forums.silverfrost.com Forum Index forums.silverfrost.com
Welcome to the Silverfrost forums
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Silverfrost FTN95: Fortran for Windows version 8.50

 
This forum is locked: you cannot post, reply to, or edit topics.   This topic is locked: you cannot edit posts or make replies.    forums.silverfrost.com Forum Index -> Announcements
View previous topic :: View next topic  
Author Message
silverfrost
Site Admin


Joined: 29 Nov 2006
Posts: 166
Location: Manchester

PostPosted: Thu May 02, 2019 9:33 pm    Post subject: Silverfrost FTN95: Fortran for Windows version 8.50 Reply with quote

We are pleased to announce the release of FTN95 version 8.50. This is available for download now to supported users.

New Features:

  • A new Silverfrost utility called EditSVG
    is provided. This can be used to form images from a number of SVG
    sources into one composite image.
  • The size of the FTN95 stack is effectively no longer limited. The default size is 32MB and this can be
    extended via the SLINK64 "stack_size <hex-value>" command or the new SLINK64 "stack <hex-value>" command
    where <hex-value> is the decimal number of decimal megabytes required.
  • Fortran 2003 standard I/O FLUSH statement added. (Win32/x64 only)
  • Fortran 2003 standard I/O routines, keyword argument IOMSG added. (Win32/x64 only)
  • Fortran 2003 standard intrinsic GET_ENVIRONMENT_VARIABLE added.
  • Fortran 2008 standard intrinsic EXECUTE_COMMAND_LINE added.
  • Routines READFLONG@ and WRITEFLONG@ have been added (x64 only). These are the same as READF@ and WRITEF@ respectively except that they take INTEGER*8 values for NBYTES and NBYTES_READ.
  • The intrinsic functions LEN and LEN_TRIM can now take a second argument for the KIND value of the result.
  • Array arguments for routines can now be ALLOCATABLE according to the Fortran 2003 standard.
  • An FTN95 ALLOCATE statement can take the keyword argument ABSOLUTE_ADDRESS=addr where addr is the INTEGER KIND(7) address obtained from the LOC of some object. (Win32/x64 only)
  • A new FTN95 "CORE" intrinsic PCORE7 has been added for use in pointer assignments of the form p=>PCORE7(addr) where p is a pointer to some object and addr is the LOC of another object of the same type.
  • Microsoft Visual Studio 2017 plugin
  • ClearWin+: The buffer size for %RE was previously limited to 32K characters. This limit has now been removed but in such cases the buffer will not be automatically updated and an explicit call to GetEditText@ will be required in order to copy the text to the buffer. If this copying takes place within an associated callback function then clearwin_info@("CURRENT_CONTROL") returns the required HWND.
  • ClearWin+: ALLOCATE(data1,ABSOLUTE_ADDRESS=addr) can be used in a callback relating to %ud. addr will be obtained via CLEARWIN_INFO@("USER_DATA") and data1 will be an object of the same kind as that used in the call to LOC.


Last edited by silverfrost on Fri May 03, 2019 8:56 am; edited 1 time in total
Back to top
View user's profile Send private message Visit poster's website
silverfrost
Site Admin


Joined: 29 Nov 2006
Posts: 166
Location: Manchester

PostPosted: Thu May 02, 2019 9:36 pm    Post subject: Reply with quote

Outline summary of bugs that have been fixed:

  • A PRIVATE TYPE in a MODULE was not private.
  • A USE ONLY attribute for a MODULE TYPE was not working correctly.
  • There was an internal compiler error with /64 and PRINT a(i,1:n).
  • /UNDEF was failing for an ALLOCATEd vector and vector subscript.
  • Various /64 /OPT bugs have been fixed.
  • An implicitly typed implied DO loop variable was being exported from a MODULE.
  • GET_COMMAND_ARGUMENT was failing when the path is set in double quotes.
  • RANDOM@ gave incorrect results when used in a small loop.
  • There was an access violation with /64 and an internal READ of a logical list.
  • COMPLEX(RESHAPE(...)) was not working correctly for REAL data.
  • An error report for a certain ALLOCATE rank failure was giving an access violation.
  • Assigning a scalar return from SUM to an array was not working correctly.
  • An array used in a WHERE mask was being interpreted as a vector subscript in the body of the construct.
  • /UNDEF was failing with the SUM intrinsic when used with ALLOCATABLE arrays.
  • There was a regression for Win32 when using SPREAD within a certain call to SUM.
  • The 64 bit version of RENAME@ was incorrectly giving an error number of zero when the file to rename did not exist.
Back to top
View user's profile Send private message Visit poster's website
Display posts from previous:   
This forum is locked: you cannot post, reply to, or edit topics.   This topic is locked: you cannot edit posts or make replies.    forums.silverfrost.com Forum Index -> Announcements All times are GMT + 1 Hour
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group