/*************************************************************************/

 /* bookloupe--check for assorted weirdnesses in a PG candidate text file */

 /*									 */

 /* Copyright 2000-2005 Jim Tinsley <jtinsley@pobox.com>			 */

 /* Copyright 2012- J. Ali Harlow <ali@juiblex.co.uk>			 */

 /*									 */

 /* This program is free software; you can redistribute it and/or modify  */

 /* it under the terms of the GNU General Public License as published by  */

 /* the Free Software Foundation; either version 2 of the License, or     */

 /* (at your option) any later version.					 */

 /*									 */

 /* This program is distributed in the hope that it will be useful,       */

 /* but WITHOUT ANY WARRANTY; without even the implied warranty of	 */

 /* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the		 */

 /* GNU General Public License for more details.				 */

 /*									 */

 /* You should have received a copy of the GNU General Public License	 */

 /* along with this program. If not, see <http://www.gnu.org/licenses/>.	 */

 /*************************************************************************/

 #include <stdio.h>

 #include <stdlib.h>

 #include <string.h>

 #include <ctype.h>

 #ifdef __WIN32__

 #include <windows.h>

 #endif

 #include <glib.h>

 #include <bl/bl.h>

 #include "bookloupe.h"

 #include "counters.h"

 #include "pending.h"

 #include "HTMLentities.h"

 gchar *charset;		/* Or NULL for auto (ISO_8859-1/ASCII or UNICODE) */

 GIConv charset_validator=(GIConv)-1;

 gchar *prevline;

 /* Common typos. */

 char *typo[] = {

     "teh", "th", "og", "fi", "ro", "adn", "yuo", "ot", "fo", "thet", "ane",

     "nad", "te", "ig", "acn",  "ahve", "alot", "anbd", "andt", "awya", "aywa",

     "bakc", "om", "btu", "byt", "cna", "cxan", "coudl", "dont", "didnt",

     "couldnt", "wouldnt", "doesnt", "shouldnt", "doign", "ehr", "hmi", "hse",

     "esle", "eyt", "fitrs", "firts", "foudn", "frmo", "fromt", "fwe", "gaurd",

     "gerat", "goign", "gruop", "haev", "hda", "hearign", "seeign", "sayign",

     "herat", "hge", "hsa", "hsi", "hte", "htere", "htese", "htey", "htis",

     "hvae", "hwich", "idae", "ihs", "iits", "int", "iwll", "iwth", "jsut",

     "loev", "sefl", "myu", "nkow", "nver", "nwe", "nwo", "ocur", "ohter",

     "omre", "onyl", "otehr", "otu", "owrk", "owuld", "peice", "peices",

     "peolpe", "peopel", "perhasp", "perhpas", "pleasent", "poeple", "porblem",

     "porblems", "rwite", "saidt", "saidh", "saids", "seh", "smae", "smoe",

     "sohw", "stnad", "stopry", "stoyr", "stpo", "tahn", "taht", "tath",

     "tehy", "tghe", "tghis", "theri", "theyll", "thgat", "thge", "thier",

     "thna", "thne", "thnig", "thnigs", "thsi", "thsoe", "thta", "timne",

     "tirne", "tkae", "tthe", "tyhat", "tyhe", "veyr", "vou", "vour", "vrey",

     "waht", "wasnt", "awtn", "watn", "wehn", "whic", "whcih", "whihc", "whta",

     "wihch", "wief", "wiht", "witha", "wiull", "wnat", "wnated", "wnats",

     "woh", "wohle", "wokr", "woudl", "wriet", "wrod", "wroet", "wroking",

     "wtih", "wuould", "wya", "yera", "yeras", "yersa", "yoiu", "youve",

     "ytou", "yuor", "abead", "ahle", "ahout", "ahove", "altbough", "balf",

     "bardly", "bas", "bave", "baving", "bebind", "beld", "belp", "belped",

     "ber", "bere", "bim", "bis", "bome", "bouse", "bowever", "buge",

     "dehates", "deht", "han", "hecause", "hecome", "heen", "hefore", "hegan",

     "hegin", "heing", "helieve", "henefit", "hetter", "hetween", "heyond",

     "hig", "higber", "huild", "huy", "hy", "jobn", "joh", "meanwbile",

     "memher", "memhers", "numher", "numhers", "perbaps", "prohlem", "puhlic",

     "witbout", "arn", "hin", "hirn", "wrok", "wroked", "amd", "aud",

     "prornise", "prornised", "modem", "bo", "heside", "chapteb", "chaptee",

     "se", ""

 };

 GTree *usertypo;

 /* Common abbreviations and other OK words not to query as typos. */

 char *okword[] = {

     "mr", "mrs", "mss", "mssrs", "ft", "pm", "st", "dr", "hmm", "h'm", "hmmm",

     "rd", "sh", "br", "pp", "hm", "cf", "jr", "sr", "vs", "lb", "lbs", "ltd",

     "pompeii","hawaii","hawaiian", "hotbed", "heartbeat", "heartbeats",

     "outbid", "outbids", "frostbite", "frostbitten", ""

 };

 /* Common abbreviations that cause otherwise unexplained periods. */

 char *abbrev[] = {

     "cent", "cents", "viz", "vol", "vols", "vid", "ed", "al", "etc", "op",

     "cit", "deg", "min", "chap", "oz", "mme", "mlle", "mssrs", ""

 };

 /*

  * Two-Letter combinations that rarely if ever start words,

  * but are common scannos or otherwise common letter combinations.

  */

 char *nostart[] = {

     "hr", "hl", "cb", "sb", "tb", "wb", "tl", "tn", "rn", "lt", "tj", ""

 };

 /*

  * Two-Letter combinations that rarely if ever end words,

  * but are common scannos or otherwise common letter combinations.

  */

 char *noend[] = {

     "cb", "gb", "pb", "sb", "tb", "wh", "fr", "br", "qu", "tw", "gl", "fl",

     "sw", "gr", "sl", "cl", "iy", ""

 };

 char *markup[] = {

     "a", "b", "big", "blockquote", "body", "br", "center", "col", "div", "em",

     "font", "h1", "h2", "h3", "h4", "h5", "h6", "head", "hr", "html", "i",

     "img", "li", "meta", "ol", "p", "pre", "small", "span", "strong", "sub",

     "sup", "table", "td", "tfoot", "thead", "title", "tr", "tt", "u", "ul", ""

 };

 char *DPmarkup[] = {

     "<sc>", "</sc>", "/*", "*/", "/#", "#/", "/$", "$/", "<tb>", ""

 };

 char *nocomma[] = {

     "the", "it's", "their", "an", "mrs", "a", "our", "that's", "its", "whose",

     "every", "i'll", "your", "my", "mr", "mrs", "mss", "mssrs", "ft", "pm",

     "st", "dr", "rd", "pp", "cf", "jr", "sr", "vs", "lb", "lbs", "ltd", "i'm",

     "during", "let", "toward", "among", ""

 };

 char *noperiod[] = {

     "every", "i'm", "during", "that's", "their", "your", "our", "my", "or",

     "and", "but", "as", "if", "the", "its", "it's", "until", "than", "whether",

     "i'll", "whose", "who", "because", "when", "let", "till", "very", "an",

     "among", "those", "into", "whom", "having", "thence", ""

 }; 

 gboolean pswit[SWITNO];  /* program switches */

 gchar *opt_charset;

 gboolean typo_compat,paranoid_compat;

 static GOptionEntry options[]={

     { "dp", 'd', 0, G_OPTION_ARG_NONE, pswit+DP_SWITCH,

       "Ignore DP-specific markup", NULL },

     { "no-dp", 0, G_OPTION_FLAG_HIDDEN|G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+DP_SWITCH,

       "Don't ignore DP-specific markup", NULL },

     { "echo", 0, G_OPTION_FLAG_HIDDEN, G_OPTION_ARG_NONE, pswit+ECHO_SWITCH,

       "Echo queried line", NULL },

     { "no-echo", 'e', G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+ECHO_SWITCH,

       "Don't echo queried line", NULL },

     { "squote", 's', 0, G_OPTION_ARG_NONE, pswit+SQUOTE_SWITCH,

       "Check single quotes", NULL },

     { "no-squote", 0, G_OPTION_FLAG_HIDDEN|G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+SQUOTE_SWITCH,

       "Don't check single quotes", NULL },

     { "typo", 0, 0, G_OPTION_ARG_NONE, pswit+TYPO_SWITCH,

       "Check common typos", NULL },

     { "no-typo", 0, G_OPTION_FLAG_HIDDEN|G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+TYPO_SWITCH,

       "Don't check common typos", NULL },

     { "qpara", 'p', 0, G_OPTION_ARG_NONE, pswit+QPARA_SWITCH,

       "Require closure of quotes on every paragraph", NULL },

     { "no-qpara", 0, G_OPTION_FLAG_HIDDEN|G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+QPARA_SWITCH,

       "Don't require closure of quotes on every paragraph", NULL },

     { "paranoid", 0, G_OPTION_FLAG_HIDDEN,

       G_OPTION_ARG_NONE, pswit+PARANOID_SWITCH,

       "Enable paranoid querying of everything", NULL },

     { "no-paranoid", 0, G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+PARANOID_SWITCH,

       "Disable paranoid querying of everything", NULL },

     { "line-end", 0, G_OPTION_FLAG_HIDDEN,

       G_OPTION_ARG_NONE, pswit+LINE_END_SWITCH,

       "Enable line end checking", NULL },

     { "no-line-end", 'l', G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+LINE_END_SWITCH,

       "Disable line end checking", NULL },

     { "overview", 'o', 0, G_OPTION_ARG_NONE, pswit+OVERVIEW_SWITCH,

       "Overview: just show counts", NULL },

     { "no-overview", 0, G_OPTION_FLAG_HIDDEN|G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+OVERVIEW_SWITCH,

       "Show individual warnings", NULL },

     { "stdout", 'y', 0, G_OPTION_ARG_NONE, pswit+STDOUT_SWITCH,

       "Output errors to stdout instead of stderr", NULL },

     { "no-stdout", 0, G_OPTION_FLAG_HIDDEN|G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+STDOUT_SWITCH,

       "Output errors to stderr instead of stdout", NULL },

     { "header", 'h', 0, G_OPTION_ARG_NONE, pswit+HEADER_SWITCH,

       "Echo header fields", NULL },

     { "no-header", 0, G_OPTION_FLAG_HIDDEN|G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+HEADER_SWITCH,

       "Don't echo header fields", NULL },

     { "markup", 'm', 0, G_OPTION_ARG_NONE, pswit+MARKUP_SWITCH,

       "Ignore markup in < >", NULL },

     { "no-markup", 0, G_OPTION_FLAG_HIDDEN|G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+MARKUP_SWITCH,

       "No special handling for markup in < >", NULL },

     { "usertypo", 'u', 0, G_OPTION_ARG_NONE, pswit+USERTYPO_SWITCH,

       "Use file of user-defined typos", NULL },

     { "no-usertypo", 0, G_OPTION_FLAG_HIDDEN|G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+USERTYPO_SWITCH,

       "Ignore file of user-defined typos", NULL },

     { "verbose", 'v', 0, G_OPTION_ARG_NONE, pswit+VERBOSE_SWITCH,

       "Verbose - list everything", NULL },

     { "no-verbose", 0, G_OPTION_FLAG_HIDDEN|G_OPTION_FLAG_REVERSE,

       G_OPTION_ARG_NONE, pswit+VERBOSE_SWITCH,

       "Switch off verbose mode", NULL },

     { "charset", 0, 0, G_OPTION_ARG_STRING, &opt_charset,

       "Set of characters valid for this ebook", "NAME" },

     { NULL }

 };

 /*

  * Options relating to configuration which make no sense from inside

  * a configuration file.

  */

 static GOptionEntry config_options[]={

     { "web", 'w', 0, G_OPTION_ARG_NONE, pswit+WEB_SWITCH,

       "Defaults for use on www upload", NULL },

     { "dump-config", 0, 0, G_OPTION_ARG_NONE, pswit+DUMP_CONFIG_SWITCH,

       "Dump current config settings", NULL },

     { NULL }

 };

 static GOptionEntry compatibility_options[]={

     { "toggle-typo", 't', 0, G_OPTION_ARG_NONE, &typo_compat,

       "Toggle checking for common typos", NULL },

     { "toggle-relaxed", 'x', 0, G_OPTION_ARG_NONE, &paranoid_compat,

       "Toggle both paranoid mode and common typos", NULL },

     { NULL }

 };

 long cnt_quote;		/* for overview mode, count of quote queries */

 long cnt_brack;		/* for overview mode, count of brackets queries */

 long cnt_bin;		/* for overview mode, count of non-ASCII queries */

 long cnt_odd;		/* for overview mode, count of odd character queries */

 long cnt_long;		/* for overview mode, count of long line errors */

 long cnt_short;		/* for overview mode, count of short line queries */

 long cnt_punct;		/* for overview mode,

 			   count of punctuation and spacing queries */

 long cnt_dash;		/* for overview mode, count of dash-related queries */

 long cnt_word;		/* for overview mode, count of word queries */

 long cnt_html;		/* for overview mode, count of html queries */

 long cnt_lineend;	/* for overview mode, count of line-end queries */

 long cnt_spacend;	/* count of lines with space at end */

 long linecnt;		/* count of total lines in the file */

 long checked_linecnt;	/* count of lines actually checked */

 void proghelp(GOptionContext *context);

 void procfile(const char *);

 gchar *running_from;

 gboolean mixdigit(const char *);

 gchar *getaword(const char *,const char **);

 char *flgets(char **,long,int);

 void postprocess_for_HTML(char *);

 char *linehasmarkup(char *);

 char *losemarkup(char *);

 gboolean tagcomp(const char *,const char *);

 void loseentities(char *);

 gboolean isroman(const char *);

 void postprocess_for_DP(char *);

 void print_as_windows_1252(const char *string);

 void print_as_utf_8(const char *string);

 GTree *qword,*qperiod;

 #ifdef __WIN32__

 UINT saved_cp;

 #endif

 gboolean set_charset(const char *name,GError **err)

 {

     /* The various UNICODE encodings all share the same character set. */

     const char *unicode_aliases[]={ "UCS-2", "UCS-2BE", "UCS-2LE", "UCS-4",

       "UCS-4BE", "UCS-4LE", "UCS2", "UCS4", "UNICODE", "UNICODEBIG",

       "UNICODELITTLE", "UTF-7", "UTF-8", "UTF-16", "UTF-16BE", "UTF-16LE",

       "UTF-32", "UTF-32BE", "UTF-32LE", "UTF7", "UTF8", "UTF16", "UTF16BE",

       "UTF16LE", "UTF32", "UTF32BE", "UTF32LE" };

     int i;

     if (charset)

 	g_free(charset);

     if (charset_validator!=(GIConv)-1)

 	g_iconv_close(charset_validator);

     if (!name || !g_strcasecmp(name,"auto"))

     {

 	charset=NULL;

 	charset_validator=(GIConv)-1;

 	return TRUE;

     }

     else

 	charset=g_strdup(name);

     for(i=0;i<G_N_ELEMENTS(unicode_aliases);i++)

 	if (!g_strcasecmp(charset,unicode_aliases[i]))

 	{

 	    g_free(charset);

 	    charset=g_strdup("UTF-8");

 	    break;

 	}

     if (!strcmp(charset,"UTF-8"))

 	charset_validator=(GIConv)-1;

     else

     {

 	charset_validator=g_iconv_open(charset,"UTF-8");

 	if (charset_validator==(GIConv)-1)

 	{

 	    g_set_error(err,G_CONVERT_ERROR,G_CONVERT_ERROR_NO_CONVERSION,

 	      "Unknown character set \"%s\"",charset);

 	    return FALSE;

 	}

     }

     return TRUE;

 }

 GKeyFile *config;

 void config_file_update(GKeyFile *kf)

 {

     int i;

     const char *s;

     gboolean sw;

     for(i=0;options[i].long_name;i++)

     {

 	if (g_str_has_prefix(options[i].long_name,"no-"))

 	    continue;

 	if (options[i].arg==G_OPTION_ARG_NONE)

 	{

 	    sw=*(gboolean *)options[i].arg_data;

 	    if (options[i].flags&G_OPTION_FLAG_REVERSE)

 		sw=!sw;

 	    g_key_file_set_boolean(kf,"options",options[i].long_name,sw);

 	}

 	else if (options[i].arg==G_OPTION_ARG_STRING)

 	{

 	    s=*(gchar **)options[i].arg_data;

 	    if (!s)

 		s="auto";

 	    g_key_file_set_string(kf,"options",options[i].long_name,s);

 	}

 	else

 	    g_assert_not_reached();

     }

 }

 void config_file_add_comments(GKeyFile *kf)

 {

     int i;

     gchar *comment;

     g_key_file_set_comment(kf,NULL,NULL," Default configuration for bookloupe",

       NULL);

     for(i=0;options[i].long_name;i++)

     {

 	if (g_str_has_prefix(options[i].long_name,"no-"))

 	    continue;

 	comment=g_strconcat(" ",options[i].description,NULL);

 	g_key_file_set_comment(kf,"options",options[i].long_name,comment,NULL);

 	g_free(comment);

     }

 }

 void dump_config(void)

 {

     gchar *s;

     if (config)

 	config_file_update(config);

     else

     {

 	config=g_key_file_new();

 	config_file_update(config);

 	config_file_add_comments(config);

     }

     s=g_key_file_to_data(config,NULL,NULL);

     if (s)

 	g_print("%s",s);

     g_free(s);

 }

 GKeyFile *read_config_file(gchar **full_path)

 {

     int i;

     GError *err=NULL;

     gchar **search_dirs;

     gchar *path;

     const char *search_path;

     GKeyFile *kf;

     kf=g_key_file_new();

     search_path=g_getenv("BOOKLOUPE_CONFIG_PATH");

     if (search_path)

     {

 #ifdef __WIN32__

 	search_dirs=g_strsplit(search_path,";",0);

 #else

 	search_dirs=g_strsplit(search_path,":",0);

 #endif

     }

     else

     {

 	search_dirs=g_new(gchar *,4);

 	search_dirs[0]=g_get_current_dir();

 	search_dirs[1]=g_strdup(running_from);

 	search_dirs[2]=g_strdup(g_get_user_config_dir());

 	search_dirs[3]=NULL;

     }

     for(i=0;search_dirs[i];i++)

     {

 	path=g_build_filename(search_dirs[i],"bookloupe.ini",NULL);

 	if (g_key_file_load_from_file(kf,path,

 	  G_KEY_FILE_KEEP_COMMENTS|G_KEY_FILE_KEEP_TRANSLATIONS,&err))

 	    break;

 	if (!g_error_matches(err,G_FILE_ERROR,G_FILE_ERROR_NOENT))

 	{

 	    g_printerr("Bookloupe: Error reading %s\n",path);

 	    g_printerr("%s\n",err->message);

 	    exit(1);

 	}

 	g_clear_error(&err);

 	g_free(path);

 	path=NULL;

     }

     if (!search_dirs[i])

     {

 	g_key_file_free(kf);

 	kf=NULL;

     }

     g_strfreev(search_dirs);

     if (full_path && kf)

 	*full_path=path;

     else

 	g_free(path);

     return kf;

 }

 void parse_config_file(void)

 {

     int i,j;

     gchar *path,*s;

     gchar **keys;

     gboolean sw;

     GError *err=NULL;

     config=read_config_file(&path);

     if (config)

 	keys=g_key_file_get_keys(config,"options",NULL,NULL);

     else

 	keys=NULL;

     if (keys)

     {

 	for(i=0;keys[i];i++)

 	{

 	    for(j=0;options[j].long_name;j++)

 	    {

 		if (g_str_has_prefix(options[j].long_name,"no-"))

 		    continue;

 		else if (!strcmp(keys[i],options[j].long_name))

 		{

 		    if (options[j].arg==G_OPTION_ARG_NONE)

 		    {

 			sw=g_key_file_get_boolean(config,"options",keys[i],

 			  &err);

 			if (err)

 			{

 			    g_printerr("Bookloupe: %s: options.%s: %s\n",

 			      path,keys[i],err->message);

 			    g_clear_error(&err);

 			}

 			else

 			{

 			    if (options[j].flags&G_OPTION_FLAG_REVERSE)

 				sw=!sw;

 			    *(gboolean *)options[j].arg_data=sw;

 			}

 			break;

 		    }

 		    else if (options[j].arg==G_OPTION_ARG_STRING)

 		    {

 			s=g_key_file_get_string(config,"options",keys[i],

 			  &err);

 			if (err)

 			{

 			    g_printerr("Bookloupe: %s: options.%s: %s\n",

 			      path,keys[i],err->message);

 			    g_clear_error(&err);

 			}

 			else

 			{

 			    g_free(*(gchar **)options[j].arg_data);

 			    if (!g_strcmp0(s,"auto"))

 			    {

 				*(gchar **)options[j].arg_data=NULL;

 				g_free(s);

 			    }

 			    else

 				*(gchar **)options[j].arg_data=s;

 			}

 			break;

 		    }

 		    else

 			g_assert_not_reached();

 		}

 	    }

 	    if (!options[j].long_name)

 		g_printerr("Bookloupe: %s: Unknown option \"%s\" ignored\n",

 		  path,keys[i]);

 	}

 	g_strfreev(keys);

     }

     if (config)

 	g_free(path);

 }

 void parse_options(int *argc,char ***argv)

 {

     GError *err=NULL;

     GOptionContext *context;

     GOptionGroup *compatibility;

     context=g_option_context_new(

       "file - look for errors in Project Gutenberg(TM) etexts");

     g_option_context_add_main_entries(context,options,NULL);

     g_option_context_add_main_entries(context,config_options,NULL);

     compatibility=g_option_group_new("compatibility",

       "Options for Compatibility with Gutcheck:",

       "Show compatibility options",NULL,NULL);

     g_option_group_add_entries(compatibility,compatibility_options);

     g_option_context_add_group(context,compatibility);

     g_option_context_set_description(context,

       "For simplicity, only the switch options which reverse the\n"

       "default configuration are listed. In most cases, both vanilla\n"

       "and \"no-\" prefixed versions are available for use.");

     if (!g_option_context_parse(context,argc,argv,&err))

     {

 	g_printerr("Bookloupe: %s\n",err->message);

 	g_printerr("Use \"%s --help\" for help\n",(*argv)[0]);

 	exit(1);

     }

     if (typo_compat)

 	pswit[TYPO_SWITCH]=!pswit[TYPO_SWITCH];

     if (paranoid_compat)

     {

 	pswit[PARANOID_SWITCH]=!pswit[PARANOID_SWITCH];

 	pswit[TYPO_SWITCH]=!pswit[TYPO_SWITCH];

     }

     /*

      * Web uploads - for the moment, this is really just a placeholder

      * until we decide what processing we really want to do on web uploads

      */

     if (pswit[WEB_SWITCH])

     {

 	/* specific override for web uploads */

 	pswit[ECHO_SWITCH]=TRUE;

 	pswit[SQUOTE_SWITCH]=FALSE;

 	pswit[TYPO_SWITCH]=TRUE;

 	pswit[QPARA_SWITCH]=FALSE;

 	pswit[PARANOID_SWITCH]=TRUE;

 	pswit[LINE_END_SWITCH]=FALSE;

 	pswit[OVERVIEW_SWITCH]=FALSE;

 	pswit[STDOUT_SWITCH]=FALSE;

 	pswit[HEADER_SWITCH]=TRUE;

 	pswit[VERBOSE_SWITCH]=FALSE;

 	pswit[MARKUP_SWITCH]=FALSE;

 	pswit[USERTYPO_SWITCH]=FALSE;

 	pswit[DP_SWITCH]=FALSE;

     }

     if (opt_charset && !set_charset(opt_charset,&err))

     {

 	g_printerr("%s\n",err->message);

 	exit(1);

     }

     if (pswit[DUMP_CONFIG_SWITCH])

     {

 	dump_config();

 	exit(0);

     }

     g_free(opt_charset);

     opt_charset=NULL;

     if (pswit[OVERVIEW_SWITCH])

 	/* just print summary; don't echo */

 	pswit[ECHO_SWITCH]=FALSE;

     if (*argc<2)

     {

 	proghelp(context);

 	exit(1);

     }

     g_option_context_free(context);

 }

 /*

  * read_user_scannos:

  *

  * Read in the user-defined stealth scanno list.

  */

 void read_user_scannos(void)

 {

     GError *err=NULL;

     gchar *usertypo_file;

     gboolean okay;

     int i;

     gsize len,nb;

     gchar *contents,*utf8,**lines;

     usertypo_file=g_strdup("bookloupe.typ");

     okay=file_get_contents_text(usertypo_file,&contents,&len,&err);

     if (g_error_matches(err,G_FILE_ERROR,G_FILE_ERROR_NOENT))

     {

 	g_clear_error(&err);

 	g_free(usertypo_file);

 	usertypo_file=g_build_filename(running_from,"bookloupe.typ",NULL);

 	okay=file_get_contents_text(usertypo_file,&contents,&len,&err);

     }

     if (g_error_matches(err,G_FILE_ERROR,G_FILE_ERROR_NOENT))

     {

 	g_clear_error(&err);

 	g_free(usertypo_file);

 	usertypo_file=g_strdup("gutcheck.typ");

 	okay=file_get_contents_text(usertypo_file,&contents,&len,&err);

     }

     if (g_error_matches(err,G_FILE_ERROR,G_FILE_ERROR_NOENT))

     {

 	g_clear_error(&err);

 	g_free(usertypo_file);

 	usertypo_file=g_build_filename(running_from,"gutcheck.typ",NULL);

 	okay=file_get_contents_text(usertypo_file,&contents,&len,&err);

     }

     if (g_error_matches(err,G_FILE_ERROR,G_FILE_ERROR_NOENT))

     {

 	g_free(usertypo_file);

 	g_print("   --> I couldn't find bookloupe.typ "

 	  "-- proceeding without user typos.\n");

 	return;

     }

     else if (!okay)

     {

 	fprintf(stderr,"%s: %s\n",usertypo_file,err->message);

 	g_free(usertypo_file);

 	g_clear_error(&err);

 	exit(1);

     }

     if (g_utf8_validate(contents,len,NULL))

     {

 	utf8=g_utf8_normalize(contents,len,G_NORMALIZE_DEFAULT_COMPOSE);

 	if (!charset)

 	    (void)set_charset("UNICODE",NULL);

     }

     else

 	utf8=g_convert(contents,len,"UTF-8","WINDOWS-1252",NULL,&nb,NULL);

     g_free(contents);

     lines=g_strsplit_set(utf8,"\r\n",0);

     g_free(utf8);

     usertypo=g_tree_new_full((GCompareDataFunc)strcmp,NULL,g_free,NULL);

     for (i=0;lines[i];i++)

 	if (*(unsigned char *)lines[i]>'!')

 	    g_tree_insert(usertypo,lines[i],GINT_TO_POINTER(1));

 	else

 	    g_free(lines[i]);

     g_free(lines);

 }

 /*

  * read_etext:

  *

  * Read an etext returning a newly allocated string containing the file

  * contents or NULL on error.

  */

 gchar *read_etext(const char *filename,GError **err)

 {

     GError *tmp_err=NULL;

     gchar *contents,*utf8;

     gsize len,bytes_read,bytes_written;

     int i,line,col;

     if (!g_file_get_contents(filename,&contents,&len,err))

 	return NULL;

     if (g_utf8_validate(contents,len,NULL))

     {

 	utf8=g_utf8_normalize(contents,len,G_NORMALIZE_DEFAULT_COMPOSE);

 	g_set_print_handler(print_as_utf_8);

 #ifdef __WIN32__

 	SetConsoleOutputCP(CP_UTF8);

 #endif

     }

     else

     {

 	utf8=g_convert(contents,len,"UTF-8","WINDOWS-1252",&bytes_read,

 	  &bytes_written,&tmp_err);

 	if (g_error_matches(tmp_err,G_CONVERT_ERROR,

 	  G_CONVERT_ERROR_ILLEGAL_SEQUENCE))

 	{

 	    line=col=1;

 	    for(i=0;i<bytes_read;i++)

 		if (contents[i]=='\n')

 		{

 		    line++;

 		    col=1;

 		}

 		else if (contents[i]!='\r')

 		    col++;

 	    g_set_error(err,G_CONVERT_ERROR,G_CONVERT_ERROR_ILLEGAL_SEQUENCE,

 	      "Input conversion failed. Byte %d at line %d, column %d is not a "

 	      "valid Windows-1252 character",

 	      ((unsigned char *)contents)[bytes_read],line,col);

 	}

 	else if (tmp_err)

 	    g_propagate_error(err,tmp_err);

 	g_set_print_handler(print_as_windows_1252);

 #ifdef __WIN32__

 	SetConsoleOutputCP(1252);

 #endif

     }

     g_free(contents);

     return utf8;

 }

 void cleanup_on_exit(void)

 {

 #ifdef __WIN32__

     SetConsoleOutputCP(saved_cp);

 #endif

 }

 int main(int argc,char **argv)

 {

 #ifdef __WIN32__

     atexit(cleanup_on_exit);

     saved_cp=GetConsoleOutputCP();

 #endif

     running_from=g_path_get_dirname(argv[0]);

     /* Paranoid checking is turned OFF, not on, by its switch */

     pswit[PARANOID_SWITCH]=TRUE;

     /* if running in paranoid mode, typo checks default to enabled */

     pswit[TYPO_SWITCH]=TRUE;

     /* Line-end checking is turned OFF, not on, by its switch */

     pswit[LINE_END_SWITCH]=TRUE;

     /* Echoing is turned OFF, not on, by its switch */

     pswit[ECHO_SWITCH]=TRUE;

     parse_config_file();

     parse_options(&argc,&argv);

     if (pswit[USERTYPO_SWITCH])

 	read_user_scannos();

     fprintf(stderr,"bookloupe: Check and report on an e-text\n");

     procfile(argv[1]);

     if (pswit[OVERVIEW_SWITCH])

     {

 	g_print("    Checked %ld lines of %ld (head+foot = %ld)\n\n",

 	  checked_linecnt,linecnt,linecnt-checked_linecnt);

 	g_print("    --------------- Queries found --------------\n");

 	if (cnt_long)

 	    g_print("    Long lines:		    %14ld\n",cnt_long);

 	if (cnt_short)

 	    g_print("    Short lines:		   %14ld\n",cnt_short);

 	if (cnt_lineend)

 	    g_print("    Line-end problems:	     %14ld\n",cnt_lineend);

 	if (cnt_word)

 	    g_print("    Common typos:		  %14ld\n",cnt_word);

 	if (cnt_quote)

 	    g_print("    Unmatched quotes:	      %14ld\n",cnt_quote);

 	if (cnt_brack)

 	    g_print("    Unmatched brackets:	    %14ld\n",cnt_brack);

 	if (cnt_bin)

 	    g_print("    Non-ASCII characters:	  %14ld\n",cnt_bin);

 	if (cnt_odd)

 	    g_print("    Proofing characters:	   %14ld\n",cnt_odd);

 	if (cnt_punct)

 	    g_print("    Punctuation & spacing queries: %14ld\n",cnt_punct);

 	if (cnt_dash)

 	    g_print("    Non-standard dashes:	   %14ld\n",cnt_dash);

 	if (cnt_html)

 	    g_print("    Possible HTML tags:	    %14ld\n",cnt_html);

 	g_print("\n");

 	g_print("    TOTAL QUERIES		  %14ld\n",

 	  cnt_quote+cnt_brack+cnt_bin+cnt_odd+cnt_long+cnt_short+cnt_punct+

 	  cnt_dash+cnt_word+cnt_html+cnt_lineend);

     }

     g_free(running_from);

     if (usertypo)

 	g_tree_unref(usertypo);

     set_charset(NULL,NULL);

     if (config)

 	g_key_file_free(config);

     return 0;

 }

 void count_dashes(const char *line,const char *dash,

   struct dash_results *results)

 {

     int i;

     gchar **tokens;

     gunichar pc,nc;

     gboolean spaced=FALSE,unspaced=FALSE,spaced2=FALSE;

     if (!*line)

 	return;

     tokens=g_strsplit(line,dash,0);

     if (tokens[1])

 	results->base++;

     for(i=1;tokens[i];i++)

     {

 	pc=g_utf8_get_char(g_utf8_prev_char(tokens[i-1]+strlen(tokens[i-1])));

 	nc=g_utf8_get_char(tokens[i]);

 	if (g_unichar_isspace(pc) || g_unichar_isspace(nc))

 	    spaced=TRUE;

 	if (g_unichar_isspace(pc) && g_unichar_isspace(nc))

 	    spaced2=TRUE;

 	else if (!g_unichar_isspace(pc) && !g_unichar_isspace(nc))

 	    unspaced=TRUE;

     }

     if (spaced)

 	results->space++;

     if (spaced2)

 	/* count of lines with em-dashes with spaces both sides */

 	results->non_PG_space++;

     if (unspaced)

 	/* count of lines with PG-type em-dashes with no spaces */

 	results->PG_space++;

     g_strfreev(tokens);

 }

 /*

  * first_pass:

  *

  * Run a first pass - verify that it's a valid PG

  * file, decide whether to report some things that

  * occur many times in the text like long or short

  * lines, non-standard dashes, etc.

  */

 struct first_pass_results *first_pass(const char *etext)

 {

     gunichar laststart=CHAR_SPACE;

     const char *s;

     gchar *lc_line;

     int i,j,lbytes,llen;

     gchar **lines;

     unsigned int lastlen=0,lastblen=0;

     long spline=0,nspline=0;

     static struct first_pass_results results={0};

     struct dash_results tmp_dash_results;

     gchar *inword;

     QuoteClass qc;

     lines=g_strsplit(etext,"\n",0);

     if (!lines[0])

     {

 	/* An empty etext has no terminators */

 	results.newlines=DOS_NEWLINES;

     }

     else if (!lines[1])

     {

 	/*

 	 * If there are no LFs, we don't have UNIX-style

 	 * terminators, but we might have OS9-style ones.

 	 */

 	results.newlines=OS9_NEWLINES;

 	g_strfreev(lines);

 	lines=g_strsplit(etext,"\r",0);

 	if (!lines[0] || !lines[1])

 	    /* Looks like we don't have any terminators at all */

 	    results.newlines=DOS_NEWLINES;

     }

     else

     {

 	/* We might have UNIX-style terminators */

 	results.newlines=UNIX_NEWLINES;

     }

     for (j=0;lines[j];j++)

     {

 	lbytes=strlen(lines[j]);

 	if (lbytes>0 && lines[j][lbytes-1]=='\r')

 	{

 	    results.newlines=DOS_NEWLINES;

 	    do

 	    {

 		lines[j][--lbytes]='\0';

 	    } while (lbytes>0 && lines[j][lbytes-1]=='\r');

 	}

 	llen=g_utf8_strlen(lines[j],lbytes);

 	linecnt++;

 	if (strstr(lines[j],"*END") && strstr(lines[j],"SMALL PRINT") &&

 	  (strstr(lines[j],"PUBLIC DOMAIN") || strstr(lines[j],"COPYRIGHT")))

 	{

 	    if (spline)

 		g_print("   --> Duplicate header?\n");

 	    spline=linecnt+1;   /* first line of non-header text, that is */

 	}

 	if (!strncmp(lines[j],"*** START",9) &&

 	  strstr(lines[j],"PROJECT GUTENBERG"))

 	{

 	    if (nspline)

 		g_print("   --> Duplicate header?\n");

 	    nspline=linecnt+1;   /* first line of non-header text, that is */

 	}

 	if (spline || nspline)

 	{

 	    lc_line=g_utf8_strdown(lines[j],lbytes);

 	    if (strstr(lc_line,"end") && strstr(lc_line,"project gutenberg"))

 	    {

 		if (strstr(lc_line,"end")<strstr(lc_line,"project gutenberg"))

 		{

 		    if (results.footerline)

 		    {

 			/* it's an old-form header - we can detect duplicates */

 			if (!nspline)

 			    g_print("   --> Duplicate footer?\n");

 		    }

 		    else

 			results.footerline=linecnt;

 		}

 	    }

 	    g_free(lc_line);

 	}

 	if (spline)

 	    results.firstline=spline;

 	if (nspline)

 	    results.firstline=nspline;  /* override with new */

 	if (results.footerline)

 	    continue;    /* don't count the boilerplate in the footer */

 	results.totlen+=llen;

 	for (s=lines[j];*s;s=g_utf8_next_char(s))

 	{

 	    if (g_utf8_get_char(s)>127)

 		results.binlen++;

 	    if (g_unichar_isalpha(g_utf8_get_char(s)))

 		results.alphalen++;

 	    if (s>lines[j])

 	    {

 		if (CHAR_IS_DQUOTE(g_utf8_get_char(s)))

 		    qc=QUOTE_CLASS(g_utf8_get_char(s));

 		else

 		    qc=INVALID_QUOTE;

 		if ((qc==CLOSING_QUOTE || qc==NEUTRAL_QUOTE) &&

 		  g_unichar_isalpha(g_utf8_get_char(g_utf8_prev_char(s))))

 		    results.endquote_count++;

 	    }

 	}

 	if (llen>2 && lastlen>2 && lastlen<SHORTEST_PG_LINE && lastblen>2 &&

 	  lastblen>SHORTEST_PG_LINE && laststart!=CHAR_SPACE)

 	    results.shortline++;

 	if (lbytes>0 &&

 	  g_utf8_get_char(g_utf8_prev_char(lines[j]+lbytes))<=CHAR_SPACE)

 	    cnt_spacend++;

 	if (strstr(lines[j],".,"))

 	    results.dotcomma++;

 	/* only count ast lines for ignoring purposes where there is */

 	/* locase text on the line */

 	if (strchr(lines[j],'*'))

 	{

 	    for (s=lines[j];*s;s=g_utf8_next_char(s))

 		if (g_unichar_islower(g_utf8_get_char(s)))

 		    break;

 	    if (*s)

 		results.astline++;

 	}

 	if (strchr(lines[j],'/'))

 	    results.fslashline++;

 	if (lbytes>0)

 	{

 	    for (s=g_utf8_prev_char(lines[j]+lbytes);

 	      s>lines[j] && g_utf8_get_char(s)<=CHAR_SPACE;

 	      s=g_utf8_prev_char(s))

 		;

 	    if (s>g_utf8_next_char(lines[j]) && g_utf8_get_char(s)=='-' &&

 	      g_utf8_get_char(g_utf8_prev_char(s))!='-')

 		results.hyphens++;

 	}

 	if (llen>LONGEST_PG_LINE)

 	    results.longline++;

 	if (llen>WAY_TOO_LONG)

 	    results.verylongline++;

 	if (strchr(lines[j],'<') && strchr(lines[j],'>'))

 	{

 	    i=(int)(strchr(lines[j],'>')-strchr(lines[j],'<')+1);

 	    if (i>0)

 		results.htmcount++;

 	    if (strstr(lines[j],"<i>"))

 		results.htmcount+=4; /* bonus marks! */

 	}

 	/* Check for spaced em-dashes */

 	memset(&tmp_dash_results,0,sizeof(tmp_dash_results));

 	count_dashes(lines[j],"--",&tmp_dash_results);

 	count_dashes(lines[j],"—",&tmp_dash_results);

 	if (tmp_dash_results.base)

 	    results.emdash.base++;

 	if (tmp_dash_results.non_PG_space)

 	    results.emdash.non_PG_space++;

 	if (tmp_dash_results.PG_space)

 	    results.emdash.PG_space++;

 	for (s=lines[j];*s;)

 	{

 	    inword=getaword(NULL,&s);

 	    if (!strcmp(inword,"hij") || !strcmp(inword,"niet")) 

 		results.Dutchcount++;

 	    if (!strcmp(inword,"dans") || !strcmp(inword,"avec")) 

 		results.Frenchcount++;

 	    if (!strcmp(inword,"0") || !strcmp(inword,"1")) 

 		results.standalone_digit++;

 	    g_free(inword);

 	}

 	/* Check for spaced dashes */

 	if (strstr(lines[j]," -") && *(strstr(lines[j]," -")+2)!='-')

 	    results.spacedash++;

 	lastblen=lastlen;

 	lastlen=llen;

 	laststart=lines[j][0];

     }

     g_strfreev(lines);

     return &results;

 }

 /*

  * report_first_pass:

  *

  * Make some snap decisions based on the first pass results.

  */

 struct warnings *report_first_pass(struct first_pass_results *results)

 {

     static struct warnings warnings={0};

     warnings.newlines=results->newlines;

     if (warnings.newlines==UNIX_NEWLINES)

 	g_print("   --> No lines in this file have a CR. Not reporting them. "

 	  "Project Gutenberg requires that all lineends be CR-LF.\n");

     else if (warnings.newlines==OS9_NEWLINES)

 	g_print("   --> No lines in this file have a LF. Not reporting them. "

 	  "Project Gutenberg requires that all lineends be CR-LF.\n");

     if (cnt_spacend>0)

 	g_print("   --> %ld lines in this file have white space at end\n",

 	  cnt_spacend);

     warnings.dotcomma=1;

     if (results->dotcomma>5)

     {

 	warnings.dotcomma=0;

 	g_print("   --> %ld lines in this file contain '.,'. "

 	  "Not reporting them.\n",results->dotcomma);

     }

     /*

      * If more than 50 lines, or one-tenth, are short,

      * don't bother reporting them.

      */

     warnings.shortline=1;

     if (results->shortline>50 || results->shortline*10>linecnt)

     {

 	warnings.shortline=0;

 	g_print("   --> %ld lines in this file are short. "

 	  "Not reporting short lines.\n",results->shortline);

     }

     /*

      * If more than 50 lines, or one-tenth, are long,

      * don't bother reporting them.

      */

     warnings.longline=1;

     if (results->longline>50 || results->longline*10>linecnt)

     {

 	warnings.longline=0;

 	g_print("   --> %ld lines in this file are long. "

 	  "Not reporting long lines.\n",results->longline);

     }

     /* If more than 10 lines contain asterisks, don't bother reporting them. */

     warnings.ast=1;

     if (results->astline>10)

     {

 	warnings.ast=0;

 	g_print("   --> %ld lines in this file contain asterisks. "

 	  "Not reporting them.\n",results->astline);

     }

     /*

      * If more than 10 lines contain forward slashes,

      * don't bother reporting them.

      */

     warnings.fslash=1;

     if (results->fslashline>10)

     {

 	warnings.fslash=0;

 	g_print("   --> %ld lines in this file contain forward slashes. "

 	  "Not reporting them.\n",results->fslashline);

     }

     /*

      * If more than 20 lines contain unpunctuated endquotes,

      * don't bother reporting them.

      */

     warnings.endquote=1;

     if (results->endquote_count>20)

     {

 	warnings.endquote=0;

 	g_print("   --> %ld lines in this file contain unpunctuated endquotes. "

 	  "Not reporting them.\n",results->endquote_count);

     }

     /*

      * If more than 15 lines contain standalone digits,

      * don't bother reporting them.

      */

     warnings.digit=1;

     if (results->standalone_digit>10)

     {

 	warnings.digit=0;

 	g_print("   --> %ld lines in this file contain standalone 0s and 1s. "

 	  "Not reporting them.\n",results->standalone_digit);

     }

     /*

      * If more than 20 lines contain hyphens at end,

      * don't bother reporting them.

      */

     warnings.hyphen=1;

     if (results->hyphens>20)

     {

 	warnings.hyphen=0;

 	g_print("   --> %ld lines in this file have hyphens at end. "

 	  "Not reporting them.\n",results->hyphens);

     }

     if (results->htmcount>20 && !pswit[MARKUP_SWITCH])

     {

 	g_print("   --> Looks like this is HTML. Switching HTML mode ON.\n");

 	pswit[MARKUP_SWITCH]=1;

     }

     if (results->verylongline>0)

 	g_print("   --> %ld lines in this file are VERY long!\n",

 	  results->verylongline);

     /*

      * If there are more non-PG spaced dashes than PG em-dashes,

      * assume it's deliberate.

      * Current PG guidelines say don't use them, but older texts do,

      * and some people insist on them whatever the guidelines say.

      */

     warnings.dash=1;

     if (results->spacedash+results->emdash.non_PG_space>

       results->emdash.PG_space)

     {

 	warnings.dash=0;

 	g_print("   --> There are %ld spaced dashes and em-dashes. "

 	  "Not reporting them.\n",

 	  results->spacedash+results->emdash.non_PG_space);

     }

     if (charset)

 	warnings.bin=0;

     else

     {

 	/* Charset ISO_8859-1/ASCII checks for compatibility with gutcheck */

 	warnings.bin=1;

 	/* If more than a quarter of characters are hi-bit, bug out. */

 	if (results->binlen*4>results->totlen)

 	{

 	    g_print("   --> This file does not appear to be ASCII. "

 	      "Terminating. Best of luck with it!\n");

 	    exit(1);

 	}

 	if (results->alphalen*4<results->totlen)

 	{

 	    g_print("   --> This file does not appear to be text. "

 	      "Terminating. Best of luck with it!\n");

 	    exit(1);

 	}

 	if (results->binlen*100>results->totlen || results->binlen>100)

 	{

 	    g_print("   --> There are a lot of foreign letters here. "

 	      "Not reporting them.\n");

 	    if (!pswit[VERBOSE_SWITCH])

 		warnings.bin=0;

 	}

     }

     warnings.isDutch=FALSE;

     if (results->Dutchcount>50)

     {

 	warnings.isDutch=TRUE;

 	g_print("   --> This looks like Dutch - "

 	  "switching off dashes and warnings for 's Middags case.\n");

     }

     warnings.isFrench=FALSE;

     if (results->Frenchcount>50)

     {

 	warnings.isFrench=TRUE;

 	g_print("   --> This looks like French - "

 	  "switching off some doublepunct.\n");

     }

     if (results->firstline && results->footerline)

 	g_print("    The PG header and footer appear to be already on.\n");

     else

     {

 	if (results->firstline)

 	    g_print("    The PG header is on - no footer.\n");

 	if (results->footerline)

 	    g_print("    The PG footer is on - no header.\n");

     }

     g_print("\n");

     if (pswit[VERBOSE_SWITCH])

     {

 	warnings.shortline=1;

 	warnings.dotcomma=1;

 	warnings.longline=1;

 	warnings.dash=1;

 	warnings.digit=1;

 	warnings.ast=1;

 	warnings.fslash=1;

 	warnings.hyphen=1;

 	warnings.endquote=1;

 	g_print("   *** Verbose output is ON -- you asked for it! ***\n");

     }

     if (warnings.isDutch)

 	warnings.dash=0;

     if (results->footerline>0 && results->firstline>0 &&

       results->footerline>results->firstline &&

       results->footerline-results->firstline<100)

     {

 	g_print("   --> I don't really know where this text starts. \n");

 	g_print("       There are no reference points.\n");

 	g_print("       I'm going to have to report the header and footer "

 	  "as well.\n");

 	results->firstline=0;

     }

     return &warnings;

 }

 /*

  * analyse_quotes:

  *

  * Look along the line, accumulate the count of quotes, and see

  * if this is an empty line - i.e. a line with nothing on it

  * but spaces.

  * If line has just spaces, period, * and/or - on it, don't

  * count it, since empty lines with asterisks or dashes to

  * separate sections are common.

  *

  * Returns: TRUE if the line is empty.

  */

 gboolean analyse_quotes(const char *aline,struct counters *counters)

 {

     int guessquote=0;

     /* assume the line is empty until proven otherwise */

     gboolean isemptyline=TRUE;

     const char *s=aline,*sprev,*snext;

     gunichar c;

     sprev=NULL;

     GError *tmp_err=NULL;

     while (*s)

     {

 	snext=g_utf8_next_char(s);

 	c=g_utf8_get_char(s);

 	if (CHAR_IS_DQUOTE(c))

 	    (void)count_quote(counters,c,QUOTE_CLASS(c),&tmp_err);

 	else if (CHAR_IS_SQUOTE(c) && pswit[SQUOTE_SWITCH])

 	{

 	    if (s==aline)

 	    {

 		/*

 		 * At start of line, it can only be a quotation mark.

 		 * Hardcode a very common exception!

 		 */

 		if (!g_str_has_prefix(snext,"tis") &&

 		  !g_str_has_prefix(snext,"Tis"))

 		    (void)count_quote(counters,c,NEUTRAL_QUOTE,&tmp_err);

 	    }

 	    else if (g_unichar_isalpha(g_utf8_get_char(sprev)) &&

 	      g_unichar_isalpha(g_utf8_get_char(snext)))

 		/* Do nothing! it's definitely an apostrophe, not a quote */

 		;

 	    /* it's outside a word - let's check it out */

 	    else if (c==CHAR_OPEN_SQUOTE || c==CHAR_LS_QUOTE ||

 	      g_unichar_isalpha(g_utf8_get_char(snext)))

 	    {

 		/* certainly looks like a quotation mark */

 		if (!g_str_has_prefix(snext,"tis") &&

 		  !g_str_has_prefix(snext,"Tis"))

 		    /* hardcode a very common exception! */

 		{

 		    if (strchr(".?!,;:",g_utf8_get_char(sprev)))

 			(void)count_quote(counters,c,NEUTRAL_QUOTE,&tmp_err);

 		    else

 			(void)count_quote(counters,c,OPENING_QUOTE,&tmp_err);

 		}

 	    }

 	    else

 	    {

 		/* now - is it a quotation mark? */

 		guessquote=0;   /* accumulate clues */

 		if (g_unichar_isalpha(g_utf8_get_char(sprev)))

 		{

 		    /* it follows a letter - could be either */

 		    guessquote++;

 		    if (g_utf8_get_char(sprev)=='s')

 		    {

 			/* looks like a plural apostrophe */

 			guessquote-=3;

 			if (g_utf8_get_char(snext)==CHAR_SPACE)

 			    /* bonus marks! */

 			    guessquote-=2;

 		    }

 		    if (innermost_quote_matches(counters,c))

 			/*

 			 * Give it the benefit of some doubt,

 			 * if a squote is already open.

 			 */

 			guessquote++;

 		    else

 			guessquote--;

 		    if (guessquote>=0)

 			(void)count_quote(counters,c,CLOSING_QUOTE,&tmp_err);

 		}

 		else

 		    /* no adjacent letter - it must be a quote of some kind */

 		    (void)count_quote(counters,c,NEUTRAL_QUOTE,&tmp_err);

 	    }

 	}

 	if (tmp_err)

 	{

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column %ld - %s\n",

 		  linecnt,g_utf8_pointer_to_offset(aline,s)+1,tmp_err->message);

 	    g_clear_error(&tmp_err);

 	}

 	if (c!=CHAR_SPACE && c!='-' && c!='.' && c!=CHAR_ASTERISK &&

 	  c!='\r' && c!='\n')

 	    isemptyline=FALSE;  /* ignore lines like  *  *  *  as spacers */

 	if (c==CHAR_UNDERSCORE)

 	    counters->c_unders++;

 	if (c==CHAR_OPEN_SBRACK)

 	{

 	    if (!matching_difference(counters,COUNTER_ILLUSTRATION) &&

 	      !matching_difference(counters,c) && s==aline &&

 	      g_str_has_prefix(s,"[Illustration:"))

 		increment_matching(counters,COUNTER_ILLUSTRATION,TRUE);

 	    else

 		increment_matching(counters,c,TRUE);

 	}

 	else if (c==CHAR_OPEN_CBRACK || c==CHAR_OPEN_RBRACK)

 	    increment_matching(counters,c,TRUE);

 	if (c==CHAR_CLOSE_SBRACK)

 	{

 	    if (!matching_count(counters,COUNTER_ILLUSTRATION,FALSE) &&

 	      !matching_difference(counters,c) && !*snext)

 		increment_matching(counters,COUNTER_ILLUSTRATION,FALSE);

 	    else

 		increment_matching(counters,c,FALSE);

 	}

 	else if (c==CHAR_CLOSE_CBRACK || c==CHAR_CLOSE_RBRACK)

 	    increment_matching(counters,c,FALSE);

 	sprev=s;

 	s=snext;

     }

     return isemptyline;

 }

 /*

  * check_for_control_characters:

  *

  * Check for invalid or questionable characters in the line

  * Anything above 127 is invalid for plain ASCII, and

  * non-printable control characters should also be flagged.

  * Tabs should generally not be there.

  */

 void check_for_control_characters(const char *aline)

 {

     gunichar c;

     const char *s;

     for (s=aline;*s;s=g_utf8_next_char(s))

     {

 	c=g_utf8_get_char(s);

 	if (c<CHAR_SPACE && c!=CHAR_LF && c!=CHAR_CR && c!=CHAR_TAB)

 	{

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column %ld - Control character %u\n",

 		  linecnt,g_utf8_pointer_to_offset(s,aline)+1,c);

 	    else

 		cnt_bin++;

 	}

     }

 }

 /*

  * check_for_odd_characters:

  *

  * Check for binary and other odd characters.

  */

 void check_for_odd_characters(const char *aline,const struct warnings *warnings,

   gboolean isemptyline)

 {

     /* Don't repeat multiple warnings on one line. */

     gboolean eInvalidChar=FALSE,eTab=FALSE,eTilde=FALSE;

     gboolean eCarat=FALSE,eFSlash=FALSE,eAst=FALSE;

     const char *s;

     gunichar c;

     gsize nb;

     gchar *t;

     for (s=aline;*s;s=g_utf8_next_char(s))

     {

 	c=g_utf8_get_char(s);

 	if (warnings->bin && !eInvalidChar &&

 	  (c<CHAR_SPACE && c!='\t' && c!='\n' || c>127))

 	{

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		if (c>127 && c<160 || c>255)

 		    g_print("    Line %ld column %ld - "

 		      "Non-ISO-8859 character %u\n",

 		      linecnt,g_utf8_pointer_to_offset(aline,s)+1,c);

 		else

 		    g_print("    Line %ld column %ld - "

 		      "Non-ASCII character %u\n",

 		      linecnt,g_utf8_pointer_to_offset(aline,s)+1,c);

 	    else

 		cnt_bin++;

 	    eInvalidChar=TRUE;

 	}

 	if (!eInvalidChar && charset)

 	{

 	    if (charset_validator==(GIConv)-1)

 	    {

 		if (!g_unichar_isdefined(c))

 		{

 		    if (pswit[ECHO_SWITCH])

 			g_print("\n%s\n",aline);

 		    if (!pswit[OVERVIEW_SWITCH])

 			g_print("    Line %ld column %ld - Unassigned UNICODE "

 			  "code point U+%04" G_GINT32_MODIFIER "X\n",

 			  linecnt,g_utf8_pointer_to_offset(aline,s)+1,c);

 		    else

 			cnt_bin++;

 		    eInvalidChar=TRUE;

 		}

 		else if (c>=0xE000 && c<=0xF8FF || c>=0xF0000 && c<=0xFFFFD ||

 		  c>=100000 && c<=0x10FFFD)

 		{

 		    if (pswit[ECHO_SWITCH])

 			g_print("\n%s\n",aline);

 		    if (!pswit[OVERVIEW_SWITCH])

 			g_print("    Line %ld column %ld - Private Use "

 			  "character U+%04" G_GINT32_MODIFIER "X\n",

 			  linecnt,g_utf8_pointer_to_offset(aline,s)+1,c);

 		    else

 			cnt_bin++;

 		    eInvalidChar=TRUE;

 		}

 	    }

 	    else

 	    {

 		t=g_convert_with_iconv(s,g_utf8_next_char(s)-s,

 		  charset_validator,NULL,&nb,NULL);

 		if (t)

 		    g_free(t);

 		else

 		{

 		    if (pswit[ECHO_SWITCH])

 			g_print("\n%s\n",aline);

 		    if (!pswit[OVERVIEW_SWITCH])

 			g_print("    Line %ld column %ld - Non-%s "

 			  "character %u\n",linecnt,

 			  g_utf8_pointer_to_offset(aline,s)+1,charset,c);

 		    else

 			cnt_bin++;

 		    eInvalidChar=TRUE;

 		}

 	    }

 	}

 	if (!eTab && c==CHAR_TAB)

 	{

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column %ld - Tab character?\n",

 		  linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 	    else

 		cnt_odd++;

 	    eTab=TRUE;

 	}

 	if (!eTilde && c==CHAR_TILDE)

 	{

 	    /*

 	     * Often used by OCR software to indicate an

 	     * unrecognizable character.

 	     */

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column %ld - Tilde character?\n",

 		  linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 	    else

 		cnt_odd++;

 	    eTilde=TRUE;

 	}

 	if (!eCarat && c==CHAR_CARAT)

 	{  

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column %ld - Carat character?\n",

 		  linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 	    else

 		cnt_odd++;

 	    eCarat=TRUE;

 	}

 	if (!eFSlash && c==CHAR_FORESLASH && warnings->fslash)

 	{  

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column %ld - Forward slash?\n",

 		  linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 	    else

 		cnt_odd++;

 	    eFSlash=TRUE;

 	}

 	/*

 	 * Report asterisks only in paranoid mode,

 	 * since they're often deliberate.

 	 */

 	if (!eAst && pswit[PARANOID_SWITCH] && warnings->ast && !isemptyline &&

 	  c==CHAR_ASTERISK)

 	{

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column %ld - Asterisk?\n",

 		  linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 	    else

 		cnt_odd++;

 	    eAst=TRUE;

 	}

     }

 }

 /*

  * check_for_long_line:

  *

  * Check for line too long.

  */

 void check_for_long_line(const char *aline)

 {

     if (g_utf8_strlen(aline,-1)>LONGEST_PG_LINE)

     {

 	if (pswit[ECHO_SWITCH])

 	    g_print("\n%s\n",aline);

 	if (!pswit[OVERVIEW_SWITCH])

 	    g_print("    Line %ld column %ld - Long line %ld\n",

 	      linecnt,g_utf8_strlen(aline,-1),g_utf8_strlen(aline,-1));

 	else

 	    cnt_long++;

     }

 }

 /*

  * check_for_short_line:

  *

  * Check for line too short.

  *

  * This one is a bit trickier to implement: we don't want to

  * flag the last line of a paragraph for being short, so we

  * have to wait until we know that our current line is a

  * "normal" line, then report the _previous_ line if it was too

  * short. We also don't want to report indented lines like

  * chapter heads or formatted quotations. We therefore keep

  * last->len as the length of the last line examined, and

  * last->blen as the length of the last but one, and try to

  * suppress unnecessary warnings by checking that both were of

  * "normal" length. We keep the first character of the last

  * line in last->start, and if it was a space, we assume that

  * the formatting is deliberate. I can't figure out a way to

  * distinguish something like a quoted verse left-aligned or

  * the header or footer of a letter from a paragraph of short

  * lines - maybe if I examined the whole paragraph, and if the

  * para has less than, say, 8 lines and if all lines are short,

  * then just assume it's OK? Need to look at some texts to see

  * how often a formula like this would get the right result.

  */

 void check_for_short_line(const char *aline,const struct line_properties *last)

 {

     if (g_utf8_strlen(aline,-1)>1 && last->len>1 &&

       last->len<SHORTEST_PG_LINE && last->blen>1 &&

       last->blen>SHORTEST_PG_LINE && last->start!=CHAR_SPACE)

     {

 	if (pswit[ECHO_SWITCH])

 	    g_print("\n%s\n",prevline);

 	if (!pswit[OVERVIEW_SWITCH])

 	    g_print("    Line %ld column %ld - Short line %ld?\n",

 	      linecnt-1,g_utf8_strlen(prevline,-1),g_utf8_strlen(prevline,-1));

 	else

 	    cnt_short++;

     }

 }

 /*

  * check_for_starting_punctuation:

  *

  * Look for punctuation other than full ellipses at start of line.

  */

 void check_for_starting_punctuation(const char *aline)

 {

     if (*aline && g_utf8_strchr(".?!,;:",-1,g_utf8_get_char(aline)) &&

       !g_str_has_prefix(aline,". . ."))

     {

 	if (pswit[ECHO_SWITCH])

 	    g_print("\n%s\n",aline);

 	if (!pswit[OVERVIEW_SWITCH])

 	    g_print("    Line %ld column 1 - Begins with punctuation?\n",

 	      linecnt);

 	else

 	    cnt_punct++;

     }

 }

 /*

  * str_emdash:

  *

  * Find the first em-dash, return a pointer to it and set <next> to the

  * character following the dash.

  */

 char *str_emdash(const char *s,const char **next)

 {

     const char *s1,*s2;

     s1=strstr(s,"--");

     s2=strstr(s,"—");

     if (!s1)

     {

 	if (s2)

 	    *next=g_utf8_next_char(s2);

 	return (char *)s2;

     }

     else if (!s2)

     {

 	*next=g_utf8_next_char(g_utf8_next_char(s1));

 	return (char *)s1;

     }

     else if (s1<s2)

     {

 	*next=g_utf8_next_char(g_utf8_next_char(s1));

 	return (char *)s1;

     }

     else

     {

 	*next=g_utf8_next_char(s2);

 	return (char *)s2;

     }

 }

 /*

  * check_for_spaced_emdash:

  *

  * Check for spaced em-dashes.

  *

  * We must check _all_ occurrences of em-dashes on the line

  * hence the loop - even if the first dash is OK

  * there may be another that's wrong later on.

  */

 void check_for_spaced_emdash(const char *aline)

 {

     const char *s,*t,*next;

     for (s=aline;t=str_emdash(s,&next);s=next)

     {

 	if (t>aline && g_utf8_get_char(g_utf8_prev_char(t))==CHAR_SPACE ||

 	  g_utf8_get_char(next)==CHAR_SPACE)

 	{

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column %ld - Spaced em-dash?\n",

 		  linecnt,g_utf8_pointer_to_offset(aline,t)+1);

 	    else

 		cnt_dash++;

 	}

     }

 }

 /*

  * check_for_spaced_dash:

  *

  * Check for spaced dashes.

  */

 void check_for_spaced_dash(const char *aline)

 {

     const char *s;

     if ((s=strstr(aline," -")))

     {

 	if (g_utf8_get_char(g_utf8_next_char(g_utf8_next_char(s)))!='-')

 	{

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column %ld - Spaced dash?\n",

 		  linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 	    else

 		cnt_dash++;

 	}

     }

     else if ((s=strstr(aline,"- ")))

     {

 	if (s==aline || g_utf8_get_char(g_utf8_prev_char(s))!='-')

 	{

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column %ld - Spaced dash?\n",

 		  linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 	    else

 		cnt_dash++;

 	}

     }

 }

 /*

  * check_for_unmarked_paragraphs:

  *

  * Check for unmarked paragraphs indicated by separate speakers.

  *

  * May well be false positive:

  * "Bravo!" "Wonderful!" called the crowd.

  * but useful all the same.

  */

 void check_for_unmarked_paragraphs(const char *aline)

 {

     const char *s;

     s=strstr(aline,"\"  \"");

     if (!s)

 	s=strstr(aline,"\" \"");

     if (s)

     {

 	if (pswit[ECHO_SWITCH])

 	    g_print("\n%s\n",aline);

 	if (!pswit[OVERVIEW_SWITCH])

 	    g_print("    Line %ld column %ld - "

 	      "Query missing paragraph break?\n",

 	      linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 	else

 	    cnt_punct++;

     }

 }

 /*

  * check_for_jeebies:

  *

  * Check for "to he" and other easy h/b errors.

  *

  * This is a very inadequate effort on the h/b problem,

  * but the phrase "to he" is always an error, whereas "to

  * be" is quite common.

  * Similarly, '"Quiet!", be said.' is a non-be error

  * "to he" is _not_ always an error!:

  *       "Where they went to he couldn't say."

  * Another false positive:

  *       What would "Cinderella" be without the . . .

  * and another: "If he wants to he can see for himself."

  */

 void check_for_jeebies(const char *aline)

 {

     const char *s;

     s=strstr(aline," be could ");

     if (!s)

 	s=strstr(aline," be would ");

     if (!s)

 	s=strstr(aline," was be ");

     if (!s)

 	s=strstr(aline," be is ");

     if (!s)

 	s=strstr(aline," is be ");

     if (!s)

 	s=strstr(aline,"\", be ");

     if (!s)

 	s=strstr(aline,"\" be ");

     if (!s)

 	s=strstr(aline,"\" be ");

     if (!s)

 	s=strstr(aline," to he ");

     if (s)

     {

 	if (pswit[ECHO_SWITCH])

 	    g_print("\n%s\n",aline);

 	if (!pswit[OVERVIEW_SWITCH])

 	    g_print("    Line %ld column %ld - Query he/be error?\n",

 	      linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 	else

 	    cnt_word++;

     }

     s=strstr(aline," the had ");

     if (!s)

 	s=strstr(aline," a had ");

     if (!s)

 	s=strstr(aline," they bad ");

     if (!s)

 	s=strstr(aline," she bad ");

     if (!s)

 	s=strstr(aline," he bad ");

     if (!s)

 	s=strstr(aline," you bad ");

     if (!s)

 	s=strstr(aline," i bad ");

     if (s)

     {

 	if (pswit[ECHO_SWITCH])

 	    g_print("\n%s\n",aline);

 	if (!pswit[OVERVIEW_SWITCH])

 	    g_print("    Line %ld column %ld - Query had/bad error?\n",

 	      linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 	else

 	    cnt_word++;

     }

     s=strstr(aline,"; hut ");

     if (!s)

 	s=strstr(aline,", hut ");

     if (s)

     {

 	if (pswit[ECHO_SWITCH])

 	    g_print("\n%s\n",aline);

 	if (!pswit[OVERVIEW_SWITCH])

 	    g_print("    Line %ld column %ld - Query hut/but error?\n",

 	      linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 	else

 	    cnt_word++;

     }

 }

 /*

  * check_for_mta_from:

  *

  * Special case - angled bracket in front of "From" placed there by an

  * MTA when sending an e-mail.

  */

 void check_for_mta_from(const char *aline)

 {

     const char *s;

     s=strstr(aline,">From");

     if (s)

     {

 	if (pswit[ECHO_SWITCH])

 	    g_print("\n%s\n",aline);

 	if (!pswit[OVERVIEW_SWITCH])

 	    g_print("    Line %ld column %ld - "

 	      "Query angled bracket with From\n",

 	      linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 	else

 	    cnt_punct++;

     }

 }

 /*

  * check_for_orphan_character:

  *

  * Check for a single character line -

  * often an overflow from bad wrapping.

  */

 void check_for_orphan_character(const char *aline)

 {

     gunichar c;

     c=g_utf8_get_char(aline);

     if (c && !*g_utf8_next_char(aline))

     {

 	if (c=='I' || c=='V' || c=='X' || c=='L' || g_unichar_isdigit(c))

 	    ; /* Nothing - ignore numerals alone on a line. */

 	else

 	{

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column 1 - Query single character line\n",

 		  linecnt);

 	    else

 		cnt_punct++;

 	}

     }

 }

 /*

  * check_for_pling_scanno:

  *

  * Check for I" - often should be !

  */

 void check_for_pling_scanno(const char *aline)

 {

     const char *s;

     s=strstr(aline," I\"");

     if (s)

     {

 	if (pswit[ECHO_SWITCH])

 	    g_print("\n%s\n",aline);

 	if (!pswit[OVERVIEW_SWITCH])

 	    g_print("    Line %ld column %ld - Query I=exclamation mark?\n",

 	      linecnt,g_utf8_pointer_to_offset(aline,s));

 	else

 	    cnt_punct++;

     }

 }

 /*

  * check_for_extra_period:

  *

  * Check for period without a capital letter. Cut-down from gutspell.

  * Only works when it happens on a single line.

  */

 void check_for_extra_period(const char *aline,const struct warnings *warnings)

 {

     const char *s,*t,*s1,*sprev;

     int i;

     gsize len;

     gboolean istypo;

     gchar *testword;

     gunichar c,nc,pc,*decomposition;

     if (pswit[PARANOID_SWITCH])

     {

 	for (t=aline;t=strstr(t,". ");)

 	{

 	    if (t==aline)

 	    {

 		t=g_utf8_next_char(t);

 		/* start of line punctuation is handled elsewhere */

 		continue;

 	    }

 	    if (!g_unichar_isalpha(g_utf8_get_char(g_utf8_prev_char(t))))

 	    {

 		t=g_utf8_next_char(t);

 		continue;

 	    }

 	    if (warnings->isDutch)

 	    {

 		/* For Frank & Jeroen -- 's Middags case */

 		gunichar c2,c3,c4,c5;

 		c2=g_utf8_get_char(g_utf8_offset_to_pointer(t,2));

 		c3=g_utf8_get_char(g_utf8_offset_to_pointer(t,3));

 		c4=g_utf8_get_char(g_utf8_offset_to_pointer(t,4));

 		c5=g_utf8_get_char(g_utf8_offset_to_pointer(t,5));

 		if (CHAR_IS_APOSTROPHE(c2) &&

 		  g_unichar_islower(c3) && c4==CHAR_SPACE &&

 		  g_unichar_isupper(c5))

 		{

 		    t=g_utf8_next_char(t);

 		    continue;

 		}

 	    }

 	    s1=g_utf8_next_char(g_utf8_next_char(t));

 	    while (*s1 && !g_unichar_isalpha(g_utf8_get_char(s1)) &&

 	      !g_unichar_isdigit(g_utf8_get_char(s1)))

 		s1=g_utf8_next_char(s1);

 	    if (g_unichar_islower(g_utf8_get_char(s1)))

 	    {

 		/* we have something to investigate */

 		istypo=TRUE;

 		/* so let's go back and find out */

 		nc=g_utf8_get_char(t);

 		s1=g_utf8_prev_char(t);

 		c=g_utf8_get_char(s1);

 		sprev=g_utf8_prev_char(s1);

 		pc=g_utf8_get_char(sprev);

 		while (s1>=aline &&

 		  (g_unichar_isalpha(c) || g_unichar_isdigit(c) ||

 		  g_unichar_isalpha(pc) && CHAR_IS_APOSTROPHE(c) &&

 		  g_unichar_isalpha(nc)))

 		{

 		    nc=c;

 		    s1=sprev;

 		    c=pc;

 		    sprev=g_utf8_prev_char(s1);

 		    pc=g_utf8_get_char(sprev);

 		}

 		s1=g_utf8_next_char(s1);

 		s=strchr(s1,'.');

 		if (s)

 		    testword=g_strndup(s1,s-s1);

 		else

 		    testword=g_strdup(s1);

 		for (i=0;*abbrev[i];i++)

 		    if (!strcmp(testword,abbrev[i]))

 			istypo=FALSE;

 		if (g_unichar_isdigit(g_utf8_get_char(testword)))

 		    istypo=FALSE;

 		if (!*g_utf8_next_char(testword))

 		    istypo=FALSE;

 		if (isroman(testword))

 		    istypo=FALSE;

 		if (istypo)

 		{

 		    istypo=FALSE;

 		    for (s=testword;*s;s=g_utf8_next_char(s))

 		    {

 			decomposition=g_unicode_canonical_decomposition(

 			  g_utf8_get_char(s),&len);

 			if (g_utf8_strchr("aeiou",-1,decomposition[0]))

 			    istypo=TRUE;

 			g_free(decomposition);

 		    }

 		}

 		if (istypo &&

 		  (pswit[VERBOSE_SWITCH] || !g_tree_lookup(qperiod,testword)))

 		{

 		    g_tree_insert(qperiod,g_strdup(testword),

 		      GINT_TO_POINTER(1));

 		    if (pswit[ECHO_SWITCH])

 			g_print("\n%s\n",aline);

 		    if (!pswit[OVERVIEW_SWITCH])

 			g_print("    Line %ld column %ld - Extra period?\n",

 			  linecnt,g_utf8_pointer_to_offset(aline,t)+1);

 		    else

 			cnt_punct++;

 		}

 		g_free(testword);

 	    }

 	    t=g_utf8_next_char(t);

 	}

     }

 }

 /*

  * check_for_following_punctuation:

  *

  * Check for words usually not followed by punctuation.

  */

 void check_for_following_punctuation(const char *aline)

 {

     int i;

     const char *s,*wordstart;

     gunichar c;

     gchar *inword,*t;

     if (pswit[TYPO_SWITCH])

     {

 	for (s=aline;*s;)

 	{

 	    wordstart=s;

 	    t=getaword(NULL,&s);

 	    if (!*t)

 	    {

 		g_free(t);

 		continue;

 	    }

 	    inword=g_utf8_strdown(t,-1);

 	    g_free(t);

 	    for (i=0;*nocomma[i];i++)

 		if (!strcmp(inword,nocomma[i]))

 		{

 		    c=g_utf8_get_char(s);

 		    if (c==',' || c==';' || c==':')

 		    {

 			if (pswit[ECHO_SWITCH])

 			    g_print("\n%s\n",aline);

 			if (!pswit[OVERVIEW_SWITCH])

 			    g_print("    Line %ld column %ld - "

 			      "Query punctuation after %s?\n",

 			      linecnt,g_utf8_pointer_to_offset(aline,s)+1,

 			      inword);

 			else

 			    cnt_punct++;

 		    }

 		}

 	    for (i=0;*noperiod[i];i++)

 		if (!strcmp(inword,noperiod[i]))

 		{

 		    c=g_utf8_get_char(s);

 		    if (c=='.' || c=='!')

 		    {

 			if (pswit[ECHO_SWITCH])

 			    g_print("\n%s\n",aline);

 			if (!pswit[OVERVIEW_SWITCH])

 			    g_print("    Line %ld column %ld - "

 			      "Query punctuation after %s?\n",

 			      linecnt,g_utf8_pointer_to_offset(aline,s)+1,

 			      inword);

 			else

 			    cnt_punct++;

 		    }

 		}

 	    g_free(inword);

 	}

     }

 }

 /*

  * check_for_typos:

  *

  * Check for commonly mistyped words, and digits like 0 for O in a word.

  * Note that somewhat confusingly, this is also where we call getaword()

  * with a non-NULL line so that it will issue warnings.

  */

 void check_for_typos(const char *aline,struct warnings *warnings)

 {

     const char *s,*t,*nt,*wordstart;

     gchar *inword;

     gunichar *decomposition;

     gchar *testword;

     int i,vowel,consonant,*dupcnt;

     gboolean isdup,istypo,alower;

     gunichar c,pc;

     long offset,len;

     gsize decomposition_len;

     for (s=aline;*s;)

     {

 	wordstart=s;

 	inword=getaword(aline,&s);

 	if (!*inword)

 	{

 	    g_free(inword);

 	    continue; /* don't bother with empty lines */

 	}

 	if (mixdigit(inword))

 	{

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column %ld - Query digit in %s\n",

 		  linecnt,g_utf8_pointer_to_offset(aline,wordstart)+1,inword);

 	    else

 		cnt_word++;

 	}

 	/*

 	 * Put the word through a series of tests for likely typos and OCR

 	 * errors.

 	 */

 	if (pswit[TYPO_SWITCH] || pswit[USERTYPO_SWITCH])

 	{

 	    istypo=FALSE;

 	    alower=FALSE;

 	    for (t=inword;*t;t=g_utf8_next_char(t))

 	    {

 		c=g_utf8_get_char(t);

 		nt=g_utf8_next_char(t);

 		/* lowercase for testing */

 		if (g_unichar_islower(c))

 		    alower=TRUE;

 		if (alower && (g_unichar_isupper(c) || g_unichar_istitle(c)))

 		{

 		    /*

 		     * We have an uppercase mid-word. However, there are

 		     * common cases:

 		     *   Mac and Mc like McGill

 		     *   French contractions like l'Abbe

 		     */

 		    offset=g_utf8_pointer_to_offset(inword,t);

 		    if (offset>0)

 			pc=g_utf8_get_char(g_utf8_prev_char(t));

 		    else

 			pc='\0';

 		    if (offset==2 && c=='m' && g_utf8_get_char(nt)=='c' ||

 		      offset==3 && c=='m' && g_utf8_get_char(nt)=='a' &&

 		      g_utf8_get_char(g_utf8_next_char(nt))=='c' ||

 		      CHAR_IS_APOSTROPHE(pc))

 			; /* do nothing! */

 		    else

 			istypo=TRUE;

 		}

 	    }

 	    testword=g_utf8_casefold(inword,-1);

 	}

 	if (pswit[TYPO_SWITCH])

 	{

 	    /*

 	     * Check for certain unlikely two-letter combinations at word

 	     * start and end.

 	     */

 	    len=g_utf8_strlen(testword,-1);

 	    if (len>1)

 	    {

 		for (i=0;*nostart[i];i++)

 		    if (g_str_has_prefix(testword,nostart[i]))

 			istypo=TRUE;

 		for (i=0;*noend[i];i++)

 		    if (g_str_has_suffix(testword,noend[i]))

 			istypo=TRUE;

 	    }

 	    /* ght is common, gbt never. Like that. */

 	    if (strstr(testword,"cb"))

 		istypo=TRUE;

 	    if (strstr(testword,"gbt"))

 		istypo=TRUE;

 	    if (strstr(testword,"pbt"))

 		istypo=TRUE;

 	    if (strstr(testword,"tbs"))

 		istypo=TRUE;

 	    if (strstr(testword,"mrn"))

 		istypo=TRUE;

 	    if (strstr(testword,"ahle"))

 		istypo=TRUE;

 	    if (strstr(testword,"ihle"))

 		istypo=TRUE;

 	    /*

 	     * "TBE" does happen - like HEARTBEAT - but uncommon.

 	     * Also "TBI" - frostbite, outbid - but uncommon.

 	     * Similarly "ii" like Hawaii, or Pompeii, and in Roman

 	     * numerals, but "ii" is a common scanno.

 	     */

 	    if (strstr(testword,"tbi"))

 		istypo=TRUE;

 	    if (strstr(testword,"tbe"))

 		istypo=TRUE;

 	    if (strstr(testword,"ii"))

 		istypo=TRUE;

 	    /*

 	     * Check for no vowels or no consonants.

 	     * If none, flag a typo.

 	     */

 	    if (!istypo && len>1)

 	    {

 		vowel=consonant=0;

 		for (t=testword;*t;t=g_utf8_next_char(t))

 		{

 		    c=g_utf8_get_char(t);

 		    decomposition=

 		      g_unicode_canonical_decomposition(c,&decomposition_len);

 		    if (c=='y' || g_unichar_isdigit(c))

 		    {

 			/* Yah, this is loose. */

 			vowel++;

 			consonant++;

 		    }

 		    else if (g_utf8_strchr("aeiou",-1,decomposition[0]))

 			vowel++;

 		    else

 			consonant++;

 		    g_free(decomposition);

 		}

 		if (!vowel || !consonant)

 		    istypo=TRUE;

 	    }

 	    /*

 	     * Now exclude the word from being reported if it's in

 	     * the okword list.

 	     */

 	    for (i=0;*okword[i];i++)

 		if (!strcmp(testword,okword[i]))

 		    istypo=FALSE;

 	    /*

 	     * What looks like a typo may be a Roman numeral.

 	     * Exclude these.

 	     */

 	    if (istypo && isroman(testword))

 		istypo=FALSE;

 	    /* Check the manual list of typos. */

 	    if (!istypo)

 		for (i=0;*typo[i];i++)

 		    if (!strcmp(testword,typo[i]))

 			istypo=TRUE;

 	    /*

 	     * Check lowercase s, l, i and m - special cases.

 	     *   "j" - often a semi-colon gone wrong.

 	     *   "d" for a missing apostrophe - he d

 	     *   "n" for "in"

 	     */

 	    if (!istypo && len==1 &&

 	      g_utf8_strchr("slmijdn",-1,g_utf8_get_char(inword)))

 		istypo=TRUE;

 	    if (istypo)

 	    {

 		dupcnt=g_tree_lookup(qword,testword);

 		if (dupcnt)

 		{

 		    (*dupcnt)++;

 		    isdup=!pswit[VERBOSE_SWITCH];

 		}

 		else

 		{

 		    dupcnt=g_new0(int,1);

 		    g_tree_insert(qword,g_strdup(testword),dupcnt);

 		    isdup=FALSE;

 		}

 		if (!isdup)

 		{

 		    if (pswit[ECHO_SWITCH])

 			g_print("\n%s\n",aline);

 		    if (!pswit[OVERVIEW_SWITCH])

 		    {

 			g_print("    Line %ld column %ld - Query word %s",

 			  linecnt,g_utf8_pointer_to_offset(aline,wordstart)+1,

 			  inword);

 			if (!pswit[VERBOSE_SWITCH])

 			    g_print(" - not reporting duplicates");

 			g_print("\n");

 		    }

 		    else

 			cnt_word++;

 		}

 	    }

 	}

 	/* check the user's list of typos */

 	if (!istypo && usertypo && g_tree_lookup(usertypo,testword))

 	{

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])  

 		g_print("    Line %ld column %ld - Query possible scanno %s\n",

 		  linecnt,g_utf8_pointer_to_offset(aline,wordstart)+2,inword);

 	}

 	if (pswit[TYPO_SWITCH] || pswit[USERTYPO_SWITCH])

 	    g_free(testword);

 	if (pswit[PARANOID_SWITCH] && warnings->digit)

 	{

 	    /* In paranoid mode, query all 0 and 1 standing alone. */

 	    if (!strcmp(inword,"0") || !strcmp(inword,"1"))

 	    {

 		if (pswit[ECHO_SWITCH])

 		    g_print("\n%s\n",aline);

 		if (!pswit[OVERVIEW_SWITCH])

 		    g_print("    Line %ld column %ld - Query standalone %s\n",

 		      linecnt,g_utf8_pointer_to_offset(aline,wordstart)+2,

 		      inword);

 		else

 		    cnt_word++;

 	    }

 	}

 	g_free(inword);

     }

 }

 /*

  * check_for_misspaced_punctuation:

  *

  * Look for added or missing spaces around punctuation and quotes.

  * If there is a punctuation character like ! with no space on

  * either side, suspect a missing!space. If there are spaces on

  * both sides , assume a typo. If we see a double quote with no

  * space or punctuation on either side of it, assume unspaced

  * quotes "like"this.

  */

 void check_for_misspaced_punctuation(const char *aline,

   struct parities *parities,gboolean isemptyline)

 {

     gboolean isacro,isellipsis;

     const char *s;

     gunichar c,nc,pc,n2c;

     int parity;

     c=g_utf8_get_char(aline);

     nc=c?g_utf8_get_char(g_utf8_next_char(aline)):0;

     for (s=g_utf8_next_char(aline);nc;s=g_utf8_next_char(s))

     {

 	pc=c;

 	c=nc;

 	nc=g_utf8_get_char(g_utf8_next_char(s));

 	/* For each character in the line after the first. */

 	if (g_utf8_strchr(".?!,;:_",-1,c))  /* if it's punctuation */

 	{

 	    /* we need to suppress warnings for acronyms like M.D. */

 	    isacro=FALSE;

 	    /* we need to suppress warnings for ellipsis . . . */

 	    isellipsis=FALSE;

 	    /*

 	     * If there are letters on both sides of it or

 	     * if it's strict punctuation followed by an alpha.

 	     */

 	    if (c!='_' && g_unichar_isalpha(nc) && (g_unichar_isalpha(pc) ||

 	      g_utf8_strchr("?!,;:",-1,c)))

 	    {

 		if (c=='.')

 		{

 		    if (g_utf8_pointer_to_offset(aline,s)>2 &&

 		      g_utf8_get_char(g_utf8_offset_to_pointer(s,-2))=='.')

 			isacro=TRUE;

 		    n2c=g_utf8_get_char(g_utf8_next_char(g_utf8_next_char(s)));

 		    if (nc && n2c=='.')

 			isacro=TRUE;

 		}

 		if (!isacro)

 		{

 		    if (pswit[ECHO_SWITCH])

 			g_print("\n%s\n",aline);

 		    if (!pswit[OVERVIEW_SWITCH])

 			g_print("    Line %ld column %ld - Missing space?\n",

 			  linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 		    else

 			cnt_punct++;

 		}

 	    }

 	    if (pc==CHAR_SPACE && (nc==CHAR_SPACE || !nc))

 	    {

 		/*

 		 * If there are spaces on both sides,

 		 * or space before and end of line.

 		 */

 		if (c=='.')

 		{

 		    if (g_utf8_pointer_to_offset(aline,s)>2 &&

 		      g_utf8_get_char(g_utf8_offset_to_pointer(s,-2))=='.')

 			isellipsis=TRUE;

 		    n2c=g_utf8_get_char(g_utf8_next_char(g_utf8_next_char(s)));

 		    if (nc && n2c=='.')

 			isellipsis=TRUE;

 		}

 		if (!isemptyline && !isellipsis)

 		{

 		    if (pswit[ECHO_SWITCH])

 			g_print("\n%s\n",aline);

 		    if (!pswit[OVERVIEW_SWITCH])

 			g_print("    Line %ld column %ld - "

 			  "Spaced punctuation?\n",linecnt,

 			  g_utf8_pointer_to_offset(aline,s)+1);

 		    else

 			cnt_punct++;

 		}

 	    }

 	}

     }

     /* Split out the characters that CANNOT be preceded by space. */

     c=g_utf8_get_char(aline);

     nc=c?g_utf8_get_char(g_utf8_next_char(aline)):0;

     for (s=g_utf8_next_char(aline);nc;s=g_utf8_next_char(s))

     {

 	pc=c;

 	c=nc;

 	nc=g_utf8_get_char(g_utf8_next_char(s));

 	/* for each character in the line after the first */

 	if (g_utf8_strchr("?!,;:",-1,c))

 	{

 	    /* if it's punctuation that _cannot_ have a space before it */

 	    if (pc==CHAR_SPACE && !isemptyline && nc!=CHAR_SPACE)

 	    {

 		/*

 		 * If nc DOES == space,

 		 * it was already reported just above.

 		 */

 		if (pswit[ECHO_SWITCH])

 		    g_print("\n%s\n",aline);

 		if (!pswit[OVERVIEW_SWITCH])

 		    g_print("    Line %ld column %ld - Spaced punctuation?\n",

 		      linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 		else

 		    cnt_punct++;

 	    }

 	}

     }

     /*

      * Special case " .X" where X is any alpha.

      * This plugs a hole in the acronym code above.

      * Inelegant, but maintainable.

      */

     c=g_utf8_get_char(aline);

     nc=c?g_utf8_get_char(g_utf8_next_char(aline)):0;

     for (s=g_utf8_next_char(aline);nc;s=g_utf8_next_char(s))

     {

 	pc=c;

 	c=nc;

 	nc=g_utf8_get_char(g_utf8_next_char(s));

 	/* for each character in the line after the first */

 	if (c=='.')

 	{

 	    /* if it's a period */

 	    if (pc==CHAR_SPACE && g_unichar_isalpha(nc))

 	    {

 		/*

 		 * If the period follows a space and

 		 * is followed by a letter.

 		 */

 		if (pswit[ECHO_SWITCH])

 		    g_print("\n%s\n",aline);

 		if (!pswit[OVERVIEW_SWITCH])

 		    g_print("    Line %ld column %ld - Spaced punctuation?\n",

 		      linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 		else

 		    cnt_punct++;

 	    }

 	}

     }

     c=g_utf8_get_char(aline);

     nc=c?g_utf8_get_char(g_utf8_next_char(aline)):0;

     for (s=g_utf8_next_char(aline);nc;s=g_utf8_next_char(s))

     {

 	pc=c;

 	c=nc;

 	nc=g_utf8_get_char(g_utf8_next_char(s));

 	/* for each character in the line after the first */

 	if (CHAR_IS_DQUOTE(c))

 	{

 	    if (!g_utf8_strchr(" _-.'`,;:!/([{?}])",-1,pc) &&

 	      !g_utf8_strchr(" _-.'`,;:!/([{?}])",-1,nc) && nc ||

 	      !g_utf8_strchr(" _-([{'`",-1,pc) && g_unichar_isalpha(nc))

 	    {

 		if (pswit[ECHO_SWITCH])

 		    g_print("\n%s\n",aline);

 		if (!pswit[OVERVIEW_SWITCH])

 		    g_print("    Line %ld column %ld - Unspaced quotes?\n",

 		      linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 		else

 		    cnt_punct++;

 	    }

 	}

     }

     /* Check parity of quotes. */

     nc=g_utf8_get_char(aline);

     for (s=aline;*s;s=g_utf8_next_char(s))

     {

 	c=nc;

 	nc=g_utf8_get_char(g_utf8_next_char(s));

 	if (CHAR_IS_DQUOTE(c))

 	{

 	    if (c==CHAR_DQUOTE)

 	    {

 		parities->dquote=!parities->dquote;

 		parity=parities->dquote;

 	    }

 	    else if (c==CHAR_LD_QUOTE)

 		parity=1;

 	    else

 		parity=0;

 	    if (!parity)

 	    {

 		/* parity even */

 		if (!g_utf8_strchr("_-.'`‘’/,;:!?)]} ",-1,nc))

 		{

 		    if (pswit[ECHO_SWITCH])

 			g_print("\n%s\n",aline);

 		    if (!pswit[OVERVIEW_SWITCH])

 			g_print("    Line %ld column %ld - "

 			  "Wrongspaced quotes?\n",

 			  linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 		    else

 			cnt_punct++;

 		}

 	    }

 	    else

 	    {

 		/* parity odd */

 		if (!g_unichar_isalpha(nc) && !g_unichar_isdigit(nc) &&

 		  !g_utf8_strchr("_-/.'`‘’([{$",-1,nc) || !nc)

 		{

 		    if (pswit[ECHO_SWITCH])

 			g_print("\n%s\n",aline);

 		    if (!pswit[OVERVIEW_SWITCH])

 			g_print("    Line %ld column %ld - "

 			  "Wrongspaced quotes?\n",

 			  linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 		    else

 			cnt_punct++;

 		}

 	    }

 	}

     }

     c=g_utf8_get_char(aline);

     if (CHAR_IS_DQUOTE(c))

     {

 	if (g_utf8_strchr(",;:!?)]} ",-1,

 	  g_utf8_get_char(g_utf8_next_char(aline))))

 	{

 	    if (pswit[ECHO_SWITCH])

 		g_print("\n%s\n",aline);

 	    if (!pswit[OVERVIEW_SWITCH])

 		g_print("    Line %ld column 1 - Wrongspaced quotes?\n",

 		  linecnt);

 	    else

 		cnt_punct++;

 	}

     }

     if (pswit[SQUOTE_SWITCH])

     {

 	nc=g_utf8_get_char(aline);

 	for (s=aline;*s;s=g_utf8_next_char(s))

 	{

 	    c=nc;

 	    nc=g_utf8_get_char(g_utf8_next_char(s));

 	    if (CHAR_IS_SQUOTE(c) && (s==aline || s>aline &&

 	      !g_unichar_isalpha(g_utf8_get_char(g_utf8_prev_char(s))) ||

 	      !g_unichar_isalpha(nc)))

 	    {

 		parities->squote=!parities->squote;

 		if (!parities->squote)

 		{

 		    /* parity even */

 		    if (!g_utf8_strchr("_-.'`/\",;:!?)]} ",-1,nc))

 		    {

 			if (pswit[ECHO_SWITCH])

 			    g_print("\n%s\n",aline);

 			if (!pswit[OVERVIEW_SWITCH])

 			    g_print("    Line %ld column %ld - "

 			      "Wrongspaced singlequotes?\n",

 			      linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 			else

 			    cnt_punct++;

 		    }

 		}

 		else

 		{

 		    /* parity odd */

 		    if (!g_unichar_isalpha(nc) && !g_unichar_isdigit(nc) &&

 		      !g_utf8_strchr("_-/\".'`",-1,nc) || !nc)

 		    {

 			if (pswit[ECHO_SWITCH])

 			    g_print("\n%s\n",aline);

 			if (!pswit[OVERVIEW_SWITCH])

 			    g_print("    Line %ld column %ld - "

 			      "Wrongspaced singlequotes?\n",

 			      linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 			else

 			    cnt_punct++;

 		    }

 		}

 	    }

 	}

     }

 }

 /*

  * str_follows_word:

  *

  * Given a position p within a string str, determine whether it follows the

  * given word. This is roughly equivalent to the regular expression (?<=\bword)

  * but has different boundary conditions.

  */

 static gboolean str_follows_word(const char *str,const char *p,const char *word)

 {

     int len=strlen(word);

     if (p-len<str)

 	return FALSE;

     else if (!g_str_has_prefix(p-len,word))

 	return FALSE;

     else if (p-len==str)

 	return TRUE;

     else

 	/* Using non-alpha as a word boundary. See UAX #29 for a better way. */

 	return !g_unichar_isalpha(g_utf8_get_char(g_utf8_prev_char(p-len)));

 }

 /*

  * check_for_double_punctuation:

  *

  * Look for double punctuation like ,. or ,,

  * Thanks to DW for the suggestion!

  * In books with references, ".," and ".;" are common

  * e.g. "etc., etc.," and vol. 1.; vol 3.;

  * OTOH, from my initial tests, there are also fairly

  * common errors. What to do? Make these cases paranoid?

  * ".," is the most common, so warnings->dotcomma is used

  * to suppress detailed reporting if it occurs often.

  * Indeed, ".," is so common after "etc" or "&c" that

  * we don't warn on these cases at all.

  */

 void check_for_double_punctuation(const char *aline,struct warnings *warnings)

 {

     const char *s;

     gunichar c,nc;

     gboolean is_query;

     nc=g_utf8_get_char(aline);

     for (s=aline;*s;s=g_utf8_next_char(s))

     {

 	c=nc;

 	nc=g_utf8_get_char(g_utf8_next_char(s));

 	/* for each punctuation character in the line */

 	if (c && nc && g_utf8_strchr(".?!,;:",-1,c) &&

 	  g_utf8_strchr(".?!,;:",-1,nc))

 	{

 	    /* followed by punctuation, it's a query, unless . . . */

 	    is_query=TRUE;

 	    if (warnings->isFrench &&

 	      (g_str_has_prefix(s,",...") || g_str_has_prefix(s,"...,") ||

 	       g_str_has_prefix(s,";...") || g_str_has_prefix(s,"...;") ||

 	       g_str_has_prefix(s,":...") || g_str_has_prefix(s,"...:") ||

 	       g_str_has_prefix(s,"!...") || g_str_has_prefix(s,"...!") ||

 	       g_str_has_prefix(s,"?...") || g_str_has_prefix(s,"...?")))

 	    {

 		s+=4;

 		nc=g_utf8_get_char(g_utf8_next_char(s));

 		is_query=FALSE;

 	    }

 	    else if (c==nc && (c=='.' || c=='?' || c=='!'))

 	    {

 		/* do nothing for .. !! and ?? which can be legit */

 		is_query=FALSE;

 	    }

 	    else if (c=='.' && nc==',')

 	    {

 		if (!warnings->dotcomma || str_follows_word(aline,s,"etc") || 

 		  str_follows_word(aline,s,"&c"))

 		    is_query=FALSE;

 	    }

 	    if (is_query)

 	    {

 		if (pswit[ECHO_SWITCH])

 		    g_print("\n%s\n",aline);

 		if (!pswit[OVERVIEW_SWITCH])

 		    g_print("    Line %ld column %ld - Double punctuation?\n",

 		      linecnt,g_utf8_pointer_to_offset(aline,s)+1);

 		else

 		    cnt_punct++;

 	    }