' makes that line a comment. CACHING The parameters on lines 1 through 6 determine how the DAP++ library will use its HTTP 1.1 cache. If the value of USE_CACHE is 1, the cache is active. A value of 0 turns off the cache. Make sure you use zero (0) and not the letter `O'. The value of MAX_CACHE_SIZE sets the maximum size of the cache in megabytes. Once the cache reaches this size, caching more objects will cause cache garbage collection. The algorithm used is to first purge the cache of any stale entries and then remove remaining entries starting with those that have the lowest hit count. Garbage collection stops when 90% of the cache has been purged. The value of MAX_CACHED_OBJ sets the maximum size of any individual object in the cache in megabytes. Objects received from a server larger than this value will not be cached even if there's room for them without purging other objects. The parameter CACHE_ROOT contains the pathname to the cache's top directory. If two or more users want to share a cache, then they must both have read and write permissions to the cache root. If the value of IGNORE_EXPIRES is 1, then Expires: headers in response documents will be ignored. The value of DEFAULT_EXPIRES sets the expiration time for any response that does not include either an Expires or Last-Modified header. The value is given in seconds; 86,400 is 24 hours. In general you should *not* ignore the Expires header; the server probably had a good reason to send it along with the response. This parameter is here for unusual situations. Note: If a Last-Modified header is returned with the response, and there's *no* Expires header, the expiration time is is 10% of the difference between the current time and the LM time or 48 hours, whichever is smaller. Note that libdap ignores the DEFAULT_EXPIRES time in this case. Any request made before the expiration time will use the cached response without validation. Any request made after the expiration time will use a conditional GET. Servers that have been upgraded to 3.2 or greater will return a 304 response if the cached response is still valid or a new response if it is not valid. If the value of ALWAYS_VALIDATE is 1, then all accesses will be validated with the origin server. A value of 0 causes libwww to use the more complex expiration and validate algorithm. CONTROLLING DATA COMPRESSION If the DEFLATE parameter is set to one (DEFLATE=1) then clients will request that servers compress data transmissions. Servers may or may not honor this request depending on their abilities. SSL VALIDATION Set VALIDATE_SSL to zero to turn off SSL server host and certificate validation. By default, SSL hosts and certificates are now validated. If you're using your own certificates and don't want to pay to have them signed by a CA or to run your own authorization site, use this to defeat SSL validation. The old libdap default behavior was to _not_ perform validation (because that was the old libcurl default) and we've included this option so that you can get that old behavior. PROXY SERVERS Note that the parameters PROXY_SERVER and NO_PROXY_FOR may be repeated to add multiple proxy servers, suppress proxy access for several hosts, etc. Lines 9 and 10 contain the parameters used to configure a proxy server. The parameter PROXY_SERVER configures a default proxy server. The format is PROXY_SERVER=<protocol>,<proxy host url> The only supported protocols are http, https, and ftp. <proxy host url> must be a full URL to the host running the proxy server. If a password is used to access the proxy server, include it in the URL using the <user:password@host> syntax as shown in the example. The NO_PROXY parameter provides a simple way to say that access to a certain host should never go through a proxy. The syntax of NO_PROXY is: NO_PROXY=<protocol>,<hostname> where <protocol> is as for PROXY_SERVER <hostname> is the name of the host, not a url. The Ancillary Information Service See README.AIS for more information about the AIS.