<?xml version="1.0" encoding="utf-8"?>
<feed xml:lang="en" xmlns="http://www.w3.org/2005/Atom"><title>Recent posts to news</title><link href="https://sourceforge.net/p/tass64/news/" rel="alternate"/><link href="https://sourceforge.net/p/tass64/news/feed.atom" rel="self"/><id>https://sourceforge.net/p/tass64/news/</id><updated>2025-05-10T05:33:04.921000Z</updated><subtitle>Recent posts to news</subtitle><entry><title>64tass 1.60.3243</title><link href="https://sourceforge.net/p/tass64/news/2025/05/64tass-1603243/" rel="alternate"/><published>2025-05-10T05:33:04.921000Z</published><updated>2025-05-10T05:33:04.921000Z</updated><author><name>Soci/Singular</name><uri>https://sourceforge.net/u/soci/</uri></author><id>https://sourceforge.net6cde0fede4d620e4a73b20f559c2b2ba05de969b</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Time to make another release as it'd be good to get the fixes out.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Fixes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Rename memalign function&lt;br/&gt;
  Caused compilation failures if such function existed in standard library&lt;/li&gt;
&lt;li&gt;Struct members offsets were still addresses sometimes&lt;br/&gt;
  This could cause wrong diagnostic messages (#71)&lt;/li&gt;
&lt;li&gt;Improve indirect alignment directives&lt;br/&gt;
  In some cases they failed even if the alignment was doable&lt;/li&gt;
&lt;li&gt;Wrong value in addressing limit messages&lt;br/&gt;
  The full address mode was displayed while only the address value is relevant&lt;/li&gt;
&lt;li&gt;Typo causing various issues on win32 &lt;br/&gt;
  Especially on newer versions. (#75)&lt;/li&gt;
&lt;li&gt;Listing file error got lost&lt;br/&gt;
  If an error occurred on creation of a listing file it wasn't reported&lt;/li&gt;
&lt;li&gt;Original definition note message even when location was not known&lt;br/&gt;
  Without the location it's useless, so it's avoided now.&lt;/li&gt;
&lt;li&gt;BBR and BBS long branches&lt;br/&gt;
  These broke at some point but are working now again.&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;Changes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;pow() now the same as power operator&lt;br/&gt;
  This also means powers of integers remain integer now if possible&lt;/li&gt;
&lt;li&gt;.from not limited to namespaces&lt;br/&gt;
  Symbols can come from dictionaries as well&lt;/li&gt;
&lt;li&gt;More specific error messages for some unlikely corner cases&lt;br/&gt;
  Out of memory errors were replaced with better fitting (and less fatal) messages.&lt;/li&gt;
&lt;li&gt;Improve first parameter of regular for loops&lt;br/&gt;
  Can be a compound assignment now, not just an assignment.&lt;/li&gt;
&lt;li&gt;hypot(a, b, c)&lt;br/&gt;
  Not restricted to two dimensions anymore&lt;/li&gt;
&lt;li&gt;log(v, base)&lt;br/&gt;
  The base of the logarithm can be specified now as an optional parameter&lt;/li&gt;
&lt;li&gt;Long branches can return now&lt;br/&gt;
  If the destination is a return then that'll be used instead of a jump.&lt;/li&gt;
&lt;li&gt;Hint on .enc and .cpu quoting&lt;br/&gt;
  There's a hint now that the argument needs to be quoted&lt;/li&gt;
&lt;li&gt;-X applies on s-record now&lt;br/&gt;
  Can be used to avoid S19 and force output to be at least S28.&lt;/li&gt;
&lt;li&gt;Searched in global scope note was removed&lt;br/&gt;
  Wasn't really useful as there's no source location it could point to.&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;New&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Symbol type&lt;br/&gt;
  Not all that new but it got a built-in name and can be converted from string&lt;/li&gt;
&lt;li&gt;Dictionaries from separate keys and values. dict(.(a, b, c), (1,2,3))&lt;br/&gt;
  This allows easier creation of enumerations from ranges of values&lt;/li&gt;
&lt;li&gt;Column oriented iterator .for loop&lt;br/&gt;
  For parallel iteration of several lists without indexing.&lt;/li&gt;
&lt;li&gt;PSH and PUL (RSH, RUL)&lt;br/&gt;
  Pseudo instructions for generic pushing and pulling (mostly registers)&lt;/li&gt;
&lt;li&gt;LDR, STR&lt;br/&gt;
  Pseudo instructions for generic load and store based on register parameter&lt;/li&gt;
&lt;li&gt;New cpu type 45gs02&lt;br/&gt;
  For mega65 projects. Please read the manual for mnemonics.&lt;/li&gt;
&lt;li&gt;--ctag-labels&lt;br/&gt;
  Some editors support it for jump to definition functionality.&lt;/li&gt;
&lt;li&gt;--cody-bin&lt;br/&gt;
  Simple output format used by Cody.&lt;/li&gt;
&lt;li&gt;--wdc-bin&lt;br/&gt;
  Output format used by WDC.&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;Possible incompatibilities&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Column oriented iterator for loop syntax&lt;br/&gt;
  Before the column oriented for loop one could get away with 'in [1,2],[3,4]'&lt;br/&gt;
  as an equivalent of 'in ([1,2],[3,4])'. As the former coma separated&lt;br/&gt;
  version is now the column oriented syntax the tuple must be explicitly&lt;br/&gt;
  written out. The latter version is backwards compatible.&lt;/li&gt;
&lt;li&gt;New pseudo instructions vs labels&lt;br/&gt;
  The new pseudo instructions might conflict with existing labels without&lt;br/&gt;
  colons. Please add colons in such cases.&lt;/li&gt;
&lt;li&gt;Long branch and always taken branch pseudo instruction changes&lt;br/&gt;
  The resulting binary might differ due to improvements. The main reason is&lt;br/&gt;
  that these might use returns now in some cases and branch merging got &lt;br/&gt;
  more conservative now.&lt;/li&gt;
&lt;/ul&gt;&lt;/div&gt;</summary></entry><entry><title>64tass 1.59.3120</title><link href="https://sourceforge.net/p/tass64/news/2023/09/64tass-1593120/" rel="alternate"/><published>2023-09-10T16:24:00.725000Z</published><updated>2023-09-10T16:24:00.725000Z</updated><author><name>Soci/Singular</name><uri>https://sourceforge.net/u/soci/</uri></author><id>https://sourceforge.netdd218f551fd6b5b9673e8b7ee3a53cf135d72a97</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;About time to make another release. There are some fixes for reported&lt;br/&gt;
problems (thanks!) and for some more I've noticed myself. For new&lt;br/&gt;
features check below.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Fixes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Line numbers didn't always show up in listing file when requested.&lt;br/&gt;
  Was a regression, there was a workaround but it's not necessary now&lt;/li&gt;
&lt;li&gt;Listing file didn't listed labels at the end of functions/macros.&lt;br/&gt;
  Now it does&lt;/li&gt;
&lt;li&gt;When .binclude failed it could cause an unexpected too many passes error.&lt;br/&gt;
  No such surprises now on filename typos.&lt;/li&gt;
&lt;li&gt;Some versions of GCC miscompiled the source with LTO optimization&lt;br/&gt;
  It resulted in an immediate crash at startup. It's worked around now.&lt;/li&gt;
&lt;li&gt;Possible issues fixed with (0, x) style indexed indirect addressing&lt;br/&gt;
  This workaround should have been done more carefully. Should work now.&lt;/li&gt;
&lt;li&gt;When outputting a specific section the output was incomplete in some cases&lt;br/&gt;
  I usually used complicated arrangements which worked so it was&lt;br/&gt;
  surprising to find that in special a simple one it didn't&lt;/li&gt;
&lt;li&gt;Parsing errors could cause unnecessary undefined label errors&lt;br/&gt;
  It does not do this any more, so less noise in error messages.&lt;/li&gt;
&lt;li&gt;Zero length concatenation of inverted bits/bytes was not correct&lt;br/&gt;
  A very special case.&lt;/li&gt;
&lt;li&gt;The order of indirect parameters in format() was wrong (broke in r2345)&lt;br/&gt;
  Haven't used it until now and was surprised to find this.&lt;/li&gt;
&lt;li&gt;The line number was at the wrong place in listing in some cases&lt;br/&gt;
  Now it's always in front where it should be.&lt;/li&gt;
&lt;li&gt;There was a complicated way to cause a crash through encoding definition&lt;br/&gt;
  One of those what would happen if experiments. Unlikely anyone&lt;br/&gt;
  encountered this.&lt;/li&gt;
&lt;li&gt;Katepart syntax file was incompatible with newer version&lt;br/&gt;
  There were silly mistakes in XML which were accepted in earlier versions.&lt;/li&gt;
&lt;li&gt;Possible crash fixed related to iterated bfor error handling&lt;br/&gt;
  The error case handling had a mistake.&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;Changes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;.align and .page directives have offset parameters now.&lt;br/&gt;
  It can be useful to shift the page windows up/down&lt;/li&gt;
&lt;li&gt;Listing time uses file modification times if possible&lt;br/&gt;
  So the time stamp can remain the same across re-compilations&lt;/li&gt;
&lt;li&gt;Improved documentation&lt;br/&gt;
  .page moved to the new alignment section. And lots of other fixes.&lt;/li&gt;
&lt;li&gt;Deprecation warning for non-spaced @b, @w and @l&lt;br/&gt;
  It's planned that more than one letter will be used after the @ symbol&lt;br/&gt;
  in the future. Once that's implemented any following non-spaced&lt;br/&gt;
  separated label or number will become part of the symbol! So it's&lt;br/&gt;
  advised now to add the missing spaces to avoid broken code after a&lt;br/&gt;
  future upgrade.&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;New&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Added -Wpriority command line option&lt;br/&gt;
  Common operator priority mistakes like &amp;gt;lbl1 != &amp;gt;lbl2 are reported now.&lt;/li&gt;
&lt;li&gt;Added -Walign command line option&lt;br/&gt;
  Wasted space on doing alignments can be diagnosed now.&lt;/li&gt;
&lt;li&gt;Added .alignblk/.endalignblk directives&lt;br/&gt;
  Alignment block for avoiding page crosses without wasting to many bytes&lt;/li&gt;
&lt;li&gt;Added .alignind and .alignpageind directives&lt;br/&gt;
  It's now possible to align further ahead than the target location&lt;/li&gt;
&lt;li&gt;Added --simple-labels command line option&lt;br/&gt;
  Label files in traditional label=$1234 style for simple parsing.&lt;/li&gt;
&lt;li&gt;Added --mesen-labels command line option&lt;br/&gt;
  Experimental Mesen label file output.&lt;/li&gt;
&lt;li&gt;Added --labels-add-prefix command line option&lt;br/&gt;
  When listing labels it's possible to add a prefix to symbols now&lt;/li&gt;
&lt;li&gt;Added --labels-section command line option&lt;br/&gt;
  Label listing can be limited to labels pointing to a certain section.&lt;/li&gt;
&lt;li&gt;Added .from directive&lt;br/&gt;
  For defining a symbol based on an already existing one somewhere else&lt;/li&gt;
&lt;li&gt;Added --dependencies-append command line option&lt;br/&gt;
  For consistency.&lt;/li&gt;
&lt;li&gt;Added .elif directive&lt;br/&gt;
  An alias for those who tend to mistype .elsif&lt;/li&gt;
&lt;li&gt;Added --c256-pgx and --c256-pgz command line options&lt;br/&gt;
  Both are simple formats and a variant of existing ones.&lt;/li&gt;
&lt;li&gt;Binary and hexadecimal can be bits instead of int.&lt;br/&gt;
  So format("%~#x", ~$2) is "~$2".&lt;/li&gt;
&lt;li&gt;Added --output-exec command line option&lt;br/&gt;
  Now the execution address can be defined for those output which have&lt;br/&gt;
  such concept&lt;/li&gt;
&lt;/ul&gt;&lt;/div&gt;</summary></entry><entry><title>64tass 1.58.2974</title><link href="https://sourceforge.net/p/tass64/news/2023/01/64tass-1582974/" rel="alternate"/><published>2023-01-11T07:36:38.851000Z</published><updated>2023-01-11T07:36:38.851000Z</updated><author><name>Soci/Singular</name><uri>https://sourceforge.net/u/soci/</uri></author><id>https://sourceforge.net356fe318b653bfd2b734bdd83ca788319cf132c5</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;This is a bugfix release due to the broken multi file output. There were&lt;br/&gt;
a bunch of encoding related work meanwhile so these are included as well.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Fixes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;There was a mistake which could cause crashes (usually on Windows)&lt;br/&gt;
  when outputting sections to multiple files.&lt;/li&gt;
&lt;li&gt;Linking to older versions of MS C library could result in unexpected&lt;br/&gt;
  behaviour when Unicode characters were attempted to be output due to&lt;br/&gt;
  differences to ISO-C. Such functions are not used anymore. The&lt;br/&gt;
  precompiled binary used a recent enough toolchain and so was not&lt;br/&gt;
  affected.&lt;/li&gt;
&lt;li&gt;Map file output not written to a file could still report a file open&lt;br/&gt;
  error if the standard output was redirected and that couldn't be written.&lt;/li&gt;
&lt;li&gt;In some cases ANSI colour sequences could end up in text output to files.&lt;br/&gt;
  This shouldn't happen now.&lt;/li&gt;
&lt;li&gt;Memory maps written to files were not added to makefile as outputs.&lt;/li&gt;
&lt;li&gt;The Notepad++ syntax file was incompatible with dark-mode as it set the&lt;br/&gt;
  text background to white instead of transparent.&lt;/li&gt;
&lt;li&gt;When running on DOS the code page was not detected with some versions&lt;br/&gt;
  of the OS.&lt;/li&gt;
&lt;li&gt;Attempting to use it on Windows versions without wide character&lt;br/&gt;
  support resulted in immediate exit. Now it can be used as normal.&lt;/li&gt;
&lt;li&gt;When trying to use options with non-ASCII characters the error message&lt;br/&gt;
  showed garbage.&lt;/li&gt;
&lt;li&gt;Coloured diagnostic messages only worked in console window but not in&lt;br/&gt;
  mintty (unless compiled for Cygwin). Now it should work with the&lt;br/&gt;
  native version as well.&lt;/li&gt;
&lt;li&gt;Depending on the C library memory map output could hang due to a&lt;br/&gt;
  mistake. (e.g. on Amiga)&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;Changes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;On Windows the file encoding for "ANSI" files were assumed to use the&lt;br/&gt;
  encoding from the current locale. However this can be different from&lt;br/&gt;
  the active code page what text editors are using. Now such files are&lt;br/&gt;
  assumed to use the active code page unless the locale is set differently&lt;br/&gt;
  in the environment. Regardless UTF-8 sources are best for portability.&lt;/li&gt;
&lt;li&gt;On Windows text output (e.g. listing) was UTF-8 or ASCII depending on&lt;br/&gt;
  the toolchain used. Now the default encoding is the active code page.&lt;br/&gt;
  This can be UTF-8 on recent systems but in any case it can be forced by&lt;br/&gt;
  setting an UTF-8 locale through the environment.&lt;/li&gt;
&lt;li&gt;Depending on C library version printable Unicode characters showed&lt;br/&gt;
  placeholders even if the encoding would allow for it (e.g. UTF-8).&lt;br/&gt;
  This is now determined internally to be consistent across platforms.&lt;/li&gt;
&lt;li&gt;The subset of Unicode characters usable in identifiers got more strict&lt;br/&gt;
  now.&lt;/li&gt;
&lt;li&gt;Updated to Unicode 13.&lt;/li&gt;
&lt;/ul&gt;&lt;/div&gt;</summary></entry><entry><title>64tass 1.57.2900</title><link href="https://sourceforge.net/p/tass64/news/2022/11/64tass-1572900/" rel="alternate"/><published>2022-11-05T12:46:37.542000Z</published><updated>2022-11-05T12:46:37.542000Z</updated><author><name>Soci/Singular</name><uri>https://sourceforge.net/u/soci/</uri></author><id>https://sourceforge.net2806825d59a5f1315a798da90d399801a67aad96</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;It's time to release a version now. People are looking for fixes which were in&lt;br/&gt;
trunk for a long time.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Fixes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Paths with parent references don't give bogus portability warnings any more&lt;br/&gt;
  On Windows the due to an oversight parent referencing paths incorrectly&lt;br/&gt;
  triggered the portability warning originally intended for detecting a&lt;br/&gt;
  filename case mismatch.&lt;/li&gt;
&lt;li&gt;Arbitrary length int right shift was wrong in some cases.&lt;br/&gt;
  Now it's correct.&lt;/li&gt;
&lt;li&gt;.bfor didn't accepted .endfor closing in some cases&lt;br/&gt;
  Now it's correct. Did always work with .next and I didn't transitioned the&lt;br/&gt;
  test sources...&lt;/li&gt;
&lt;li&gt;(0, x) is now recognized as indexed indirect for opcodes&lt;br/&gt;
  The parser of course did recognize (0,x) just fine for opcodes. However&lt;br/&gt;
  writing (0, x) is a tuple and so does something else.  There's a workaround&lt;br/&gt;
  now to recognize such a construct as indexed indirect instead. This should&lt;br/&gt;
  avoid bad surprises for those unaware that the indexing operator should not&lt;br/&gt;
  have a white space there.&lt;/li&gt;
&lt;li&gt;Writing to standard output suppresses messages in all cases&lt;br/&gt;
  This only worked for output files but now for everything (e.g. listing&lt;br/&gt;
  to "-" didn't suppressed messages while --output "-" did).&lt;/li&gt;
&lt;li&gt;The TASM compatibility text macro parameter expansion was wrong&lt;br/&gt;
  It didn't remove the quotes from strings. It broke quite long ago, now&lt;br/&gt;
  works again.&lt;/li&gt;
&lt;li&gt;Output append didn't reproduce gaps working correctly in some cases&lt;br/&gt;
  Padding was corrected and now it does.&lt;/li&gt;
&lt;li&gt;Output could have garbage in gaps on non-POSIX conforming systems&lt;br/&gt;
  Unlikely anyone got that far in porting but gaps should be correctly&lt;br/&gt;
  filled with zeros now&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;Changes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;JMP and JML are distinct now&lt;br/&gt;
  JMP could be unexpectedly upgraded to JML in some cases (e.g. bank 0 address in&lt;br/&gt;
  another program bank) which is bad as the program bank will be changed when&lt;br/&gt;
  that happens. Now it's consistent with JSR/JSL.&lt;/li&gt;
&lt;li&gt;label = 1,2,3 is the tuple (1,2,3) now&lt;br/&gt;
  That makes it actually useful for something. Before it was an object&lt;br/&gt;
  with one couldn't do much with.&lt;/li&gt;
&lt;li&gt;.switch and .if blocks labels have size now&lt;br/&gt;
  Everything else with regular code blocks worked that way already.&lt;/li&gt;
&lt;li&gt;%c in format() applies padding now&lt;br/&gt;
  For consistency with the rest of the format parameters.&lt;/li&gt;
&lt;li&gt;Oversize integer diagnostic in bytes and bits now&lt;br/&gt;
  It was in bits all the time but if the limit was set in bytes it makes&lt;br/&gt;
  sense to give the error in bytes as well.&lt;/li&gt;
&lt;li&gt;Compound assignments can now change variables in parent scope&lt;br/&gt;
  Without this using .bwhile is a pain as the loop condition variable is&lt;br/&gt;
  usually declared out of the loop (and so the scope).&lt;/li&gt;
&lt;li&gt;The memory map was improved or at least is different now.&lt;br/&gt;
  With the former memory map there were lots of redundancy if one used&lt;br/&gt;
  sections. It was even worse with nested sections. The new memory map&lt;br/&gt;
  tries to display overlaps more compact. However for most people with simple&lt;br/&gt;
  projects the difference would be the length display in decimal on the left&lt;br/&gt;
  and that gaps between data areas are shown now.&lt;/li&gt;
&lt;li&gt;Per output memory map, also writable to file.&lt;br/&gt;
  The memory map was global before. Now if a specific section is output&lt;br/&gt;
  then only the relevant parts of the map is shown. This also means that&lt;br/&gt;
  if there are multiple output files then there are as many memory maps.&lt;br/&gt;
  As a side effect .virtual blocks aren't displayed any more as these&lt;br/&gt;
  don't really belong to an output file.&lt;/li&gt;
&lt;li&gt;The .enc directive now requires an expression&lt;br/&gt;
  In the past one could write the name of encodings without quotes. Soon there&lt;br/&gt;
  was a need to allow labels to hold a name so there was a warning to quote the&lt;br/&gt;
  names for the past 5 years or so. Now that custom encodings can be assigned&lt;br/&gt;
  to labels there's an even greater need to be able to use labels. Therefore&lt;br/&gt;
  the old style unquoted encoding names will be now interpreted as labels and&lt;br/&gt;
  will likely fail to be found. Please quote the old name or as a workaround&lt;br/&gt;
  define the missing labels with a string of the encoding name (e.g screen =&lt;br/&gt;
  "screen").&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;New&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;.symbol in namespace&lt;br/&gt;
  Now it's possible to check if a variable exists in a namespace. May be useful&lt;br/&gt;
  to check if a variable is defined. No, this does not work with constant&lt;br/&gt;
  definitions.&lt;/li&gt;
&lt;li&gt;Function parameter conversion&lt;br/&gt;
  This allows to have a parameter like "data : binary" which calls&lt;br/&gt;
  binary() and its result is assigned to the parameter. This allows to&lt;br/&gt;
  load data relative to the caller's file and not relative to the file the&lt;br/&gt;
  function is located in.&lt;/li&gt;
&lt;li&gt;!in and !== operator&lt;br/&gt;
  Instead of !(a in b) now it's possible to write a !in b. Also the ===&lt;br/&gt;
  operator got its !== pair.&lt;/li&gt;
&lt;li&gt;Multi operand opcodes do broadcast now&lt;br/&gt;
  It was only working for single operand ones so far, now it's consistent.&lt;/li&gt;
&lt;li&gt;Bit strings can be iterated now&lt;br/&gt;
  Just like character and byte strings did already&lt;/li&gt;
&lt;li&gt;::= reassign operator&lt;br/&gt;
  This can change a variable in parent scope. Unlike := which instead&lt;br/&gt;
  would create/update a variable in the current scope instead.&lt;/li&gt;
&lt;li&gt;Better reporting of unexpected characters&lt;br/&gt;
  Strings should be quoted with straight quotes. Using fancy quotes did&lt;br/&gt;
  result in an error but it was difficult to make sense of it as the&lt;br/&gt;
  difference is hard to spot. Now the diagnostic is more clear.&lt;/li&gt;
&lt;li&gt;--mos-hex for MOS Technology HEX format&lt;br/&gt;
  There was Intel and Motorola before but not MOS yet. There are 6502 platforms&lt;br/&gt;
  needing this.&lt;/li&gt;
&lt;li&gt;--list-append added&lt;br/&gt;
  Now listing can be appended too, just in case.&lt;/li&gt;
&lt;li&gt;.encode/.endencode encoding area&lt;br/&gt;
  This allows to have local encoding definitions. Such definitions can be&lt;br/&gt;
  assigned to a label to allow reuse in other areas. Can be used as string&lt;br/&gt;
  conversion function as well.&lt;/li&gt;
&lt;li&gt;Added .tdef for easier encoding definition&lt;br/&gt;
  It was not easy to define translation for randomly scattered Unicode&lt;br/&gt;
  characters so far as .cdef was made for ranges like A-Z. One had to&lt;br/&gt;
  define a bunch of single character ranges but such a workaround can be&lt;br/&gt;
  replaced now.&lt;/li&gt;
&lt;/ul&gt;&lt;/div&gt;</summary></entry><entry><title>64tass 1.56.2625</title><link href="https://sourceforge.net/p/tass64/news/2021/04/64tass-1562625/" rel="alternate"/><published>2021-04-25T21:23:47.091000Z</published><updated>2021-04-25T21:23:47.091000Z</updated><author><name>Soci/Singular</name><uri>https://sourceforge.net/u/soci/</uri></author><id>https://sourceforge.neteeffcfbaadaa5267837e7f0057adb96319e8d601</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;The last release got more than a year old now and there were&lt;br/&gt;
plenty of pending corrections to be released. Some planned&lt;br/&gt;
features still didn't made it in. Maybe next time.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Fixes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;.fill/.align multi byte pattern fix&lt;br/&gt;
  Broke due to changes in r1845.&lt;/li&gt;
&lt;li&gt;Wrong diagnostic for incomplete byte string&lt;br/&gt;
  The missing closing quote was pointed at the wrong place&lt;/li&gt;
&lt;li&gt;Incomplete diagnostic message&lt;br/&gt;
  The opcode was not displayed in some cases&lt;/li&gt;
&lt;li&gt;Diagnostic locations fixed&lt;br/&gt;
  In some cases the location was wrong&lt;/li&gt;
&lt;li&gt;Duplicate diagnostic messages avoided&lt;br/&gt;
  In some cases redundant diagnostics was output&lt;/li&gt;
&lt;li&gt;Crash while .function declaration and macro init parameters&lt;br/&gt;
  Found by Pierre&lt;/li&gt;
&lt;li&gt;Improved filename portability checks&lt;br/&gt;
  In some cases case insensitive filenames were not recognized&lt;/li&gt;
&lt;li&gt;Diagnostics of non-portable filename characters improved&lt;br/&gt;
  Now points at the problematic character.&lt;/li&gt;
&lt;li&gt;Functions did not execute as loop and other conditions&lt;br/&gt;
  Now they do.&lt;/li&gt;
&lt;li&gt;Can't calculate errors related to functions&lt;br/&gt;
  In some cases these were incorrectly reported&lt;/li&gt;
&lt;li&gt;4510 missed had () instead of (),z addressing mode&lt;br/&gt;
  Now same as 65CE02 in this regard&lt;/li&gt;
&lt;li&gt;Improved format() diagnostics&lt;br/&gt;
  It's more strict now&lt;/li&gt;
&lt;li&gt;Floating point bitops didn't work as intended&lt;br/&gt;
  Actually in some cases they did work but mostly not&lt;/li&gt;
&lt;li&gt;Dot eor operator works again in TASM compatible mode&lt;br/&gt;
  Got broken due to syntax clash with hexadecimal floats&lt;/li&gt;
&lt;li&gt;Wrong message text for invalid unary operators&lt;br/&gt;
  Tried to diagnose them like binary by mistake&lt;/li&gt;
&lt;li&gt;Improved syntax highlighting&lt;br/&gt;
  In some case it didn't highlight well on vim&lt;/li&gt;
&lt;li&gt;Byte string compare didn't work properly&lt;br/&gt;
  In case a byte string was inverted.&lt;/li&gt;
&lt;li&gt;In some cases .null failed to detect zeros&lt;br/&gt;
  For example for ~$ff&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;Changes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Comment nesting&lt;br/&gt;
  Directives are not recognized now in commented block&lt;/li&gt;
&lt;li&gt;Negative size an error now, size extension a warning&lt;br/&gt;
  If an offset reduces a memory area beyond 0 it's a fault now&lt;br/&gt;
  (instead of 0) and extension beyond the original size a warning.&lt;br/&gt;
  Controllable through -Wsize-large.&lt;/li&gt;
&lt;li&gt;PC bank wrapping gone now&lt;br/&gt;
  Seemed like a good idea but crossing banks with long data gave&lt;br/&gt;
  unexpected results for many people.&lt;/li&gt;
&lt;li&gt;VICE label de-duplication is gone now&lt;br/&gt;
  Was intended to avoid warnings on import but actually breaks&lt;br/&gt;
  symbolic breakpoints if the needed label gets deduplicated...&lt;/li&gt;
&lt;li&gt;-Wswitch-case gone now&lt;br/&gt;
  Could cause unexpected side effects which is a no-no for diagnostics&lt;/li&gt;
&lt;li&gt;Ident renamed to symbol&lt;br/&gt;
  For less confusing diagnostic messages&lt;/li&gt;
&lt;li&gt;Long values not displayed in listing by default&lt;br/&gt;
  They cluttered up the listing. They are still displayed in verbose&lt;br/&gt;
  mode.&lt;/li&gt;
&lt;li&gt;Dictionary keys must be identical and not equal&lt;br/&gt;
  Otherwise bit strings can't be matched properly for example.&lt;/li&gt;
&lt;li&gt;Hex float format from format() gone now&lt;br/&gt;
  Seemed like it's easy to add without much effort but it was not cross&lt;br/&gt;
  platform.&lt;/li&gt;
&lt;li&gt;Bit string extraction operators now return bit string&lt;br/&gt;
  Did return bytes for a long while but now I realized those may not&lt;br/&gt;
  compare as expected in some cases.&lt;/li&gt;
&lt;li&gt;Multi line expressions are gone now&lt;br/&gt;
  Seemed like a good idea back then but they cause more problems than&lt;br/&gt;
  they solve. They were never documented due to all the unsolved edge&lt;br/&gt;
  cases anyway.&lt;/li&gt;
&lt;li&gt;Duplicated label files do not append automatically&lt;br/&gt;
  It must be done now explicitly with --labels-append&lt;/li&gt;
&lt;li&gt;Unnamed values not listed in listing any more&lt;br/&gt;
  There's no label so makes no sense&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;New&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Immediate operand to branches&lt;br/&gt;
  It's now possible to write 'bne #0' (for self modifying code)&lt;/li&gt;
&lt;li&gt;Hexadecimal byte string literal separation with spaces&lt;br/&gt;
  Space is accepted now in hexadecimal byte strings like x'ab cd'&lt;/li&gt;
&lt;li&gt;Various byte string functions added&lt;br/&gt;
  Like byte(), word() and friends&lt;/li&gt;
&lt;li&gt;New --output-append parameter&lt;br/&gt;
  For appending the output to a file instead of overwriting it&lt;/li&gt;
&lt;li&gt;New --no-output option&lt;br/&gt;
  For cases where no output file is needed&lt;/li&gt;
&lt;li&gt;Symbol definition syntax&lt;br/&gt;
  May be defined with a dot prefix. Useful as dictionary keys.&lt;/li&gt;
&lt;li&gt;New .sfunction directive&lt;br/&gt;
  For defining single line functions which only do calculations&lt;/li&gt;
&lt;li&gt;Alternative end directives&lt;br/&gt;
  Bit more verbose but more consistent and easier to remember.&lt;br/&gt;
  E.g. .endpage vs. .endp&lt;/li&gt;
&lt;li&gt;Identity operator (===)&lt;br/&gt;
  For exact matching&lt;/li&gt;
&lt;li&gt;--no-error and --error-append&lt;br/&gt;
  To not save or append error messages&lt;/li&gt;
&lt;li&gt;Broadcasting conditional operator is '??'&lt;br/&gt;
  Sometimes broadcasting is needed but that's not compatible with the&lt;br/&gt;
  regular '?' conditional operator&lt;/li&gt;
&lt;li&gt;Diagnostics for interrupted compilation&lt;br/&gt;
  Now it's displayed where it happened for improved infinite loop&lt;br/&gt;
  debugging.&lt;/li&gt;
&lt;li&gt;--labels-append&lt;br/&gt;
  For appending to label files.&lt;/li&gt;
&lt;/ul&gt;&lt;/div&gt;</summary></entry><entry><title>64tass 1.55.2200</title><link href="https://sourceforge.net/p/tass64/news/2020/04/64tass-1552200/" rel="alternate"/><published>2020-04-07T19:43:08.573000Z</published><updated>2020-04-07T19:43:08.573000Z</updated><author><name>Soci/Singular</name><uri>https://sourceforge.net/u/soci/</uri></author><id>https://sourceforge.netc9d4cf1703611b35bca2243db664ea7d5a88969c</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;This is a bug fix release to fix the unintended performance degradation on some&lt;br/&gt;
systems like BSDs.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Fixes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Revert buffering changes&lt;br/&gt;
  These caused really slow performance on some systems which I forgot to test.&lt;/li&gt;
&lt;li&gt;Makefile name escaping&lt;br/&gt;
  It's not the same as for shells and is more limited. So names with spaces&lt;br/&gt;
  should work now but otherwise it's advised to avoid incompatible file naming.&lt;/li&gt;
&lt;li&gt;Bold black is not be visible on black background&lt;br/&gt;
  Changed to cyan.&lt;/li&gt;
&lt;li&gt;Unicode parameter file names didn't work on some platforms&lt;br/&gt;
  Forgot a conversion so only UTF-8 worked&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;New&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Coloured output on windows console and DOS.&lt;br/&gt;
  This was available for a while but not for these platforms. Real DOS needs&lt;br/&gt;
  ANSI loaded and TERM set.&lt;/li&gt;
&lt;/ul&gt;&lt;/div&gt;</summary></entry><entry><title>64tass 1.55.2176</title><link href="https://sourceforge.net/p/tass64/news/2020/03/64tass-1552176/" rel="alternate"/><published>2020-03-23T20:39:27.642000Z</published><updated>2020-03-23T20:39:27.642000Z</updated><author><name>Soci/Singular</name><uri>https://sourceforge.net/u/soci/</uri></author><id>https://sourceforge.netc3e5da5f9c90e8bf05d8a2b6062938f4f15f1de7</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;There were quite many changes in trunk so it's time to release another version.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Changes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Removed -Wtype-mixing &lt;br/&gt;
  It was a transitional warning just in case someone relied on an undocumented&lt;br/&gt;
  tuple/list mixing which changed earlier.&lt;/li&gt;
&lt;li&gt;Floor rounding is used by default now, added -Wfloat-round&lt;br/&gt;
  The old method was to truncate but it was not consistent with the integer&lt;br/&gt;
  division and was problematic when using floating point to represent&lt;br/&gt;
  fixed point numbers. The difference only shows for negative numbers.&lt;br/&gt;
  Just to be on the safe side there's a new warning now for cases where&lt;br/&gt;
  the programmer could use one of the rounding functions explicitly to avoid&lt;br/&gt;
  a rounding ambiguity.&lt;/li&gt;
&lt;li&gt;Handling of address space wraps in calculations&lt;br/&gt;
  There are various address space wraps depending on the CPU. As an example the&lt;br/&gt;
  zero page may wrap or may not with indexed addressing. 64tass now tries to&lt;br/&gt;
  wrap when an offset is applied to an address to keep it in the same address&lt;br/&gt;
  space. E.g. a label at $10 in the expression lda label-$20,x results in&lt;br/&gt;
  lda $f0,x for a 6502 but lda $fffff0,x for a 65816. In all cases there are&lt;br/&gt;
  warnings issued for such an ambiguity. If it's intentional it may be&lt;br/&gt;
  suppressed but usually it's not.&lt;/li&gt;
&lt;li&gt;-Wmem-wrap and -Wpc-wrap renamed now to -Wwrap-mem and -Wwrap-pc&lt;br/&gt;
  To make them in line with the other bunch of new -Wwrap-something switches.&lt;br/&gt;
  Please update your build scripts. They won't fail just yet but may in future.&lt;/li&gt;
&lt;li&gt;Only --export-labels references labels&lt;br/&gt;
  Listing labels incorrectly referenced .proc blocks and so just by doing a&lt;br/&gt;
  label list unused .proc blocks got suddenly compiled in. This is unexpected&lt;br/&gt;
  and should not happen for a simple label list therefore it won't any more. If&lt;br/&gt;
  otherwise unused .proc blocks are needed in the label listing please add the&lt;br/&gt;
  new --export-labels option to get them listed and compiled.&lt;/li&gt;
&lt;li&gt;Star is finally a label now&lt;br/&gt;
  Before it was only a number but as the theory goes it should be the&lt;br/&gt;
  memory location at the beginning of the line. This means label = * + 1&lt;br/&gt;
  constructs are now correctly refering to memory locations.&lt;/li&gt;
&lt;li&gt;Variables are not listed in VICE label list any more&lt;br/&gt;
  They are ambiguous as may be assigned multiple times and usually don't refer&lt;br/&gt;
  to memory locations&lt;/li&gt;
&lt;li&gt;Dictionaries are order preserving and iterable now&lt;br/&gt;
  Earlier order was undefined.&lt;/li&gt;
&lt;li&gt;I/O error messages or in C locale now&lt;br/&gt;
  As the rest of the messages aren't localized either.&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;New&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Add -Wmacro-prefix&lt;br/&gt;
  Warns about macro calls without any prefix in front&lt;/li&gt;
&lt;li&gt;Add WDM opcode&lt;br/&gt;
  It was requested for 65816&lt;/li&gt;
&lt;li&gt;Added .while and .bwhile&lt;br/&gt;
  Could be done with .for before but it's cleaner this way&lt;/li&gt;
&lt;li&gt;Add .breakif and .continueif&lt;br/&gt;
  Simplifies .if .break .fi and .if .continue .fi sequences&lt;/li&gt;
&lt;li&gt;Functions got a namespace&lt;br/&gt;
  The function itself can hold symbols now so there's no need for an&lt;br/&gt;
  additional label just to store data across invocations. That is&lt;br/&gt;
  "functionname.stuff := 4" works now.&lt;/li&gt;
&lt;li&gt;Lists can be used as keys in dictionaries&lt;br/&gt;
  On query they need to be put into a list of course&lt;/li&gt;
&lt;li&gt;CPA alias was added for CMP&lt;br/&gt;
  For consistency with CPX and CPY.&lt;/li&gt;
&lt;li&gt;Added --macro-caret-diag&lt;br/&gt;
  Normally the diagnostic message is displayed along with the line causing it.&lt;br/&gt;
  There was an option to suppress the line display for those who don't want&lt;br/&gt;
  this but that makes debugging of macro expansions hard as the source file only&lt;br/&gt;
  contains the unexpanded macro. This option is a middle ground to display&lt;br/&gt;
  faulty lines only for macros.&lt;/li&gt;
&lt;li&gt;Added --vice-labels-numeric&lt;br/&gt;
  Normally only memory locations are included to avoid numeric constant&lt;br/&gt;
  definitions in the listing but if you have a lot of label = $3000 style&lt;br/&gt;
  definitions this is for you.&lt;/li&gt;
&lt;li&gt;Dictionaries are concatenable now&lt;br/&gt;
  There was hack to add entries by using the splat feature but the simpler&lt;br/&gt;
  {1:2}..{2:3} works now too.&lt;/li&gt;
&lt;li&gt;Dictionaries are sortable&lt;br/&gt;
  Now that the order is preserved and they are iterable this makes sense&lt;/li&gt;
&lt;li&gt;Multiple output files can be generated now&lt;br/&gt;
  By using multiple "--output" parameters. The output format and the section used&lt;br/&gt;
  may be varied of course.&lt;/li&gt;
&lt;li&gt;Broadcast works with format() as well&lt;br/&gt;
  That is now it's now possible to format values in a list and the result is a&lt;br/&gt;
  list of formatted values&lt;/li&gt;
&lt;li&gt;Added --make-phony command line option&lt;br/&gt;
  This is useful for automatic dependency generation purposes for Makefiles.&lt;/li&gt;
&lt;li&gt;Added .with/.endwith&lt;br/&gt;
  This allows to access symbols of another namespace without changing the&lt;br/&gt;
  current one. Unlike .namespace/.endn which does that.&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;Fixes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Now @b and @w works for branches&lt;br/&gt;
  It can be used to prevent automatic long branches where it'd be bad&lt;/li&gt;
&lt;li&gt;Fix AHX/SHA mode&lt;br/&gt;
  Wrong addressing mode was used&lt;/li&gt;
&lt;li&gt;Leading zeros warning message suppression was corrected&lt;br/&gt;
  Due to a copy mistake it was -Wimmediate instead of -Wleading-zeros&lt;/li&gt;
&lt;li&gt;Improve double definition error handling&lt;br/&gt;
  It worked before as well but caused so additional errors&lt;/li&gt;
&lt;li&gt;The "Too early to reference" error is gone now&lt;br/&gt;
  Now nothing is to early to reference&lt;/li&gt;
&lt;li&gt;Listing now contains returned values&lt;br/&gt;
  The results of function return values are in the listing now&lt;/li&gt;
&lt;li&gt;Unused .proc labels not in listing any more&lt;br/&gt;
  If unused technically the don't exists and should not be there&lt;/li&gt;
&lt;li&gt;Improved macro diagnostic message position for macros&lt;br/&gt;
  In many cases now the diagnostic messages are issued at the macro invocation&lt;br/&gt;
  instead in the macro body. At least when it seems like the problem is due to&lt;br/&gt;
  wrong parametrization.&lt;/li&gt;
&lt;li&gt;Addressing mode and size messages improved&lt;br/&gt;
  They display the opcode now which can't have that addressing mode or size&lt;/li&gt;
&lt;li&gt;Diagnostic backtrace now omitted for trivial cases&lt;br/&gt;
  This cuts back on the useless clutter for diagnostic messages in a multi file&lt;br/&gt;
  project&lt;/li&gt;
&lt;li&gt;Inverse command line options added for consistency&lt;br/&gt;
  For example there was a --quiet but no --no-quiet.&lt;/li&gt;
&lt;li&gt;VICE label lists omit duplicate entries now&lt;br/&gt;
  The monitor can only display one of them&lt;/li&gt;
&lt;li&gt;Various diagnostic message location improvements&lt;br/&gt;
  Some diagnostic messages pointed to the directive instead of it's parameter&lt;/li&gt;
&lt;li&gt;Functions may be created anywhere&lt;br/&gt;
  There were limitations on where functions may be defined but it's sorted now.&lt;/li&gt;
&lt;li&gt;Struct offsets are numbers now&lt;br/&gt;
  Of course they are but they were addresses earlier which caused -Wimmediate&lt;br/&gt;
  to not trigger on them.&lt;/li&gt;
&lt;li&gt;In some cases a missing .ends was not reported&lt;br/&gt;
  Now it is&lt;/li&gt;
&lt;/ul&gt;&lt;/div&gt;</summary></entry><entry><title>64tass 1.54.1900</title><link href="https://sourceforge.net/p/tass64/news/2019/02/64tass-1541900/" rel="alternate"/><published>2019-02-17T22:26:28.620000Z</published><updated>2019-02-17T22:26:28.620000Z</updated><author><name>Soci/Singular</name><uri>https://sourceforge.net/u/soci/</uri></author><id>https://sourceforge.netcc05a9c7502a192dea7ecf6223dcc44d26e657f9</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;Quick bugfix release to sort important regressions for file numbering, label files and *=/.offs.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Fixes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;File number in line numbered list file was wrong&lt;br/&gt;
  The parent was shown instead (which is not very useful)&lt;/li&gt;
&lt;li&gt;Empty data or code lines vs. --no-source&lt;br/&gt;
  Empty data lines are ommited now from the listing file&lt;/li&gt;
&lt;li&gt;Undefined symbol messages could be wrong in macros&lt;br/&gt;
  In some special cases it ended up as garbage&lt;/li&gt;
&lt;li&gt;Open/close directives shouldn't work across files&lt;br/&gt;
  Still worked with .include and the result was confusing&lt;/li&gt;
&lt;li&gt;VICE label file generation crash on unconvertable address&lt;br/&gt;
  In the unlikely case like 0,x and similar&lt;/li&gt;
&lt;li&gt;.offs now does what C64 TASM does&lt;br/&gt;
  This fixes a regression for old-school *=/.offs relocations&lt;/li&gt;
&lt;/ul&gt;&lt;/div&gt;</summary></entry><entry><title>64tass 1.54.1864</title><link href="https://sourceforge.net/p/tass64/news/2019/02/64tass-1541864/" rel="alternate"/><published>2019-02-03T22:29:56.857000Z</published><updated>2019-02-03T22:29:56.857000Z</updated><author><name>Soci/Singular</name><uri>https://sourceforge.net/u/soci/</uri></author><id>https://sourceforge.net21308348429209b8da1a59ed01e55813f4a20407</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;This release took a while as I wanted to test some of the features in practice first and those projects took a lot of time. Also some parts had to be rewritten as the implementation wasn't quite right. Then in the testing phase I've found a lot of problems and I wanted to solve them before doing a release.&lt;/p&gt;
&lt;p&gt;Anyway it's finally out now and I think it's a good improvement over r1515. If not then please let me know. &lt;/p&gt;
&lt;p&gt;&lt;strong&gt;New&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Add -Waltmode to show unavailable addressing mode fallbacks&lt;br/&gt;
   Data bank addressing may be bad sometimes instead of direct page&lt;/li&gt;
&lt;li&gt;Add ".for var in sequence" style iteration loops&lt;br/&gt;
  Looping over lists elements got easier now&lt;/li&gt;
&lt;li&gt;Add .namespace/.endn directives&lt;br/&gt;
  Now it's possible to switch to namespaces temporarily&lt;/li&gt;
&lt;li&gt;Add -Wno-page&lt;br/&gt;
  Now the .page/.endp error can be turned into a warning or off&lt;/li&gt;
&lt;li&gt;Add --output-section&lt;br/&gt;
  Output can be limited to a specific section instead of dumping everything&lt;/li&gt;
&lt;li&gt;Add binary() function&lt;br/&gt;
  This enables reading file content as a byte string for easier processing&lt;/li&gt;
&lt;li&gt;Add .virtual/.endv&lt;br/&gt;
  For compiling to an address but without producing any output&lt;/li&gt;
&lt;li&gt;.align works in structures now&lt;br/&gt;
  It aligns from the start of structure&lt;/li&gt;
&lt;li&gt;Folding operator "..." added&lt;br/&gt;
  Summing, minimum, maximum, etc. of list elements is easier now.&lt;/li&gt;
&lt;li&gt;@w and @b size override now works with immediate addressing&lt;br/&gt;
  For consistency, this is more useful for 65816&lt;/li&gt;
&lt;li&gt;Hexadecimal representation of byte strings x"1234"&lt;br/&gt;
  More straight forward to type than bytes($3412) and with "proper" ordering&lt;/li&gt;
&lt;li&gt;z85 style byte strings representation z"0S*"&lt;br/&gt;
  More compact than x"1234". For label files and to embed binary data in source as text.&lt;/li&gt;
&lt;li&gt;Add .bfor and .brept&lt;br/&gt;
  Scoped variant of .for/.rept which handles non-anonymous labels better&lt;/li&gt;
&lt;li&gt;Conditional assignment ":?=" added&lt;br/&gt;
  Creates a new variable when symbol undefined otherwise keeps old value&lt;/li&gt;
&lt;li&gt;Lazy fault handling&lt;br/&gt;
  A failed calculation is not a problem if the result is not used&lt;/li&gt;
&lt;li&gt;Negative .binary offsets&lt;br/&gt;
  To allow offsets based from end of file&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;Fixes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Data directives (.byte) must have at least one argument&lt;br/&gt;
  Confusing behaviour, no argument is likely a mistake&lt;/li&gt;
&lt;li&gt;Missing closing directives didn't auto close&lt;br/&gt;
  This resulted in confusing error messages or worse&lt;/li&gt;
&lt;li&gt;Improved directive missing messages&lt;br/&gt;
  It's not just "something is expected" but a bit more verbose&lt;/li&gt;
&lt;li&gt;Undefined sections could cause too many passes&lt;br/&gt;
  Implementation bug&lt;/li&gt;
&lt;li&gt;Optimizer suggestions are not shown for macros&lt;br/&gt;
  How to apply multiple suggestions for a single line?&lt;/li&gt;
&lt;li&gt;Exported symbols (-l) were reported as unused&lt;br/&gt;
  They are used for exporting after all...&lt;/li&gt;
&lt;li&gt;Compilation failed due to wcwidth&lt;br/&gt;
  Some platform had interesting defines preventing compilation&lt;/li&gt;
&lt;li&gt;Improve unused checks&lt;br/&gt;
  Was not reporting correctly in some cases (like macro invocation)&lt;/li&gt;
&lt;li&gt;Compilation warning reported by Compyx&lt;br/&gt;
  A static keyword was missing.&lt;/li&gt;
&lt;li&gt;Sometimes compilation ended too early&lt;br/&gt;
  In some sections related cases not enough passes were made&lt;/li&gt;
&lt;li&gt;Describe the "command line from file" feature better&lt;br/&gt;
  A single paragraph was not enough&lt;/li&gt;
&lt;li&gt;Improved missing macro argument message&lt;br/&gt;
  The old one was not verbose enough&lt;/li&gt;
&lt;li&gt;Redundant compilation warning (#33)&lt;br/&gt;
  Didn't delete enough ;)&lt;/li&gt;
&lt;li&gt;Section end with union&lt;br/&gt;
  Section end was not right in union&lt;/li&gt;
&lt;li&gt;Anonymous labels vs. conditional compilation&lt;br/&gt;
  Make clear in documentation how this works&lt;/li&gt;
&lt;li&gt;65CE02 stack relative addressing fix&lt;br/&gt;
  Unlike a 65816 it's signed.&lt;/li&gt;
&lt;li&gt;VICE label generation was not generic enough&lt;br/&gt;
  Some members where not listed.&lt;/li&gt;
&lt;li&gt;Improved -Wshadow&lt;br/&gt;
  Due to the .namespace directive it was not good enough any more&lt;/li&gt;
&lt;li&gt;WIN64 compile fixes&lt;br/&gt;
  Some format strings are unavailable on mingw64&lt;/li&gt;
&lt;li&gt;Incorrect warning about labels not on left&lt;br/&gt;
  Colon suffix was not enough&lt;/li&gt;
&lt;li&gt;str()/repr() improvement&lt;br/&gt;
  Some types have better string representations now&lt;/li&gt;
&lt;li&gt;Wrong type messages replaced with better conversion messages&lt;br/&gt;
  The old message was not too helpful&lt;/li&gt;
&lt;li&gt;65CE02 long branch fix second try&lt;br/&gt;
  There were still cases where the wrong distance was calculated&lt;/li&gt;
&lt;li&gt;Structs and unions didn't instantiate using a dot&lt;br/&gt;
  It was documented but didn't work&lt;/li&gt;
&lt;li&gt;Update to Unicode 11&lt;br/&gt;
  Upgraded tables to the latest version&lt;/li&gt;
&lt;li&gt;Suppress repeated caret lines in messages&lt;br/&gt;
  One is enough if all the others are the same&lt;/li&gt;
&lt;li&gt;.dstruct/.dunion double message on not defined&lt;br/&gt;
  No additional syntax error message any more&lt;/li&gt;
&lt;li&gt;Improve error positions&lt;br/&gt;
  Still some errors were not on the right character in line&lt;/li&gt;
&lt;li&gt;Only report file errors once&lt;br/&gt;
  It's enough to know once if a file is missing&lt;/li&gt;
&lt;li&gt;Improve syntax error messages&lt;br/&gt;
  Mostly replaced by 'expected' messages now&lt;/li&gt;
&lt;li&gt;Some messages lacked spacing at the end&lt;br/&gt;
  Now it's there&lt;/li&gt;
&lt;li&gt;Indexing precedence fixed&lt;br/&gt;
  Broke in r1315, oops.&lt;/li&gt;
&lt;li&gt;Faster inplace operations&lt;br/&gt;
  Most noticable for list appending or concatenate&lt;/li&gt;
&lt;li&gt;Anonymous labels vs. scopes&lt;br/&gt;
  Now they's truly bound to scope like normal labels&lt;/li&gt;
&lt;li&gt;Runaway macro/function recursions are stopped faster now&lt;br/&gt;
  Certain recursion faults were difficult to debug&lt;/li&gt;
&lt;li&gt;Wrong negated bit string shifting&lt;br/&gt;
  ~%10 &amp;lt;&amp;lt; 1 was ~%100 instead of ~%101&lt;/li&gt;
&lt;li&gt;Wrong negated byte string&lt;br/&gt;
  -x'00' was ~x'ff' instead of x'00' (integer value different)&lt;/li&gt;
&lt;/ul&gt;&lt;/div&gt;</summary></entry><entry><title>64tass 1.53.1515</title><link href="https://sourceforge.net/p/tass64/news/2017/05/64tass-1531515/" rel="alternate"/><published>2017-05-01T12:31:30.913000Z</published><updated>2017-05-01T12:31:30.913000Z</updated><author><name>Soci/Singular</name><uri>https://sourceforge.net/u/soci/</uri></author><id>https://sourceforge.net2047e33e159e56aab4b4a5050b2b8377f93b0b0a</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;It's about time to release another version as many changes were done and people were waiting for fixes already sitting in trunk for a while now. Some ideas didn't made it into this version as the implementation would have delayed the release even further, maybe next time.&lt;/p&gt;
&lt;p&gt;Changes since the last version:&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Compatibility&lt;/strong&gt;&lt;/p&gt;
&lt;p&gt;I'm not happy about these "annoying" changes and hope it's easy to update the affected code.&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;
&lt;p&gt;Deprecation of '^' operator&lt;br/&gt;
Code needs to be updated to use the format("%d", value) function for converting&lt;br/&gt;
values to decimal strings. The reason is that the "standard" use of this&lt;br/&gt;
operator is to return the bank byte in almost all other 65816 assemblers. This&lt;br/&gt;
function is available with the grave accent operator currently, as '^' was&lt;br/&gt;
already taken at the time. Plan is to migrate to the standard variant.  Old&lt;br/&gt;
code with '^' for decimal strings conversion will of course still compile now.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Deprecation of non-string encoding names&lt;br/&gt;
The original ".enc" directive used symbols for identifying encodings. The&lt;br/&gt;
problem with this is that it's not parametrizable so the encoding can't be an&lt;br/&gt;
expression or variable. There was no other way back then as strings were not&lt;br/&gt;
supported at the time but it would be nice to allow such use. In this version&lt;br/&gt;
the encoding names may be quoted for future compatibility but no expressions&lt;br/&gt;
are available yet otherwise not yet updated code wouldn't compile.&lt;/p&gt;
&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;New&lt;/strong&gt;&lt;/p&gt;
&lt;p&gt;Please check the manual for further details.&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;
&lt;p&gt;Keys without values in dict (sets)&lt;br/&gt;
For quicker "in" than in list&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Compound assignment also in .for&lt;br/&gt;
Now "+=" and friends work in ".for"&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Add &amp;amp;&amp;amp;= and ||= compound operators&lt;br/&gt;
For consistency.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Check for file name portability issues (-Wno-portable)&lt;br/&gt;
On multi platform development file names can cause compilation problems&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Find unused labels with -Wunused&lt;br/&gt;
Some unused labels may remain in code which can be pointed out&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Check inconsistent symbol references with -Wcase-symbol&lt;br/&gt;
For case insensitive but case matching symbols&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Add &amp;lt;? and ?&amp;gt; operators (smaller/min, greater/max)&lt;br/&gt;
New operator for range limiting&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Check for double .case values with -Wswitch-case&lt;br/&gt;
Multiple use of case value does not compile multiple times and may be a problem&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Check for missing # with -Wimmediate&lt;br/&gt;
For checking numeric constant opcodes without a '#' which may be a typos.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Turn on a check with -Werror=xxx too&lt;br/&gt;
Less options needed as -Werror=xxx assumes -Wxxx.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Command line arguments from file with @file&lt;br/&gt;
Frequently used options may be stored in a file&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Scopes and addresses included in VICE label listing now&lt;br/&gt;
Now more symbols are available in VICE&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Listing of labels of specific scopes only into files (--root-labels=)&lt;br/&gt;
Exporting of symbols only in a particular scope is possible now.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Dict indexable with lists now&lt;br/&gt;
For translation uses&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Improve some error messages by displaying the invalid values&lt;br/&gt;
So that there's less guessing what the wrong value was&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Keep integers when rounding&lt;br/&gt;
floor() and friends converted to float unnecessarily&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Add sort() to sort lists&lt;br/&gt;
For sorting&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;'-' can be a variable as well&lt;br/&gt;
For consistency.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Optimizer warns about ror/rol which could be lsr/asl&lt;br/&gt;
C is clear, so there's no point shifting it in.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Warn for float comparison problems -Wfloat-compare&lt;br/&gt;
Float comparisons may be wrong, especially for equality due to limited precision&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;format() improvements (integer precision, etc.)&lt;br/&gt;
Some format() features were missing or wrong.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Predefine pi constant&lt;br/&gt;
In case rad(180) needs too much typing...&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Assign function parameters as variables instead of constants&lt;br/&gt;
So that they may be modified.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Check for leading zeros -Wleading-zeros&lt;br/&gt;
There's no octal number support, so this may be an error&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Check for aliases with -Walias&lt;br/&gt;
For consistent use of opcodes.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Improve message of const as variable&lt;br/&gt;
It was a bit confusing&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Improved non-negative message on range errors&lt;br/&gt;
Make clear that the problem is with negative numbers not with number of bits&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Notes for common pitfalls -Wpitfall&lt;br/&gt;
Tips for some commonly encountered "problems".&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Long branch warnings are possible again -Wlong-branch&lt;br/&gt;
Long time ago this was there but was temporarily removed due to internal changes&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Optimizer warns about constant results, duplicate sir/sac&lt;br/&gt;
Calculations with constant values are pointless&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Optimizer proposes use of stz on 65c02&lt;br/&gt;
Storing zero, so why may use the specific instruction for that as well&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Update to Unicode 9.0&lt;br/&gt;
Including width tables this time&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Optimizer proposes use of bra&lt;br/&gt;
For always taken conditional branches&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Better out of memory error handling&lt;br/&gt;
For common cases it shows now where it happened.&lt;/p&gt;
&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;&lt;strong&gt;Fixes&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;
&lt;p&gt;Fix a few error positions which were missing&lt;br/&gt;
Some error messages still didn't point to the cause...&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Reduce fault messages for .endu/.ends&lt;br/&gt;
Too many messages&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Fix a possible null pointer crash for unknown address (fuzzed by Pierre)&lt;br/&gt;
Implementation bug&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Fix possible crash root namespace copying (fuzzed by Pierre)&lt;br/&gt;
Implementation bug&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Fix no argument indexing crash (fuzzed by Pierre)&lt;br/&gt;
Implementation bug&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Crash while handling struct values (fuzzed by Pierre)&lt;br/&gt;
Implementation bug&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Some special section nesting is not allowed (fuzzed by Pierre)&lt;br/&gt;
Implementation bug. The fix caused some regression but it should be fixed as well.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Wrong memory use on long division (fuzzed by Pierre)&lt;br/&gt;
Implementation bug&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Missing type checks at a few places (fuzzed by Pierre)&lt;br/&gt;
Implementation bug&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Fix some return leaks&lt;br/&gt;
Return where not freed&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Compile fixes with VS&lt;br/&gt;
Compiles now with newer VS without messing with code&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Display address without mode correctly&lt;br/&gt;
Did not display that an address is an address&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Some notes were still displayed without warning/errors&lt;br/&gt;
Some suppressed messages still had their notes displayed&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Stdin/stdout not to be listed in makefiles&lt;br/&gt;
Pointless to do so, they are not files.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Improve operations on addresses&lt;br/&gt;
Especially needed without mode&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Implement code offsets&lt;br/&gt;
To preserve code reference&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;More operations on gap&lt;br/&gt;
For .struct defaults with calculations&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Fix integer limit checks&lt;br/&gt;
It was done wrong.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Correct function call from macro line&lt;br/&gt;
Didn't work well.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Fixed scoped macro highlighting in vim syntax&lt;br/&gt;
Was not handled in the syntax file&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Fix error messages on undefined section&lt;br/&gt;
Messages were wrong.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Fix display of colon list with defaults&lt;br/&gt;
It was wrong.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Unnecessary bytes needed message after another error&lt;br/&gt;
Duplicate message for something which is not there&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;.eor was not applied to all fields in listing&lt;br/&gt;
Listing showed the wrong hex values except for the opcode&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;label *=*+1 works again with warning -Wstar-equal&lt;br/&gt;
Broke when compound operators were added, now works again.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Unpack code in .text&lt;br/&gt;
Inconsistent behaviour&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Fix some negative truncations/non-negative only values&lt;br/&gt;
Some negative values didn't supposed to be truncated.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Restore bold output after invalid char inverted output&lt;br/&gt;
Bold was missing after inversion.&lt;/p&gt;
&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;Colours for all messages&lt;br/&gt;
Command line parameter errors didn't had it yet.&lt;/p&gt;
&lt;/li&gt;
&lt;/ul&gt;&lt;/div&gt;</summary></entry></feed>