git-svn: trunk@3917
Török Edvin authored on 2008/07/08 04:10:50... | ... |
@@ -1,3 +1,7 @@ |
1 |
+Mon Jul 7 21:42:18 EEST 2008 (edwin) |
|
2 |
+------------------------------------- |
|
3 |
+ * build system, libclamav/jsparse: add Javascript normalizer (not yet enabled) |
|
4 |
+ |
|
1 | 5 |
Mon Jul 7 15:41:02 CEST 2008 (tk) |
2 | 6 |
---------------------------------- |
3 | 7 |
* libclamav/ole2_extract.c, sigtool: make sigtool compatible with the new |
... | ... |
@@ -867,6 +867,7 @@ LIBTOOL |
867 | 867 |
VERSIONSCRIPTFLAG |
868 | 868 |
VERSIONSCRIPT_TRUE |
869 | 869 |
VERSIONSCRIPT_FALSE |
870 |
+GPERF |
|
870 | 871 |
LCHECK |
871 | 872 |
MAINTAINER_MODE_TRUE |
872 | 873 |
MAINTAINER_MODE_FALSE |
... | ... |
@@ -4767,7 +4768,7 @@ ia64-*-hpux*) |
4767 | 4767 |
;; |
4768 | 4768 |
*-*-irix6*) |
4769 | 4769 |
# Find out which ABI we are using. |
4770 |
- echo '#line 4770 "configure"' > conftest.$ac_ext |
|
4770 |
+ echo '#line 4771 "configure"' > conftest.$ac_ext |
|
4771 | 4771 |
if { (eval echo "$as_me:$LINENO: \"$ac_compile\"") >&5 |
4772 | 4772 |
(eval $ac_compile) 2>&5 |
4773 | 4773 |
ac_status=$? |
... | ... |
@@ -6857,11 +6858,11 @@ else |
6857 | 6857 |
-e 's:.*FLAGS}\{0,1\} :&$lt_compiler_flag :; t' \ |
6858 | 6858 |
-e 's: [^ ]*conftest\.: $lt_compiler_flag&:; t' \ |
6859 | 6859 |
-e 's:$: $lt_compiler_flag:'` |
6860 |
- (eval echo "\"\$as_me:6860: $lt_compile\"" >&5) |
|
6860 |
+ (eval echo "\"\$as_me:6861: $lt_compile\"" >&5) |
|
6861 | 6861 |
(eval "$lt_compile" 2>conftest.err) |
6862 | 6862 |
ac_status=$? |
6863 | 6863 |
cat conftest.err >&5 |
6864 |
- echo "$as_me:6864: \$? = $ac_status" >&5 |
|
6864 |
+ echo "$as_me:6865: \$? = $ac_status" >&5 |
|
6865 | 6865 |
if (exit $ac_status) && test -s "$ac_outfile"; then |
6866 | 6866 |
# The compiler can only warn and ignore the option if not recognized |
6867 | 6867 |
# So say no if there are warnings other than the usual output. |
... | ... |
@@ -7147,11 +7148,11 @@ else |
7147 | 7147 |
-e 's:.*FLAGS}\{0,1\} :&$lt_compiler_flag :; t' \ |
7148 | 7148 |
-e 's: [^ ]*conftest\.: $lt_compiler_flag&:; t' \ |
7149 | 7149 |
-e 's:$: $lt_compiler_flag:'` |
7150 |
- (eval echo "\"\$as_me:7150: $lt_compile\"" >&5) |
|
7150 |
+ (eval echo "\"\$as_me:7151: $lt_compile\"" >&5) |
|
7151 | 7151 |
(eval "$lt_compile" 2>conftest.err) |
7152 | 7152 |
ac_status=$? |
7153 | 7153 |
cat conftest.err >&5 |
7154 |
- echo "$as_me:7154: \$? = $ac_status" >&5 |
|
7154 |
+ echo "$as_me:7155: \$? = $ac_status" >&5 |
|
7155 | 7155 |
if (exit $ac_status) && test -s "$ac_outfile"; then |
7156 | 7156 |
# The compiler can only warn and ignore the option if not recognized |
7157 | 7157 |
# So say no if there are warnings other than the usual output. |
... | ... |
@@ -7251,11 +7252,11 @@ else |
7251 | 7251 |
-e 's:.*FLAGS}\{0,1\} :&$lt_compiler_flag :; t' \ |
7252 | 7252 |
-e 's: [^ ]*conftest\.: $lt_compiler_flag&:; t' \ |
7253 | 7253 |
-e 's:$: $lt_compiler_flag:'` |
7254 |
- (eval echo "\"\$as_me:7254: $lt_compile\"" >&5) |
|
7254 |
+ (eval echo "\"\$as_me:7255: $lt_compile\"" >&5) |
|
7255 | 7255 |
(eval "$lt_compile" 2>out/conftest.err) |
7256 | 7256 |
ac_status=$? |
7257 | 7257 |
cat out/conftest.err >&5 |
7258 |
- echo "$as_me:7258: \$? = $ac_status" >&5 |
|
7258 |
+ echo "$as_me:7259: \$? = $ac_status" >&5 |
|
7259 | 7259 |
if (exit $ac_status) && test -s out/conftest2.$ac_objext |
7260 | 7260 |
then |
7261 | 7261 |
# The compiler can only warn and ignore the option if not recognized |
... | ... |
@@ -9628,7 +9629,7 @@ else |
9628 | 9628 |
lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2 |
9629 | 9629 |
lt_status=$lt_dlunknown |
9630 | 9630 |
cat > conftest.$ac_ext <<EOF |
9631 |
-#line 9631 "configure" |
|
9631 |
+#line 9632 "configure" |
|
9632 | 9632 |
#include "confdefs.h" |
9633 | 9633 |
|
9634 | 9634 |
#if HAVE_DLFCN_H |
... | ... |
@@ -9728,7 +9729,7 @@ else |
9728 | 9728 |
lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2 |
9729 | 9729 |
lt_status=$lt_dlunknown |
9730 | 9730 |
cat > conftest.$ac_ext <<EOF |
9731 |
-#line 9731 "configure" |
|
9731 |
+#line 9732 "configure" |
|
9732 | 9732 |
#include "confdefs.h" |
9733 | 9733 |
|
9734 | 9734 |
#if HAVE_DLFCN_H |
... | ... |
@@ -11570,6 +11571,12 @@ else |
11570 | 11570 |
fi |
11571 | 11571 |
|
11572 | 11572 |
|
11573 |
+# it is not fatal if gperf is missing |
|
11574 |
+ |
|
11575 |
+GPERF=${GPERF-"${am_missing_run}gperf"} |
|
11576 |
+ |
|
11577 |
+ |
|
11578 |
+ |
|
11573 | 11579 |
|
11574 | 11580 |
|
11575 | 11581 |
|
... | ... |
@@ -20179,6 +20186,7 @@ LIBTOOL!$LIBTOOL$ac_delim |
20179 | 20179 |
VERSIONSCRIPTFLAG!$VERSIONSCRIPTFLAG$ac_delim |
20180 | 20180 |
VERSIONSCRIPT_TRUE!$VERSIONSCRIPT_TRUE$ac_delim |
20181 | 20181 |
VERSIONSCRIPT_FALSE!$VERSIONSCRIPT_FALSE$ac_delim |
20182 |
+GPERF!$GPERF$ac_delim |
|
20182 | 20183 |
LCHECK!$LCHECK$ac_delim |
20183 | 20184 |
MAINTAINER_MODE_TRUE!$MAINTAINER_MODE_TRUE$ac_delim |
20184 | 20185 |
MAINTAINER_MODE_FALSE!$MAINTAINER_MODE_FALSE$ac_delim |
... | ... |
@@ -20212,7 +20220,7 @@ LIBOBJS!$LIBOBJS$ac_delim |
20212 | 20212 |
LTLIBOBJS!$LTLIBOBJS$ac_delim |
20213 | 20213 |
_ACEOF |
20214 | 20214 |
|
20215 |
- if test `sed -n "s/.*$ac_delim\$/X/p" conf$$subs.sed | grep -c X` = 36; then |
|
20215 |
+ if test `sed -n "s/.*$ac_delim\$/X/p" conf$$subs.sed | grep -c X` = 37; then |
|
20216 | 20216 |
break |
20217 | 20217 |
elif $ac_last_try; then |
20218 | 20218 |
{ { echo "$as_me:$LINENO: error: could not make $CONFIG_STATUS" >&5 |
... | ... |
@@ -289,6 +289,10 @@ EOF |
289 | 289 |
AC_SUBST([VERSIONSCRIPTFLAG]) |
290 | 290 |
AM_CONDITIONAL([VERSIONSCRIPT], test "x$ac_cv_ld_version_script" = "xyes") |
291 | 291 |
|
292 |
+# it is not fatal if gperf is missing |
|
293 |
+AM_MISSING_PROG(GPERF, gperf) |
|
294 |
+AC_SUBST(GPERF) |
|
295 |
+ |
|
292 | 296 |
AC_CHECK_HEADERS([stdint.h unistd.h sys/int_types.h dlfcn.h inttypes.h sys/inttypes.h memory.h ndir.h stdlib.h strings.h string.h sys/mman.h sys/param.h sys/stat.h sys/types.h malloc.h poll.h limits.h sys/filio.h sys/uio.h termios.h stdbool.h pwd.h grp.h]) |
293 | 297 |
AC_CHECK_HEADER([syslog.h],AC_DEFINE([USE_SYSLOG],1,[use syslog]),) |
294 | 298 |
|
... | ... |
@@ -188,7 +188,10 @@ libclamav_la_SOURCES = \ |
188 | 188 |
textnorm.c \ |
189 | 189 |
textnorm.h \ |
190 | 190 |
dlp.c \ |
191 |
- dlp.h |
|
191 |
+ dlp.h \ |
|
192 |
+ jsparse/js-norm.c \ |
|
193 |
+ jsparse/js-norm.h \ |
|
194 |
+ jsparse/lexglobal.h |
|
192 | 195 |
|
193 | 196 |
libclamav_internal_utils_la_SOURCES=str.c \ |
194 | 197 |
str.h \ |
... | ... |
@@ -199,4 +202,33 @@ libclamav_internal_utils_la_LDFLAGS=-static |
199 | 199 |
libclamav_internal_utils_la_CFLAGS=-DCLI_MEMFUNSONLY |
200 | 200 |
lib_LTLIBRARIES = libclamav.la |
201 | 201 |
noinst_LTLIBRARIES = libclamav_internal_utils.la |
202 |
-EXTRA_DIST = regex/engine.c libclamav.map |
|
202 |
+EXTRA_DIST = regex/engine.c libclamav.map jsparse-keywords.gperf \ |
|
203 |
+ jsparse-operators.h jsparse-keywords.h jsparse/future_reserved_words\ |
|
204 |
+ jsparse/keywords jsparse/special_keywords jsparse/operators.gperf |
|
205 |
+ |
|
206 |
+if MAINTAINER_MODE |
|
207 |
+BUILT_SOURCES=jsparse/generated/operators.h jsparse/generated/keywords.h jsparse-keywords.gperf |
|
208 |
+ |
|
209 |
+GPERF_FLAGS=-E -t -L ANSI-C -C -F ', TOK_ERROR' -c |
|
210 |
+ |
|
211 |
+jsparse-keywords.gperf: jsparse/keywords.list jsparse/future_reserved_words.list jsparse/special_keywords.list |
|
212 |
+ echo -e "struct keyword { const char *name; int val; };\n%%" >keywords-g-tmp |
|
213 |
+ for i in `cat @srcdir@/jsparse/keywords.list`; do j=`echo $$i |tr \[a-z\] \[A-Z\]`; echo "$$i, TOK_$$j" >>keywords-g-tmp; done |
|
214 |
+ for i in `cat @srcdir@/jsparse/future_reserved_words.list`; do echo "$$i, TOK_FUTURE_RESERVED_WORD" >>keywords-g-tmp; done |
|
215 |
+ cat @srcdir@/jsparse/special_keywords.list >>keywords-g-tmp |
|
216 |
+ mv keywords-g-tmp $@ |
|
217 |
+ |
|
218 |
+jsparse/generated/operators.h: jsparse/operators.gperf |
|
219 |
+ $(GPERF) $(GPERF_FLAGS) -H op_hash -N in_op_set -W oplist $< >operators-tmp-g |
|
220 |
+ grep -v '^#line' <operators-tmp-g | sed -e 's/^const struct/static const struct/' -e 's/register //g' >operators-tmp |
|
221 |
+ rm operators-tmp-g |
|
222 |
+ mv operators-tmp @srcdir@/jsparse/generated/operators.h |
|
223 |
+ |
|
224 |
+jsparse/generated/keywords.h: jsparse-keywords.gperf |
|
225 |
+ $(GPERF) $(GPERF_FLAGS) $< >keywords-tmp-g |
|
226 |
+ grep -v '^#line' <keywords-tmp-g | sed -e 's/^const struct/static const struct/' -e 's/register //g' >keywords-tmp |
|
227 |
+ rm keywords-tmp-g |
|
228 |
+ mv keywords-tmp @srcdir@/jsparse/generated/keywords.h |
|
229 |
+ |
|
230 |
+CLEANFILES=jsparse-keywords.gperf @srcdir@/jsparse/generated/operators.h @srcdir@/jsparse/generated/keywords.h |
|
231 |
+endif |
... | ... |
@@ -89,7 +89,7 @@ am_libclamav_la_OBJECTS = matcher-ac.lo matcher-bm.lo matcher.lo \ |
89 | 89 |
infblock.lo pdf.lo spin.lo yc.lo elf.lo sis.lo uuencode.lo \ |
90 | 90 |
phishcheck.lo phish_domaincheck_db.lo phish_whitelist.lo \ |
91 | 91 |
regex_list.lo mspack.lo cab.lo entconv.lo hashtab.lo dconf.lo \ |
92 |
- lzma_iface.lo explode.lo textnorm.lo dlp.lo |
|
92 |
+ lzma_iface.lo explode.lo textnorm.lo dlp.lo js-norm.lo |
|
93 | 93 |
libclamav_la_OBJECTS = $(am_libclamav_la_OBJECTS) |
94 | 94 |
libclamav_la_LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) \ |
95 | 95 |
$(LIBTOOLFLAGS) --mode=link $(CCLD) $(AM_CFLAGS) $(CFLAGS) \ |
... | ... |
@@ -165,6 +165,7 @@ EGREP = @EGREP@ |
165 | 165 |
EXEEXT = @EXEEXT@ |
166 | 166 |
FRESHCLAM_LIBS = @FRESHCLAM_LIBS@ |
167 | 167 |
GETENT = @GETENT@ |
168 |
+GPERF = @GPERF@ |
|
168 | 169 |
GREP = @GREP@ |
169 | 170 |
HAVE_LIBGMP = @HAVE_LIBGMP@ |
170 | 171 |
INSTALL = @INSTALL@ |
... | ... |
@@ -423,7 +424,10 @@ libclamav_la_SOURCES = \ |
423 | 423 |
textnorm.c \ |
424 | 424 |
textnorm.h \ |
425 | 425 |
dlp.c \ |
426 |
- dlp.h |
|
426 |
+ dlp.h \ |
|
427 |
+ jsparse/js-norm.c \ |
|
428 |
+ jsparse/js-norm.h \ |
|
429 |
+ jsparse/lexglobal.h |
|
427 | 430 |
|
428 | 431 |
libclamav_internal_utils_la_SOURCES = str.c \ |
429 | 432 |
str.h \ |
... | ... |
@@ -434,8 +438,15 @@ libclamav_internal_utils_la_LDFLAGS = -static |
434 | 434 |
libclamav_internal_utils_la_CFLAGS = -DCLI_MEMFUNSONLY |
435 | 435 |
lib_LTLIBRARIES = libclamav.la |
436 | 436 |
noinst_LTLIBRARIES = libclamav_internal_utils.la |
437 |
-EXTRA_DIST = regex/engine.c libclamav.map |
|
438 |
-all: all-recursive |
|
437 |
+EXTRA_DIST = regex/engine.c libclamav.map jsparse-keywords.gperf \ |
|
438 |
+ jsparse-operators.h jsparse-keywords.h jsparse/future_reserved_words\ |
|
439 |
+ jsparse/keywords jsparse/special_keywords jsparse/operators.gperf |
|
440 |
+ |
|
441 |
+@MAINTAINER_MODE_TRUE@BUILT_SOURCES = jsparse/generated/operators.h jsparse/generated/keywords.h jsparse-keywords.gperf |
|
442 |
+@MAINTAINER_MODE_TRUE@GPERF_FLAGS = -E -t -L ANSI-C -C -F ', TOK_ERROR' -c |
|
443 |
+@MAINTAINER_MODE_TRUE@CLEANFILES = jsparse-keywords.gperf @srcdir@/jsparse/generated/operators.h @srcdir@/jsparse/generated/keywords.h |
|
444 |
+all: $(BUILT_SOURCES) |
|
445 |
+ $(MAKE) $(AM_MAKEFLAGS) all-recursive |
|
439 | 446 |
|
440 | 447 |
.SUFFIXES: |
441 | 448 |
.SUFFIXES: .c .lo .o .obj |
... | ... |
@@ -536,6 +547,7 @@ distclean-compile: |
536 | 536 |
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/infblock.Plo@am__quote@ |
537 | 537 |
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/inflate64.Plo@am__quote@ |
538 | 538 |
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/is_tar.Plo@am__quote@ |
539 |
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/js-norm.Plo@am__quote@ |
|
539 | 540 |
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/libclamav_internal_utils_la-others.Plo@am__quote@ |
540 | 541 |
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/libclamav_internal_utils_la-str.Plo@am__quote@ |
541 | 542 |
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/line.Plo@am__quote@ |
... | ... |
@@ -666,6 +678,13 @@ infblock.lo: nsis/infblock.c |
666 | 666 |
@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@ |
667 | 667 |
@am__fastdepCC_FALSE@ $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o infblock.lo `test -f 'nsis/infblock.c' || echo '$(srcdir)/'`nsis/infblock.c |
668 | 668 |
|
669 |
+js-norm.lo: jsparse/js-norm.c |
|
670 |
+@am__fastdepCC_TRUE@ $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT js-norm.lo -MD -MP -MF $(DEPDIR)/js-norm.Tpo -c -o js-norm.lo `test -f 'jsparse/js-norm.c' || echo '$(srcdir)/'`jsparse/js-norm.c |
|
671 |
+@am__fastdepCC_TRUE@ mv -f $(DEPDIR)/js-norm.Tpo $(DEPDIR)/js-norm.Plo |
|
672 |
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ source='jsparse/js-norm.c' object='js-norm.lo' libtool=yes @AMDEPBACKSLASH@ |
|
673 |
+@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@ |
|
674 |
+@am__fastdepCC_FALSE@ $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o js-norm.lo `test -f 'jsparse/js-norm.c' || echo '$(srcdir)/'`jsparse/js-norm.c |
|
675 |
+ |
|
669 | 676 |
libclamav_internal_utils_la-str.lo: str.c |
670 | 677 |
@am__fastdepCC_TRUE@ $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) --mode=compile $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(libclamav_internal_utils_la_CFLAGS) $(CFLAGS) -MT libclamav_internal_utils_la-str.lo -MD -MP -MF $(DEPDIR)/libclamav_internal_utils_la-str.Tpo -c -o libclamav_internal_utils_la-str.lo `test -f 'str.c' || echo '$(srcdir)/'`str.c |
671 | 678 |
@am__fastdepCC_TRUE@ mv -f $(DEPDIR)/libclamav_internal_utils_la-str.Tpo $(DEPDIR)/libclamav_internal_utils_la-str.Plo |
... | ... |
@@ -877,14 +896,16 @@ distdir: $(DISTFILES) |
877 | 877 |
fi; \ |
878 | 878 |
done |
879 | 879 |
check-am: all-am |
880 |
-check: check-recursive |
|
880 |
+check: $(BUILT_SOURCES) |
|
881 |
+ $(MAKE) $(AM_MAKEFLAGS) check-recursive |
|
881 | 882 |
all-am: Makefile $(LTLIBRARIES) $(HEADERS) |
882 | 883 |
installdirs: installdirs-recursive |
883 | 884 |
installdirs-am: |
884 | 885 |
for dir in "$(DESTDIR)$(libdir)" "$(DESTDIR)$(includedir)"; do \ |
885 | 886 |
test -z "$$dir" || $(MKDIR_P) "$$dir"; \ |
886 | 887 |
done |
887 |
-install: install-recursive |
|
888 |
+install: $(BUILT_SOURCES) |
|
889 |
+ $(MAKE) $(AM_MAKEFLAGS) install-recursive |
|
888 | 890 |
install-exec: install-exec-recursive |
889 | 891 |
install-data: install-data-recursive |
890 | 892 |
uninstall: uninstall-recursive |
... | ... |
@@ -901,6 +922,7 @@ install-strip: |
901 | 901 |
mostlyclean-generic: |
902 | 902 |
|
903 | 903 |
clean-generic: |
904 |
+ -test -z "$(CLEANFILES)" || rm -f $(CLEANFILES) |
|
904 | 905 |
|
905 | 906 |
distclean-generic: |
906 | 907 |
-test -z "$(CONFIG_CLEAN_FILES)" || rm -f $(CONFIG_CLEAN_FILES) |
... | ... |
@@ -908,6 +930,7 @@ distclean-generic: |
908 | 908 |
maintainer-clean-generic: |
909 | 909 |
@echo "This command is intended for maintainers to use" |
910 | 910 |
@echo "it deletes files that may require special tools to rebuild." |
911 |
+ -test -z "$(BUILT_SOURCES)" || rm -f $(BUILT_SOURCES) |
|
911 | 912 |
clean: clean-recursive |
912 | 913 |
|
913 | 914 |
clean-am: clean-generic clean-libLTLIBRARIES clean-libtool \ |
... | ... |
@@ -987,6 +1010,25 @@ uninstall-am: uninstall-includeHEADERS uninstall-libLTLIBRARIES |
987 | 987 |
tags tags-recursive uninstall uninstall-am \ |
988 | 988 |
uninstall-includeHEADERS uninstall-libLTLIBRARIES |
989 | 989 |
|
990 |
+ |
|
991 |
+@MAINTAINER_MODE_TRUE@jsparse-keywords.gperf: jsparse/keywords.list jsparse/future_reserved_words.list jsparse/special_keywords.list |
|
992 |
+@MAINTAINER_MODE_TRUE@ echo -e "struct keyword { const char *name; int val; };\n%%" >keywords-g-tmp |
|
993 |
+@MAINTAINER_MODE_TRUE@ for i in `cat @srcdir@/jsparse/keywords.list`; do j=`echo $$i |tr \[a-z\] \[A-Z\]`; echo "$$i, TOK_$$j" >>keywords-g-tmp; done |
|
994 |
+@MAINTAINER_MODE_TRUE@ for i in `cat @srcdir@/jsparse/future_reserved_words.list`; do echo "$$i, TOK_FUTURE_RESERVED_WORD" >>keywords-g-tmp; done |
|
995 |
+@MAINTAINER_MODE_TRUE@ cat @srcdir@/jsparse/special_keywords.list >>keywords-g-tmp |
|
996 |
+@MAINTAINER_MODE_TRUE@ mv keywords-g-tmp $@ |
|
997 |
+ |
|
998 |
+@MAINTAINER_MODE_TRUE@jsparse/generated/operators.h: jsparse/operators.gperf |
|
999 |
+@MAINTAINER_MODE_TRUE@ $(GPERF) $(GPERF_FLAGS) -H op_hash -N in_op_set -W oplist $< >operators-tmp-g |
|
1000 |
+@MAINTAINER_MODE_TRUE@ grep -v '^#line' <operators-tmp-g | sed -e 's/^const struct/static const struct/' -e 's/register //g' >operators-tmp |
|
1001 |
+@MAINTAINER_MODE_TRUE@ rm operators-tmp-g |
|
1002 |
+@MAINTAINER_MODE_TRUE@ mv operators-tmp @srcdir@/jsparse/generated/operators.h |
|
1003 |
+ |
|
1004 |
+@MAINTAINER_MODE_TRUE@jsparse/generated/keywords.h: jsparse-keywords.gperf |
|
1005 |
+@MAINTAINER_MODE_TRUE@ $(GPERF) $(GPERF_FLAGS) $< >keywords-tmp-g |
|
1006 |
+@MAINTAINER_MODE_TRUE@ grep -v '^#line' <keywords-tmp-g | sed -e 's/^const struct/static const struct/' -e 's/register //g' >keywords-tmp |
|
1007 |
+@MAINTAINER_MODE_TRUE@ rm keywords-tmp-g |
|
1008 |
+@MAINTAINER_MODE_TRUE@ mv keywords-tmp @srcdir@/jsparse/generated/keywords.h |
|
990 | 1009 |
# Tell versions [3.59,3.63) of GNU make to not export all variables. |
991 | 1010 |
# Otherwise a system limit (for SysV at least) may be exceeded. |
992 | 1011 |
.NOEXPORT: |
993 | 1012 |
new file mode 100644 |
... | ... |
@@ -0,0 +1,31 @@ |
0 |
+abstract |
|
1 |
+boolean |
|
2 |
+byte |
|
3 |
+char |
|
4 |
+class |
|
5 |
+const |
|
6 |
+debugger |
|
7 |
+double |
|
8 |
+enum |
|
9 |
+export |
|
10 |
+extends |
|
11 |
+final |
|
12 |
+float |
|
13 |
+goto |
|
14 |
+implements |
|
15 |
+import |
|
16 |
+int |
|
17 |
+interface |
|
18 |
+long |
|
19 |
+native |
|
20 |
+package |
|
21 |
+private |
|
22 |
+protected |
|
23 |
+public |
|
24 |
+short |
|
25 |
+static |
|
26 |
+super |
|
27 |
+synchronized |
|
28 |
+throws |
|
29 |
+transient |
|
30 |
+volatile |
0 | 31 |
new file mode 100644 |
... | ... |
@@ -0,0 +1,194 @@ |
0 |
+/* ANSI-C code produced by gperf version 3.0.3 */ |
|
1 |
+/* Command-line: gperf -E -t -L ANSI-C -C -F ', TOK_ERROR' -c jsparse-keywords.gperf */ |
|
2 |
+/* Computed positions: -k'1-2' */ |
|
3 |
+ |
|
4 |
+#if !((' ' == 32) && ('!' == 33) && ('"' == 34) && ('#' == 35) \ |
|
5 |
+ && ('%' == 37) && ('&' == 38) && ('\'' == 39) && ('(' == 40) \ |
|
6 |
+ && (')' == 41) && ('*' == 42) && ('+' == 43) && (',' == 44) \ |
|
7 |
+ && ('-' == 45) && ('.' == 46) && ('/' == 47) && ('0' == 48) \ |
|
8 |
+ && ('1' == 49) && ('2' == 50) && ('3' == 51) && ('4' == 52) \ |
|
9 |
+ && ('5' == 53) && ('6' == 54) && ('7' == 55) && ('8' == 56) \ |
|
10 |
+ && ('9' == 57) && (':' == 58) && (';' == 59) && ('<' == 60) \ |
|
11 |
+ && ('=' == 61) && ('>' == 62) && ('?' == 63) && ('A' == 65) \ |
|
12 |
+ && ('B' == 66) && ('C' == 67) && ('D' == 68) && ('E' == 69) \ |
|
13 |
+ && ('F' == 70) && ('G' == 71) && ('H' == 72) && ('I' == 73) \ |
|
14 |
+ && ('J' == 74) && ('K' == 75) && ('L' == 76) && ('M' == 77) \ |
|
15 |
+ && ('N' == 78) && ('O' == 79) && ('P' == 80) && ('Q' == 81) \ |
|
16 |
+ && ('R' == 82) && ('S' == 83) && ('T' == 84) && ('U' == 85) \ |
|
17 |
+ && ('V' == 86) && ('W' == 87) && ('X' == 88) && ('Y' == 89) \ |
|
18 |
+ && ('Z' == 90) && ('[' == 91) && ('\\' == 92) && (']' == 93) \ |
|
19 |
+ && ('^' == 94) && ('_' == 95) && ('a' == 97) && ('b' == 98) \ |
|
20 |
+ && ('c' == 99) && ('d' == 100) && ('e' == 101) && ('f' == 102) \ |
|
21 |
+ && ('g' == 103) && ('h' == 104) && ('i' == 105) && ('j' == 106) \ |
|
22 |
+ && ('k' == 107) && ('l' == 108) && ('m' == 109) && ('n' == 110) \ |
|
23 |
+ && ('o' == 111) && ('p' == 112) && ('q' == 113) && ('r' == 114) \ |
|
24 |
+ && ('s' == 115) && ('t' == 116) && ('u' == 117) && ('v' == 118) \ |
|
25 |
+ && ('w' == 119) && ('x' == 120) && ('y' == 121) && ('z' == 122) \ |
|
26 |
+ && ('{' == 123) && ('|' == 124) && ('}' == 125) && ('~' == 126)) |
|
27 |
+/* The character set is not based on ISO-646. */ |
|
28 |
+#error "gperf generated tables don't work with this execution character set. Please report a bug to <bug-gnu-gperf@gnu.org>." |
|
29 |
+#endif |
|
30 |
+ |
|
31 |
+struct keyword { const char *name; int val; }; |
|
32 |
+/* maximum key range = 100, duplicates = 0 */ |
|
33 |
+ |
|
34 |
+#ifdef __GNUC__ |
|
35 |
+__inline |
|
36 |
+#else |
|
37 |
+#ifdef __cplusplus |
|
38 |
+inline |
|
39 |
+#endif |
|
40 |
+#endif |
|
41 |
+static unsigned int |
|
42 |
+hash (const char *str, unsigned int len) |
|
43 |
+{ |
|
44 |
+ static const unsigned char asso_values[] = |
|
45 |
+ { |
|
46 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
47 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
48 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
49 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
50 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
51 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
52 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
53 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
54 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
55 |
+ 103, 103, 103, 103, 103, 103, 103, 30, 5, 0, |
|
56 |
+ 5, 0, 10, 50, 35, 5, 103, 103, 25, 55, |
|
57 |
+ 0, 20, 35, 103, 0, 40, 15, 5, 45, 55, |
|
58 |
+ 45, 50, 103, 103, 103, 103, 103, 103, 103, 103, |
|
59 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
60 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
61 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
62 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
63 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
64 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
65 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
66 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
67 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
68 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
69 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
70 |
+ 103, 103, 103, 103, 103, 103, 103, 103, 103, 103, |
|
71 |
+ 103, 103, 103, 103, 103, 103 |
|
72 |
+ }; |
|
73 |
+ return len + asso_values[(unsigned char)str[1]] + asso_values[(unsigned char)str[0]]; |
|
74 |
+} |
|
75 |
+ |
|
76 |
+#ifdef __GNUC__ |
|
77 |
+__inline |
|
78 |
+#ifdef __GNUC_STDC_INLINE__ |
|
79 |
+__attribute__ ((__gnu_inline__)) |
|
80 |
+#endif |
|
81 |
+#endif |
|
82 |
+static const struct keyword * |
|
83 |
+in_word_set (const char *str, unsigned int len) |
|
84 |
+{ |
|
85 |
+ enum |
|
86 |
+ { |
|
87 |
+ TOTAL_KEYWORDS = 59, |
|
88 |
+ MIN_WORD_LENGTH = 2, |
|
89 |
+ MAX_WORD_LENGTH = 12, |
|
90 |
+ MIN_HASH_VALUE = 3, |
|
91 |
+ MAX_HASH_VALUE = 102 |
|
92 |
+ }; |
|
93 |
+ |
|
94 |
+ static const struct keyword wordlist[] = |
|
95 |
+ { |
|
96 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
97 |
+ {"new", TOK_NEW}, |
|
98 |
+ {"enum", TOK_FUTURE_RESERVED_WORD}, |
|
99 |
+ {"", TOK_ERROR}, |
|
100 |
+ {"return", TOK_RETURN}, |
|
101 |
+ {"in", TOK_IN}, |
|
102 |
+ {"int", TOK_FUTURE_RESERVED_WORD}, |
|
103 |
+ {"null", TOK_NULL}, |
|
104 |
+ {"break", TOK_BREAK}, |
|
105 |
+ {"delete", TOK_DELETE}, |
|
106 |
+ {"default", TOK_DEFAULT}, |
|
107 |
+ {"debugger", TOK_FUTURE_RESERVED_WORD}, |
|
108 |
+ {"interface", TOK_FUTURE_RESERVED_WORD}, |
|
109 |
+ {"instanceof", TOK_INSTANCEOF}, |
|
110 |
+ {"", TOK_ERROR}, |
|
111 |
+ {"if", TOK_IF}, |
|
112 |
+ {"try", TOK_TRY}, |
|
113 |
+ {"true", TOK_TRUE}, |
|
114 |
+ {"final", TOK_FUTURE_RESERVED_WORD}, |
|
115 |
+ {"", TOK_ERROR}, |
|
116 |
+ {"finally", TOK_FINALLY}, |
|
117 |
+ {"function", TOK_FUNCTION}, |
|
118 |
+ {"transient", TOK_FUTURE_RESERVED_WORD}, |
|
119 |
+ {"const", TOK_FUTURE_RESERVED_WORD}, |
|
120 |
+ {"", TOK_ERROR}, |
|
121 |
+ {"do", TOK_DO}, |
|
122 |
+ {"continue", TOK_CONTINUE}, |
|
123 |
+ {"else", TOK_ELSE}, |
|
124 |
+ {"class", TOK_FUTURE_RESERVED_WORD}, |
|
125 |
+ {"double", TOK_FUTURE_RESERVED_WORD}, |
|
126 |
+ {"boolean", TOK_FUTURE_RESERVED_WORD}, |
|
127 |
+ {"for", TOK_FOR}, |
|
128 |
+ {"case", TOK_CASE}, |
|
129 |
+ {"catch", TOK_CATCH}, |
|
130 |
+ {"native", TOK_FUTURE_RESERVED_WORD}, |
|
131 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
132 |
+ {"char", TOK_FUTURE_RESERVED_WORD}, |
|
133 |
+ {"float", TOK_FUTURE_RESERVED_WORD}, |
|
134 |
+ {"", TOK_ERROR}, |
|
135 |
+ {"private", TOK_FUTURE_RESERVED_WORD}, |
|
136 |
+ {"abstract", TOK_FUTURE_RESERVED_WORD}, |
|
137 |
+ {"protected", TOK_FUTURE_RESERVED_WORD}, |
|
138 |
+ {"false", TOK_FALSE}, |
|
139 |
+ {"public", TOK_FUTURE_RESERVED_WORD}, |
|
140 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
141 |
+ {"long", TOK_FUTURE_RESERVED_WORD}, |
|
142 |
+ {"super", TOK_FUTURE_RESERVED_WORD}, |
|
143 |
+ {"export", TOK_FUTURE_RESERVED_WORD}, |
|
144 |
+ {"extends", TOK_FUTURE_RESERVED_WORD}, |
|
145 |
+ {"", TOK_ERROR}, |
|
146 |
+ {"this", TOK_THIS}, |
|
147 |
+ {"throw", TOK_THROW}, |
|
148 |
+ {"throws", TOK_FUTURE_RESERVED_WORD}, |
|
149 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
150 |
+ {"byte", TOK_FUTURE_RESERVED_WORD}, |
|
151 |
+ {"", TOK_ERROR}, |
|
152 |
+ {"static", TOK_FUTURE_RESERVED_WORD}, |
|
153 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
154 |
+ {"with", TOK_WITH}, |
|
155 |
+ {"", TOK_ERROR}, |
|
156 |
+ {"import", TOK_FUTURE_RESERVED_WORD}, |
|
157 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
158 |
+ {"void", TOK_VOID}, |
|
159 |
+ {"implements", TOK_FUTURE_RESERVED_WORD}, |
|
160 |
+ {"typeof", TOK_TYPEOF}, |
|
161 |
+ {"package", TOK_FUTURE_RESERVED_WORD}, |
|
162 |
+ {"volatile", TOK_FUTURE_RESERVED_WORD}, |
|
163 |
+ {"goto", TOK_FUTURE_RESERVED_WORD}, |
|
164 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
165 |
+ {"var", TOK_VAR}, |
|
166 |
+ {"", TOK_ERROR}, |
|
167 |
+ {"short", TOK_FUTURE_RESERVED_WORD}, |
|
168 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
169 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
170 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
171 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
172 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
173 |
+ {"while", TOK_WHILE}, |
|
174 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
175 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
176 |
+ {"switch", TOK_SWITCH}, |
|
177 |
+ {"synchronized", TOK_FUTURE_RESERVED_WORD} |
|
178 |
+ }; |
|
179 |
+ |
|
180 |
+ if (len <= MAX_WORD_LENGTH && len >= MIN_WORD_LENGTH) |
|
181 |
+ { |
|
182 |
+ int key = hash (str, len); |
|
183 |
+ |
|
184 |
+ if (key <= MAX_HASH_VALUE && key >= 0) |
|
185 |
+ { |
|
186 |
+ const char *s = wordlist[key].name; |
|
187 |
+ |
|
188 |
+ if (*str == *s && !strncmp (str + 1, s + 1, len - 1) && s[len] == '\0') |
|
189 |
+ return &wordlist[key]; |
|
190 |
+ } |
|
191 |
+ } |
|
192 |
+ return 0; |
|
193 |
+} |
0 | 194 |
new file mode 100644 |
... | ... |
@@ -0,0 +1,186 @@ |
0 |
+/* ANSI-C code produced by gperf version 3.0.3 */ |
|
1 |
+/* Command-line: gperf -E -t -L ANSI-C -C -F ', TOK_ERROR' -c -H op_hash -N in_op_set -W oplist ../../../trunk/libclamav/jsparse/operators.gperf */ |
|
2 |
+/* Computed positions: -k'1,$' */ |
|
3 |
+ |
|
4 |
+#if !((' ' == 32) && ('!' == 33) && ('"' == 34) && ('#' == 35) \ |
|
5 |
+ && ('%' == 37) && ('&' == 38) && ('\'' == 39) && ('(' == 40) \ |
|
6 |
+ && (')' == 41) && ('*' == 42) && ('+' == 43) && (',' == 44) \ |
|
7 |
+ && ('-' == 45) && ('.' == 46) && ('/' == 47) && ('0' == 48) \ |
|
8 |
+ && ('1' == 49) && ('2' == 50) && ('3' == 51) && ('4' == 52) \ |
|
9 |
+ && ('5' == 53) && ('6' == 54) && ('7' == 55) && ('8' == 56) \ |
|
10 |
+ && ('9' == 57) && (':' == 58) && (';' == 59) && ('<' == 60) \ |
|
11 |
+ && ('=' == 61) && ('>' == 62) && ('?' == 63) && ('A' == 65) \ |
|
12 |
+ && ('B' == 66) && ('C' == 67) && ('D' == 68) && ('E' == 69) \ |
|
13 |
+ && ('F' == 70) && ('G' == 71) && ('H' == 72) && ('I' == 73) \ |
|
14 |
+ && ('J' == 74) && ('K' == 75) && ('L' == 76) && ('M' == 77) \ |
|
15 |
+ && ('N' == 78) && ('O' == 79) && ('P' == 80) && ('Q' == 81) \ |
|
16 |
+ && ('R' == 82) && ('S' == 83) && ('T' == 84) && ('U' == 85) \ |
|
17 |
+ && ('V' == 86) && ('W' == 87) && ('X' == 88) && ('Y' == 89) \ |
|
18 |
+ && ('Z' == 90) && ('[' == 91) && ('\\' == 92) && (']' == 93) \ |
|
19 |
+ && ('^' == 94) && ('_' == 95) && ('a' == 97) && ('b' == 98) \ |
|
20 |
+ && ('c' == 99) && ('d' == 100) && ('e' == 101) && ('f' == 102) \ |
|
21 |
+ && ('g' == 103) && ('h' == 104) && ('i' == 105) && ('j' == 106) \ |
|
22 |
+ && ('k' == 107) && ('l' == 108) && ('m' == 109) && ('n' == 110) \ |
|
23 |
+ && ('o' == 111) && ('p' == 112) && ('q' == 113) && ('r' == 114) \ |
|
24 |
+ && ('s' == 115) && ('t' == 116) && ('u' == 117) && ('v' == 118) \ |
|
25 |
+ && ('w' == 119) && ('x' == 120) && ('y' == 121) && ('z' == 122) \ |
|
26 |
+ && ('{' == 123) && ('|' == 124) && ('}' == 125) && ('~' == 126)) |
|
27 |
+/* The character set is not based on ISO-646. */ |
|
28 |
+#error "gperf generated tables don't work with this execution character set. Please report a bug to <bug-gnu-gperf@gnu.org>." |
|
29 |
+#endif |
|
30 |
+ |
|
31 |
+struct operator { |
|
32 |
+ const char *name; |
|
33 |
+ int val; |
|
34 |
+}; |
|
35 |
+/* maximum key range = 121, duplicates = 0 */ |
|
36 |
+ |
|
37 |
+#ifdef __GNUC__ |
|
38 |
+__inline |
|
39 |
+#else |
|
40 |
+#ifdef __cplusplus |
|
41 |
+inline |
|
42 |
+#endif |
|
43 |
+#endif |
|
44 |
+static unsigned int |
|
45 |
+op_hash (const char *str, unsigned int len) |
|
46 |
+{ |
|
47 |
+ static const unsigned char asso_values[] = |
|
48 |
+ { |
|
49 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
50 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
51 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
52 |
+ 122, 122, 122, 50, 122, 122, 122, 31, 40, 122, |
|
53 |
+ 122, 122, 21, 30, 122, 25, 122, 16, 122, 122, |
|
54 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 45, 122, |
|
55 |
+ 10, 5, 0, 35, 122, 122, 122, 122, 122, 122, |
|
56 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
57 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
58 |
+ 122, 122, 122, 122, 60, 122, 122, 122, 122, 122, |
|
59 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
60 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
61 |
+ 122, 122, 122, 122, 20, 122, 15, 122, 122, 122, |
|
62 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
63 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
64 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
65 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
66 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
67 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
68 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
69 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
70 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
71 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
72 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
73 |
+ 122, 122, 122, 122, 122, 122, 122, 122, 122, 122, |
|
74 |
+ 122, 122, 122, 122, 122, 122 |
|
75 |
+ }; |
|
76 |
+ return len + asso_values[(unsigned char)str[len - 1]] + asso_values[(unsigned char)str[0]]; |
|
77 |
+} |
|
78 |
+ |
|
79 |
+#ifdef __GNUC__ |
|
80 |
+__inline |
|
81 |
+#ifdef __GNUC_STDC_INLINE__ |
|
82 |
+__attribute__ ((__gnu_inline__)) |
|
83 |
+#endif |
|
84 |
+#endif |
|
85 |
+static const struct operator * |
|
86 |
+in_op_set (const char *str, unsigned int len) |
|
87 |
+{ |
|
88 |
+ enum |
|
89 |
+ { |
|
90 |
+ TOTAL_KEYWORDS = 39, |
|
91 |
+ MIN_WORD_LENGTH = 1, |
|
92 |
+ MAX_WORD_LENGTH = 4, |
|
93 |
+ MIN_HASH_VALUE = 1, |
|
94 |
+ MAX_HASH_VALUE = 121 |
|
95 |
+ }; |
|
96 |
+ |
|
97 |
+ static const struct operator oplist[] = |
|
98 |
+ { |
|
99 |
+ {"", TOK_ERROR}, |
|
100 |
+ {">", TOK_GREATER}, |
|
101 |
+ {">>", TOK_SHIFT_RIGHT}, |
|
102 |
+ {">>>", TOK_DOUBLESHIFT_RIGHT}, |
|
103 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
104 |
+ {">=", TOK_GREATEREQUAL}, |
|
105 |
+ {">>=", TOK_ASSIGNMENT_OPERATOR_NOEQUAL}, |
|
106 |
+ {">>>=", TOK_ASSIGNMENT_OPERATOR_NOEQUAL}, |
|
107 |
+ {"", TOK_ERROR}, |
|
108 |
+ {"=", TOK_EQUAL}, |
|
109 |
+ {"==", TOK_EQUAL_EQUAL}, |
|
110 |
+ {"===", TOK_TRIPLE_EQUAL}, |
|
111 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
112 |
+ {"<=", TOK_LESSEQUAL}, |
|
113 |
+ {"<<=", TOK_ASSIGNMENT_OPERATOR_NOEQUAL}, |
|
114 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
115 |
+ {"<", TOK_LESS}, |
|
116 |
+ {"<<", TOK_SHIFT_LEFT}, |
|
117 |
+ {"/=", TOK_ASSIGNMENT_OPERATOR_NOEQUAL}, |
|
118 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
119 |
+ {"|=", TOK_ASSIGNMENT_OPERATOR_NOEQUAL}, |
|
120 |
+ {"*=", TOK_ASSIGNMENT_OPERATOR_NOEQUAL}, |
|
121 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
122 |
+ {"~", TOK_TILDE}, |
|
123 |
+ {"-=", TOK_ASSIGNMENT_OPERATOR_NOEQUAL}, |
|
124 |
+ {"/", TOK_DIVIDE}, |
|
125 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
126 |
+ {"+=", TOK_ASSIGNMENT_OPERATOR_NOEQUAL}, |
|
127 |
+ {"%=", TOK_ASSIGNMENT_OPERATOR_NOEQUAL}, |
|
128 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
129 |
+ {"|", TOK_OR}, |
|
130 |
+ {"||", TOK_OR_OR}, |
|
131 |
+ {"*", TOK_MULTIPLY}, |
|
132 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
133 |
+ {"&=", TOK_ASSIGNMENT_OPERATOR_NOEQUAL}, |
|
134 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
135 |
+ {"-", TOK_MINUS}, |
|
136 |
+ {"--", TOK_MINUSMINUS}, |
|
137 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
138 |
+ {"", TOK_ERROR}, |
|
139 |
+ {"!=", TOK_NOT_EQUAL}, |
|
140 |
+ {"!==", TOK_NOT_DOUBLEEQUAL}, |
|
141 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
142 |
+ {"+", TOK_PLUS}, |
|
143 |
+ {"++", TOK_PLUSPLUS}, |
|
144 |
+ {"%", TOK_PERCENT}, |
|
145 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
146 |
+ {"^=", TOK_ASSIGNMENT_OPERATOR_NOEQUAL}, |
|
147 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
148 |
+ {"?", TOK_QUESTIONMARK}, |
|
149 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
150 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
151 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
152 |
+ {"&", TOK_AND}, |
|
153 |
+ {"&&", TOK_AND_AND}, |
|
154 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
155 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
156 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
157 |
+ {":", TOK_COLON}, |
|
158 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
159 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
160 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
161 |
+ {"!", TOK_EXCLAMATION}, |
|
162 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
163 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
164 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
165 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
166 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
167 |
+ {"", TOK_ERROR}, {"", TOK_ERROR}, {"", TOK_ERROR}, |
|
168 |
+ {"", TOK_ERROR}, |
|
169 |
+ {"^", TOK_XOR} |
|
170 |
+ }; |
|
171 |
+ |
|
172 |
+ if (len <= MAX_WORD_LENGTH && len >= MIN_WORD_LENGTH) |
|
173 |
+ { |
|
174 |
+ int key = op_hash (str, len); |
|
175 |
+ |
|
176 |
+ if (key <= MAX_HASH_VALUE && key >= 0) |
|
177 |
+ { |
|
178 |
+ const char *s = oplist[key].name; |
|
179 |
+ |
|
180 |
+ if (*str == *s && !strncmp (str + 1, s + 1, len - 1) && s[len] == '\0') |
|
181 |
+ return &oplist[key]; |
|
182 |
+ } |
|
183 |
+ } |
|
184 |
+ return 0; |
|
185 |
+} |
0 | 186 |
new file mode 100644 |
... | ... |
@@ -0,0 +1,1672 @@ |
0 |
+/* |
|
1 |
+ * Javascript normalizer. |
|
2 |
+ * |
|
3 |
+ * Copyright (C) 2008 Sourcefire, Inc. |
|
4 |
+ * |
|
5 |
+ * Authors: Török Edvin |
|
6 |
+ * |
|
7 |
+ * This program is free software; you can redistribute it and/or modify |
|
8 |
+ * it under the terms of the GNU General Public License version 2 as |
|
9 |
+ * published by the Free Software Foundation. |
|
10 |
+ * |
|
11 |
+ * This program is distributed in the hope that it will be useful, |
|
12 |
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
13 |
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
14 |
+ * GNU General Public License for more details. |
|
15 |
+ * |
|
16 |
+ * You should have received a copy of the GNU General Public License |
|
17 |
+ * along with this program; if not, write to the Free Software |
|
18 |
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, |
|
19 |
+ * MA 02110-1301, USA. |
|
20 |
+ */ |
|
21 |
+ |
|
22 |
+#include <stdio.h> |
|
23 |
+#include <unistd.h> |
|
24 |
+#include <sys/types.h> |
|
25 |
+#include <sys/stat.h> |
|
26 |
+#include <fcntl.h> |
|
27 |
+#include <stdlib.h> |
|
28 |
+#include <stdint.h> |
|
29 |
+#include <string.h> |
|
30 |
+#include <ctype.h> |
|
31 |
+#include <assert.h> |
|
32 |
+#define BUFS 65536 |
|
33 |
+#include "lexglobal.h" |
|
34 |
+#include "hashtab.h" |
|
35 |
+#include "others.h" |
|
36 |
+#include "js-norm.h" |
|
37 |
+#include "jsparse/generated/operators.h" |
|
38 |
+#include "jsparse/generated/keywords.h" |
|
39 |
+ |
|
40 |
+/* ----------- tokenizer ---------------- */ |
|
41 |
+enum tokenizer_state { |
|
42 |
+ Initial, |
|
43 |
+ MultilineComment, |
|
44 |
+ SinglelineComment, |
|
45 |
+ Number, |
|
46 |
+ DoubleQString, |
|
47 |
+ SingleQString, |
|
48 |
+ Identifier |
|
49 |
+}; |
|
50 |
+ |
|
51 |
+struct text_buffer { |
|
52 |
+ char *data; |
|
53 |
+ size_t pos; |
|
54 |
+ size_t capacity; |
|
55 |
+}; |
|
56 |
+ |
|
57 |
+typedef struct scanner { |
|
58 |
+ enum tokenizer_state state; |
|
59 |
+ struct text_buffer buf; |
|
60 |
+ const char *yytext; |
|
61 |
+ size_t yylen; |
|
62 |
+ const char *in; |
|
63 |
+ size_t insize; |
|
64 |
+ size_t pos; |
|
65 |
+} *yyscan_t; |
|
66 |
+ |
|
67 |
+typedef int YY_BUFFER_STATE; |
|
68 |
+ |
|
69 |
+static int yylex( YYSTYPE *lvalp, yyscan_t ); |
|
70 |
+static void yy_delete_buffer( YY_BUFFER_STATE, yyscan_t); |
|
71 |
+static YY_BUFFER_STATE yy_scan_bytes( const char *, size_t, yyscan_t scanner ); |
|
72 |
+static const char *yyget_text ( yyscan_t scanner ); |
|
73 |
+static int yyget_leng ( yyscan_t scanner ); |
|
74 |
+static int yylex_init ( yyscan_t * ptr_yy_globals ) ; |
|
75 |
+static void yyset_debug (int debug_flag ,yyscan_t yyscanner ); |
|
76 |
+static int yylex_destroy ( yyscan_t yyscanner ) ; |
|
77 |
+/* ----------- tokenizer end ---------------- */ |
|
78 |
+ |
|
79 |
+enum fsm_state { |
|
80 |
+ Base, |
|
81 |
+ InsideVar, |
|
82 |
+ InsideInitializer, |
|
83 |
+ WaitFunctionName, |
|
84 |
+ WaitParameterList, |
|
85 |
+ InsideFunctionDecl |
|
86 |
+}; |
|
87 |
+ |
|
88 |
+struct scope { |
|
89 |
+ struct hashtable id_map; |
|
90 |
+ struct scope *parent;/* hierarchy */ |
|
91 |
+ struct scope *nxt;/* all scopes kept in a list so we can easily free all of them */ |
|
92 |
+ enum fsm_state fsm_state; |
|
93 |
+ int last_token; |
|
94 |
+ unsigned int brackets; |
|
95 |
+ unsigned int blocks; |
|
96 |
+}; |
|
97 |
+ |
|
98 |
+struct tokens { |
|
99 |
+ yystype *data; |
|
100 |
+ size_t cnt; |
|
101 |
+ size_t capacity; |
|
102 |
+}; |
|
103 |
+ |
|
104 |
+/* state for the current JS file being parsed */ |
|
105 |
+struct parser_state { |
|
106 |
+ unsigned long var_uniq; |
|
107 |
+ unsigned long syntax_errors; |
|
108 |
+ struct scope *global; |
|
109 |
+ struct scope *current; |
|
110 |
+ struct scope *list; |
|
111 |
+ yyscan_t scanner; |
|
112 |
+ struct tokens tokens; |
|
113 |
+}; |
|
114 |
+ |
|
115 |
+static struct scope* scope_new(struct parser_state *state) |
|
116 |
+{ |
|
117 |
+ struct scope *parent = state->current; |
|
118 |
+ struct scope *s = cli_calloc(1, sizeof(*s)); |
|
119 |
+ if(!s) |
|
120 |
+ return NULL; |
|
121 |
+ if(hashtab_init(&s->id_map, 10) < 0) { |
|
122 |
+ free(s); |
|
123 |
+ return NULL; |
|
124 |
+ } |
|
125 |
+ s->parent = parent; |
|
126 |
+ s->fsm_state = Base; |
|
127 |
+ s->nxt = state->list; |
|
128 |
+ state->list = s; |
|
129 |
+ state->current = s; |
|
130 |
+ return s; |
|
131 |
+} |
|
132 |
+ |
|
133 |
+static struct scope* scope_done(struct scope *s) |
|
134 |
+{ |
|
135 |
+ struct scope* parent = s->parent; |
|
136 |
+ /* TODO: have a hashtab_destroy */ |
|
137 |
+ hashtab_clear(&s->id_map); |
|
138 |
+ free(s->id_map.htable); |
|
139 |
+ free(s); |
|
140 |
+ return parent; |
|
141 |
+} |
|
142 |
+ |
|
143 |
+/* transitions: |
|
144 |
+ * Base --(VAR)--> InsideVar |
|
145 |
+ * InsideVar --(Identifier)-->InsideInitializer |
|
146 |
+ * InsideVar --(anything_else) --> POP (to Base) |
|
147 |
+ * InsideInitializer --(COMMA)--> POP (to InsideVar) |
|
148 |
+ * InsideInitializer | InsideVar --(SEMICOLON) --> POP (to Base) |
|
149 |
+ * InsideInitializer --(BRACKET_OPEN) --> WaitBrClose |
|
150 |
+ * InsideInitializer --(PAR_OPEN) --> WaitParClose |
|
151 |
+ * WaitBrClose --(BRACKET_OPEN) --> increase depth |
|
152 |
+ * WaitBrClose --(BRACKET_CLOSE) --> POP |
|
153 |
+ * WaitParClose --(PAR_CLOSE) --> POP |
|
154 |
+ * WaitParClose --(PAR_OPEN) --> increase depth |
|
155 |
+ */ |
|
156 |
+ |
|
157 |
+/* Base --(VAR)--> PUSH, to InsideVar |
|
158 |
+ * InsideVar --(Identifier)--> InsideInitializer |
|
159 |
+ * InsideVar --(ELSE)--> POP, inc. syntax_errors |
|
160 |
+ * InsideInitializer --(COMMA)--> POP (to InsideVar) |
|
161 |
+ * --(BRACKET_OPEN)--> inc bracket_counter |
|
162 |
+ * --(PAR_OPEN)--> inc par_counter |
|
163 |
+ * --(BRACKET_CLOSE) --> dec bracket_counter |
|
164 |
+ * --(PAR_CLOSE)--> dec par_counter |
|
165 |
+ * --(VAR)--> PUSH, to InsideVar (if bracket_counter != 0 || par_counter != 0) |
|
166 |
+ * --> POP, to InsideVar, inc. syntax_errors (if bracket_counter == 0 && par_counter == 0) |
|
167 |
+ * POP only allowed if bracket_counter == 0 && par_counter == 0 |
|
168 |
+ * |
|
169 |
+ * InsideInitializer acts differently, make it only a flag |
|
170 |
+ * .................... |
|
171 |
+ * |
|
172 |
+ * Pushing, Poping is done when entering / exiting function scopes, |
|
173 |
+ * tracking { and function ( is done by the function scope tracker too. |
|
174 |
+ * |
|
175 |
+ * we only need to track brackets. |
|
176 |
+ */ |
|
177 |
+ |
|
178 |
+ |
|
179 |
+/* |
|
180 |
+ * var x = document; |
|
181 |
+ * x.writeln(...); |
|
182 |
+ * |
|
183 |
+ * ^we must not normalize member method names |
|
184 |
+ */ |
|
185 |
+ |
|
186 |
+/* |
|
187 |
+ * Variables are declared at function scope, and their initial value is |
|
188 |
+ * undefined. At the point where the initializer is, and from there on the value |
|
189 |
+ * is defined. |
|
190 |
+ * |
|
191 |
+ * { doesn't introduce a new variable scope, they are in function's scope too |
|
192 |
+ * |
|
193 |
+ * function foo() { |
|
194 |
+ * alert(x); -> x exists, undefined |
|
195 |
+ * var x=5; |
|
196 |
+ * alert(x); -> x exists, =5 |
|
197 |
+ * } |
|
198 |
+ * |
|
199 |
+ * vs. |
|
200 |
+ * |
|
201 |
+ * function bar() { |
|
202 |
+ * alert(x);//error, x not declared |
|
203 |
+ * x=5; |
|
204 |
+ * } |
|
205 |
+ * |
|
206 |
+ * vs. |
|
207 |
+ * |
|
208 |
+ * but we can declare variables without var, only valid if we use them after |
|
209 |
+ * assigning. |
|
210 |
+ * |
|
211 |
+ * function foobar() { |
|
212 |
+ * x=5; |
|
213 |
+ * alert(x);//x is defined, value is 5 |
|
214 |
+ * } |
|
215 |
+ * |
|
216 |
+ * other examples: |
|
217 |
+ * function foo2() { |
|
218 |
+ * alert(x); -> x exists, undefined |
|
219 |
+ * { |
|
220 |
+ * var x=5; -> x equals to 5 |
|
221 |
+ * } |
|
222 |
+ * alert(x); -> x is 5 |
|
223 |
+ * } |
|
224 |
+ * |
|
225 |
+ * function foo3() { |
|
226 |
+ * var x=4; -> x exists, equals to 4 |
|
227 |
+ * alert(x); -> x exists, equals to 4 |
|
228 |
+ * { |
|
229 |
+ * var x=5; -> x equals to 5 |
|
230 |
+ * } |
|
231 |
+ * alert(x); -> x is 5 |
|
232 |
+ * } |
|
233 |
+ * |
|
234 |
+ * function bar3() { |
|
235 |
+ * //same as foo3 |
|
236 |
+ * var x=4; |
|
237 |
+ * alert(x); |
|
238 |
+ * { |
|
239 |
+ * x=5; |
|
240 |
+ * } |
|
241 |
+ * alert(x); |
|
242 |
+ * } |
|
243 |
+ * |
|
244 |
+ */ |
|
245 |
+ |
|
246 |
+ |
|
247 |
+static const char* scope_declare(struct scope *s, const char *token, const size_t len, struct parser_state *state) |
|
248 |
+{ |
|
249 |
+ const struct element *el = hashtab_insert(&s->id_map, token, len, state->var_uniq++); |
|
250 |
+ /* hashtab_insert either finds an already existing entry, or allocates a |
|
251 |
+ * new one, we return the allocated string */ |
|
252 |
+ return el ? el->key : NULL; |
|
253 |
+} |
|
254 |
+ |
|
255 |
+static const char* scope_use(struct scope *s, const char *token, const size_t len) |
|
256 |
+{ |
|
257 |
+ const struct element *el = hashtab_find(&s->id_map, token, len); |
|
258 |
+ if(el) { |
|
259 |
+ /* identifier already found in current scope, |
|
260 |
+ * return here to avoid overwriting uniq id */ |
|
261 |
+ return el->key; |
|
262 |
+ } |
|
263 |
+ /* identifier not yet in current scope's hashtab, add with ID -1. |
|
264 |
+ * Later if we find a declaration it will automatically assign a uniq ID |
|
265 |
+ * to it. If not, we'll know that we have to push ID == -1 tokens to an |
|
266 |
+ * outer scope.*/ |
|
267 |
+ el = hashtab_insert(&s->id_map, token, len, -1); |
|
268 |
+ return el ? el->key : NULL; |
|
269 |
+} |
|
270 |
+ |
|
271 |
+static long scope_lookup(struct scope *s, const char *token, const size_t len) |
|
272 |
+{ |
|
273 |
+ while(s) { |
|
274 |
+ const struct element *el = hashtab_find(&s->id_map, token, len); |
|
275 |
+ if(el && el->data != -1) { |
|
276 |
+ return el->data; |
|
277 |
+ } |
|
278 |
+ /* not found in current scope, try in outer scope */ |
|
279 |
+ s = s->parent; |
|
280 |
+ } |
|
281 |
+ return -1; |
|
282 |
+} |
|
283 |
+ |
|
284 |
+static int tokens_ensure_capacity(struct tokens *tokens, size_t cap) |
|
285 |
+{ |
|
286 |
+ if(tokens->capacity < cap) { |
|
287 |
+ tokens->capacity = cap + 1024; |
|
288 |
+ tokens->data = cli_realloc2(tokens->data, tokens->capacity * sizeof(*tokens->data)); |
|
289 |
+ if(!tokens->data) |
|
290 |
+ return CL_EMEM; |
|
291 |
+ } |
|
292 |
+ return CL_SUCCESS; |
|
293 |
+} |
|
294 |
+ |
|
295 |
+static int add_token(struct parser_state *state, const yystype *token) |
|
296 |
+{ |
|
297 |
+ if(tokens_ensure_capacity(&state->tokens, state->tokens.cnt + 1) == -1) |
|
298 |
+ return -1; |
|
299 |
+ state->tokens.data[state->tokens.cnt++] = *token; |
|
300 |
+ return 0; |
|
301 |
+} |
|
302 |
+ |
|
303 |
+struct buf { |
|
304 |
+ size_t pos; |
|
305 |
+ int outfd; |
|
306 |
+ char buf[65536]; |
|
307 |
+}; |
|
308 |
+ |
|
309 |
+static inline int buf_outc(char c, struct buf *buf) |
|
310 |
+{ |
|
311 |
+ if(buf->pos >= sizeof(buf->buf)) { |
|
312 |
+ if(write(buf->outfd, buf->buf, sizeof(buf->buf)) != sizeof(buf->buf)) |
|
313 |
+ return CL_EIO; |
|
314 |
+ buf->pos = 0; |
|
315 |
+ } |
|
316 |
+ buf->buf[buf->pos++] = c; |
|
317 |
+ return CL_SUCCESS; |
|
318 |
+} |
|
319 |
+ |
|
320 |
+static inline int buf_outs(const char *s, struct buf *buf) |
|
321 |
+{ |
|
322 |
+ const size_t buf_len = sizeof(buf->buf); |
|
323 |
+ size_t len = strlen(s); |
|
324 |
+ while(buf->pos + len > buf_len) { |
|
325 |
+ memcpy(buf->buf + buf->pos, s, buf_len - buf->pos); |
|
326 |
+ len -= (buf_len - buf->pos); |
|
327 |
+ if(write(buf->outfd, buf->buf, buf_len) < 0) |
|
328 |
+ return CL_EIO; |
|
329 |
+ buf->pos = 0; |
|
330 |
+ } |
|
331 |
+ memcpy(buf->buf + buf->pos, s, len); |
|
332 |
+ buf->pos += len; |
|
333 |
+ return CL_SUCCESS; |
|
334 |
+} |
|
335 |
+ |
|
336 |
+static inline void output_space(char last, char current, struct buf *out) |
|
337 |
+{ |
|
338 |
+ if(isalnum(last) && isalnum(current)) |
|
339 |
+ buf_outc(' ', out); |
|
340 |
+} |
|
341 |
+ |
|
342 |
+ |
|
343 |
+/* return class of last character */ |
|
344 |
+static char output_token(const yystype *token, struct scope *scope, struct buf *out, char lastchar) |
|
345 |
+{ |
|
346 |
+ char sbuf[128]; |
|
347 |
+ const char *s = TOKEN_GET(token, cstring); |
|
348 |
+ /* TODO: use a local buffer, instead of FILE* */ |
|
349 |
+ switch(token->type) { |
|
350 |
+ case TOK_StringLiteral: |
|
351 |
+ output_space(lastchar,'"', out); |
|
352 |
+ buf_outc('"', out); |
|
353 |
+ if(s) { |
|
354 |
+ buf_outs(s, out); |
|
355 |
+ } |
|
356 |
+ buf_outc('"', out); |
|
357 |
+ return '\"'; |
|
358 |
+ case TOK_NumericInt: |
|
359 |
+ output_space(lastchar,'0', out); |
|
360 |
+ snprintf(sbuf, sizeof(sbuf), "%ld", TOKEN_GET(token, ival)); |
|
361 |
+ buf_outs(sbuf, out); |
|
362 |
+ return '0'; |
|
363 |
+ case TOK_NumericFloat: |
|
364 |
+ output_space(lastchar,'0', out); |
|
365 |
+ snprintf(sbuf, sizeof(sbuf), "%e", TOKEN_GET(token, dval)); |
|
366 |
+ buf_outs(sbuf, out); |
|
367 |
+ return '0'; |
|
368 |
+ case TOK_IDENTIFIER_NAME: |
|
369 |
+ /* TODO: lookup identifier name here, and normalize it |
|
370 |
+ * */ |
|
371 |
+ output_space(lastchar,'a', out); |
|
372 |
+ if(s) { |
|
373 |
+ long id = scope_lookup(scope, s, strlen(s)); |
|
374 |
+ if(id == -1) { |
|
375 |
+ /* identifier not normalized */ |
|
376 |
+ buf_outs(s, out); |
|
377 |
+ } else { |
|
378 |
+ snprintf(sbuf, sizeof(sbuf), "n%03ld",id); |
|
379 |
+ buf_outs(sbuf, out); |
|
380 |
+ } |
|
381 |
+ } |
|
382 |
+ return 'a'; |
|
383 |
+ case TOK_FUNCTION: |
|
384 |
+ /*TODO: output function name */ |
|
385 |
+ output_space(lastchar,'a', out); |
|
386 |
+ buf_outs("function",out); |
|
387 |
+ return 'a'; |
|
388 |
+ default: |
|
389 |
+ if(s) { |
|
390 |
+ const size_t len = strlen(s); |
|
391 |
+ output_space(lastchar,s[0], out); |
|
392 |
+ buf_outs(s, out); |
|
393 |
+ return len ? s[len-1] : '\0'; |
|
394 |
+ } |
|
395 |
+ return '\0'; |
|
396 |
+ } |
|
397 |
+} |
|
398 |
+ |
|
399 |
+/* |
|
400 |
+ * We can't delete the scope as soon as we see a }, because |
|
401 |
+ * we still need the hashmap from it. |
|
402 |
+ * |
|
403 |
+ * If we would normalize all the identifiers, and output when a scope is closed, |
|
404 |
+ * then it would be impossible to normalize calls to other functions. |
|
405 |
+ * |
|
406 |
+ * So we need to keep all scopes in memory, to do this instead of scope_done, we |
|
407 |
+ * simply just set current = current->parent when a scope is closed. |
|
408 |
+ * We keep a list of all scopes created in parser_state-> When we parsed |
|
409 |
+ * everything, we output everything, and then delete all scopes. |
|
410 |
+ * |
|
411 |
+ * We also need to know where to switch scopes on the second pass, so for |
|
412 |
+ * TOK_FUNCTION types we will use another pointer, that points to the scope |
|
413 |
+ * (added to yystype's union). |
|
414 |
+ * |
|
415 |
+ * We lookup the identifier in the scope (using scope_lookup, it looks in parent |
|
416 |
+ * scopes too), if ID is found then output (n%3d, Id), |
|
417 |
+ * otherwise output the identifier as is. |
|
418 |
+ * |
|
419 |
+ * To make it easier to match sigs, we do a xfrm : |
|
420 |
+ * 'function ID1 (..'. => 'n%3d = function (...' |
|
421 |
+ */ |
|
422 |
+ |
|
423 |
+/* |
|
424 |
+ * we'll add all identifier to the scope's map |
|
425 |
+ * those that are not decl. will have initial ID -1 |
|
426 |
+ * if we later see a decl for it in same scope, it'll automatically get a |
|
427 |
+ * correct ID. |
|
428 |
+ * |
|
429 |
+ * When parsing of local scope is done, we take any ID -1 identifiers, |
|
430 |
+ * and push them up one level (careful not to overwrite existing IDs). |
|
431 |
+ * |
|
432 |
+ * it would be nice if the tokens would contain a link to the entry in the |
|
433 |
+ * hashtab, a link that automatically gets updated when the element is moved |
|
434 |
+ * (pushed up). This would prevent subsequent lookups in the map, |
|
435 |
+ * when we want to output the tokens. |
|
436 |
+ * There is no easy way to do that, so we just do another lookup |
|
437 |
+ * |
|
438 |
+ */ |
|
439 |
+ |
|
440 |
+/* |
|
441 |
+ * This actually works, redefining foo: |
|
442 |
+ * function foo() { |
|
443 |
+ * var foo=5; alert(foo); |
|
444 |
+ * } |
|
445 |
+ * So we can't treat function names just as any other identifier? |
|
446 |
+ * We can, because you can no longer call foo, if you redefined it as a var. |
|
447 |
+ * So if we rename both foo-s with same name, it will have same behaviour. |
|
448 |
+ * |
|
449 |
+ * This means that a new scope should begin after function, and not after |
|
450 |
+ * function ... (. |
|
451 |
+ */ |
|
452 |
+ |
|
453 |
+static void scope_free_all(struct scope *p) |
|
454 |
+{ |
|
455 |
+ struct scope *nxt; |
|
456 |
+ do { |
|
457 |
+ nxt = p->nxt; |
|
458 |
+ scope_done(p); |
|
459 |
+ p = nxt; |
|
460 |
+ } while(p); |
|
461 |
+} |
|
462 |
+ |
|
463 |
+void cli_strtokenize(char *buffer, const char delim, const size_t token_count, const char **tokens); |
|
464 |
+static int match_parameters(const yystype *tokens, const char ** param_names, size_t count) |
|
465 |
+{ |
|
466 |
+ size_t i,j=0; |
|
467 |
+ if(tokens[0].type != TOK_PAR_OPEN) |
|
468 |
+ return -1; |
|
469 |
+ i=1; |
|
470 |
+ while(count--) { |
|
471 |
+ const char *token_val = TOKEN_GET(&tokens[i], cstring); |
|
472 |
+ if(tokens[i].type != TOK_IDENTIFIER_NAME || |
|
473 |
+ !token_val || |
|
474 |
+ strcmp(token_val, param_names[j++])) |
|
475 |
+ return -1; |
|
476 |
+ ++i; |
|
477 |
+ if((count && tokens[i].type != TOK_COMMA) |
|
478 |
+ || (!count && tokens[i].type != TOK_PAR_CLOSE)) |
|
479 |
+ return -1; |
|
480 |
+ ++i; |
|
481 |
+ } |
|
482 |
+ return 0; |
|
483 |
+} |
|
484 |
+ |
|
485 |
+static const char *de_packer_3[] = {"p","a","c","k","e","r"}; |
|
486 |
+static const char *de_packer_2[] = {"p","a","c","k","e","d"}; |
|
487 |
+ |
|
488 |
+ |
|
489 |
+#ifndef MAX |
|
490 |
+#define MAX(a, b) ((a)>(b) ? (a) : (b)) |
|
491 |
+#endif |
|
492 |
+ |
|
493 |
+static inline char *textbuffer_done(yyscan_t scanner) |
|
494 |
+{ |
|
495 |
+ /* free unusued memory */ |
|
496 |
+ char *str = cli_realloc(scanner->buf.data, scanner->buf.pos); |
|
497 |
+ if(!str) { |
|
498 |
+ str = scanner->buf.data; |
|
499 |
+ } |
|
500 |
+ scanner->yytext = scanner->buf.data; |
|
501 |
+ scanner->yylen = scanner->buf.pos - 1; |
|
502 |
+ memset(&scanner->buf, 0, sizeof(scanner->buf)); |
|
503 |
+ return str; |
|
504 |
+} |
|
505 |
+ |
|
506 |
+static inline int textbuffer_ensure_capacity(struct text_buffer *txtbuf, size_t len) |
|
507 |
+{ |
|
508 |
+ if (txtbuf->pos + len > txtbuf->capacity) { |
|
509 |
+ char *d; |
|
510 |
+ txtbuf->capacity = MAX(txtbuf->pos + len, txtbuf->capacity + 4096); |
|
511 |
+ d = cli_realloc(txtbuf->data, txtbuf->capacity); |
|
512 |
+ if(!d) |
|
513 |
+ return -1; |
|
514 |
+ txtbuf->data = d; |
|
515 |
+ } |
|
516 |
+ return 0; |
|
517 |
+} |
|
518 |
+ |
|
519 |
+static inline void textbuffer_append_len(struct text_buffer *txtbuf, const char *s, size_t len) |
|
520 |
+{ |
|
521 |
+ textbuffer_ensure_capacity(txtbuf, len); |
|
522 |
+ memcpy(&txtbuf->data[txtbuf->pos], s, len); |
|
523 |
+ txtbuf->pos += len; |
|
524 |
+} |
|
525 |
+ |
|
526 |
+ |
|
527 |
+static inline void textbuffer_append(struct text_buffer *txtbuf, const char *s) |
|
528 |
+{ |
|
529 |
+ size_t len = strlen(s); |
|
530 |
+ textbuffer_append_len(txtbuf, s, len); |
|
531 |
+} |
|
532 |
+ |
|
533 |
+static inline void textbuffer_putc(struct text_buffer *txtbuf, const char c) |
|
534 |
+{ |
|
535 |
+ textbuffer_ensure_capacity(txtbuf, 1); |
|
536 |
+ txtbuf->data[txtbuf->pos++] = c; |
|
537 |
+} |
|
538 |
+#define MODULE "JS-Norm: " |
|
539 |
+ |
|
540 |
+static void free_token(yystype *token) |
|
541 |
+{ |
|
542 |
+ if(token->vtype == vtype_string) { |
|
543 |
+ free(token->val.string); |
|
544 |
+ token->val.string = NULL; |
|
545 |
+ } |
|
546 |
+} |
|
547 |
+ |
|
548 |
+static int replace_token_range(struct tokens *dst, size_t start, size_t end, const struct tokens *with) |
|
549 |
+{ |
|
550 |
+ const size_t len = with ? with->cnt : 0; |
|
551 |
+ size_t i; |
|
552 |
+ cli_dbgmsg(MODULE "Replacing tokens %lu - %lu with %lu tokens\n",start, end, len); |
|
553 |
+ if(start >= dst->cnt || end > dst->cnt) |
|
554 |
+ return -1; |
|
555 |
+ for(i=start;i<end;i++) { |
|
556 |
+ free_token(&dst->data[i]); |
|
557 |
+ } |
|
558 |
+ if(tokens_ensure_capacity(dst, dst->cnt - (end-start) + len) < 0) |
|
559 |
+ return CL_EMEM; |
|
560 |
+ memmove(&dst->data[start+len], &dst->data[end], (dst->cnt - end) * sizeof(dst->data[0])); |
|
561 |
+ if(with && len > 0) { |
|
562 |
+ memcpy(&dst->data[start], with->data, len * sizeof(dst->data[0])); |
|
563 |
+ } |
|
564 |
+ dst->cnt = dst->cnt - (end-start) + len; |
|
565 |
+ return CL_SUCCESS; |
|
566 |
+} |
|
567 |
+ |
|
568 |
+static int append_tokens(struct tokens *dst, const struct tokens *src) |
|
569 |
+{ |
|
570 |
+ if(!dst || !src) |
|
571 |
+ return CL_ENULLARG; |
|
572 |
+ if(!dst->cnt) |
|
573 |
+ return CL_SUCCESS; |
|
574 |
+ if(tokens_ensure_capacity(dst, dst->cnt + src->cnt) == -1) |
|
575 |
+ return CL_EMEM; |
|
576 |
+ cli_dbgmsg(MODULE "Appending %lu tokens\n", src->cnt); |
|
577 |
+ memcpy(&dst->data[dst->cnt], src->data, src->cnt * sizeof(dst->data[0])); |
|
578 |
+ dst->cnt += src->cnt; |
|
579 |
+ return CL_SUCCESS; |
|
580 |
+} |
|
581 |
+ |
|
582 |
+static void decode_de(yystype *params[], struct text_buffer *txtbuf) |
|
583 |
+{ |
|
584 |
+ const char *p = TOKEN_GET(params[0], cstring); |
|
585 |
+ const long a = TOKEN_GET(params[1], ival); |
|
586 |
+ /*const char *c = params[2];*/ |
|
587 |
+ char *k = TOKEN_GET(params[3], string); |
|
588 |
+ /*const char *r = params[5];*/ |
|
589 |
+ |
|
590 |
+ unsigned val=0; |
|
591 |
+ unsigned nsplit = 0; |
|
592 |
+ const char* o; |
|
593 |
+ const char **tokens; |
|
594 |
+ |
|
595 |
+ memset(txtbuf, 0, sizeof(*txtbuf)); |
|
596 |
+ if(!p || !k ) |
|
597 |
+ return; |
|
598 |
+ for(o = k; *o; o++) if(*o == '|') nsplit++; |
|
599 |
+ nsplit++; |
|
600 |
+ tokens = malloc(sizeof(char*)*nsplit); |
|
601 |
+ if(!tokens) { |
|
602 |
+ return; |
|
603 |
+ } |
|
604 |
+ cli_strtokenize(k,'|',nsplit, tokens); |
|
605 |
+ |
|
606 |
+ do { |
|
607 |
+ while(*p && !isalnum(*p)) { |
|
608 |
+ if(*p=='\\' && (p[1] == '\'' || p[1] == '\"')) |
|
609 |
+ p++; |
|
610 |
+ else |
|
611 |
+ textbuffer_putc(txtbuf, *p++); |
|
612 |
+ } |
|
613 |
+ if(!*p) break; |
|
614 |
+ val = 0; |
|
615 |
+ o = p; |
|
616 |
+ while(*p && isalnum(*p)) { |
|
617 |
+ unsigned x; |
|
618 |
+ unsigned char v = *p++; |
|
619 |
+ /* TODO: use a table here */ |
|
620 |
+ if(v >= 'a') x = 10+v-'a'; |
|
621 |
+ else if(v >= 'A') x = 36+v-'A'; |
|
622 |
+ else x = v-'0'; |
|
623 |
+ val = val*a+x; |
|
624 |
+ } |
|
625 |
+ if(val >= nsplit || !tokens[val] || !tokens[val][0]) |
|
626 |
+ while(o!=p) |
|
627 |
+ textbuffer_putc(txtbuf, *o++); |
|
628 |
+ else textbuffer_append(txtbuf, tokens[val]); |
|
629 |
+ } while (*p); |
|
630 |
+ free(tokens); |
|
631 |
+ textbuffer_append(txtbuf, "\0"); |
|
632 |
+} |
|
633 |
+ |
|
634 |
+struct decode_result { |
|
635 |
+ struct text_buffer txtbuf; |
|
636 |
+ size_t pos_begin; |
|
637 |
+ size_t pos_end; |
|
638 |
+ unsigned append:1; /* 0: tokens are replaced with new token(s), |
|
639 |
+ 1: old tokens are deleted, new ones appended at the end */ |
|
640 |
+}; |
|
641 |
+ |
|
642 |
+static void handle_de(yystype *tokens, size_t start, const size_t cnt, const char *name, struct decode_result *res) |
|
643 |
+{ |
|
644 |
+ /* find function decl. end */ |
|
645 |
+ size_t i, nesting = 1, j; |
|
646 |
+ yystype* parameters [6]; |
|
647 |
+ const size_t parameters_cnt = 6; |
|
648 |
+ |
|
649 |
+ for(i=start;i < cnt; i++) { |
|
650 |
+ if(tokens[i].type == TOK_FUNCTION) { |
|
651 |
+ if(TOKEN_GET(&tokens[i], scope)) |
|
652 |
+ nesting++; |
|
653 |
+ else |
|
654 |
+ nesting--; |
|
655 |
+ if(!nesting) |
|
656 |
+ break; |
|
657 |
+ } |
|
658 |
+ } |
|
659 |
+ if(nesting) |
|
660 |
+ return; |
|
661 |
+ if(name) { |
|
662 |
+ /* find call to function */ |
|
663 |
+ for(;i+2 < cnt; i++) { |
|
664 |
+ const char* token_val = TOKEN_GET(&tokens[i], cstring); |
|
665 |
+ if(tokens[i].type == TOK_IDENTIFIER_NAME && |
|
666 |
+ token_val && |
|
667 |
+ !strcmp(name, token_val) && |
|
668 |
+ tokens[i+1].type == TOK_PAR_OPEN) { |
|
669 |
+ |
|
670 |
+ i += 2; |
|
671 |
+ for(j = 0;j < parameters_cnt && i < cnt;j++) { |
|
672 |
+ parameters[j] = &tokens[i++]; |
|
673 |
+ if(j != parameters_cnt-1) |
|
674 |
+ while (tokens[i].type != TOK_COMMA && i < cnt) i++; |
|
675 |
+ else |
|
676 |
+ while (tokens[i].type != TOK_PAR_CLOSE && i < cnt) i++; |
|
677 |
+ i++; |
|
678 |
+ } |
|
679 |
+ if(j == parameters_cnt) |
|
680 |
+ decode_de(parameters, &res->txtbuf); |
|
681 |
+ } |
|
682 |
+ } |
|
683 |
+ } else { |
|
684 |
+ while(i<cnt && tokens[i].type != TOK_PAR_OPEN) i++; |
|
685 |
+ ++i; |
|
686 |
+ if(i >= cnt) return; |
|
687 |
+ /* TODO: move this v to another func */ |
|
688 |
+ for(j = 0;j < parameters_cnt && i < cnt;j++) { |
|
689 |
+ parameters[j] = &tokens[i++]; |
|
690 |
+ if(j != parameters_cnt-1) |
|
691 |
+ while (tokens[i].type != TOK_COMMA && i < cnt) i++; |
|
692 |
+ else |
|
693 |
+ while (tokens[i].type != TOK_PAR_CLOSE && i < cnt) i++; |
|
694 |
+ i++; |
|
695 |
+ } |
|
696 |
+ if(j == parameters_cnt) |
|
697 |
+ decode_de(parameters, &res->txtbuf); |
|
698 |
+ } |
|
699 |
+ res->pos_begin = parameters[0] - tokens; |
|
700 |
+ res->pos_end = parameters[parameters_cnt-1] - tokens + 1; |
|
701 |
+ if(tokens[res->pos_end].type == TOK_BRACKET_OPEN && |
|
702 |
+ tokens[res->pos_end+1].type == TOK_BRACKET_CLOSE && |
|
703 |
+ tokens[res->pos_end+2].type == TOK_PAR_CLOSE) |
|
704 |
+ res->pos_end += 3; /* {}) */ |
|
705 |
+ else |
|
706 |
+ res->pos_end++; /* ) */ |
|
707 |
+} |
|
708 |
+ |
|
709 |
+/* --------- this should be in str.c -------------------------------- */ |
|
710 |
+static const int hex_chars[256] = { |
|
711 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
712 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
713 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
714 |
+ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9,-1,-1, -1,-1,-1,-1, |
|
715 |
+ -1,10,11,12, 13,14,15,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
716 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
717 |
+ -1,10,11,12, 13,14,15,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
718 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
719 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
720 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
721 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
722 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
723 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
724 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
725 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
726 |
+ -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, |
|
727 |
+}; |
|
728 |
+ |
|
729 |
+static inline int cli_hex2int(const char c) |
|
730 |
+{ |
|
731 |
+ return hex_chars[(const unsigned char)c]; |
|
732 |
+} |
|
733 |
+ |
|
734 |
+static inline size_t output_utf8(uint16_t u, unsigned char* dst) |
|
735 |
+{ |
|
736 |
+ if(!u) { |
|
737 |
+ *dst = 0x1; /* don't add \0, add \1 instead */ |
|
738 |
+ return 1; |
|
739 |
+ } |
|
740 |
+ if(u < 0x80) { |
|
741 |
+ *dst = u&0xff; |
|
742 |
+ return 1; |
|
743 |
+ } |
|
744 |
+ if(u < 0x800) { |
|
745 |
+ *dst++ = 0xc0 | (u>>6); /* 110yyyyy */ |
|
746 |
+ *dst = 0x80 | (u & 0x3f); /* 10zzzzzz */ |
|
747 |
+ return 2; |
|
748 |
+ } |
|
749 |
+ /* u < 0x10000 because we only handle utf-16, |
|
750 |
+ * values in range 0xd800 - 0xdfff aren't valid, but we don't check for |
|
751 |
+ * that*/ |
|
752 |
+ *dst++ = 0xe0 | (u>>12); /* 1110xxxx */ |
|
753 |
+ *dst++ = 0x80 | ((u>>6)&0x3f); /* 10yyyyyy */ |
|
754 |
+ *dst = 0x80 | (u & 0x3f); /* 10zzzzzz */ |
|
755 |
+ return 3; |
|
756 |
+} |
|
757 |
+ |
|
758 |
+static void textbuffer_append_normalize(struct text_buffer *buf, const char *str, size_t len) |
|
759 |
+{ |
|
760 |
+ size_t i; |
|
761 |
+ for(i=0;i < len;i++) { |
|
762 |
+ char c = str[i]; |
|
763 |
+ if (c == '\\' && i+1 < len) { |
|
764 |
+ i++; |
|
765 |
+ switch (str[i]) { |
|
766 |
+ case '0': |
|
767 |
+ c = 0; |
|
768 |
+ break; |
|
769 |
+ case 'b': |
|
770 |
+ c = 8; |
|
771 |
+ break; |
|
772 |
+ case 't': |
|
773 |
+ c = 9; |
|
774 |
+ break; |
|
775 |
+ case 'n': |
|
776 |
+ c = 10; |
|
777 |
+ break; |
|
778 |
+ case 'v': |
|
779 |
+ c = 11; |
|
780 |
+ break; |
|
781 |
+ case 'f': |
|
782 |
+ c = 12; |
|
783 |
+ break; |
|
784 |
+ case 'r': |
|
785 |
+ c=13; |
|
786 |
+ break; |
|
787 |
+ case 'x': |
|
788 |
+ if(i+2 < len) |
|
789 |
+ c = (cli_hex2int(str[i+1])<<4)|cli_hex2int(str[i+2]); |
|
790 |
+ i += 2; |
|
791 |
+ break; |
|
792 |
+ case 'u': |
|
793 |
+ if(i+4 < len) { |
|
794 |
+ uint16_t u = (cli_hex2int(str[i+1])<<12) | (cli_hex2int(str[i+2])<<8) | |
|
795 |
+ (cli_hex2int(str[i+3])<<4) | cli_hex2int(str[i+4]); |
|
796 |
+ textbuffer_ensure_capacity(buf, 4); |
|
797 |
+ buf->pos += output_utf8(u, (unsigned char*)buf->data); |
|
798 |
+ i += 4; |
|
799 |
+ continue; |
|
800 |
+ } |
|
801 |
+ break; |
|
802 |
+ default: |
|
803 |
+ c = str[i]; |
|
804 |
+ break; |
|
805 |
+ } |
|
806 |
+ } |
|
807 |
+ if(!c) c = 1; /* we don't insert \0 */ |
|
808 |
+ textbuffer_putc(buf, c); |
|
809 |
+ } |
|
810 |
+} |
|
811 |
+ |
|
812 |
+ |
|
813 |
+static char *cli_unescape(const char *str) |
|
814 |
+{ |
|
815 |
+ char *R; |
|
816 |
+ size_t k, i=0; |
|
817 |
+ const size_t len = strlen(str); |
|
818 |
+ /* unescaped string is at most as long as original, |
|
819 |
+ * it will usually be shorter */ |
|
820 |
+ R = cli_malloc(len + 1); |
|
821 |
+ for(k=0;k < len;k++) { |
|
822 |
+ unsigned char c = str[k]; |
|
823 |
+ if (str[k] == '%') { |
|
824 |
+ if(k+5 >= len || str[k+1] != 'u' || !isxdigit(str[k+2]) || !isxdigit(str[k+3]) |
|
825 |
+ || !isxdigit(str[k+4]) || !isxdigit(str[k+5])) { |
|
826 |
+ if(k+2 < len && isxdigit(str[k+1]) && isxdigit(str[k+2])) { |
|
827 |
+ c = (cli_hex2int(str[k+1])<<4) | cli_hex2int(str[k+2]); |
|
828 |
+ k += 2; |
|
829 |
+ } |
|
830 |
+ } else { |
|
831 |
+ uint16_t u = (cli_hex2int(str[k+2])<<12) | (cli_hex2int(str[k+3])<<8) | |
|
832 |
+ (cli_hex2int(str[k+4])<<4) | cli_hex2int(str[k+5]); |
|
833 |
+ i += output_utf8(u, (unsigned char*)&R[i]); |
|
834 |
+ k += 5; |
|
835 |
+ continue; |
|
836 |
+ } |
|
837 |
+ } |
|
838 |
+ if(!c) c = 1; /* don't add \0 */ |
|
839 |
+ /* TODO: if c >= 0x80 output UTF-8, and do the same in |
|
840 |
+ * normalize_string, and interpret the full %u sequence ! */ |
|
841 |
+ R[i++] = c; |
|
842 |
+ } |
|
843 |
+ R[i++] = '\0'; |
|
844 |
+ R = cli_realloc(R, i); |
|
845 |
+ return R; |
|
846 |
+} |
|
847 |
+ |
|
848 |
+/* ------------ end of str.c ----------------- */ |
|
849 |
+ |
|
850 |
+static int handle_unescape(struct tokens *tokens, size_t start, const size_t cnt) |
|
851 |
+{ |
|
852 |
+ if(tokens->data[start].type == TOK_StringLiteral) { |
|
853 |
+ char *R; |
|
854 |
+ struct tokens new_tokens; |
|
855 |
+ yystype tok; |
|
856 |
+ |
|
857 |
+ R = cli_unescape(TOKEN_GET(&tokens->data[start], cstring)); |
|
858 |
+ tok.type = TOK_StringLiteral; |
|
859 |
+ TOKEN_SET(&tok, string, R); |
|
860 |
+ new_tokens.capacity = new_tokens.cnt = 1; |
|
861 |
+ new_tokens.data = &tok; |
|
862 |
+ if(replace_token_range(tokens, start-2, start+2, &new_tokens) < 0) |
|
863 |
+ return CL_EMEM; |
|
864 |
+ } |
|
865 |
+ return CL_SUCCESS; |
|
866 |
+} |
|
867 |
+ |
|
868 |
+ |
|
869 |
+/* scriptasylum dot com's JS encoder */ |
|
870 |
+static void handle_df(const yystype *tokens, size_t start, const size_t cnt, struct decode_result *res) |
|
871 |
+{ |
|
872 |
+ char *str, *s1; |
|
873 |
+ size_t len, s1_len, i; |
|
874 |
+ unsigned char clast; |
|
875 |
+ char *R; |
|
876 |
+ |
|
877 |
+ if(tokens[start].type != TOK_StringLiteral) |
|
878 |
+ return; |
|
879 |
+ str = TOKEN_GET(&tokens[start], string); |
|
880 |
+ if(!str) |
|
881 |
+ return; |
|
882 |
+ len = strlen(str); |
|
883 |
+ clast = str[len-1] - '0'; |
|
884 |
+ |
|
885 |
+ str[len-1] = '\0'; |
|
886 |
+ s1 = cli_unescape(str); |
|
887 |
+ s1_len = strlen(s1); |
|
888 |
+ for(i=0;i<s1_len;i++) { |
|
889 |
+ s1[i] -= clast; |
|
890 |
+ } |
|
891 |
+ R = cli_unescape(s1); |
|
892 |
+ free(s1); |
|
893 |
+ res->pos_begin = start-2; |
|
894 |
+ res->pos_end = start+2; |
|
895 |
+ res->txtbuf.data = R; |
|
896 |
+ res->txtbuf.pos = strlen(R); |
|
897 |
+ res->append = 1; |
|
898 |
+} |
|
899 |
+ |
|
900 |
+ |
|
901 |
+ |
|
902 |
+static void handle_eval(struct tokens *tokens, size_t start, struct decode_result *res) |
|
903 |
+{ |
|
904 |
+ res->txtbuf.data = TOKEN_GET(&tokens->data[start], string); |
|
905 |
+ if(res->txtbuf.data && tokens->data[start+1].type == TOK_PAR_CLOSE) { |
|
906 |
+ TOKEN_SET(&tokens->data[start], string, NULL); |
|
907 |
+ res->txtbuf.pos = strlen(res->txtbuf.data); |
|
908 |
+ res->pos_begin = start-2; |
|
909 |
+ res->pos_end = start+2; |
|
910 |
+ } |
|
911 |
+} |
|
912 |
+ |
|
913 |
+static void run_folders(struct tokens *tokens) |
|
914 |
+{ |
|
915 |
+ size_t i; |
|
916 |
+ |
|
917 |
+ for(i = 0; i < tokens->cnt; i++) { |
|
918 |
+ const char *cstring = TOKEN_GET(&tokens->data[i], cstring); |
|
919 |
+ if(i+2 < tokens->cnt && tokens->data[i].type == TOK_IDENTIFIER_NAME && |
|
920 |
+ cstring && |
|
921 |
+ !strcmp("unescape", cstring) && tokens->data[i+1].type == TOK_PAR_OPEN) { |
|
922 |
+ |
|
923 |
+ handle_unescape(tokens, i+2, tokens->cnt); |
|
924 |
+ } |
|
925 |
+ } |
|
926 |
+} |
|
927 |
+ |
|
928 |
+static inline int state_update_scope(struct parser_state *state, const yystype *token) |
|
929 |
+{ |
|
930 |
+ if(token->type == TOK_FUNCTION) { |
|
931 |
+ struct scope *scope = TOKEN_GET(token, scope); |
|
932 |
+ if(scope) { |
|
933 |
+ state->current = scope; |
|
934 |
+ } |
|
935 |
+ else { |
|
936 |
+ /* dummy token marking function end */ |
|
937 |
+ if(state->current->parent) |
|
938 |
+ state->current = state->current->parent; |
|
939 |
+ /* don't output this token, it is just a dummy marker */ |
|
940 |
+ return 0; |
|
941 |
+ } |
|
942 |
+ } |
|
943 |
+ return 1; |
|
944 |
+} |
|
945 |
+ |
|
946 |
+static void run_decoders(struct parser_state *state) |
|
947 |
+{ |
|
948 |
+ size_t i; |
|
949 |
+ const char* name; |
|
950 |
+ struct tokens *tokens = &state->tokens; |
|
951 |
+ |
|
952 |
+ for(i = 0; i < tokens->cnt; i++) { |
|
953 |
+ const char *cstring = TOKEN_GET(&tokens->data[i], cstring); |
|
954 |
+ struct decode_result res; |
|
955 |
+ res.pos_begin = res.pos_end = 0; |
|
956 |
+ res.append = 0; |
|
957 |
+ if(tokens->data[i].type == TOK_FUNCTION && i+13 < tokens->cnt) { |
|
958 |
+ name = NULL; |
|
959 |
+ ++i; |
|
960 |
+ if(tokens->data[i].type == TOK_IDENTIFIER_NAME) { |
|
961 |
+ name = cstring; |
|
962 |
+ ++i; |
|
963 |
+ } |
|
964 |
+ if(match_parameters(&tokens->data[i], de_packer_3, sizeof(de_packer_3)/sizeof(de_packer_3[0])) != -1 |
|
965 |
+ || match_parameters(&tokens->data[i], de_packer_2, sizeof(de_packer_2)/sizeof(de_packer_2[0])) != -1) { |
|
966 |
+ /* find function decl. end */ |
|
967 |
+ handle_de(tokens->data, i, tokens->cnt, name, &res); |
|
968 |
+ } |
|
969 |
+ } else if(i+2 < tokens->cnt && tokens->data[i].type == TOK_IDENTIFIER_NAME && |
|
970 |
+ cstring && |
|
971 |
+ !strcmp("dF", cstring) && tokens->data[i+1].type == TOK_PAR_OPEN) { |
|
972 |
+ /* TODO: also match signature of dF function (possibly |
|
973 |
+ * declared using unescape */ |
|
974 |
+ |
|
975 |
+ handle_df(tokens->data, i+2, tokens->cnt, &res); |
|
976 |
+ } else if(i+2 < tokens->cnt && tokens->data[i].type == TOK_IDENTIFIER_NAME && |
|
977 |
+ cstring && |
|
978 |
+ !strcmp("eval", cstring) && tokens->data[i+1].type == TOK_PAR_OPEN) { |
|
979 |
+ handle_eval(tokens, i+2, &res); |
|
980 |
+ } |
|
981 |
+ if(res.pos_end > res.pos_begin) { |
|
982 |
+ struct tokens parent_tokens; |
|
983 |
+ if(res.pos_end < tokens->cnt && tokens->data[res.pos_end].type == TOK_SEMICOLON) |
|
984 |
+ res.pos_end++; |
|
985 |
+ parent_tokens = state->tokens;/* save current tokens */ |
|
986 |
+ /* initialize embedded context */ |
|
987 |
+ memset(&state->tokens, 0, sizeof(state->tokens)); |
|
988 |
+ cli_js_process_buffer(state, res.txtbuf.data, res.txtbuf.pos); |
|
989 |
+ free(res.txtbuf.data); |
|
990 |
+ /* state->tokens still refers to the embedded/nested context |
|
991 |
+ * here */ |
|
992 |
+ if(!res.append) { |
|
993 |
+ replace_token_range(&parent_tokens, res.pos_begin, res.pos_end, &state->tokens); |
|
994 |
+ } else { |
|
995 |
+ /* delete tokens */ |
|
996 |
+ replace_token_range(&parent_tokens, res.pos_begin, res.pos_end, NULL); |
|
997 |
+ append_tokens(&parent_tokens, &state->tokens); |
|
998 |
+ } |
|
999 |
+ /* end of embedded context, restore tokens state */ |
|
1000 |
+ free(state->tokens.data); |
|
1001 |
+ state->tokens = parent_tokens; |
|
1002 |
+ } |
|
1003 |
+ state_update_scope(state, &state->tokens.data[i]); |
|
1004 |
+ } |
|
1005 |
+} |
|
1006 |
+ |
|
1007 |
+void cli_js_parse_done(struct parser_state* state) |
|
1008 |
+{ |
|
1009 |
+ run_folders(&state->tokens); |
|
1010 |
+ run_decoders(state); |
|
1011 |
+ |
|
1012 |
+ yylex_destroy(state->scanner); |
|
1013 |
+ state->global = NULL; /* make this state invalid for parsing */ |
|
1014 |
+} |
|
1015 |
+ |
|
1016 |
+ |
|
1017 |
+void cli_js_output(struct parser_state *state) |
|
1018 |
+{ |
|
1019 |
+ unsigned i; |
|
1020 |
+ struct buf buf; |
|
1021 |
+ char lastchar = '\0'; |
|
1022 |
+ buf.pos = 0; |
|
1023 |
+ buf.outfd = STDOUT_FILENO; |
|
1024 |
+ state->current = state->global; |
|
1025 |
+ for(i = 0; i < state->tokens.cnt; i++) { |
|
1026 |
+ if(state_update_scope(state, &state->tokens.data[i])) |
|
1027 |
+ lastchar = output_token(&state->tokens.data[i], state->current, &buf, lastchar); |
|
1028 |
+ } |
|
1029 |
+ if(write(buf.outfd, buf.buf, buf.pos) < 0) { |
|
1030 |
+ cli_dbgmsg(MODULE "I/O error"); |
|
1031 |
+ } |
|
1032 |
+} |
|
1033 |
+ |
|
1034 |
+void cli_js_destroy(struct parser_state *state) |
|
1035 |
+{ |
|
1036 |
+ size_t i; |
|
1037 |
+ scope_free_all(state->list); |
|
1038 |
+ for(i=0;i<state->tokens.cnt;i++) { |
|
1039 |
+ free_token(&state->tokens.data[i]); |
|
1040 |
+ } |
|
1041 |
+ free(state->tokens.data); |
|
1042 |
+} |
|
1043 |
+ |
|
1044 |
+/* buffer is html-normlike "chunk", if original file is bigger than buffer, |
|
1045 |
+ * we rewind to a space, so we'll know that tokens won't be broken in half at |
|
1046 |
+ * the end of a buffer. All tokens except string-literals of course. |
|
1047 |
+ * So we can assume that after the buffer there is either a space, EOF, or a |
|
1048 |
+ * chunk of text not containing whitespace at all (for which we care only if its |
|
1049 |
+ * a stringliteral)*/ |
|
1050 |
+void cli_js_process_buffer(struct parser_state *state, const char *buf, size_t n) |
|
1051 |
+{ |
|
1052 |
+ struct scope* current = state->current; |
|
1053 |
+ YYSTYPE val; |
|
1054 |
+ int yv; |
|
1055 |
+ YY_BUFFER_STATE yyb; |
|
1056 |
+ |
|
1057 |
+ if(!state->global) { |
|
1058 |
+ /* this state has either not been initialized, |
|
1059 |
+ * or cli_js_parse_done() was already called on it */ |
|
1060 |
+ cli_warnmsg(MODULE "invalid state"); |
|
1061 |
+ return; |
|
1062 |
+ } |
|
1063 |
+ yyb = yy_scan_bytes(buf, n, state->scanner); |
|
1064 |
+ memset(&val, 0, sizeof(val)); |
|
1065 |
+ val.vtype = vtype_undefined; |
|
1066 |
+ /* on EOF yylex will return 0 */ |
|
1067 |
+ while( (yv=yylex(&val, state->scanner)) != 0) |
|
1068 |
+ { |
|
1069 |
+ const char *text; |
|
1070 |
+ size_t leng; |
|
1071 |
+ |
|
1072 |
+ val.type = yv; |
|
1073 |
+ switch(yv) { |
|
1074 |
+ case TOK_VAR: |
|
1075 |
+ current->fsm_state = InsideVar; |
|
1076 |
+ break; |
|
1077 |
+ case TOK_IDENTIFIER_NAME: |
|
1078 |
+ text = yyget_text(state->scanner); |
|
1079 |
+ leng = yyget_leng(state->scanner); |
|
1080 |
+ if(current->last_token == TOK_DOT) { |
|
1081 |
+ /* this is a member name, don't normalize |
|
1082 |
+ */ |
|
1083 |
+ TOKEN_SET(&val, string, cli_strdup(text)); |
|
1084 |
+ val.type = TOK_UNNORM_IDENTIFIER; |
|
1085 |
+ } else { |
|
1086 |
+ switch(current->fsm_state) { |
|
1087 |
+ case WaitParameterList: |
|
1088 |
+ state->syntax_errors++; |
|
1089 |
+ /* fall through */ |
|
1090 |
+ case Base: |
|
1091 |
+ case InsideInitializer: |
|
1092 |
+ TOKEN_SET(&val, cstring, scope_use(current, text, leng)); |
|
1093 |
+ break; |
|
1094 |
+ case InsideVar: |
|
1095 |
+ case InsideFunctionDecl: |
|
1096 |
+ TOKEN_SET(&val, cstring, scope_declare(current, text, leng, state)); |
|
1097 |
+ current->fsm_state = InsideInitializer; |
|
1098 |
+ current->brackets = 0; |
|
1099 |
+ break; |
|
1100 |
+ case WaitFunctionName: |
|
1101 |
+ TOKEN_SET(&val, cstring, scope_declare(current, text, leng, state)); |
|
1102 |
+ current->fsm_state = WaitParameterList; |
|
1103 |
+ break; |
|
1104 |
+ } |
|
1105 |
+ } |
|
1106 |
+ break; |
|
1107 |
+ case TOK_PAR_OPEN: |
|
1108 |
+ switch(current->fsm_state) { |
|
1109 |
+ case WaitFunctionName: |
|
1110 |
+ /* TODO: function name is null */ |
|
1111 |
+ /* fallthrough */ |
|
1112 |
+ case WaitParameterList: |
|
1113 |
+ current->fsm_state = InsideFunctionDecl; |
|
1114 |
+ break; |
|
1115 |
+ default: |
|
1116 |
+ /* noop */ |
|
1117 |
+ break; |
|
1118 |
+ } |
|
1119 |
+ break; |
|
1120 |
+ case TOK_PAR_CLOSE: |
|
1121 |
+ switch(current->fsm_state) { |
|
1122 |
+ case WaitFunctionName: |
|
1123 |
+ state->syntax_errors++; |
|
1124 |
+ break; |
|
1125 |
+ case WaitParameterList: |
|
1126 |
+ current->fsm_state = Base; |
|
1127 |
+ break; |
|
1128 |
+ default: |
|
1129 |
+ /* noop */ |
|
1130 |
+ break; |
|
1131 |
+ } |
|
1132 |
+ break; |
|
1133 |
+ case TOK_CURLY_BRACE_OPEN: |
|
1134 |
+ switch(current->fsm_state) { |
|
1135 |
+ case WaitFunctionName: |
|
1136 |
+ /* TODO: function name is null */ |
|
1137 |
+ /* fallthrough */ |
|
1138 |
+ case WaitParameterList: |
|
1139 |
+ case InsideFunctionDecl: |
|
1140 |
+ /* in a syntactically correct |
|
1141 |
+ * file, we would already be in |
|
1142 |
+ * the Base state when we see a { |
|
1143 |
+ */ |
|
1144 |
+ current->fsm_state = Base; |
|
1145 |
+ /* fall-through */ |
|
1146 |
+ case InsideVar: |
|
1147 |
+ case InsideInitializer: |
|
1148 |
+ state->syntax_errors++; |
|
1149 |
+ /* fall-through */ |
|
1150 |
+ case Base: |
|
1151 |
+ default: |
|
1152 |
+ current->blocks++; |
|
1153 |
+ break; |
|
1154 |
+ } |
|
1155 |
+ break; |
|
1156 |
+ case TOK_CURLY_BRACE_CLOSE: |
|
1157 |
+ if(current->blocks > 0) |
|
1158 |
+ current->blocks--; |
|
1159 |
+ else |
|
1160 |
+ state->syntax_errors++; |
|
1161 |
+ if(!current->blocks) { |
|
1162 |
+ if(current->parent) { |
|
1163 |
+ /* add dummy FUNCTION token to |
|
1164 |
+ * mark function end */ |
|
1165 |
+ TOKEN_SET(&val, cstring, "}"); |
|
1166 |
+ add_token(state, &val); |
|
1167 |
+ TOKEN_SET(&val, scope, NULL); |
|
1168 |
+ val.type = TOK_FUNCTION; |
|
1169 |
+ |
|
1170 |
+ state->current = current = current->parent; |
|
1171 |
+ } else{ |
|
1172 |
+ /* extra } */ |
|
1173 |
+ state->syntax_errors++; |
|
1174 |
+ } |
|
1175 |
+ } |
|
1176 |
+ break; |
|
1177 |
+ case TOK_BRACKET_OPEN: |
|
1178 |
+ current->brackets++; |
|
1179 |
+ break; |
|
1180 |
+ case TOK_BRACKET_CLOSE: |
|
1181 |
+ if(current->brackets > 0) |
|
1182 |
+ current->brackets--; |
|
1183 |
+ else |
|
1184 |
+ state->syntax_errors++; |
|
1185 |
+ break; |
|
1186 |
+ case TOK_COMMA: |
|
1187 |
+ if (current->fsm_state == InsideInitializer && current->brackets == 0 && current->blocks == 0) { |
|
1188 |
+ /* initializer ended only if we |
|
1189 |
+ * encountered a comma, and [] are |
|
1190 |
+ * balanced. |
|
1191 |
+ * This avoids switching state on: |
|
1192 |
+ * var x = [4,y,u];*/ |
|
1193 |
+ current->fsm_state = InsideVar; |
|
1194 |
+ } |
|
1195 |
+ break; |
|
1196 |
+ case TOK_SEMICOLON: |
|
1197 |
+ if (current->brackets == 0 && current->blocks == 0) { |
|
1198 |
+ /* avoid switching state on unbalanced []: |
|
1199 |
+ * var x = [test;testi]; */ |
|
1200 |
+ current->fsm_state = Base; |
|
1201 |
+ } |
|
1202 |
+ break; |
|
1203 |
+ case TOK_FUNCTION: |
|
1204 |
+ current = scope_new(state); |
|
1205 |
+ current->fsm_state = WaitFunctionName; |
|
1206 |
+ TOKEN_SET(&val, scope, state->current); |
|
1207 |
+ break; |
|
1208 |
+ case TOK_StringLiteral: |
|
1209 |
+ if(state->tokens.data[state->tokens.cnt-1].type == TOK_PLUS) { |
|
1210 |
+ /* see if can fold */ |
|
1211 |
+ yystype *prev_string = &state->tokens.data[state->tokens.cnt-2]; |
|
1212 |
+ if(prev_string->type == TOK_StringLiteral) { |
|
1213 |
+ char *str = TOKEN_GET(prev_string, string); |
|
1214 |
+ size_t str_len = strlen(str); |
|
1215 |
+ |
|
1216 |
+ text = yyget_text(state->scanner); |
|
1217 |
+ leng = yyget_leng(state->scanner); |
|
1218 |
+ |
|
1219 |
+ |
|
1220 |
+ /* delete TOK_PLUS */ |
|
1221 |
+ free_token(&state->tokens.data[--state->tokens.cnt]); |
|
1222 |
+ |
|
1223 |
+ str = cli_realloc(str, str_len + leng + 1); |
|
1224 |
+ strncpy(str+str_len, text+1, leng); |
|
1225 |
+ str[str_len + leng] = '\0'; |
|
1226 |
+ TOKEN_SET(prev_string, string, str); |
|
1227 |
+ free(val.val.string); |
|
1228 |
+ memset(&val, 0, sizeof(val)); |
|
1229 |
+ val.vtype = vtype_undefined; |
|
1230 |
+ continue; |
|
1231 |
+ } |
|
1232 |
+ } |
|
1233 |
+ break; |
|
1234 |
+ } |
|
1235 |
+ if(val.vtype == vtype_undefined) { |
|
1236 |
+ text = yyget_text(state->scanner); |
|
1237 |
+ /* TODO: tokenizer should set it to point to a constant |
|
1238 |
+ * string, it currently doesn't do that for operators ,;:=... */ |
|
1239 |
+ TOKEN_SET(&val, string, cli_strdup(text)); |
|
1240 |
+ abort(); |
|
1241 |
+ } |
|
1242 |
+ add_token(state, &val); |
|
1243 |
+ current->last_token = yv; |
|
1244 |
+ memset(&val, 0, sizeof(val)); |
|
1245 |
+ val.vtype = vtype_undefined; |
|
1246 |
+ } |
|
1247 |
+ yy_delete_buffer(yyb, state->scanner); |
|
1248 |
+} |
|
1249 |
+ |
|
1250 |
+int cli_js_init(struct parser_state *state) |
|
1251 |
+{ |
|
1252 |
+ if(!state) |
|
1253 |
+ return CL_ENULLARG; |
|
1254 |
+ memset(state, 0, sizeof(*state)); |
|
1255 |
+ if(!scope_new(state)) { |
|
1256 |
+ return CL_EMEM; |
|
1257 |
+ } |
|
1258 |
+ state->global = state->current; |
|
1259 |
+ |
|
1260 |
+ if(yylex_init(&state->scanner)) { |
|
1261 |
+ scope_done(state->global); |
|
1262 |
+ return CL_EMEM; |
|
1263 |
+ } |
|
1264 |
+ yyset_debug(1, state->scanner); |
|
1265 |
+ return CL_SUCCESS; |
|
1266 |
+} |
|
1267 |
+ |
|
1268 |
+int main(int argc,char** argv) |
|
1269 |
+{ |
|
1270 |
+ int n; |
|
1271 |
+ char buf[BUFS+2]; |
|
1272 |
+ struct parser_state state; |
|
1273 |
+ |
|
1274 |
+ /*cli_debug_flag=1;*/ |
|
1275 |
+ printf("Enter javascript:\n"); |
|
1276 |
+ printf(" Terminate with ^D\n"); |
|
1277 |
+ |
|
1278 |
+ cli_js_init(&state); |
|
1279 |
+ while ( ( n=read(fileno(stdin), buf, BUFS )) > 0) |
|
1280 |
+ { |
|
1281 |
+ /*buf[n] = '\0';*/ |
|
1282 |
+ cli_js_process_buffer(&state, buf, n); |
|
1283 |
+ } |
|
1284 |
+ cli_js_parse_done(&state); |
|
1285 |
+ cli_js_output(&state); |
|
1286 |
+ cli_js_destroy(&state); |
|
1287 |
+ return 0; |
|
1288 |
+} |
|
1289 |
+ |
|
1290 |
+/* TODO: special identifiers in global scope (document, ...) |
|
1291 |
+ * |
|
1292 |
+ * avoid extra strdup: |
|
1293 |
+ * - when string is going to be folded |
|
1294 |
+ * - normalize_string, and hashtab_insert - avoid one |
|
1295 |
+ * |
|
1296 |
+ * decoded stuff should be parsed in the correct context (not a global one!) |
|
1297 |
+ * |
|
1298 |
+ * make unescape, packers handling more generic |
|
1299 |
+ * memory leaks - manul check |
|
1300 |
+ * check for allocation failure everywhere |
|
1301 |
+ * limits |
|
1302 |
+ * security check |
|
1303 |
+ * */ |
|
1304 |
+ |
|
1305 |
+ |
|
1306 |
+/*-------------- tokenizer ---------------------*/ |
|
1307 |
+enum char_class { |
|
1308 |
+ Whitespace, |
|
1309 |
+ Slash, |
|
1310 |
+ Operator, |
|
1311 |
+ DQuote, |
|
1312 |
+ SQuote, |
|
1313 |
+ Digit, |
|
1314 |
+ IdStart, |
|
1315 |
+ BracketOpen = TOK_BRACKET_OPEN, |
|
1316 |
+ BracketClose = TOK_BRACKET_CLOSE, |
|
1317 |
+ Comma = TOK_COMMA, |
|
1318 |
+ CurlyOpen = TOK_CURLY_BRACE_OPEN, |
|
1319 |
+ CurlyClose = TOK_CURLY_BRACE_CLOSE, |
|
1320 |
+ ParOpen = TOK_PAR_OPEN, |
|
1321 |
+ ParClose = TOK_PAR_CLOSE, |
|
1322 |
+ Dot = TOK_DOT, |
|
1323 |
+ SemiColon = TOK_SEMICOLON, |
|
1324 |
+ Nop |
|
1325 |
+}; |
|
1326 |
+ |
|
1327 |
+#define SL Slash |
|
1328 |
+#define DG Digit |
|
1329 |
+#define DQ DQuote |
|
1330 |
+#define SQ SQuote |
|
1331 |
+#define ID IdStart |
|
1332 |
+#define OP Operator |
|
1333 |
+#define WS Whitespace |
|
1334 |
+#define BO BracketOpen |
|
1335 |
+#define BC BracketClose |
|
1336 |
+#define CM Comma |
|
1337 |
+#define CO CurlyOpen |
|
1338 |
+#define CC CurlyClose |
|
1339 |
+#define PO ParOpen |
|
1340 |
+#define PC ParClose |
|
1341 |
+#define DT Dot |
|
1342 |
+#define SC SemiColon |
|
1343 |
+#define NA Nop |
|
1344 |
+ |
|
1345 |
+static const enum char_class ctype[256] = { |
|
1346 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, WS, WS, WS, NA, WS, NA, NA, |
|
1347 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1348 |
+ WS, OP, DQ, NA, ID, OP, OP, SQ, PO, PC, OP, OP, CM, OP, DT, SL, |
|
1349 |
+ DG, DG, DG, DG, DG, DG, DG, DG, DG, DG, OP, SC, OP, OP, OP, OP, |
|
1350 |
+ NA, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, |
|
1351 |
+ ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, BO, ID, BC, OP, ID, |
|
1352 |
+ NA, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, |
|
1353 |
+ ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, CO, OP, CC, OP, NA, |
|
1354 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1355 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1356 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1357 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1358 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1359 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1360 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1361 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA |
|
1362 |
+}; |
|
1363 |
+ |
|
1364 |
+static const enum char_class id_ctype[256] = { |
|
1365 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1366 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1367 |
+ NA, NA, NA, NA, ID, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1368 |
+ ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, NA, NA, NA, NA, NA, NA, |
|
1369 |
+ NA, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, |
|
1370 |
+ ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, NA, OP, NA, NA, ID, |
|
1371 |
+ NA, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, |
|
1372 |
+ ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, ID, NA, NA, NA, NA, NA, |
|
1373 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1374 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1375 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1376 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1377 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1378 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1379 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1380 |
+ NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, |
|
1381 |
+}; |
|
1382 |
+ |
|
1383 |
+#define CASE_SPECIAL_CHAR(C, S) case C: TOKEN_SET(lvalp, cstring, (S)); return cClass; |
|
1384 |
+ |
|
1385 |
+#define BUF_KEEP_SIZE 32768 |
|
1386 |
+ |
|
1387 |
+static void textbuf_clean(struct text_buffer *buf) |
|
1388 |
+{ |
|
1389 |
+ if(buf->capacity > BUF_KEEP_SIZE) { |
|
1390 |
+ buf->data = cli_realloc(buf->data, BUF_KEEP_SIZE); |
|
1391 |
+ buf->capacity = BUF_KEEP_SIZE; |
|
1392 |
+ } |
|
1393 |
+ buf->pos = 0; |
|
1394 |
+} |
|
1395 |
+ |
|
1396 |
+ |
|
1397 |
+static inline int parseString(YYSTYPE *lvalp, yyscan_t scanner, const char q, |
|
1398 |
+ enum tokenizer_state tostate) |
|
1399 |
+{ |
|
1400 |
+ size_t len; |
|
1401 |
+ /* look for " terminating the string */ |
|
1402 |
+ const char *start = &scanner->in[scanner->pos], *end = start; |
|
1403 |
+ do { |
|
1404 |
+ const size_t siz = &scanner->in[scanner->insize] - end; |
|
1405 |
+ end = memchr(end, q, siz); |
|
1406 |
+ if(end && end[-1] == '\\') { |
|
1407 |
+ ++end; |
|
1408 |
+ continue; |
|
1409 |
+ } |
|
1410 |
+ break; |
|
1411 |
+ } while (1); |
|
1412 |
+ len = (end && end > start) ? end - start : scanner->insize - scanner->pos; |
|
1413 |
+ textbuffer_append_normalize(&scanner->buf, start, len); |
|
1414 |
+ if(end) { |
|
1415 |
+ /* skip over end quote */ |
|
1416 |
+ scanner->pos += len + 1; |
|
1417 |
+ textbuffer_putc(&scanner->buf, '\0'); |
|
1418 |
+ TOKEN_SET(lvalp, string, textbuffer_done(scanner)); |
|
1419 |
+ scanner->state = Initial; |
|
1420 |
+ assert(lvalp->val.string); |
|
1421 |
+ return TOK_StringLiteral; |
|
1422 |
+ } else { |
|
1423 |
+ scanner->pos += len; |
|
1424 |
+ /* unfinished string */ |
|
1425 |
+ scanner->state = tostate; |
|
1426 |
+ return 0; |
|
1427 |
+ } |
|
1428 |
+} |
|
1429 |
+ |
|
1430 |
+static inline int parseDQString(YYSTYPE *lvalp, yyscan_t scanner) |
|
1431 |
+{ |
|
1432 |
+ return parseString(lvalp, scanner, '"', DoubleQString); |
|
1433 |
+} |
|
1434 |
+ |
|
1435 |
+static inline int parseSQString(YYSTYPE *lvalp, yyscan_t scanner) |
|
1436 |
+{ |
|
1437 |
+ return parseString(lvalp, scanner, '\'', SingleQString); |
|
1438 |
+} |
|
1439 |
+ |
|
1440 |
+static inline int parseNumber(YYSTYPE *lvalp, yyscan_t scanner) |
|
1441 |
+{ |
|
1442 |
+ const unsigned char *in = (const unsigned char*)scanner->in; |
|
1443 |
+ int is_float = 0; |
|
1444 |
+ while(scanner->pos < scanner->insize) { |
|
1445 |
+ unsigned char c = in[scanner->pos++]; |
|
1446 |
+ if(isdigit(c)) { |
|
1447 |
+ textbuffer_putc(&scanner->buf, c); |
|
1448 |
+ continue; |
|
1449 |
+ } |
|
1450 |
+ if(c =='.' && !is_float) { |
|
1451 |
+ is_float = 1; |
|
1452 |
+ textbuffer_putc(&scanner->buf, '.'); |
|
1453 |
+ continue; |
|
1454 |
+ } |
|
1455 |
+ if((c=='e' || c=='E') && is_float) { |
|
1456 |
+ textbuffer_putc(&scanner->buf, c); |
|
1457 |
+ if(scanner->pos < scanner->insize) { |
|
1458 |
+ c = in[scanner->pos++]; |
|
1459 |
+ if(c == '+' || c == '-' || isdigit(c)) { |
|
1460 |
+ textbuffer_putc(&scanner->buf, c); |
|
1461 |
+ continue; |
|
1462 |
+ } |
|
1463 |
+ } |
|
1464 |
+ } |
|
1465 |
+ scanner->pos--; |
|
1466 |
+ textbuffer_putc(&scanner->buf, '\0'); |
|
1467 |
+ scanner->state = Initial; |
|
1468 |
+ if(is_float) { |
|
1469 |
+ TOKEN_SET(lvalp, dval, atof(scanner->buf.data)); |
|
1470 |
+ return TOK_NumericFloat; |
|
1471 |
+ } else { |
|
1472 |
+ TOKEN_SET(lvalp, ival, atoi(scanner->buf.data)); |
|
1473 |
+ return TOK_NumericInt; |
|
1474 |
+ } |
|
1475 |
+ } |
|
1476 |
+ scanner->state = Number; |
|
1477 |
+ return 0; |
|
1478 |
+} |
|
1479 |
+ |
|
1480 |
+static inline int parseId(YYSTYPE *lvalp, yyscan_t scanner) |
|
1481 |
+{ |
|
1482 |
+ const struct keyword *kw; |
|
1483 |
+ const unsigned char *in = (const unsigned char*)scanner->in; |
|
1484 |
+ scanner->state = Initial; |
|
1485 |
+ while(scanner->pos < scanner->insize) { |
|
1486 |
+ unsigned char c = in[scanner->pos++]; |
|
1487 |
+ enum char_class cClass = id_ctype[c]; |
|
1488 |
+ switch(cClass) { |
|
1489 |
+ case IdStart: |
|
1490 |
+ textbuffer_putc(&scanner->buf, c); |
|
1491 |
+ break; |
|
1492 |
+ case Operator: |
|
1493 |
+ /* the table contains OP only for \ */ |
|
1494 |
+ assert(c == '\\'); |
|
1495 |
+ if(scanner->pos < scanner->insize && |
|
1496 |
+ in[scanner->pos++] == 'u') { |
|
1497 |
+ textbuffer_putc(&scanner->buf, c); |
|
1498 |
+ break; |
|
1499 |
+ } |
|
1500 |
+ /* else fallthrough */ |
|
1501 |
+ default: |
|
1502 |
+ /* character is no longer part of identifier */ |
|
1503 |
+ textbuffer_putc(&scanner->buf, '\0'); |
|
1504 |
+ scanner->pos--; |
|
1505 |
+ kw = in_word_set(scanner->buf.data, scanner->buf.pos-1); |
|
1506 |
+ if(kw) { |
|
1507 |
+ /* we got a keyword */ |
|
1508 |
+ TOKEN_SET(lvalp, cstring, kw->name); |
|
1509 |
+ return kw->val; |
|
1510 |
+ } |
|
1511 |
+ /* it is not a keyword, just an identifier */ |
|
1512 |
+ TOKEN_SET(lvalp, cstring, NULL); |
|
1513 |
+ return TOK_IDENTIFIER_NAME; |
|
1514 |
+ } |
|
1515 |
+ } |
|
1516 |
+ scanner->state = Identifier; |
|
1517 |
+ return 0; |
|
1518 |
+} |
|
1519 |
+ |
|
1520 |
+#ifndef MIN |
|
1521 |
+#define MIN(a,b) ((a)<(b) ? (a):(b)) |
|
1522 |
+#endif |
|
1523 |
+ |
|
1524 |
+static int parseOperator(YYSTYPE *lvalp, yyscan_t scanner) |
|
1525 |
+{ |
|
1526 |
+ size_t len = MIN(5, scanner->insize - scanner->pos); |
|
1527 |
+ while(len) { |
|
1528 |
+ const struct operator *kw = in_op_set(&scanner->in[scanner->pos], len); |
|
1529 |
+ if(kw) { |
|
1530 |
+ TOKEN_SET(lvalp, cstring, kw->name); |
|
1531 |
+ scanner->pos += len; |
|
1532 |
+ return kw->val; |
|
1533 |
+ } |
|
1534 |
+ len--; |
|
1535 |
+ } |
|
1536 |
+ scanner->pos++; |
|
1537 |
+ TOKEN_SET(lvalp, cstring, NULL); |
|
1538 |
+ return TOK_ERROR; |
|
1539 |
+} |
|
1540 |
+ |
|
1541 |
+static int yylex_init(yyscan_t *scanner) |
|
1542 |
+{ |
|
1543 |
+ *scanner = cli_calloc(1, sizeof(**scanner)); |
|
1544 |
+ return *scanner ? 0 : -1; |
|
1545 |
+} |
|
1546 |
+ |
|
1547 |
+static int yylex_destroy(yyscan_t scanner) |
|
1548 |
+{ |
|
1549 |
+ free(scanner->buf.data); |
|
1550 |
+ free(scanner); |
|
1551 |
+ return 0; |
|
1552 |
+} |
|
1553 |
+ |
|
1554 |
+static int yy_scan_bytes(const char *p, size_t len, yyscan_t scanner) |
|
1555 |
+{ |
|
1556 |
+ scanner->in = p; |
|
1557 |
+ scanner->insize = len; |
|
1558 |
+ scanner->pos = 0; |
|
1559 |
+ return 0; |
|
1560 |
+} |
|
1561 |
+ |
|
1562 |
+static void yyset_debug (int debug_flag ,yyscan_t yyscanner ) |
|
1563 |
+{ |
|
1564 |
+ /* TODO */ |
|
1565 |
+} |
|
1566 |
+ |
|
1567 |
+static void yy_delete_buffer( YY_BUFFER_STATE yyb, yyscan_t scanner) |
|
1568 |
+{ |
|
1569 |
+ /* TODO */ |
|
1570 |
+} |
|
1571 |
+ |
|
1572 |
+static const char *yyget_text(yyscan_t scanner) |
|
1573 |
+{ |
|
1574 |
+ assert(scanner->buf.data || scanner->yytext); |
|
1575 |
+ return scanner->yytext ? scanner->yytext : scanner->buf.data; |
|
1576 |
+} |
|
1577 |
+ |
|
1578 |
+static int yyget_leng(yyscan_t scanner) |
|
1579 |
+{ |
|
1580 |
+ /* we have a \0 too */ |
|
1581 |
+ return scanner->yylen ? scanner->yylen : (scanner->buf.pos > 0 ? scanner->buf.pos - 1 : 0); |
|
1582 |
+} |
|
1583 |
+ |
|
1584 |
+static int yylex(YYSTYPE *lvalp, yyscan_t scanner) |
|
1585 |
+{ |
|
1586 |
+ const size_t len = scanner->insize; |
|
1587 |
+ const unsigned char *in = (const unsigned char*)scanner->in; |
|
1588 |
+ unsigned char lookahead; |
|
1589 |
+ enum char_class cClass; |
|
1590 |
+ |
|
1591 |
+ scanner->yytext = NULL; |
|
1592 |
+ scanner->yylen = 0; |
|
1593 |
+ while(scanner->pos < scanner->insize) { |
|
1594 |
+ switch(scanner->state) { |
|
1595 |
+ case Initial: |
|
1596 |
+ textbuf_clean(&scanner->buf); |
|
1597 |
+ cClass = ctype[in[scanner->pos++]]; |
|
1598 |
+ switch(cClass) { |
|
1599 |
+ case Whitespace: |
|
1600 |
+ /* eat whitespace */ |
|
1601 |
+ continue; |
|
1602 |
+ case Slash: |
|
1603 |
+ if(scanner->pos < len) { |
|
1604 |
+ lookahead = in[scanner->pos]; |
|
1605 |
+ switch(lookahead) { |
|
1606 |
+ case '*': |
|
1607 |
+ scanner->state = MultilineComment; |
|
1608 |
+ scanner->pos++; |
|
1609 |
+ continue; |
|
1610 |
+ case '/': |
|
1611 |
+ scanner->state = SinglelineComment; |
|
1612 |
+ scanner->pos++; |
|
1613 |
+ continue; |
|
1614 |
+ } |
|
1615 |
+ } |
|
1616 |
+ --scanner->pos; |
|
1617 |
+ return parseOperator(lvalp, scanner); |
|
1618 |
+ case Operator: |
|
1619 |
+ --scanner->pos; |
|
1620 |
+ return parseOperator(lvalp, scanner); |
|
1621 |
+ case DQuote: |
|
1622 |
+ return parseDQString(lvalp, scanner); |
|
1623 |
+ case SQuote: |
|
1624 |
+ return parseSQString(lvalp, scanner); |
|
1625 |
+ case Digit: |
|
1626 |
+ --scanner->pos; |
|
1627 |
+ return parseNumber(lvalp, scanner); |
|
1628 |
+ case IdStart: |
|
1629 |
+ --scanner->pos; |
|
1630 |
+ return parseId(lvalp,scanner); |
|
1631 |
+ CASE_SPECIAL_CHAR(BracketOpen, "["); |
|
1632 |
+ CASE_SPECIAL_CHAR(BracketClose, "]"); |
|
1633 |
+ CASE_SPECIAL_CHAR(Comma, ","); |
|
1634 |
+ CASE_SPECIAL_CHAR(CurlyOpen, "{"); |
|
1635 |
+ CASE_SPECIAL_CHAR(CurlyClose, "}"); |
|
1636 |
+ CASE_SPECIAL_CHAR(ParOpen, "("); |
|
1637 |
+ CASE_SPECIAL_CHAR(ParClose, ")"); |
|
1638 |
+ CASE_SPECIAL_CHAR(Dot, "."); |
|
1639 |
+ CASE_SPECIAL_CHAR(SemiColon, ";"); |
|
1640 |
+ case Nop: |
|
1641 |
+ continue; |
|
1642 |
+ } |
|
1643 |
+ break; |
|
1644 |
+ case DoubleQString: |
|
1645 |
+ return parseString(lvalp, scanner, '"', DoubleQString); |
|
1646 |
+ case SingleQString: |
|
1647 |
+ return parseString(lvalp, scanner, '\'', DoubleQString); |
|
1648 |
+ case Identifier: |
|
1649 |
+ return parseId(lvalp, scanner); |
|
1650 |
+ case MultilineComment: |
|
1651 |
+ while(scanner->pos+1 < scanner->insize) { |
|
1652 |
+ if(in[scanner->pos] == '*' && in[scanner->pos+1] == '/') |
|
1653 |
+ break; |
|
1654 |
+ scanner->pos++; |
|
1655 |
+ } |
|
1656 |
+ scanner->state = Initial; |
|
1657 |
+ break; |
|
1658 |
+ case Number: |
|
1659 |
+ return parseNumber(lvalp, scanner); |
|
1660 |
+ case SinglelineComment: |
|
1661 |
+ while(scanner->pos < scanner->insize) { |
|
1662 |
+ if(in[scanner->pos] == '\n') |
|
1663 |
+ break; |
|
1664 |
+ scanner->pos++; |
|
1665 |
+ } |
|
1666 |
+ scanner->state = Initial; |
|
1667 |
+ break; |
|
1668 |
+ } |
|
1669 |
+ } |
|
1670 |
+ return 0; |
|
1671 |
+} |
0 | 1672 |
new file mode 100644 |
... | ... |
@@ -0,0 +1,29 @@ |
0 |
+/* |
|
1 |
+ * Javascript normalizer. |
|
2 |
+ * |
|
3 |
+ * Copyright (C) 2008 Sourcefire, Inc. |
|
4 |
+ * |
|
5 |
+ * Authors: Török Edvin |
|
6 |
+ * |
|
7 |
+ * This program is free software; you can redistribute it and/or modify |
|
8 |
+ * it under the terms of the GNU General Public License version 2 as |
|
9 |
+ * published by the Free Software Foundation. |
|
10 |
+ * |
|
11 |
+ * This program is distributed in the hope that it will be useful, |
|
12 |
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
13 |
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
14 |
+ * GNU General Public License for more details. |
|
15 |
+ * |
|
16 |
+ * You should have received a copy of the GNU General Public License |
|
17 |
+ * along with this program; if not, write to the Free Software |
|
18 |
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, |
|
19 |
+ * MA 02110-1301, USA. |
|
20 |
+ */ |
|
21 |
+ |
|
22 |
+struct parser_state; |
|
23 |
+int cli_js_init(struct parser_state *state); |
|
24 |
+void cli_js_process_buffer(struct parser_state *state, const char *buf, size_t n); |
|
25 |
+void cli_js_parse_done(struct parser_state* state); |
|
26 |
+void cli_js_output(struct parser_state *state); |
|
27 |
+void cli_js_destroy(struct parser_state *state); |
|
28 |
+ |
0 | 25 |
new file mode 100644 |
... | ... |
@@ -0,0 +1,146 @@ |
0 |
+/* |
|
1 |
+ * Javascript normalizer. |
|
2 |
+ * |
|
3 |
+ * Copyright (C) 2008 Sourcefire, Inc. |
|
4 |
+ * |
|
5 |
+ * Authors: Török Edvin |
|
6 |
+ * |
|
7 |
+ * This program is free software; you can redistribute it and/or modify |
|
8 |
+ * it under the terms of the GNU General Public License version 2 as |
|
9 |
+ * published by the Free Software Foundation. |
|
10 |
+ * |
|
11 |
+ * This program is distributed in the hope that it will be useful, |
|
12 |
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
13 |
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
14 |
+ * GNU General Public License for more details. |
|
15 |
+ * |
|
16 |
+ * You should have received a copy of the GNU General Public License |
|
17 |
+ * along with this program; if not, write to the Free Software |
|
18 |
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, |
|
19 |
+ * MA 02110-1301, USA. |
|
20 |
+ */ |
|
21 |
+#ifndef YYSTYPE |
|
22 |
+enum token_type { |
|
23 |
+ TOK_FUTURE_RESERVED_WORD=1, |
|
24 |
+ TOK_ERROR, |
|
25 |
+ TOK_IDENTIFIER_NAME, |
|
26 |
+ TOK_TRUE, |
|
27 |
+ TOK_FALSE, |
|
28 |
+ TOK_NULL, |
|
29 |
+ TOK_BRACKET_OPEN, |
|
30 |
+ TOK_BRACKET_CLOSE, |
|
31 |
+ TOK_COMMA, |
|
32 |
+ TOK_CURLY_BRACE_OPEN, |
|
33 |
+ TOK_CURLY_BRACE_CLOSE, |
|
34 |
+ TOK_PAR_OPEN, |
|
35 |
+ TOK_PAR_CLOSE, |
|
36 |
+ TOK_DOT, |
|
37 |
+ TOK_SEMICOLON, |
|
38 |
+ TOK_COLON, |
|
39 |
+ TOK_NEW, |
|
40 |
+ TOK_NumericInt, |
|
41 |
+ TOK_NumericFloat, |
|
42 |
+ TOK_StringLiteral, |
|
43 |
+ TOK_REGULAR_EXPRESSION_LITERAL, |
|
44 |
+ TOK_THIS, |
|
45 |
+ TOK_PLUSPLUS, |
|
46 |
+ TOK_MINUSMINUS, |
|
47 |
+ TOK_DELETE, |
|
48 |
+ TOK_VOID, |
|
49 |
+ TOK_TYPEOF, |
|
50 |
+ TOK_MINUS, |
|
51 |
+ TOK_TILDE, |
|
52 |
+ TOK_EXCLAMATION, |
|
53 |
+ TOK_MULTIPLY, |
|
54 |
+ TOK_DIVIDE, |
|
55 |
+ TOK_PERCENT, |
|
56 |
+ TOK_PLUS, |
|
57 |
+ TOK_SHIFT_LEFT, |
|
58 |
+ TOK_SHIFT_RIGHT, |
|
59 |
+ TOK_DOUBLESHIFT_RIGHT, |
|
60 |
+ TOK_LESS, |
|
61 |
+ TOK_GREATER, |
|
62 |
+ TOK_LESSEQUAL, |
|
63 |
+ TOK_GREATEREQUAL, |
|
64 |
+ TOK_INSTANCEOF, |
|
65 |
+ TOK_IN, |
|
66 |
+ TOK_EQUAL_EQUAL, |
|
67 |
+ TOK_NOT_EQUAL, |
|
68 |
+ TOK_TRIPLE_EQUAL, |
|
69 |
+ TOK_NOT_DOUBLEEQUAL, |
|
70 |
+ TOK_AND, |
|
71 |
+ TOK_XOR, |
|
72 |
+ TOK_OR, |
|
73 |
+ TOK_AND_AND, |
|
74 |
+ TOK_OR_OR, |
|
75 |
+ TOK_QUESTIONMARK, |
|
76 |
+ TOK_EQUAL, |
|
77 |
+ TOK_ASSIGNMENT_OPERATOR_NOEQUAL, |
|
78 |
+ TOK_VAR, |
|
79 |
+ TOK_IF, |
|
80 |
+ TOK_ELSE, |
|
81 |
+ TOK_DO, |
|
82 |
+ TOK_WHILE, |
|
83 |
+ TOK_FOR, |
|
84 |
+ TOK_CONTINUE, |
|
85 |
+ TOK_BREAK, |
|
86 |
+ TOK_RETURN, |
|
87 |
+ TOK_WITH, |
|
88 |
+ TOK_SWITCH, |
|
89 |
+ TOK_CASE, |
|
90 |
+ TOK_DEFAULT, |
|
91 |
+ TOK_THROW, |
|
92 |
+ TOK_TRY, |
|
93 |
+ TOK_CATCH, |
|
94 |
+ TOK_FINALLY, |
|
95 |
+ TOK_FUNCTION, |
|
96 |
+ TOK_UNNORM_IDENTIFIER |
|
97 |
+}; |
|
98 |
+ |
|
99 |
+enum val_type { |
|
100 |
+ vtype_undefined, |
|
101 |
+ vtype_cstring, |
|
102 |
+ vtype_string, |
|
103 |
+ vtype_scope, |
|
104 |
+ vtype_dval, |
|
105 |
+ vtype_ival |
|
106 |
+}; |
|
107 |
+ |
|
108 |
+typedef struct token { |
|
109 |
+ union { |
|
110 |
+ const char *cstring; |
|
111 |
+ char *string; |
|
112 |
+ struct scope *scope;/* for function */ |
|
113 |
+ double dval; |
|
114 |
+ long ival; |
|
115 |
+ } val; |
|
116 |
+ enum token_type type; |
|
117 |
+ enum val_type vtype; |
|
118 |
+} yystype; |
|
119 |
+ |
|
120 |
+/* inline functions to access the structure to ensure type safety */ |
|
121 |
+ |
|
122 |
+#define TOKEN_SET(DST, VTYPE, VAL) do {\ |
|
123 |
+ (DST)->vtype = vtype_##VTYPE ; (DST)->val.VTYPE = (VAL); \ |
|
124 |
+} while(0); |
|
125 |
+ |
|
126 |
+#define cstring_invalid NULL |
|
127 |
+#define string_invalid NULL |
|
128 |
+#define scope_invalid NULL |
|
129 |
+/* there isn't really an invalid double, or long value, but we don't care |
|
130 |
+ * about those values anyway, so -1 will be fine here */ |
|
131 |
+#define dval_invalid -1 |
|
132 |
+#define ival_invalid -1 |
|
133 |
+ |
|
134 |
+/* compatible if same type, or if we request a const char* instead of char*, |
|
135 |
+ * but not viceversa! */ |
|
136 |
+static int vtype_compatible(enum val_type orig, enum val_type req) |
|
137 |
+{ |
|
138 |
+ return orig == req || (orig == vtype_string && req == vtype_cstring); |
|
139 |
+} |
|
140 |
+#define COMPATIBLE(SRC, VTYPE) ((SRC)->vtype == vtype_##VTYPE || ) |
|
141 |
+ |
|
142 |
+#define TOKEN_GET(SRC, VTYPE) (vtype_compatible((SRC)->vtype, vtype_##VTYPE) ? (SRC)->val.VTYPE : VTYPE##_invalid) |
|
143 |
+ |
|
144 |
+#define YYSTYPE yystype |
|
145 |
+#endif |
0 | 146 |
new file mode 100644 |
... | ... |
@@ -0,0 +1,44 @@ |
0 |
+struct operator { |
|
1 |
+ const char *name; |
|
2 |
+ int val; |
|
3 |
+}; |
|
4 |
+%% |
|
5 |
+=, TOK_EQUAL |
|
6 |
++, TOK_PLUS |
|
7 |
+===, TOK_TRIPLE_EQUAL |
|
8 |
+>>>, TOK_DOUBLESHIFT_RIGHT |
|
9 |
+!==, TOK_NOT_DOUBLEEQUAL |
|
10 |
+<<, TOK_SHIFT_LEFT |
|
11 |
+<=, TOK_LESSEQUAL |
|
12 |
+==, TOK_EQUAL_EQUAL |
|
13 |
+>=, TOK_GREATEREQUAL |
|
14 |
+>>, TOK_SHIFT_RIGHT |
|
15 |
+||, TOK_OR_OR |
|
16 |
+--, TOK_MINUSMINUS |
|
17 |
+!=, TOK_NOT_EQUAL |
|
18 |
+&&, TOK_AND_AND |
|
19 |
+*=, TOK_ASSIGNMENT_OPERATOR_NOEQUAL |
|
20 |
+/=, TOK_ASSIGNMENT_OPERATOR_NOEQUAL |
|
21 |
+"%=", TOK_ASSIGNMENT_OPERATOR_NOEQUAL |
|
22 |
++=, TOK_ASSIGNMENT_OPERATOR_NOEQUAL |
|
23 |
+-=, TOK_ASSIGNMENT_OPERATOR_NOEQUAL |
|
24 |
+<<=, TOK_ASSIGNMENT_OPERATOR_NOEQUAL |
|
25 |
+>>=, TOK_ASSIGNMENT_OPERATOR_NOEQUAL |
|
26 |
+>>>=, TOK_ASSIGNMENT_OPERATOR_NOEQUAL |
|
27 |
+&=, TOK_ASSIGNMENT_OPERATOR_NOEQUAL |
|
28 |
+^=, TOK_ASSIGNMENT_OPERATOR_NOEQUAL |
|
29 |
+|=, TOK_ASSIGNMENT_OPERATOR_NOEQUAL |
|
30 |
+&, TOK_AND |
|
31 |
+:, TOK_COLON |
|
32 |
+/, TOK_DIVIDE |
|
33 |
+!, TOK_EXCLAMATION |
|
34 |
+>, TOK_GREATER |
|
35 |
+<, TOK_LESS |
|
36 |
+-, TOK_MINUS |
|
37 |
+*, TOK_MULTIPLY |
|
38 |
+|, TOK_OR |
|
39 |
+"%", TOK_PERCENT |
|
40 |
+?, TOK_QUESTIONMARK |
|
41 |
+~, TOK_TILDE |
|
42 |
+^, TOK_XOR |