pkgsrc/lang/gauche/distinfo

13 lines
646 B
Text
Raw Normal View History

$NetBSD: distinfo,v 1.20 2006/03/31 18:35:22 jlam Exp $
Update gauche to 0.8.6. Patch provided by Kenji Hisazumi. From this release we don't use external Boehm GC library because gauche needs GC to be built with special compilation options. New features in 0.8.6: * New modules: + dbi: Database independent access layer, providing unified access to various relational databases. You need separate "driver" packages to access the actual RDBMS. There are a few driver packages available at http://www.kahua.org/cgi-bin/kahua.fcgi/kahua-web/show/dev/DBI/. Note: If you have been using the separate dbi module, make sure you remove it before using the new dbi and dbd modules. You can find the old dbi.scm under somewhere like /usr/local/share/gauche/site/lib (the actualy directory depends on the configuration when you've installed the dbi module). + util.relation: A framework to work with relations (as defined by Codd). The result of database access via dbi is represened as a relation. + text.sql: SQL parser/constructor. Full features are not implemented yet, but used in dbi module for prepared queries. * New SRFIs: + SRFI-40 (Library of streams) as util.stream. + SRFI-43 (vector library) as srfi-43. + SRFI-45 (Primitives for Expressing Iterative Lazy Algorithms) : built-in. * New built-in proceduers global-variable-bound? and glboal-variable-ref. The former supersedes symbol-bound? ( symbol-bound? is now deprecated and will go away in the future releases. Code that uses symbol-bound? should change it to global-variable-bound?. ). The latter removes some need of using eval just to peek the value of the global variable. * New regexp procedures: regexp-replace*, regexp-replace-all*, regexp-case-fold?. * Stack overflow handling is largely improved. You can see better performance if your script frequently oveflows the stack. 0.8.5 was a maintainance release. 0.8.4: Gauche 0.8.4: The compiler and VM have been rewritten. Now Gauche runs faster with less memory (as fast as 1.9x, or cosumes 0.7x memory, in best cases of our tests. But your mileage may vary.) The compiler now does simple closure optimization, so the typical loop-by-local-closure style code will get the advantage. On the other hand, you won't see much gain in OO-heavy or library-heavy programs. Other changes: * New features: + srfi-42 (Eager comprehension) is supported. + srfi-55 (require-extension) is supported. + A simple sampling profiler is implemented to help tuning programs. Check out "Profiling and tuning" section of the reference manual. The profiler may not be available on all platforms. + We provide an experimenal Windows/MinGW binary package for the convenience. See download page.
2005-12-09 16:38:54 +01:00
SHA1 (Gauche-0.8.6.tgz) = 8ce200011da2eff925c8ccf65502529df379f2b8
RMD160 (Gauche-0.8.6.tgz) = e6986c2602926b6a5c16a41838212bd041a4fbae
Size (Gauche-0.8.6.tgz) = 2881800 bytes
SHA1 (patch-aa) = 21aa160a43ce7eb0de636dd422c8d9b59b71f8ee
SHA1 (patch-ab) = 7c05e4e266078903871e13084c217ba6930275ce
SHA1 (patch-ac) = 7e5e0056f6e1cf685edf8dd0bd5b7ea87b05348c
SHA1 (patch-ad) = 25d6ed22d3a3d8d0360575ab8be5c5f1f0cf447e
SHA1 (patch-ae) = 0f76f4ba32b21d9c34335713e17bac70b8f8f9fd
SHA1 (patch-af) = 3f269e0c65830eeed9c5723cda02a9a66a8aa475
SHA1 (patch-ag) = 97d19d33e2ae59bb894716bd289206b9b12ed38c