hyung-hwan/hawk
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fixed a parser bug when handling the print/printf statement without redirection.

Browse Source 
fixed a bug where a single dot was converted to 0 when passed via -v (e.g. -vTDIR=.)
added more test cases
This commit is contained in:
hyung-hwan 2020-12-01 10:22:17 +00:00
parent 35829a524b
commit 2dd6029064
20 changed files with 64957 additions and 50 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
hawk/Makefile.in
View File

@ -170,8 +170,8 @@ am__DIST_COMMON = $(srcdir)/Makefile.in $(top_srcdir)/ac/ar-lib \
$(top_srcdir)/ac/config.sub $(top_srcdir)/ac/install-sh \
$(top_srcdir)/ac/ltmain.sh $(top_srcdir)/ac/missing \
$(top_srcdir)/pkgs/hawk.spec.in ac/ar-lib ac/compile \
ac/config.guess ac/config.sub ac/install-sh ac/ltmain.sh \
ac/missing
ac/config.guess ac/config.sub ac/depcomp ac/install-sh \
ac/ltmain.sh ac/missing
DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
distdir = $(PACKAGE)-$(VERSION)
top_distdir = $(distdir)
@ -354,7 +354,6 @@ pdfdir = @pdfdir@
prefix = @prefix@
program_transform_name = @program_transform_name@
psdir = @psdir@
runstatedir = @runstatedir@
sbindir = @sbindir@
sharedstatedir = @sharedstatedir@
srcdir = @srcdir@

1
hawk/bin/Makefile.in
View File

@ -323,7 +323,6 @@ pdfdir = @pdfdir@
prefix = @prefix@
program_transform_name = @program_transform_name@
psdir = @psdir@
runstatedir = @runstatedir@
sbindir = @sbindir@
sharedstatedir = @sharedstatedir@
srcdir = @srcdir@

24
hawk/configure vendored
View File

@ -783,7 +783,6 @@ infodir
docdir
oldincludedir
includedir
runstatedir
localstatedir
sharedstatedir
sysconfdir
@ -884,7 +883,6 @@ datadir='${datarootdir}'
sysconfdir='${prefix}/etc'
sharedstatedir='${prefix}/com'
localstatedir='${prefix}/var'
runstatedir='${localstatedir}/run'
includedir='${prefix}/include'
oldincludedir='/usr/include'
docdir='${datarootdir}/doc/${PACKAGE_TARNAME}'
@ -1137,15 +1135,6 @@ do
| -silent | --silent | --silen | --sile | --sil)
silent=yes ;;
-runstatedir | --runstatedir | --runstatedi | --runstated \
| --runstate | --runstat | --runsta | --runst | --runs \
| --run | --ru | --r)
ac_prev=runstatedir ;;
-runstatedir=* | --runstatedir=* | --runstatedi=* | --runstated=* \
| --runstate=* | --runstat=* | --runsta=* | --runst=* | --runs=* \
| --run=* | --ru=* | --r=*)
runstatedir=$ac_optarg ;;
-sbindir | --sbindir | --sbindi | --sbind | --sbin | --sbi | --sb)
ac_prev=sbindir ;;
-sbindir=* | --sbindir=* | --sbindi=* | --sbind=* | --sbin=* \
@ -1283,7 +1272,7 @@ fi
for ac_var in exec_prefix prefix bindir sbindir libexecdir datarootdir \
datadir sysconfdir sharedstatedir localstatedir includedir \
oldincludedir docdir infodir htmldir dvidir pdfdir psdir \
libdir localedir mandir runstatedir
libdir localedir mandir
do
eval ac_val=\$$ac_var
# Remove trailing slashes.
@ -1436,7 +1425,6 @@ Fine tuning of the installation directories:
--sysconfdir=DIR read-only single-machine data [PREFIX/etc]
--sharedstatedir=DIR modifiable architecture-independent data [PREFIX/com]
--localstatedir=DIR modifiable single-machine data [PREFIX/var]
--runstatedir=DIR modifiable per-process data [LOCALSTATEDIR/run]
--libdir=DIR object code libraries [EPREFIX/lib]
--includedir=DIR C header files [PREFIX/include]
--oldincludedir=DIR C header files for non-gcc [/usr/include]
@ -18080,7 +18068,7 @@ else
We can't simply define LARGE_OFF_T to be 9223372036854775807,
since some C++ compilers masquerading as C compilers
incorrectly reject 9223372036854775807. */
#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
&& LARGE_OFF_T % 2147483647 == 1)
? 1 : -1];
@ -18126,7 +18114,7 @@ else
We can't simply define LARGE_OFF_T to be 9223372036854775807,
since some C++ compilers masquerading as C compilers
incorrectly reject 9223372036854775807. */
#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
&& LARGE_OFF_T % 2147483647 == 1)
? 1 : -1];
@ -18150,7 +18138,7 @@ rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
We can't simply define LARGE_OFF_T to be 9223372036854775807,
since some C++ compilers masquerading as C compilers
incorrectly reject 9223372036854775807. */
#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
&& LARGE_OFF_T % 2147483647 == 1)
? 1 : -1];
@ -18195,7 +18183,7 @@ else
We can't simply define LARGE_OFF_T to be 9223372036854775807,
since some C++ compilers masquerading as C compilers
incorrectly reject 9223372036854775807. */
#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
&& LARGE_OFF_T % 2147483647 == 1)
? 1 : -1];
@ -18219,7 +18207,7 @@ rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
We can't simply define LARGE_OFF_T to be 9223372036854775807,
since some C++ compilers masquerading as C compilers
incorrectly reject 9223372036854775807. */
#define LARGE_OFF_T ((((off_t) 1 << 31) << 31) - 1 + (((off_t) 1 << 31) << 31))
#define LARGE_OFF_T (((off_t) 1 << 62) - 1 + ((off_t) 1 << 62))
int off_t_is_large[(LARGE_OFF_T % 2147483629 == 721
&& LARGE_OFF_T % 2147483647 == 1)
? 1 : -1];

1
hawk/lib/Makefile.in
View File

@ -561,7 +561,6 @@ pdfdir = @pdfdir@
prefix = @prefix@
program_transform_name = @program_transform_name@
psdir = @psdir@
runstatedir = @runstatedir@
sbindir = @sbindir@
sharedstatedir = @sharedstatedir@
srcdir = @srcdir@

1
hawk/lib/hawk-prv.h
View File

@ -208,6 +208,7 @@ typedef struct hawk_tok_t hawk_tok_t;
struct hawk_tok_t
{
int type;
int flags;
hawk_ooecs_t* name;
hawk_loc_t loc;
};

34
hawk/lib/parse.c
View File

@ -69,6 +69,7 @@
#define FMT_EUNDEF HAWK_T("undefined identifier '%.*js'")
#define FMT_EXKWNR HAWK_T("'%.*js' not recognized")
#define TOK_FLAGS_LPAREN_CLOSER (1 << 0)
enum tok_t
{
@ -3117,7 +3118,7 @@ static hawk_nde_t* parse_print (hawk_t* hawk, const hawk_loc_t* xloc)
}
eloc = hawk->tok.loc;
args = parse_expr_withdc (hawk, &eloc);
args = parse_expr_withdc(hawk, &eloc);
if (args == HAWK_NULL) goto oops;
args_tail = args;
@ -3150,7 +3151,7 @@ static hawk_nde_t* parse_print (hawk_t* hawk, const hawk_loc_t* xloc)
}
eloc = hawk->tok.loc;
args_tail->next = parse_expr_withdc (hawk, &eloc);
args_tail->next = parse_expr_withdc(hawk, &eloc);
if (args_tail->next == HAWK_NULL) goto oops;
tail_prev = args_tail;
@ -3184,7 +3185,7 @@ static hawk_nde_t* parse_print (hawk_t* hawk, const hawk_loc_t* xloc)
* print 1, 2, (3 > 4) > 5;
* print 1, 2, (3 > 4) > 5 + 6;
*/
if (in_parens == 1 && hawk->ptok.type == TOK_RPAREN &&
if (in_parens == 1 && hawk->ptok.type == TOK_RPAREN && (hawk->ptok.flags & TOK_FLAGS_LPAREN_CLOSER) &&
hawk->parse.lparen_last_closed == opening_lparen_seq)
{
in_parens = 2; /* confirmed */
@ -3192,6 +3193,10 @@ static hawk_nde_t* parse_print (hawk_t* hawk, const hawk_loc_t* xloc)
if (in_parens != 2 && gm_in_parens != 2 && args_tail->type == HAWK_NDE_EXP_BIN)
{
/* check if the expression ends with an output redirector
* and take out the part after the redirector and make it
* the output target */
int i;
hawk_nde_exp_t* ep = (hawk_nde_exp_t*)args_tail;
static struct
@ -3233,8 +3238,7 @@ static hawk_nde_t* parse_print (hawk_t* hawk, const hawk_loc_t* xloc)
tmp = args_tail;
if (tail_prev != HAWK_NULL)
tail_prev->next = ep->left;
if (tail_prev) tail_prev->next = ep->left;
else args = ep->left;
out = ep->right;
@ -3261,7 +3265,7 @@ static hawk_nde_t* parse_print (hawk_t* hawk, const hawk_loc_t* xloc)
if (get_token(hawk) <= -1) goto oops;
eloc = hawk->tok.loc;
out = parse_expr_withdc (hawk, &eloc);
out = parse_expr_withdc(hawk, &eloc);
if (out == HAWK_NULL) goto oops;
}
}
@ -3272,8 +3276,8 @@ static hawk_nde_t* parse_print (hawk_t* hawk, const hawk_loc_t* xloc)
goto oops;
}
nde = (hawk_nde_print_t*) hawk_callocmem (hawk, HAWK_SIZEOF(*nde));
if (nde == HAWK_NULL)
nde = (hawk_nde_print_t*)hawk_callocmem(hawk, HAWK_SIZEOF(*nde));
if (HAWK_UNLIKELY(!nde))
{
ADJERR_LOC (hawk, xloc);
goto oops;
@ -3621,7 +3625,7 @@ static hawk_nde_t* parse_expr (hawk_t* hawk, const hawk_loc_t* xloc)
{
hawk_loc_t eloc;
eloc = hawk->tok.loc;
y = parse_expr_withdc (hawk, &eloc);
y = parse_expr_withdc(hawk, &eloc);
}
if (y == HAWK_NULL)
{
@ -4944,8 +4948,9 @@ static hawk_nde_t* parse_primary_lparen (hawk_t* hawk, const hawk_loc_t* xloc)
}
/* remember the sequence number of the left parenthesis
* that' been just closed by the matching right parenthesis */
* that's been just closed by the matching right parenthesis */
hawk->parse.lparen_last_closed = opening_lparen_seq;
hawk->tok.flags |= TOK_FLAGS_LPAREN_CLOSER; /* indicate that this RPAREN is closing LPAREN */
if (get_token(hawk) <= -1) goto oops;
@ -5151,7 +5156,7 @@ static hawk_nde_t* parse_primary_nopipe (hawk_t* hawk, const hawk_loc_t* xloc)
(MATCH(hawk,TOK_PRINT) || MATCH(hawk,TOK_PRINTF)))
{
if (get_token(hawk) <= -1) return HAWK_NULL;
return parse_print (hawk, xloc);
return parse_print(hawk, xloc);
}
/* valid expression introducer is expected */
@ -6635,6 +6640,7 @@ retry:
while (n >= 1);
hawk_ooecs_clear (tok->name);
tok->flags = 0;
tok->loc.file = hawk->sio.last.file;
tok->loc.line = hawk->sio.last.line;
tok->loc.colm = hawk->sio.last.colm;
@ -6895,6 +6901,7 @@ retry:
static int get_token (hawk_t* hawk)
{
hawk->ptok.type = hawk->tok.type;
hawk->ptok.flags = hawk->tok.flags;
hawk->ptok.loc.file = hawk->tok.loc.file;
hawk->ptok.loc.line = hawk->tok.loc.line;
hawk->ptok.loc.colm = hawk->tok.loc.colm;
@ -6903,6 +6910,7 @@ static int get_token (hawk_t* hawk)
if (HAWK_OOECS_LEN(hawk->ntok.name) > 0)
{
hawk->tok.type = hawk->ntok.type;
hawk->tok.flags = hawk->ntok.flags;
hawk->tok.loc.file = hawk->ntok.loc.file;
hawk->tok.loc.line = hawk->ntok.loc.line;
hawk->tok.loc.colm = hawk->ntok.loc.colm;
@ -6913,7 +6921,7 @@ static int get_token (hawk_t* hawk)
return 0;
}
return get_token_into (hawk, &hawk->tok);
return get_token_into(hawk, &hawk->tok);
}
static int preget_token (hawk_t* hawk)
@ -6956,7 +6964,7 @@ static int preget_token (hawk_t* hawk)
{
/* if there is no token pre-read, we get a new
* token and place it to hawk->ntok. */
return get_token_into (hawk, &hawk->ntok);
return get_token_into(hawk, &hawk->ntok);
}
}

6
hawk/lib/val.c
View File

@ -724,10 +724,13 @@ hawk_val_t* hawk_rtx_makenumorstrvalwithuchars (hawk_rtx_t* rtx, const hawk_uch_
hawk_int_t l;
hawk_flt_t r;
if (ptr[0] == '.' && len == 1) goto make_str;
x = hawk_uchars_to_num(HAWK_OOCHARS_TO_NUM_MAKE_OPTION(1, 1, HAWK_RTX_IS_STRIPSTRSPC_ON(rtx), 0), ptr, len, &l, &r);
if (x == 0) return hawk_rtx_makeintval(rtx, l);
else if (x >= 1) return hawk_rtx_makefltval(rtx, r);
make_str:
return hawk_rtx_makestrvalwithuchars(rtx, ptr, len);
}
@ -737,10 +740,13 @@ hawk_val_t* hawk_rtx_makenumorstrvalwithbchars (hawk_rtx_t* rtx, const hawk_bch_
hawk_int_t l;
hawk_flt_t r;
if (ptr[0] == '.' && len == 1) goto make_str;
x = hawk_bchars_to_num(HAWK_OOCHARS_TO_NUM_MAKE_OPTION(1, 1, HAWK_RTX_IS_STRIPSTRSPC_ON(rtx), 0), ptr, len, &l, &r);
if (x == 0) return hawk_rtx_makeintval(rtx, l);
else if (x >= 1) return hawk_rtx_makefltval(rtx, r);
make_str:
return hawk_rtx_makestrvalwithbchars(rtx, ptr, len);
}

1
hawk/mod/Makefile.in
View File

@ -379,7 +379,6 @@ pdfdir = @pdfdir@
prefix = @prefix@
program_transform_name = @program_transform_name@
psdir = @psdir@
runstatedir = @runstatedir@
sbindir = @sbindir@
sharedstatedir = @sharedstatedir@
srcdir = @srcdir@

1
hawk/samples/Makefile.in
View File

@ -353,7 +353,6 @@ pdfdir = @pdfdir@
prefix = @prefix@
program_transform_name = @program_transform_name@
psdir = @psdir@
runstatedir = @runstatedir@
sbindir = @sbindir@
sharedstatedir = @sharedstatedir@
srcdir = @srcdir@

8
hawk/t/Makefile.am
View File

@ -13,9 +13,11 @@ LDFLAGS_COMMON=-L$(abs_builddir)/../lib -L$(libdir)
## for more information.
LIBADD_COMMON = ../lib/libhawk.la $(LIBM)
check_SCRIPTS = h-001.hawk h-002.hawk
check_SCRIPTS = h-001.hawk h-002.hawk h-003.hawk
##noinst_SCRIPTS = $(check_SCRIPTS)
EXTRA_DIST = $(check_SCRIPTS) ensure.inc
EXTRA_DIST = $(check_SCRIPTS) ensure.inc \
journal-toc.hawk journal-toc.in journal-toc.out journal-toc-html.out \
bibtex-to-html.hawk bibtex-to-html.out
check_PROGRAMS = t-001 t-002 t-003 t-004 t-005 t-006
@ -57,4 +59,4 @@ TESTS = $(check_PROGRAMS) $(check_SCRIPTS)
TEST_EXTENSIONS = .hawk
HAWK_LOG_COMPILER = ../bin/hawk
AM_HAWK_LOG_FLAGS = --modlibdirs=../lib/.libs:../mod/.libs -f
AM_HAWK_LOG_FLAGS = -vTDIR=${abs_srcdir} --modlibdirs=../lib/.libs:../mod/.libs -f

10
hawk/t/Makefile.in
View File

@ -558,7 +558,6 @@ pdfdir = @pdfdir@
prefix = @prefix@
program_transform_name = @program_transform_name@
psdir = @psdir@
runstatedir = @runstatedir@
sbindir = @sbindir@
sharedstatedir = @sharedstatedir@
srcdir = @srcdir@
@ -578,8 +577,11 @@ CPPFLAGS_COMMON = \
CFLAGS_COMMON =
LDFLAGS_COMMON = -L$(abs_builddir)/../lib -L$(libdir)
LIBADD_COMMON = ../lib/libhawk.la $(LIBM)
check_SCRIPTS = h-001.hawk h-002.hawk
EXTRA_DIST = $(check_SCRIPTS) ensure.inc
check_SCRIPTS = h-001.hawk h-002.hawk h-003.hawk
EXTRA_DIST = $(check_SCRIPTS) ensure.inc \
journal-toc.hawk journal-toc.in journal-toc.out journal-toc-html.out \
bibtex-to-html.hawk bibtex-to-html.out
t_001_SOURCES = t-001.c t.h
t_001_CPPFLAGS = $(CPPFLAGS_COMMON)
t_002_SOURCES = t-002.c t.h
@ -610,7 +612,7 @@ t_006_LDADD = $(LIBADD_COMMON)
TESTS = $(check_PROGRAMS) $(check_SCRIPTS)
TEST_EXTENSIONS = .hawk
HAWK_LOG_COMPILER = ../bin/hawk
AM_HAWK_LOG_FLAGS = --modlibdirs=../lib/.libs:../mod/.libs -f
AM_HAWK_LOG_FLAGS = -vTDIR=${abs_srcdir} --modlibdirs=../lib/.libs:../mod/.libs -f
all: all-am
.SUFFIXES:

545
hawk/t/bibtex-to-html.hawk Normal file
View File

@ -0,0 +1,545 @@
# http://www.netlib.org/bibnet/tools/software/bibtex-to-html.awk
#
### ====================================================================
### @Awk-file{
### author = "Nelson H. F. Beebe",
### version = "1.02",
### date = "05 July 1997",
### time = "12:04:52 MDT",
### filename = "bibtex-to-html.awk",
### address = "Center for Scientific Computing
### Department of Mathematics
### University of Utah
### Salt Lake City, UT 84112
### USA",
### telephone = "+1 801 581 5254",
### FAX = "+1 801 581 4148",
### URL = "http://www.math.utah.edu/~beebe",
### checksum = "08699 482 2173 18348",
### email = "beebe@math.utah.edu (Internet)",
### codetable = "ISO/ASCII",
### keywords = "bibliography, BibTeX, HTML, World-Wide Web,
### WWW",
### supported = "yes",
### docstring = "This program converts BibTeX bibliographies
### to HTML, suitable for viewing on the
### World-Wide Web.
###
### The level of HTML produced is version 3.2,
### adopted 14-Jan-1997, and defined in the SGML
### document type definition (DTD) available at
###
### http://www.w3.org/MarkUp/Wilbur/HTML32.dtd
###
### and documented at
###
### http://www.w3.org/MarkUp/Wilbur/
### http://www.w3.org/TR/REC-html32.html
###
### HTML markup is added to provide hypertext
### links for:
###
### * all URLs in the BibTeX file, both in
### comments, and inside string values;
### * all bibliography entry crossref
### values;
### * all \cite{} references;
### * all @String{name = "value"} names.
###
### In addition, every BibTeX citation label in
### @Entry lines, and every @String name, will
### be marked as an HTML label, allowing
### hypertext links to each from elsewhere in
### the same HTML file, or from other HTML
### files. In particular, every bibliography
### entry can be directly referenced by
### hypertext links from anywhere on the
### Internet.
###
### Each such linkable-name will be displayed
### in bold text to draw attention to the fact
### that it can be directly referenced by a
### suitable URL. In principle, this should be
### an option that WWW browsers provide, but
### none that I have used currently do.
###
### Although no browsers to my knowledge yet
### provide the capability of partial
### downloading of HTML files, the possibility
### has been discussed for future versions of
### the HTTP protocol. Such support would make
### it possible to construct bibliographies in
### electronic documents as links to large
### bibliography database files, without the
### browser having to load the entire database,
### but just individual entries. Since these
### in turn can have URLs that point to other
### electronic sources of the publication, a
### reader could easily follow links from a
### publication to a bibliography and then to
### abstracts and to the complete original
### text. Some journals, such as the Digital
### Technical Journal (electronically accessible
### at http://www.digital.com:80/info/DTJ/home.html),
### already could offer this possibility.
###
### The Web browser user will see material that
### looks just like normal BibTeX entries,
### except that some portions may be
### highlighted to indicate hypertext links.
### However, window cut-and-paste actions will
### recover a BibTeX entry in a form suitable
### for pasting into another BibTeX file,
### without any need for further editing.
###
### This program assumes that the BibTeX
### bibliography is formatted in the style
### produced by bibclean, and that embedded
### URLs and "key = stringname" pairs are coded
### on a single line, so that simple pattern
### matching suffices to recognize text in need
### of additional HTML markup.
###
### Usage:
### nawk -f bibtex-to-html.awk \
### [-v PREFIX=prefix] [-v SUFFIX=suffix] \
### BibTeX-file(s)
###
### An input file with a filename of the form
### abc.xyz is output to a file named
### PREFIXabcSUFFIX. The default PREFIX is
### empty, and the default SUFFIX is ".html".
###
### If no file names are specified on the
### command line, then the PREFIX and SUFFIX
### settings are ignored, and input is read
### from stdin, and output is written to
### stdout, so that the program can be used in
### a UNIX pipeline.
###
### In the current version, no provision is
### made for splitting the output files into
### smaller pieces to speed network file
### transfer. While this would improve browser
### responsiveness over slow network
### connections, it would also significantly
### complicate hypertext link generation for
### this program, and seriously damage browser
### search capability within the bibliography
### file. Perhaps the solution will come in
### (a) browsers' adopting the netscape browser
### practice of displaying data as soon as
### enough to fill a screen is available, and
### (b) faster network connections.
###
### In the TUG bibliography collection at
### ftp://ftp.math.utah.edu/, bibliography
### file sizes range from 3K to 4700K, with an
### average of 370K. These are rather large,
### since typical WWW file sizes need to be
### about 16K or less for good responsiveness.
###
### The checksum field above contains a CRC-16
### checksum as the first value, followed by the
### equivalent of the standard UNIX wc (word
### count) utility output of lines, words, and
### characters. This is produced by Robert
### Solovay's checksum utility.",
### }
### ====================================================================
BEGIN \
{
######################################################################
VERSION = "1.02 [05-Jul-1997]" # <-- NB: Change this with each update!
######################################################################
PROGRAM = "bibtex-to-html"
UNSET_FILENAME = "/dev/unset"
LASTFILENAME = UNSET_FILENAME
_last_input_filename = UNSET_FILENAME
if (SUFFIX == "")
SUFFIX = ".html"
USER = ENVIRON["USER"]
if (USER == "")
USER = ENVIRON["LOGNAME"]
if (USER == "")
USER = "????"
"hostname" | getline HOSTNAME
"date" | getline DATE
# [01-Aug-2019] ypcat no longer available: replace by getent
# ("ypcat passwd | grep '^" USER ":' | awk -F: '{print $5}'") | getline PERSONAL_NAME
("getent passwd " USER " | " ARGV[0] " -F: '{print $5}'") | getline PERSONAL_NAME
if (PERSONAL_NAME == "")
##("grep '^" USER ":' /etc/passwd | awk -F: '{print $5}'") | getline PERSONAL_NAME
("grep '^" USER ":' /etc/passwd | " ARGV[0] " -F: '{print $5}'") | getline PERSONAL_NAME
# NB: " has become &#34; before this pattern is used
CROSSREF_EQUALS_LABEL_PATTERN = "^[ \t]*crossref[ \t]*=[ \t]*&#34;"
# Pattern to match a line like this:
# %%% email = "beebe at math.utah.edu (Internet)",
BIBTEX_EMAIL_PATTERN = "= &#34;[A-Za-z0-9-]+ at [A-Za-z0-9.-]+"
BIBTEX_EMAIL_OFFSET = 7 # was 8 before &quot; became &#34;
BIBTEX_EMAIL_PREFIX = "mailto:"
BIBTEX_EMAIL_SAVE_LABEL = 0
##CITE_PATTERN = "\\\\cite{[^}]+}"
CITE_PATTERN = "\\\\cite\\{[^\\}]+}"
CITE_OFFSET = 6
CITE_PREFIX = ""
CITE_SAVE_LABEL = 1
EMAIL_PATTERN = "[A-Za-z0-9-]+@[A-Za-z0-9.-]+"
EMAIL_OFFSET = 0
EMAIL_PREFIX = "mailto:"
EMAIL_SAVE_LABEL = 0
# See Nelson H. F. Beebe, ``Bibliography prettyprinting
# and syntax checking'', TUGboat 14(3), 222-222, October
# (1993), and 14(4), 395--419, December (1993) for the
# syntax of BibTeX names used here in ENTRY_PATTERN,
# KEY_EQUALS_NAME_PATTERN and STRING_PATTERN.
##ENTRY_PATTERN = "^[ \t]*@[ \t]*[A-Za-z][A-Za-z0-9:.+/'-]*[ \t]*{[A-Za-z][A-Za-z0-9:.+/'-]*,[ \t]*$"
ENTRY_PATTERN = "^[ \t]*@[ \t]*[A-Za-z][A-Za-z0-9:.+/'-]*[ \t]*\\{[A-Za-z][A-Za-z0-9:.+/'-]*,[ \t]*$"
KEY_EQUALS_NAME_PATTERN = "^[ \t]*[A-Za-z][A-Za-z0-9:.+/'-]*[ \t]*=[ \t]*[A-Za-z]"
##STRING_PATTERN = "^@[Ss][Tt][Rr][Ii][Nn][gG]{[A-Za-z][A-Za-z0-9:.+/'-]*"
STRING_PATTERN = "^@[Ss][Tt][Rr][Ii][Nn][gG]\\{[A-Za-z][A-Za-z0-9:.+/'-]*"
STRING_OFFSET = 8
STRING_PREFIX = ""
STRING_SAVE_LABEL = 1
# According to Internet RFC 1614 (May 1994), a URL is
# defined in the document T. Berners-Lee, ``Uniform
# Resource Locators'', March 1993, available at URL
# ftp://info.cern.ch/pub/ietf/url4.ps. Unfortunately,
# that address is no longer valid. However, I was able to
# track down pointers from http://www.w3.org/ to locate a
# suitable description in Internet RFC 1630 (June 1994).
# NB: We additionally disallow & in a URL because it is
# needed in SGML entities "&name;". We also disallow =
# and | because these are commonly used in \path=...= and
# \path|...| strings in BibTeX files. These restrictions
# could be removed if we went to the trouble of first
# encoding these special characters in %xy hexadecimal
# format, but they are rare enough that I am not going to
# do so for now. The worst that will happen from this
# decision is that an occasional URL in a BibTeX file will
# be missing a surrounding anchor.
URL_PATTERN = "[A-Za-z]+://[^ \",&=|]+"
URL_OFFSET = 0
URL_PREFIX = ""
URL_SAVE_LABEL = 0
# [24-May-2016] support for background coloring of block comments
IN_BLOCK_COMMENT = 0
}
# Each line receives identical processing.
{ do_line() }
END \
{
if (LASTFILENAME != UNSET_FILENAME)
end_file(LASTFILENAME)
}
function add_entry(array,value)
{
if (value in array)
array[value] = array[value] " " FNR
else
array[value] = FNR
}
function anchor(s,type,pattern,offset,prefix,save_label, name,rstart,rlength)
{
# Add anchors <A type="....">...</A> around text in s matching
# pattern. A non-zero offset discards that many characters from
# the start of the match, allowing the pattern to contain leading
# context which goes outside the anchored region. The prefix is
# attached to the start of the matched string, inside the value
# quotes in the anchor.
if (match(s,pattern))
{
rstart = RSTART # need private copies of these globals because
rlength = RLENGTH # recursion will change them
rstart += offset # adjust by offset to discard leading
rlength -= offset # context in pattern
name = substr(s,rstart,rlength)
sub(/ +at +/,"@",name) # reduce "user at host" to "user@host"
s = substr(s,1,rstart-1) \
"<A " type "=\"" prefix name "\">" \
((type == "NAME") ? "<STRONG>" : "") \
substr(s,rstart,rlength) \
((type == "NAME") ? "</STRONG>" : "") \
"</A>" \
anchor(substr(s,rstart+rlength),type,pattern,offset,prefix,save)
if (save_label)
{
if (type == "HREF")
add_entry(label_hrefs, name)
else if (type == "NAME")
add_entry(label_names, name)
}
}
return (s)
}
function begin_file( f)
{
f = output_filename(FILENAME)
## NB: If Transitional is eliminated in DOCTYPE, background coloring is lost! Why?
slash_pos = str::rindex(FILENAME, "/");
BASE_FILENAME = (slash_pos > 0)? str::substr(FILENAME, slash_pos + 1): FILENAME;
print "<!-- -*-html-*- -->" > f
print "" > f
## print "<!-- " FILENAME " -->" > f
print "<!-- " BASE_FILENAME " -->" > f
print "<!-- WARNING: Do NOT edit this file. It was converted from -->" > f
print "<!-- BibTeX format to HTML by " PROGRAM " version " VERSION " -->" > f
## print "<!-- on " DATE " -->" > f
## print "<!-- for " PERSONAL_NAME " (" USER "@" HOSTNAME ") -->" > f
print "" > f
print "<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.0 Transitional//EN\" \"http://www.w3.org/TR/1998/REC-html40-19980424/loose.dtd\">" > f
print "" > f
print "" > f
print "<HTML>" > f
print " <HEAD>" > f
print " <META HTTP-EQUIV=\"Content-Type\" CONTENT=\"text/html; charset=iso-8859-1\">"> f
print " <TITLE>" > f
## print " BibTeX bibliography " FILENAME > f
print " BibTeX bibliography " BASE_FILENAME > f
print " </TITLE>" > f
print " <LINK REV=\"made\" HREF=\"mailto:" USER "@" HOSTNAME "\">" > f
print " <LINK HREF=\"http://www.math.utah.edu/pub/tex/bib/tugbib.css\" TYPE=\"text/css\" REL=\"stylesheet\">" > f
print " </HEAD>" > f
print "" > f
print " <BODY>" > f
print " <DIV ALIGN=\"right\">" > f
print " <A HREF=\"http://validator.w3.org/check/referer\">" > f
print " <IMG ALIGN=\"MIDDLE\" BORDER=\"0\" SRC=\"/images/valid-html40.png\" ALT=\"Valid HTML 4.0!\" HEIGHT=\"31\" WIDTH=\"88\">" > f
print " </A>" > f
print " <A HREF=\"http://jigsaw.w3.org/css-validator/check/referer\">" > f
print " <IMG ALIGN=\"MIDDLE\" BORDER=\"0\" SRC=\"/images/valid-css.gif\" ALT=\"Valid CSS!\" HEIGHT=\"31\" WIDTH=\"88\">" > f
print " </A>" > f
print " </DIV>" > f
print "<PRE>" > f
clear_array(label_names)
clear_array(label_hrefs)
}
function check_for_file_change()
{
if (LASTFILENAME != FILENAME)
{
if (LASTFILENAME != UNSET_FILENAME)
{
end_file(LASTFILENAME)
if (LASTFILENAME != "/dev/stdout")
close (output_filename(LASTFILENAME))
}
LASTFILENAME = FILENAME
begin_file()
}
}
function check_refs( label)
{
for (label in label_hrefs)
{
if (!(label in label_names))
warning("undefined label " label " at line(s) " label_hrefs[label])
}
}
function clear_array(array, key)
{
for (key in array)
delete array[key]
}
function end_file(filename, f)
{
f = output_filename(filename)
print "</PRE>" > f
print " </BODY>" > f
print "</HTML>" > f
check_refs()
}
function do_cite(s, k,n,labels,t)
{
n = split(substr(s,RSTART + CITE_OFFSET,RLENGTH - 1 - CITE_OFFSET),labels,",")
t = substr(s,1,RSTART+CITE_OFFSET-1)
for (k = 1; k <= n; ++k)
{
t = t ((k > 1) ? "," : "") "<A HREF=\"#" labels[k] "\">" labels[k] "</A>"
add_entry(label_hrefs, labels[k])
}
t = t substr(s,RSTART + RLENGTH - 1)
return (t)
}
function do_line( n,name,s)
{
s = protect_SGML_characters($0)
if (match(s,STRING_PATTERN)) # remember name from @String{name = "value"}
{
name = substr(s,RSTART + STRING_OFFSET,RLENGTH - STRING_OFFSET)
string_name[name] = 1
# print "DEBUG 1: name =", name >"/dev/stderr"
}
if (match(s,/^%+[ \t]*email[ \t]*=/)) # special handling because BibTeX does not allow @ in comments
s = anchor(s,"HREF",BIBTEX_EMAIL_PATTERN,BIBTEX_EMAIL_OFFSET,BIBTEX_EMAIL_PREFIX,\
BIBTEX_EMAIL_SAVE_LABEL)
else
s = anchor(s,"HREF",EMAIL_PATTERN,EMAIL_OFFSET,EMAIL_PREFIX,EMAIL_SAVE_LABEL)
s = anchor(s,"HREF",URL_PATTERN,URL_OFFSET,URL_PREFIX,URL_SAVE_LABEL)
s = anchor(s,"NAME",STRING_PATTERN,STRING_OFFSET,STRING_PREFIX,STRING_SAVE_LABEL)
if (match(s,CITE_PATTERN))
s = do_cite(s)
if (match(s,ENTRY_PATTERN)) # then have ``@Entry{label,''
{
n = index(s,"{")
name = substr(s,n+1)
gsub(/^[ \t]*/,"",name) # trim optional leading space
gsub(/,[ \t]*$/,"",name) # trim trailing comma and optional space
# print "DEBUG 2: name =", name >"/dev/stderr"
s = substr(s,1,n) \
"<A NAME=\"" name "\"><STRONG>" name "</STRONG></A>" \
substr(s,n+1+length(name))
add_entry(label_names, name)
}
else if (match(s,KEY_EQUALS_NAME_PATTERN)) # then have ``key = name''
{
name = substr(s,RSTART+RLENGTH-1)
sub(/,?[ \t]*$/,"",name) # trim optional trailing comma and space
# print "DEBUG 3: name =", name >"/dev/stderr"
if (name in string_name) # then we have a definition of this name
{
s = substr(s,1,RSTART+RLENGTH-2) \
"<A HREF=\"#" name "\">" name "</A>" substr(s,RSTART+RLENGTH-1+length(name))
add_entry(label_hrefs, name)
}
}
else if (match(s,CROSSREF_EQUALS_LABEL_PATTERN)) # then have `` crossref = "label"''
{
name = substr(s,RSTART+RLENGTH)
sub(/&#34;,?[ \t]*$/,"",name) # trim trailing quote and optional comma and space
# print "DEBUG 4: name =", name >"/dev/stderr"
s = substr(s,1,RSTART+RLENGTH-1) \
"<A HREF=\"#" name "\">" name "</A>" substr(s,RSTART+RLENGTH+length(name))
add_entry(label_hrefs, name)
}
check_for_file_change()
if ( (s ~ "^%") && !IN_BLOCK_COMMENT)
{
printf("</PRE><PRE CLASS=\"blockcomment\">") > output_filename(FILENAME)
IN_BLOCK_COMMENT = 1
}
else if ( (s !~ "^%") && IN_BLOCK_COMMENT)
{
printf("</PRE><PRE>") > output_filename(FILENAME)
IN_BLOCK_COMMENT = 0
}
print s >output_filename(FILENAME)
}
function output_filename(input_filename)
{
## HAWK - for use in t/h-003.hawk
return "/dev/stdout";
if (input_filename != _last_input_filename)
{ # optimization: we cache last function result for speed
_last_input_filename = input_filename
sub(/.[^.]*$/,"",input_filename)
if ((input_filename == "") || (input_filename == "/dev/stdin"))
_last_output_filename = "/dev/stdout"
else
_last_output_filename = PREFIX input_filename SUFFIX
}
return (_last_output_filename)
}
function protect_SGML_characters(s)
{
gsub(/&/,"\\&amp;",s) # NB: this one MUST be first
gsub(/</,"\\&lt;",s)
gsub(/>/,"\\&gt;",s)
## [24-May-2016] with the change from HTML 3.2 to 4.0, we can use &quot; again!
## gsub(/\"/,"\\&#34;",s) # this was &quot; in earlier HTML
# versions, including the HTML 3.2
# draft, but was stupidly eliminated in
# the final HTML 3.2 version: see
# http://www.w3.org/pub/WWW/MarkUp/Wilbur/
# in the section ``What happened to &quot;?''
gsub(/\"/,"\\&quot;",s)
return (s)
}
function warning(message)
{
# print FILENAME ":" FNR ":%%" message >"/dev/stderr"
#
# The only place that we need warning messages above is
# checkrefs(), after the current file has been closed, and a new
# one started, so we need LASTFILENAME instead of FILENAME here,
# and we omit the FNR, since we have no record of it for
# LASTFILENAME.
print LASTFILENAME ":%%" message >"/dev/stderr"
}

27252
hawk/t/bibtex-to-html.out Normal file
View File

File diff suppressed because it is too large Load Diff

33
hawk/t/h-002.hawk
View File

@ -22,7 +22,7 @@ function main()
ensure (foo2[1], 1, @SCRIPTNAME, @SCRIPTLINE);
ensure (foo3[1], ini, @SCRIPTNAME, @SCRIPTLINE);
ensure (foo4[1], ini, @SCRIPTNAME, @SCRIPTLINE);
}
}
## gsub
{
@ -279,7 +279,9 @@ function main()
{
@local a;
# split, str::split, str::splita
@local a, str;
ensure (split("Here===Is=Some=====Data", a, "=+"), 4, @SCRIPTNAME, @SCRIPTLINE);
ensure (a[1] === "Here", 1, @SCRIPTNAME, @SCRIPTLINE);
ensure (a[2] === "Is", 1, @SCRIPTNAME, @SCRIPTLINE);
@ -315,14 +317,37 @@ function main()
ensure (str::splita("Here===Is=Some=====Data", a, ""), 23, @SCRIPTNAME, @SCRIPTLINE);
ensure (hawk::typename(a), "array",@SCRIPTNAME, @SCRIPTLINE);
ensure (split("Here Is Some Data", a, / /), 7, @SCRIPTNAME, @SCRIPTLINE);
ensure (split("Here Is Some Data", a, " "), 4, @SCRIPTNAME, @SCRIPTLINE);
ensure (str::split("Here Is Some Data", a, / /), 7, @SCRIPTNAME, @SCRIPTLINE);
ensure (str::split("Here Is Some Data", a, " "), 4, @SCRIPTNAME, @SCRIPTLINE);
ensure (a[1] === "Here", 1, @SCRIPTNAME, @SCRIPTLINE);
ensure (a[2] === "Is", 1, @SCRIPTNAME, @SCRIPTLINE);
ensure (a[3] === "Some", 1, @SCRIPTNAME, @SCRIPTLINE);
ensure (a[4] === "Data", 1, @SCRIPTNAME, @SCRIPTLINE);
str = "a b\t\tc d";
ensure (str::splita(str, a, " "), 4, @SCRIPTNAME, @SCRIPTLINE);
ensure (a[1], "a", @SCRIPTNAME, @SCRIPTLINE);
ensure (a[2], "b", @SCRIPTNAME, @SCRIPTLINE);
ensure (a[3], "c", @SCRIPTNAME, @SCRIPTLINE);
ensure (a[4], "d", @SCRIPTNAME, @SCRIPTLINE);
ensure (str::splita(str, a, / /), 5, @SCRIPTNAME, @SCRIPTLINE);
ensure (a[1], "a", @SCRIPTNAME, @SCRIPTLINE);
ensure (a[2], "", @SCRIPTNAME, @SCRIPTLINE);
ensure (a[3], "", @SCRIPTNAME, @SCRIPTLINE);
ensure (a[4], "b\t\tc", @SCRIPTNAME, @SCRIPTLINE);
ensure (a[5], "d", @SCRIPTNAME, @SCRIPTLINE);
}
{
## back reference in a regular expression
ensure (("a2b" ~ /(a)\12b/), 0, @SCRIPTNAME, @SCRIPTLINE);
ensure (("aa2b" ~ /(a)\12b/), 1, @SCRIPTNAME, @SCRIPTLINE);
ensure (("aaa2b" ~ /(a)\12b/), 1, @SCRIPTNAME, @SCRIPTLINE);
}
print "SUCCESS";
}

76
hawk/t/h-003.hawk Normal file
View File

@ -0,0 +1,76 @@
@pragma entry main
@pragma implicit off
@include "ensure.inc";
function are_files_identical(a, b)
{
@local f1, f2, x, y, diff;
f1 = sys::open(a, sys::O_RDONLY);
if (f1 <= -1)
{
printf ("ERROR: unable to open %s\n", a);
return -1;
}
f2 = sys::open(b, sys::O_RDONLY);
if (f2 <= -1)
{
sys::close (a);
printf ("ERROR: unable to open %s\n", b);
return -1;
}
diff = 0;
while (sys::read(f1, x, 1) > 0)
{
if (sys::read(f2, y, 1) <= 0 || x !== y)
{
diff = 1;
break;
}
}
if (sys::read(f2, y, 1) > 0) diff = 1;
sys::close (f2);
sys::close (f1);
return !diff;
}
function run_test (x, more_opts, in_name, out_name)
{
@local cmd, inf, expf, outf, same;
if (hawk::isnil(in_name)) in_name = x;
if (hawk::isnil(out_name)) out_name = x;
inf = sprintf("%s/%s.in", TDIR, in_name);
expf = sprintf("%s/%s.out", TDIR, out_name);
outf = sprintf("/tmp/%s.%d.out", out_name, sys::getpid());
##print TDIR, inf, expf, outf;
cmd=sprintf("%s %s -f %s/%s.hawk %s > %s", ARGV[0], more_opts, TDIR, x, inf, outf);
##print cmd;
system (cmd);
same = are_files_identical(expf, outf);
if (same <= 0)
{
## don't delete the output file for review.
printf ("FAILURE: %s - %s and %s differ\n", x, expf, outf);
exit (1);
}
sys::unlink (outf);
}
function main()
{
run_test ("journal-toc");
run_test ("journal-toc", "-vHTML=1", "journal-toc", "journal-toc-html");
run_test ("bibtex-to-html", "", "journal-toc", "bibtex-to-html");
print "SUCCESS";
}

5090
hawk/t/journal-toc-html.out Normal file
View File

File diff suppressed because it is too large Load Diff

993
hawk/t/journal-toc.hawk Normal file
View File

@ -0,0 +1,993 @@
# http://www.netlib.org/bibnet/tools/software/journal-toc.awk
#
### ====================================================================
### @Awk-file{
### author = "Nelson H. F. Beebe",
### version = "1.00",
### date = "09 October 1996",
### time = "15:57:06 MDT",
### filename = "journal-toc.awk",
### address = "Center for Scientific Computing
### Department of Mathematics
### University of Utah
### Salt Lake City, UT 84112
### USA",
### telephone = "+1 801 581 5254",
### FAX = "+1 801 581 4148",
### URL = "http://www.math.utah.edu/~beebe",
### checksum = "25092 977 3357 26493",
### email = "beebe@math.utah.edu (Internet)",
### codetable = "ISO/ASCII",
### keywords = "BibTeX, bibliography, HTML, journal table of
### contents",
### supported = "yes",
### docstring = "Create a journal cover table of contents from
### <at>Article{...} entries in a journal BibTeX
### .bib file for checking the bibliography
### database against the actual journal covers.
### The output can be either plain text, or HTML.
###
### Usage:
### bibclean -max-width 0 BibTeX-file(s) | \
### bibsort -byvolume | \
### awk -f journal-toc.awk \
### [-v HTML=nnn] [-v INDENT=nnn] \
### [-v BIBFILEURL=url] >foo.toc
###
### or if the bibliography is already sorted
### by volume,
###
### bibclean -max-width 0 BibTeX-file(s) | \
### awk -f journal-toc.awk \
### [-v HTML=nnn] [-v INDENT=nnn] \
### [-v BIBFILEURL=url] >foo.toc
###
### A non-zero value of the command-line option,
### HTML=nnn, results in HTML output instead of
### the default plain ASCII text (corresponding
### to HTML=0). The
###
### The INDENT=nnn command-line option specifies
### the number of blanks to indent each logical
### level of HTML. The default is INDENT=4.
### INDENT=0 suppresses indentation. The INDENT
### option has no effect when the default HTML=0
### (plain text output) option is in effect.
###
### When HTML output is selected, the
### BIBFILEURL=url command-line option provides a
### way to request hypertext links from table of
### contents page numbers to the complete BibTeX
### entry for the article. These links are
### created by appending a sharp (#) and the
### citation label to the BIBFILEURL value, which
### conforms with the practice of
### bibtex-to-html.awk.
###
### The HTML output form may be useful as a more
### compact representation of journal article
### bibliography data than the original BibTeX
### file provides. Of course, the
### table-of-contents format provides less
### information, and is considerably more
### troublesome for a computer program to parse.
###
### When URL key values are provided, they will
### be used to create hypertext links around
### article titles. This supports journals that
### provide article contents on the World-Wide
### Web.
###
### For parsing simplicity, this program requires
### that BibTeX
###
### key = "value"
###
### and
###
### @String{name = "value"}
###
### specifications be entirely contained on
### single lines, which is readily provided by
### the `bibclean -max-width 0' filter. It also
### requires that bibliography entries begin and
### end at the start of a line, and that
### quotation marks, rather than balanced braces,
### delimit string values. This is a
### conventional format that again can be
### guaranteed by bibclean.
###
### This program requires `new' awk, as described
### in the book
###
### Alfred V. Aho, Brian W. Kernighan, and
### Peter J. Weinberger,
### ``The AWK Programming Language'',
### Addison-Wesley (1988), ISBN
### 0-201-07981-X,
###
### such as provided by programs named (GNU)
### gawk, nawk, and recent AT&T awk.
###
### The checksum field above contains a CRC-16
### checksum as the first value, followed by the
### equivalent of the standard UNIX wc (word
### count) utility output of lines, words, and
### characters. This is produced by Robert
### Solovay's checksum utility.",
### }
### ====================================================================
BEGIN { initialize() }
/^ *@ *[Ss][Tt][Rr][Ii][Nn][Gg] *\{/ { do_String(); next }
/^ *@ *[Pp][Rr][Ee][Aa][Mm][Bb][Ll][Ee]/ { next }
/^ *@ *[Aa][Rr][Tt][Ii][Cc][Ll][Ee]/ { do_Article(); next }
/^ *@/ { do_Other(); next }
/^ *author *= *\"/ { do_author(); next }
/^ *journal *= */ { do_journal(); next }
/^ *volume *= *\"/ { do_volume(); next }
/^ *number *= *\"/ { do_number(); next }
/^ *year *= *\"/ { do_year(); next }
/^ *month *= */ { do_month(); next }
/^ *title *= *\"/ { do_title(); next }
/^ *pages *= *\"/ { do_pages(); next }
/^ *URL *= *\"/ { do_URL(); next }
/^ *} *$/ { if (In_Article) do_end_entry(); next }
END { terminate() }
########################################################################
# NB: The programming conventions for variables in this program are: #
# UPPERCASE global constants and user options #
# Initialuppercase global variables #
# lowercase local variables #
# Any deviation is an error! #
########################################################################
function do_Article()
{
In_Article = 1
Citation_label = $0
sub(/^[^\{]*\{/,"",Citation_label)
sub(/ *, *$/,"",Citation_label)
Author = ""
Title = ""
Journal = ""
Volume = ""
Number = ""
Month = ""
Year = ""
Pages = ""
Url = ""
}
function do_author()
{
Author = TeX_to_HTML(get_value($0))
}
function do_end_entry( k,n,parts)
{
n = split(Author,parts," and ")
if (Last_number != Number)
do_new_issue()
for (k = 1; k < n; ++k)
print_toc_line(parts[k] " and", "", "")
Title_prefix = html_begin_title()
Title_suffix = html_end_title()
if (html_length(Title) <= (MAX_TITLE_CHARS + MIN_LEADERS)) # complete title fits on line
print_toc_line(parts[n], Title, html_begin_pages() Pages html_end_pages())
else # need to split long title over multiple lines
do_long_title(parts[n], Title, html_begin_pages() Pages html_end_pages())
}
function do_journal()
{
if ($0 ~ /[=] *"/) # have journal = "quoted journal name",
Journal = get_value($0)
else # have journal = journal-abbreviation,
{
Journal = get_abbrev($0)
if (Journal in String) # replace abbrev by its expansion
Journal = String[Journal]
}
gsub(/\\-/,"",Journal) # remove discretionary hyphens
}
function do_long_title(author,title,pages, last_title,n)
{
title = trim(title) # discard leading and trailing space
while (length(title) > 0)
{
n = html_breakpoint(title,MAX_TITLE_CHARS+MIN_LEADERS)
last_title = substr(title,1,n)
title = substr(title,n+1)
sub(/^ +/,"",title) # discard any leading space
print_toc_line(author, last_title, (length(title) == 0) ? pages : "")
author = ""
}
}
function do_month( k,n,parts)
{
Month = ($0 ~ /[=] *"/) ? get_value($0) : get_abbrev($0)
gsub(/[\"]/,"",Month)
gsub(/ *# *\\slash *# */," / ",Month)
gsub(/ *# *-+ *# */," / ",Month)
n = split(Month,parts," */ *")
Month = ""
for (k = 1; k <= n; ++k)
Month = Month ((k > 1) ? " / " : "") \
((parts[k] in Month_expansion) ? Month_expansion[parts[k]] : parts[k])
}
function do_new_issue()
{
Last_number = Number
if (HTML)
{
if (Last_volume != Volume)
{
Last_volume = Volume
print_line(prefix(2) "<BR>")
}
html_end_toc()
html_begin_issue()
print_line(prefix(2) Journal "<BR>")
}
else
{
print_line("")
print_line(Journal)
}
print_line(strip_html(vol_no_month_year()))
if (HTML)
{
html_end_issue()
html_toc_entry()
html_begin_toc()
}
else
print_line("")
}
function do_number()
{
Number = get_value($0)
}
function do_Other()
{
In_Article = 0
}
function do_pages()
{
Pages = get_value($0)
sub(/--[?][?]/,"",Pages)
}
function do_String()
{
sub(/^[^\{]*\{/,"",$0) # discard up to and including open brace
sub(/\} *$/,"",$0) # discard from optional whitespace and trailing brace to end of line
String[get_key($0)] = get_value($0)
}
function do_title()
{
Title = TeX_to_HTML(get_value($0))
}
function do_URL( parts)
{
Url = get_value($0)
split(Url,parts,"[,;]") # in case we have multiple URLs
Url = trim(parts[1])
}
function do_volume()
{
Volume = get_value($0)
}
function do_year()
{
Year = get_value($0)
}
function get_abbrev(s)
{ # return abbrev from ``key = abbrev,''
sub(/^[^=]*= */,"",s) # discard text up to start of non-blank value
sub(/ *,? *$/,"",s) # discard trailing optional whitspace, quote,
# optional comma, and optional space
return (s)
}
function get_key(s)
{ # return kay from ``key = "value",''
sub(/^ */,"",s) # discard leading space
sub(/ *=.*$/,"",s) # discard everthing after key
return (s)
}
function get_value(s)
{ # return value from ``key = "value",''
sub(/^[^\"]*\" */,"",s) # discard text up to start of non-blank value
sub(/ *\",? *$/,"",s) # discard trailing optional whitspace, quote,
# optional comma, and optional space
return (s)
}
function html_accents(s)
{
if (index(s,"\\") > 0) # important optimization
{
# Convert common lower-case accented letters according to the
# table on p. 169 of in Peter Flynn's ``The World Wide Web
# Handbook'', International Thomson Computer Press, 1995, ISBN
# 1-85032-205-8. The official table of ISO Latin 1 SGML
# entities used in HTML can be found in the file
# /usr/local/lib/html-check/lib/ISOlat1.sgml (your path
# may differ).
gsub(/\{\\\a}/, "\\&agrave;", s)
gsub(/\{\\'a}/, "\\&aacute;", s)
gsub(/\{\\[\^]a}/,"\\&acirc;", s)
gsub(/\{\\~a}/, "\\&atilde;", s)
##gsub(/\{\\\"a\}/, "\\&auml;", s)
gsub(/\{\\\"\{a\}\}/, "\\&auml;", s)
gsub(/\{\\aa}/, "\\&aring;", s)
gsub(/\{\\ae}/, "\\&aelig;", s)
gsub(/\{\\c\{c\}}/,"\\&ccedil;", s)
gsub(/\{\\\e}/, "\\&egrave;", s)
gsub(/\{\\'e}/, "\\&eacute;", s)
gsub(/\{\\[\^]e}/,"\\&ecirc;", s)
gsub(/\{\\\"e}/, "\\&euml;", s)
gsub(/\{\\\i}/, "\\&igrave;", s)
gsub(/\{\\'i}/, "\\&iacute;", s)
gsub(/\{\\[\^]i}/,"\\&icirc;", s)
gsub(/\{\\\"i}/, "\\&iuml;", s)
# ignore eth and thorn
gsub(/\{\\~n}/, "\\&ntilde;", s)
gsub(/\{\\\o}/, "\\&ograve;", s)
gsub(/\{\\'o}/, "\\&oacute;", s)
gsub(/\{\\[\^]o}/, "\\&ocirc;", s)
gsub(/\{\\~o}/, "\\&otilde;", s)
##gsub(/\{\\\"o}/, "\\&ouml;", s)
gsub(/\{\\\"\{o\}}/, "\\&ouml;", s)
gsub(/\{\\o}/, "\\&oslash;", s)
gsub(/\{\\\u}/, "\\&ugrave;", s)
gsub(/\{\\'u}/, "\\&uacute;", s)
gsub(/\{\\[\^]u}/,"\\&ucirc;", s)
##gsub(/\{\\\"u}/, "\\&uuml;", s)
gsub(/\{\\\"\{u\}\}/, "\\&uuml;", s)
gsub(/\{\\'y}/, "\\&yacute;", s)
gsub(/\{\\\"y}/, "\\&yuml;", s)
# Now do the same for upper-case accents
gsub(/\{\\\A}/, "\\&Agrave;", s)
gsub(/\{\\'A}/, "\\&Aacute;", s)
gsub(/\{\\[\^]A}/, "\\&Acirc;", s)
gsub(/\{\\~A}/, "\\&Atilde;", s)
##gsub(/\{\\\"A}/, "\\&Auml;", s)
gsub(/\{\\\"\{A\}\}/, "\\&Auml;", s)
gsub(/\{\\AA}/, "\\&Aring;", s)
gsub(/\{\\AE}/, "\\&AElig;", s)
gsub(/\{\\c\{C\}}/,"\\&Ccedil;", s)
gsub(/\{\\\e}/, "\\&Egrave;", s)
gsub(/\{\\'E}/, "\\&Eacute;", s)
gsub(/\{\\[\^]E}/, "\\&Ecirc;", s)
gsub(/\{\\\"E}/, "\\&Euml;", s)
gsub(/\{\\\I}/, "\\&Igrave;", s)
gsub(/\{\\'I}/, "\\&Iacute;", s)
gsub(/\{\\[\^]I}/, "\\&Icirc;", s)
gsub(/\{\\\"I}/, "\\&Iuml;", s)
# ignore eth and thorn
gsub(/\{\\~N}/, "\\&Ntilde;", s)
gsub(/\{\\\O}/, "\\&Ograve;", s)
gsub(/\{\\'O}/, "\\&Oacute;", s)
gsub(/\{\\[\^]O}/, "\\&Ocirc;", s)
gsub(/\{\\~O}/, "\\&Otilde;", s)
##gsub(/\{\\\"O}/, "\\&Ouml;", s)
gsub(/\{\\\"\{O\}\}/, "\\&Ouml;", s)
gsub(/\{\\O}/, "\\&Oslash;", s)
gsub(/\{\\\U}/, "\\&Ugrave;", s)
gsub(/\{\\'U}/, "\\&Uacute;", s)
gsub(/\{\\[\^]U}/, "\\&Ucirc;", s)
##gsub(/\{\\\"U}/, "\\&Uuml;", s)
gsub(/\{\\\"\{U\}\}/, "\\&Uuml;", s)
gsub(/\{\\'Y}/, "\\&Yacute;", s)
gsub(/\{\\ss}/, "\\&szlig;", s)
# Others not mentioned in Flynn's book
gsub(/\{\\'\\i}/,"\\&iacute;", s)
gsub(/\{\\'\\j}/,"j", s)
}
return (s)
}
function html_begin_issue()
{
print_line("")
print_line(prefix(2) "<HR>")
print_line("")
print_line(prefix(2) "<H1>")
print_line(prefix(3) "<A NAME=\"" html_label() "\">")
}
function html_begin_pages()
{
return ((HTML && (BIBFILEURL != "")) ? ("<A HREF=\"" BIBFILEURL "#" Citation_label "\">") : "")
}
function html_begin_pre()
{
In_PRE = 1
print_line("<PRE>")
}
function html_begin_title()
{
return ((HTML && (Url != "")) ? ("<A HREF=\"" Url "\">") : "")
}
function html_begin_toc()
{
html_end_toc()
html_begin_pre()
}
function html_body( k)
{
for (k = 1; k <= BodyLines; ++k)
print Body[k]
}
function html_breakpoint(title,maxlength, break_after,k)
{
# Return the largest character position in title AFTER which we
# can break the title across lines, without exceeding maxlength
# visible characters.
if (html_length(title) > maxlength) # then need to split title across lines
{
# In the presence of HTML markup, the initialization of
# k here is complicated, because we need to advance it
# until html_length(title) is at least maxlength,
# without invoking the expensive html_length() function
# too frequently. The need to split the title makes the
# alternative of delayed insertion of HTML markup much
# more complicated.
break_after = 0
for (k = min(maxlength,length(title)); k < length(title); ++k)
{
if (substr(title,k+1,1) == " ")
{ # could break after position k
if (html_length(substr(title,1,k)) <= maxlength)
break_after = k
else # advanced too far, retreat back to last break_after
break
}
}
if (break_after == 0) # no breakpoint found by forward scan
{ # so switch to backward scan
for (k = min(maxlength,length(title)) - 1; \
(k > 0) && (substr(title,k+1,1) != " "); --k)
; # find space at which to break title
if (k < 1) # no break point found
k = length(title) # so must print entire string
}
else
k = break_after
}
else # title fits on one line
k = length(title)
return (k)
}
function html_end_issue()
{
print_line(prefix(3) "</A>")
print_line(prefix(2) "</H1>")
}
function html_end_pages()
{
return ((HTML && (BIBFILEURL != "")) ? "</A>" : "")
}
function html_end_pre()
{
if (In_PRE)
{
print_line("</PRE>")
In_PRE = 0
}
}
function html_end_title()
{
return ((HTML && (Url != "")) ? "</A>" : "")
}
function html_end_toc()
{
html_end_pre()
}
function html_fonts(s, arg,control_word,k,level,n,open_brace)
{
open_brace = index(s,"{")
if (open_brace > 0) # important optimization
{
level = 1
for (k = open_brace + 1; (level != 0) && (k <= length(s)); ++k)
{
if (substr(s,k,1) == "{")
level++
else if (substr(s,k,1) == "}")
level--
}
# {...} is now found at open_brace ... (k-1)
for (control_word in Font_decl_map) # look for {\xxx ...}
{
if (substr(s,open_brace+1,length(control_word)+1) ~ \
("\\" control_word "[^A-Za-z]"))
{
n = open_brace + 1 + length(control_word)
arg = trim(substr(s,n,k - n))
if (Font_decl_map[control_word] == "toupper") # arg -> ARG
arg = toupper(arg)
else if (Font_decl_map[control_word] != "") # arg -> <TAG>arg</TAG>
arg = "<" Font_decl_map[control_word] ">" arg "</" Font_decl_map[control_word] ">"
return (substr(s,1,open_brace-1) arg html_fonts(substr(s,k)))
}
}
for (control_word in Font_cmd_map) # look for \xxx{...}
{
if (substr(s,open_brace - length(control_word),length(control_word)) ~ \
("\\" control_word))
{
n = open_brace + 1
arg = trim(substr(s,n,k - n))
if (Font_cmd_map[control_word] == "toupper") # arg -> ARG
arg = toupper(arg)
else if (Font_cmd_map[control_word] != "") # arg -> <TAG>arg</TAG>
arg = "<" Font_cmd_map[control_word] ">" arg "</" Font_cmd_map[control_word] ">"
n = open_brace - length(control_word) - 1
return (substr(s,1,n) arg html_fonts(substr(s,k)))
}
}
}
return (s)
}
function html_header()
{
USER = ENVIRON["USER"]
if (USER == "")
USER = ENVIRON["LOGNAME"]
if (USER == "")
USER = "????"
"hostname" | getline HOSTNAME
"date" | getline DATE
##("ypcat passwd | grep '^" USER ":' | awk -F: '{print $5}'") | getline PERSONAL_NAME
("getent passwd " USER " | awk -F: '{print $5}'") | getline PERSONAL_NAME
if (PERSONAL_NAME == "")
("grep '^" USER ":' /etc/passwd | awk -F: '{print $5}'") | getline PERSONAL_NAME
print "<!-- WARNING: Do NOT edit this file. It was converted from -->"
print "<!-- BibTeX format to HTML by journal-toc.awk version " VERSION_NUMBER " " VERSION_DATE " -->"
##print "<!-- on " DATE " -->"
##print "<!-- for " PERSONAL_NAME " (" USER "@" HOSTNAME ") -->"
print ""
print ""
print "<!DOCTYPE HTML public \"-//IETF//DTD HTML//EN\">"
print ""
print "<HTML>"
print prefix(1) "<HEAD>"
print prefix(2) "<TITLE>"
print prefix(3) Journal
print prefix(2) "</TITLE>"
print prefix(2) "<LINK REV=\"made\" HREF=\"mailto:" USER "@" HOSTNAME "\">"
print prefix(1) "</HEAD>"
print ""
print prefix(1) "<BODY>"
}
function html_label( label)
{
label = Volume "(" Number "):" Month ":" Year
gsub(/[^A-Za-z0-9():,;.\/\-]/,"",label)
return (label)
}
function html_length(s)
{ # Return visible length of s, ignoring any HTML markup
if (HTML)
{
gsub(/<\/?[^>]*>/,"",s) # remove SGML tags
gsub(/&[A-Za-z0-9]+;/,"",s) # remove SGML entities
}
return (length(s))
}
function html_toc()
{
print prefix(2) "<H1>"
print prefix(3) "Table of contents for issues of " Journal
print prefix(2) "</H1>"
print HTML_TOC
}
function html_toc_entry()
{
HTML_TOC = HTML_TOC " <A HREF=\"#" html_label() "\">"
HTML_TOC = HTML_TOC vol_no_month_year()
HTML_TOC = HTML_TOC "</A><BR>" "\n"
}
function html_trailer()
{
html_end_pre()
print prefix(1) "</BODY>"
print "</HTML>"
}
function initialize()
{
# NB: Update these when the program changes
VERSION_DATE = "[09-Oct-1996]"
VERSION_NUMBER = "1.00"
HTML = (HTML == "") ? 0 : (0 + HTML)
if (INDENT == "")
INDENT = 4
if (HTML == 0)
INDENT = 0 # indentation suppressed in ASCII mode
LEADERS = " . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ."
MAX_TITLE_CHARS = 36 # 36 produces a 79-char output line when there is
# just an initial page number. If this is
# increased, the LEADERS string may need to be
# lengthened.
MIN_LEADERS = 4 # Minimum number of characters from LEADERS
# required when leaders are used. The total
# number of characters that can appear in a
# title line is MAX_TITLE_CHARS + MIN_LEADERS.
# Leaders are omitted when the title length is
# between MAX_TITLE_CHARS and this sum.
MIN_LEADERS_SPACE = " " # must be at least MIN_LEADERS characters long
Month_expansion["jan"] = "January"
Month_expansion["feb"] = "February"
Month_expansion["mar"] = "March"
Month_expansion["apr"] = "April"
Month_expansion["may"] = "May"
Month_expansion["jun"] = "June"
Month_expansion["jul"] = "July"
Month_expansion["aug"] = "August"
Month_expansion["sep"] = "September"
Month_expansion["oct"] = "October"
Month_expansion["nov"] = "November"
Month_expansion["dec"] = "December"
Font_cmd_map["\\emph"] = "EM"
Font_cmd_map["\\textbf"] = "B"
Font_cmd_map["\\textit"] = "I"
Font_cmd_map["\\textmd"] = ""
Font_cmd_map["\\textrm"] = ""
Font_cmd_map["\\textsc"] = "toupper"
Font_cmd_map["\\textsl"] = "I"
Font_cmd_map["\\texttt"] = "t"
Font_cmd_map["\\textup"] = ""
Font_decl_map["\\bf"] = "B"
Font_decl_map["\\em"] = "EM"
Font_decl_map["\\it"] = "I"
Font_decl_map["\\rm"] = ""
Font_decl_map["\\sc"] = "toupper"
Font_decl_map["\\sf"] = ""
Font_decl_map["\\tt"] = "TT"
Font_decl_map["\\itshape"] = "I"
Font_decl_map["\\upshape"] = ""
Font_decl_map["\\slshape"] = "I"
Font_decl_map["\\scshape"] = "toupper"
Font_decl_map["\\mdseries"] = ""
Font_decl_map["\\bfseries"] = "B"
Font_decl_map["\\rmfamily"] = ""
Font_decl_map["\\sffamily"] = ""
Font_decl_map["\\ttfamily"] = "TT"
}
function min(a,b)
{
return (a < b) ? a : b
}
function prefix(level)
{
# Return a prefix of up to 60 blanks
if (In_PRE)
return ("")
else
return (substr(" ", \
1, INDENT * level))
}
function print_line(line)
{
if (HTML) # must buffer in memory so that we can accumulate TOC
Body[++BodyLines] = line
else
print line
}
function print_toc_line(author,title,pages, extra,leaders,n,t)
{
# When we have a multiline title, the hypertext link goes only
# on the first line. A multiline hypertext link looks awful
# because of long underlines under the leading indentation.
if (pages == "") # then no leaders needed in title lines other than last one
t = sprintf("%31s %s%s%s", author, Title_prefix, title, Title_suffix)
else # last title line, with page number
{
n = html_length(title) # potentially expensive
extra = n % 2 # extra space for aligned leader dots
if (n <= MAX_TITLE_CHARS) # then need leaders
leaders = substr(LEADERS, 1, MAX_TITLE_CHARS + MIN_LEADERS - extra - \
min(MAX_TITLE_CHARS,n))
else # title (almost) fills line, so no leaders
leaders = substr(MIN_LEADERS_SPACE,1, \
(MAX_TITLE_CHARS + MIN_LEADERS - extra - n))
t = sprintf("%31s %s%s%s%s%s %4s", \
author, Title_prefix, title, Title_suffix, \
(extra ? " " : ""), leaders, pages)
}
Title_prefix = "" # forget any hypertext
Title_suffix = "" # link material
# Efficency note: an earlier version accumulated the body in a
# single scalar like this: "Body = Body t". Profiling revealed
# this statement as the major hot spot, and the change to array
# storage made the program more than twice as fast. This
# suggests that awk might benefit from an optimization of
# "s = s t" that uses realloc() instead of malloc().
if (HTML)
Body[++BodyLines] = t
else
print t
}
function protect_SGML_characters(s)
{
gsub(/&/,"\\&amp;",s) # NB: this one MUST be first
gsub(/</,"\\&lt;",s)
gsub(/>/,"\\&gt;",s)
gsub(/\"/,"\\&quot;",s)
##gsub(/\"/,"\\&#34;",s)
return (s)
}
function strip_braces(s, k)
{ # strip non-backslashed braces from s and return the result
return (strip_char(strip_char(s,"{"),"}"))
}
function strip_char(s,c, k)
{ # strip non-backslashed instances of c from s, and return the result
k = index(s,c)
if (k > 0) # then found the character
{
if (substr(s,k-1,1) != "\\") # then not backslashed char
s = substr(s,1,k-1) strip_char(substr(s,k+1),c) # so remove it (recursively)
else # preserve backslashed char
s = substr(s,1,k) strip_char(s,k+1,c)
}
return (s)
}
function strip_html(s)
{
gsub(/<\/?[^>]*>/,"",s)
return (s)
}
function terminate()
{
if (HTML)
{
html_end_pre()
HTML = 0 # NB: stop line buffering
html_header()
html_toc()
html_body()
html_trailer()
}
}
function TeX_to_HTML(s, k,n,parts)
{
# First convert the four SGML reserved characters to SGML entities
if (HTML)
{
gsub(/>/, "\\&gt;", s)
gsub(/</, "\\&lt;", s)
##gsub(/"/, "\\&quot;", s)
}
gsub(/[$][$]/,"$$",s) # change display math to triple dollars for split
n = split(s,parts,/[$]/)# split into non-math (odd) and math (even) parts
s = ""
for (k = 1; k <= n; ++k) # unbrace non-math part, leaving math mode intact
s = s ((k > 1) ? "$" : "") \
((k % 2) ? strip_braces(TeX_to_HTML_nonmath(parts[k])) : \
TeX_to_HTML_math(parts[k]))
gsub(/[$][$][$]/,"$$",s) # restore display math
if (HTML)
{
gsub(/"/, "\\&quot;", s)
}
return (s)
}
function TeX_to_HTML_math(s)
{
# Mostly a dummy for now, but HTML 3 could support some math translation
gsub(/\\&/,"\\&amp;",s) # reduce TeX ampersands to SGML entities
return (s)
}
function TeX_to_HTML_nonmath(s)
{
if (index(s,"\\") > 0) # important optimization
{
gsub(/\\slash +/,"/",s) # replace TeX slashes with conventional ones
gsub(/ *\\emdash +/," --- ",s) # replace BibNet emdashes with conventional ones
gsub(/\\%/,"%",s) # reduce TeX percents to conventional ones
gsub(/\\[$]/,"$",s) # reduce TeX dollars to conventional ones
gsub(/\\#/,"#",s) # reduce TeX sharps to conventional ones
if (HTML) # translate TeX markup to HTML
{
gsub(/\\&/,"\\&amp;",s) # reduce TeX ampersands to SGML entities
s = html_accents(s)
s = html_fonts(s)
}
else # plain ASCII text output: discard all TeX markup
{
gsub(/\\\&/, "\\&", s) # reduce TeX ampersands to conventional ones
gsub(/\\[a-z][a-z] +/,"",s) # remove TeX font changes
gsub(/\\[^A-Za-z]/,"",s) # remove remaining TeX control symbols
}
}
return (s)
}
function trim(s)
{
gsub(/^[ \t]+/,"",s)
gsub(/[ \t]+$/,"",s)
return (s)
}
function vol_no_month_year()
{
return ("Volume " wrap(Volume) ", Number " wrap(Number) ", " wrap(Month) ", " wrap(Year))
}
function wrap(value)
{
return (HTML ? ("<STRONG>" value "</STRONG>") : value)
}

27220
hawk/t/journal-toc.in Normal file
View File

File diff suppressed because it is too large Load Diff

3705
hawk/t/journal-toc.out Normal file
View File

File diff suppressed because it is too large Load Diff

1
hawk/tools/Makefile.in
View File

@ -330,7 +330,6 @@ pdfdir = @pdfdir@
prefix = @prefix@
program_transform_name = @program_transform_name@
psdir = @psdir@
runstatedir = @runstatedir@
sbindir = @sbindir@
sharedstatedir = @sharedstatedir@
srcdir = @srcdir@