| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
might incur ncurses dependencies on some platforms, and ncurses just bumped
its shlib.
Some packages were bumped twice now, sorry for that.
|
|
|
|
Thank you, wiz@.
|
|
CHangelog:
* Changes in opensource COBOL 1.4.0J
** New features
*** Supports for Microsoft Windows ( Visual Studio and CL compiler )
Added win32 directory in opensource COBOL package, and prepared Solution Files for Visual Studio.
*** Enhanced CUI features
(1) Add a new function "CBL_OC_KEISEN".
This displays a KEISEN ( vertical or horizontal ruled lines ) on the screen.
(2) Supports SJIS multi-byte character input and output on the screen.
** Changes
*** Added VBISAM to the opensource COBOL package
*** Fixed not to abandon REPLACE after following COPY REPLACING has done
*** Though old style STOP statement maybe skipped, it should be still at least recognized as non-nil statement on empty check
*** Stop editing initial value of national-edited item
*** Changed NATIONAL comparison to be independent on COLLATING SEQUENCE setting
*** Fixed memory leaks in decoding national words
*** Fixed behavior of COPY REPLACING LEADING/TRAILING according to the standard
*** Fixed undecoded national item names appear in runtime messages
*** Fixed undecoded national item names appear in compilation messages
*** Prefixing/Suffixing should have no effect on FILLER item
*** Fixed compilation error on some SOURCE FORMAT FREE usage
*** Fixed bug in field subscript boundary check in cond expression
*** Preserve system time when date is modified by COB_DATE
*** Avoid reporting wrong section/paragraph name in error message
*** Fixed bug in reference modification boundary check in MOVE statement
*** Fixed initialization bug of ZonedDecimal variable defined as SIGN SEPARATE
*** Fixed initialization bug of ZonedDecimal variable when using -fsign-ebcdic option
** Other Updates
*** added and enhanced some tests
*** updated message catalog for 'ja'
*** bug fix some problem
|
|
|
|
Fix PR pkg/48659 from Yasushi Oshima.
* Enable UTF-8 support
Changelog:
* Changes in opensource COBOL 1.3.2J
** New features
*** Merged OpenCOBOL 1.1CE updates.
*** Thanks to The OpenCOBOL Project! ***
*** New compiler warning option: -Wcompat
(1) multiply after division in a numeric expression
(2) GO TO over the SECTION boundary
(3) X, NX Literal usage
*** Supports for compatibility issues found in existing COBOL programs:
*** "COPY JOINING AS PREFIX/SUFFIX", "COPY PREFIXING/SUFFIXING"
*** 'OR' in "GREATER/LESS THAN OR EQUAL" to optional
*** Treat full-width spaces as blank(token delimiter) in source text
*** accept data description without period
*** accept SPECIAL-NAMES without period
*** accept IO-CONTROL without period
*** accept syntax of INDEXED before KEY in OCCURS clause
*** ND, NX Literal, and N, NC Literal in SJIS mode
*** RESERVE "NO" in SELECT statement - syntactically accepted, no functionality
*** cheating DATE value by environment variable (set COB_DATE=yyyy/mm/dd)
*** put sort/merge result record count on console (when set COB_VERBOSE=Y)
*** implicitly substitute WRITE with REWRITE on the file opened in I-O mode (when set COB_IO_ASSUME_REWRITE=Y)
*** treat sign-nibble 'C' as 'F' in unsigned packed numeric value, when also type checking.
*** added another device naming format in assign clause (conf.parameter assign-clause: jph1)
*** UPSI 0-7 aliased to SWITCH1-8
*** INVALID KEY for sequential READ/WRITE (conf.parameter use-invalidkey-handler-on-status34)
*** COBOL68 style COPY semantics in data description clause (conf.parameter cobol68-copy-in-data-description)
*** accept key field name on right hand side of WHEN expression in SEARCH statement (conf.parameter allow-search-key-in-rhs)
*** added optional "IS" for SORT KEY spec. syntax
*** ignore "RECORD CONTAINS" (conf.parameter ignore-invalid-record-contains)
** Changes
*** omit checking field reference in DATA RECORDS
*** improve parser error recovery and source line-number in messages
** Other Updates
*** added some tests
*** updated message catalog for 'ja'
*** fixed bug in decording multibytes chars in section/paragraph name in compile messages
-----------------------------------------------------------------------
* Changes in opensource COBOL 1.3.1J
** New features
*** Enhanced SPLIT KEY support also accepts multiple key items specified for RECORD KEY.
*** Added SPLIT KEY support to START statement.
*** Added compilation warning that SPLIT KEY is currently only available on VB-ISAM
(and other ISAM compatibles).
*** Conditional compiler directive $IF.
*** Automatic file creation on opening non exisiting file for I-O, EXTEND.
*** Added DELETE FILE statement.
** Bug fixes
*** Fixed the problem that COPY statement ignores the last line of each copybook when
it does not end with line break. Actually same as opensource COBOL 1.2J, applied
again to repair corruption in recent changes.
*** Restored compilation warning for KEY phrase in READ statement on the SEQUENTIAL
mode file. It seemed to have been corrupted in recent changes.
*** Fixed problem that DYNAMIC phrase has no effect when -assign_external option is
specified.
*** Fixed problem that some options are never passed to READ routine in USERFH.
*** Fixed problem of build time causes error when no curses library given.
*** Fixed some segfault on compilation error.
*** Fixed problem eliminates record length information on READing, SORTing variable
length record.
*** Fixed internal error on long Japanese word in PROGRAM-ID.
|
|
|
|
opensource COBOL is based on OpenCOBOL compiler and has some
extensions used in Japan. It translate COBOL programs to C code
and compiles it using GCC.
This package tracks opensource COBOL UTF-8 version.
|
