Author: bernhard
Date: Sat Oct 8 04:55:03 2005
New Revision: 9406
Modified:
trunk/languages/m4/ChangeLog
trunk/languages/m4/TODO.pod
trunk/languages/m4/src/input.pir
trunk/languages/m4/src/m4.pir
trunk/languages/m4/src/macro.pir
trunk/languages/m4/t/basic/003_getopt.t
trunk/languages/m4/t/regex/004_pge.t
Log:
Call it 'Parrot m4 0.0.14'.
Fix m4/t/regex/004_pge.t.
Add state['state'] and state['rulesub'],
and thus rely less on naming convention
Modified: trunk/languages/m4/ChangeLog
==============================================================================
--- trunk/languages/m4/ChangeLog (original)
+++ trunk/languages/m4/ChangeLog Sat Oct 8 04:55:03 2005
@@ -1,6 +1,12 @@
# $Id$
2005-06-29 Bernhard Schmalhofer
+ * Make it partly work with Parrot 0.3.0
+ * Skip testing of loading frozen states
+ * Fix t/regex/004_pge.t
+ * Call it Parrot m4 0.0.14
+
+2005-06-29 Bernhard Schmalhofer
* Switch from PCRE to PGE
* Call it Parrot m4 0.0.13
Modified: trunk/languages/m4/TODO.pod
==============================================================================
--- trunk/languages/m4/TODO.pod (original)
+++ trunk/languages/m4/TODO.pod Sat Oct 8 04:55:03 2005
@@ -231,6 +231,8 @@ by chapters in the GNU m4 documentation.
=over
+=item Needs to be fixed, as the rx_* ops are gone
+
=item more tests
=back
Modified: trunk/languages/m4/src/input.pir
==============================================================================
--- trunk/languages/m4/src/input.pir (original)
+++ trunk/languages/m4/src/input.pir Sat Oct 8 04:55:03 2005
@@ -6,7 +6,7 @@ input.pir - Setting up input and reading
=head1 DESCRIPTION
-Copyright: 2004 Bernhard Schmalhofer. All Rights Reserved.
+Copyright: 2004-2005 Bernhard Schmalhofer. All Rights Reserved.
CVS Info: $Id$
History: Ported from GNU m4 1.4
References: http://www.gnu.org/software/m4/m4.html
@@ -15,17 +15,33 @@ References: http://www.gnu.org/software/
=head2 void input_init( Hash state )
-Initialise some stacks and some regexes
+Initialise some stacks and put them into the Hash state['stack']:
-'token_stack'
-'input_stack' contains files, strings and macro definitions
-'wrapup_stack'
-'word_rulesub' recognizes TOKEN_WORD.
-'string_rulesub' recognizes TOKEN_STRING
-'simple_rulesub' recognizes TOKEN_SIMPLE
-'comment_rulesub' recognizes comments, returned as TOKEN_SIMPLE
+=over 4
-TODO: recognize nested quoted strings
+=item state['stack';'token'] not used yet
+
+=item state['stack';'input'] contains files, strings and macro definitions
+
+=item state['stack';'wrapup'] not used yet
+
+=back
+
+Initialize some PGE rulesubs and put them into the Hash state['rulesub']:
+
+=over 4
+
+=item state['rulesub';'word'] recognizes TOKEN_WORD.
+
+=item state['rulesub';'string'] recognizes TOKEN_STRING
+
+=item state['rulesub';'simple'] recognizes TOKEN_SIMPLE
+
+=item state['rulesub';'simple'] recognizes comments, returned as TOKEN_SIMPLE
+
+=back
+
+TOOO: recognize nested quoted strings
=cut
@@ -35,27 +51,35 @@ TODO: recognize nested quoted strings
.param pmc state
# setup of stacks
+ .local pmc stack_in_state
+ stack_in_state = new Hash
+ state['stack'] = stack_in_state
.local pmc empty_array
empty_array = new ResizablePMCArray
- state['token_stack'] = empty_array
+ stack_in_state['token'] = empty_array
empty_array = new ResizablePMCArray
- state['input_stack'] = empty_array
+ stack_in_state['input'] = empty_array
empty_array = new ResizablePMCArray
- state['wrapup_stack'] = empty_array
+ stack_in_state['wrapup'] = empty_array
# setup of some rules
# these rules should be kept in sync with t/regex/002_tokens.t
+ # TODO: Use named or positional captures
+ # TODO: rulesub for frozen state
.local pmc p6rule
find_global p6rule, "PGE", "p6rule"
+ .local pmc rulesub_in_state
+ rulesub_in_state = new Hash
+ state['rulesub'] = rulesub_in_state
.local pmc rulesub
- rulesub = p6rule( "^<[_a..zA..Z]><[_a..zA..Z0..9]>*" )
- state['word_rulesub'] = rulesub
+ rulesub = p6rule( "^<[_a..zA..Z]><[_a..zA..Z0..9>]>*" )
+ rulesub_in_state['word'] = rulesub
rulesub = p6rule( "^`<-[`]>*'" )
- state['string_rulesub'] = rulesub
+ rulesub_in_state['string'] = rulesub
rulesub = p6rule( "^<-[`#_a..zA..Z]>" )
- state['simple_rulesub'] = rulesub
+ rulesub_in_state['simple'] = rulesub
rulesub = p6rule( "^\#\N*\n" )
- state['comment_rulesub'] = rulesub
+ rulesub_in_state['comment'] = rulesub
.end
@@ -86,10 +110,10 @@ TODO: open these files and complain when
input_string = read in, 50000
close in
- # state['input_stack'] has been created in input_init
+ # state['stack';'input'] has been created in input_init
# TODO: seperate input blocks for every file
.local pmc input_stack
- input_stack = state['input_stack']
+ input_stack = state['stack';'input']
.local pmc input_block
.local int stack_size
stack_size = input_stack
@@ -113,7 +137,7 @@ TODO: open these files and complain when
=head2 string next_token( Hash state )
Parse and return a single token from the input stream. A token can
-either be TOKEN_EOF, if the input_stack is empty; it can be TOKEN_STRING
+either be TOKEN_EOF, if the state['stack';'input'] is empty; it can be
TOKEN_STRING
for a quoted string; TOKEN_WORD for something that is a potential macro
name; and TOKEN_SIMPLE for any single character that is not a part of
any of the previous types.
@@ -128,7 +152,7 @@ Uses regular expressions for finding tok
.param pmc state
.local pmc input_stack
- input_stack = state['input_stack']
+ input_stack = state['stack';'input']
.local pmc input_block
input_block = shift input_stack
.local string input_string
@@ -144,26 +168,26 @@ Uses regular expressions for finding tok
# look for 'TOKEN_SIMPLE'
# read a whole bunch of non-macro and non-word charcters
- rulesub = state['simple_rulesub']
+ rulesub = state['rulesub';'simple']
token_type = 'TOKEN_SIMPLE'
match = rulesub( input_string )
if match goto MATCH
# look for comments and return it as 'TOKEN_SIMPLE'
- rulesub = state['comment_rulesub']
+ rulesub = state['rulesub';'comment']
token_type = 'TOKEN_SIMPLE'
match = rulesub( input_string )
if match goto MATCH
# look for 'TOKEN_STRING'
- rulesub = state['string_rulesub']
+ rulesub = state['rulesub';'string']
token_type = 'TOKEN_STRING'
match = rulesub( input_string )
if match goto MATCH
# look for 'TOKEN_WORD'
# this will be checked for macro substitution
- rulesub = state['word_rulesub']
+ rulesub = state['rulesub';'word']
token_type = 'TOKEN_WORD'
match = rulesub( input_string )
if match goto MATCH
Modified: trunk/languages/m4/src/m4.pir
==============================================================================
--- trunk/languages/m4/src/m4.pir (original)
+++ trunk/languages/m4/src/m4.pir Sat Oct 8 04:55:03 2005
@@ -134,7 +134,7 @@ Looks at the command line arguments and
# Was '--version' passed ?
is_defined = defined opt["version"]
unless is_defined goto NO_VERSION_FLAG
- print "Parrot m4 0.0.13\n"
+ print "Parrot m4 0.0.14\n"
end
NO_VERSION_FLAG:
@@ -255,7 +255,7 @@ NO_UNIMPLEMENTED_OPTION:
.local pmc m4_eval_compiler_lib
m4_eval_compiler_lib = loadlib "m4_eval_compiler"
- # init of input structures, creates state['input_stack']
+ # init of input structures, creates state['stack';'input']
input_init( state )
# TODO: init of output structures
Modified: trunk/languages/m4/src/macro.pir
==============================================================================
--- trunk/languages/m4/src/macro.pir (original)
+++ trunk/languages/m4/src/macro.pir Sat Oct 8 04:55:03 2005
@@ -77,7 +77,7 @@ the text are just copied to the output.
blind_no_args = symbol['blind_no_args']
unless blind_no_args goto EXPAND_MACRO
.local string input_string
- input_string = state['input_stack';0;'string']
+ input_string = state['stack';'input';0;'string']
.local int first_char, open_parenthesis
first_char = ord input_string
open_parenthesis = ord '('
@@ -138,9 +138,9 @@ NESTING_LIMIT_NOT_REACHED_YET:
.local string text
( text ) = call_macro( state, symbol, arguments )
.local string input_string
- input_string = state['input_stack';0;'string']
+ input_string = state['stack';'input';0;'string']
input_string = text . input_string
- state['input_stack';0;'string'] = input_string
+ state['stack';'input';0;'string'] = input_string
expansion_level = state['expansion_level']
dec expansion_level
@@ -207,7 +207,7 @@ Collect all the arguments to a call of a
# Thus we need to remenber the start and the length of these two captures
.local int cnt_stack
.local string input_string
- input_string = state['input_stack';0;'string']
+ input_string = state['stack';'input';0;'string']
# We need a '(' at beginning of string
.local int index_opening
@@ -250,5 +250,5 @@ NO_MORE_ARGS:
substr input_string, 0, index_closing, ''
NOT_A_ARGUMENT_LIST:
- state['input_stack';0;'string'] = input_string
+ state['stack';'input';0;'string'] = input_string
.end
Modified: trunk/languages/m4/t/basic/003_getopt.t
==============================================================================
--- trunk/languages/m4/t/basic/003_getopt.t (original)
+++ trunk/languages/m4/t/basic/003_getopt.t Sat Oct 8 04:55:03 2005
@@ -34,7 +34,7 @@ END_OUT
#--------------------------------------------
$real_out = `$parrot_m4 --version 2>&1`;
is( $real_out, << 'END_OUT', '--version' );
-Parrot m4 0.0.13
+Parrot m4 0.0.14
END_OUT
Modified: trunk/languages/m4/t/regex/004_pge.t
==============================================================================
--- trunk/languages/m4/t/regex/004_pge.t (original)
+++ trunk/languages/m4/t/regex/004_pge.t Sat Oct 8 04:55:03 2005
@@ -33,6 +33,7 @@ use Parrot::Test tests => 1;
match = word_rulesub(target) # execute rule on target string
match_loop:
+ if_null match, match_fail # if match fails stop
unless match goto match_fail # if match fails stop
print "match succeeded\n"
@@ -43,19 +44,14 @@ match_loop:
match_fail:
print "match failed\n"
- .return()
.end
END_PIR
match succeeded
: <Hello @ 0> 0
match succeeded
-: <Hell @ 0> 0
-match succeeded
-: <Hel @ 0> 0
-match succeeded
-: <He @ 0> 0
+: <Hello @ 0> 0
match succeeded
-: <H @ 0> 0
+: <Hello @ 0> 0
match failed
OUTPUT
}
