added more documentation with some code fix

qse/doc/page/sed.doc

@@ -3,9 +3,10 @@
 @section sed_intro INTRODUCTION
 
 The sed stream editor is a non-interactive text editing tool commonly used
-on Unix environment. Sed reads text from an input stream, applies a set of 
+on Unix environment. Sed reads text from an input stream, stores it to
-editing commands, and writes the results to an output stream. Typically, 
+pattern space, manipulates the pattern space by applying a set of editing
-the input and output streams are a console or a file. 
+commands, and writes the pattern space to an output stream. Typically, the
+input and output streams are a console or a file. 
 
 @b QSE provides an embeddable stream editor that supports most of 
 the conventional sed commands and implements input and output streams as a 
@@ -28,7 +29,8 @@ A line selector selects input lines to apply a command to and has the following
 forms:
 - address - specify a single address
 - address,address - specify an address range 
-- start~step - specify a starting line and a step
+- start~step - specify a starting line and a step. 
+               #QSE_SED_CLASSIC disables this form.
 
 An @b address is a line number, a regular expression, or a dollar sign ($) 
 while a @b start and a @b step is a line number. 
@@ -58,15 +60,15 @@ lines except the first line.
 A command without a line selector is applied to all input lines; 
 A command with a single address is applied to an input line that matches 
 the address; A command with an address range is applied to all input 
-lines within the range, inclusive; A command with a starting line and 
+lines within the range, inclusive; A command with a start and a step is
-a step is applied to every @b step'th line starting from the line @b start.
+applied to every @b step'th line starting from the line @b start.
 
 Here is the summary of the commands.
 
 - <b># comment text</b>
-The text beginning from # to the line end is ignored by @b sed;
+The text beginning from # to the line end is ignored; # in a line following
-# in a line following <b>a \\</b>, <b>i \\</b>, and <b>c \\</b> is treated
+<b>a \\</b>, <b>i \\</b>, and <b>c \\</b> is treated literally and does not
-literally and does not introduce a comment.
+introduce a comment.
 - <b>: label</b>
 A label can be composed of letters, digits, periods, hyphens, and underlines.
 It remembers a target label for @b b and @b t commands and prohibits a line
@@ -74,6 +76,55 @@ selector.
 - <b>{</b>
 The left curly bracket forms a command group where you can nest other 
 commands. It should be paired with an ending }.
-- <b>q, Q</b>
+- <b>q</b>
-Both commands terminates the exection of commands. 
+Terminates the exection of commands. Upon termination, it prints the pattern
+space if #QSE_SED_QUIET is not set.
+- <b>Q</b>
+Terminates the exection of commands quietly.
+- <b>a \\ \n text</b>
+Stores @b text into an append buffer which is printed after the pattern 
+space for each input line. If #QSE_SED_CLASSIC is specified, an address range
+is not allowed in the line selector.
+- <b>i \\ \n text</b>
+Inserts @b text into an insert buffer which is printed before the pattern
+space for each input line. If #QSE_SED_CLASSIC is specified, an address range
+is not allowed in the line selector.
+- <b>c \\ \n text</b>
+If a single line is selected for the command (i.e. no line selector, a single
+address line selector, or a start~step line selector is specified), it changes
+pattern space to @b text and skips all subsequent commands for the line.
+If an address range is specified, it deletes pattern space and skips 
+subsequent commands for all input lines but the last, and changes pattern
+space to @b text and skips subsequent commands for the last line.
+- <b>d</b>
+Deletes pattern space and skips subsequent commands for each selected line.
+- <b>D</b>
+Deletes the first line of pattern space. If the pattern space is emptied,
+subsequent commands are skipped. Otherwise, the commands from the first are 
+reapplied to the current pattern space.
+- <b>=</b>
+Prints the current line number. If #QSE_SED_CLASSIC is speccified, an address
+range is not allowed in the line selector.
+- <b>p</b>
+Prints pattern space.
+- <b>P</b>
+Prints the first line of pattern space.
+- <b>l</b>
+Prints pattern space in a visually unambiguous form.
+- <b>h</b> 
+Copies pattern space to hold space
+- <b>H</b> 
+Appends pattern space to hold space
+- <b>g</b> 
+Copies hold space to pattern space
+- <b>G</b> 
+Appends hold space to pattern space
+- <b>x</b> 
+Exchanges pattern space and hold space
+- <b>n</b>
+Prints pattern space and read the next line from the input stream to fill
+pattern space.
+- <b>N</b>
+Prints pattern space and read the next line from the input stream to append it
+to pattern space with a newline inserted.
 */

qse/include/qse/cmn/opt.h

@@ -1,5 +1,5 @@
 /*
- * $Id: opt.h 189 2009-06-07 06:23:53Z hyunghwan.chung $
+ * $Id: opt.h 190 2009-06-07 12:11:56Z hyunghwan.chung $
  *
    Copyright 2006-2009 Chung, Hyung-Hwan.
 
@@ -22,6 +22,11 @@
 #include <qse/types.h>
 #include <qse/macros.h>
 
+/** @file
+ * <qse/cmn/opt.h> defines functions and data structures to process 
+ * command-line arguments. 
+ */
+
 typedef struct qse_opt_t qse_opt_t;
 typedef struct qse_opt_lng_t qse_opt_lng_t;
 
@@ -56,12 +61,24 @@ extern "C" {
 #endif
 
 /**
- * The qse_getopt() function returns QSE_CHAR_EOF when it finishes processing
+ * The qse_getopt() function processes the @a argc command-line arguments
- * command line options. The return values of QSE_T('?') indicates an error.
+ * pointed to by @a argv as configured in @a opt. It can process two 
+ * different option styles: a single character starting with '-', and a 
+ * long name starting with '--'. 
+ *
+ * A character in @a opt.str is treated as a single character option. Should
+ * it require a parameter, specify ':' after it.
+ *
+ * Two special returning option characters indicate special error conditions. 
+ * - @b ? indicates a bad option stored in the @a opt->opt field.
+ * - @b : indicates a bad parameter for an option stored in the 
+ *      @a opt->opt field.
+ *
+ * @return an option character on success, QSE_CHAR_EOF on no more options.
  */
 qse_cint_t qse_getopt (
-	int                argc  /* argument count */, 
+	int                argc, /* argument count */ 
-	qse_char_t* const* argv  /* argument array */,
+	qse_char_t* const* argv, /* argument array */
 	qse_opt_t*         opt   /* option configuration */
 );
 

qse/lib/sed/sed.c

@@ -52,7 +52,7 @@ static const qse_char_t* dflerrstr (qse_sed_t* sed, qse_sed_errnum_t errnum)
 		QSE_T("failed to compile regular expression '${0}'"),
 		QSE_T("failed to match regular expression"),
 		QSE_T("address 1 prohibited for '${0}'"),
-		QSE_T("address 2 prohibited"),
+		QSE_T("address 2 prohibited for '${0}'"),
 		QSE_T("address 2 missing or invalid"),
 		QSE_T("newline expected"),
 		QSE_T("backslash expected"),
@@ -1139,7 +1139,10 @@ static int get_command (qse_sed_t* sed, qse_sed_cmd_t* cmd)
 			if (sed->option & QSE_SED_CLASSIC &&
 			    cmd->a2.type != QSE_SED_ADR_NONE)
 			{
-				SETERR0 (sed, QSE_SED_EA2PHB, sed->src.lnum);
+				SETERR1 (
+					sed, QSE_SED_EA2PHB,
+					sed->src.lnum, &cmd->type, 1
+				);
 				return -1;
 			}
 
@@ -1149,6 +1152,16 @@ static int get_command (qse_sed_t* sed, qse_sed_cmd_t* cmd)
 
 		case QSE_T('a'):
 		case QSE_T('i'):
+			if (sed->option & QSE_SED_CLASSIC &&
+			    cmd->a2.type != QSE_SED_ADR_NONE)
+			{
+				qse_char_t tmpc = c;
+				SETERR1 (
+					sed, QSE_SED_EA2PHB,
+					sed->src.lnum, &tmpc, 1
+				);
+				return -1;
+			}
 		case QSE_T('c'):
 		{
 			cmd->type = c;
@@ -1186,16 +1199,21 @@ static int get_command (qse_sed_t* sed, qse_sed_cmd_t* cmd)
 			if (sed->option & QSE_SED_CLASSIC &&
 			    cmd->a2.type != QSE_SED_ADR_NONE)
 			{
-				SETERR0 (sed, QSE_SED_EA2PHB, sed->src.lnum);
+				qse_char_t tmpc = c;
+				SETERR1 (
+					sed, QSE_SED_EA2PHB,
+					sed->src.lnum, &tmpc, 1
+				);
 				return -1;
 			}
-		case QSE_T('p'):
-		case QSE_T('P'):
-		case QSE_T('l'):
 
 		case QSE_T('d'):
 		case QSE_T('D'):
 
+		case QSE_T('p'):
+		case QSE_T('P'):
+		case QSE_T('l'):
+
 		case QSE_T('h'):
 		case QSE_T('H'):
 		case QSE_T('g'):
@@ -1326,7 +1344,8 @@ int qse_sed_comp (qse_sed_t* sed, const qse_char_t* sptr, qse_size_t slen)
 						return -1;
 					}
 				}
-				else if (delim == QSE_T('~'))
+				else if (!(sed->option & QSE_SED_CLASSIC) && 
+				         (delim == QSE_T('~')))
 				{
 					if (cmd->a1.type != QSE_SED_ADR_LINE || 
 					    cmd->a2.type != QSE_SED_ADR_LINE)

qse/test/sed/sed01.c

@@ -138,6 +138,22 @@ static int handle_args (int argc, qse_char_t* argv[])
 				print_usage (QSE_STDERR, argc, argv);
 				return -1;
 
+			case QSE_T('?'):
+				qse_fprintf (QSE_STDERR, 
+					QSE_T("ERROR: bad option - %c\n"),
+					opt.opt
+				);
+				print_usage (QSE_STDERR, argc, argv);
+				return -1;
+
+			case QSE_T(':'):
+				qse_fprintf (QSE_STDERR, 
+					QSE_T("ERROR: bad parameter for %c\n"),
+					opt.opt
+				);
+				print_usage (QSE_STDERR, argc, argv);
+				return -1;
+
 			case QSE_T('h'):
 				print_usage (QSE_STDOUT, argc, argv);
 				return 0;
@@ -156,12 +172,13 @@ static int handle_args (int argc, qse_char_t* argv[])
 	if (opt.ind < argc) g_script = argv[opt.ind++];
 	if (opt.ind < argc) g_infile = argv[opt.ind++];
 
-	if (g_script == QSE_NULL)
+	if (g_script == QSE_NULL || opt.ind < argc)
 	{
 		print_usage (QSE_STDERR, argc, argv);
 		return -1;
 	}
 
+
 	return 1;
 }