Branch: Tag:

2008-05-07

2008-05-07 10:41:23 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Added __func__ and CompilerEnvironment/PikeCompiler.
Some cleanups.
Fixed several typos.
Added lots of white space.

Rev: CHANGES:1.90

2:   ----------------------------------------------------------------------      o Added ABI selection. +     It's now possible to select whether to compile in 32bit or 64bit    mode at configure time by using the --with-abi option.      o New syntax to index from the end in range operations. -  +     A "<" can be added before an index in the [..] operator to count    from the end instead. This is convenient to e.g. chop off the last    element in an array: a[..<1].    - o Several declarations and definitions (most notably the debug and + o Improved support for embedding. +  +  Several declarations and definitions (most notably the debug and    runtime flags) have moved from main.h to pike_embed.h, in an attempt    to add support for embedding.      o Major compiler API cleanups. -  +     The Pike compiler is now executing in a pike function context    (rather than in an efun context), and it is possible to customize    some of its behaviour via inherit (rather than via handler objects).    As a consequence the compiler is now much closer to being thread-safe.      o New `[..] operator function. -  +     Range operations have been separated from the `[] operator function    and is handled by a new `[..] which provides greater control in how    the range bounds are specified. For compatibility, if there is no
29:    FIXME: Describe relation with Pike.INDEX_FROM_BEG/END & OPEN_BOUND.      o Generalized this and this_program. +     It is now possible to refer to inherits in objects. Example: -  +     class A {    int a;    void foo() { werror("A\n"); }
40:    void foo() { werror("B\n"); }    A::this_program get_a() { return A::this; }    } +     In the above B()->get_a() will return an object with two symbols,    'a' and 'foo', but B()->get_a()->foo() will still write "B\n".      o Added support for getters and setters. -  +     It is now possible to simulate variables with functions. Example: -  +     class A {    private int a;    int `b() { return a; } // Getter for the symbol b.
60:    werror("%d\n", a->b); // Calls `b().      o Unicode escapes. +     Pike now understands the common way to escape unicode chars, using    \uxxxx and \Uxxxxxxxx escapes. These escapes works both in string    and character literals and in the preprocessor. The latter means
68:    code charset.      o New (stricter) type checker for function calls. +     The type checker for function calls is now based on the concept    of currification. This should provide for error messages that    are more easily understood. It also is much better att typechecking
76:    Pike.get_first_arg_type(), Pike.low_check_call() and Pike.get_return_type().      o Added generic attributes for types. +     Provides a method to hook into the type checker, so that it is    possible to make custom type checking.    No syntax yet.    Added such a type checker for sprintf().      o Stricter typing of strings. -  +     The string type may now have an (optional) value range. Example: -  +     string(0..255) bytes;      o Stdio.* -  +     Stdio.cp can now work recursively in a directory tree.    Stdio.cp now keeps permissions when copying.    Added Stdio.recursive_mv which works on every OS and also when the
101:    NOTE!    NOTE! Changed the argument for backend callbacks!    NOTE! -  NOTE! The argument is now a struct Backend_struct * when called at entry (was NULL). +  NOTE! The argument is now a struct Backend_struct * when called +  NOTE! at entry (was NULL).    NOTE! The argument is now NULL when called at exit (was 1).    NOTE!      o bin/precompile.pike -  +     Voidable pointer types are no longer promoted to mixed.      o cpp -  +     The preprocessor now supports macro expansion in the #include and #string    directives.      o PIKE_PORTABLE_BYTECODE -  +     --with-portable-bytecode is now the default. Pike programs that have    been dumped on one architecture now can be decoded on another.      o Math.Matrix -  Multiplication were bugged and gave B*A instead of A*B. Now fixed. +     -  +  Multiplication was bugged and gave B*A instead of A*B. Now fixed. +    o Destruct reason passed to lfun::destroy. -  +     lfun::destroy now receives an integer flag that tells why the object    is being destructed, e.g. if it was an explicit call to destroy(),    running out of references, or reaped by the garbage collector.
132:    Also added support for aliased symbols.      o Improved support for mixin. +     The Pike compiler now supports mixin for symbols that have been    declared static.      o Implicit and explicit create(). -  +     The compiler now supports defining classes with both an implicit    and an explicit create().      o Warnings for unused private symbols. -  +     The compiler now checks that all symbols that have been declared    private actually are used.      o Unicode -  +     Case information and the Unicode module are updated to Unicode    5.0.0.      o protected -  +     The modifier protected is now an alias for the modifier static.      o Pike_fp->context -  +     Pike_fp->context is now a pointer to the current struct inherit    rather than a copy of it. This allows for easier access to    inherited symbols in C-code.
159:   Extensions and New Functions      o __attribute__ and __deprecated__ +     It's now possible to set custom attributes on types. This is    currently used to improve the argument checking for sprintf()    and related functions, and for marking symbols as deprecated.    -  + o __func__ +  +  The symbol __func__ now evaluates to the name of the current +  function. Note that this name currently can differ from the +  declared name in case of local functions (ie lambdas). Note +  also that __func__ behaves like a litteral string, so implicit +  string concatenation is supported. eg: +  +  error("Error in " __func__ ".\n"); +    o ADT.BitBuffer       Added read() method that reads whole bytes from the buffer and
215:       - Timezones are now read from /etc/localtime, if available.    + o CompilerEnvironment & CompilerEnvironment()->PikeCompiler +  +  The Pike compiler has been refactored to be more object-oriented +  and more transparent. It is now possible to customize the compiler +  by overloading functions in the above two classes. +    o Debug       The new function count_objects() will return the different kinds of    objects existing within the Pike process. Useful when trying to    pinpoint a leak of Pike objects.    -  + o DefaultCompilerEnvironment +  +  The CompilerEnvironment object used by Pike internally. +    o Error       The new function mkerror() will normalize any thrown value into a
267:    Int.inf is an object that can be used as an infinitly large integer.      o Locale.Charset +     FIXME: Rewrite    Remapped and documented the use of the private character space.    ISO-IR non-spacers are now converted into combiners.
301:    strings.      o Mysql +     Unicode support. FIXME: Document in more detail. C.f. check-in 1.25    of lib/modules/Sql.pmod/mysql.pike.      o Odbc -  +     FIXME: Rewrite    Now supports UnixODBC properly.    Fixed various issues with Unicode.
315:    performance. The #string directives should be handled correctly now.      o Parser.RCS +     FIXME: Rewrite    Handle broken RCS files better.   
326:    - Fixed crash bug.      o Parser.XML.Tree +     FIXME: Rewrite    Added rendering of DTD ENTITY nodes.    The renderer now knows about the internal_subset attribute for the
433:    The new AsyncClient class implements an asynchronous XMLRPC client.      o Regexp.PCRE.Widestring +     This feature is now reported in the basic feature list (pike    --features).   
513:    added to an RDF set.      o sprintf +  +  sprintf() and friends now attempt to perform compile-time argument +  checking. +     sprintf() now attempts to lookup the name of the program when    formatting types of objects and programs.      o gethrtime, gethrvtime, gauge -  +     Added support for POSIX style timers using clock_gettime(3). Notably    this fixes nice high resolution thread local cpu time and monotonic    real time on reasonably modern Linux systems.
533:   o ADT.Trie      o ADT.Set +     ADT.Set implements a datatype for sets. These sets behave much    like multisets, except that they are restricted to containing only    one instance of each member value.
543:    and code readability.      o Arg +     The new argument parser module allows for Getopt style arugment    parsing, but with a much simpler and object oriented API.   
571:    }      o Protocols.DNS_SD +     This module provides an interface to DNS Service Discovery. The    functionality of DNS-SD is described at <http://www.dns-sd.org/>.   
589:    of the Service class is valid.      o Bittorrent.Tracker +     Bittorrent tracker with support for scraping and UDP extension.      o Protocols.HTTP.Server.Proxy -  +     A simple HTTP proxy.      o Standards.TLD -  +     Country domains and other TLDs according to IANA. Useful when    parsing log information or providing live information about clients    connecting to your server.      o Tools.Shoot -  +     Several new tests have been added to benchmark and improve on    various aspects of Pike. ReplaceParallel and ReplaceSerial compares    the time it takes to replace multiple substrings of a string in a
611:    task, to remove XML tags from a string.      o Web.CGI +     Provides a CGI interface on the callee side. Retrieves information    from environment variables and populates the variables in the    Request object.
640:    source file or starting Pike with -V7.6.      o Array.transpose_old +     This function is removed.      o _Charset -  +     This internal module has had most of its content removed.    -  +  FIXME: Que?????????? +    o Crypto -  The old crypto functions from Pike 7.4 has been removed. These +  +  The old crypto functions from Pike 7.4 have been removed. These    functions produced a warning when used in Pike 7.6.      o Debug.describe_program -  +     The API for this debug function has changed.      o Image.Image -  +     The functions select_colors(), map_closest(), map_fast() and    map_fs() has been removed. Use Image.Colortable operations instead.      o Parser.XML -  +     The XML parsers are now stricter in verifying the correctness    of the XML. The function compat_allow_errors can be called in the    create method of the Parser.XML.Simple and Parser.XML.Validating
664:    be created with PARSE_COMPAT_ALLOW_ERRORS_7_6 as flag.      o Protocols.LDAP.client +     The "dn" field wasn't properly utf-8 decoded in 7.6 and earlier. If    your application does it yourself, you need to use the compatibility    version of this class.      o spider.XML -  +     The spider module no longer contains the XML parser. The functions    isbasechar(), iscombiningchar(), isdigit(), isextender(),    isfirstnamechar(), ishexchar(), isidiographic(), isletter(),    isnamechar() and isspace() has also been moved to the Parser module.      o Sql.Sql -  +     Pike will no longer create a .column entry in SQL query responses if    there is no table name.      o Standards.UUID -  Functions new() and new_string() has been removed. Use +  +  Functions new() and new_string() have been removed. Use    make_version1(-1)->encode() and make_version1(-1)->str() instead.      o Stdio -  +     The functions read_file(), read_bytes(), write_file() and -  append_file() now always throw errors upon errors, to allow easier -  use as errno doesn't have to be checked. read_file() and -  read_bytes() still returns 0 if the file does not exists. +  append_file() now always throw errors on error conditions, +  to allow easier use as errno doesn't have to be checked. +  read_file() and read_bytes() still return 0 if the file +  does not exist.    - o The modules Mird, Perl, Pipe and Ssleay have been removed. + o The modules Mird, Perl and Ssleay have been removed.         -------------------- NOTES------------