remove unused TODO, BUGS files

diff --git a/BUGS b/BUGS
deleted file mode 100644 (file)
index f456b69..0000000
--- a/BUGS
+++ /dev/null
@@ -1,107 +0,0 @@
-This is a list of bugs still in the queue at lex.sf.net at the time we closed
-out the project and moved it to flex.sf.net.
-
--------------------------------------------------------------
-
-Some strict compilers warn about a few internal flex variables signedness. They
-are bogus warnings and can be ignored, but people send in reports nonethless.
-
--------------------------------------------------------------
-
-The initializer of the yy_transition array in the
-generated scanner
-contains fewer entries than the declared size of the array.
-
-Examples include yy_transition[6504] with 6250 entries,
-yy_transition[13215] with 12961 entries. This looks
-like it
-is always 254 fewer entries than the declared size.
-
-This bug is present in flex 2.5.4a as well. It appears to be harmless.
-
--------------------------------------------------------------
-The examples in the chapter "Generating C++ Scanners"
-contain suspicious code. Attached is a patch that
-corrects this, and after these
-modifications this example compiles and works.
-
--------------------------------------------------------------
-
-C++ scanners derived from the yyFlexLexer base class
-will not compile with flex-2.5.31 because the
-<FlexLexer.h> automatically gets included into the
-scanner twice. Because yyFlexLexer is now defined by
-default even if no prefix is specified, including
-FlexLexer.h twice causes the class yyFlexLexer to be
-declared twice. In flex-2.5.4 because yyFlexLexer was
-not defined by flex in the scanner code, including
-FlexLexer.h more than once only declared yyFlexLexer
-once. I appreciate that this is because of the M4
-additions to flex, but I can not find a way to stop
-flex defining yyFlexLexer if it is not needed.
-
-Here is an example of a class that will not compile:
-
-derived_lexer.h:
-
-#ifndef __derived_lexer__
-#define __derived_lexer__
-
-#include <FlexLexer.h>
-
-class derived_lexer : public yyFlexLexer
-{
-public:
-derived_lexer(std::istream* arg_yyin = 0) :
-yyFlexLexer(arg_yyin){}
-int yylex();
-int x;
-};
-#endif
-
-derived_lexer.l:
-
-%{
-#include "derived_lexer.h"
-%}
-
-%option yyclass="derived_lexer"
-
-%%
-[0-9]+ {
-x = atoi(yytext);
-}
-%%
-
-main.cpp:
-
-#include "derived_lexer.h"
-#include <fstream>
-
-int main()
-{
-std::ifstream input;
-input.open("input");
-derived_lexer lexer(&input);
-lexer.yylex();
-}
-
--------------------------------------------------------------
-
-Hi, the anomally is that if I generate a c++ parser it will not make 
-yy_scan_buffer and friends.
-this is happenning on flex version 2.5.4.
-Is this the intent, if so how do I make a c++ parser read from my buffer?
-
-P.S. in c++ it will only generate:
-#ifndef YY_NO_SCAN_BUFFER
-#endif
-
-
-#ifndef YY_NO_SCAN_STRING
-#endif
-
-
-#ifndef YY_NO_SCAN_BYTES
-#endif
--------------------------------------------------------------

diff --git a/TODO b/TODO
deleted file mode 100644 (file)
index 702a6db..0000000
--- a/TODO
+++ /dev/null
@@ -1,66 +0,0 @@
-* the manual:
-
-** do an end-to-end proofread of the manual (this is under way, but is
-   going slowly)
-
-** pretty up the dvi output; overflows, etc.
-
-** faq
-
-*** clean up the faqs section. The information is good; the texinfo
-   could use some touching up.
-
-*** index the faq entries
-
-*** mention that it's possible to use a variable to scan matching
-    brackets, nested comments etc.
-
-*** include something about lexing/parsing fortran
-
-** create a section on flex design, features, etc.
-
-* getext
-
-** make sure all flex modules use gettext translation facilities
-
-*subdirectories
-
-** in examples/manual, integrate the Makefile.examples into the
-    Makefile.am
-
-* test suite
-
-** integrate the test suite into automake's framework (note that the
-   test suite can be run from the top level directory with "make
-   check". Still, we want to get it completely under automake's control.)
-
-** make test suite more complete
-
-* generic coding
-
-** move as much skeleton code as possible out of gen.c and into
-  flex.skl
-
-** figure out whether we want to add the capability to have
-  auto-generated backout rules
-
-** token-type and token buffer support
-
-** check if we still need to #undef macros at the end of a header
-
-** merge yylineno into support for location tracking
-
-** bug where yylineno is not decremented on REJECT
-
-** bug where yylineno is counted in trailing context
-
-* C++
-
-** have a separate skeleton for c++
-
-** revisit the C++ API. We get requests to make it more complete.
-
-Local Variables:
-Mode: text
-mode: outline-minor
-End: