Version 0.8.6 released
21 December 2009 Andreas Rumpf
The version jump from 0.8.2 to 0.8.6 acknowledges the fact that all development of the compiler is now done in Nimrod.
- The pragmas
- Method call syntax for iterators works again (
for x in lines.split()).
- Fixed a typo in
removeDirfor POSIX that lead to an infinite recursion.
- The compiler now checks that module filenames are valid identifiers.
- Empty patterns for the
dynlibpragma are now possible.
os.parseCmdLinereturned wrong results for trailing whitespace.
- Inconsequent tuple usage (using the same tuple with and without named fields) does not crash the code generator anymore.
- A better error message is provided when the loading of a proc within a dynamic lib fails.
system.containsfor open arrays.
- The PEG module now supports the search loop operator
SAD|INDis allowed before any kind of closing bracket. This allows for more flexible source code formatting.
- The compiler now uses a bind table for symbol lookup within a
bindcontext. (See manual.html#templates for details.)
discard """my long comment"""is now optimized away.
--floatChecks: on|offswitches and pragmas for better debugging of floating point operations. (See manual.html#types-pre-defined-floating-point-types for details.)
- The manual has been improved. (Many thanks to Philippe Lhoste!)
Changes affecting backwards compatibility
- The compiler does not skip the linking step anymore even if no file has changed.
("", ".xyz", "")instead of
("", "", ".xyz"). So filenames starting with a dot are handled differently.
strutils.split(s: string, seps: set[char])never yields the empty string anymore. This behaviour is probably more appropriate for whitespace splitting.
- The compiler now stops after the
--versioncommand line switch.
- Removed support for enum inheritance in the parser; enum inheritance has never been documented anyway.
system.E_basehas now the type
string, instead of
cstring. This improves memory safety.