X-Git-Url: http://git.indexdata.com/?p=yaz-moved-to-github.git;a=blobdiff_plain;f=CHANGELOG;h=7f27f4ed6388cadd9a15206f434d05bd45c0c054;hp=259eb2bafd00e28f6acc2d02f4fb6a5e1d84392a;hb=42eee1a00a498bd7a106363d85f804d163d3cfb6;hpb=97e0cd0e1341f4b75216dec7c95f7a4627f4e63d diff --git a/CHANGELOG b/CHANGELOG index 259eb2b..7f27f4e 100644 --- a/CHANGELOG +++ b/CHANGELOG @@ -1,5 +1,75 @@ Possible compatibility problems with earlier versions marked with '*'. +Fixed bug in client.c which caused a crash when a scanResponse didn't +contain positionOfTerm. + +* Fixed inconsistency in the ScanResponse/ListEntries protocol. This +* will cause an error in a client if a non-YAZ target sends both a +* list of Scan entries and non-surrogate diagnostics. + +Retrieval module enhancements. Tag sets may be typed in the reference +to it. From the .abs-file the "tagset" directive takes a third optional +integer type for the tag set referenced. From a .tag-file the "include" +directive takes a third optional type as well. The old "type" directive +in the tag set itself is still recognized but acts as the default type +for the tag set. + +Backend interface change. Individual Scan terms returned from bend_scan +may also be tagged as Surrogate Diagnostics. Refer to struct scan_entry +in include/backend.h. + +Retrieval module enhancements. Multiple tag sets and attribute sets may +be specified in abstract syntax specs. Revised the error/warning for +logging messages reported by the retrieval module when reading the +various spec files. + +Object Identifier system (oid_..) changed. The API is backwards +compatible but the oid-routines now allocates new OID's when needed. +Raw OID's may be specified in oid_getvalbyname. + +YAZ now uses GNU configure to generate Makefile(s). + +* Minor changes in some of the member names of DeleteResultSetRequest and +DeleteResultSetResponse. See include/proto.h. + +* Changed some C definitions regarding EXPLAIN record syntax (prt-exp.h) +and added member 'languages' in TargetInfo (which was missing). + +Changed the way attribute sets are handled in the retrieval +module. These are now cached, just like the abstract syntaxes +(schemas). + +Changed YAZ so that it links with compiled ASN.1. + +Added Explain-schema definitions and mapping routines. The following +Explain categories have been implemented: AttributeDetails, +AttributeSetInfo, DatabaseInfo, CategoryList and TargetInfo. + +Fixed minor bug in Windows version of Server Library. + +--- 1.4pl2+ 1998/4/17 (Released with Zebra) + +Added Extended Services to server (backend.h). Thanks to Charles Woodfield. + +Added EXTERNAL: UNIverse Resource Report (non-standard). + +Implemened odr_enum and odr_set_of functions to handle ASN.1 types +ENUMERATED and SET OF. + +* Added 'const' to some of the char pointer arguments for the +functions, data1_read_node, data1_getelementbytagname, +data1_insert_taggeddata, data1_getesetbyname, data1_getelementbyname, +data1_get_absyn. + +Fixed bug in data1_insert_taggeddata - the last_child member of +parent wasn't initialised. + +* Changed data1_read_record and data1_read_node so that the supplied +"SGML"-buffer is read-only. In particular the resulting data1 - tree +doesn't refer to any parts of the "SGML"-buffer. + +Fixed bug in server library (Windows version of statserv_remove). + * Changed code so that it compiles as C++. The type definition of Odr_fun which is a member of Odr_arm was changed. The result is, that an explicit typecast is needed for the fun pointer when