diff options
28 files changed, 4869 insertions, 75 deletions
diff --git a/winsup/cygwin/ChangeLog b/winsup/cygwin/ChangeLog index 6c73935fb..f88f45bf5 100644 --- a/winsup/cygwin/ChangeLog +++ b/winsup/cygwin/ChangeLog @@ -1,3 +1,22 @@ +2001-12-19 Christopher Faylor <cgf@redhat.com> + + * Makefile.in (VPATH): Add regex directory. + (NM): new variable. + (OBSOLETE_FUNCTIONS): Ditto. + (NEW_FUNCTIONS): Ditto. + (install-headers): Install regex.h. + (install-man): New target. + (install): Use new target. + (DLL_OFILES): Add v8_reg* stuff. + (new-cygwin1.dll): Eliminate stamp-cygwin-lib creation. + (libcygwin.a): Remove obsolete functions from import lib. Add new functions. + * configure.in: Detect 'nm' tool. + * configure: Regenerate. + * cygwin.din: Export posix_reg* functions. Eliminate export of v8 reg* functions. + This is now handled in object files themselves. + * regex/*: New files. + * regexp/v8_*.c: New files, renamed from non v8_ equivalents. + 2001-12-17 Corinna Vinschen <corinna@vinschen.de> * include/getopt.h: Don't define getopt_long() and friends when diff --git a/winsup/cygwin/Makefile.in b/winsup/cygwin/Makefile.in index ab49440b8..78ea72156 100644 --- a/winsup/cygwin/Makefile.in +++ b/winsup/cygwin/Makefile.in @@ -14,7 +14,7 @@ srcdir:=@srcdir@ objdir:=. CONFIG_DIR:=$(srcdir)/config/@CONFIG_DIR@ -VPATH:=$(srcdir):$(CONFIG_DIR):$(srcdir)/regexp:$(srcdir)/lib +VPATH:=$(srcdir):$(CONFIG_DIR):$(srcdir)/regex:$(srcdir)/regexp:$(srcdir)/lib target_alias:=@target_alias@ build_alias:=@build_alias@ @@ -44,8 +44,8 @@ INSTALL_PROGRAM:=@INSTALL_PROGRAM@ # # --enable options from configure # -MT_SAFE = @MT_SAFE@ -DEFS = @DEFS@ +MT_SAFE:=@MT_SAFE@ +DEFS:=@DEFS@ CC:=@CC@ # FIXME: Which is it, CC or CC_FOR_TARGET? @@ -65,7 +65,8 @@ LD:=@LD@ DLLTOOL:=@DLLTOOL@ WINDRES:=@WINDRES@ AS:=@AS@ -LDSCRIPT=cygwin.sc +NM:=@NM@ +LDSCRIPT:=cygwin.sc # # Include common definitions for winsup directory @@ -124,14 +125,21 @@ DLL_OFILES:=assert.o autoload.o cygheap.o dcrt0.o debug.o delqueue.o dir.o \ fhandler_tape.o fhandler_termios.o fhandler_tty.o fhandler_windows.o \ fhandler_zero.o fork.o glob.o grp.o heap.o init.o ioctl.o localtime.o \ malloc.o miscfuncs.o mmap.o net.o ntea.o passwd.o path.o pinfo.o pipe.o \ - poll.o pthread.o regexp.o regerror.o regsub.o registry.o resource.o \ - scandir.o sched.o sec_acl.o sec_helper.o security.o select.o shared.o \ - shortcut.o signal.o sigproc.o smallprint.o spawn.o strace.o strsep.o \ - sync.o syscalls.o sysconf.o syslog.o termios.o thread.o times.o tty.o \ - uinfo.o uname.o wait.o wincap.o window.o \ + poll.o pthread.o regcomp.o regerror.o regexec.o regfree.o registry.o \ + resource.o scandir.o sched.o sec_acl.o sec_helper.o security.o \ + select.o shared.o shortcut.o signal.o sigproc.o smallprint.o spawn.o \ + strace.o strsep.o sync.o syscalls.o sysconf.o syslog.o termios.o \ + thread.o times.o tty.o uinfo.o uname.o v8_regexp.o v8_regerror.o \ + v8_regsub.o wait.o wincap.o window.o \ $(EXTRA_DLL_OFILES) $(EXTRA_OFILES) $(MALLOC_OFILES) $(MT_SAFE_OBJECTS) -GMON_OFILES:= gmon.o mcount.o profil.o +GMON_OFILES:=gmon.o mcount.o profil.o + +OBSOLETE_FUNCTIONS:=regcomp regerror regexec regfree regsub +NEW_FUNCTIONS:=regcomp posix_regcomp \ + regerror posix_regerror \ + regexec posix_regexec \ + regfree posix_regfree .PHONY: all force dll_ofiles install all_target install_target all_host install_host \ install install_target install-libs install-headers @@ -150,7 +158,7 @@ all_host: new-$(LIB_NAME) cygrun.exe force: -install: install-libs install-headers $(install_host) $(install_target) +install: install-libs install-headers install-man $(install_host) $(install_target) install-libs: $(LIB_NAME) $(INSTALL_DATA) new-$(DLL_NAME) $(bindir)/$(DLL_NAME); \ @@ -165,43 +173,54 @@ install-headers: $(INSTALL_DATA) $$i $(tooldir)/$$sub/`basename $$i` ; \ done ; \ done ; \ - $(INSTALL_DATA) regexp/regexp.h $(tooldir)/include/regexp.h + $(INSTALL_DATA) regex/regex.h $(tooldir)/include/regex.h + +install-man: + cd $(srcdir); \ + for i in `find . -type f -name '*.2'`; do \ + $(INSTALL_DATA) $$i $(tooldir)/man/man2/`basename $$i` ; \ + done; \ + for i in `find . -type f -name '*.3'`; do \ + $(INSTALL_DATA) $$i $(tooldir)/man/man3/`basename $$i` ; \ + done; \ + for i in `find . -type f -name '*.5'`; do \ + $(INSTALL_DATA) $$i $(tooldir)/man/man5/`basename $$i` ; \ + done; \ + for i in `find . -type f -name '*.7'`; do \ + $(INSTALL_DATA) $$i $(tooldir)/man/man7/`basename $$i` ; \ + done install_host: clean: - -rm -f *.o *.dll *.a *.exp junk *.base version.cc regexp/*.o winver_stamp *.exe *.d + -rm -f *.o *.dll *.a *.exp junk *.base version.cc regexp/*.o winver_stamp *.exe *.d *stamp* maintainer-clean realclean: clean @echo "This command is intended for maintainers to use;" @echo "it deletes files that may require special tools to rebuild." -rm -fr configure -# Rule to build libcygwin.a - -# Rule to make stub library used by testsuite - -# dependency set to $(LIB_NAME) to accommodate make -j2. -# Otherwise dlltool gets confused. cgf (11-16-2000) -new-$(LIB_NAME): $(LIB_NAME) - $(DLLTOOL) --as=$(AS) --dllname new-$(DLL_NAME) --def $(DEF_FILE) --output-lib new-templib.a - $(AR) rcv new-templib.a $(LIBCOS) - mv new-templib.a new-$(LIB_NAME) # Rule to build cygwin.dll - new-$(DLL_NAME): $(LDSCRIPT) $(DLL_OFILES) $(DEF_FILE) $(DLL_IMPORTS) $(LIBC) $(LIBM) Makefile winver_stamp $(CXX) $(CXXFLAGS) -nostdlib -Wl,-T$(firstword $^) -Wl,--out-implib,cygdll.a -shared -o $@ \ -e $(DLL_ENTRY) $(DEF_FILE) $(DLL_OFILES) version.o winver.o \ $(DLL_IMPORTS) $(MALLOC_OBJ) $(LIBM) $(LIBC) \ -lstdc++ -lgcc -lshell32 -luuid - @rm -f stamp-cygwin-lib -$(LIB_NAME): new-$(DLL_NAME) $(LIBCOS) stamp-cygwin-lib - (echo create $(LIB_NAME); echo addlib cygdll.a; echo addmod $(LIBCOS); echo save) | $(AR) -M +# Rule to build libcygwin.a +$(LIB_NAME): rmsym newsym new-$(DLL_NAME) $(LIBCOS) + /bin/sh ${word 1,$^} ./cygdll.a $(NM) $(AR) $(RANLIB) $(OBSOLETE_FUNCTIONS) || exit 0 + /bin/sh ${word 2,$^} ./cygdll.a $(AS) $(AR) $(RANLIB) $(NEW_FUNCTIONS) || exit 0 + (echo create $(LIB_NAME); echo addmod $(LIBCOS); echo addlib cygdll.a; echo save) | $(AR) -M -stamp-cygwin-lib: - @touch stamp-cygwin-lib +# Rule to make stub library used by testsuite +# dependency set to $(LIB_NAME) to accommodate make -j2. +# Otherwise dlltool gets confused. cgf (11-16-2000) +new-$(LIB_NAME): $(LIB_NAME) + $(DLLTOOL) --as=$(AS) --dllname new-$(DLL_NAME) --def $(DEF_FILE) --output-lib new-templib.a + $(AR) rcv new-templib.a $(LIBCOS) + mv new-templib.a new-$(LIB_NAME) dll_ofiles: $(DLL_OFILES) diff --git a/winsup/cygwin/configure b/winsup/cygwin/configure index 11311edda..3f41211fb 100755 --- a/winsup/cygwin/configure +++ b/winsup/cygwin/configure @@ -1274,10 +1274,77 @@ else fi fi +# Extract the first word of "${ac_tool_prefix}nm", so it can be a program name with args. +set dummy ${ac_tool_prefix}nm; ac_word=$2 +echo $ac_n "checking for $ac_word""... $ac_c" 1>&6 +echo "configure:1281: checking for $ac_word" >&5 +if eval "test \"`echo '$''{'ac_cv_prog_NM'+set}'`\" = set"; then + echo $ac_n "(cached) $ac_c" 1>&6 +else + if test -n "$NM"; then + ac_cv_prog_NM="$NM" # Let the user override the test. +else + IFS="${IFS= }"; ac_save_ifs="$IFS"; IFS=":" + ac_dummy="$PATH" + for ac_dir in $ac_dummy; do + test -z "$ac_dir" && ac_dir=. + if test -f $ac_dir/$ac_word; then + ac_cv_prog_NM="${ac_tool_prefix}nm" + break + fi + done + IFS="$ac_save_ifs" +fi +fi +NM="$ac_cv_prog_NM" +if test -n "$NM"; then + echo "$ac_t""$NM" 1>&6 +else + echo "$ac_t""no" 1>&6 +fi + + +if test -z "$ac_cv_prog_NM"; then +if test -n "$ac_tool_prefix"; then + # Extract the first word of "nm", so it can be a program name with args. +set dummy nm; ac_word=$2 +echo $ac_n "checking for $ac_word""... $ac_c" 1>&6 +echo "configure:1313: checking for $ac_word" >&5 +if eval "test \"`echo '$''{'ac_cv_prog_NM'+set}'`\" = set"; then + echo $ac_n "(cached) $ac_c" 1>&6 +else + if test -n "$NM"; then + ac_cv_prog_NM="$NM" # Let the user override the test. +else + IFS="${IFS= }"; ac_save_ifs="$IFS"; IFS=":" + ac_dummy="$PATH" + for ac_dir in $ac_dummy; do + test -z "$ac_dir" && ac_dir=. + if test -f $ac_dir/$ac_word; then + ac_cv_prog_NM="nm" + break + fi + done + IFS="$ac_save_ifs" + test -z "$ac_cv_prog_NM" && ac_cv_prog_NM="nm" +fi +fi +NM="$ac_cv_prog_NM" +if test -n "$NM"; then + echo "$ac_t""$NM" 1>&6 +else + echo "$ac_t""no" 1>&6 +fi + +else + NM="nm" +fi +fi + # Extract the first word of "${ac_tool_prefix}dlltool", so it can be a program name with args. set dummy ${ac_tool_prefix}dlltool; ac_word=$2 echo $ac_n "checking for $ac_word""... $ac_c" 1>&6 -echo "configure:1281: checking for $ac_word" >&5 +echo "configure:1348: checking for $ac_word" >&5 if eval "test \"`echo '$''{'ac_cv_prog_DLLTOOL'+set}'`\" = set"; then echo $ac_n "(cached) $ac_c" 1>&6 else @@ -1309,7 +1376,7 @@ if test -n "$ac_tool_prefix"; then # Extract the first word of "dlltool", so it can be a program name with args. set dummy dlltool; ac_word=$2 echo $ac_n "checking for $ac_word""... $ac_c" 1>&6 -echo "configure:1313: checking for $ac_word" >&5 +echo "configure:1380: checking for $ac_word" >&5 if eval "test \"`echo '$''{'ac_cv_prog_DLLTOOL'+set}'`\" = set"; then echo $ac_n "(cached) $ac_c" 1>&6 else @@ -1344,7 +1411,7 @@ fi # Extract the first word of "${ac_tool_prefix}windres", so it can be a program name with args. set dummy ${ac_tool_prefix}windres; ac_word=$2 echo $ac_n "checking for $ac_word""... $ac_c" 1>&6 -echo "configure:1348: checking for $ac_word" >&5 +echo "configure:1415: checking for $ac_word" >&5 if eval "test \"`echo '$''{'ac_cv_prog_WINDRES'+set}'`\" = set"; then echo $ac_n "(cached) $ac_c" 1>&6 else @@ -1376,7 +1443,7 @@ if test -n "$ac_tool_prefix"; then # Extract the first word of "windres", so it can be a program name with args. set dummy windres; ac_word=$2 echo $ac_n "checking for $ac_word""... $ac_c" 1>&6 -echo "configure:1380: checking for $ac_word" >&5 +echo "configure:1447: checking for $ac_word" >&5 if eval "test \"`echo '$''{'ac_cv_prog_WINDRES'+set}'`\" = set"; then echo $ac_n "(cached) $ac_c" 1>&6 else @@ -1410,7 +1477,7 @@ fi echo $ac_n "checking how to run the C preprocessor""... $ac_c" 1>&6 -echo "configure:1414: checking how to run the C preprocessor" >&5 +echo "configure:1481: checking how to run the C preprocessor" >&5 # On Suns, sometimes $CPP names a directory. if test -n "$CPP" && test -d "$CPP"; then CPP= @@ -1425,13 +1492,13 @@ else # On the NeXT, cc -E runs the code through the compiler's parser, # not just through cpp. cat > conftest.$ac_ext <<EOF -#line 1429 "configure" +#line 1496 "configure" #include "confdefs.h" #include <assert.h> Syntax Error EOF ac_try="$ac_cpp conftest.$ac_ext >/dev/null 2>conftest.out" -{ (eval echo configure:1435: \"$ac_try\") 1>&5; (eval $ac_try) 2>&5; } +{ (eval echo configure:1502: \"$ac_try\") 1>&5; (eval $ac_try) 2>&5; } ac_err=`grep -v '^ *+' conftest.out | grep -v "^conftest.${ac_ext}\$"` if test -z "$ac_err"; then : @@ -1442,13 +1509,13 @@ else rm -rf conftest* CPP="${CC-cc} -E -traditional-cpp" cat > conftest.$ac_ext <<EOF -#line 1446 "configure" +#line 1513 "configure" #include "confdefs.h" #include <assert.h> Syntax Error EOF ac_try="$ac_cpp conftest.$ac_ext >/dev/null 2>conftest.out" -{ (eval echo configure:1452: \"$ac_try\") 1>&5; (eval $ac_try) 2>&5; } +{ (eval echo configure:1519: \"$ac_try\") 1>&5; (eval $ac_try) 2>&5; } ac_err=`grep -v '^ *+' conftest.out | grep -v "^conftest.${ac_ext}\$"` if test -z "$ac_err"; then : @@ -1459,13 +1526,13 @@ else rm -rf conftest* CPP="${CC-cc} -nologo -E" cat > conftest.$ac_ext <<EOF -#line 1463 "configure" +#line 1530 "configure" #include "confdefs.h" #include <assert.h> Syntax Error EOF ac_try="$ac_cpp conftest.$ac_ext >/dev/null 2>conftest.out" -{ (eval echo configure:1469: \"$ac_try\") 1>&5; (eval $ac_try) 2>&5; } +{ (eval echo configure:1536: \"$ac_try\") 1>&5; (eval $ac_try) 2>&5; } ac_err=`grep -v '^ *+' conftest.out | grep -v "^conftest.${ac_ext}\$"` if test -z "$ac_err"; then : @@ -1492,19 +1559,19 @@ echo "$ac_t""$CPP" 1>&6 # The Ultrix 4.2 mips builtin alloca declared by alloca.h only works # for constant arguments. Useless! echo $ac_n "checking for working alloca.h""... $ac_c" 1>&6 -echo "configure:1496: checking for working alloca.h" >&5 +echo "configure:1563: checking for working alloca.h" >&5 if eval "test \"`echo '$''{'ac_cv_header_alloca_h'+set}'`\" = set"; then echo $ac_n "(cached) $ac_c" 1>&6 else cat > conftest.$ac_ext <<EOF -#line 1501 "configure" +#line 1568 "configure" #include "confdefs.h" #include <alloca.h> int main() { char *p = alloca(2 * sizeof(int)); ; return 0; } EOF -if { (eval echo configure:1508: \"$ac_link\") 1>&5; (eval $ac_link) 2>&5; } && test -s conftest${ac_exeext}; then +if { (eval echo configure:1575: \"$ac_link\") 1>&5; (eval $ac_link) 2>&5; } && test -s conftest${ac_exeext}; then rm -rf conftest* ac_cv_header_alloca_h=yes else @@ -1525,12 +1592,12 @@ EOF fi echo $ac_n "checking for alloca""... $ac_c" 1>&6 -echo "configure:1529: checking for alloca" >&5 +echo "configure:1596: checking for alloca" >&5 if eval "test \"`echo '$''{'ac_cv_func_alloca_works'+set}'`\" = set"; then echo $ac_n "(cached) $ac_c" 1>&6 else cat > conftest.$ac_ext <<EOF -#line 1534 "configure" +#line 1601 "configure" #include "confdefs.h" #ifdef __GNUC__ @@ -1558,7 +1625,7 @@ int main() { char *p = (char *) alloca(1); ; return 0; } EOF -if { (eval echo configure:1562: \"$ac_link\") 1>&5; (eval $ac_link) 2>&5; } && test -s conftest${ac_exeext}; then +if { (eval echo configure:1629: \"$ac_link\") 1>&5; (eval $ac_link) 2>&5; } && test -s conftest${ac_exeext}; then rm -rf conftest* ac_cv_func_alloca_works=yes else @@ -1590,12 +1657,12 @@ EOF echo $ac_n "checking whether alloca needs Cray hooks""... $ac_c" 1>&6 -echo "configure:1594: checking whether alloca needs Cray hooks" >&5 +echo "configure:1661: checking whether alloca needs Cray hooks" >&5 if eval "test \"`echo '$''{'ac_cv_os_cray'+set}'`\" = set"; then echo $ac_n "(cached) $ac_c" 1>&6 else cat > conftest.$ac_ext <<EOF -#line 1599 "configure" +#line 1666 "configure" #include "confdefs.h" #if defined(CRAY) && ! defined(CRAY2) webecray @@ -1620,12 +1687,12 @@ echo "$ac_t""$ac_cv_os_cray" 1>&6 if test $ac_cv_os_cray = yes; then for ac_func in _getb67 GETB67 getb67; do echo $ac_n "checking for $ac_func""... $ac_c" 1>&6 -echo "configure:1624: checking for $ac_func" >&5 +echo "configure:1691: checking for $ac_func" >&5 if eval "test \"`echo '$''{'ac_cv_func_$ac_func'+set}'`\" = set"; then echo $ac_n "(cached) $ac_c" 1>&6 else cat > conftest.$ac_ext <<EOF -#line 1629 "configure" +#line 1696 "configure" #include "confdefs.h" /* System header to define __stub macros and hopefully few prototypes, which can conflict with char $ac_func(); below. */ @@ -1648,7 +1715,7 @@ $ac_func(); ; return 0; } EOF -if { (eval echo configure:1652: \"$ac_link\") 1>&5; (eval $ac_link) 2>&5; } && test -s conftest${ac_exeext}; then +if { (eval echo configure:1719: \"$ac_link\") 1>&5; (eval $ac_link) 2>&5; } && test -s conftest${ac_exeext}; then rm -rf conftest* eval "ac_cv_func_$ac_func=yes" else @@ -1675,7 +1742,7 @@ done fi echo $ac_n "checking stack direction for C alloca""... $ac_c" 1>&6 -echo "configure:1679: checking stack direction for C alloca" >&5 +echo "configure:1746: checking stack direction for C alloca" >&5 if eval "test \"`echo '$''{'ac_cv_c_stack_direction'+set}'`\" = set"; then echo $ac_n "(cached) $ac_c" 1>&6 else @@ -1683,7 +1750,7 @@ else ac_cv_c_stack_direction=0 else cat > conftest.$ac_ext <<EOF -#line 1687 "configure" +#line 1754 "configure" #include "confdefs.h" find_stack_direction () { @@ -1702,7 +1769,7 @@ main () exit (find_stack_direction() < 0); } EOF -if { (eval echo configure:1706: \"$ac_link\") 1>&5; (eval $ac_link) 2>&5; } && test -s conftest${ac_exeext} && (./conftest; exit) 2>/dev/null +if { (eval echo configure:1773: \"$ac_link\") 1>&5; (eval $ac_link) 2>&5; } && test -s conftest${ac_exeext} && (./conftest; exit) 2>/dev/null then ac_cv_c_stack_direction=1 else @@ -1724,7 +1791,7 @@ EOF fi echo $ac_n "checking whether ${MAKE-make} sets \${MAKE}""... $ac_c" 1>&6 -echo "configure:1728: checking whether ${MAKE-make} sets \${MAKE}" >&5 +echo "configure:1795: checking whether ${MAKE-make} sets \${MAKE}" >&5 set dummy ${MAKE-make}; ac_make=`echo "$2" | sed 'y%./+-%__p_%'` if eval "test \"`echo '$''{'ac_cv_prog_make_${ac_make}_set'+set}'`\" = set"; then echo $ac_n "(cached) $ac_c" 1>&6 @@ -1761,7 +1828,7 @@ ac_link='${CXX-g++} -o conftest${ac_exeext} $CXXFLAGS $CPPFLAGS $LDFLAGS conftes cross_compiling=$ac_cv_prog_cxx_cross cat > conftest.$ac_ext <<EOF -#line 1765 "configure" +#line 1832 "configure" #include "confdefs.h" #include <string.h> @@ -1774,7 +1841,7 @@ int main() { ; return 0; } EOF -if { (eval echo configure:1778: \"$ac_compile\") 1>&5; (eval $ac_compile) 2>&5; }; then +if { (eval echo configure:1845: \"$ac_compile\") 1>&5; (eval $ac_compile) 2>&5; }; then rm -rf conftest* use_builtin_memset=yes else @@ -1899,7 +1966,7 @@ esac echo $ac_n "checking if newlib is part of the build tree""... $ac_c" 1>&6 -echo "configure:1903: checking if newlib is part of the build tree" >&5 +echo "configure:1970: checking if newlib is part of the build tree" >&5 EXE_LDFLAGS= if test -d ../newlib @@ -1914,9 +1981,9 @@ fi if test x"$EXE_LDFLAGS" = x then echo $ac_n "checking if installed newlib needed""... $ac_c" 1>&6 -echo "configure:1918: checking if installed newlib needed" >&5 +echo "configure:1985: checking if installed newlib needed" >&5 cat > conftest.$ac_ext <<EOF -#line 1920 "configure" +#line 1987 "configure" #include "confdefs.h" int main() { @@ -1924,7 +1991,7 @@ int main() { ; return 0; } EOF -if { (eval echo configure:1928: \"$ac_link\") 1>&5; (eval $ac_link) 2>&5; } && test -s conftest${ac_exeext}; then +if { (eval echo configure:1995: \"$ac_link\") 1>&5; (eval $ac_link) 2>&5; } && test -s conftest${ac_exeext}; then rm -rf conftest* echo "$ac_t""no" 1>&6 @@ -2115,6 +2182,7 @@ s%@AR@%$AR%g s%@AS@%$AS%g s%@RANLIB@%$RANLIB%g s%@LD@%$LD%g +s%@NM@%$NM%g s%@DLLTOOL@%$DLLTOOL%g s%@WINDRES@%$WINDRES%g s%@CPP@%$CPP%g diff --git a/winsup/cygwin/configure.in b/winsup/cygwin/configure.in index 8a20f084b..3934992a7 100644 --- a/winsup/cygwin/configure.in +++ b/winsup/cygwin/configure.in @@ -91,6 +91,8 @@ AC_CHECK_TOOL(RANLIB, ranlib, ranlib) dnl C_SUBST(RANLIB) AC_CHECK_TOOL(LD, ld, ld) dnl C_SUBST(LD) +AC_CHECK_TOOL(NM, nm, nm) +dnl C_SUBST(NM) AC_CHECK_TOOL(DLLTOOL, dlltool, dlltool) dnl C_SUBST(DLLTOOL) AC_CHECK_TOOL(WINDRES, windres, windres) diff --git a/winsup/cygwin/cygwin.din b/winsup/cygwin/cygwin.din index dae5fd9d3..41da8fa4e 100644 --- a/winsup/cygwin/cygwin.din +++ b/winsup/cygwin/cygwin.din @@ -594,16 +594,10 @@ readv _readv = readv realloc = export_realloc _realloc = export_realloc -regcomp -_regcomp = regcomp -regexec -_regexec = regexec -regerror -_regerror = regerror -regfree -_regfree = regfree -regsub -_regsub = regsub +posix_regcomp +posix_regerror +posix_regexec +posix_regfree remainder _remainder = remainder remainderf @@ -910,6 +904,7 @@ wctomb _wctomb = wctomb write _write = write +barfly = write writev _writev = writev y0 diff --git a/winsup/cygwin/newsym b/winsup/cygwin/newsym new file mode 100755 index 000000000..688a8c6f8 --- /dev/null +++ b/winsup/cygwin/newsym @@ -0,0 +1,24 @@ +#!/bin/sh -x +lib=$1; shift +as=$1; shift +ar=$1; shift +ranlib=$1; shift +rm -rf newsym.dir +trap "rm -rf newsym.dir" 0 1 2 15 +mkdir newsym.dir +while [ -n "$1" ]; do + newsym=$1; shift + oldsym=$1; shift + cat <<EOF > newsym.dir/$newsym.s + .section .idata$6 + .extern __imp__$oldsym + .extern __head_cygwin1_dll + .section .text + .global _$newsym +_$newsym: + jmp *__imp__$oldsym +EOF + $as -o newsym.dir/$newsym.o newsym.dir/$newsym.s +done +$ar cru $lib newsym.dir/*.o +$ranlib $lib diff --git a/winsup/cygwin/regex/COPYRIGHT b/winsup/cygwin/regex/COPYRIGHT new file mode 100644 index 000000000..30c1f7a48 --- /dev/null +++ b/winsup/cygwin/regex/COPYRIGHT @@ -0,0 +1,20 @@ +Copyright 1992, 1993, 1994, 1997 Henry Spencer. All rights reserved. +This software is not subject to any license of the American Telephone +and Telegraph Company or of the Regents of the University of California. + +Permission is granted to anyone to use this software for any purpose on +any computer system, and to alter it and redistribute it, subject +to the following restrictions: + +1. The author is not responsible for the consequences of use of this + software, no matter how awful, even if they arise from flaws in it. + +2. The origin of this software must not be misrepresented, either by + explicit claim or by omission. Since few users ever read sources, + credits must appear in the documentation. + +3. Altered versions must be plainly marked as such, and must not be + misrepresented as being the original software. Since few users + ever read sources, credits must appear in the documentation. + +4. This notice may not be removed or altered. diff --git a/winsup/cygwin/regex/cclass.h b/winsup/cygwin/regex/cclass.h new file mode 100644 index 000000000..7ddb44afb --- /dev/null +++ b/winsup/cygwin/regex/cclass.h @@ -0,0 +1,31 @@ +/* character-class table */ +static struct cclass { + const char *name; + const char *chars; + const char *multis; +} cclasses[] = { + {"alnum", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz\ +0123456789", ""}, + {"alpha", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz", + ""}, + {"blank", " \t", ""}, + {"cntrl", "\007\b\t\n\v\f\r\1\2\3\4\5\6\16\17\20\21\22\23\24\ + \25\26\27\30\31\32\33\34\35\36\37\177", ""}, + {"digit", "0123456789", ""}, + {"graph", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz\ + 0123456789!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + ""}, + {"lower", "abcdefghijklmnopqrstuvwxyz", + ""}, + {"print", "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz\ + 0123456789!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~ ", + ""}, + {"punct", "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + ""}, + {"space", "\t\n\v\f\r ", ""}, + {"upper", "ABCDEFGHIJKLMNOPQRSTUVWXYZ", + ""}, + {"xdigit", "0123456789ABCDEFabcdef", + ""}, + {NULL, 0, ""} +}; diff --git a/winsup/cygwin/regex/cname.h b/winsup/cygwin/regex/cname.h new file mode 100644 index 000000000..1000f603e --- /dev/null +++ b/winsup/cygwin/regex/cname.h @@ -0,0 +1,102 @@ +/* character-name table */ +static struct cname { + const char *name; + char code; +} cnames[] = { + {"NUL", '\0'}, + {"SOH", '\001'}, + {"STX", '\002'}, + {"ETX", '\003'}, + {"EOT", '\004'}, + {"ENQ", '\005'}, + {"ACK", '\006'}, + {"BEL", '\007'}, + {"alert", '\007'}, + {"BS", '\010'}, + {"backspace", '\b'}, + {"HT", '\011'}, + {"tab", '\t'}, + {"LF", '\012'}, + {"newline", '\n'}, + {"VT", '\013'}, + {"vertical-tab", '\v'}, + {"FF", '\014'}, + {"form-feed", '\f'}, + {"CR", '\015'}, + {"carriage-return", '\r'}, + {"SO", '\016'}, + {"SI", '\017'}, + {"DLE", '\020'}, + {"DC1", '\021'}, + {"DC2", '\022'}, + {"DC3", '\023'}, + {"DC4", '\024'}, + {"NAK", '\025'}, + {"SYN", '\026'}, + {"ETB", '\027'}, + {"CAN", '\030'}, + {"EM", '\031'}, + {"SUB", '\032'}, + {"ESC", '\033'}, + {"IS4", '\034'}, + {"FS", '\034'}, + {"IS3", '\035'}, + {"GS", '\035'}, + {"IS2", '\036'}, + {"RS", '\036'}, + {"IS1", '\037'}, + {"US", '\037'}, + {"space", ' '}, + {"exclamation-mark", '!'}, + {"quotation-mark", '"'}, + {"number-sign", '#'}, + {"dollar-sign", '$'}, + {"percent-sign", '%'}, + {"ampersand", '&'}, + {"apostrophe", '\''}, + {"left-parenthesis", '('}, + {"right-parenthesis", ')'}, + {"asterisk", '*'}, + {"plus-sign", '+'}, + {"comma", ','}, + {"hyphen", '-'}, + {"hyphen-minus", '-'}, + {"period", '.'}, + {"full-stop", '.'}, + {"slash", '/'}, + {"solidus", '/'}, + {"zero", '0'}, + {"one", '1'}, + {"two", '2'}, + {"three", '3'}, + {"four", '4'}, + {"five", '5'}, + {"six", '6'}, + {"seven", '7'}, + {"eight", '8'}, + {"nine", '9'}, + {"colon", ':'}, + {"semicolon", ';'}, + {"less-than-sign", '<'}, + {"equals-sign", '='}, + {"greater-than-sign", '>'}, + {"question-mark", '?'}, + {"commercial-at", '@'}, + {"left-square-bracket", '['}, + {"backslash", '\\'}, + {"reverse-solidus", '\\'}, + {"right-square-bracket", ']'}, + {"circumflex", '^'}, + {"circumflex-accent", '^'}, + {"underscore", '_'}, + {"low-line", '_'}, + {"grave-accent", '`'}, + {"left-brace", '{'}, + {"left-curly-bracket", '{'}, + {"vertical-line", '|'}, + {"right-brace", '}'}, + {"right-curly-bracket", '}'}, + {"tilde", '~'}, + {"DEL", '\177'}, + {NULL, 0} +}; diff --git a/winsup/cygwin/regex/engine.c b/winsup/cygwin/regex/engine.c new file mode 100644 index 000000000..919fe3f64 --- /dev/null +++ b/winsup/cygwin/regex/engine.c @@ -0,0 +1,1019 @@ +/* + * The matching engine and friends. This file is #included by regexec.c + * after suitable #defines of a variety of macros used herein, so that + * different state representations can be used without duplicating masses + * of code. + */ + +#ifdef SNAMES +#define matcher smatcher +#define fast sfast +#define slow sslow +#define dissect sdissect +#define backref sbackref +#define step sstep +#define print sprint +#define at sat +#define match smat +#endif +#ifdef LNAMES +#define matcher lmatcher +#define fast lfast +#define slow lslow +#define dissect ldissect +#define backref lbackref +#define step lstep +#define print lprint +#define at lat +#define match lmat +#endif + +/* another structure passed up and down to avoid zillions of parameters */ +struct match { + struct re_guts *g; + int eflags; + regmatch_t *pmatch; /* [nsub+1] (0 element unused) */ + char *offp; /* offsets work from here */ + char *beginp; /* start of string -- virtual NUL precedes */ + char *endp; /* end of string -- virtual NUL here */ + char *coldp; /* can be no match starting before here */ + char **lastpos; /* [nplus+1] */ + STATEVARS; + states st; /* current states */ + states fresh; /* states for a fresh start */ + states tmp; /* temporary */ + states empty; /* empty set of states */ +}; + +#include "engine.ih" + +#ifdef REDEBUG +#define SP(t, s, c) print(m, t, s, c, stdout) +#define AT(t, p1, p2, s1, s2) at(m, t, p1, p2, s1, s2) +#define NOTE(str) { if (m->eflags®_TRACE) printf("=%s\n", (str)); } +#else +#define SP(t, s, c) /* nothing */ +#define AT(t, p1, p2, s1, s2) /* nothing */ +#define NOTE(s) /* nothing */ +#endif + +/* + - matcher - the actual matching engine + == static int matcher(register struct re_guts *g, char *string, \ + == size_t nmatch, regmatch_t pmatch[], int eflags); + */ +static int /* 0 success, REG_NOMATCH failure */ +matcher(g, string, nmatch, pmatch, eflags) +register struct re_guts *g; +char *string; +size_t nmatch; +regmatch_t pmatch[]; +int eflags; +{ + register char *endp; + register int i; + struct match mv; + register struct match *m = &mv; + register char *dp; + const register sopno gf = g->firststate+1; /* +1 for OEND */ + const register sopno gl = g->laststate; + char *start; + char *stop; + + /* simplify the situation where possible */ + if (g->cflags®_NOSUB) + nmatch = 0; + if (eflags®_STARTEND) { + start = string + pmatch[0].rm_so; + stop = string + pmatch[0].rm_eo; + } else { + start = string; + stop = start + strlen(start); + } + if (stop < start) + return(REG_INVARG); + + /* prescreening; this does wonders for this rather slow code */ + if (g->must != NULL) { + for (dp = start; dp < stop; dp++) + if (*dp == g->must[0] && stop - dp >= g->mlen && + memcmp(dp, g->must, (size_t)g->mlen) == 0) + break; + if (dp == stop) /* we didn't find g->must */ + return(REG_NOMATCH); + } + + /* match struct setup */ + m->g = g; + m->eflags = eflags; + m->pmatch = NULL; + m->lastpos = NULL; + m->offp = string; + m->beginp = start; + m->endp = stop; + STATESETUP(m, 4); + SETUP(m->st); + SETUP(m->fresh); + SETUP(m->tmp); + SETUP(m->empty); + CLEAR(m->empty); + + /* this loop does only one repetition except for backrefs */ + for (;;) { + endp = fast(m, start, stop, gf, gl); + if (endp == NULL) { /* a miss */ + STATETEARDOWN(m); + return(REG_NOMATCH); + } + if (nmatch == 0 && !g->backrefs) + break; /* no further info needed */ + + /* where? */ + assert(m->coldp != NULL); + for (;;) { + NOTE("finding start"); + endp = slow(m, m->coldp, stop, gf, gl); + if (endp != NULL) + break; + assert(m->coldp < m->endp); + m->coldp++; + } + if (nmatch == 1 && !g->backrefs) + break; /* no further info needed */ + + /* oh my, he wants the subexpressions... */ + if (m->pmatch == NULL) + m->pmatch = (regmatch_t *)malloc((m->g->nsub + 1) * + sizeof(regmatch_t)); + if (m->pmatch == NULL) { + STATETEARDOWN(m); + return(REG_ESPACE); + } + for (i = 1; i <= m->g->nsub; i++) + m->pmatch[i].rm_so = m->pmatch[i].rm_eo = -1; + if (!g->backrefs && !(m->eflags®_BACKR)) { + NOTE("dissecting"); + dp = dissect(m, m->coldp, endp, gf, gl); + } else { + if (g->nplus > 0 && m->lastpos == NULL) + m->lastpos = (char **)malloc((g->nplus+1) * + sizeof(char *)); + if (g->nplus > 0 && m->lastpos == NULL) { + free(m->pmatch); + STATETEARDOWN(m); + return(REG_ESPACE); + } + NOTE("backref dissect"); + dp = backref(m, m->coldp, endp, gf, gl, (sopno)0); + } + if (dp != NULL) + break; + + /* uh-oh... we couldn't find a subexpression-level match */ + assert(g->backrefs); /* must be back references doing it */ + assert(g->nplus == 0 || m->lastpos != NULL); + for (;;) { + if (dp != NULL || endp <= m->coldp) + break; /* defeat */ + NOTE("backoff"); + endp = slow(m, m->coldp, endp-1, gf, gl); + if (endp == NULL) + break; /* defeat */ + /* try it on a shorter possibility */ +#ifndef NDEBUG + for (i = 1; i <= m->g->nsub; i++) { + assert(m->pmatch[i].rm_so == -1); + assert(m->pmatch[i].rm_eo == -1); + } +#endif + NOTE("backoff dissect"); + dp = backref(m, m->coldp, endp, gf, gl, (sopno)0); + } + assert(dp == NULL || dp == endp); + if (dp != NULL) /* found a shorter one */ + break; + + /* despite initial appearances, there is no match here */ + NOTE("false alarm"); + start = m->coldp + 1; /* recycle starting later */ + assert(start <= stop); + } + + /* fill in the details if requested */ + if (nmatch > 0) { + pmatch[0].rm_so = m->coldp - m->offp; + pmatch[0].rm_eo = endp - m->offp; + } + if (nmatch > 1) { + assert(m->pmatch != NULL); + for (i = 1; i < nmatch; i++) + if (i <= m->g->nsub) + pmatch[i] = m->pmatch[i]; + else { + pmatch[i].rm_so = -1; + pmatch[i].rm_eo = -1; + } + } + + if (m->pmatch != NULL) + free((char *)m->pmatch); + if (m->lastpos != NULL) + free((char *)m->lastpos); + STATETEARDOWN(m); + return(0); +} + +/* + - dissect - figure out what matched what, no back references + == static char *dissect(register struct match *m, char *start, \ + == char *stop, sopno startst, sopno stopst); + */ +static char * /* == stop (success) always */ +dissect(m, start, stop, startst, stopst) +register struct match *m; +char *start; +char *stop; +sopno startst; +sopno stopst; +{ + register int i; + register sopno ss; /* start sop of current subRE */ + register sopno es; /* end sop of current subRE */ + register char *sp; /* start of string matched by it */ + register char *stp; /* string matched by it cannot pass here */ + register char *rest; /* start of rest of string */ + register char *tail; /* string unmatched by rest of RE */ + register sopno ssub; /* start sop of subsubRE */ + register sopno esub; /* end sop of subsubRE */ + register char *ssp; /* start of string matched by subsubRE */ + register char *sep; /* end of string matched by subsubRE */ + register char *oldssp; /* previous ssp */ + register char *dp; + + AT("diss", start, stop, startst, stopst); + sp = start; + for (ss = startst; ss < stopst; ss = es) { + /* identify end of subRE */ + es = ss; + switch (OP(m->g->strip[es])) { + case OPLUS_: + case OQUEST_: + es += OPND(m->g->strip[es]); + break; + case OCH_: + while (OP(m->g->strip[es]) != O_CH) + es += OPND(m->g->strip[es]); + break; + } + es++; + + /* figure out what it matched */ + switch (OP(m->g->strip[ss])) { + case OEND: + assert(nope); + break; + case OCHAR: + sp++; + break; + case OBOL: + case OEOL: + case OBOW: + case OEOW: + break; + case OANY: + case OANYOF: + sp++; + break; + case OBACK_: + case O_BACK: + assert(nope); + break; + /* cases where length of match is hard to find */ + case OQUEST_: + stp = stop; + for (;;) { + /* how long could this one be? */ + rest = slow(m, sp, stp, ss, es); + assert(rest != NULL); /* it did match */ + /* could the rest match the rest? */ + tail = slow(m, rest, stop, es, stopst); + if (tail == stop) + break; /* yes! */ + /* no -- try a shorter match for this one */ + stp = rest - 1; + assert(stp >= sp); /* it did work */ + } + ssub = ss + 1; + esub = es - 1; + /* did innards match? */ + if (slow(m, sp, rest, ssub, esub) != NULL) { + dp = dissect(m, sp, rest, ssub, esub); + assert(dp == rest); + } else /* no */ + assert(sp == rest); + sp = rest; + break; + case OPLUS_: + stp = stop; + for (;;) { + /* how long could this one be? */ + rest = slow(m, sp, stp, ss, es); + assert(rest != NULL); /* it did match */ + /* could the rest match the rest? */ + tail = slow(m, rest, stop, es, stopst); + if (tail == stop) + break; /* yes! */ + /* no -- try a shorter match for this one */ + stp = rest - 1; + assert(stp >= sp); /* it did work */ + } + ssub = ss + 1; + esub = es - 1; + ssp = sp; + oldssp = ssp; + for (;;) { /* find last match of innards */ + sep = slow(m, ssp, rest, ssub, esub); + if (sep == NULL || sep == ssp) + break; /* failed or matched null */ + oldssp = ssp; /* on to next try */ + ssp = sep; + } + if (sep == NULL) { + /* last successful match */ + sep = ssp; + ssp = oldssp; + } + assert(sep == rest); /* must exhaust substring */ + assert(slow(m, ssp, sep, ssub, esub) == rest); + dp = dissect(m, ssp, sep, ssub, esub); + assert(dp == sep); + sp = rest; + break; + case OCH_: + stp = stop; + for (;;) { + /* how long could this one be? */ + rest = slow(m, sp, stp, ss, es); + assert(rest != NULL); /* it did match */ + /* could the rest match the rest? */ + tail = slow(m, rest, stop, es, stopst); + if (tail == stop) + break; /* yes! */ + /* no -- try a shorter match for this one */ + stp = rest - 1; + assert(stp >= sp); /* it did work */ + } + ssub = ss + 1; + esub = ss + OPND(m->g->strip[ss]) - 1; + assert(OP(m->g->strip[esub]) == OOR1); + for (;;) { /* find first matching branch */ + if (slow(m, sp, rest, ssub, esub) == rest) + break; /* it matched all of it */ + /* that one missed, try next one */ + assert(OP(m->g->strip[esub]) == OOR1); + esub++; + assert(OP(m->g->strip[esub]) == OOR2); + ssub = esub + 1; + esub += OPND(m->g->strip[esub]); + if (OP(m->g->strip[esub]) == OOR2) + esub--; + else + assert(OP(m->g->strip[esub]) == O_CH); + } + dp = dissect(m, sp, rest, ssub, esub); + assert(dp == rest); + sp = rest; + break; + case O_PLUS: + case O_QUEST: + case OOR1: + case OOR2: + case O_CH: + assert(nope); + break; + case OLPAREN: + i = OPND(m->g->strip[ss]); + assert(0 < i && i <= m->g->nsub); + m->pmatch[i].rm_so = sp - m->offp; + break; + case ORPAREN: + i = OPND(m->g->strip[ss]); + assert(0 < i && i <= m->g->nsub); + m->pmatch[i].rm_eo = sp - m->offp; + break; + default: /* uh oh */ + assert(nope); + break; + } + } + + assert(sp == stop); + return(sp); +} + +/* + - backref - figure out what matched what, figuring in back references + == static char *backref(register struct match *m, char *start, \ + == char *stop, sopno startst, sopno stopst, sopno lev); + */ +static char * /* == stop (success) or NULL (failure) */ +backref(m, start, stop, startst, stopst, lev) +register struct match *m; +char *start; +char *stop; +sopno startst; +sopno stopst; +sopno lev; /* PLUS nesting level */ +{ + register int i; + register sopno ss; /* start sop of current subRE */ + register char *sp; /* start of string matched by it */ + register sopno ssub; /* start sop of subsubRE */ + register sopno esub; /* end sop of subsubRE */ + register char *ssp; /* start of string matched by subsubRE */ + register char *dp; + register size_t len; + register int hard; + register sop s; + register regoff_t offsave; + register cset *cs; + + AT("back", start, stop, startst, stopst); + sp = start; + + /* get as far as we can with easy stuff */ + hard = 0; + for (ss = startst; !hard && ss < stopst; ss++) + switch (OP(s = m->g->strip[ss])) { + case OCHAR: + if (sp == stop || *sp++ != (char)OPND(s)) + return(NULL); + break; + case OANY: + if (sp == stop) + return(NULL); + sp++; + break; + case OANYOF: + cs = &m->g->sets[OPND(s)]; + if (sp == stop || !CHIN(cs, *sp++)) + return(NULL); + break; + case OBOL: + if ( (sp == m->beginp && !(m->eflags®_NOTBOL)) || + (sp < m->endp && *(sp-1) == '\n' && + (m->g->cflags®_NEWLINE)) ) + { /* yes */ } + else + return(NULL); + break; + case OEOL: + if ( (sp == m->endp && !(m->eflags®_NOTEOL)) || + (sp < m->endp && *sp == '\n' && + (m->g->cflags®_NEWLINE)) ) + { /* yes */ } + else + return(NULL); + break; + case OBOW: + if (( (sp == m->beginp && !(m->eflags®_NOTBOL)) || + (sp < m->endp && *(sp-1) == '\n' && + (m->g->cflags®_NEWLINE)) || + (sp > m->beginp && + !ISWORD(*(sp-1))) ) && + (sp < m->endp && ISWORD(*sp)) ) + { /* yes */ } + else + return(NULL); + break; + case OEOW: + if (( (sp == m->endp && !(m->eflags®_NOTEOL)) || + (sp < m->endp && *sp == '\n' && + (m->g->cflags®_NEWLINE)) || + (sp < m->endp && !ISWORD(*sp)) ) && + (sp > m->beginp && ISWORD(*(sp-1))) ) + { /* yes */ } + else + return(NULL); + break; + case O_QUEST: + break; + case OOR1: /* matches null but needs to skip */ + ss++; + s = m->g->strip[ss]; + do { + assert(OP(s) == OOR2); + ss += OPND(s); + } while (OP(s = m->g->strip[ss]) != O_CH); + /* note that the ss++ gets us past the O_CH */ + break; + default: /* have to make a choice */ + hard = 1; + break; + } + if (!hard) { /* that was it! */ + if (sp != stop) + return(NULL); + return(sp); + } + ss--; /* adjust for the for's final increment */ + + /* the hard stuff */ + AT("hard", sp, stop, ss, stopst); + s = m->g->strip[ss]; + switch (OP(s)) { + case OBACK_: /* the vilest depths */ + i = OPND(s); + assert(0 < i && i <= m->g->nsub); + if (m->pmatch[i].rm_eo == -1) + return(NULL); + assert(m->pmatch[i].rm_so != -1); + len = m->pmatch[i].rm_eo - m->pmatch[i].rm_so; + assert(stop - m->beginp >= len); + if (sp > stop - len) + return(NULL); /* not enough left to match */ + ssp = m->offp + m->pmatch[i].rm_so; + if (memcmp(sp, ssp, len) != 0) + return(NULL); + while (m->g->strip[ss] != SOP(O_BACK, i)) + ss++; + return(backref(m, sp+len, stop, ss+1, stopst, lev)); + break; + case OQUEST_: /* to null or not */ + dp = backref(m, sp, stop, ss+1, stopst, lev); + if (dp != NULL) + return(dp); /* not */ + return(backref(m, sp, stop, ss+OPND(s)+1, stopst, lev)); + break; + case OPLUS_: + assert(m->lastpos != NULL); + assert(lev+1 <= m->g->nplus); + m->lastpos[lev+1] = sp; + return(backref(m, sp, stop, ss+1, stopst, lev+1)); + break; + case O_PLUS: + if (sp == m->lastpos[lev]) /* last pass matched null */ + return(backref(m, sp, stop, ss+1, stopst, lev-1)); + /* try another pass */ + m->lastpos[lev] = sp; + dp = backref(m, sp, stop, ss-OPND(s)+1, stopst, lev); + if (dp == NULL) + return(backref(m, sp, stop, ss+1, stopst, lev-1)); + else + return(dp); + break; + case OCH_: /* find the right one, if any */ + ssub = ss + 1; + esub = ss + OPND(s) - 1; + assert(OP(m->g->strip[esub]) == OOR1); + for (;;) { /* find first matching branch */ + dp = backref(m, sp, stop, ssub, esub, lev); + if (dp != NULL) + return(dp); + /* that one missed, try next one */ + if (OP(m->g->strip[esub]) == O_CH) + return(NULL); /* there is none */ + esub++; + assert(OP(m->g->strip[esub]) == OOR2); + ssub = esub + 1; + esub += OPND(m->g->strip[esub]); + if (OP(m->g->strip[esub]) == OOR2) + esub--; + else + assert(OP(m->g->strip[esub]) == O_CH); + } + break; + case OLPAREN: /* must undo assignment if rest fails */ + i = OPND(s); + assert(0 < i && i <= m->g->nsub); + offsave = m->pmatch[i].rm_so; + m->pmatch[i].rm_so = sp - m->offp; + dp = backref(m, sp, stop, ss+1, stopst, lev); + if (dp != NULL) + return(dp); + m->pmatch[i].rm_so = offsave; + return(NULL); + break; + case ORPAREN: /* must undo assignment if rest fails */ + i = OPND(s); + assert(0 < i && i <= m->g->nsub); + offsave = m->pmatch[i].rm_eo; + m->pmatch[i].rm_eo = sp - m->offp; + dp = backref(m, sp, stop, ss+1, stopst, lev); + if (dp != NULL) + return(dp); + m->pmatch[i].rm_eo = offsave; + return(NULL); + break; + default: /* uh oh */ + assert(nope); + break; + } + + /* "can't happen" */ + assert(nope); + /* NOTREACHED */ + return((char *)NULL); /* dummy */ +} + +/* + - fast - step through the string at top speed + == static char *fast(register struct match *m, char *start, \ + == char *stop, sopno startst, sopno stopst); + */ +static char * /* where tentative match ended, or NULL */ +fast(m, start, stop, startst, stopst) +register struct match *m; +char *start; +char *stop; +sopno startst; +sopno stopst; +{ + register states st = m->st; + register states fresh = m->fresh; + register states tmp = m->tmp; + register char *p = start; + register int c = (start == m->beginp) ? OUT : *(start-1); + register int lastc; /* previous c */ + register int flagch; + register int i; + register char *coldp; /* last p after which no match was underway */ + + CLEAR(st); + SET1(st, startst); + st = step(m->g, startst, stopst, st, NOTHING, st); + ASSIGN(fresh, st); + SP("start", st, *p); + coldp = NULL; + for (;;) { + /* next character */ + lastc = c; + c = (p == m->endp) ? OUT : *p; + if (EQ(st, fresh)) + coldp = p; + + /* is there an EOL and/or BOL between lastc and c? */ + flagch = '\0'; + i = 0; + if ( (lastc == '\n' && m->g->cflags®_NEWLINE) || + (lastc == OUT && !(m->eflags®_NOTBOL)) ) { + flagch = BOL; + i = m->g->nbol; + } + if ( (c == '\n' && m->g->cflags®_NEWLINE) || + (c == OUT && !(m->eflags®_NOTEOL)) ) { + flagch = (flagch == BOL) ? BOLEOL : EOL; + i += m->g->neol; + } + if (i != 0) { + for (; i > 0; i--) + st = step(m->g, startst, stopst, st, flagch, st); + SP("boleol", st, c); + } + + /* how about a word boundary? */ + if ( (flagch == BOL || (lastc != OUT && !ISWORD(lastc))) && + (c != OUT && ISWORD(c)) ) { + flagch = BOW; + } + if ( (lastc != OUT && ISWORD(lastc)) && + (flagch == EOL || (c != OUT && !ISWORD(c))) ) { + flagch = EOW; + } + if (flagch == BOW || flagch == EOW) { + st = step(m->g, startst, stopst, st, flagch, st); + SP("boweow", st, c); + } + + /* are we done? */ + if (ISSET(st, stopst) || p == stop) + break; /* NOTE BREAK OUT */ + + /* no, we must deal with this character */ + ASSIGN(tmp, st); + ASSIGN(st, fresh); + assert(c != OUT); + st = step(m->g, startst, stopst, tmp, c, st); + SP("aft", st, c); + assert(EQ(step(m->g, startst, stopst, st, NOTHING, st), st)); + p++; + } + + assert(coldp != NULL); + m->coldp = coldp; + if (ISSET(st, stopst)) + return(p+1); + else + return(NULL); +} + +/* + - slow - step through the string more deliberately + == static char *slow(register struct match *m, char *start, \ + == char *stop, sopno startst, sopno stopst); + */ +static char * /* where it ended */ +slow(m, start, stop, startst, stopst) +register struct match *m; +char *start; +char *stop; +sopno startst; +sopno stopst; +{ + register states st = m->st; + register states empty = m->empty; + register states tmp = m->tmp; + register char *p = start; + register int c = (start == m->beginp) ? OUT : *(start-1); + register int lastc; /* previous c */ + register int flagch; + register int i; + register char *matchp; /* last p at which a match ended */ + + AT("slow", start, stop, startst, stopst); + CLEAR(st); + SET1(st, startst); + SP("sstart", st, *p); + st = step(m->g, startst, stopst, st, NOTHING, st); + matchp = NULL; + for (;;) { + /* next character */ + lastc = c; + c = (p == m->endp) ? OUT : *p; + + /* is there an EOL and/or BOL between lastc and c? */ + flagch = '\0'; + i = 0; + if ( (lastc == '\n' && m->g->cflags®_NEWLINE) || + (lastc == OUT && !(m->eflags®_NOTBOL)) ) { + flagch = BOL; + i = m->g->nbol; + } + if ( (c == '\n' && m->g->cflags®_NEWLINE) || + (c == OUT && !(m->eflags®_NOTEOL)) ) { + flagch = (flagch == BOL) ? BOLEOL : EOL; + i += m->g->neol; + } + if (i != 0) { + for (; i > 0; i--) + st = step(m->g, startst, stopst, st, flagch, st); + SP("sboleol", st, c); + } + + /* how about a word boundary? */ + if ( (flagch == BOL || (lastc != OUT && !ISWORD(lastc))) && + (c != OUT && ISWORD(c)) ) { + flagch = BOW; + } + if ( (lastc != OUT && ISWORD(lastc)) && + (flagch == EOL || (c != OUT && !ISWORD(c))) ) { + flagch = EOW; + } + if (flagch == BOW || flagch == EOW) { + st = step(m->g, startst, stopst, st, flagch, st); + SP("sboweow", st, c); + } + + /* are we done? */ + if (ISSET(st, stopst)) + matchp = p; + if (EQ(st, empty) || p == stop) + break; /* NOTE BREAK OUT */ + + /* no, we must deal with this character */ + ASSIGN(tmp, st); + ASSIGN(st, empty); + assert(c != OUT); + st = step(m->g, startst, stopst, tmp, c, st); + SP("saft", st, c); + assert(EQ(step(m->g, startst, stopst, st, NOTHING, st), st)); + p++; + } + + return(matchp); +} + + +/* + - step - map set of states reachable before char to set reachable after + == static states step(register struct re_guts *g, sopno start, sopno stop, \ + == register states bef, int ch, register states aft); + == #define BOL (OUT+1) + == #define EOL (BOL+1) + == #define BOLEOL (BOL+2) + == #define NOTHING (BOL+3) + == #define BOW (BOL+4) + == #define EOW (BOL+5) + == #define CODEMAX (BOL+5) // highest code used + == #define NONCHAR(c) ((c) > CHAR_MAX) + == #define NNONCHAR (CODEMAX-CHAR_MAX) + */ +static states +step(g, start, stop, bef, ch, aft) +register struct re_guts *g; +sopno start; /* start state within strip */ +sopno stop; /* state after stop state within strip */ +register states bef; /* states reachable before */ +int ch; /* character or NONCHAR code */ +register states aft; /* states already known reachable after */ +{ + register cset *cs; + register sop s; + register sopno pc; + register onestate here; /* note, macros know this name */ + register sopno look; + register long i; + + for (pc = start, INIT(here, pc); pc != stop; pc++, INC(here)) { + s = g->strip[pc]; + switch (OP(s)) { + case OEND: + assert(pc == stop-1); + break; + case OCHAR: + /* only characters can match */ + assert(!NONCHAR(ch) || ch != (char)OPND(s)); + if (ch == (char)OPND(s)) + FWD(aft, bef, 1); + break; + case OBOL: + if (ch == BOL || ch == BOLEOL) + FWD(aft, bef, 1); + break; + case OEOL: + if (ch == EOL || ch == BOLEOL) + FWD(aft, bef, 1); + break; + case OBOW: + if (ch == BOW) + FWD(aft, bef, 1); + break; + case OEOW: + if (ch == EOW) + FWD(aft, bef, 1); + break; + case OANY: + if (!NONCHAR(ch)) + FWD(aft, bef, 1); + break; + case OANYOF: + cs = &g->sets[OPND(s)]; + if (!NONCHAR(ch) && CHIN(cs, ch)) + FWD(aft, bef, 1); + break; + case OBACK_: /* ignored here */ + case O_BACK: + FWD(aft, aft, 1); + break; + case OPLUS_: /* forward, this is just an empty */ + FWD(aft, aft, 1); + break; + case O_PLUS: /* both forward and back */ + FWD(aft, aft, 1); + i = ISSETBACK(aft, OPND(s)); + BACK(aft, aft, OPND(s)); + if (!i && ISSETBACK(aft, OPND(s))) { + /* oho, must reconsider loop body */ + pc -= OPND(s) + 1; + INIT(here, pc); + } + break; + case OQUEST_: /* two branches, both forward */ + FWD(aft, aft, 1); + FWD(aft, aft, OPND(s)); + break; + case O_QUEST: /* just an empty */ + FWD(aft, aft, 1); + break; + case OLPAREN: /* not significant here */ + case ORPAREN: + FWD(aft, aft, 1); + break; + case OCH_: /* mark the first two branches */ + FWD(aft, aft, 1); + assert(OP(g->strip[pc+OPND(s)]) == OOR2); + FWD(aft, aft, OPND(s)); + break; + case OOR1: /* done a branch, find the O_CH */ + if (ISSTATEIN(aft, here)) { + for (look = 1; + OP(s = g->strip[pc+look]) != O_CH; + look += OPND(s)) + assert(OP(s) == OOR2); + FWD(aft, aft, look); + } + break; + case OOR2: /* propagate OCH_'s marking */ + FWD(aft, aft, 1); + if (OP(g->strip[pc+OPND(s)]) != O_CH) { + assert(OP(g->strip[pc+OPND(s)]) == OOR2); + FWD(aft, aft, OPND(s)); + } + break; + case O_CH: /* just empty */ + FWD(aft, aft, 1); + break; + default: /* ooooops... */ + assert(nope); + break; + } + } + + return(aft); +} + +#ifdef REDEBUG +/* + - print - print a set of states + == #ifdef REDEBUG + == static void print(struct match *m, char *caption, states st, \ + == int ch, FILE *d); + == #endif + */ +static void +print(m, caption, st, ch, d) +struct match *m; +char *caption; +states st; +int ch; +FILE *d; +{ + register struct re_guts *g = m->g; + register int i; + register int first = 1; + + if (!(m->eflags®_TRACE)) + return; + + fprintf(d, "%s", caption); + if (ch != '\0') + fprintf(d, " %s", pchar(ch)); + for (i = 0; i < g->nstates; i++) + if (ISSET(st, i)) { + fprintf(d, "%s%d", (first) ? "\t" : ", ", i); + first = 0; + } + fprintf(d, "\n"); +} + +/* + - at - print current situation + == #ifdef REDEBUG + == static void at(struct match *m, char *title, char *start, char *stop, \ + == sopno startst, sopno stopst); + == #endif + */ +static void +at(m, title, start, stop, startst, stopst) +struct match *m; +char *title; +char *start; +char *stop; +sopno startst; +sopno stopst; +{ + if (!(m->eflags®_TRACE)) + return; + + printf("%s %s-", title, pchar(*start)); + printf("%s ", pchar(*stop)); + printf("%ld-%ld\n", (long)startst, (long)stopst); +} + +#ifndef PCHARDONE +#define PCHARDONE /* never again */ +/* + - pchar - make a character printable + == #ifdef REDEBUG + == static char *pchar(int ch); + == #endif + * + * Is this identical to regchar() over in debug.c? Well, yes. But a + * duplicate here avoids having a debugging-capable regexec.o tied to + * a matching debug.o, and this is convenient. It all disappears in + * the non-debug compilation anyway, so it doesn't matter much. + */ +static char * /* -> representation */ +pchar(ch) +int ch; +{ + static char pbuf[10]; + + if (isprint(ch) || ch == ' ') + sprintf(pbuf, "%c", ch); + else + sprintf(pbuf, "\\%o", ch); + return(pbuf); +} +#endif +#endif + +#undef matcher +#undef fast +#undef slow +#undef dissect +#undef backref +#undef step +#undef print +#undef at +#undef match diff --git a/winsup/cygwin/regex/engine.ih b/winsup/cygwin/regex/engine.ih new file mode 100644 index 000000000..cc98334e7 --- /dev/null +++ b/winsup/cygwin/regex/engine.ih @@ -0,0 +1,35 @@ +/* ========= begin header generated by ./mkh ========= */ +#ifdef __cplusplus +extern "C" { +#endif + +/* === engine.c === */ +static int matcher(register struct re_guts *g, char *string, size_t nmatch, regmatch_t pmatch[], int eflags); +static char *dissect(register struct match *m, char *start, char *stop, sopno startst, sopno stopst); +static char *backref(register struct match *m, char *start, char *stop, sopno startst, sopno stopst, sopno lev); +static char *fast(register struct match *m, char *start, char *stop, sopno startst, sopno stopst); +static char *slow(register struct match *m, char *start, char *stop, sopno startst, sopno stopst); +static states step(register struct re_guts *g, sopno start, sopno stop, register states bef, int ch, register states aft); +#define BOL (OUT+1) +#define EOL (BOL+1) +#define BOLEOL (BOL+2) +#define NOTHING (BOL+3) +#define BOW (BOL+4) +#define EOW (BOL+5) +#define CODEMAX (BOL+5) /* highest code used */ +#define NONCHAR(c) ((c) > CHAR_MAX) +#define NNONCHAR (CODEMAX-CHAR_MAX) +#ifdef REDEBUG +static void print(struct match *m, char *caption, states st, int ch, FILE *d); +#endif +#ifdef REDEBUG +static void at(struct match *m, char *title, char *start, char *stop, sopno startst, sopno stopst); +#endif +#ifdef REDEBUG +static char *pchar(int ch); +#endif + +#ifdef __cplusplus +} +#endif +/* ========= end header generated by ./mkh ========= */ diff --git a/winsup/cygwin/regex/mkh b/winsup/cygwin/regex/mkh new file mode 100755 index 000000000..252b246c7 --- /dev/null +++ b/winsup/cygwin/regex/mkh @@ -0,0 +1,76 @@ +#! /bin/sh +# mkh - pull headers out of C source +PATH=/bin:/usr/bin ; export PATH + +# egrep pattern to pick out marked lines +egrep='^ =([ ]|$)' + +# Sed program to process marked lines into lines for the header file. +# The markers have already been removed. Two things are done here: removal +# of backslashed newlines, and some fudging of comments. The first is done +# because -o needs to have prototypes on one line to strip them down. +# Getting comments into the output is tricky; we turn C++-style // comments +# into /* */ comments, after altering any existing */'s to avoid trouble. +peel=' /\\$/N + /\\\n[ ]*/s///g + /\/\//s;\*/;* /;g + /\/\//s;//\(.*\);/*\1 */;' + +for a +do + case "$a" in + -o) # old (pre-function-prototype) compiler + # add code to comment out argument lists + peel="$peel + "'/^\([^#\/][^\/]*[a-zA-Z0-9_)]\)(\(.*\))/s;;\1(/*\2*/);' + shift + ;; + -b) # funny Berkeley __P macro + peel="$peel + "'/^\([^#\/][^\/]*[a-zA-Z0-9_)]\)(\(.*\))/s;;\1 __P((\2));' + shift + ;; + -s) # compiler doesn't like `static foo();' + # add code to get rid of the `static' + peel="$peel + "'/^static[ ][^\/]*[a-zA-Z0-9_)](.*)/s;static.;;' + shift + ;; + -p) # private declarations + egrep='^ ==([ ]|$)' + shift + ;; + -i) # wrap in #ifndef, argument is name + ifndef="$2" + shift ; shift + ;; + *) break + ;; + esac +done + +if test " $ifndef" != " " +then + echo "#ifndef $ifndef" + echo "#define $ifndef /* never again */" +fi +echo "/* ========= begin header generated by $0 ========= */" +echo '#ifdef __cplusplus' +echo 'extern "C" {' +echo '#endif' +for f +do + echo + echo "/* === $f === */" + egrep "$egrep" $f | sed 's/^ ==*[ ]//;s/^ ==*$//' | sed "$peel" + echo +done +echo '#ifdef __cplusplus' +echo '}' +echo '#endif' +echo "/* ========= end header generated by $0 ========= */" +if test " $ifndef" != " " +then + echo "#endif" +fi +exit 0 diff --git a/winsup/cygwin/regex/regcomp.c b/winsup/cygwin/regex/regcomp.c new file mode 100644 index 000000000..34c1f2e5c --- /dev/null +++ b/winsup/cygwin/regex/regcomp.c @@ -0,0 +1,1546 @@ +#include "winsup.h" +#include <sys/types.h> +#include <stdio.h> +#include <string.h> +#include <ctype.h> +#include <limits.h> +#include <stdlib.h> +#include "regex.h" + +#include "utils.h" +#include "regex2.h" + +#include "cclass.h" +#include "cname.h" + +/* + * parse structure, passed up and down to avoid global variables and + * other clumsinesses + */ +struct parse { + char *next; /* next character in RE */ + char *end; /* end of string (-> NUL normally) */ + int error; /* has an error been seen? */ + sop *strip; /* malloced strip */ + sopno ssize; /* malloced strip size (allocated) */ + sopno slen; /* malloced strip length (used) */ + int ncsalloc; /* number of csets allocated */ + struct re_guts *g; +# define NPAREN 10 /* we need to remember () 1-9 for back refs */ + sopno pbegin[NPAREN]; /* -> ( ([0] unused) */ + sopno pend[NPAREN]; /* -> ) ([0] unused) */ +}; + +#include "regcomp.ih" + +static char nuls[10]; /* place to point scanner in event of error */ + +/* + * macros for use with parse structure + * BEWARE: these know that the parse structure is named `p' !!! + */ +#define PEEK() (*p->next) +#define PEEK2() (*(p->next+1)) +#define MORE() (p->next < p->end) +#define MORE2() (p->next+1 < p->end) +#define SEE(c) (MORE() && PEEK() == (c)) +#define SEETWO(a, b) (MORE() && MORE2() && PEEK() == (a) && PEEK2() == (b)) +#define EAT(c) ((SEE(c)) ? (NEXT(), 1) : 0) +#define EATTWO(a, b) ((SEETWO(a, b)) ? (NEXT2(), 1) : 0) +#define NEXT() (p->next++) +#define NEXT2() (p->next += 2) +#define NEXTn(n) (p->next += (n)) +#define GETNEXT() (*p->next++) +#define SETERROR(e) seterr(p, (e)) +#define REQUIRE(co, e) ((co) || SETERROR(e)) +#define MUSTSEE(c, e) (REQUIRE(MORE() && PEEK() == (c), e)) +#define MUSTEAT(c, e) (REQUIRE(MORE() && GETNEXT() == (c), e)) +#define MUSTNOTSEE(c, e) (REQUIRE(!MORE() || PEEK() != (c), e)) +#define EMIT(op, sopnd) doemit(p, (sop)(op), (size_t)(sopnd)) +#define INSERT(op, pos) doinsert(p, (sop)(op), HERE()-(pos)+1, pos) +#define AHEAD(pos) dofwd(p, pos, HERE()-(pos)) +#define ASTERN(sop, pos) EMIT(sop, HERE()-pos) +#define HERE() (p->slen) +#define THERE() (p->slen - 1) +#define THERETHERE() (p->slen - 2) +#define DROP(n) (p->slen -= (n)) + +#ifndef NDEBUG +static int never = 0; /* for use in asserts; shuts lint up */ +#else +#define never 0 /* some <assert.h>s have bugs too */ +#endif + +/* + - regcomp - interface for parser and compilation + = extern int regcomp(regex_t *, const char *, int); + = #define REG_BASIC 0000 + = #define REG_EXTENDED 0001 + = #define REG_ICASE 0002 + = #define REG_NOSUB 0004 + = #define REG_NEWLINE 0010 + = #define REG_NOSPEC 0020 + = #define REG_PEND 0040 + = #define REG_DUMP 0200 + */ +int /* 0 success, otherwise REG_something */ +regcomp(preg, pattern, cflags) +regex_t *preg; +const char *pattern; +int cflags; +{ + struct parse pa; + register struct re_guts *g; + register struct parse *p = &pa; + register int i; + register size_t len; +#ifdef REDEBUG +# define GOODFLAGS(f) (f) +#else +# define GOODFLAGS(f) ((f)&~REG_DUMP) +#endif + + cflags = GOODFLAGS(cflags); + if ((cflags®_EXTENDED) && (cflags®_NOSPEC)) + return(REG_INVARG); + + if (cflags®_PEND) { + if (preg->re_endp < pattern) + return(REG_INVARG); + len = preg->re_endp - pattern; + } else + len = strlen((char *)pattern); + + /* do the mallocs early so failure handling is easy */ + g = (struct re_guts *)malloc(sizeof(struct re_guts) + + (NC-1)*sizeof(cat_t)); + if (g == NULL) + return(REG_ESPACE); + p->ssize = len/(size_t)2*(size_t)3 + (size_t)1; /* ugh */ + p->strip = (sop *)malloc(p->ssize * sizeof(sop)); + p->slen = 0; + if (p->strip == NULL) { + free((char *)g); + return(REG_ESPACE); + } + + /* set things up */ + p->g = g; + p->next = (char *)pattern; /* convenience; we do not modify it */ + p->end = p->next + len; + p->error = 0; + p->ncsalloc = 0; + for (i = 0; i < NPAREN; i++) { + p->pbegin[i] = 0; + p->pend[i] = 0; + } + g->csetsize = NC; + g->sets = NULL; + g->setbits = NULL; + g->ncsets = 0; + g->cflags = cflags; + g->iflags = 0; + g->nbol = 0; + g->neol = 0; + g->must = NULL; + g->mlen = 0; + g->nsub = 0; + g->ncategories = 1; /* category 0 is "everything else" */ + g->categories = &g->catspace[-(CHAR_MIN)]; + (void) memset((char *)g->catspace, 0, NC*sizeof(cat_t)); + g->backrefs = 0; + + /* do it */ + EMIT(OEND, 0); + g->firststate = THERE(); + if (cflags®_EXTENDED) + p_ere(p, OUT); + else if (cflags®_NOSPEC) + p_str(p); + else + p_bre(p, OUT, OUT); + EMIT(OEND, 0); + g->laststate = THERE(); + + /* tidy up loose ends and fill things in */ + categorize(p, g); + stripsnug(p, g); + findmust(p, g); + g->nplus = pluscount(p, g); + g->magic = MAGIC2; + preg->re_nsub = g->nsub; + preg->re_g = g; + preg->re_magic = MAGIC1; +#ifndef REDEBUG + /* not debugging, so can't rely on the assert() in regexec() */ + if (g->iflags&BAD) + SETERROR(REG_ASSERT); +#endif + + /* win or lose, we're done */ + if (p->error != 0) /* lose */ + regfree(preg); + return(p->error); +} + +/* + - p_ere - ERE parser top level, concatenation and alternation + == static void p_ere(register struct parse *p, int stop); + */ +static void +p_ere(p, stop) +register struct parse *p; +int stop; /* character this ERE should end at */ +{ + register char c; + register sopno prevback = 0; + register sopno prevfwd = 0; + register sopno conc = 0; + register int first = 1; /* is this the first alternative? */ + + for (;;) { + /* do a bunch of concatenated expressions */ + conc = HERE(); + while (MORE() && (c = PEEK()) != '|' && c != stop) + p_ere_exp(p); + REQUIRE(HERE() != conc, REG_EMPTY); /* require nonempty */ + + if (!EAT('|')) + break; /* NOTE BREAK OUT */ + + if (first) { + INSERT(OCH_, conc); /* offset is wrong */ + prevfwd = conc; + prevback = conc; + first = 0; + } + ASTERN(OOR1, prevback); + prevback = THERE(); + AHEAD(prevfwd); /* fix previous offset */ + prevfwd = HERE(); + EMIT(OOR2, 0); /* offset is very wrong */ + } + + if (!first) { /* tail-end fixups */ + AHEAD(prevfwd); + ASTERN(O_CH, prevback); + } + + assert(!MORE() || SEE(stop)); +} + +/* + - p_ere_exp - parse one subERE, an atom possibly followed by a repetition op + == static void p_ere_exp(register struct parse *p); + */ +static void +p_ere_exp(p) +register struct parse *p; +{ + register char c; + register sopno pos; + register int count; + register int count2; + register sopno subno; + int wascaret = 0; + + assert(MORE()); /* caller should have ensured this */ + c = GETNEXT(); + + pos = HERE(); + switch (c) { + case '(': + REQUIRE(MORE(), REG_EPAREN); + p->g->nsub++; + subno = p->g->nsub; + if (subno < NPAREN) + p->pbegin[subno] = HERE(); + EMIT(OLPAREN, subno); + if (!SEE(')')) + p_ere(p, ')'); + if (subno < NPAREN) { + p->pend[subno] = HERE(); + assert(p->pend[subno] != 0); + } + EMIT(ORPAREN, subno); + MUSTEAT(')', REG_EPAREN); + break; +#ifndef POSIX_MISTAKE + case ')': /* happens only if no current unmatched ( */ + /* + * You may ask, why the ifndef? Because I didn't notice + * this until slightly too late for 1003.2, and none of the + * other 1003.2 regular-expression reviewers noticed it at + * all. So an unmatched ) is legal POSIX, at least until + * we can get it fixed. + */ + SETERROR(REG_EPAREN); + break; +#endif + case '^': + EMIT(OBOL, 0); + p->g->iflags |= USEBOL; + p->g->nbol++; + wascaret = 1; + break; + case '$': + EMIT(OEOL, 0); + p->g->iflags |= USEEOL; + p->g->neol++; + break; + case '|': + SETERROR(REG_EMPTY); + break; + case '*': + case '+': + case '?': + SETERROR(REG_BADRPT); + break; + case '.': + if (p->g->cflags®_NEWLINE) + nonnewline(p); + else + EMIT(OANY, 0); + break; + case '[': + p_bracket(p); + break; + case '\\': + REQUIRE(MORE(), REG_EESCAPE); + c = GETNEXT(); + ordinary(p, c); + break; + case '{': /* okay as ordinary except if digit follows */ + REQUIRE(!MORE() || !isdigit(PEEK()), REG_BADRPT); + /* FALLTHROUGH */ + default: + ordinary(p, c); + break; + } + + if (!MORE()) + return; + c = PEEK(); + /* we call { a repetition if followed by a digit */ + if (!( c == '*' || c == '+' || c == '?' || + (c == '{' && MORE2() && isdigit(PEEK2())) )) + return; /* no repetition, we're done */ + NEXT(); + + REQUIRE(!wascaret, REG_BADRPT); + switch (c) { + case '*': /* implemented as +? */ + /* this case does not require the (y|) trick, noKLUDGE */ + INSERT(OPLUS_, pos); + ASTERN(O_PLUS, pos); + INSERT(OQUEST_, pos); + ASTERN(O_QUEST, pos); + break; + case '+': + INSERT(OPLUS_, pos); + ASTERN(O_PLUS, pos); + break; + case '?': + /* KLUDGE: emit y? as (y|) until subtle bug gets fixed */ + INSERT(OCH_, pos); /* offset slightly wrong */ + ASTERN(OOR1, pos); /* this one's right */ + AHEAD(pos); /* fix the OCH_ */ + EMIT(OOR2, 0); /* offset very wrong... */ + AHEAD(THERE()); /* ...so fix it */ + ASTERN(O_CH, THERETHERE()); + break; + case '{': + count = p_count(p); + if (EAT(',')) { + if (isdigit(PEEK())) { + count2 = p_count(p); + REQUIRE(count <= count2, REG_BADBR); + } else /* single number with comma */ + count2 = INFINITY; + } else /* just a single number */ + count2 = count; + repeat(p, pos, count, count2); + if (!EAT('}')) { /* error heuristics */ + while (MORE() && PEEK() != '}') + NEXT(); + REQUIRE(MORE(), REG_EBRACE); + SETERROR(REG_BADBR); + } + break; + } + + if (!MORE()) + return; + c = PEEK(); + if (!( c == '*' || c == '+' || c == '?' || + (c == '{' && MORE2() && isdigit(PEEK2())) ) ) + return; + SETERROR(REG_BADRPT); +} + +/* + - p_str - string (no metacharacters) "parser" + == static void p_str(register struct parse *p); + */ +static void +p_str(p) +register struct parse *p; +{ + REQUIRE(MORE(), REG_EMPTY); + while (MORE()) + ordinary(p, GETNEXT()); +} + +/* + - p_bre - BRE parser top level, anchoring and concatenation + == static void p_bre(register struct parse *p, register int end1, \ + == register int end2); + * Giving end1 as OUT essentially eliminates the end1/end2 check. + * + * This implementation is a bit of a kludge, in that a trailing $ is first + * taken as an ordinary character and then revised to be an anchor. The + * only undesirable side effect is that '$' gets included as a character + * category in such cases. This is fairly harmless; not worth fixing. + * The amount of lookahead needed to avoid this kludge is excessive. + */ +static void +p_bre(p, end1, end2) +register struct parse *p; +register int end1; /* first terminating character */ +register int end2; /* second terminating character */ +{ + register sopno start = HERE(); + register int first = 1; /* first subexpression? */ + register int wasdollar = 0; + + if (EAT('^')) { + EMIT(OBOL, 0); + p->g->iflags |= USEBOL; + p->g->nbol++; + } + while (MORE() && !SEETWO(end1, end2)) { + wasdollar = p_simp_re(p, first); + first = 0; + } + if (wasdollar) { /* oops, that was a trailing anchor */ + DROP(1); + EMIT(OEOL, 0); + p->g->iflags |= USEEOL; + p->g->neol++; + } + + REQUIRE(HERE() != start, REG_EMPTY); /* require nonempty */ +} + +/* + - p_simp_re - parse a simple RE, an atom possibly followed by a repetition + == static int p_simp_re(register struct parse *p, int starordinary); + */ +static int /* was the simple RE an unbackslashed $? */ +p_simp_re(p, starordinary) +register struct parse *p; +int starordinary; /* is a leading * an ordinary character? */ +{ + register int c; + register int count; + register int count2; + register sopno pos; + register int i; + register sopno subno; +# define BACKSL (1<<CHAR_BIT) + + pos = HERE(); /* repetion op, if any, covers from here */ + + assert(MORE()); /* caller should have ensured this */ + c = GETNEXT(); + if (c == '\\') { + REQUIRE(MORE(), REG_EESCAPE); + c = BACKSL | (unsigned char)GETNEXT(); + } + switch (c) { + case '.': + if (p->g->cflags®_NEWLINE) + nonnewline(p); + else + EMIT(OANY, 0); + break; + case '[': + p_bracket(p); + break; + case BACKSL|'{': + SETERROR(REG_BADRPT); + break; + case BACKSL|'(': + p->g->nsub++; + subno = p->g->nsub; + if (subno < NPAREN) + p->pbegin[subno] = HERE(); + EMIT(OLPAREN, subno); + /* the MORE here is an error heuristic */ + if (MORE() && !SEETWO('\\', ')')) + p_bre(p, '\\', ')'); + if (subno < NPAREN) { + p->pend[subno] = HERE(); + assert(p->pend[subno] != 0); + } + EMIT(ORPAREN, subno); + REQUIRE(EATTWO('\\', ')'), REG_EPAREN); + break; + case BACKSL|')': /* should not get here -- must be user */ + case BACKSL|'}': + SETERROR(REG_EPAREN); + break; + case BACKSL|'1': + case BACKSL|'2': + case BACKSL|'3': + case BACKSL|'4': + case BACKSL|'5': + case BACKSL|'6': + case BACKSL|'7': + case BACKSL|'8': + case BACKSL|'9': + i = (c&~BACKSL) - '0'; + assert(i < NPAREN); + if (p->pend[i] != 0) { + assert(i <= p->g->nsub); + EMIT(OBACK_, i); + assert(p->pbegin[i] != 0); + assert(OP(p->strip[p->pbegin[i]]) == OLPAREN); + assert(OP(p->strip[p->pend[i]]) == ORPAREN); + (void) dupl(p, p->pbegin[i]+1, p->pend[i]); + EMIT(O_BACK, i); + } else + SETERROR(REG_ESUBREG); + p->g->backrefs = 1; + break; + case '*': + REQUIRE(starordinary, REG_BADRPT); + /* FALLTHROUGH */ + default: + ordinary(p, (char)c); /* takes off BACKSL, if any */ + break; + } + + if (EAT('*')) { /* implemented as +? */ + /* this case does not require the (y|) trick, noKLUDGE */ + INSERT(OPLUS_, pos); + ASTERN(O_PLUS, pos); + INSERT(OQUEST_, pos); + ASTERN(O_QUEST, pos); + } else if (EATTWO('\\', '{')) { + count = p_count(p); + if (EAT(',')) { + if (MORE() && isdigit(PEEK())) { + count2 = p_count(p); + REQUIRE(count <= count2, REG_BADBR); + } else /* single number with comma */ + count2 = INFINITY; + } else /* just a single number */ + count2 = count; + repeat(p, pos, count, count2); + if (!EATTWO('\\', '}')) { /* error heuristics */ + while (MORE() && !SEETWO('\\', '}')) + NEXT(); + REQUIRE(MORE(), REG_EBRACE); + SETERROR(REG_BADBR); + } + } else if (c == (unsigned char)'$') /* $ (but not \$) ends it */ + return(1); + + return(0); +} + +/* + - p_count - parse a repetition count + == static int p_count(register struct parse *p); + */ +static int /* the value */ +p_count(p) +register struct parse *p; +{ + register int count = 0; + register int ndigits = 0; + + while (MORE() && isdigit(PEEK()) && count <= DUPMAX) { + count = count*10 + (GETNEXT() - '0'); + ndigits++; + } + + REQUIRE(ndigits > 0 && count <= DUPMAX, REG_BADBR); + return(count); +} + +/* + - p_bracket - parse a bracketed character list + == static void p_bracket(register struct parse *p); + * + * Note a significant property of this code: if the allocset() did SETERROR, + * no set operations are done. + */ +static void +p_bracket(p) +register struct parse *p; +{ + register cset *cs = allocset(p); + register int invert = 0; + + /* Dept of Truly Sickening Special-Case Kludges */ + if (p->next + 5 < p->end && strncmp(p->next, "[:<:]]", 6) == 0) { + EMIT(OBOW, 0); + NEXTn(6); + return; + } + if (p->next + 5 < p->end && strncmp(p->next, "[:>:]]", 6) == 0) { + EMIT(OEOW, 0); + NEXTn(6); + return; + } + + if (EAT('^')) + invert++; /* make note to invert set at end */ + if (EAT(']')) + CHadd(cs, ']'); + else if (EAT('-')) + CHadd(cs, '-'); + while (MORE() && PEEK() != ']' && !SEETWO('-', ']')) + p_b_term(p, cs); + if (EAT('-')) + CHadd(cs, '-'); + MUSTEAT(']', REG_EBRACK); + + if (p->error != 0) /* don't mess things up further */ + return; + + if (p->g->cflags®_ICASE) { + register int i; + register int ci; + + for (i = p->g->csetsize - 1; i >= 0; i--) + if (CHIN(cs, i) && isalpha(i)) { + ci = othercase(i); + if (ci != i) + CHadd(cs, ci); + } + if (cs->multis != NULL) + mccase(p, cs); + } + if (invert) { + register int i; + + for (i = p->g->csetsize - 1; i >= 0; i--) + if (CHIN(cs, i)) + CHsub(cs, i); + else + CHadd(cs, i); + if (p->g->cflags®_NEWLINE) + CHsub(cs, '\n'); + if (cs->multis != NULL) + mcinvert(p, cs); + } + + assert(cs->multis == NULL); /* xxx */ + + if (nch(p, cs) == 1) { /* optimize singleton sets */ + ordinary(p, firstch(p, cs)); + freeset(p, cs); + } else + EMIT(OANYOF, freezeset(p, cs)); +} + +/* + - p_b_term - parse one term of a bracketed character list + == static void p_b_term(register struct parse *p, register cset *cs); + */ +static void +p_b_term(p, cs) +register struct parse *p; +register cset *cs; +{ + register char c; + register char start, finish; + register int i; + + /* classify what we've got */ + switch ((MORE()) ? PEEK() : '\0') { + case '[': + c = (MORE2()) ? PEEK2() : '\0'; + break; + case '-': + SETERROR(REG_ERANGE); + return; /* NOTE RETURN */ + break; + default: + c = '\0'; + break; + } + + switch (c) { + case ':': /* character class */ + NEXT2(); + REQUIRE(MORE(), REG_EBRACK); + c = PEEK(); + REQUIRE(c != '-' && c != ']', REG_ECTYPE); + p_b_cclass(p, cs); + REQUIRE(MORE(), REG_EBRACK); + REQUIRE(EATTWO(':', ']'), REG_ECTYPE); + break; + case '=': /* equivalence class */ + NEXT2(); + REQUIRE(MORE(), REG_EBRACK); + c = PEEK(); + REQUIRE(c != '-' && c != ']', REG_ECOLLATE); + p_b_eclass(p, cs); + REQUIRE(MORE(), REG_EBRACK); + REQUIRE(EATTWO('=', ']'), REG_ECOLLATE); + break; + default: /* symbol, ordinary character, or range */ +/* xxx revision needed for multichar stuff */ + start = p_b_symbol(p); + if (SEE('-') && MORE2() && PEEK2() != ']') { + /* range */ + NEXT(); + if (EAT('-')) + finish = '-'; + else + finish = p_b_symbol(p); + } else + finish = start; +/* xxx what about signed chars here... */ + REQUIRE(start <= finish, REG_ERANGE); + for (i = start; i <= finish; i++) + CHadd(cs, i); + break; + } +} + +/* + - p_b_cclass - parse a character-class name and deal with it + == static void p_b_cclass(register struct parse *p, register cset *cs); + */ +static void +p_b_cclass(p, cs) +register struct parse *p; +register cset *cs; +{ + register char *sp = p->next; + register struct cclass *cp; + register size_t len; + register const char *u; + register char c; + + while (MORE() && isalpha(PEEK())) + NEXT(); + len = p->next - sp; + for (cp = cclasses; cp->name != NULL; cp++) + if (strncmp(cp->name, sp, len) == 0 && cp->name[len] == '\0') + break; + if (cp->name == NULL) { + /* oops, didn't find it */ + SETERROR(REG_ECTYPE); + return; + } + + u = cp->chars; + while ((c = *u++) != '\0') + CHadd(cs, c); + for (u = cp->multis; *u != '\0'; u += strlen(u) + 1) + MCadd(p, cs, u); +} + +/* + - p_b_eclass - parse an equivalence-class name and deal with it + == static void p_b_eclass(register struct parse *p, register cset *cs); + * + * This implementation is incomplete. xxx + */ +static void +p_b_eclass(p, cs) +register struct parse *p; +register cset *cs; +{ + register char c; + + c = p_b_coll_elem(p, '='); + CHadd(cs, c); +} + +/* + - p_b_symbol - parse a character or [..]ed multicharacter collating symbol + == static char p_b_symbol(register struct parse *p); + */ +static char /* value of symbol */ +p_b_symbol(p) +register struct parse *p; +{ + register char value; + + REQUIRE(MORE(), REG_EBRACK); + if (!EATTWO('[', '.')) + return(GETNEXT()); + + /* collating symbol */ + value = p_b_coll_elem(p, '.'); + REQUIRE(EATTWO('.', ']'), REG_ECOLLATE); + return(value); +} + +/* + - p_b_coll_elem - parse a collating-element name and look it up + == static char p_b_coll_elem(register struct parse *p, int endc); + */ +static char /* value of collating element */ +p_b_coll_elem(p, endc) +register struct parse *p; +int endc; /* name ended by endc,']' */ +{ + register char *sp = p->next; + register struct cname *cp; + register int len; + + while (MORE() && !SEETWO(endc, ']')) + NEXT(); + if (!MORE()) { + SETERROR(REG_EBRACK); + return(0); + } + len = p->next - sp; + for (cp = cnames; cp->name != NULL; cp++) + if (strncmp(cp->name, sp, len) == 0 && cp->name[len] == '\0') + return(cp->code); /* known name */ + if (len == 1) + return(*sp); /* single character */ + SETERROR(REG_ECOLLATE); /* neither */ + return(0); +} + +/* + - othercase - return the case counterpart of an alphabetic + == static char othercase(int ch); + */ +static char /* if no counterpart, return ch */ +othercase(ch) +int ch; +{ + assert(isalpha(ch)); + if (isupper(ch)) + return(tolower(ch)); + else if (islower(ch)) + return(toupper(ch)); + else /* peculiar, but could happen */ + return(ch); +} + +/* + - bothcases - emit a dualcase version of a two-case character + == static void bothcases(register struct parse *p, int ch); + * + * Boy, is this implementation ever a kludge... + */ +static void +bothcases(p, ch) +register struct parse *p; +int ch; +{ + register char *oldnext = p->next; + register char *oldend = p->end; + char bracket[3]; + + assert(othercase(ch) != ch); /* p_bracket() would recurse */ + p->next = bracket; + p->end = bracket+2; + bracket[0] = ch; + bracket[1] = ']'; + bracket[2] = '\0'; + p_bracket(p); + assert(p->next == bracket+2); + p->next = oldnext; + p->end = oldend; +} + +/* + - ordinary - emit an ordinary character + == static void ordinary(register struct parse *p, register int ch); + */ +static void +ordinary(p, ch) +register struct parse *p; +register int ch; +{ + register cat_t *cap = p->g->categories; + + if ((p->g->cflags®_ICASE) && isalpha(ch) && othercase(ch) != ch) + bothcases(p, ch); + else { + EMIT(OCHAR, (unsigned char)ch); + if (cap[ch] == 0) + cap[ch] = p->g->ncategories++; + } +} + +/* + - nonnewline - emit REG_NEWLINE version of OANY + == static void nonnewline(register struct parse *p); + * + * Boy, is this implementation ever a kludge... + */ +static void +nonnewline(p) +register struct parse *p; +{ + register char *oldnext = p->next; + register char *oldend = p->end; + char bracket[4]; + + p->next = bracket; + p->end = bracket+3; + bracket[0] = '^'; + bracket[1] = '\n'; + bracket[2] = ']'; + bracket[3] = '\0'; + p_bracket(p); + assert(p->next == bracket+3); + p->next = oldnext; + p->end = oldend; +} + +/* + - repeat - generate code for a bounded repetition, recursively if needed + == static void repeat(register struct parse *p, sopno start, int from, int to); + */ +static void +repeat(p, start, from, to) +register struct parse *p; +sopno start; /* operand from here to end of strip */ +int from; /* repeated from this number */ +int to; /* to this number of times (maybe INFINITY) */ +{ + register sopno finish = HERE(); +# define N 2 +# define INF 3 +# define REP(f, t) ((f)*8 + (t)) +# define MAP(n) (((n) <= 1) ? (n) : ((n) == INFINITY) ? INF : N) + register sopno copy; + + if (p->error != 0) /* head off possible runaway recursion */ + return; + + assert(from <= to); + + switch (REP(MAP(from), MAP(to))) { + case REP(0, 0): /* must be user doing this */ + DROP(finish-start); /* drop the operand */ + break; + case REP(0, 1): /* as x{1,1}? */ + case REP(0, N): /* as x{1,n}? */ + case REP(0, INF): /* as x{1,}? */ + /* KLUDGE: emit y? as (y|) until subtle bug gets fixed */ + INSERT(OCH_, start); /* offset is wrong... */ + repeat(p, start+1, 1, to); + ASTERN(OOR1, start); + AHEAD(start); /* ... fix it */ + EMIT(OOR2, 0); + AHEAD(THERE()); + ASTERN(O_CH, THERETHERE()); + break; + case REP(1, 1): /* trivial case */ + /* done */ + break; + case REP(1, N): /* as x?x{1,n-1} */ + /* KLUDGE: emit y? as (y|) until subtle bug gets fixed */ + INSERT(OCH_, start); + ASTERN(OOR1, start); + AHEAD(start); + EMIT(OOR2, 0); /* offset very wrong... */ + AHEAD(THERE()); /* ...so fix it */ + ASTERN(O_CH, THERETHERE()); + copy = dupl(p, start+1, finish+1); + assert(copy == finish+4); + repeat(p, copy, 1, to-1); + break; + case REP(1, INF): /* as x+ */ + INSERT(OPLUS_, start); + ASTERN(O_PLUS, start); + break; + case REP(N, N): /* as xx{m-1,n-1} */ + copy = dupl(p, start, finish); + repeat(p, copy, from-1, to-1); + break; + case REP(N, INF): /* as xx{n-1,INF} */ + copy = dupl(p, start, finish); + repeat(p, copy, from-1, to); + break; + default: /* "can't happen" */ + SETERROR(REG_ASSERT); /* just in case */ + break; + } +} + +/* + - seterr - set an error condition + == static int seterr(register struct parse *p, int e); + */ +static int /* useless but makes type checking happy */ +seterr(p, e) +register struct parse *p; +int e; +{ + if (p->error == 0) /* keep earliest error condition */ + p->error = e; + p->next = nuls; /* try to bring things to a halt */ + p->end = nuls; + return(0); /* make the return value well-defined */ +} + +/* + - allocset - allocate a set of characters for [] + == static cset *allocset(register struct parse *p); + */ +static cset * +allocset(p) +register struct parse *p; +{ + register int no = p->g->ncsets++; + register size_t nc; + register size_t nbytes; + register cset *cs; + register size_t css = (size_t)p->g->csetsize; + register int i; + + if (no >= p->ncsalloc) { /* need another column of space */ + p->ncsalloc += CHAR_BIT; + nc = p->ncsalloc; + assert(nc % CHAR_BIT == 0); + nbytes = nc / CHAR_BIT * css; + if (p->g->sets == NULL) + p->g->sets = (cset *)malloc(nc * sizeof(cset)); + else + p->g->sets = (cset *)realloc((char *)p->g->sets, + nc * sizeof(cset)); + if (p->g->setbits == NULL) + p->g->setbits = (uch *)malloc(nbytes); + else { + p->g->setbits = (uch *)realloc((char *)p->g->setbits, + nbytes); + /* xxx this isn't right if setbits is now NULL */ + for (i = 0; i < no; i++) + p->g->sets[i].ptr = p->g->setbits + css*(i/CHAR_BIT); + } + if (p->g->sets != NULL && p->g->setbits != NULL) + (void) memset((char *)p->g->setbits + (nbytes - css), + 0, css); + else { + no = 0; + SETERROR(REG_ESPACE); + /* caller's responsibility not to do set ops */ + } + } + + assert(p->g->sets != NULL); /* xxx */ + cs = &p->g->sets[no]; + cs->ptr = p->g->setbits + css*((no)/CHAR_BIT); + cs->mask = 1 << ((no) % CHAR_BIT); + cs->hash = 0; + cs->smultis = 0; + cs->multis = NULL; + + return(cs); +} + +/* + - freeset - free a now-unused set + == static void freeset(register struct parse *p, register cset *cs); + */ +static void +freeset(p, cs) +register struct parse *p; +register cset *cs; +{ + register int i; + register cset *top = &p->g->sets[p->g->ncsets]; + register size_t css = (size_t)p->g->csetsize; + + for (i = 0; i < css; i++) + CHsub(cs, i); + if (cs == top-1) /* recover only the easy case */ + p->g->ncsets--; +} + +/* + - freezeset - final processing on a set of characters + == static int freezeset(register struct parse *p, register cset *cs); + * + * The main task here is merging identical sets. This is usually a waste + * of time (although the hash code minimizes the overhead), but can win + * big if REG_ICASE is being used. REG_ICASE, by the way, is why the hash + * is done using addition rather than xor -- all ASCII [aA] sets xor to + * the same value! + */ +static int /* set number */ +freezeset(p, cs) +register struct parse *p; +register cset *cs; +{ + register uch h = cs->hash; + register int i; + register cset *top = &p->g->sets[p->g->ncsets]; + register cset *cs2; + register size_t css = (size_t)p->g->csetsize; + + /* look for an earlier one which is the same */ + for (cs2 = &p->g->sets[0]; cs2 < top; cs2++) + if (cs2->hash == h && cs2 != cs) { + /* maybe */ + for (i = 0; i < css; i++) + if (!!CHIN(cs2, i) != !!CHIN(cs, i)) + break; /* no */ + if (i == css) + break; /* yes */ + } + + if (cs2 < top) { /* found one */ + freeset(p, cs); + cs = cs2; + } + + return((int)(cs - p->g->sets)); +} + +/* + - firstch - return first character in a set (which must have at least one) + == static int firstch(register struct parse *p, register cset *cs); + */ +static int /* character; there is no "none" value */ +firstch(p, cs) +register struct parse *p; +register cset *cs; +{ + register int i; + register size_t css = (size_t)p->g->csetsize; + + for (i = 0; i < css; i++) + if (CHIN(cs, i)) + return((char)i); + assert(never); + return(0); /* arbitrary */ +} + +/* + - nch - number of characters in a set + == static int nch(register struct parse *p, register cset *cs); + */ +static int +nch(p, cs) +register struct parse *p; +register cset *cs; +{ + register int i; + register size_t css = (size_t)p->g->csetsize; + register int n = 0; + + for (i = 0; i < css; i++) + if (CHIN(cs, i)) + n++; + return(n); +} + +/* + - mcadd - add a collating element to a cset + == static void mcadd(register struct parse *p, register cset *cs, \ + == register char *cp); + */ +static void +mcadd(p, cs, cp) +register struct parse *p; +register cset *cs; +register const char *cp; +{ + register size_t oldend = cs->smultis; + + cs->smultis += strlen(cp) + 1; + if (cs->multis == NULL) + cs->multis = malloc(cs->smultis); + else + cs->multis = realloc(cs->multis, cs->smultis); + if (cs->multis == NULL) { + SETERROR(REG_ESPACE); + return; + } + + (void) strcpy(cs->multis + oldend - 1, cp); + cs->multis[cs->smultis - 1] = '\0'; +} + +/* + - mcinvert - invert the list of collating elements in a cset + == static void mcinvert(register struct parse *p, register cset *cs); + * + * This would have to know the set of possibilities. Implementation + * is deferred. + */ +static void +mcinvert(p, cs) +register struct parse *p; +register cset *cs; +{ + assert(cs->multis == NULL); /* xxx */ +} + +/* + - mccase - add case counterparts of the list of collating elements in a cset + == static void mccase(register struct parse *p, register cset *cs); + * + * This would have to know the set of possibilities. Implementation + * is deferred. + */ +static void +mccase(p, cs) +register struct parse *p; +register cset *cs; +{ + assert(cs->multis == NULL); /* xxx */ +} + +/* + - isinsets - is this character in any sets? + == static int isinsets(register struct re_guts *g, int c); + */ +static int /* predicate */ +isinsets(g, c) +register struct re_guts *g; +int c; +{ + register uch *col; + register int i; + register int ncols = (g->ncsets+(CHAR_BIT-1)) / CHAR_BIT; + register unsigned uc = (unsigned char)c; + + for (i = 0, col = g->setbits; i < ncols; i++, col += g->csetsize) + if (col[uc] != 0) + return(1); + return(0); +} + +/* + - samesets - are these two characters in exactly the same sets? + == static int samesets(register struct re_guts *g, int c1, int c2); + */ +static int /* predicate */ +samesets(g, c1, c2) +register struct re_guts *g; +int c1; +int c2; +{ + register uch *col; + register int i; + register int ncols = (g->ncsets+(CHAR_BIT-1)) / CHAR_BIT; + register unsigned uc1 = (unsigned char)c1; + register unsigned uc2 = (unsigned char)c2; + + for (i = 0, col = g->setbits; i < ncols; i++, col += g->csetsize) + if (col[uc1] != col[uc2]) + return(0); + return(1); +} + +/* + - categorize - sort out character categories + == static void categorize(struct parse *p, register struct re_guts *g); + */ +static void +categorize(p, g) +struct parse *p; +register struct re_guts *g; +{ + register cat_t *cats = g->categories; + register int c; + register int c2; + register cat_t cat; + + /* avoid making error situations worse */ + if (p->error != 0) + return; + + for (c = CHAR_MIN; c <= CHAR_MAX; c++) + if (cats[c] == 0 && isinsets(g, c)) { + cat = g->ncategories++; + cats[c] = cat; + for (c2 = c+1; c2 <= CHAR_MAX; c2++) + if (cats[c2] == 0 && samesets(g, c, c2)) + cats[c2] = cat; + } +} + +/* + - dupl - emit a duplicate of a bunch of sops + == static sopno dupl(register struct parse *p, sopno start, sopno finish); + */ +static sopno /* start of duplicate */ +dupl(p, start, finish) +register struct parse *p; +sopno start; /* from here */ +sopno finish; /* to this less one */ +{ + register sopno ret = HERE(); + register sopno len = finish - start; + + assert(finish >= start); + if (len == 0) + return(ret); + enlarge(p, p->ssize + len); /* this many unexpected additions */ + assert(p->ssize >= p->slen + len); + (void) memcpy((char *)(p->strip + p->slen), + (char *)(p->strip + start), (size_t)len*sizeof(sop)); + p->slen += len; + return(ret); +} + +/* + - doemit - emit a strip operator + == static void doemit(register struct parse *p, sop op, size_t opnd); + * + * It might seem better to implement this as a macro with a function as + * hard-case backup, but it's just too big and messy unless there are + * some changes to the data structures. Maybe later. + */ +static void +doemit(p, op, opnd) +register struct parse *p; +sop op; +size_t opnd; +{ + /* avoid making error situations worse */ + if (p->error != 0) + return; + + /* deal with oversize operands ("can't happen", more or less) */ + assert(opnd < 1<<OPSHIFT); + + /* deal with undersized strip */ + if (p->slen >= p->ssize) + enlarge(p, (p->ssize+1) / 2 * 3); /* +50% */ + assert(p->slen < p->ssize); + + /* finally, it's all reduced to the easy case */ + p->strip[p->slen++] = SOP(op, opnd); +} + +/* + - doinsert - insert a sop into the strip + == static void doinsert(register struct parse *p, sop op, size_t opnd, sopno pos); + */ +static void +doinsert(p, op, opnd, pos) +register struct parse *p; +sop op; +size_t opnd; +sopno pos; +{ + register sopno sn; + register sop s; + register int i; + + /* avoid making error situations worse */ + if (p->error != 0) + return; + + sn = HERE(); + EMIT(op, opnd); /* do checks, ensure space */ + assert(HERE() == sn+1); + s = p->strip[sn]; + + /* adjust paren pointers */ + assert(pos > 0); + for (i = 1; i < NPAREN; i++) { + if (p->pbegin[i] >= pos) { + p->pbegin[i]++; + } + if (p->pend[i] >= pos) { + p->pend[i]++; + } + } + + memmove((char *)&p->strip[pos+1], (char *)&p->strip[pos], + (HERE()-pos-1)*sizeof(sop)); + p->strip[pos] = s; +} + +/* + - dofwd - complete a forward reference + == static void dofwd(register struct parse *p, sopno pos, sop value); + */ +static void +dofwd(p, pos, value) +register struct parse *p; +register sopno pos; +sop value; +{ + /* avoid making error situations worse */ + if (p->error != 0) + return; + + assert(value < 1<<OPSHIFT); + p->strip[pos] = OP(p->strip[pos]) | value; +} + +/* + - enlarge - enlarge the strip + == static void enlarge(register struct parse *p, sopno size); + */ +static void +enlarge(p, size) +register struct parse *p; +register sopno size; +{ + register sop *sp; + + if (p->ssize >= size) + return; + + sp = (sop *)realloc(p->strip, size*sizeof(sop)); + if (sp == NULL) { + SETERROR(REG_ESPACE); + return; + } + p->strip = sp; + p->ssize = size; +} + +/* + - stripsnug - compact the strip + == static void stripsnug(register struct parse *p, register struct re_guts *g); + */ +static void +stripsnug(p, g) +register struct parse *p; +register struct re_guts *g; +{ + g->nstates = p->slen; + g->strip = (sop *)realloc((char *)p->strip, p->slen * sizeof(sop)); + if (g->strip == NULL) { + SETERROR(REG_ESPACE); + g->strip = p->strip; + } +} + +/* + - findmust - fill in must and mlen with longest mandatory literal string + == static void findmust(register struct parse *p, register struct re_guts *g); + * + * This algorithm could do fancy things like analyzing the operands of | + * for common subsequences. Someday. This code is simple and finds most + * of the interesting cases. + * + * Note that must and mlen got initialized during setup. + */ +static void +findmust(p, g) +struct parse *p; +register struct re_guts *g; +{ + register sop *scan; + sop *start = NULL; + register sop *newstart = NULL; + register sopno newlen; + register sop s; + register char *cp; + register sopno i; + + /* avoid making error situations worse */ + if (p->error != 0) + return; + + /* find the longest OCHAR sequence in strip */ + newlen = 0; + scan = g->strip + 1; + do { + s = *scan++; + switch (OP(s)) { + case OCHAR: /* sequence member */ + if (newlen == 0) /* new sequence */ + newstart = scan - 1; + newlen++; + break; + case OPLUS_: /* things that don't break one */ + case OLPAREN: + case ORPAREN: + break; + case OQUEST_: /* things that must be skipped */ + case OCH_: + scan--; + do { + scan += OPND(s); + s = *scan; + /* assert() interferes w debug printouts */ + if (OP(s) != O_QUEST && OP(s) != O_CH && + OP(s) != OOR2) { + g->iflags |= BAD; + return; + } + } while (OP(s) != O_QUEST && OP(s) != O_CH); + /* fallthrough */ + default: /* things that break a sequence */ + if (newlen > g->mlen) { /* ends one */ + start = newstart; + g->mlen = newlen; + } + newlen = 0; + break; + } + } while (OP(s) != OEND); + + if (g->mlen == 0) /* there isn't one */ + return; + + /* turn it into a character string */ + g->must = malloc((size_t)g->mlen + 1); + if (g->must == NULL) { /* argh; just forget it */ + g->mlen = 0; + return; + } + cp = g->must; + scan = start; + for (i = g->mlen; i > 0; i--) { + while (OP(s = *scan++) != OCHAR) + continue; + assert(cp < g->must + g->mlen); + *cp++ = (char)OPND(s); + } + assert(cp == g->must + g->mlen); + *cp++ = '\0'; /* just on general principles */ +} + +/* + - pluscount - count + nesting + == static sopno pluscount(register struct parse *p, register struct re_guts *g); + */ +static sopno /* nesting depth */ +pluscount(p, g) +struct parse *p; +register struct re_guts *g; +{ + register sop *scan; + register sop s; + register sopno plusnest = 0; + register sopno maxnest = 0; + + if (p->error != 0) + return(0); /* there may not be an OEND */ + + scan = g->strip + 1; + do { + s = *scan++; + switch (OP(s)) { + case OPLUS_: + plusnest++; + break; + case O_PLUS: + if (plusnest > maxnest) + maxnest = plusnest; + plusnest--; + break; + } + } while (OP(s) != OEND); + if (plusnest != 0) + g->iflags |= BAD; + return(maxnest); +} diff --git a/winsup/cygwin/regex/regcomp.ih b/winsup/cygwin/regex/regcomp.ih new file mode 100644 index 000000000..e16d5ab2f --- /dev/null +++ b/winsup/cygwin/regex/regcomp.ih @@ -0,0 +1,48 @@ +/* ========= begin header generated by ./mkh ========= */ +#ifdef __cplusplus +extern "C" { +#endif + +/* === regcomp.c === */ +static void p_ere(register struct parse *p, int stop); +static void p_ere_exp(register struct parse *p); +static void p_str(register struct parse *p); +static void p_bre(register struct parse *p, register int end1, register int end2); +static int p_simp_re(register struct parse *p, int starordinary); +static int p_count(register struct parse *p); +static void p_bracket(register struct parse *p); +static void p_b_term(register struct parse *p, register cset *cs); +static void p_b_cclass(register struct parse *p, register cset *cs); +static void p_b_eclass(register struct parse *p, register cset *cs); +static char p_b_symbol(register struct parse *p); +static char p_b_coll_elem(register struct parse *p, int endc); +static char othercase(int ch); +static void bothcases(register struct parse *p, int ch); +static void ordinary(register struct parse *p, register int ch); +static void nonnewline(register struct parse *p); +static void repeat(register struct parse *p, sopno start, int from, int to); +static int seterr(register struct parse *p, int e); +static cset *allocset(register struct parse *p); +static void freeset(register struct parse *p, register cset *cs); +static int freezeset(register struct parse *p, register cset *cs); +static int firstch(register struct parse *p, register cset *cs); +static int nch(register struct parse *p, register cset *cs); +static void mcadd(register struct parse *p, register cset *cs, register const char *cp); +static void mcinvert(register struct parse *p, register cset *cs); +static void mccase(register struct parse *p, register cset *cs); +static int isinsets(register struct re_guts *g, int c); +static int samesets(register struct re_guts *g, int c1, int c2); +static void categorize(struct parse *p, register struct re_guts *g); +static sopno dupl(register struct parse *p, sopno start, sopno finish); +static void doemit(register struct parse *p, sop op, size_t opnd); +static void doinsert(register struct parse *p, sop op, size_t opnd, sopno pos); +static void dofwd(register struct parse *p, sopno pos, sop value); +static void enlarge(register struct parse *p, sopno size); +static void stripsnug(register struct parse *p, register struct re_guts *g); +static void findmust(register struct parse *p, register struct re_guts *g); +static sopno pluscount(register struct parse *p, register struct re_guts *g); + +#ifdef __cplusplus +} +#endif +/* ========= end header generated by ./mkh ========= */ diff --git a/winsup/cygwin/regex/regerror.c b/winsup/cygwin/regex/regerror.c new file mode 100644 index 000000000..4d14ab231 --- /dev/null +++ b/winsup/cygwin/regex/regerror.c @@ -0,0 +1,127 @@ +#include "winsup.h" +#include <sys/types.h> +#include <stdio.h> +#include <string.h> +#include <ctype.h> +#include <limits.h> +#include <stdlib.h> +#include "regex.h" + +#include "utils.h" +#include "regerror.ih" + +/* + = #define REG_OKAY 0 + = #define REG_NOMATCH 1 + = #define REG_BADPAT 2 + = #define REG_ECOLLATE 3 + = #define REG_ECTYPE 4 + = #define REG_EESCAPE 5 + = #define REG_ESUBREG 6 + = #define REG_EBRACK 7 + = #define REG_EPAREN 8 + = #define REG_EBRACE 9 + = #define REG_BADBR 10 + = #define REG_ERANGE 11 + = #define REG_ESPACE 12 + = #define REG_BADRPT 13 + = #define REG_EMPTY 14 + = #define REG_ASSERT 15 + = #define REG_INVARG 16 + = #define REG_ATOI 255 // convert name to number (!) + = #define REG_ITOA 0400 // convert number to name (!) + */ +static struct rerr { + int code; + const char *name; + const char *explain; +} rerrs[] = { + {REG_OKAY, "REG_OKAY", "no errors detected"}, + {REG_NOMATCH, "REG_NOMATCH", "regexec() failed to match"}, + {REG_BADPAT, "REG_BADPAT", "invalid regular expression"}, + {REG_ECOLLATE, "REG_ECOLLATE", "invalid collating element"}, + {REG_ECTYPE, "REG_ECTYPE", "invalid character class"}, + {REG_EESCAPE, "REG_EESCAPE", "trailing backslash (\\)"}, + {REG_ESUBREG, "REG_ESUBREG", "invalid backreference number"}, + {REG_EBRACK, "REG_EBRACK", "brackets ([ ]) not balanced"}, + {REG_EPAREN, "REG_EPAREN", "parentheses not balanced"}, + {REG_EBRACE, "REG_EBRACE", "braces not balanced"}, + {REG_BADBR, "REG_BADBR", "invalid repetition count(s)"}, + {REG_ERANGE, "REG_ERANGE", "invalid character range"}, + {REG_ESPACE, "REG_ESPACE", "out of memory"}, + {REG_BADRPT, "REG_BADRPT", "repetition-operator operand invalid"}, + {REG_EMPTY, "REG_EMPTY", "empty (sub)expression"}, + {REG_ASSERT, "REG_ASSERT", "\"can't happen\" -- you found a bug"}, + {REG_INVARG, "REG_INVARG", "invalid argument to regex routine"}, + {-1, "", "*** unknown regexp error code ***"}, +}; + +/* + - regerror - the interface to error numbers + = extern size_t regerror(int, const regex_t *, char *, size_t); + */ +/* ARGSUSED */ +size_t +regerror(errcode, preg, errbuf, errbuf_size) +int errcode; +const regex_t *preg; +char *errbuf; +size_t errbuf_size; +{ + register struct rerr *r; + register size_t len; + register int target = errcode &~ REG_ITOA; + register const char *s; + char convbuf[50]; + + if (errcode == REG_ATOI) + s = regatoi(preg, convbuf); + else { + for (r = rerrs; r->code >= 0; r++) + if (r->code == target) + break; + + if (errcode®_ITOA) { + if (r->code >= 0) + (void) strcpy(convbuf, r->name); + else + sprintf(convbuf, "REG_0x%x", target); + assert(strlen(convbuf) < sizeof(convbuf)); + s = convbuf; + } else + s = r->explain; + } + + len = strlen(s) + 1; + if (errbuf_size > 0) { + if (errbuf_size > len) + (void) strcpy(errbuf, s); + else { + (void) strncpy(errbuf, s, errbuf_size-1); + errbuf[errbuf_size-1] = '\0'; + } + } + + return(len); +} + +/* + - regatoi - internal routine to implement REG_ATOI + == static char *regatoi(const regex_t *preg, char *localbuf); + */ +static const char * +regatoi(preg, localbuf) +const regex_t *preg; +char *localbuf; +{ + register struct rerr *r; + + for (r = rerrs; r->code >= 0; r++) + if (strcmp(r->name, preg->re_endp) == 0) + break; + if (r->code < 0) + return("0"); + + sprintf(localbuf, "%d", r->code); + return(localbuf); +} diff --git a/winsup/cygwin/regex/regerror.ih b/winsup/cygwin/regex/regerror.ih new file mode 100644 index 000000000..d428544e3 --- /dev/null +++ b/winsup/cygwin/regex/regerror.ih @@ -0,0 +1,12 @@ +/* ========= begin header generated by ./mkh ========= */ +#ifdef __cplusplus +extern "C" { +#endif + +/* === regerror.c === */ +static const char *regatoi(const regex_t *preg, char *localbuf); + +#ifdef __cplusplus +} +#endif +/* ========= end header generated by ./mkh ========= */ diff --git a/winsup/cygwin/regex/regex.3 b/winsup/cygwin/regex/regex.3 new file mode 100644 index 000000000..bc747096d --- /dev/null +++ b/winsup/cygwin/regex/regex.3 @@ -0,0 +1,509 @@ +.TH REGEX 3 "25 Sept 1997" +.BY "Henry Spencer" +.de ZR +.\" one other place knows this name: the SEE ALSO section +.IR regex (7) \\$1 +.. +.SH NAME +regcomp, regexec, regerror, regfree \- regular-expression library +.SH SYNOPSIS +.ft B +.\".na +#include <sys/types.h> +.br +#include <regex.h> +.HP 10 +int regcomp(regex_t\ *preg, const\ char\ *pattern, int\ cflags); +.HP +int\ regexec(const\ regex_t\ *preg, const\ char\ *string, +size_t\ nmatch, regmatch_t\ pmatch[], int\ eflags); +.HP +size_t\ regerror(int\ errcode, const\ regex_t\ *preg, +char\ *errbuf, size_t\ errbuf_size); +.HP +void\ regfree(regex_t\ *preg); +.\".ad +.ft +.SH DESCRIPTION +These routines implement POSIX 1003.2 regular expressions (``RE''s); +see +.ZR . +.I Regcomp +compiles an RE written as a string into an internal form, +.I regexec +matches that internal form against a string and reports results, +.I regerror +transforms error codes from either into human-readable messages, +and +.I regfree +frees any dynamically-allocated storage used by the internal form +of an RE. +.PP +The header +.I <regex.h> +declares two structure types, +.I regex_t +and +.IR regmatch_t , +the former for compiled internal forms and the latter for match reporting. +It also declares the four functions, +a type +.IR regoff_t , +and a number of constants with names starting with ``REG_''. +.PP +.I Regcomp +compiles the regular expression contained in the +.I pattern +string, +subject to the flags in +.IR cflags , +and places the results in the +.I regex_t +structure pointed to by +.IR preg . +.I Cflags +is the bitwise OR of zero or more of the following flags: +.IP REG_EXTENDED \w'REG_EXTENDED'u+2n +Compile modern (``extended'') REs, +rather than the obsolete (``basic'') REs that +are the default. +.IP REG_BASIC +This is a synonym for 0, +provided as a counterpart to REG_EXTENDED to improve readability. +This is an extension, +compatible with but not specified by POSIX 1003.2, +and should be used with +caution in software intended to be portable to other systems. +.IP REG_NOSPEC +Compile with recognition of all special characters turned off. +All characters are thus considered ordinary, +so the ``RE'' is a literal string. +This is an extension, +compatible with but not specified by POSIX 1003.2, +and should be used with +caution in software intended to be portable to other systems. +REG_EXTENDED and REG_NOSPEC may not be used +in the same call to +.IR regcomp . +.IP REG_ICASE +Compile for matching that ignores upper/lower case distinctions. +See +.ZR . +.IP REG_NOSUB +Compile for matching that need only report success or failure, +not what was matched. +.IP REG_NEWLINE +Compile for newline-sensitive matching. +By default, newline is a completely ordinary character with no special +meaning in either REs or strings. +With this flag, +`[^' bracket expressions and `.' never match newline, +a `^' anchor matches the null string after any newline in the string +in addition to its normal function, +and the `$' anchor matches the null string before any newline in the +string in addition to its normal function. +.IP REG_PEND +The regular expression ends, +not at the first NUL, +but just before the character pointed to by the +.I re_endp +member of the structure pointed to by +.IR preg . +The +.I re_endp +member is of type +.IR const\ char\ * . +This flag permits inclusion of NULs in the RE; +they are considered ordinary characters. +This is an extension, +compatible with but not specified by POSIX 1003.2, +and should be used with +caution in software intended to be portable to other systems. +.PP +When successful, +.I regcomp +returns 0 and fills in the structure pointed to by +.IR preg . +One member of that structure +(other than +.IR re_endp ) +is publicized: +.IR re_nsub , +of type +.IR size_t , +contains the number of parenthesized subexpressions within the RE +(except that the value of this member is undefined if the +REG_NOSUB flag was used). +If +.I regcomp +fails, it returns a non-zero error code; +see DIAGNOSTICS. +.PP +.I Regexec +matches the compiled RE pointed to by +.I preg +against the +.IR string , +subject to the flags in +.IR eflags , +and reports results using +.IR nmatch , +.IR pmatch , +and the returned value. +The RE must have been compiled by a previous invocation of +.IR regcomp . +The compiled form is not altered during execution of +.IR regexec , +so a single compiled RE can be used simultaneously by multiple threads. +.PP +By default, +the NUL-terminated string pointed to by +.I string +is considered to be the text of an entire line, +with the NUL indicating the end of the line. +(That is, +any other end-of-line marker is considered to have been removed +and replaced by the NUL.) +The +.I eflags +argument is the bitwise OR of zero or more of the following flags: +.IP REG_NOTBOL \w'REG_STARTEND'u+2n +The first character of +the string +is not the beginning of a line, so the `^' anchor should not match before it. +This does not affect the behavior of newlines under REG_NEWLINE. +.IP REG_NOTEOL +The NUL terminating +the string +does not end a line, so the `$' anchor should not match before it. +This does not affect the behavior of newlines under REG_NEWLINE. +.IP REG_STARTEND +The string is considered to start at +\fIstring\fR\ + \fIpmatch\fR[0].\fIrm_so\fR +and to have a terminating NUL located at +\fIstring\fR\ + \fIpmatch\fR[0].\fIrm_eo\fR +(there need not actually be a NUL at that location), +regardless of the value of +.IR nmatch . +See below for the definition of +.IR pmatch +and +.IR nmatch . +This is an extension, +compatible with but not specified by POSIX 1003.2, +and should be used with +caution in software intended to be portable to other systems. +Note that a non-zero \fIrm_so\fR does not imply REG_NOTBOL; +REG_STARTEND affects only the location of the string, +not how it is matched. +.PP +See +.ZR +for a discussion of what is matched in situations where an RE or a +portion thereof could match any of several substrings of +.IR string . +.PP +Normally, +.I regexec +returns 0 for success and the non-zero code REG_NOMATCH for failure. +Other non-zero error codes may be returned in exceptional situations; +see DIAGNOSTICS. +.PP +If REG_NOSUB was specified in the compilation of the RE, +or if +.I nmatch +is 0, +.I regexec +ignores the +.I pmatch +argument (but see below for the case where REG_STARTEND is specified). +Otherwise, +.I pmatch +points to an array of +.I nmatch +structures of type +.IR regmatch_t . +Such a structure has at least the members +.I rm_so +and +.IR rm_eo , +both of type +.I regoff_t +(a signed arithmetic type at least as large as an +.I off_t +and a +.IR ssize_t ), +containing respectively the offset of the first character of a substring +and the offset of the first character after the end of the substring. +Offsets are measured from the beginning of the +.I string +argument given to +.IR regexec . +An empty substring is denoted by equal offsets, +both indicating the character following the empty substring. +.PP +The 0th member of the +.I pmatch +array is filled in to indicate what substring of +.I string +was matched by the entire RE. +Remaining members report what substring was matched by parenthesized +subexpressions within the RE; +member +.I i +reports subexpression +.IR i , +with subexpressions counted (starting at 1) by the order of their opening +parentheses in the RE, left to right. +Unused entries in the array\(emcorresponding either to subexpressions that +did not participate in the match at all, or to subexpressions that do not +exist in the RE (that is, \fIi\fR\ > \fIpreg\fR\->\fIre_nsub\fR)\(emhave both +.I rm_so +and +.I rm_eo +set to \-1. +If a subexpression participated in the match several times, +the reported substring is the last one it matched. +(Note, as an example in particular, that when the RE `(b*)+' matches `bbb', +the parenthesized subexpression matches the three `b's and then +an infinite number of empty strings following the last `b', +so the reported substring is one of the empties.) +.PP +If REG_STARTEND is specified, +.I pmatch +must point to at least one +.I regmatch_t +(even if +.I nmatch +is 0 or REG_NOSUB was specified), +to hold the input offsets for REG_STARTEND. +Use for output is still entirely controlled by +.IR nmatch ; +if +.I nmatch +is 0 or REG_NOSUB was specified, +the value of +.IR pmatch [0] +will not be changed by a successful +.IR regexec . +.PP +.I Regerror +maps a non-zero +.I errcode +from either +.I regcomp +or +.I regexec +to a human-readable, printable message. +If +.I preg +is non-NULL, +the error code should have arisen from use of +the +.I regex_t +pointed to by +.IR preg , +and if the error code came from +.IR regcomp , +it should have been the result from the most recent +.I regcomp +using that +.IR regex_t . +.RI ( Regerror +may be able to supply a more detailed message using information +from the +.IR regex_t .) +.I Regerror +places the NUL-terminated message into the buffer pointed to by +.IR errbuf , +limiting the length (including the NUL) to at most +.I errbuf_size +bytes. +If the whole message won't fit, +as much of it as will fit before the terminating NUL is supplied. +In any case, +the returned value is the size of buffer needed to hold the whole +message (including terminating NUL). +If +.I errbuf_size +is 0, +.I errbuf +is ignored but the return value is still correct. +.PP +If the +.I errcode +given to +.I regerror +is first ORed with REG_ITOA, +the ``message'' that results is the printable name of the error code, +e.g. ``REG_NOMATCH'', +rather than an explanation thereof. +If +.I errcode +is REG_ATOI, +then +.I preg +shall be non-NULL and the +.I re_endp +member of the structure it points to +must point to the printable name of an error code; +in this case, the result in +.I errbuf +is the decimal digits of +the numeric value of the error code +(0 if the name is not recognized). +REG_ITOA and REG_ATOI are intended primarily as debugging facilities; +they are extensions, +compatible with but not specified by POSIX 1003.2, +and should be used with +caution in software intended to be portable to other systems. +Be warned also that they are considered experimental and changes are possible. +.PP +.I Regfree +frees any dynamically-allocated storage associated with the compiled RE +pointed to by +.IR preg . +The remaining +.I regex_t +is no longer a valid compiled RE +and the effect of supplying it to +.I regexec +or +.I regerror +is undefined. +.PP +None of these functions references global variables except for tables +of constants; +all are safe for use from multiple threads if the arguments are safe. +.SH IMPLEMENTATION CHOICES +There are a number of decisions that 1003.2 leaves up to the implementor, +either by explicitly saying ``undefined'' or by virtue of them being +forbidden by the RE grammar. +This implementation treats them as follows. +.PP +See +.ZR +for a discussion of the definition of case-independent matching. +.PP +There is no particular limit on the length of REs, +except insofar as memory is limited. +Memory usage is approximately linear in RE size, and largely insensitive +to RE complexity, except for bounded repetitions. +See BUGS for one short RE using them +that will run almost any system out of memory. +.PP +A backslashed character other than one specifically given a magic meaning +by 1003.2 (such magic meanings occur only in obsolete [``basic''] REs) +is taken as an ordinary character. +.PP +Any unmatched [ is a REG_EBRACK error. +.PP +Equivalence classes cannot begin or end bracket-expression ranges. +The endpoint of one range cannot begin another. +.PP +RE_DUP_MAX, the limit on repetition counts in bounded repetitions, is 255. +.PP +A repetition operator (?, *, +, or bounds) cannot follow another +repetition operator. +A repetition operator cannot begin an expression or subexpression +or follow `^' or `|'. +.PP +`|' cannot appear first or last in a (sub)expression or after another `|', +i.e. an operand of `|' cannot be an empty subexpression. +An empty parenthesized subexpression, `()', is legal and matches an +empty (sub)string. +An empty string is not a legal RE. +.PP +A `{' followed by a digit is considered the beginning of bounds for a +bounded repetition, which must then follow the syntax for bounds. +A `{' \fInot\fR followed by a digit is considered an ordinary character. +.PP +`^' and `$' beginning and ending subexpressions in obsolete (``basic'') +REs are anchors, not ordinary characters. +.SH SEE ALSO +grep(1), regex(7) +.PP +POSIX 1003.2, sections 2.8 (Regular Expression Notation) +and +B.5 (C Binding for Regular Expression Matching). +.SH DIAGNOSTICS +Non-zero error codes from +.I regcomp +and +.I regexec +include the following: +.PP +.nf +.ta \w'REG_ECOLLATE'u+3n +REG_NOMATCH regexec() failed to match +REG_BADPAT invalid regular expression +REG_ECOLLATE invalid collating element +REG_ECTYPE invalid character class +REG_EESCAPE \e applied to unescapable character +REG_ESUBREG invalid backreference number +REG_EBRACK brackets [ ] not balanced +REG_EPAREN parentheses ( ) not balanced +REG_EBRACE braces { } not balanced +REG_BADBR invalid repetition count(s) in { } +REG_ERANGE invalid character range in [ ] +REG_ESPACE ran out of memory +REG_BADRPT ?, *, or + operand invalid +REG_EMPTY empty (sub)expression +REG_ASSERT ``can't happen''\(emyou found a bug +REG_INVARG invalid argument, e.g. negative-length string +.fi +.SH HISTORY +Written by Henry Spencer, +henry@zoo.toronto.edu. +.SH BUGS +This is an alpha release with known defects. +Please report problems. +.PP +There is one known functionality bug. +The implementation of internationalization is incomplete: +the locale is always assumed to be the default one of 1003.2, +and only the collating elements etc. of that locale are available. +.PP +The back-reference code is subtle and doubts linger about its correctness +in complex cases. +.PP +.I Regexec +performance is poor. +This will improve with later releases. +.I Nmatch +exceeding 0 is expensive; +.I nmatch +exceeding 1 is worse. +.I Regexec +is largely insensitive to RE complexity \fIexcept\fR that back +references are massively expensive. +RE length does matter; in particular, there is a strong speed bonus +for keeping RE length under about 30 characters, +with most special characters counting roughly double. +.PP +.I Regcomp +implements bounded repetitions by macro expansion, +which is costly in time and space if counts are large +or bounded repetitions are nested. +An RE like, say, +`((((a{1,100}){1,100}){1,100}){1,100}){1,100}' +will (eventually) run almost any existing machine out of swap space. +.PP +There are suspected problems with response to obscure error conditions. +Notably, +certain kinds of internal overflow, +produced only by truly enormous REs or by multiply nested bounded repetitions, +are probably not handled well. +.PP +Due to a mistake in 1003.2, things like `a)b' are legal REs because `)' is +a special character only in the presence of a previous unmatched `('. +This can't be fixed until the spec is fixed. +.PP +The standard's definition of back references is vague. +For example, does +`a\e(\e(b\e)*\e2\e)*d' match `abbbd'? +Until the standard is clarified, +behavior in such cases should not be relied on. +.PP +The implementation of word-boundary matching is a bit of a kludge, +and bugs may lurk in combinations of word-boundary matching and anchoring. diff --git a/winsup/cygwin/regex/regex.7 b/winsup/cygwin/regex/regex.7 new file mode 100644 index 000000000..0fa180269 --- /dev/null +++ b/winsup/cygwin/regex/regex.7 @@ -0,0 +1,235 @@ +.TH REGEX 7 "25 Oct 1995" +.BY "Henry Spencer" +.SH NAME +regex \- POSIX 1003.2 regular expressions +.SH DESCRIPTION +Regular expressions (``RE''s), +as defined in POSIX 1003.2, come in two forms: +modern REs (roughly those of +.IR egrep ; +1003.2 calls these ``extended'' REs) +and obsolete REs (roughly those of +.IR ed ; +1003.2 ``basic'' REs). +Obsolete REs mostly exist for backward compatibility in some old programs; +they will be discussed at the end. +1003.2 leaves some aspects of RE syntax and semantics open; +`\(dg' marks decisions on these aspects that +may not be fully portable to other 1003.2 implementations. +.PP +A (modern) RE is one\(dg or more non-empty\(dg \fIbranches\fR, +separated by `|'. +It matches anything that matches one of the branches. +.PP +A branch is one\(dg or more \fIpieces\fR, concatenated. +It matches a match for the first, followed by a match for the second, etc. +.PP +A piece is an \fIatom\fR possibly followed +by a single\(dg `*', `+', `?', or \fIbound\fR. +An atom followed by `*' matches a sequence of 0 or more matches of the atom. +An atom followed by `+' matches a sequence of 1 or more matches of the atom. +An atom followed by `?' matches a sequence of 0 or 1 matches of the atom. +.PP +A \fIbound\fR is `{' followed by an unsigned decimal integer, +possibly followed by `,' +possibly followed by another unsigned decimal integer, +always followed by `}'. +The integers must lie between 0 and RE_DUP_MAX (255\(dg) inclusive, +and if there are two of them, the first may not exceed the second. +An atom followed by a bound containing one integer \fIi\fR +and no comma matches +a sequence of exactly \fIi\fR matches of the atom. +An atom followed by a bound +containing one integer \fIi\fR and a comma matches +a sequence of \fIi\fR or more matches of the atom. +An atom followed by a bound +containing two integers \fIi\fR and \fIj\fR matches +a sequence of \fIi\fR through \fIj\fR (inclusive) matches of the atom. +.PP +An atom is a regular expression enclosed in `()' (matching a match for the +regular expression), +an empty set of `()' (matching the null string)\(dg, +a \fIbracket expression\fR (see below), `.' +(matching any single character), `^' (matching the null string at the +beginning of a line), `$' (matching the null string at the +end of a line), a `\e' followed by one of the characters +`^.[$()|*+?{\e' +(matching that character taken as an ordinary character), +a `\e' followed by any other character\(dg +(matching that character taken as an ordinary character, +as if the `\e' had not been present\(dg), +or a single character with no other significance (matching that character). +A `{' followed by a character other than a digit is an ordinary +character, not the beginning of a bound\(dg. +It is illegal to end an RE with `\e'. +.PP +A \fIbracket expression\fR is a list of characters enclosed in `[]'. +It normally matches any single character from the list (but see below). +If the list begins with `^', +it matches any single character +(but see below) \fInot\fR from the rest of the list. +If two characters in the list are separated by `\-', this is shorthand +for the full \fIrange\fR of characters between those two (inclusive) in the +collating sequence, +e.g. `[0\-9]' in ASCII matches any decimal digit. +It is illegal\(dg for two ranges to share an +endpoint, e.g. `a\-c\-e'. +Ranges are very collating-sequence-dependent, +and portable programs should avoid relying on them. +.PP +To include a literal `]' in the list, make it the first character +(following a possible `^'). +To include a literal `\-', make it the first or last character, +or the second endpoint of a range. +To use a literal `\-' as the first endpoint of a range, +enclose it in `[.' and `.]' to make it a collating element (see below). +With the exception of these and some combinations using `[' (see next +paragraphs), all other special characters, including `\e', lose their +special significance within a bracket expression. +.PP +Within a bracket expression, a collating element (a character, +a multi-character sequence that collates as if it were a single character, +or a collating-sequence name for either) +enclosed in `[.' and `.]' stands for the +sequence of characters of that collating element. +The sequence is a single element of the bracket expression's list. +A bracket expression containing a multi-character collating element +can thus match more than one character, +e.g. if the collating sequence includes a `ch' collating element, +then the RE `[[.ch.]]*c' matches the first five characters +of `chchcc'. +.PP +Within a bracket expression, a collating element enclosed in `[=' and +`=]' is an equivalence class, standing for the sequences of characters +of all collating elements equivalent to that one, including itself. +(If there are no other equivalent collating elements, +the treatment is as if the enclosing delimiters were `[.' and `.]'.) +For example, if o and \o'o^' are the members of an equivalence class, +then `[[=o=]]', `[[=\o'o^'=]]', and `[o\o'o^']' are all synonymous. +An equivalence class may not\(dg be an endpoint +of a range. +.PP +Within a bracket expression, the name of a \fIcharacter class\fR enclosed +in `[:' and `:]' stands for the list of all characters belonging to that +class. +Standard character class names are: +.PP +.RS +.nf +.ta 3c 6c 9c +alnum digit punct +alpha graph space +blank lower upper +cntrl print xdigit +.fi +.RE +.PP +These stand for the character classes defined in +.IR ctype (3). +A locale may provide others. +A character class may not be used as an endpoint of a range. +.PP +There are two special cases\(dg of bracket expressions: +the bracket expressions `[[:<:]]' and `[[:>:]]' match the null string at +the beginning and end of a word respectively. +A word is defined as a sequence of +word characters +which is neither preceded nor followed by +word characters. +A word character is an +.I alnum +character (as defined by +.IR ctype (3)) +or an underscore. +This is an extension, +compatible with but not specified by POSIX 1003.2, +and should be used with +caution in software intended to be portable to other systems. +.PP +In the event that an RE could match more than one substring of a given +string, +the RE matches the one starting earliest in the string. +If the RE could match more than one substring starting at that point, +it matches the longest. +Subexpressions also match the longest possible substrings, subject to +the constraint that the whole match be as long as possible, +with subexpressions starting earlier in the RE taking priority over +ones starting later. +Note that higher-level subexpressions thus take priority over +their lower-level component subexpressions. +.PP +Match lengths are measured in characters, not collating elements. +A null string is considered longer than no match at all. +For example, +`bb*' matches the three middle characters of `abbbc', +`(wee|week)(knights|nights)' matches all ten characters of `weeknights', +when `(.*).*' is matched against `abc' the parenthesized subexpression +matches all three characters, and +when `(a*)*' is matched against `bc' both the whole RE and the parenthesized +subexpression match the null string. +.PP +If case-independent matching is specified, +the effect is much as if all case distinctions had vanished from the +alphabet. +When an alphabetic that exists in multiple cases appears as an +ordinary character outside a bracket expression, it is effectively +transformed into a bracket expression containing both cases, +e.g. `x' becomes `[xX]'. +When it appears inside a bracket expression, all case counterparts +of it are added to the bracket expression, so that (e.g.) `[x]' +becomes `[xX]' and `[^x]' becomes `[^xX]'. +.PP +No particular limit is imposed on the length of REs\(dg. +Programs intended to be portable should not employ REs longer +than 256 bytes, +as an implementation can refuse to accept such REs and remain +POSIX-compliant. +.PP +Obsolete (``basic'') regular expressions differ in several respects. +`|', `+', and `?' are ordinary characters and there is no equivalent +for their functionality. +The delimiters for bounds are `\e{' and `\e}', +with `{' and `}' by themselves ordinary characters. +The parentheses for nested subexpressions are `\e(' and `\e)', +with `(' and `)' by themselves ordinary characters. +`^' is an ordinary character except at the beginning of the +RE or\(dg the beginning of a parenthesized subexpression, +`$' is an ordinary character except at the end of the +RE or\(dg the end of a parenthesized subexpression, +and `*' is an ordinary character if it appears at the beginning of the +RE or the beginning of a parenthesized subexpression +(after a possible leading `^'). +Finally, there is one new type of atom, a \fIback reference\fR: +`\e' followed by a non-zero decimal digit \fId\fR +matches the same sequence of characters +matched by the \fId\fRth parenthesized subexpression +(numbering subexpressions by the positions of their opening parentheses, +left to right), +so that (e.g.) `\e([bc]\e)\e1' matches `bb' or `cc' but not `bc'. +.SH SEE ALSO +regex(3) +.PP +POSIX 1003.2, section 2.8 (Regular Expression Notation). +.SH HISTORY +Written by Henry Spencer, based on the 1003.2 spec. +.SH BUGS +Having two kinds of REs is a botch. +.PP +The current 1003.2 spec says that `)' is an ordinary character in +the absence of an unmatched `('; +this was an unintentional result of a wording error, +and change is likely. +Avoid relying on it. +.PP +Back references are a dreadful botch, +posing major problems for efficient implementations. +They are also somewhat vaguely defined +(does +`a\e(\e(b\e)*\e2\e)*d' match `abbbd'?). +Avoid using them. +.PP +1003.2's specification of case-independent matching is vague. +The ``one case implies all cases'' definition given above +is current consensus among implementors as to the right interpretation. +.PP +The syntax for word boundaries is incredibly ugly. diff --git a/winsup/cygwin/regex/regex.h b/winsup/cygwin/regex/regex.h new file mode 100644 index 000000000..0ef0ace36 --- /dev/null +++ b/winsup/cygwin/regex/regex.h @@ -0,0 +1,83 @@ +#ifndef _REGEX_H_ +#define _REGEX_H_ /* never again */ +#include <sys/types.h> +/* ========= begin header generated by ./mkh ========= */ +#ifdef __cplusplus +extern "C" { +#endif + +#ifdef __INSIDE_CYGWIN__ +#define regcomp posix_regcomp +#define regerror posix_regerror +#define regexec posix_regexec +#define regfree posix_regfree +#endif + +/* === regex2.h === */ +typedef off_t regoff_t; +typedef struct { + int re_magic; + size_t re_nsub; /* number of parenthesized subexpressions */ + const char *re_endp; /* end pointer for REG_PEND */ + struct re_guts *re_g; /* none of your business :-) */ +} regex_t; +typedef struct { + regoff_t rm_so; /* start of match */ + regoff_t rm_eo; /* end of match */ +} regmatch_t; + + +/* === regcomp.c === */ +extern int regcomp(regex_t *, const char *, int); +#define REG_BASIC 0000 +#define REG_EXTENDED 0001 +#define REG_ICASE 0002 +#define REG_NOSUB 0004 +#define REG_NEWLINE 0010 +#define REG_NOSPEC 0020 +#define REG_PEND 0040 +#define REG_DUMP 0200 + + +/* === regerror.c === */ +#define REG_OKAY 0 +#define REG_NOMATCH 1 +#define REG_BADPAT 2 +#define REG_ECOLLATE 3 +#define REG_ECTYPE 4 +#define REG_EESCAPE 5 +#define REG_ESUBREG 6 +#define REG_EBRACK 7 +#define REG_EPAREN 8 +#define REG_EBRACE 9 +#define REG_BADBR 10 +#define REG_ERANGE 11 +#define REG_ESPACE 12 +#define REG_BADRPT 13 +#define REG_EMPTY 14 +#define REG_ASSERT 15 +#define REG_INVARG 16 +#define REG_ATOI 255 /* convert name to number (!) */ +#define REG_ITOA 0400 /* convert number to name (!) */ +extern size_t regerror(int, const regex_t *, char *, size_t); + + +/* === regexec.c === */ +extern int regexec(const regex_t *, const char *, size_t, regmatch_t [], int); +#define REG_NOTBOL 00001 +#define REG_NOTEOL 00002 +#define REG_STARTEND 00004 +#define REG_TRACE 00400 /* tracing of execution */ +#define REG_LARGE 01000 /* force large representation */ +#define REG_BACKR 02000 /* force use of backref code */ + + +/* === regfree.c === */ +extern void regfree(regex_t *); + + +#ifdef __cplusplus +} +#endif +/* ========= end header generated by ./mkh ========= */ +#endif diff --git a/winsup/cygwin/regex/regex2.h b/winsup/cygwin/regex/regex2.h new file mode 100644 index 000000000..58fd8d8a4 --- /dev/null +++ b/winsup/cygwin/regex/regex2.h @@ -0,0 +1,134 @@ +/* + * First, the stuff that ends up in the outside-world include file + = typedef off_t regoff_t; + = typedef struct { + = int re_magic; + = size_t re_nsub; // number of parenthesized subexpressions + = const char *re_endp; // end pointer for REG_PEND + = struct re_guts *re_g; // none of your business :-) + = } regex_t; + = typedef struct { + = regoff_t rm_so; // start of match + = regoff_t rm_eo; // end of match + = } regmatch_t; + */ +/* + * internals of regex_t + */ +#define MAGIC1 ((('r'^0200)<<8) | 'e') + +/* + * The internal representation is a *strip*, a sequence of + * operators ending with an endmarker. (Some terminology etc. is a + * historical relic of earlier versions which used multiple strips.) + * Certain oddities in the representation are there to permit running + * the machinery backwards; in particular, any deviation from sequential + * flow must be marked at both its source and its destination. Some + * fine points: + * + * - OPLUS_ and O_PLUS are *inside* the loop they create. + * - OQUEST_ and O_QUEST are *outside* the bypass they create. + * - OCH_ and O_CH are *outside* the multi-way branch they create, while + * OOR1 and OOR2 are respectively the end and the beginning of one of + * the branches. Note that there is an implicit OOR2 following OCH_ + * and an implicit OOR1 preceding O_CH. + * + * In state representations, an operator's bit is on to signify a state + * immediately *preceding* "execution" of that operator. + */ +typedef long sop; /* strip operator */ +typedef long sopno; +#define OPRMASK 0x7c000000 +#define OPDMASK 0x03ffffff +#define OPSHIFT (26) +#define OP(n) ((n)&OPRMASK) +#define OPND(n) ((n)&OPDMASK) +#define SOP(op, opnd) ((op)|(opnd)) +/* operators meaning operand */ +/* (back, fwd are offsets) */ +#define OEND (1<<OPSHIFT) /* endmarker - */ +#define OCHAR (2<<OPSHIFT) /* character unsigned char */ +#define OBOL (3<<OPSHIFT) /* left anchor - */ +#define OEOL (4<<OPSHIFT) /* right anchor - */ +#define OANY (5<<OPSHIFT) /* . - */ +#define OANYOF (6<<OPSHIFT) /* [...] set number */ +#define OBACK_ (7<<OPSHIFT) /* begin \d paren number */ +#define O_BACK (8<<OPSHIFT) /* end \d paren number */ +#define OPLUS_ (9<<OPSHIFT) /* + prefix fwd to suffix */ +#define O_PLUS (10<<OPSHIFT) /* + suffix back to prefix */ +#define OQUEST_ (11<<OPSHIFT) /* ? prefix fwd to suffix */ +#define O_QUEST (12<<OPSHIFT) /* ? suffix back to prefix */ +#define OLPAREN (13<<OPSHIFT) /* ( fwd to ) */ +#define ORPAREN (14<<OPSHIFT) /* ) back to ( */ +#define OCH_ (15<<OPSHIFT) /* begin choice fwd to OOR2 */ +#define OOR1 (16<<OPSHIFT) /* | pt. 1 back to OOR1 or OCH_ */ +#define OOR2 (17<<OPSHIFT) /* | pt. 2 fwd to OOR2 or O_CH */ +#define O_CH (18<<OPSHIFT) /* end choice back to OOR1 */ +#define OBOW (19<<OPSHIFT) /* begin word - */ +#define OEOW (20<<OPSHIFT) /* end word - */ + +/* + * Structure for [] character-set representation. Character sets are + * done as bit vectors, grouped 8 to a byte vector for compactness. + * The individual set therefore has both a pointer to the byte vector + * and a mask to pick out the relevant bit of each byte. A hash code + * simplifies testing whether two sets could be identical. + * + * This will get trickier for multicharacter collating elements. As + * preliminary hooks for dealing with such things, we also carry along + * a string of multi-character elements, and decide the size of the + * vectors at run time. + */ +typedef struct { + uch *ptr; /* -> uch [csetsize] */ + uch mask; /* bit within array */ + uch hash; /* hash code */ + size_t smultis; + char *multis; /* -> char[smulti] ab\0cd\0ef\0\0 */ +} cset; +/* note that CHadd and CHsub are unsafe, and CHIN doesn't yield 0/1 */ +#define CHadd(cs, c) ((cs)->ptr[(uch)(c)] |= (cs)->mask, (cs)->hash += (c)) +#define CHsub(cs, c) ((cs)->ptr[(uch)(c)] &= ~(cs)->mask, (cs)->hash -= (c)) +#define CHIN(cs, c) ((cs)->ptr[(uch)(c)] & (cs)->mask) +#define MCadd(p, cs, cp) mcadd(p, cs, cp) /* regcomp() internal fns */ +#define MCsub(p, cs, cp) mcsub(p, cs, cp) +#define MCin(p, cs, cp) mcin(p, cs, cp) + +/* stuff for character categories */ +typedef unsigned char cat_t; + +/* + * main compiled-expression structure + */ +struct re_guts { + int magic; +# define MAGIC2 ((('R'^0200)<<8)|'E') + sop *strip; /* malloced area for strip */ + int csetsize; /* number of bits in a cset vector */ + int ncsets; /* number of csets in use */ + cset *sets; /* -> cset [ncsets] */ + uch *setbits; /* -> uch[csetsize][ncsets/CHAR_BIT] */ + int cflags; /* copy of regcomp() cflags argument */ + sopno nstates; /* = number of sops */ + sopno firststate; /* the initial OEND (normally 0) */ + sopno laststate; /* the final OEND */ + int iflags; /* internal flags */ +# define USEBOL 01 /* used ^ */ +# define USEEOL 02 /* used $ */ +# define BAD 04 /* something wrong */ + int nbol; /* number of ^ used */ + int neol; /* number of $ used */ + int ncategories; /* how many character categories */ + cat_t *categories; /* ->catspace[-CHAR_MIN] */ + char *must; /* match must contain this string */ + int mlen; /* length of must */ + size_t nsub; /* copy of re_nsub */ + int backrefs; /* does it use back references? */ + sopno nplus; /* how deep does it nest +s? */ + /* catspace must be last */ + cat_t catspace[1]; /* actually [NC] */ +}; + +/* misc utilities */ +#define OUT (CHAR_MAX+1) /* a non-character value */ +#define ISWORD(c) (isalnum(c) || (c) == '_') diff --git a/winsup/cygwin/regex/regexec.c b/winsup/cygwin/regex/regexec.c new file mode 100644 index 000000000..5ab72e00e --- /dev/null +++ b/winsup/cygwin/regex/regexec.c @@ -0,0 +1,141 @@ +/* + * the outer shell of regexec() + * + * This file includes engine.c *twice*, after muchos fiddling with the + * macros that code uses. This lets the same code operate on two different + * representations for state sets. + */ +#include "winsup.h" +#include <sys/types.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <limits.h> +#include <ctype.h> +#include "regex.h" + +#include "utils.h" +#include "regex2.h" + +#ifdef lint +static int nope = 0; /* for use in asserts; shuts lint up */ +#endif + +/* macros for manipulating states, small version */ +#define states unsigned +#define states1 unsigned /* for later use in regexec() decision */ +#define CLEAR(v) ((v) = 0) +#define SET0(v, n) ((v) &= ~((unsigned)1 << (n))) +#define SET1(v, n) ((v) |= (unsigned)1 << (n)) +#define ISSET(v, n) ((v) & ((unsigned)1 << (n))) +#define ASSIGN(d, s) ((d) = (s)) +#define EQ(a, b) ((a) == (b)) +#define STATEVARS int dummy /* dummy version */ +#define STATESETUP(m, n) /* nothing */ +#define STATETEARDOWN(m) /* nothing */ +#define SETUP(v) ((v) = 0) +#define onestate unsigned +#define INIT(o, n) ((o) = (unsigned)1 << (n)) +#define INC(o) ((o) <<= 1) +#define ISSTATEIN(v, o) ((v) & (o)) +/* some abbreviations; note that some of these know variable names! */ +/* do "if I'm here, I can also be there" etc without branches */ +#define FWD(dst, src, n) ((dst) |= ((unsigned)(src)&(here)) << (n)) +#define BACK(dst, src, n) ((dst) |= ((unsigned)(src)&(here)) >> (n)) +#define ISSETBACK(v, n) ((v) & ((unsigned)here >> (n))) +/* function names */ +#define SNAMES /* engine.c looks after details */ + +#include "engine.c" + +/* now undo things */ +#undef states +#undef CLEAR +#undef SET0 +#undef SET1 +#undef ISSET +#undef ASSIGN +#undef EQ +#undef STATEVARS +#undef STATESETUP +#undef STATETEARDOWN +#undef SETUP +#undef onestate +#undef INIT +#undef INC +#undef ISSTATEIN +#undef FWD +#undef BACK +#undef ISSETBACK +#undef SNAMES + +/* macros for manipulating states, large version */ +#define states char * +#define CLEAR(v) memset(v, 0, m->g->nstates) +#define SET0(v, n) ((v)[n] = 0) +#define SET1(v, n) ((v)[n] = 1) +#define ISSET(v, n) ((v)[n]) +#define ASSIGN(d, s) memcpy(d, s, m->g->nstates) +#define EQ(a, b) (memcmp(a, b, m->g->nstates) == 0) +#define STATEVARS int vn; char *space +#define STATESETUP(m, nv) { (m)->space = malloc((nv)*(m)->g->nstates); \ + if ((m)->space == NULL) return(REG_ESPACE); \ + (m)->vn = 0; } +#define STATETEARDOWN(m) { free((m)->space); } +#define SETUP(v) ((v) = &m->space[m->vn++ * m->g->nstates]) +#define onestate int +#define INIT(o, n) ((o) = (n)) +#define INC(o) ((o)++) +#define ISSTATEIN(v, o) ((v)[o]) +/* some abbreviations; note that some of these know variable names! */ +/* do "if I'm here, I can also be there" etc without branches */ +#define FWD(dst, src, n) ((dst)[here+(n)] |= (src)[here]) +#define BACK(dst, src, n) ((dst)[here-(n)] |= (src)[here]) +#define ISSETBACK(v, n) ((v)[here - (n)]) +/* function names */ +#define LNAMES /* flag */ + +#include "engine.c" + +/* + - regexec - interface for matching + = extern int regexec(const regex_t *, const char *, size_t, \ + = regmatch_t [], int); + = #define REG_NOTBOL 00001 + = #define REG_NOTEOL 00002 + = #define REG_STARTEND 00004 + = #define REG_TRACE 00400 // tracing of execution + = #define REG_LARGE 01000 // force large representation + = #define REG_BACKR 02000 // force use of backref code + * + * We put this here so we can exploit knowledge of the state representation + * when choosing which matcher to call. Also, by this point the matchers + * have been prototyped. + */ +int /* 0 success, REG_NOMATCH failure */ +regexec(preg, string, nmatch, pmatch, eflags) +const regex_t *preg; +const char *string; +size_t nmatch; +regmatch_t pmatch[]; +int eflags; +{ + register struct re_guts *g = preg->re_g; +#ifdef REDEBUG +# define GOODFLAGS(f) (f) +#else +# define GOODFLAGS(f) ((f)&(REG_NOTBOL|REG_NOTEOL|REG_STARTEND)) +#endif + + if (preg->re_magic != MAGIC1 || g->magic != MAGIC2) + return(REG_BADPAT); + assert(!(g->iflags&BAD)); + if (g->iflags&BAD) /* backstop for no-debug case */ + return(REG_BADPAT); + eflags = GOODFLAGS(eflags); + + if (g->nstates <= CHAR_BIT*sizeof(states1) && !(eflags®_LARGE)) + return(smatcher(g, (char *)string, nmatch, pmatch, eflags)); + else + return(lmatcher(g, (char *)string, nmatch, pmatch, eflags)); +} diff --git a/winsup/cygwin/regex/regfree.c b/winsup/cygwin/regex/regfree.c new file mode 100644 index 000000000..b8be143d8 --- /dev/null +++ b/winsup/cygwin/regex/regfree.c @@ -0,0 +1,38 @@ +#include "winsup.h" +#include <sys/types.h> +#include <stdio.h> +#include <stdlib.h> +#include "regex.h" + +#include "utils.h" +#include "regex2.h" + +/* + - regfree - free everything + = extern void regfree(regex_t *); + */ +void +regfree(preg) +regex_t *preg; +{ + register struct re_guts *g; + + if (preg->re_magic != MAGIC1) /* oops */ + return; /* nice to complain, but hard */ + + g = preg->re_g; + if (g == NULL || g->magic != MAGIC2) /* oops again */ + return; + preg->re_magic = 0; /* mark it invalid */ + g->magic = 0; /* mark it invalid */ + + if (g->strip != NULL) + free((char *)g->strip); + if (g->sets != NULL) + free((char *)g->sets); + if (g->setbits != NULL) + free((char *)g->setbits); + if (g->must != NULL) + free(g->must); + free((char *)g); +} diff --git a/winsup/cygwin/regex/tests b/winsup/cygwin/regex/tests new file mode 100644 index 000000000..e4d928dad --- /dev/null +++ b/winsup/cygwin/regex/tests @@ -0,0 +1,477 @@ +# regular expression test set +# Lines are at least three fields, separated by one or more tabs. "" stands +# for an empty field. First field is an RE. Second field is flags. If +# C flag given, regcomp() is expected to fail, and the third field is the +# error name (minus the leading REG_). +# +# Otherwise it is expected to succeed, and the third field is the string to +# try matching it against. If there is no fourth field, the match is +# expected to fail. If there is a fourth field, it is the substring that +# the RE is expected to match. If there is a fifth field, it is a comma- +# separated list of what the subexpressions should match, with - indicating +# no match for that one. In both the fourth and fifth fields, a (sub)field +# starting with @ indicates that the (sub)expression is expected to match +# a null string followed by the stuff after the @; this provides a way to +# test where null strings match. The character `N' in REs and strings +# is newline, `S' is space, `T' is tab, `Z' is NUL. +# +# The full list of flags: +# - placeholder, does nothing +# b RE is a BRE, not an ERE +# & try it as both an ERE and a BRE +# C regcomp() error expected, third field is error name +# i REG_ICASE +# m ("mundane") REG_NOSPEC +# s REG_NOSUB (not really testable) +# n REG_NEWLINE +# ^ REG_NOTBOL +# $ REG_NOTEOL +# # REG_STARTEND (see below) +# p REG_PEND +# +# For REG_STARTEND, the start/end offsets are those of the substring +# enclosed in (). + +# basics +a & a a +abc & abc abc +abc|de - abc abc +a|b|c - abc a + +# parentheses and perversions thereof +a(b)c - abc abc +a\(b\)c b abc abc +a( C EPAREN +a( b a( a( +a\( - a( a( +a\( bC EPAREN +a\(b bC EPAREN +a(b C EPAREN +a(b b a(b a(b +# gag me with a right parenthesis -- 1003.2 goofed here (my fault, partly) +a) - a) a) +) - ) ) +# end gagging (in a just world, those *should* give EPAREN) +a) b a) a) +a\) bC EPAREN +\) bC EPAREN +a()b - ab ab +a\(\)b b ab ab + +# anchoring and REG_NEWLINE +^abc$ & abc abc +a^b - a^b +a^b b a^b a^b +a$b - a$b +a$b b a$b a$b +^ & abc @abc +$ & abc @ +^$ & "" @ +$^ - "" @ +\($\)\(^\) b "" @ +# stop retching, those are legitimate (although disgusting) +^^ - "" @ +$$ - "" @ +b$ & abNc +b$ &n abNc b +^b$ & aNbNc +^b$ &n aNbNc b +^$ &n aNNb @Nb +^$ n abc +^$ n abcN @ +$^ n aNNb @Nb +\($\)\(^\) bn aNNb @Nb +^^ n^ aNNb @Nb +$$ n aNNb @NN +^a ^ a +a$ $ a +^a ^n aNb +^b ^n aNb b +a$ $n bNa +b$ $n bNa b +a*(^b$)c* - b b +a*\(^b$\)c* b b b + +# certain syntax errors and non-errors +| C EMPTY +| b | | +* C BADRPT +* b * * ++ C BADRPT +? C BADRPT +"" &C EMPTY +() - abc @abc +\(\) b abc @abc +a||b C EMPTY +|ab C EMPTY +ab| C EMPTY +(|a)b C EMPTY +(a|)b C EMPTY +(*a) C BADRPT +(+a) C BADRPT +(?a) C BADRPT +({1}a) C BADRPT +\(\{1\}a\) bC BADRPT +(a|*b) C BADRPT +(a|+b) C BADRPT +(a|?b) C BADRPT +(a|{1}b) C BADRPT +^* C BADRPT +^* b * * +^+ C BADRPT +^? C BADRPT +^{1} C BADRPT +^\{1\} bC BADRPT + +# metacharacters, backslashes +a.c & abc abc +a[bc]d & abd abd +a\*c & a*c a*c +a\\b & a\b a\b +a\\\*b & a\*b a\*b +a\bc & abc abc +a\ &C EESCAPE +a\\bc & a\bc a\bc +\{ bC BADRPT +a\[b & a[b a[b +a[b &C EBRACK +# trailing $ is a peculiar special case for the BRE code +a$ & a a +a$ & a$ +a\$ & a +a\$ & a$ a$ +a\\$ & a +a\\$ & a$ +a\\$ & a\$ +a\\$ & a\ a\ + +# back references, ugh +a\(b\)\2c bC ESUBREG +a\(b\1\)c bC ESUBREG +a\(b*\)c\1d b abbcbbd abbcbbd bb +a\(b*\)c\1d b abbcbd +a\(b*\)c\1d b abbcbbbd +^\(.\)\1 b abc +a\([bc]\)\1d b abcdabbd abbd b +a\(\([bc]\)\2\)*d b abbccd abbccd +a\(\([bc]\)\2\)*d b abbcbd +# actually, this next one probably ought to fail, but the spec is unclear +a\(\(b\)*\2\)*d b abbbd abbbd +# here is a case that no NFA implementation does right +\(ab*\)[ab]*\1 b ababaaa ababaaa a +# check out normal matching in the presence of back refs +\(a\)\1bcd b aabcd aabcd +\(a\)\1bc*d b aabcd aabcd +\(a\)\1bc*d b aabd aabd +\(a\)\1bc*d b aabcccd aabcccd +\(a\)\1bc*[ce]d b aabcccd aabcccd +^\(a\)\1b\(c\)*cd$ b aabcccd aabcccd + +# ordinary repetitions +ab*c & abc abc +ab+c - abc abc +ab?c - abc abc +a\(*\)b b a*b a*b +a\(**\)b b ab ab +a\(***\)b bC BADRPT +*a b *a *a +**a b a a +***a bC BADRPT + +# the dreaded bounded repetitions +{ & { { +{abc & {abc {abc +{1 C BADRPT +{1} C BADRPT +a{b & a{b a{b +a{1}b - ab ab +a\{1\}b b ab ab +a{1,}b - ab ab +a\{1,\}b b ab ab +a{1,2}b - aab aab +a\{1,2\}b b aab aab +a{1 C EBRACE +a\{1 bC EBRACE +a{1a C EBRACE +a\{1a bC EBRACE +a{1a} C BADBR +a\{1a\} bC BADBR +a{,2} - a{,2} a{,2} +a\{,2\} bC BADBR +a{,} - a{,} a{,} +a\{,\} bC BADBR +a{1,x} C BADBR +a\{1,x\} bC BADBR +a{1,x C EBRACE +a\{1,x bC EBRACE +a{300} C BADBR +a\{300\} bC BADBR +a{1,0} C BADBR +a\{1,0\} bC BADBR +ab{0,0}c - abcac ac +ab\{0,0\}c b abcac ac +ab{0,1}c - abcac abc +ab\{0,1\}c b abcac abc +ab{0,3}c - abbcac abbc +ab\{0,3\}c b abbcac abbc +ab{1,1}c - acabc abc +ab\{1,1\}c b acabc abc +ab{1,3}c - acabc abc +ab\{1,3\}c b acabc abc +ab{2,2}c - abcabbc abbc +ab\{2,2\}c b abcabbc abbc +ab{2,4}c - abcabbc abbc +ab\{2,4\}c b abcabbc abbc +((a{1,10}){1,10}){1,10} - a a a,a + +# multiple repetitions +a** &C BADRPT +a++ C BADRPT +a?? C BADRPT +a*+ C BADRPT +a*? C BADRPT +a+* C BADRPT +a+? C BADRPT +a?* C BADRPT +a?+ C BADRPT +a{1}{1} C BADRPT +a*{1} C BADRPT +a+{1} C BADRPT +a?{1} C BADRPT +a{1}* C BADRPT +a{1}+ C BADRPT +a{1}? C BADRPT +a*{b} - a{b} a{b} +a\{1\}\{1\} bC BADRPT +a*\{1\} bC BADRPT +a\{1\}* bC BADRPT + +# brackets, and numerous perversions thereof +a[b]c & abc abc +a[ab]c & abc abc +a[^ab]c & adc adc +a[]b]c & a]c a]c +a[[b]c & a[c a[c +a[-b]c & a-c a-c +a[^]b]c & adc adc +a[^-b]c & adc adc +a[b-]c & a-c a-c +a[b &C EBRACK +a[] &C EBRACK +a[1-3]c & a2c a2c +a[3-1]c &C ERANGE +a[1-3-5]c &C ERANGE +a[[.-.]--]c & a-c a-c +a[1- &C ERANGE +a[[. &C EBRACK +a[[.x &C EBRACK +a[[.x. &C EBRACK +a[[.x.] &C EBRACK +a[[.x.]] & ax ax +a[[.x,.]] &C ECOLLATE +a[[.one.]]b & a1b a1b +a[[.notdef.]]b &C ECOLLATE +a[[.].]]b & a]b a]b +a[[:alpha:]]c & abc abc +a[[:notdef:]]c &C ECTYPE +a[[: &C EBRACK +a[[:alpha &C EBRACK +a[[:alpha:] &C EBRACK +a[[:alpha,:] &C ECTYPE +a[[:]:]]b &C ECTYPE +a[[:-:]]b &C ECTYPE +a[[:alph:]] &C ECTYPE +a[[:alphabet:]] &C ECTYPE +[[:alnum:]]+ - -%@a0X- a0X +[[:alpha:]]+ - -%@aX0- aX +[[:blank:]]+ - aSSTb SST +[[:cntrl:]]+ - aNTb NT +[[:digit:]]+ - a019b 019 +[[:graph:]]+ - Sa%bS a%b +[[:lower:]]+ - AabC ab +[[:print:]]+ - NaSbN aSb +[[:punct:]]+ - S%-&T %-& +[[:space:]]+ - aSNTb SNT +[[:upper:]]+ - aBCd BC +[[:xdigit:]]+ - p0f3Cq 0f3C +a[[=b=]]c & abc abc +a[[= &C EBRACK +a[[=b &C EBRACK +a[[=b= &C EBRACK +a[[=b=] &C EBRACK +a[[=b,=]] &C ECOLLATE +a[[=one=]]b & a1b a1b + +# complexities +a(((b)))c - abc abc +a(b|(c))d - abd abd +a(b*|c)d - abbd abbd +# just gotta have one DFA-buster, of course +a[ab]{20} - aaaaabaaaabaaaabaaaab aaaaabaaaabaaaabaaaab +# and an inline expansion in case somebody gets tricky +a[ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab] - aaaaabaaaabaaaabaaaab aaaaabaaaabaaaabaaaab +# and in case somebody just slips in an NFA... +a[ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab][ab](wee|week)(knights|night) - aaaaabaaaabaaaabaaaabweeknights aaaaabaaaabaaaabaaaabweeknights +# fish for anomalies as the number of states passes 32 +12345678901234567890123456789 - a12345678901234567890123456789b 12345678901234567890123456789 +123456789012345678901234567890 - a123456789012345678901234567890b 123456789012345678901234567890 +1234567890123456789012345678901 - a1234567890123456789012345678901b 1234567890123456789012345678901 +12345678901234567890123456789012 - a12345678901234567890123456789012b 12345678901234567890123456789012 +123456789012345678901234567890123 - a123456789012345678901234567890123b 123456789012345678901234567890123 +# and one really big one, beyond any plausible word width +1234567890123456789012345678901234567890123456789012345678901234567890 - a1234567890123456789012345678901234567890123456789012345678901234567890b 1234567890123456789012345678901234567890123456789012345678901234567890 +# fish for problems as brackets go past 8 +[ab][cd][ef][gh][ij][kl][mn] - xacegikmoq acegikm +[ab][cd][ef][gh][ij][kl][mn][op] - xacegikmoq acegikmo +[ab][cd][ef][gh][ij][kl][mn][op][qr] - xacegikmoqy acegikmoq +[ab][cd][ef][gh][ij][kl][mn][op][q] - xacegikmoqy acegikmoq + +# subtleties of matching +abc & xabcy abc +a\(b\)?c\1d b acd +aBc i Abc Abc +a[Bc]*d i abBCcd abBCcd +0[[:upper:]]1 &i 0a1 0a1 +0[[:lower:]]1 &i 0A1 0A1 +a[^b]c &i abc +a[^b]c &i aBc +a[^b]c &i adc adc +[a]b[c] - abc abc +[a]b[a] - aba aba +[abc]b[abc] - abc abc +[abc]b[abd] - abd abd +a(b?c)+d - accd accd +(wee|week)(knights|night) - weeknights weeknights +(we|wee|week|frob)(knights|night|day) - weeknights weeknights +a[bc]d - xyzaaabcaababdacd abd +a[ab]c - aaabc abc +abc s abc abc +a* & b @b + +# Let's have some fun -- try to match a C comment. +# first the obvious, which looks okay at first glance... +/\*.*\*/ - /*x*/ /*x*/ +# but... +/\*.*\*/ - /*x*/y/*z*/ /*x*/y/*z*/ +# okay, we must not match */ inside; try to do that... +/\*([^*]|\*[^/])*\*/ - /*x*/ /*x*/ +/\*([^*]|\*[^/])*\*/ - /*x*/y/*z*/ /*x*/ +# but... +/\*([^*]|\*[^/])*\*/ - /*x**/y/*z*/ /*x**/y/*z*/ +# and a still fancier version, which does it right (I think)... +/\*([^*]|\*+[^*/])*\*+/ - /*x*/ /*x*/ +/\*([^*]|\*+[^*/])*\*+/ - /*x*/y/*z*/ /*x*/ +/\*([^*]|\*+[^*/])*\*+/ - /*x**/y/*z*/ /*x**/ +/\*([^*]|\*+[^*/])*\*+/ - /*x****/y/*z*/ /*x****/ +/\*([^*]|\*+[^*/])*\*+/ - /*x**x*/y/*z*/ /*x**x*/ +/\*([^*]|\*+[^*/])*\*+/ - /*x***x/y/*z*/ /*x***x/y/*z*/ + +# subexpressions +.* - abc abc - +a(b)(c)d - abcd abcd b,c +a(((b)))c - abc abc b,b,b +a(b|(c))d - abd abd b,- +a(b*|c|e)d - abbd abbd bb +a(b*|c|e)d - acd acd c +a(b*|c|e)d - ad ad @d +a(b?)c - abc abc b +a(b?)c - ac ac @c +a(b+)c - abc abc b +a(b+)c - abbbc abbbc bbb +a(b*)c - ac ac @c +(a|ab)(bc([de]+)f|cde) - abcdef abcdef a,bcdef,de +# the regression tester only asks for 9 subexpressions +a(b)(c)(d)(e)(f)(g)(h)(i)(j)k - abcdefghijk abcdefghijk b,c,d,e,f,g,h,i,j +a(b)(c)(d)(e)(f)(g)(h)(i)(j)(k)l - abcdefghijkl abcdefghijkl b,c,d,e,f,g,h,i,j,k +a([bc]?)c - abc abc b +a([bc]?)c - ac ac @c +a([bc]+)c - abc abc b +a([bc]+)c - abcc abcc bc +a([bc]+)bc - abcbc abcbc bc +a(bb+|b)b - abb abb b +a(bbb+|bb+|b)b - abb abb b +a(bbb+|bb+|b)b - abbb abbb bb +a(bbb+|bb+|b)bb - abbb abbb b +(.*).* - abcdef abcdef abcdef +(a*)* - bc @b @b + +# do we get the right subexpression when it is used more than once? +a(b|c)*d - ad ad - +a(b|c)*d - abcd abcd c +a(b|c)+d - abd abd b +a(b|c)+d - abcd abcd c +a(b|c?)+d - ad ad @d +a(b|c?)+d - abcd abcd @d +a(b|c){0,0}d - ad ad - +a(b|c){0,1}d - ad ad - +a(b|c){0,1}d - abd abd b +a(b|c){0,2}d - ad ad - +a(b|c){0,2}d - abcd abcd c +a(b|c){0,}d - ad ad - +a(b|c){0,}d - abcd abcd c +a(b|c){1,1}d - abd abd b +a(b|c){1,1}d - acd acd c +a(b|c){1,2}d - abd abd b +a(b|c){1,2}d - abcd abcd c +a(b|c){1,}d - abd abd b +a(b|c){1,}d - abcd abcd c +a(b|c){2,2}d - acbd acbd b +a(b|c){2,2}d - abcd abcd c +a(b|c){2,4}d - abcd abcd c +a(b|c){2,4}d - abcbd abcbd b +a(b|c){2,4}d - abcbcd abcbcd c +a(b|c){2,}d - abcd abcd c +a(b|c){2,}d - abcbd abcbd b +a(b+|((c)*))+d - abd abd @d,@d,- +a(b+|((c)*))+d - abcd abcd @d,@d,- + +# check out the STARTEND option +[abc] &# a(b)c b +[abc] &# a(d)c +[abc] &# a(bc)d b +[abc] &# a(dc)d c +. &# a()c +b.*c &# b(bc)c bc +b.* &# b(bc)c bc +.*c &# b(bc)c bc + +# plain strings, with the NOSPEC flag +abc m abc abc +abc m xabcy abc +abc m xyz +a*b m aba*b a*b +a*b m ab +"" mC EMPTY + +# cases involving NULs +aZb & a a +aZb &p a +aZb &p# (aZb) aZb +aZ*b &p# (ab) ab +a.b &# (aZb) aZb +a.* &# (aZb)c aZb + +# word boundaries (ick) +[[:<:]]a & a a +[[:<:]]a & ba +[[:<:]]a & -a a +a[[:>:]] & a a +a[[:>:]] & ab +a[[:>:]] & a- a +[[:<:]]a.c[[:>:]] & axcd-dayc-dazce-abc abc +[[:<:]]a.c[[:>:]] & axcd-dayc-dazce-abc-q abc +[[:<:]]a.c[[:>:]] & axc-dayc-dazce-abc axc +[[:<:]]b.c[[:>:]] & a_bxc-byc_d-bzc-q bzc +[[:<:]].x..[[:>:]] & y_xa_-_xb_y-_xc_-axdc _xc_ +[[:<:]]a_b[[:>:]] & x_a_b + +# past problems, and suspected problems +(A[1])|(A[2])|(A[3])|(A[4])|(A[5])|(A[6])|(A[7])|(A[8])|(A[9])|(A[A]) - A1 A1 +abcdefghijklmnop i abcdefghijklmnop abcdefghijklmnop +abcdefghijklmnopqrstuv i abcdefghijklmnopqrstuv abcdefghijklmnopqrstuv +(ALAK)|(ALT[AB])|(CC[123]1)|(CM[123]1)|(GAMC)|(LC[23][EO ])|(SEM[1234])|(SL[ES][12])|(SLWW)|(SLF )|(SLDT)|(VWH[12])|(WH[34][EW])|(WP1[ESN]) - CC11 CC11 +CC[13]1|a{21}[23][EO][123][Es][12]a{15}aa[34][EW]aaaaaaa[X]a - CC11 CC11 +Char \([a-z0-9_]*\)\[.* b Char xyz[k Char xyz[k xyz +a?b - ab ab +-\{0,1\}[0-9]*$ b -5 -5 +a*a*a*a*a*a*a* & aaaaaa aaaaaa diff --git a/winsup/cygwin/regex/utils.h b/winsup/cygwin/regex/utils.h new file mode 100644 index 000000000..1a997ac8f --- /dev/null +++ b/winsup/cygwin/regex/utils.h @@ -0,0 +1,22 @@ +/* utility definitions */ +#ifdef _POSIX2_RE_DUP_MAX +#define DUPMAX _POSIX2_RE_DUP_MAX +#else +#define DUPMAX 255 +#endif +#define INFINITY (DUPMAX + 1) +#define NC (CHAR_MAX - CHAR_MIN + 1) +typedef unsigned char uch; + +/* switch off assertions (if not already off) if no REDEBUG */ +#ifndef REDEBUG +#ifndef NDEBUG +#define NDEBUG /* no assertions please */ +#endif +#endif +#include <assert.h> + +/* for old systems with bcopy() but no memmove() */ +#ifdef USEBCOPY +#define memmove(d, s, c) bcopy(s, d, c) +#endif diff --git a/winsup/cygwin/regexp/regerror.c b/winsup/cygwin/regexp/v8_regerror.c index 0f6159bb7..56d63ff2f 100644 --- a/winsup/cygwin/regexp/regerror.c +++ b/winsup/cygwin/regexp/v8_regerror.c @@ -12,7 +12,7 @@ details. */ #include "regexp.h" #include <stdio.h> -void +void __declspec(dllexport) regerror(const char *s __attribute__ ((unused))) { #ifdef ERRAVAIL diff --git a/winsup/cygwin/regexp/regexp.c b/winsup/cygwin/regexp/v8_regexp.c index 2f60d6463..fef1267a0 100644 --- a/winsup/cygwin/regexp/regexp.c +++ b/winsup/cygwin/regexp/v8_regexp.c @@ -204,7 +204,7 @@ STATIC int strcspn (char *, char *); * Beware that the optimization-preparation code in here knows about some * of the structure of the compiled regexp. */ -regexp * +regexp * __declspec(dllexport) regcomp(exp) const char *exp; { @@ -788,7 +788,7 @@ STATIC char *regprop __P((char *)); /* - regexec - match a regexp against a string */ -int +int __declspec(dllexport) regexec(prog, string) register const regexp *prog; register const char *string; diff --git a/winsup/cygwin/regexp/regsub.c b/winsup/cygwin/regexp/v8_regsub.c index e544072c1..aa95b876a 100644 --- a/winsup/cygwin/regexp/regsub.c +++ b/winsup/cygwin/regexp/v8_regsub.c @@ -25,6 +25,7 @@ static char *rcsid = "$Id$"; #endif /* not lint */ #endif +#include "winsup.h" #include "regexp.h" #include <stdio.h> #include <string.h> @@ -39,7 +40,7 @@ static char *rcsid = "$Id$"; /* - regsub - perform substitutions after a regexp match */ -void +void __declspec(dllexport) regsub(prog, source, dest) const regexp *prog; const char *source; diff --git a/winsup/cygwin/rmsym b/winsup/cygwin/rmsym new file mode 100755 index 000000000..685802684 --- /dev/null +++ b/winsup/cygwin/rmsym @@ -0,0 +1,11 @@ +#!/bin/sh +lib=$1; shift +nm=$1; shift +ar=$1; shift +ranlib=$1; shift +grepit=`echo $* | sed 's/ /|__imp__/g'` +[ -n "$grepit" ] && grepit="__imp__$grepit" +objs=`$nm $lib | awk -F: '/^d[0-9]*.o:/ {obj=$1} '"/$grepit/"'{print obj}'` +[ -n "$objs" ] || exit 1 +$ar d $lib $objs +$ranlib $lib |