Revision: 1039
http://vcs.pcre.org/viewvc?view=rev&revision=1039
Author: ph10
Date: 2012-09-13 17:39:03 +0100 (Thu, 13 Sep 2012)
Log Message:
-----------
Allow multiple -o in pcregrep, and add --om-separator.
Modified Paths:
--------------
code/trunk/ChangeLog
code/trunk/RunGrepTest
code/trunk/doc/pcregrep.1
code/trunk/pcregrep.c
code/trunk/testdata/grepoutput
Modified: code/trunk/ChangeLog
===================================================================
--- code/trunk/ChangeLog 2012-09-12 09:03:23 UTC (rev 1038)
+++ code/trunk/ChangeLog 2012-09-13 16:39:03 UTC (rev 1039)
@@ -76,7 +76,12 @@
17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
+18. If the --only-matching (-o) option in pcregrep is specified multiple
+ times, each one causes appropriate output. For example, -o1 -o2 outputs the
+ substrings matched by the 1st and 2nd capturing parentheses. A separating
+ string can be specified by --om-separator (default empty).
+
Version 8.31 06-July-2012
-------------------------
Modified: code/trunk/RunGrepTest
===================================================================
--- code/trunk/RunGrepTest 2012-09-12 09:03:23 UTC (rev 1038)
+++ code/trunk/RunGrepTest 2012-09-13 16:39:03 UTC (rev 1039)
@@ -478,6 +478,15 @@
(cd $srcdir; $valgrind $pcregrep -L -r --include grepinput --exclude-from $builddir/testtemp1 --exclude-from=$builddir/testtemp2 --exclude-dir='^\.' 'fox' ./testdata | sort) >>testtry
echo "RC=$?" >>testtry
+echo "---------------------------- Test 100 ------------------------------" >>testtry
+(cd $srcdir; $valgrind $pcregrep -Ho2 --only-matching=1 -o3 '(\w+) binary (\w+)(\.)?' ./testdata/grepinput) >>testtry
+echo "RC=$?" >>testtry
+
+echo "---------------------------- Test 101 ------------------------------" >>testtry
+(cd $srcdir; $valgrind $pcregrep -o3 -Ho2 -o12 --only-matching=1 -o3 --colour=always --om-separator='/' '(\w+) binary (\w+)(\.)?' ./testdata/grepinput) >>testtry
+echo "RC=$?" >>testtry
+
+
# Now compare the results.
$cf $srcdir/testdata/grepoutput testtry
Modified: code/trunk/doc/pcregrep.1
===================================================================
--- code/trunk/doc/pcregrep.1 2012-09-12 09:03:23 UTC (rev 1038)
+++ code/trunk/doc/pcregrep.1 2012-09-13 16:39:03 UTC (rev 1039)
@@ -1,4 +1,4 @@
-.TH PCREGREP 1 "15 August 2012" "PCRE 8.32"
+.TH PCREGREP 1 "13 September 2012" "PCRE 8.32"
.SH NAME
pcregrep - a grep with Perl-compatible regular expressions.
.SH SYNOPSIS
@@ -70,8 +70,8 @@
of the line.
.P
This behaviour means that the order in which multiple patterns are specified
-can affect the output when one of the above options is used. This is no longer
-the same behaviour as GNU grep, which now manages to display earlier matches
+can affect the output when one of the above options is used. This is no longer
+the same behaviour as GNU grep, which now manages to display earlier matches
for later patterns (as long as there is no overlap).
.P
Patterns that can match an empty string are accepted, but empty string
@@ -209,7 +209,7 @@
single pattern that starts with a hyphen. When \fB-e\fP is used, no argument
pattern is taken from the command line; all arguments are treated as file
names. There is no limit to the number of patterns. They are applied to each
-line in the order in which they are defined until one matches.
+line in the order in which they are defined until one matches.
.sp
If \fB-f\fP is used with \fB-e\fP, the command line patterns are matched first,
followed by the patterns from the file(s), independent of the order in which
@@ -227,7 +227,7 @@
obtained from \fB--file-list\fP, or by scanning a directory. The pattern is a
PCRE regular expression, and is matched against the final component of the file
name, not the entire path. The \fB-F\fP, \fB-w\fP, and \fB-x\fP options do not
-apply to this pattern. The option may be given any number of times in order to
+apply to this pattern. The option may be given any number of times in order to
specify multiple patterns. If a file name matches both an \fB--include\fP
and an \fB--exclude\fP pattern, it is excluded. There is no short form for this
option.
@@ -337,7 +337,7 @@
\fB--include-from=\fP\fIfilename\fP
Treat each non-empty line of the file as the data for an \fB--include\fP
option. What constitutes a newline for this purpose is the operating system's
-default. The \fB--newline\fP option has no effect on this option. This option
+default. The \fB--newline\fP option has no effect on this option. This option
may be given any number of times; all the files are read.
.TP
\fB--include-dir\fP=\fIpattern\fP
@@ -454,8 +454,8 @@
makes it possible to use \fBpcregrep\fP to scan files that have come from other
environments without having to modify their line endings. If the data that is
being scanned does not agree with the convention set by this option,
-\fBpcregrep\fP may behave in strange ways. Note that this option does not
-apply to files specified by the \fB-f\fP, \fB--exclude-from\fP, or
+\fBpcregrep\fP may behave in strange ways. Note that this option does not
+apply to files specified by the \fB-f\fP, \fB--exclude-from\fP, or
\fB--include-from\fP options, which are expected to use the operating system's
standard newline sequence.
.TP
@@ -485,14 +485,23 @@
.TP
\fB-o\fP\fInumber\fP, \fB--only-matching\fP=\fInumber\fP
Show only the part of the line that matched the capturing parentheses of the
-given number. Up to 32 capturing parentheses are supported. Because these
-options can be given without an argument (see above), if an argument is
-present, it must be given in the same shell item, for example, -o3 or
---only-matching=2. The comments given for the non-argument case above also
-apply to this case. If the specified capturing parentheses do not exist in the
-pattern, or were not set in the match, nothing is output unless the file name
-or line number are being printed.
+given number. Up to 32 capturing parentheses are supported, and -o0 is
+equivalent to \fB-o\fP without a number. Because these options can be given
+without an argument (see above), if an argument is present, it must be given in
+the same shell item, for example, -o3 or --only-matching=2. The comments given
+for the non-argument case above also apply to this case. If the specified
+capturing parentheses do not exist in the pattern, or were not set in the
+match, nothing is output unless the file name or line number are being printed.
+.sp
+If this option is given multiple times, multiple substrings are output, in the
+order the options are given. For example, -o3 -o1 -o3 causes the substrings
+matched by capturing parentheses 3 and 1 and then 3 again to be output. By
+default, there is no separator (but see the next option).
.TP
+\fB--om-separator\fP=\fItext\fP
+Specify a separating string for multiple occurrences of \fB-o\fP. The default
+is an empty string. Separating strings are never coloured.
+.TP
\fB-q\fP, \fB--quiet\fP
Work quietly, that is, display nothing except error messages. The exit
status indicates whether or not any matches were found.
@@ -514,7 +523,7 @@
.TP
\fB-u\fP, \fB--utf-8\fP
Operate in UTF-8 mode. This option is available only if PCRE has been compiled
-with UTF-8 support. All patterns (including those for any \fB--exclude\fP and
+with UTF-8 support. All patterns (including those for any \fB--exclude\fP and
\fB--include\fP options) and all subject lines that are scanned must be valid
strings of UTF-8 characters.
.TP
@@ -574,7 +583,7 @@
\fB--xxx-regexp\fP (GNU terminology) is also available as \fB--xxx-regex\fP
(PCRE terminology). However, the \fB--file-list\fP, \fB--file-offsets\fP,
\fB--include-dir\fP, \fB--line-offsets\fP, \fB--locale\fP, \fB--match-limit\fP,
-\fB-M\fP, \fB--multiline\fP, \fB-N\fP, \fB--newline\fP,
+\fB-M\fP, \fB--multiline\fP, \fB-N\fP, \fB--newline\fP, \fB--om-separator\fP,
\fB--recursion-limit\fP, \fB-u\fP, and \fB--utf-8\fP options are specific to
\fBpcregrep\fP, as is the use of the \fB--only-matching\fP option with a
capturing parentheses number.
@@ -665,6 +674,6 @@
.rs
.sp
.nf
-Last updated: 15 August 2012
+Last updated: 13 September 2012
Copyright (c) 1997-2012 University of Cambridge.
.fi
Modified: code/trunk/pcregrep.c
===================================================================
--- code/trunk/pcregrep.c 2012-09-12 09:03:23 UTC (rev 1038)
+++ code/trunk/pcregrep.c 2012-09-13 16:39:03 UTC (rev 1039)
@@ -143,6 +143,7 @@
static char *locale = NULL;
static char *main_buffer = NULL;
static char *newline = NULL;
+static char *om_separator = (char *)"";
static char *stdin_name = (char *)"(standard input)";
static const unsigned char *pcretables = NULL;
@@ -163,7 +164,6 @@
static int DEE_action = DEE_READ;
static int error_count = 0;
static int filenames = FN_DEFAULT;
-static int only_matching = -1;
static int pcre_options = 0;
static int process_options = 0;
@@ -188,9 +188,29 @@
static BOOL omit_zero_count = FALSE;
static BOOL resource_error = FALSE;
static BOOL quiet = FALSE;
+static BOOL show_only_matching = FALSE;
static BOOL silent = FALSE;
static BOOL utf8 = FALSE;
+/* Structure for list of --only-matching capturing numbers. */
+
+typedef struct omstr {
+ struct omstr *next;
+ int groupnum;
+} omstr;
+
+static omstr *only_matching = NULL;
+static omstr *only_matching_last = NULL;
+
+/* Structure for holding the two variables that describe a number chain. */
+
+typedef struct omdatastr {
+ omstr **anchor;
+ omstr **lastptr;
+} omdatastr;
+
+static omdatastr only_matching_data = { &only_matching, &only_matching_last };
+
/* Structure for list of file names (for -f and --{in,ex}clude-from) */
typedef struct fnstr {
@@ -264,7 +284,7 @@
/* Structure for options and list of them */
enum { OP_NODATA, OP_STRING, OP_OP_STRING, OP_NUMBER, OP_LONGNUMBER,
- OP_OP_NUMBER, OP_PATLIST, OP_FILELIST, OP_BINFILES };
+ OP_OP_NUMBER, OP_OP_NUMBERS, OP_PATLIST, OP_FILELIST, OP_BINFILES };
typedef struct option_item {
int type;
@@ -297,6 +317,7 @@
#define N_BINARY_FILES (-18)
#define N_EXCLUDE_FROM (-19)
#define N_INCLUDE_FROM (-20)
+#define N_OM_SEPARATOR (-21)
static option_item optionlist[] = {
{ OP_NODATA, N_NULL, NULL, "", "terminate options" },
@@ -337,7 +358,8 @@
{ OP_NODATA, 'M', NULL, "multiline", "run in multiline mode" },
{ OP_STRING, 'N', &newline, "newline=type", "set newline type (CR, LF, CRLF, ANYCRLF or ANY)" },
{ OP_NODATA, 'n', NULL, "line-number", "print line number with output lines" },
- { OP_OP_NUMBER, 'o', &only_matching, "only-matching=n", "show only the part of the line that matched" },
+ { OP_OP_NUMBERS, 'o', &only_matching_data, "only-matching=n", "show only the part of the line that matched" },
+ { OP_STRING, N_OM_SEPARATOR, &om_separator, "om-separator=text", "set separator for multiple -o output" },
{ OP_NODATA, 'q', NULL, "quiet", "suppress output, just set return code" },
{ OP_NODATA, 'r', NULL, "recursive", "recursively scan sub-directories" },
{ OP_PATLIST, N_EXCLUDE,&exclude_patdata, "exclude=pattern","exclude matching files when recursing" },
@@ -393,6 +415,30 @@
/*************************************************
+* Exit from the program *
+*************************************************/
+
+/* If there has been a resource error, give a suitable message.
+
+Argument: the return code
+Returns: does not return
+*/
+
+static void
+pcregrep_exit(int rc)
+{
+if (resource_error)
+ {
+ fprintf(stderr, "pcregrep: Error %d, %d or %d means that a resource limit "
+ "was exceeded.\n", PCRE_ERROR_MATCHLIMIT, PCRE_ERROR_RECURSIONLIMIT,
+ PCRE_ERROR_JIT_STACKLIMIT);
+ fprintf(stderr, "pcregrep: Check your regex for nested unlimited loops.\n");
+ }
+exit(rc);
+}
+
+
+/*************************************************
* Add item to chain of patterns *
*************************************************/
@@ -403,7 +449,7 @@
s pattern string to add
after if not NULL points to item to insert after
-Returns: new pattern block, or NULL after malloc failure
+Returns: new pattern block
*/
static patstr *
@@ -413,7 +459,7 @@
if (p == NULL)
{
fprintf(stderr, "pcregrep: malloc failed\n");
- return NULL;
+ pcregrep_exit(2);
}
if (strlen(s) > MAXPATLEN)
{
@@ -481,31 +527,6 @@
/*************************************************
-* Exit from the program *
-*************************************************/
-
-/* If there has been a resource error, give a suitable message.
-
-Argument: the return code
-Returns: does not return
-*/
-
-static void
-pcregrep_exit(int rc)
-{
-if (resource_error)
- {
- fprintf(stderr, "pcregrep: Error %d, %d or %d means that a resource limit "
- "was exceeded.\n", PCRE_ERROR_MATCHLIMIT, PCRE_ERROR_RECURSIONLIMIT,
- PCRE_ERROR_JIT_STACKLIMIT);
- fprintf(stderr, "pcregrep: Check your regex for nested unlimited loops.\n");
- }
-
-exit(rc);
-}
-
-
-/*************************************************
* OS-specific functions *
*************************************************/
@@ -779,6 +800,98 @@
/*************************************************
+* Usage function *
+*************************************************/
+
+static int
+usage(int rc)
+{
+option_item *op;
+fprintf(stderr, "Usage: pcregrep [-");
+for (op = optionlist; op->one_char != 0; op++)
+ {
+ if (op->one_char > 0) fprintf(stderr, "%c", op->one_char);
+ }
+fprintf(stderr, "] [long options] [pattern] [files]\n");
+fprintf(stderr, "Type `pcregrep --help' for more information and the long "
+ "options.\n");
+return rc;
+}
+
+
+
+/*************************************************
+* Help function *
+*************************************************/
+
+static void
+help(void)
+{
+option_item *op;
+
+printf("Usage: pcregrep [OPTION]... [PATTERN] [FILE1 FILE2 ...]\n");
+printf("Search for PATTERN in each FILE or standard input.\n");
+printf("PATTERN must be present if neither -e nor -f is used.\n");
+printf("\"-\" can be used as a file name to mean STDIN.\n");
+
+#ifdef SUPPORT_LIBZ
+printf("Files whose names end in .gz are read using zlib.\n");
+#endif
+
+#ifdef SUPPORT_LIBBZ2
+printf("Files whose names end in .bz2 are read using bzlib2.\n");
+#endif
+
+#if defined SUPPORT_LIBZ || defined SUPPORT_LIBBZ2
+printf("Other files and the standard input are read as plain files.\n\n");
+#else
+printf("All files are read as plain files, without any interpretation.\n\n");
+#endif
+
+printf("Example: pcregrep -i 'hello.*world' menu.h main.c\n\n");
+printf("Options:\n");
+
+for (op = optionlist; op->one_char != 0; op++)
+ {
+ int n;
+ char s[4];
+
+ /* Two options were accidentally implemented and documented with underscores
+ instead of hyphens in their names, something that was not noticed for quite a
+ few releases. When fixing this, I left the underscored versions in the list
+ in case people were using them. However, we don't want to display them in the
+ help data. There are no other options that contain underscores, and we do not
+ expect ever to implement such options. Therefore, just omit any option that
+ contains an underscore. */
+
+ if (strchr(op->long_name, '_') != NULL) continue;
+
+ if (op->one_char > 0 && (op->long_name)[0] == 0)
+ n = 31 - printf(" -%c", op->one_char);
+ else
+ {
+ if (op->one_char > 0) sprintf(s, "-%c,", op->one_char);
+ else strcpy(s, " ");
+ n = 31 - printf(" %s --%s", s, op->long_name);
+ }
+
+ if (n < 1) n = 1;
+ printf("%.*s%s\n", n, " ", op->help_text);
+ }
+
+printf("\nNumbers may be followed by K or M, e.g. --buffer-size=100K.\n");
+printf("The default value for --buffer-size is %d.\n", PCREGREP_BUFSIZE);
+printf("When reading patterns or file names from a file, trailing white\n");
+printf("space is removed and blank lines are ignored.\n");
+printf("The maximum size of any pattern is %d bytes.\n", MAXPATLEN);
+
+printf("\nWith no FILEs, read standard input. If fewer than two FILEs given, assume -h.\n");
+printf("Exit status is 0 if any matches, 1 if no matches, and 2 if trouble.\n");
+}
+
+
+
+/*************************************************
* Test exclude/includes *
*************************************************/
@@ -818,6 +931,99 @@
/*************************************************
+* Decode integer argument value *
+*************************************************/
+
+/* Integer arguments can be followed by K or M. Avoid the use of strtoul()
+because SunOS4 doesn't have it. This is used only for unpicking arguments, so
+just keep it simple.
+
+Arguments:
+ option_data the option data string
+ op the option item (for error messages)
+ longop TRUE if option given in long form
+
+Returns: a long integer
+*/
+
+static long int
+decode_number(char *option_data, option_item *op, BOOL longop)
+{
+unsigned long int n = 0;
+char *endptr = option_data;
+while (*endptr != 0 && isspace((unsigned char)(*endptr))) endptr++;
+while (isdigit((unsigned char)(*endptr)))
+ n = n * 10 + (int)(*endptr++ - '0');
+if (toupper(*endptr) == 'K')
+ {
+ n *= 1024;
+ endptr++;
+ }
+else if (toupper(*endptr) == 'M')
+ {
+ n *= 1024*1024;
+ endptr++;
+ }
+
+if (*endptr != 0) /* Error */
+ {
+ if (longop)
+ {
+ char *equals = strchr(op->long_name, '=');
+ int nlen = (equals == NULL)? (int)strlen(op->long_name) :
+ (int)(equals - op->long_name);
+ fprintf(stderr, "pcregrep: Malformed number \"%s\" after --%.*s\n",
+ option_data, nlen, op->long_name);
+ }
+ else
+ fprintf(stderr, "pcregrep: Malformed number \"%s\" after -%c\n",
+ option_data, op->one_char);
+ pcregrep_exit(usage(2));
+ }
+
+return n;
+}
+
+
+
+/*************************************************
+* Add item to a chain of numbers *
+*************************************************/
+
+/* Used to add an item onto a chain, or just return an unconnected item if the
+"after" argument is NULL.
+
+Arguments:
+ n the number to add
+ after if not NULL points to item to insert after
+
+Returns: new number block
+*/
+
+static omstr *
+add_number(int n, omstr *after)
+{
+omstr *om = (omstr *)malloc(sizeof(omstr));
+
+if (om == NULL)
+ {
+ fprintf(stderr, "pcregrep: malloc failed\n");
+ pcregrep_exit(2);
+ }
+om->next = NULL;
+om->groupnum = n;
+
+if (after != NULL)
+ {
+ om->next = after->next;
+ after->next = om;
+ }
+return om;
+}
+
+
+
+/*************************************************
* Read one line of input *
*************************************************/
@@ -1001,7 +1207,7 @@
case 0x2029: /* Unicode PS */
*lenptr = 3;
return p;
-#endif /* Not EBCDIC */
+#endif /* Not EBCDIC */
default:
break;
@@ -1099,11 +1305,11 @@
case '\v': /* VT */
case '\f': /* FF */
case '\r': /* CR */
-#ifndef EBCDIE
+#ifndef EBCDIE
case 0x85: /* Unicode NEL */
case 0x2028: /* Unicode LS */
case 0x2029: /* Unicode PS */
-#endif /* Not EBCDIC */
+#endif /* Not EBCDIC */
return p;
default:
@@ -1412,8 +1618,9 @@
}
#endif
- /* We come back here after a match when the -o option (only_matching) is set,
- in order to find any further matches in the same line. */
+ /* We come back here after a match when show_only_matching is set, in order
+ to find any further matches in the same line. This applies to
+ --only-matching, --file-offsets, and --line-offsets. */
ONLY_MATCHING_RESTART:
@@ -1460,40 +1667,64 @@
else if (quiet) return 0;
- /* The --only-matching option prints just the substring that matched, or a
- captured portion of it, as long as this string is not empty, and the
- --file-offsets and --line-offsets options output offsets for the matching
- substring (they both force --only-matching = 0). None of these options
- prints any context. Afterwards, adjust the start and then jump back to look
- for further matches in the same line. If we are in invert mode, however,
- nothing is printed and we do not restart - this could still be useful
- because the return code is set. */
+ /* The --only-matching option prints just the substring that matched,
+ and/or one or more captured portions of it, as long as these strings are
+ not empty. The --file-offsets and --line-offsets options output offsets for
+ the matching substring (all three set show_only_matching). None of these
+ mutually exclusive options prints any context. Afterwards, adjust the start
+ and then jump back to look for further matches in the same line. If we are
+ in invert mode, however, nothing is printed and we do not restart - this
+ could still be useful because the return code is set. */
- else if (only_matching >= 0)
+ else if (show_only_matching)
{
if (!invert)
{
if (printname != NULL) fprintf(stdout, "%s:", printname);
if (number) fprintf(stdout, "%d:", linenumber);
+
+ /* Handle --line-offsets */
+
if (line_offsets)
fprintf(stdout, "%d,%d\n", (int)(matchptr + offsets[0] - ptr),
offsets[1] - offsets[0]);
+
+ /* Handle --file-offsets */
+
else if (file_offsets)
fprintf(stdout, "%d,%d\n",
(int)(filepos + matchptr + offsets[0] - ptr),
offsets[1] - offsets[0]);
- else if (only_matching < mrc)
+
+ /* Handle --only-matching, which may occur many times */
+
+ else
{
- int plen = offsets[2*only_matching + 1] - offsets[2*only_matching];
- if (plen > 0)
+ BOOL printed = FALSE;
+ omstr *om;
+
+ for (om = only_matching; om != NULL; om = om->next)
{
- if (do_colour) fprintf(stdout, "%c[%sm", 0x1b, colour_string);
- FWRITE(matchptr + offsets[only_matching*2], 1, plen, stdout);
- if (do_colour) fprintf(stdout, "%c[00m", 0x1b);
- fprintf(stdout, "\n");
+ int n = om->groupnum;
+ if (n < mrc)
+ {
+ int plen = offsets[2*n + 1] - offsets[2*n];
+ if (plen > 0)
+ {
+ if (printed) fprintf(stdout, "%s", om_separator);
+ if (do_colour) fprintf(stdout, "%c[%sm", 0x1b, colour_string);
+ FWRITE(matchptr + offsets[n*2], 1, plen, stdout);
+ if (do_colour) fprintf(stdout, "%c[00m", 0x1b);
+ printed = TRUE;
+ }
+ }
}
+
+ if (printed || printname != NULL || number) fprintf(stdout, "\n");
}
- else if (printname != NULL || number) fprintf(stdout, "\n");
+
+ /* Prepare to repeat to find the next match */
+
match = FALSE;
if (line_buffered) fflush(stdout);
rc = 0; /* Had some success */
@@ -1756,7 +1987,7 @@
/* End of file; print final "after" lines if wanted; do_after_lines sets
hyphenpending if it prints something. */
-if (only_matching < 0 && !count_only)
+if (!show_only_matching && !count_only)
{
do_after_lines(lastmatchnumber, lastmatchrestart, endptr, printname);
hyphenpending |= endhyphenpending;
@@ -2003,102 +2234,7 @@
-
/*************************************************
-* Usage function *
-*************************************************/
-
-static int
-usage(int rc)
-{
-option_item *op;
-fprintf(stderr, "Usage: pcregrep [-");
-for (op = optionlist; op->one_char != 0; op++)
- {
- if (op->one_char > 0) fprintf(stderr, "%c", op->one_char);
- }
-fprintf(stderr, "] [long options] [pattern] [files]\n");
-fprintf(stderr, "Type `pcregrep --help' for more information and the long "
- "options.\n");
-return rc;
-}
-
-
-
-
-/*************************************************
-* Help function *
-*************************************************/
-
-static void
-help(void)
-{
-option_item *op;
-
-printf("Usage: pcregrep [OPTION]... [PATTERN] [FILE1 FILE2 ...]\n");
-printf("Search for PATTERN in each FILE or standard input.\n");
-printf("PATTERN must be present if neither -e nor -f is used.\n");
-printf("\"-\" can be used as a file name to mean STDIN.\n");
-
-#ifdef SUPPORT_LIBZ
-printf("Files whose names end in .gz are read using zlib.\n");
-#endif
-
-#ifdef SUPPORT_LIBBZ2
-printf("Files whose names end in .bz2 are read using bzlib2.\n");
-#endif
-
-#if defined SUPPORT_LIBZ || defined SUPPORT_LIBBZ2
-printf("Other files and the standard input are read as plain files.\n\n");
-#else
-printf("All files are read as plain files, without any interpretation.\n\n");
-#endif
-
-printf("Example: pcregrep -i 'hello.*world' menu.h main.c\n\n");
-printf("Options:\n");
-
-for (op = optionlist; op->one_char != 0; op++)
- {
- int n;
- char s[4];
-
- /* Two options were accidentally implemented and documented with underscores
- instead of hyphens in their names, something that was not noticed for quite a
- few releases. When fixing this, I left the underscored versions in the list
- in case people were using them. However, we don't want to display them in the
- help data. There are no other options that contain underscores, and we do not
- expect ever to implement such options. Therefore, just omit any option that
- contains an underscore. */
-
- if (strchr(op->long_name, '_') != NULL) continue;
-
- if (op->one_char > 0 && (op->long_name)[0] == 0)
- n = 31 - printf(" -%c", op->one_char);
- else
- {
- if (op->one_char > 0) sprintf(s, "-%c,", op->one_char);
- else strcpy(s, " ");
- n = 31 - printf(" %s --%s", s, op->long_name);
- }
-
- if (n < 1) n = 1;
- printf("%.*s%s\n", n, " ", op->help_text);
- }
-
-printf("\nNumbers may be followed by K or M, e.g. --buffer-size=100K.\n");
-printf("The default value for --buffer-size is %d.\n", PCREGREP_BUFSIZE);
-printf("When reading patterns or file names from a file, trailing white\n");
-printf("space is removed and blank lines are ignored.\n");
-printf("The maximum size of any pattern is %d bytes.\n", MAXPATLEN);
-
-printf("\nWith no FILEs, read standard input. If fewer than two FILEs given, assume -h.\n");
-printf("Exit status is 0 if any matches, 1 if no matches, and 2 if trouble.\n");
-}
-
-
-
-
-/*************************************************
* Handle a single-letter, no data option *
*************************************************/
@@ -2123,7 +2259,12 @@
case 'L': filenames = FN_NOMATCH_ONLY; break;
case 'M': multiline = TRUE; options |= PCRE_MULTILINE|PCRE_FIRSTLINE; break;
case 'n': number = TRUE; break;
- case 'o': only_matching = 0; break;
+
+ case 'o':
+ only_matching_last = add_number(0, only_matching_last);
+ if (only_matching == NULL) only_matching = only_matching_last;
+ break;
+
case 'q': quiet = TRUE; break;
case 'r': dee_action = dee_RECURSE; break;
case 's': silent = TRUE; break;
@@ -2509,6 +2650,7 @@
{
char *s = argv[i] + 1;
longop = FALSE;
+
while (*s != 0)
{
for (op = optionlist; op->one_char != 0; op++)
@@ -2521,29 +2663,35 @@
*s, argv[i]);
pcregrep_exit(usage(2));
}
+
+ option_data = s+1;
+
+ /* Break out if this is the last character in the string; it's handled
+ below like a single multi-char option. */
+
+ if (*option_data == 0) break;
- /* Check for a single-character option that has data: OP_OP_NUMBER
- is used for one that either has a numerical number or defaults, i.e. the
- data is optional. If a digit follows, there is data; if not, carry on
+ /* Check for a single-character option that has data: OP_OP_NUMBER(S)
+ are used for ones that either have a numerical number or defaults, i.e.
+ the data is optional. If a digit follows, there is data; if not, carry on
with other single-character options in the same string. */
- option_data = s+1;
- if (op->type == OP_OP_NUMBER)
+ if (op->type == OP_OP_NUMBER || op->type == OP_OP_NUMBERS)
{
if (isdigit((unsigned char)s[1])) break;
}
- else /* Check for end or a dataless option */
+ else /* Check for an option with data */
{
- if (op->type != OP_NODATA || s[1] == 0) break;
+ if (op->type != OP_NODATA) break;
}
/* Handle a single-character option with no data, then loop for the
next character in the string. */
-
+
pcre_options = handle_option(*s++, pcre_options);
}
}
-
+
/* At this point we should have op pointing to a matched option. If the type
is NO_DATA, it means that there is no data, and the option might set
something in the PCRE options. */
@@ -2554,13 +2702,14 @@
continue;
}
- /* If the option type is OP_OP_STRING or OP_OP_NUMBER, it's an option that
+ /* If the option type is OP_OP_STRING or OP_OP_NUMBER(S), it's an option that
either has a value or defaults to something. It cannot have data in a
separate item. At the moment, the only such options are "colo(u)r",
"only-matching", and Jeffrey Friedl's special -S debugging option. */
-
+
if (*option_data == 0 &&
- (op->type == OP_OP_STRING || op->type == OP_OP_NUMBER))
+ (op->type == OP_OP_STRING || op->type == OP_OP_NUMBER ||
+ op->type == OP_OP_NUMBERS))
{
switch (op->one_char)
{
@@ -2569,7 +2718,8 @@
break;
case 'o':
- only_matching = 0;
+ only_matching_last = add_number(0, only_matching_last);
+ if (only_matching == NULL) only_matching = only_matching_last;
break;
#ifdef JFRIEDL_DEBUG
@@ -2593,17 +2743,28 @@
option_data = argv[++i];
}
+ /* If the option type is OP_OP_NUMBERS, the value is a number that is to be
+ added to a chain of numbers. */
+
+ if (op->type == OP_OP_NUMBERS)
+ {
+ unsigned long int n = decode_number(option_data, op, longop);
+ omdatastr *omd = (omdatastr *)op->dataptr;
+ *(omd->lastptr) = add_number((int)n, *(omd->lastptr));
+ if (*(omd->anchor) == NULL) *(omd->anchor) = *(omd->lastptr);
+ }
+
/* If the option type is OP_PATLIST, it's the -e option, or one of the
include/exclude options, which can be called multiple times to create lists
of patterns. */
- if (op->type == OP_PATLIST)
- {
- patdatastr *pd = (patdatastr *)op->dataptr;
- *(pd->lastptr) = add_pattern(option_data, *(pd->lastptr));
- if (*(pd->lastptr) == NULL) goto EXIT2;
- if (*(pd->anchor) == NULL) *(pd->anchor) = *(pd->lastptr);
- }
+ else if (op->type == OP_PATLIST)
+ {
+ patdatastr *pd = (patdatastr *)op->dataptr;
+ *(pd->lastptr) = add_pattern(option_data, *(pd->lastptr));
+ if (*(pd->lastptr) == NULL) goto EXIT2;
+ if (*(pd->anchor) == NULL) *(pd->anchor) = *(pd->lastptr);
+ }
/* If the option type is OP_FILELIST, it's one of the options that names a
file. */
@@ -2644,53 +2805,18 @@
}
}
- /* Otherwise, deal with single string or numeric data values. */
+ /* Otherwise, deal with a single string or numeric data value. */
else if (op->type != OP_NUMBER && op->type != OP_LONGNUMBER &&
op->type != OP_OP_NUMBER)
{
*((char **)op->dataptr) = option_data;
}
-
- /* Avoid the use of strtoul() because SunOS4 doesn't have it. This is used
- only for unpicking arguments, so just keep it simple. */
-
else
{
- unsigned long int n = 0;
- char *endptr = option_data;
- while (*endptr != 0 && isspace((unsigned char)(*endptr))) endptr++;
- while (isdigit((unsigned char)(*endptr)))
- n = n * 10 + (int)(*endptr++ - '0');
- if (toupper(*endptr) == 'K')
- {
- n *= 1024;
- endptr++;
- }
- else if (toupper(*endptr) == 'M')
- {
- n *= 1024*1024;
- endptr++;
- }
- if (*endptr != 0)
- {
- if (longop)
- {
- char *equals = strchr(op->long_name, '=');
- int nlen = (equals == NULL)? (int)strlen(op->long_name) :
- (int)(equals - op->long_name);
- fprintf(stderr, "pcregrep: Malformed number \"%s\" after --%.*s\n",
- option_data, nlen, op->long_name);
- }
- else
- fprintf(stderr, "pcregrep: Malformed number \"%s\" after -%c\n",
- option_data, op->one_char);
- pcregrep_exit(usage(2));
- }
- if (op->type == OP_LONGNUMBER)
- *((unsigned long int *)op->dataptr) = n;
- else
- *((int *)op->dataptr) = n;
+ unsigned long int n = decode_number(option_data, op, longop);
+ if (op->type == OP_LONGNUMBER) *((unsigned long int *)op->dataptr) = n;
+ else *((int *)op->dataptr) = n;
}
}
@@ -2704,9 +2830,10 @@
}
/* Only one of --only-matching, --file-offsets, or --line-offsets is permitted.
-However, the latter two set only_matching. */
+However, all three set show_only_matching because they display, each in their
+own way, only the data that has matched. */
-if ((only_matching >= 0 && (file_offsets || line_offsets)) ||
+if ((only_matching != NULL && (file_offsets || line_offsets)) ||
(file_offsets && line_offsets))
{
fprintf(stderr, "pcregrep: Cannot mix --only-matching, --file-offsets "
@@ -2714,7 +2841,8 @@
pcregrep_exit(usage(2));
}
-if (file_offsets || line_offsets) only_matching = 0;
+if (only_matching != NULL || file_offsets || line_offsets)
+ show_only_matching = TRUE;
/* If a locale has not been provided as an option, see if the LC_CTYPE or
LC_ALL environment variable is set, and if so, use it. */
@@ -2880,7 +3008,7 @@
goto EXIT2;
}
-/* Study the regular expressions, as we will be running them many times. If an
+/* Study the regular expressions, as we will be running them many times. If an
extra block is needed for a limit, set PCRE_STUDY_EXTRA_NEEDED so that one is
returned, even if studying produces no data. */
@@ -2911,7 +3039,7 @@
}
/* If --match-limit or --recursion-limit was set, put the value(s) into the
-pcre_extra block for each pattern. There will always be an extra block because
+pcre_extra block for each pattern. There will always be an extra block because
of the use of PCRE_STUDY_EXTRA_NEEDED above. */
for (cp = patterns; cp != NULL; cp = cp->next)
@@ -2921,7 +3049,7 @@
cp->hint->flags |= PCRE_EXTRA_MATCH_LIMIT;
cp->hint->match_limit = match_limit;
}
-
+
if (match_limit_recursion > 0)
{
cp->hint->flags |= PCRE_EXTRA_MATCH_LIMIT_RECURSION;
@@ -3033,6 +3161,13 @@
free_file_chain(pattern_files);
free_file_chain(file_lists);
+while (only_matching != NULL)
+ {
+ omstr *this = only_matching;
+ only_matching = this->next;
+ free(this);
+ }
+
pcregrep_exit(rc);
EXIT2:
Modified: code/trunk/testdata/grepoutput
===================================================================
--- code/trunk/testdata/grepoutput 2012-09-12 09:03:23 UTC (rev 1038)
+++ code/trunk/testdata/grepoutput 2012-09-13 16:39:03 UTC (rev 1039)
@@ -695,3 +695,13 @@
./testdata/grepinput3
./testdata/grepinputx
RC=0
+---------------------------- Test 100 ------------------------------
+./testdata/grepinput:zerothe.
+./testdata/grepinput:zeroa
+./testdata/grepinput:zerothe.
+RC=0
+---------------------------- Test 101 ------------------------------
+./testdata/grepinput:?[1;31m.?[00m/?[1;31mzero?[00m/?[1;31mthe?[00m/?[1;31m.?[00m
+./testdata/grepinput:?[1;31mzero?[00m/?[1;31ma?[00m
+./testdata/grepinput:?[1;31m.?[00m/?[1;31mzero?[00m/?[1;31mthe?[00m/?[1;31m.?[00m
+RC=0
