bookloupe-testing: bookloupe/bookloupe.c@f2b7980222cf (annotated)

ali@0	1	/*************************************************************************/
ali@40	2	/* bookloupe--check for assorted weirdnesses in a PG candidate text file */
ali@68	3	/* */
ali@68	4	/* Copyright 2000-2005 Jim Tinsley <jtinsley@pobox.com> */
ali@68	5	/* Copyright 2012- J. Ali Harlow <ali@juiblex.co.uk> */
ali@68	6	/* */
ali@0	7	/* This program is free software; you can redistribute it and/or modify */
ali@0	8	/* it under the terms of the GNU General Public License as published by */
ali@0	9	/* the Free Software Foundation; either version 2 of the License, or */
ali@68	10	/* (at your option) any later version. */
ali@68	11	/* */
ali@0	12	/* This program is distributed in the hope that it will be useful, */
ali@68	13	/* but WITHOUT ANY WARRANTY; without even the implied warranty of */
ali@68	14	/* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the */
ali@68	15	/* GNU General Public License for more details. */
ali@68	16	/* */
ali@68	17	/* You should have received a copy of the GNU General Public License */
ali@68	18	/* along with this program. If not, see <http://www.gnu.org/licenses/>. */
ali@0	19	/*************************************************************************/
ali@0	20
ali@0	21	#include <stdio.h>
ali@0	22	#include <stdlib.h>
ali@0	23	#include <string.h>
ali@0	24	#include <ctype.h>
ali@73	25	#ifdef __WIN32__
ali@73	26	#include <windows.h>
ali@73	27	#endif
ali@69	28	#include <glib.h>
ali@69	29	#include <bl/bl.h>
ali@99	30	#include "bookloupe.h"
ali@99	31	#include "counters.h"
ali@103	32	#include "pending.h"
ali@71	33	#include "HTMLentities.h"
ali@0	34
ali@69	35	gchar *prevline;
ali@0	36
ali@40	37	/* Common typos. */
ali@40	38	char *typo[] = {
ali@40	39	"teh", "th", "og", "fi", "ro", "adn", "yuo", "ot", "fo", "thet", "ane",
ali@40	40	"nad", "te", "ig", "acn", "ahve", "alot", "anbd", "andt", "awya", "aywa",
ali@40	41	"bakc", "om", "btu", "byt", "cna", "cxan", "coudl", "dont", "didnt",
ali@40	42	"couldnt", "wouldnt", "doesnt", "shouldnt", "doign", "ehr", "hmi", "hse",
ali@40	43	"esle", "eyt", "fitrs", "firts", "foudn", "frmo", "fromt", "fwe", "gaurd",
ali@40	44	"gerat", "goign", "gruop", "haev", "hda", "hearign", "seeign", "sayign",
ali@40	45	"herat", "hge", "hsa", "hsi", "hte", "htere", "htese", "htey", "htis",
ali@40	46	"hvae", "hwich", "idae", "ihs", "iits", "int", "iwll", "iwth", "jsut",
ali@40	47	"loev", "sefl", "myu", "nkow", "nver", "nwe", "nwo", "ocur", "ohter",
ali@40	48	"omre", "onyl", "otehr", "otu", "owrk", "owuld", "peice", "peices",
ali@40	49	"peolpe", "peopel", "perhasp", "perhpas", "pleasent", "poeple", "porblem",
ali@40	50	"porblems", "rwite", "saidt", "saidh", "saids", "seh", "smae", "smoe",
ali@40	51	"sohw", "stnad", "stopry", "stoyr", "stpo", "tahn", "taht", "tath",
ali@40	52	"tehy", "tghe", "tghis", "theri", "theyll", "thgat", "thge", "thier",
ali@40	53	"thna", "thne", "thnig", "thnigs", "thsi", "thsoe", "thta", "timne",
ali@40	54	"tirne", "tkae", "tthe", "tyhat", "tyhe", "veyr", "vou", "vour", "vrey",
ali@40	55	"waht", "wasnt", "awtn", "watn", "wehn", "whic", "whcih", "whihc", "whta",
ali@40	56	"wihch", "wief", "wiht", "witha", "wiull", "wnat", "wnated", "wnats",
ali@40	57	"woh", "wohle", "wokr", "woudl", "wriet", "wrod", "wroet", "wroking",
ali@40	58	"wtih", "wuould", "wya", "yera", "yeras", "yersa", "yoiu", "youve",
ali@40	59	"ytou", "yuor", "abead", "ahle", "ahout", "ahove", "altbough", "balf",
ali@40	60	"bardly", "bas", "bave", "baving", "bebind", "beld", "belp", "belped",
ali@40	61	"ber", "bere", "bim", "bis", "bome", "bouse", "bowever", "buge",
ali@40	62	"dehates", "deht", "han", "hecause", "hecome", "heen", "hefore", "hegan",
ali@40	63	"hegin", "heing", "helieve", "henefit", "hetter", "hetween", "heyond",
ali@40	64	"hig", "higber", "huild", "huy", "hy", "jobn", "joh", "meanwbile",
ali@40	65	"memher", "memhers", "numher", "numhers", "perbaps", "prohlem", "puhlic",
ali@40	66	"witbout", "arn", "hin", "hirn", "wrok", "wroked", "amd", "aud",
ali@40	67	"prornise", "prornised", "modem", "bo", "heside", "chapteb", "chaptee",
ali@40	68	"se", ""
ali@40	69	};
ali@0	70
ali@69	71	GTree *usertypo;
ali@0	72
ali@40	73	/* Common abbreviations and other OK words not to query as typos. */
ali@40	74	char *okword[] = {
ali@40	75	"mr", "mrs", "mss", "mssrs", "ft", "pm", "st", "dr", "hmm", "h'm", "hmmm",
ali@40	76	"rd", "sh", "br", "pp", "hm", "cf", "jr", "sr", "vs", "lb", "lbs", "ltd",
ali@40	77	"pompeii","hawaii","hawaiian", "hotbed", "heartbeat", "heartbeats",
ali@40	78	"outbid", "outbids", "frostbite", "frostbitten", ""
ali@40	79	};
ali@0	80
ali@40	81	/* Common abbreviations that cause otherwise unexplained periods. */
ali@40	82	char *abbrev[] = {
ali@40	83	"cent", "cents", "viz", "vol", "vols", "vid", "ed", "al", "etc", "op",
ali@40	84	"cit", "deg", "min", "chap", "oz", "mme", "mlle", "mssrs", ""
ali@40	85	};
ali@0	86
ali@40	87	/*
ali@40	88	* Two-Letter combinations that rarely if ever start words,
ali@40	89	* but are common scannos or otherwise common letter combinations.
ali@40	90	*/
ali@40	91	char *nostart[] = {
ali@40	92	"hr", "hl", "cb", "sb", "tb", "wb", "tl", "tn", "rn", "lt", "tj", ""
ali@40	93	};
ali@0	94
ali@40	95	/*
ali@40	96	* Two-Letter combinations that rarely if ever end words,
ali@40	97	* but are common scannos or otherwise common letter combinations.
ali@40	98	*/
ali@40	99	char *noend[] = {
ali@40	100	"cb", "gb", "pb", "sb", "tb", "wh", "fr", "br", "qu", "tw", "gl", "fl",
ali@40	101	"sw", "gr", "sl", "cl", "iy", ""
ali@40	102	};
ali@0	103
ali@40	104	char *markup[] = {
ali@40	105	"a", "b", "big", "blockquote", "body", "br", "center", "col", "div", "em",
ali@40	106	"font", "h1", "h2", "h3", "h4", "h5", "h6", "head", "hr", "html", "i",
ali@40	107	"img", "li", "meta", "ol", "p", "pre", "small", "span", "strong", "sub",
ali@40	108	"sup", "table", "td", "tfoot", "thead", "title", "tr", "tt", "u", "ul", ""
ali@40	109	};
ali@0	110
ali@40	111	char *DPmarkup[] = {
ali@40	112	"<sc>", "</sc>", "/", "/", "/#", "#/", "/$", "$/", "<tb>", ""
ali@40	113	};
ali@0	114
ali@40	115	char *nocomma[] = {
ali@40	116	"the", "it's", "their", "an", "mrs", "a", "our", "that's", "its", "whose",
ali@40	117	"every", "i'll", "your", "my", "mr", "mrs", "mss", "mssrs", "ft", "pm",
ali@40	118	"st", "dr", "rd", "pp", "cf", "jr", "sr", "vs", "lb", "lbs", "ltd", "i'm",
ali@40	119	"during", "let", "toward", "among", ""
ali@40	120	};
ali@0	121
ali@40	122	char *noperiod[] = {
ali@40	123	"every", "i'm", "during", "that's", "their", "your", "our", "my", "or",
ali@40	124	"and", "but", "as", "if", "the", "its", "it's", "until", "than", "whether",
ali@40	125	"i'll", "whose", "who", "because", "when", "let", "till", "very", "an",
ali@40	126	"among", "those", "into", "whom", "having", "thence", ""
ali@40	127	};
ali@0	128
ali@69	129	gboolean pswit[SWITNO]; /* program switches */
ali@0	130
ali@177	131	gboolean typo_compat,paranoid_compat;
ali@177	132
ali@69	133	static GOptionEntry options[]={
ali@69	134	{ "dp", 'd', 0, G_OPTION_ARG_NONE, pswit+DP_SWITCH,
ali@69	135	"Ignore DP-specific markup", NULL },
ali@177	136	{ "no-dp", 0, G_OPTION_FLAG_HIDDEN\|G_OPTION_FLAG_REVERSE,
ali@177	137	G_OPTION_ARG_NONE, pswit+DP_SWITCH,
ali@177	138	"Don't ignore DP-specific markup", NULL },
ali@177	139	{ "echo", 0, G_OPTION_FLAG_HIDDEN, G_OPTION_ARG_NONE, pswit+ECHO_SWITCH,
ali@177	140	"Echo queried line", NULL },
ali@177	141	{ "no-echo", 'e', G_OPTION_FLAG_REVERSE,
ali@177	142	G_OPTION_ARG_NONE, pswit+ECHO_SWITCH,
ali@69	143	"Don't echo queried line", NULL },
ali@69	144	{ "squote", 's', 0, G_OPTION_ARG_NONE, pswit+SQUOTE_SWITCH,
ali@69	145	"Check single quotes", NULL },
ali@177	146	{ "no-squote", 0, G_OPTION_FLAG_HIDDEN\|G_OPTION_FLAG_REVERSE,
ali@177	147	G_OPTION_ARG_NONE, pswit+SQUOTE_SWITCH,
ali@177	148	"Don't check single quotes", NULL },
ali@177	149	{ "typo", 0, 0, G_OPTION_ARG_NONE, pswit+TYPO_SWITCH,
ali@69	150	"Check common typos", NULL },
ali@177	151	{ "no-typo", 0, G_OPTION_FLAG_HIDDEN\|G_OPTION_FLAG_REVERSE,
ali@177	152	G_OPTION_ARG_NONE, pswit+TYPO_SWITCH,
ali@177	153	"Don't check common typos", NULL },
ali@69	154	{ "qpara", 'p', 0, G_OPTION_ARG_NONE, pswit+QPARA_SWITCH,
ali@69	155	"Require closure of quotes on every paragraph", NULL },
ali@177	156	{ "no-qpara", 0, G_OPTION_FLAG_HIDDEN\|G_OPTION_FLAG_REVERSE,
ali@177	157	G_OPTION_ARG_NONE, pswit+QPARA_SWITCH,
ali@177	158	"Don't require closure of quotes on every paragraph", NULL },
ali@177	159	{ "paranoid", 0, G_OPTION_FLAG_HIDDEN,
ali@177	160	G_OPTION_ARG_NONE, pswit+PARANOID_SWITCH,
ali@177	161	"Enable paranoid querying of everything", NULL },
ali@177	162	{ "no-paranoid", 0, G_OPTION_FLAG_REVERSE,
ali@177	163	G_OPTION_ARG_NONE, pswit+PARANOID_SWITCH,
ali@69	164	"Disable paranoid querying of everything", NULL },
ali@177	165	{ "line-end", 0, G_OPTION_FLAG_HIDDEN,
ali@177	166	G_OPTION_ARG_NONE, pswit+LINE_END_SWITCH,
ali@177	167	"Enable line end checking", NULL },
ali@177	168	{ "no-line-end", 'l', G_OPTION_FLAG_REVERSE,
ali@177	169	G_OPTION_ARG_NONE, pswit+LINE_END_SWITCH,
ali@177	170	"Diable line end checking", NULL },
ali@69	171	{ "overview", 'o', 0, G_OPTION_ARG_NONE, pswit+OVERVIEW_SWITCH,
ali@69	172	"Overview: just show counts", NULL },
ali@177	173	{ "no-overview", 0, G_OPTION_FLAG_HIDDEN\|G_OPTION_FLAG_REVERSE,
ali@177	174	G_OPTION_ARG_NONE, pswit+OVERVIEW_SWITCH,
ali@177	175	"Show individual warnings", NULL },
ali@69	176	{ "stdout", 'y', 0, G_OPTION_ARG_NONE, pswit+STDOUT_SWITCH,
ali@69	177	"Output errors to stdout instead of stderr", NULL },
ali@177	178	{ "no-stdout", 0, G_OPTION_FLAG_HIDDEN\|G_OPTION_FLAG_REVERSE,
ali@177	179	G_OPTION_ARG_NONE, pswit+STDOUT_SWITCH,
ali@177	180	"Output errors to stderr instead of stdout", NULL },
ali@69	181	{ "header", 'h', 0, G_OPTION_ARG_NONE, pswit+HEADER_SWITCH,
ali@69	182	"Echo header fields", NULL },
ali@177	183	{ "no-header", 0, G_OPTION_FLAG_HIDDEN\|G_OPTION_FLAG_REVERSE,
ali@177	184	G_OPTION_ARG_NONE, pswit+HEADER_SWITCH,
ali@177	185	"Don't echo header fields", NULL },
ali@69	186	{ "markup", 'm', 0, G_OPTION_ARG_NONE, pswit+MARKUP_SWITCH,
ali@69	187	"Ignore markup in < >", NULL },
ali@177	188	{ "no-markup", 0, G_OPTION_FLAG_HIDDEN\|G_OPTION_FLAG_REVERSE,
ali@177	189	G_OPTION_ARG_NONE, pswit+MARKUP_SWITCH,
ali@177	190	"No special handling for markup in < >", NULL },
ali@69	191	{ "usertypo", 'u', 0, G_OPTION_ARG_NONE, pswit+USERTYPO_SWITCH,
ali@69	192	"Use file of user-defined typos", NULL },
ali@177	193	{ "no-usertypo", 0, G_OPTION_FLAG_HIDDEN\|G_OPTION_FLAG_REVERSE,
ali@177	194	G_OPTION_ARG_NONE, pswit+USERTYPO_SWITCH,
ali@177	195	"Ignore file of user-defined typos", NULL },
ali@177	196	{ "verbose", 'v', 0, G_OPTION_ARG_NONE, pswit+VERBOSE_SWITCH,
ali@177	197	"Verbose - list everything", NULL },
ali@177	198	{ "no-verbose", 0, G_OPTION_FLAG_HIDDEN\|G_OPTION_FLAG_REVERSE,
ali@177	199	G_OPTION_ARG_NONE, pswit+VERBOSE_SWITCH,
ali@177	200	"Switch off verbose mode", NULL },
ali@177	201	{ NULL }
ali@177	202	};
ali@177	203
ali@177	204	/*
ali@177	205	* Options relating to configuration which make no sense from inside
ali@177	206	* a configuration file.
ali@177	207	*/
ali@177	208
ali@177	209	static GOptionEntry config_options[]={
ali@69	210	{ "web", 'w', 0, G_OPTION_ARG_NONE, pswit+WEB_SWITCH,
ali@69	211	"Defaults for use on www upload", NULL },
ali@177	212	{ "dump-config", 0, 0, G_OPTION_ARG_NONE, pswit+DUMP_CONFIG_SWITCH,
ali@177	213	"Dump current config settings", NULL },
ali@177	214	{ NULL }
ali@177	215	};
ali@177	216
ali@177	217	static GOptionEntry compatibility_options[]={
ali@177	218	{ "toggle-typo", 't', 0, G_OPTION_ARG_NONE, &typo_compat,
ali@177	219	"Toggle checking for common typos", NULL },
ali@177	220	{ "toggle-relaxed", 'x', 0, G_OPTION_ARG_NONE, &paranoid_compat,
ali@177	221	"Toggle both paranoid mode and common typos", NULL },
ali@69	222	{ NULL }
ali@69	223	};
ali@0	224
ali@142	225	long cnt_quote; /* for overview mode, count of quote queries */
ali@68	226	long cnt_brack; /* for overview mode, count of brackets queries */
ali@68	227	long cnt_bin; /* for overview mode, count of non-ASCII queries */
ali@68	228	long cnt_odd; /* for overview mode, count of odd character queries */
ali@68	229	long cnt_long; /* for overview mode, count of long line errors */
ali@68	230	long cnt_short; /* for overview mode, count of short line queries */
ali@68	231	long cnt_punct; /* for overview mode,
ali@68	232	count of punctuation and spacing queries */
ali@68	233	long cnt_dash; /* for overview mode, count of dash-related queries */
ali@68	234	long cnt_word; /* for overview mode, count of word queries */
ali@68	235	long cnt_html; /* for overview mode, count of html queries */
ali@68	236	long cnt_lineend; /* for overview mode, count of line-end queries */
ali@68	237	long cnt_spacend; /* count of lines with space at end */
ali@68	238	long linecnt; /* count of total lines in the file */
ali@68	239	long checked_linecnt; /* count of lines actually checked */
ali@0	240
ali@69	241	void proghelp(GOptionContext *context);
ali@69	242	void procfile(const char *);
ali@0	243
ali@69	244	gchar *running_from;
ali@0	245
ali@70	246	gboolean mixdigit(const char *);
ali@69	247	gchar getaword(const char *);
ali@69	248	char flgets(char *,long);
ali@0	249	void postprocess_for_HTML(char *);
ali@0	250	char linehasmarkup(char );
ali@0	251	char losemarkup(char );
ali@70	252	gboolean tagcomp(const char ,const char );
ali@71	253	void loseentities(char *);
ali@69	254	gboolean isroman(const char *);
ali@0	255	void postprocess_for_DP(char *);
ali@72	256	void print_as_windows_1252(const char *string);
ali@72	257	void print_as_utf_8(const char *string);
ali@0	258
ali@69	259	GTree qword,qperiod;
ali@68	260
ali@73	261	#ifdef __WIN32__
ali@73	262	UINT saved_cp;
ali@73	263	#endif
ali@73	264
ali@177	265	GKeyFile *config;
ali@177	266
ali@177	267	void config_file_update(GKeyFile *kf)
ali@177	268	{
ali@177	269	int i;
ali@177	270	gboolean sw;
ali@177	271	for(i=0;options[i].long_name;i++)
ali@177	272	{
ali@177	273	if (g_str_has_prefix(options[i].long_name,"no-"))
ali@177	274	continue;
ali@177	275	if (options[i].arg==G_OPTION_ARG_NONE)
ali@177	276	{
ali@177	277	sw=(gboolean )options[i].arg_data;
ali@177	278	if (options[i].flags&G_OPTION_FLAG_REVERSE)
ali@177	279	sw=!sw;
ali@177	280	g_key_file_set_boolean(kf,"options",options[i].long_name,sw);
ali@177	281	}
ali@177	282	else
ali@177	283	g_assert_not_reached();
ali@177	284	}
ali@177	285	}
ali@177	286
ali@177	287	void config_file_add_comments(GKeyFile *kf)
ali@177	288	{
ali@177	289	int i;
ali@177	290	gchar *comment;
ali@177	291	g_key_file_set_comment(kf,NULL,NULL," Default configuration for bookloupe",
ali@177	292	NULL);
ali@177	293	for(i=0;options[i].long_name;i++)
ali@177	294	{
ali@177	295	if (g_str_has_prefix(options[i].long_name,"no-"))
ali@177	296	continue;
ali@177	297	comment=g_strconcat(" ",options[i].description,NULL);
ali@177	298	g_key_file_set_comment(kf,"options",options[i].long_name,comment,NULL);
ali@177	299	g_free(comment);
ali@177	300	}
ali@177	301	}
ali@177	302
ali@177	303	void dump_config(void)
ali@177	304	{
ali@177	305	gchar *s;
ali@177	306	if (config)
ali@177	307	config_file_update(config);
ali@177	308	else
ali@177	309	{
ali@177	310	config=g_key_file_new();
ali@177	311	config_file_update(config);
ali@177	312	config_file_add_comments(config);
ali@177	313	}
ali@177	314	s=g_key_file_to_data(config,NULL,NULL);
ali@177	315	if (s)
ali@177	316	g_print("%s",s);
ali@177	317	g_free(s);
ali@177	318	}
ali@177	319
ali@177	320	GKeyFile read_config_file(gchar *full_path)
ali@177	321	{
ali@177	322	int i;
ali@177	323	GError *err=NULL;
ali@177	324	gchar **search_dirs;
ali@177	325	gchar *path;
ali@177	326	const char *search_path;
ali@177	327	GKeyFile *kf;
ali@177	328	kf=g_key_file_new();
ali@177	329	search_path=g_getenv("BOOKLOUPE_CONFIG_PATH");
ali@177	330	if (search_path)
ali@177	331	{
ali@177	332	#ifdef __WIN32__
ali@177	333	search_dirs=g_strsplit(search_path,";",0);
ali@177	334	#else
ali@177	335	search_dirs=g_strsplit(search_path,":",0);
ali@177	336	#endif
ali@177	337	}
ali@177	338	else
ali@177	339	{
ali@177	340	search_dirs=g_new(gchar *,4);
ali@177	341	search_dirs[0]=g_get_current_dir();
ali@177	342	search_dirs[1]=g_strdup(running_from);
ali@177	343	search_dirs[2]=g_strdup(g_get_user_config_dir());
ali@177	344	search_dirs[3]=NULL;
ali@177	345	}
ali@177	346	for(i=0;search_dirs[i];i++)
ali@177	347	{
ali@177	348	path=g_build_filename(search_dirs[i],"bookloupe.ini",NULL);
ali@177	349	if (g_key_file_load_from_file(kf,path,
ali@177	350	G_KEY_FILE_KEEP_COMMENTS\|G_KEY_FILE_KEEP_TRANSLATIONS,&err))
ali@177	351	break;
ali@177	352	if (!g_error_matches(err,G_FILE_ERROR,G_FILE_ERROR_NOENT))
ali@177	353	{
ali@177	354	g_printerr("Bookloupe: Error reading %s\n",path);
ali@177	355	g_printerr("%s\n",err->message);
ali@177	356	exit(1);
ali@177	357	}
ali@177	358	g_clear_error(&err);
ali@177	359	g_free(path);
ali@177	360	path=NULL;
ali@177	361	}
ali@177	362	if (!search_dirs[i])
ali@177	363	{
ali@177	364	g_key_file_free(kf);
ali@177	365	kf=NULL;
ali@177	366	}
ali@177	367	g_strfreev(search_dirs);
ali@177	368	if (full_path && kf)
ali@177	369	*full_path=path;
ali@177	370	else
ali@177	371	g_free(path);
ali@177	372	return kf;
ali@177	373	}
ali@177	374
ali@177	375	void parse_config_file(void)
ali@177	376	{
ali@177	377	int i,j;
ali@177	378	gchar *path;
ali@177	379	gchar **keys;
ali@177	380	gboolean sw;
ali@177	381	GError *err=NULL;
ali@177	382	config=read_config_file(&path);
ali@177	383	if (config)
ali@177	384	keys=g_key_file_get_keys(config,"options",NULL,NULL);
ali@177	385	else
ali@177	386	keys=NULL;
ali@177	387	if (keys)
ali@177	388	{
ali@177	389	for(i=0;keys[i];i++)
ali@177	390	{
ali@177	391	for(j=0;options[j].long_name;j++)
ali@177	392	{
ali@177	393	if (g_str_has_prefix(options[j].long_name,"no-"))
ali@177	394	continue;
ali@177	395	else if (!strcmp(keys[i],options[j].long_name))
ali@177	396	{
ali@177	397	if (options[j].arg==G_OPTION_ARG_NONE)
ali@177	398	{
ali@177	399	sw=g_key_file_get_boolean(config,"options",keys[i],
ali@177	400	&err);
ali@177	401	if (err)
ali@177	402	{
ali@177	403	g_printerr("Bookloupe: %s: options.%s: %s\n",
ali@177	404	path,keys[i],err->message);
ali@177	405	g_clear_error(&err);
ali@177	406	}
ali@177	407	if (options[j].flags&G_OPTION_FLAG_REVERSE)
ali@177	408	sw=!sw;
ali@177	409	(gboolean )options[j].arg_data=sw;
ali@177	410	break;
ali@177	411	}
ali@177	412	else
ali@177	413	g_assert_not_reached();
ali@177	414	}
ali@177	415	}
ali@177	416	if (!options[j].long_name)
ali@177	417	g_printerr("Bookloupe: %s: Unknown option \"%s\" ignored\n",
ali@177	418	path,keys[i]);
ali@177	419	}
ali@177	420	g_strfreev(keys);
ali@177	421	}
ali@177	422	if (config)
ali@177	423	g_free(path);
ali@177	424	}
ali@177	425
ali@69	426	void parse_options(int argc,char **argv)
ali@0	427	{
ali@69	428	GError *err=NULL;
ali@69	429	GOptionContext *context;
ali@177	430	GOptionGroup *compatibility;
ali@69	431	context=g_option_context_new(
ali@177	432	"file - look for errors in Project Gutenberg(TM) etexts");
ali@69	433	g_option_context_add_main_entries(context,options,NULL);
ali@177	434	g_option_context_add_main_entries(context,config_options,NULL);
ali@177	435	compatibility=g_option_group_new("compatibility",
ali@177	436	"Options for Compatibility with Gutcheck:",
ali@177	437	"Show compatibility options",NULL,NULL);
ali@177	438	g_option_group_add_entries(compatibility,compatibility_options);
ali@177	439	g_option_context_add_group(context,compatibility);
ali@177	440	g_option_context_set_description(context,
ali@177	441	"For simplicity, only the switch options which reverse the\n"
ali@177	442	"default configuration are listed. In most cases, both vanilla\n"
ali@177	443	"and \"no-\" prefixed versions are available for use.");
ali@69	444	if (!g_option_context_parse(context,argc,argv,&err))
ali@69	445	{
ali@69	446	g_printerr("Bookloupe: %s\n",err->message);
ali@69	447	g_printerr("Use \"%s --help\" for help\n",(*argv)[0]);
ali@69	448	exit(1);
ali@69	449	}
ali@177	450	if (typo_compat)
ali@69	451	pswit[TYPO_SWITCH]=!pswit[TYPO_SWITCH];
ali@177	452	if (paranoid_compat)
ali@177	453	{
ali@177	454	pswit[PARANOID_SWITCH]=!pswit[PARANOID_SWITCH];
ali@177	455	pswit[TYPO_SWITCH]=!pswit[TYPO_SWITCH];
ali@177	456	}
ali@40	457	/*
ali@40	458	* Web uploads - for the moment, this is really just a placeholder
ali@40	459	* until we decide what processing we really want to do on web uploads
ali@40	460	*/
ali@40	461	if (pswit[WEB_SWITCH])
ali@40	462	{
ali@40	463	/* specific override for web uploads */
ali@69	464	pswit[ECHO_SWITCH]=TRUE;
ali@69	465	pswit[SQUOTE_SWITCH]=FALSE;
ali@69	466	pswit[TYPO_SWITCH]=TRUE;
ali@69	467	pswit[QPARA_SWITCH]=FALSE;
ali@69	468	pswit[PARANOID_SWITCH]=TRUE;
ali@69	469	pswit[LINE_END_SWITCH]=FALSE;
ali@69	470	pswit[OVERVIEW_SWITCH]=FALSE;
ali@69	471	pswit[STDOUT_SWITCH]=FALSE;
ali@69	472	pswit[HEADER_SWITCH]=TRUE;
ali@69	473	pswit[VERBOSE_SWITCH]=FALSE;
ali@69	474	pswit[MARKUP_SWITCH]=FALSE;
ali@69	475	pswit[USERTYPO_SWITCH]=FALSE;
ali@69	476	pswit[DP_SWITCH]=FALSE;
ali@40	477	}
ali@177	478	if (pswit[DUMP_CONFIG_SWITCH])
ali@177	479	{
ali@177	480	dump_config();
ali@177	481	exit(0);
ali@177	482	}
ali@177	483	if (pswit[OVERVIEW_SWITCH])
ali@177	484	/* just print summary; don't echo */
ali@177	485	pswit[ECHO_SWITCH]=FALSE;
ali@69	486	if (*argc<2)
ali@40	487	{
ali@69	488	proghelp(context);
ali@69	489	exit(1);
ali@40	490	}
ali@69	491	g_option_context_free(context);
ali@69	492	}
ali@69	493
ali@69	494	/*
ali@69	495	* read_user_scannos:
ali@69	496	*
ali@69	497	* Read in the user-defined stealth scanno list.
ali@69	498	*/
ali@69	499	void read_user_scannos(void)
ali@69	500	{
ali@69	501	GError *err=NULL;
ali@69	502	gchar *usertypo_file;
ali@69	503	gboolean okay;
ali@69	504	int i;
ali@70	505	gsize len,nb;
ali@70	506	gchar contents,utf8,**lines;
ali@69	507	usertypo_file=g_strdup("bookloupe.typ");
ali@69	508	okay=file_get_contents_text(usertypo_file,&contents,&len,&err);
ali@69	509	if (g_error_matches(err,G_FILE_ERROR,G_FILE_ERROR_NOENT))
ali@69	510	{
ali@69	511	g_clear_error(&err);
ali@69	512	g_free(usertypo_file);
ali@69	513	usertypo_file=g_build_filename(running_from,"bookloupe.typ",NULL);
ali@69	514	okay=file_get_contents_text(usertypo_file,&contents,&len,&err);
ali@69	515	}
ali@69	516	if (g_error_matches(err,G_FILE_ERROR,G_FILE_ERROR_NOENT))
ali@69	517	{
ali@69	518	g_clear_error(&err);
ali@69	519	g_free(usertypo_file);
ali@69	520	usertypo_file=g_strdup("gutcheck.typ");
ali@69	521	okay=file_get_contents_text(usertypo_file,&contents,&len,&err);
ali@69	522	}
ali@69	523	if (g_error_matches(err,G_FILE_ERROR,G_FILE_ERROR_NOENT))
ali@69	524	{
ali@69	525	g_clear_error(&err);
ali@69	526	g_free(usertypo_file);
ali@69	527	usertypo_file=g_build_filename(running_from,"gutcheck.typ",NULL);
ali@69	528	okay=file_get_contents_text(usertypo_file,&contents,&len,&err);
ali@69	529	}
ali@69	530	if (g_error_matches(err,G_FILE_ERROR,G_FILE_ERROR_NOENT))
ali@69	531	{
ali@69	532	g_free(usertypo_file);
ali@70	533	g_print(" --> I couldn't find bookloupe.typ "
ali@69	534	"-- proceeding without user typos.\n");
ali@69	535	return;
ali@69	536	}
ali@69	537	else if (!okay)
ali@69	538	{
ali@69	539	fprintf(stderr,"%s: %s\n",usertypo_file,err->message);
ali@69	540	g_free(usertypo_file);
ali@69	541	g_clear_error(&err);
ali@69	542	exit(1);
ali@69	543	}
ali@72	544	if (g_utf8_validate(contents,len,NULL))
ali@72	545	utf8=g_utf8_normalize(contents,len,G_NORMALIZE_DEFAULT_COMPOSE);
ali@72	546	else
ali@72	547	utf8=g_convert(contents,len,"UTF-8","WINDOWS-1252",NULL,&nb,NULL);
ali@70	548	g_free(contents);
ali@70	549	lines=g_strsplit_set(utf8,"\r\n",0);
ali@70	550	g_free(utf8);
ali@69	551	usertypo=g_tree_new_full((GCompareDataFunc)strcmp,NULL,g_free,NULL);
ali@69	552	for (i=0;lines[i];i++)
ali@69	553	if ((unsigned char )lines[i]>'!')
ali@69	554	g_tree_insert(usertypo,lines[i],GINT_TO_POINTER(1));
ali@69	555	else
ali@69	556	g_free(lines[i]);
ali@69	557	g_free(lines);
ali@69	558	}
ali@69	559
ali@69	560	/*
ali@69	561	* read_etext:
ali@69	562	*
ali@69	563	* Read an etext returning a newly allocated string containing the file
ali@69	564	* contents or NULL on error.
ali@69	565	*/
ali@69	566	gchar read_etext(const char filename,GError **err)
ali@69	567	{
ali@76	568	GError *tmp_err=NULL;
ali@70	569	gchar contents,utf8;
ali@76	570	gsize len,bytes_read,bytes_written;
ali@76	571	int i,line,col;
ali@69	572	if (!g_file_get_contents(filename,&contents,&len,err))
ali@69	573	return NULL;
ali@72	574	if (g_utf8_validate(contents,len,NULL))
ali@72	575	{
ali@72	576	utf8=g_utf8_normalize(contents,len,G_NORMALIZE_DEFAULT_COMPOSE);
ali@72	577	g_set_print_handler(print_as_utf_8);
ali@73	578	#ifdef __WIN32__
ali@73	579	SetConsoleOutputCP(CP_UTF8);
ali@73	580	#endif
ali@72	581	}
ali@72	582	else
ali@72	583	{
ali@76	584	utf8=g_convert(contents,len,"UTF-8","WINDOWS-1252",&bytes_read,
ali@76	585	&bytes_written,&tmp_err);
ali@76	586	if (g_error_matches(tmp_err,G_CONVERT_ERROR,
ali@76	587	G_CONVERT_ERROR_ILLEGAL_SEQUENCE))
ali@76	588	{
ali@76	589	line=col=1;
ali@76	590	for(i=0;i<bytes_read;i++)
ali@76	591	if (contents[i]=='\n')
ali@76	592	{
ali@76	593	line++;
ali@76	594	col=1;
ali@76	595	}
ali@76	596	else if (contents[i]!='\r')
ali@76	597	col++;
ali@76	598	g_set_error(err,G_CONVERT_ERROR,G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
ali@76	599	"Input conversion failed. Byte %d at line %d, column %d is not a "
ali@76	600	"valid Windows-1252 character",
ali@76	601	((unsigned char *)contents)[bytes_read],line,col);
ali@76	602	}
ali@76	603	else if (tmp_err)
ali@76	604	g_propagate_error(err,tmp_err);
ali@72	605	g_set_print_handler(print_as_windows_1252);
ali@73	606	#ifdef __WIN32__
ali@73	607	SetConsoleOutputCP(1252);
ali@73	608	#endif
ali@72	609	}
ali@70	610	g_free(contents);
ali@70	611	return utf8;
ali@69	612	}
ali@69	613
ali@73	614	void cleanup_on_exit(void)
ali@73	615	{
ali@73	616	#ifdef __WIN32__
ali@73	617	SetConsoleOutputCP(saved_cp);
ali@73	618	#endif
ali@73	619	}
ali@73	620
ali@69	621	int main(int argc,char **argv)
ali@69	622	{
ali@73	623	#ifdef __WIN32__
ali@73	624	atexit(cleanup_on_exit);
ali@73	625	saved_cp=GetConsoleOutputCP();
ali@73	626	#endif
ali@69	627	running_from=g_path_get_dirname(argv[0]);
ali@177	628	/* Paranoid checking is turned OFF, not on, by its switch */
ali@177	629	pswit[PARANOID_SWITCH]=TRUE;
ali@177	630	/* if running in paranoid mode, typo checks default to enabled */
ali@177	631	pswit[TYPO_SWITCH]=TRUE;
ali@177	632	/* Line-end checking is turned OFF, not on, by its switch */
ali@177	633	pswit[LINE_END_SWITCH]=TRUE;
ali@177	634	/* Echoing is turned OFF, not on, by its switch */
ali@177	635	pswit[ECHO_SWITCH]=TRUE;
ali@177	636	parse_config_file();
ali@69	637	parse_options(&argc,&argv);
ali@40	638	if (pswit[USERTYPO_SWITCH])
ali@69	639	read_user_scannos();
ali@40	640	fprintf(stderr,"bookloupe: Check and report on an e-text\n");
ali@69	641	procfile(argv[1]);
ali@40	642	if (pswit[OVERVIEW_SWITCH])
ali@40	643	{
ali@70	644	g_print(" Checked %ld lines of %ld (head+foot = %ld)\n\n",
ali@40	645	checked_linecnt,linecnt,linecnt-checked_linecnt);
ali@70	646	g_print(" --------------- Queries found --------------\n");
ali@68	647	if (cnt_long)
ali@70	648	g_print(" Long lines: %14ld\n",cnt_long);
ali@68	649	if (cnt_short)
ali@70	650	g_print(" Short lines: %14ld\n",cnt_short);
ali@68	651	if (cnt_lineend)
ali@70	652	g_print(" Line-end problems: %14ld\n",cnt_lineend);
ali@68	653	if (cnt_word)
ali@70	654	g_print(" Common typos: %14ld\n",cnt_word);
ali@142	655	if (cnt_quote)
ali@142	656	g_print(" Unmatched quotes: %14ld\n",cnt_quote);
ali@68	657	if (cnt_brack)
ali@70	658	g_print(" Unmatched brackets: %14ld\n",cnt_brack);
ali@68	659	if (cnt_bin)
ali@70	660	g_print(" Non-ASCII characters: %14ld\n",cnt_bin);
ali@68	661	if (cnt_odd)
ali@70	662	g_print(" Proofing characters: %14ld\n",cnt_odd);
ali@68	663	if (cnt_punct)
ali@70	664	g_print(" Punctuation & spacing queries: %14ld\n",cnt_punct);
ali@68	665	if (cnt_dash)
ali@70	666	g_print(" Non-standard dashes: %14ld\n",cnt_dash);
ali@68	667	if (cnt_html)
ali@70	668	g_print(" Possible HTML tags: %14ld\n",cnt_html);
ali@70	669	g_print("\n");
ali@70	670	g_print(" TOTAL QUERIES %14ld\n",
ali@142	671	cnt_quote+cnt_brack+cnt_bin+cnt_odd+cnt_long+cnt_short+cnt_punct+
ali@142	672	cnt_dash+cnt_word+cnt_html+cnt_lineend);
ali@40	673	}
ali@69	674	g_free(running_from);
ali@69	675	if (usertypo)
ali@69	676	g_tree_unref(usertypo);
ali@177	677	if (config)
ali@177	678	g_key_file_free(config);
ali@40	679	return 0;
ali@0	680	}
ali@0	681
ali@147	682	void count_dashes(const char line,const char dash,
ali@147	683	struct dash_results *results)
ali@147	684	{
ali@147	685	int i;
ali@147	686	gchar **tokens;
ali@147	687	gunichar pc,nc;
ali@147	688	gboolean spaced=FALSE,unspaced=FALSE,spaced2=FALSE;
ali@147	689	if (!*line)
ali@147	690	return;
ali@147	691	tokens=g_strsplit(line,dash,0);
ali@147	692	if (tokens[1])
ali@147	693	results->base++;
ali@147	694	for(i=1;tokens[i];i++)
ali@147	695	{
ali@147	696	pc=g_utf8_get_char(g_utf8_prev_char(tokens[i-1]+strlen(tokens[i-1])));
ali@147	697	nc=g_utf8_get_char(tokens[i]);
ali@147	698	if (g_unichar_isspace(pc) \|\| g_unichar_isspace(nc))
ali@147	699	spaced=TRUE;
ali@147	700	if (g_unichar_isspace(pc) && g_unichar_isspace(nc))
ali@147	701	spaced2=TRUE;
ali@147	702	else if (!g_unichar_isspace(pc) && !g_unichar_isspace(nc))
ali@147	703	unspaced=TRUE;
ali@147	704	}
ali@147	705	if (spaced)
ali@147	706	results->space++;
ali@147	707	if (spaced2)
ali@147	708	/* count of lines with em-dashes with spaces both sides */
ali@147	709	results->non_PG_space++;
ali@147	710	if (unspaced)
ali@147	711	/* count of lines with PG-type em-dashes with no spaces */
ali@147	712	results->PG_space++;
ali@147	713	g_strfreev(tokens);
ali@147	714	}
ali@147	715
ali@40	716	/*
ali@41	717	* first_pass:
ali@40	718	*
ali@41	719	* Run a first pass - verify that it's a valid PG
ali@41	720	* file, decide whether to report some things that
ali@41	721	* occur many times in the text like long or short
ali@41	722	* lines, non-standard dashes, etc.
ali@40	723	*/
ali@69	724	struct first_pass_results first_pass(const char etext)
ali@0	725	{
ali@70	726	gunichar laststart=CHAR_SPACE;
ali@54	727	const char *s;
ali@69	728	gchar *lc_line;
ali@70	729	int i,j,lbytes,llen;
ali@69	730	gchar **lines;
ali@41	731	unsigned int lastlen=0,lastblen=0;
ali@41	732	long spline=0,nspline=0;
ali@41	733	static struct first_pass_results results={0};
ali@147	734	struct dash_results tmp_dash_results;
ali@69	735	gchar *inword;
ali@142	736	QuoteClass qc;
ali@69	737	lines=g_strsplit(etext,"\n",0);
ali@69	738	for (j=0;lines[j];j++)
ali@40	739	{
ali@70	740	lbytes=strlen(lines[j]);
ali@82	741	while (lbytes>0 && lines[j][lbytes-1]=='\r')
ali@70	742	lines[j][--lbytes]='\0';
ali@70	743	llen=g_utf8_strlen(lines[j],lbytes);
ali@68	744	linecnt++;
ali@69	745	if (strstr(lines[j],"*END") && strstr(lines[j],"SMALL PRINT") &&
ali@69	746	(strstr(lines[j],"PUBLIC DOMAIN") \|\| strstr(lines[j],"COPYRIGHT")))
ali@40	747	{
ali@68	748	if (spline)
ali@70	749	g_print(" --> Duplicate header?\n");
ali@68	750	spline=linecnt+1; /* first line of non-header text, that is */
ali@40	751	}
ali@69	752	if (!strncmp(lines[j],"*** START",9) &&
ali@69	753	strstr(lines[j],"PROJECT GUTENBERG"))
ali@40	754	{
ali@68	755	if (nspline)
ali@70	756	g_print(" --> Duplicate header?\n");
ali@68	757	nspline=linecnt+1; /* first line of non-header text, that is */
ali@40	758	}
ali@68	759	if (spline \|\| nspline)
ali@40	760	{
ali@70	761	lc_line=g_utf8_strdown(lines[j],lbytes);
ali@69	762	if (strstr(lc_line,"end") && strstr(lc_line,"project gutenberg"))
ali@40	763	{
ali@69	764	if (strstr(lc_line,"end")<strstr(lc_line,"project gutenberg"))
ali@40	765	{
ali@68	766	if (results.footerline)
ali@40	767	{
ali@40	768	/* it's an old-form header - we can detect duplicates */
ali@68	769	if (!nspline)
ali@70	770	g_print(" --> Duplicate footer?\n");
ali@40	771	}
ali@68	772	else
ali@68	773	results.footerline=linecnt;
ali@40	774	}
ali@40	775	}
ali@69	776	g_free(lc_line);
ali@40	777	}
ali@68	778	if (spline)
ali@41	779	results.firstline=spline;
ali@68	780	if (nspline)
ali@41	781	results.firstline=nspline; /* override with new */
ali@68	782	if (results.footerline)
ali@40	783	continue; /* don't count the boilerplate in the footer */
ali@68	784	results.totlen+=llen;
ali@70	785	for (s=lines[j];*s;s=g_utf8_next_char(s))
ali@40	786	{
ali@70	787	if (g_utf8_get_char(s)>127)
ali@41	788	results.binlen++;
ali@70	789	if (g_unichar_isalpha(g_utf8_get_char(s)))
ali@41	790	results.alphalen++;
ali@142	791	if (s>lines[j])
ali@142	792	{
ali@142	793	if (CHAR_IS_DQUOTE(g_utf8_get_char(s)))
ali@142	794	qc=QUOTE_CLASS(g_utf8_get_char(s));
ali@142	795	else
ali@142	796	qc=INVALID_QUOTE;
ali@142	797	if ((qc==CLOSING_QUOTE \|\| qc==NEUTRAL_QUOTE) &&
ali@147	798	g_unichar_isalpha(g_utf8_get_char(g_utf8_prev_char(s))))
ali@142	799	results.endquote_count++;
ali@142	800	}
ali@40	801	}
ali@69	802	if (llen>2 && lastlen>2 && lastlen<SHORTEST_PG_LINE && lastblen>2 &&
ali@69	803	lastblen>SHORTEST_PG_LINE && laststart!=CHAR_SPACE)
ali@41	804	results.shortline++;
ali@70	805	if (lbytes>0 &&
ali@70	806	g_utf8_get_char(g_utf8_prev_char(lines[j]+lbytes))<=CHAR_SPACE)
ali@40	807	cnt_spacend++;
ali@69	808	if (strstr(lines[j],".,"))
ali@41	809	results.dotcomma++;
ali@68	810	/* only count ast lines for ignoring purposes where there is */
ali@68	811	/* locase text on the line */
ali@69	812	if (strchr(lines[j],'*'))
ali@40	813	{
ali@70	814	for (s=lines[j];*s;s=g_utf8_next_char(s))
ali@70	815	if (g_unichar_islower(g_utf8_get_char(s)))
ali@68	816	break;
ali@70	817	if (*s)
ali@41	818	results.astline++;
ali@40	819	}
ali@69	820	if (strchr(lines[j],'/'))
ali@68	821	results.fslashline++;
ali@82	822	if (lbytes>0)
ali@82	823	{
ali@82	824	for (s=g_utf8_prev_char(lines[j]+lbytes);
ali@82	825	s>lines[j] && g_utf8_get_char(s)<=CHAR_SPACE;
ali@82	826	s=g_utf8_prev_char(s))
ali@82	827	;
ali@82	828	if (s>g_utf8_next_char(lines[j]) && g_utf8_get_char(s)=='-' &&
ali@82	829	g_utf8_get_char(g_utf8_prev_char(s))!='-')
ali@82	830	results.hyphens++;
ali@82	831	}
ali@68	832	if (llen>LONGEST_PG_LINE)
ali@41	833	results.longline++;
ali@68	834	if (llen>WAY_TOO_LONG)
ali@41	835	results.verylongline++;
ali@69	836	if (strchr(lines[j],'<') && strchr(lines[j],'>'))
ali@40	837	{
ali@69	838	i=(int)(strchr(lines[j],'>')-strchr(lines[j],'<')+1);
ali@68	839	if (i>0)
ali@68	840	results.htmcount++;
ali@69	841	if (strstr(lines[j],"<i>"))
ali@41	842	results.htmcount+=4; /* bonus marks! */
ali@40	843	}
ali@68	844	/* Check for spaced em-dashes */
ali@147	845	memset(&tmp_dash_results,0,sizeof(tmp_dash_results));
ali@147	846	count_dashes(lines[j],"--",&tmp_dash_results);
ali@147	847	count_dashes(lines[j],"—",&tmp_dash_results);
ali@147	848	if (tmp_dash_results.base)
ali@147	849	results.emdash.base++;
ali@147	850	if (tmp_dash_results.non_PG_space)
ali@147	851	results.emdash.non_PG_space++;
ali@147	852	if (tmp_dash_results.PG_space)
ali@147	853	results.emdash.PG_space++;
ali@69	854	for (s=lines[j];*s;)
ali@40	855	{
ali@69	856	inword=getaword(&s);
ali@68	857	if (!strcmp(inword,"hij") \|\| !strcmp(inword,"niet"))
ali@68	858	results.Dutchcount++;
ali@68	859	if (!strcmp(inword,"dans") \|\| !strcmp(inword,"avec"))
ali@68	860	results.Frenchcount++;
ali@68	861	if (!strcmp(inword,"0") \|\| !strcmp(inword,"1"))
ali@68	862	results.standalone_digit++;
ali@69	863	g_free(inword);
ali@40	864	}
ali@68	865	/* Check for spaced dashes */
ali@69	866	if (strstr(lines[j]," -") && *(strstr(lines[j]," -")+2)!='-')
ali@41	867	results.spacedash++;
ali@68	868	lastblen=lastlen;
ali@69	869	lastlen=llen;
ali@69	870	laststart=lines[j][0];
ali@40	871	}
ali@69	872	g_strfreev(lines);
ali@41	873	return &results;
ali@41	874	}
ali@41	875
ali@42	876	/*
ali@42	877	* report_first_pass:
ali@42	878	*
ali@42	879	* Make some snap decisions based on the first pass results.
ali@42	880	*/
ali@42	881	struct warnings report_first_pass(struct first_pass_results results)
ali@42	882	{
ali@42	883	static struct warnings warnings={0};
ali@42	884	if (cnt_spacend>0)
ali@70	885	g_print(" --> %ld lines in this file have white space at end\n",
ali@42	886	cnt_spacend);
ali@42	887	warnings.dotcomma=1;
ali@42	888	if (results->dotcomma>5)
ali@42	889	{
ali@68	890	warnings.dotcomma=0;
ali@70	891	g_print(" --> %ld lines in this file contain '.,'. "
ali@42	892	"Not reporting them.\n",results->dotcomma);
ali@42	893	}
ali@42	894	/*
ali@42	895	* If more than 50 lines, or one-tenth, are short,
ali@42	896	* don't bother reporting them.
ali@42	897	*/
ali@42	898	warnings.shortline=1;
ali@42	899	if (results->shortline>50 \|\| results->shortline*10>linecnt)
ali@42	900	{
ali@68	901	warnings.shortline=0;
ali@70	902	g_print(" --> %ld lines in this file are short. "
ali@42	903	"Not reporting short lines.\n",results->shortline);
ali@42	904	}
ali@42	905	/*
ali@42	906	* If more than 50 lines, or one-tenth, are long,
ali@42	907	* don't bother reporting them.
ali@42	908	*/
ali@42	909	warnings.longline=1;
ali@42	910	if (results->longline>50 \|\| results->longline*10>linecnt)
ali@42	911	{
ali@68	912	warnings.longline=0;
ali@70	913	g_print(" --> %ld lines in this file are long. "
ali@42	914	"Not reporting long lines.\n",results->longline);
ali@42	915	}
ali@42	916	/* If more than 10 lines contain asterisks, don't bother reporting them. */
ali@42	917	warnings.ast=1;
ali@42	918	if (results->astline>10)
ali@42	919	{
ali@68	920	warnings.ast=0;
ali@70	921	g_print(" --> %ld lines in this file contain asterisks. "
ali@42	922	"Not reporting them.\n",results->astline);
ali@42	923	}
ali@42	924	/*
ali@42	925	* If more than 10 lines contain forward slashes,
ali@42	926	* don't bother reporting them.
ali@42	927	*/
ali@42	928	warnings.fslash=1;
ali@42	929	if (results->fslashline>10)
ali@42	930	{
ali@68	931	warnings.fslash=0;
ali@70	932	g_print(" --> %ld lines in this file contain forward slashes. "
ali@42	933	"Not reporting them.\n",results->fslashline);
ali@42	934	}
ali@42	935	/*
ali@42	936	* If more than 20 lines contain unpunctuated endquotes,
ali@42	937	* don't bother reporting them.
ali@42	938	*/
ali@42	939	warnings.endquote=1;
ali@42	940	if (results->endquote_count>20)
ali@42	941	{
ali@68	942	warnings.endquote=0;
ali@70	943	g_print(" --> %ld lines in this file contain unpunctuated endquotes. "
ali@42	944	"Not reporting them.\n",results->endquote_count);
ali@42	945	}
ali@42	946	/*
ali@42	947	* If more than 15 lines contain standalone digits,
ali@42	948	* don't bother reporting them.
ali@42	949	*/
ali@42	950	warnings.digit=1;
ali@42	951	if (results->standalone_digit>10)
ali@42	952	{
ali@68	953	warnings.digit=0;
ali@70	954	g_print(" --> %ld lines in this file contain standalone 0s and 1s. "
ali@42	955	"Not reporting them.\n",results->standalone_digit);
ali@42	956	}
ali@42	957	/*
ali@42	958	* If more than 20 lines contain hyphens at end,
ali@42	959	* don't bother reporting them.
ali@42	960	*/
ali@42	961	warnings.hyphen=1;
ali@42	962	if (results->hyphens>20)
ali@42	963	{
ali@68	964	warnings.hyphen=0;
ali@70	965	g_print(" --> %ld lines in this file have hyphens at end. "
ali@42	966	"Not reporting them.\n",results->hyphens);
ali@42	967	}
ali@42	968	if (results->htmcount>20 && !pswit[MARKUP_SWITCH])
ali@42	969	{
ali@70	970	g_print(" --> Looks like this is HTML. Switching HTML mode ON.\n");
ali@68	971	pswit[MARKUP_SWITCH]=1;
ali@42	972	}
ali@42	973	if (results->verylongline>0)
ali@70	974	g_print(" --> %ld lines in this file are VERY long!\n",
ali@42	975	results->verylongline);
ali@42	976	/*
ali@42	977	* If there are more non-PG spaced dashes than PG em-dashes,
ali@42	978	* assume it's deliberate.
ali@42	979	* Current PG guidelines say don't use them, but older texts do,
ali@42	980	* and some people insist on them whatever the guidelines say.
ali@42	981	*/
ali@42	982	warnings.dash=1;
ali@147	983	if (results->spacedash+results->emdash.non_PG_space>
ali@147	984	results->emdash.PG_space)
ali@42	985	{
ali@68	986	warnings.dash=0;
ali@70	987	g_print(" --> There are %ld spaced dashes and em-dashes. "
ali@42	988	"Not reporting them.\n",
ali@147	989	results->spacedash+results->emdash.non_PG_space);
ali@42	990	}
ali@42	991	/* If more than a quarter of characters are hi-bit, bug out. */
ali@42	992	warnings.bin=1;
ali@42	993	if (results->binlen*4>results->totlen)
ali@42	994	{
ali@70	995	g_print(" --> This file does not appear to be ASCII. "
ali@42	996	"Terminating. Best of luck with it!\n");
ali@68	997	exit(1);
ali@42	998	}
ali@42	999	if (results->alphalen*4<results->totlen)
ali@42	1000	{
ali@70	1001	g_print(" --> This file does not appear to be text. "
ali@42	1002	"Terminating. Best of luck with it!\n");
ali@68	1003	exit(1);
ali@42	1004	}
ali@42	1005	if (results->binlen*100>results->totlen \|\| results->binlen>100)
ali@42	1006	{
ali@70	1007	g_print(" --> There are a lot of foreign letters here. "
ali@42	1008	"Not reporting them.\n");
ali@68	1009	warnings.bin=0;
ali@42	1010	}
ali@69	1011	warnings.isDutch=FALSE;
ali@42	1012	if (results->Dutchcount>50)
ali@42	1013	{
ali@69	1014	warnings.isDutch=TRUE;
ali@70	1015	g_print(" --> This looks like Dutch - "
ali@42	1016	"switching off dashes and warnings for 's Middags case.\n");
ali@42	1017	}
ali@69	1018	warnings.isFrench=FALSE;
ali@42	1019	if (results->Frenchcount>50)
ali@42	1020	{
ali@69	1021	warnings.isFrench=TRUE;
ali@70	1022	g_print(" --> This looks like French - "
ali@42	1023	"switching off some doublepunct.\n");
ali@42	1024	}
ali@42	1025	if (results->firstline && results->footerline)
ali@70	1026	g_print(" The PG header and footer appear to be already on.\n");
ali@42	1027	else
ali@42	1028	{
ali@68	1029	if (results->firstline)
ali@70	1030	g_print(" The PG header is on - no footer.\n");
ali@68	1031	if (results->footerline)
ali@70	1032	g_print(" The PG footer is on - no header.\n");
ali@42	1033	}
ali@70	1034	g_print("\n");
ali@42	1035	if (pswit[VERBOSE_SWITCH])
ali@42	1036	{
ali@68	1037	warnings.bin=1;
ali@68	1038	warnings.shortline=1;
ali@68	1039	warnings.dotcomma=1;
ali@68	1040	warnings.longline=1;
ali@68	1041	warnings.dash=1;
ali@68	1042	warnings.digit=1;
ali@68	1043	warnings.ast=1;
ali@68	1044	warnings.fslash=1;
ali@68	1045	warnings.hyphen=1;
ali@68	1046	warnings.endquote=1;
ali@70	1047	g_print(" * Verbose output is ON -- you asked for it! *\n");
ali@42	1048	}
ali@42	1049	if (warnings.isDutch)
ali@68	1050	warnings.dash=0;
ali@42	1051	if (results->footerline>0 && results->firstline>0 &&
ali@42	1052	results->footerline>results->firstline &&
ali@42	1053	results->footerline-results->firstline<100)
ali@42	1054	{
ali@70	1055	g_print(" --> I don't really know where this text starts. \n");
ali@70	1056	g_print(" There are no reference points.\n");
ali@70	1057	g_print(" I'm going to have to report the header and footer "
ali@42	1058	"as well.\n");
ali@68	1059	results->firstline=0;
ali@42	1060	}
ali@42	1061	return &warnings;
ali@42	1062	}
ali@42	1063
ali@43	1064	/*
ali@43	1065	* analyse_quotes:
ali@43	1066	*
ali@43	1067	* Look along the line, accumulate the count of quotes, and see
ali@43	1068	* if this is an empty line - i.e. a line with nothing on it
ali@43	1069	* but spaces.
ali@43	1070	* If line has just spaces, period, * and/or - on it, don't
ali@43	1071	* count it, since empty lines with asterisks or dashes to
ali@43	1072	* separate sections are common.
ali@43	1073	*
ali@69	1074	* Returns: TRUE if the line is empty.
ali@43	1075	*/
ali@164	1076	gboolean analyse_quotes(const char aline,struct counters counters)
ali@43	1077	{
ali@68	1078	int guessquote=0;
ali@69	1079	/* assume the line is empty until proven otherwise */
ali@69	1080	gboolean isemptyline=TRUE;
ali@70	1081	const char s=aline,sprev,*snext;
ali@70	1082	gunichar c;
ali@70	1083	sprev=NULL;
ali@142	1084	GError *tmp_err=NULL;
ali@43	1085	while (*s)
ali@43	1086	{
ali@70	1087	snext=g_utf8_next_char(s);
ali@70	1088	c=g_utf8_get_char(s);
ali@142	1089	if (CHAR_IS_DQUOTE(c))
ali@142	1090	(void)count_quote(counters,c,QUOTE_CLASS(c),&tmp_err);
ali@142	1091	else if (CHAR_IS_SQUOTE(c) && pswit[SQUOTE_SWITCH])
ali@43	1092	{
ali@43	1093	if (s==aline)
ali@43	1094	{
ali@43	1095	/*
ali@142	1096	* At start of line, it can only be a quotation mark.
ali@43	1097	* Hardcode a very common exception!
ali@43	1098	*/
ali@70	1099	if (!g_str_has_prefix(snext,"tis") &&
ali@70	1100	!g_str_has_prefix(snext,"Tis"))
ali@142	1101	(void)count_quote(counters,c,NEUTRAL_QUOTE,&tmp_err);
ali@43	1102	}
ali@70	1103	else if (g_unichar_isalpha(g_utf8_get_char(sprev)) &&
ali@70	1104	g_unichar_isalpha(g_utf8_get_char(snext)))
ali@43	1105	/* Do nothing! it's definitely an apostrophe, not a quote */
ali@43	1106	;
ali@43	1107	/* it's outside a word - let's check it out */
ali@99	1108	else if (c==CHAR_OPEN_SQUOTE \|\| c==CHAR_LS_QUOTE \|\|
ali@70	1109	g_unichar_isalpha(g_utf8_get_char(snext)))
ali@43	1110	{
ali@142	1111	/* certainly looks like a quotation mark */
ali@70	1112	if (!g_str_has_prefix(snext,"tis") &&
ali@70	1113	!g_str_has_prefix(snext,"Tis"))
ali@43	1114	/* hardcode a very common exception! */
ali@142	1115	{
ali@142	1116	if (strchr(".?!,;:",g_utf8_get_char(sprev)))
ali@142	1117	(void)count_quote(counters,c,NEUTRAL_QUOTE,&tmp_err);
ali@142	1118	else
ali@142	1119	(void)count_quote(counters,c,OPENING_QUOTE,&tmp_err);
ali@142	1120	}
ali@43	1121	}
ali@43	1122	else
ali@43	1123	{
ali@142	1124	/* now - is it a quotation mark? */
ali@43	1125	guessquote=0; /* accumulate clues */
ali@70	1126	if (g_unichar_isalpha(g_utf8_get_char(sprev)))
ali@43	1127	{
ali@43	1128	/* it follows a letter - could be either */
ali@43	1129	guessquote++;
ali@70	1130	if (g_utf8_get_char(sprev)=='s')
ali@43	1131	{
ali@43	1132	/* looks like a plural apostrophe */
ali@43	1133	guessquote-=3;
ali@70	1134	if (g_utf8_get_char(snext)==CHAR_SPACE)
ali@70	1135	/* bonus marks! */
ali@43	1136	guessquote-=2;
ali@43	1137	}
ali@142	1138	if (innermost_quote_matches(counters,c))
ali@142	1139	/*
ali@142	1140	* Give it the benefit of some doubt,
ali@142	1141	* if a squote is already open.
ali@142	1142	*/
ali@142	1143	guessquote++;
ali@142	1144	else
ali@142	1145	guessquote--;
ali@142	1146	if (guessquote>=0)
ali@142	1147	(void)count_quote(counters,c,CLOSING_QUOTE,&tmp_err);
ali@43	1148	}
ali@43	1149	else
ali@142	1150	/* no adjacent letter - it must be a quote of some kind */
ali@142	1151	(void)count_quote(counters,c,NEUTRAL_QUOTE,&tmp_err);
ali@43	1152	}
ali@43	1153	}
ali@142	1154	if (tmp_err)
ali@142	1155	{
ali@142	1156	if (pswit[ECHO_SWITCH])
ali@142	1157	g_print("\n%s\n",aline);
ali@142	1158	if (!pswit[OVERVIEW_SWITCH])
ali@142	1159	g_print(" Line %ld column %ld - %s\n",
ali@142	1160	linecnt,g_utf8_pointer_to_offset(aline,s)+1,tmp_err->message);
ali@142	1161	g_clear_error(&tmp_err);
ali@142	1162	}
ali@70	1163	if (c!=CHAR_SPACE && c!='-' && c!='.' && c!=CHAR_ASTERISK &&
ali@70	1164	c!='\r' && c!='\n')
ali@69	1165	isemptyline=FALSE; /* ignore lines like * * * as spacers */
ali@70	1166	if (c==CHAR_UNDERSCORE)
ali@43	1167	counters->c_unders++;
ali@103	1168	if (c==CHAR_OPEN_SBRACK)
ali@103	1169	{
ali@103	1170	if (!matching_difference(counters,COUNTER_ILLUSTRATION) &&
ali@103	1171	!matching_difference(counters,c) && s==aline &&
ali@103	1172	g_str_has_prefix(s,"[Illustration:"))
ali@103	1173	increment_matching(counters,COUNTER_ILLUSTRATION,TRUE);
ali@103	1174	else
ali@103	1175	increment_matching(counters,c,TRUE);
ali@103	1176	}
ali@103	1177	else if (c==CHAR_OPEN_CBRACK \|\| c==CHAR_OPEN_RBRACK)
ali@99	1178	increment_matching(counters,c,TRUE);
ali@103	1179	if (c==CHAR_CLOSE_SBRACK)
ali@103	1180	{
ali@103	1181	if (!matching_count(counters,COUNTER_ILLUSTRATION,FALSE) &&
ali@103	1182	!matching_difference(counters,c) && !*snext)
ali@103	1183	increment_matching(counters,COUNTER_ILLUSTRATION,FALSE);
ali@103	1184	else
ali@103	1185	increment_matching(counters,c,FALSE);
ali@103	1186	}
ali@103	1187	else if (c==CHAR_CLOSE_CBRACK \|\| c==CHAR_CLOSE_RBRACK)
ali@99	1188	increment_matching(counters,c,FALSE);
ali@70	1189	sprev=s;
ali@70	1190	s=snext;
ali@43	1191	}
ali@43	1192	return isemptyline;
ali@43	1193	}
ali@43	1194
ali@41	1195	/*
ali@67	1196	* check_for_control_characters:
ali@67	1197	*
ali@67	1198	* Check for invalid or questionable characters in the line
ali@67	1199	* Anything above 127 is invalid for plain ASCII, and
ali@67	1200	* non-printable control characters should also be flagged.
ali@67	1201	* Tabs should generally not be there.
ali@67	1202	*/
ali@67	1203	void check_for_control_characters(const char *aline)
ali@67	1204	{
ali@70	1205	gunichar c;
ali@67	1206	const char *s;
ali@70	1207	for (s=aline;*s;s=g_utf8_next_char(s))
ali@67	1208	{
ali@70	1209	c=g_utf8_get_char(s);
ali@67	1210	if (c<CHAR_SPACE && c!=CHAR_LF && c!=CHAR_CR && c!=CHAR_TAB)
ali@67	1211	{
ali@67	1212	if (pswit[ECHO_SWITCH])
ali@70	1213	g_print("\n%s\n",aline);
ali@67	1214	if (!pswit[OVERVIEW_SWITCH])
ali@70	1215	g_print(" Line %ld column %ld - Control character %u\n",
ali@70	1216	linecnt,g_utf8_pointer_to_offset(s,aline)+1,c);
ali@67	1217	else
ali@67	1218	cnt_bin++;
ali@67	1219	}
ali@67	1220	}
ali@67	1221	}
ali@67	1222
ali@67	1223	/*
ali@44	1224	* check_for_odd_characters:
ali@44	1225	*
ali@44	1226	* Check for binary and other odd characters.
ali@44	1227	*/
ali@44	1228	void check_for_odd_characters(const char aline,const struct warnings warnings,
ali@69	1229	gboolean isemptyline)
ali@44	1230	{
ali@44	1231	/* Don't repeat multiple warnings on one line. */
ali@70	1232	gboolean eNon_A=FALSE,eTab=FALSE,eTilde=FALSE;
ali@70	1233	gboolean eCarat=FALSE,eFSlash=FALSE,eAst=FALSE;
ali@44	1234	const char *s;
ali@70	1235	gunichar c;
ali@70	1236	for (s=aline;*s;s=g_utf8_next_char(s))
ali@44	1237	{
ali@70	1238	c=g_utf8_get_char(s);
ali@70	1239	if (!eNon_A && (c<CHAR_SPACE && c!='\t' && c!='\n' \|\| c>127))
ali@44	1240	{
ali@44	1241	if (pswit[ECHO_SWITCH])
ali@70	1242	g_print("\n%s\n",aline);
ali@44	1243	if (!pswit[OVERVIEW_SWITCH])
ali@70	1244	if (c>127 && c<160 \|\| c>255)
ali@70	1245	g_print(" Line %ld column %ld - "
ali@70	1246	"Non-ISO-8859 character %u\n",
ali@70	1247	linecnt,g_utf8_pointer_to_offset(aline,s)+1,c);
ali@44	1248	else
ali@70	1249	g_print(" Line %ld column %ld - "
ali@70	1250	"Non-ASCII character %u\n",
ali@70	1251	linecnt,g_utf8_pointer_to_offset(aline,s)+1,c);
ali@44	1252	else
ali@44	1253	cnt_bin++;
ali@70	1254	eNon_A=TRUE;
ali@44	1255	}
ali@70	1256	if (!eTab && c==CHAR_TAB)
ali@44	1257	{
ali@44	1258	if (pswit[ECHO_SWITCH])
ali@70	1259	g_print("\n%s\n",aline);
ali@44	1260	if (!pswit[OVERVIEW_SWITCH])
ali@70	1261	g_print(" Line %ld column %ld - Tab character?\n",
ali@70	1262	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@44	1263	else
ali@44	1264	cnt_odd++;
ali@70	1265	eTab=TRUE;
ali@44	1266	}
ali@70	1267	if (!eTilde && c==CHAR_TILDE)
ali@44	1268	{
ali@44	1269	/*
ali@44	1270	* Often used by OCR software to indicate an
ali@44	1271	* unrecognizable character.
ali@44	1272	*/
ali@44	1273	if (pswit[ECHO_SWITCH])
ali@70	1274	g_print("\n%s\n",aline);
ali@44	1275	if (!pswit[OVERVIEW_SWITCH])
ali@70	1276	g_print(" Line %ld column %ld - Tilde character?\n",
ali@70	1277	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@44	1278	else
ali@44	1279	cnt_odd++;
ali@70	1280	eTilde=TRUE;
ali@44	1281	}
ali@70	1282	if (!eCarat && c==CHAR_CARAT)
ali@44	1283	{
ali@44	1284	if (pswit[ECHO_SWITCH])
ali@70	1285	g_print("\n%s\n",aline);
ali@44	1286	if (!pswit[OVERVIEW_SWITCH])
ali@70	1287	g_print(" Line %ld column %ld - Carat character?\n",
ali@70	1288	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@44	1289	else
ali@44	1290	cnt_odd++;
ali@70	1291	eCarat=TRUE;
ali@44	1292	}
ali@70	1293	if (!eFSlash && c==CHAR_FORESLASH && warnings->fslash)
ali@44	1294	{
ali@44	1295	if (pswit[ECHO_SWITCH])
ali@70	1296	g_print("\n%s\n",aline);
ali@44	1297	if (!pswit[OVERVIEW_SWITCH])
ali@70	1298	g_print(" Line %ld column %ld - Forward slash?\n",
ali@70	1299	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@44	1300	else
ali@44	1301	cnt_odd++;
ali@70	1302	eFSlash=TRUE;
ali@44	1303	}
ali@44	1304	/*
ali@44	1305	* Report asterisks only in paranoid mode,
ali@44	1306	* since they're often deliberate.
ali@44	1307	*/
ali@44	1308	if (!eAst && pswit[PARANOID_SWITCH] && warnings->ast && !isemptyline &&
ali@70	1309	c==CHAR_ASTERISK)
ali@44	1310	{
ali@44	1311	if (pswit[ECHO_SWITCH])
ali@70	1312	g_print("\n%s\n",aline);
ali@44	1313	if (!pswit[OVERVIEW_SWITCH])
ali@70	1314	g_print(" Line %ld column %ld - Asterisk?\n",
ali@70	1315	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@44	1316	else
ali@44	1317	cnt_odd++;
ali@70	1318	eAst=TRUE;
ali@44	1319	}
ali@44	1320	}
ali@44	1321	}
ali@44	1322
ali@44	1323	/*
ali@45	1324	* check_for_long_line:
ali@45	1325	*
ali@45	1326	* Check for line too long.
ali@45	1327	*/
ali@45	1328	void check_for_long_line(const char *aline)
ali@45	1329	{
ali@70	1330	if (g_utf8_strlen(aline,-1)>LONGEST_PG_LINE)
ali@45	1331	{
ali@45	1332	if (pswit[ECHO_SWITCH])
ali@70	1333	g_print("\n%s\n",aline);
ali@45	1334	if (!pswit[OVERVIEW_SWITCH])
ali@70	1335	g_print(" Line %ld column %ld - Long line %ld\n",
ali@70	1336	linecnt,g_utf8_strlen(aline,-1),g_utf8_strlen(aline,-1));
ali@45	1337	else
ali@45	1338	cnt_long++;
ali@45	1339	}
ali@45	1340	}
ali@45	1341
ali@45	1342	/*
ali@45	1343	* check_for_short_line:
ali@45	1344	*
ali@45	1345	* Check for line too short.
ali@45	1346	*
ali@45	1347	* This one is a bit trickier to implement: we don't want to
ali@45	1348	* flag the last line of a paragraph for being short, so we
ali@45	1349	* have to wait until we know that our current line is a
ali@45	1350	* "normal" line, then report the _previous_ line if it was too
ali@45	1351	* short. We also don't want to report indented lines like
ali@45	1352	* chapter heads or formatted quotations. We therefore keep
ali@45	1353	* last->len as the length of the last line examined, and
ali@45	1354	* last->blen as the length of the last but one, and try to
ali@45	1355	* suppress unnecessary warnings by checking that both were of
ali@45	1356	* "normal" length. We keep the first character of the last
ali@45	1357	* line in last->start, and if it was a space, we assume that
ali@45	1358	* the formatting is deliberate. I can't figure out a way to
ali@45	1359	* distinguish something like a quoted verse left-aligned or
ali@45	1360	* the header or footer of a letter from a paragraph of short
ali@45	1361	* lines - maybe if I examined the whole paragraph, and if the
ali@45	1362	* para has less than, say, 8 lines and if all lines are short,
ali@45	1363	* then just assume it's OK? Need to look at some texts to see
ali@45	1364	* how often a formula like this would get the right result.
ali@45	1365	*/
ali@45	1366	void check_for_short_line(const char aline,const struct line_properties last)
ali@45	1367	{
ali@70	1368	if (g_utf8_strlen(aline,-1)>1 && last->len>1 &&
ali@70	1369	last->len<SHORTEST_PG_LINE && last->blen>1 &&
ali@70	1370	last->blen>SHORTEST_PG_LINE && last->start!=CHAR_SPACE)
ali@45	1371	{
ali@45	1372	if (pswit[ECHO_SWITCH])
ali@70	1373	g_print("\n%s\n",prevline);
ali@45	1374	if (!pswit[OVERVIEW_SWITCH])
ali@70	1375	g_print(" Line %ld column %ld - Short line %ld?\n",
ali@70	1376	linecnt-1,g_utf8_strlen(prevline,-1),g_utf8_strlen(prevline,-1));
ali@45	1377	else
ali@45	1378	cnt_short++;
ali@45	1379	}
ali@45	1380	}
ali@45	1381
ali@45	1382	/*
ali@46	1383	* check_for_starting_punctuation:
ali@46	1384	*
ali@46	1385	* Look for punctuation other than full ellipses at start of line.
ali@46	1386	*/
ali@46	1387	void check_for_starting_punctuation(const char *aline)
ali@46	1388	{
ali@70	1389	if (*aline && g_utf8_strchr(".?!,;:",-1,g_utf8_get_char(aline)) &&
ali@70	1390	!g_str_has_prefix(aline,". . ."))
ali@46	1391	{
ali@46	1392	if (pswit[ECHO_SWITCH])
ali@70	1393	g_print("\n%s\n",aline);
ali@46	1394	if (!pswit[OVERVIEW_SWITCH])
ali@70	1395	g_print(" Line %ld column 1 - Begins with punctuation?\n",
ali@46	1396	linecnt);
ali@46	1397	else
ali@46	1398	cnt_punct++;
ali@46	1399	}
ali@46	1400	}
ali@46	1401
ali@46	1402	/*
ali@147	1403	* str_emdash:
ali@147	1404	*
ali@147	1405	* Find the first em-dash, return a pointer to it and set <next> to the
ali@147	1406	* character following the dash.
ali@147	1407	*/
ali@147	1408	char str_emdash(const char s,const char **next)
ali@147	1409	{
ali@147	1410	const char s1,s2;
ali@147	1411	s1=strstr(s,"--");
ali@147	1412	s2=strstr(s,"—");
ali@147	1413	if (!s1)
ali@147	1414	{
ali@147	1415	if (s2)
ali@147	1416	*next=g_utf8_next_char(s2);
ali@147	1417	return (char *)s2;
ali@147	1418	}
ali@147	1419	else if (!s2)
ali@147	1420	{
ali@147	1421	*next=g_utf8_next_char(g_utf8_next_char(s1));
ali@147	1422	return (char *)s1;
ali@147	1423	}
ali@147	1424	else if (s1<s2)
ali@147	1425	{
ali@147	1426	*next=g_utf8_next_char(g_utf8_next_char(s1));
ali@147	1427	return (char *)s1;
ali@147	1428	}
ali@147	1429	else
ali@147	1430	{
ali@147	1431	*next=g_utf8_next_char(s2);
ali@147	1432	return (char *)s2;
ali@147	1433	}
ali@147	1434	}
ali@147	1435
ali@147	1436	/*
ali@47	1437	* check_for_spaced_emdash:
ali@47	1438	*
ali@47	1439	* Check for spaced em-dashes.
ali@47	1440	*
ali@147	1441	* We must check _all_ occurrences of em-dashes on the line
ali@147	1442	* hence the loop - even if the first dash is OK
ali@47	1443	* there may be another that's wrong later on.
ali@47	1444	*/
ali@47	1445	void check_for_spaced_emdash(const char *aline)
ali@47	1446	{
ali@70	1447	const char s,t,*next;
ali@147	1448	for (s=aline;t=str_emdash(s,&next);s=next)
ali@47	1449	{
ali@70	1450	if (t>aline && g_utf8_get_char(g_utf8_prev_char(t))==CHAR_SPACE \|\|
ali@70	1451	g_utf8_get_char(next)==CHAR_SPACE)
ali@47	1452	{
ali@47	1453	if (pswit[ECHO_SWITCH])
ali@70	1454	g_print("\n%s\n",aline);
ali@47	1455	if (!pswit[OVERVIEW_SWITCH])
ali@70	1456	g_print(" Line %ld column %ld - Spaced em-dash?\n",
ali@70	1457	linecnt,g_utf8_pointer_to_offset(aline,t)+1);
ali@47	1458	else
ali@47	1459	cnt_dash++;
ali@47	1460	}
ali@47	1461	}
ali@47	1462	}
ali@47	1463
ali@47	1464	/*
ali@47	1465	* check_for_spaced_dash:
ali@47	1466	*
ali@47	1467	* Check for spaced dashes.
ali@47	1468	*/
ali@47	1469	void check_for_spaced_dash(const char *aline)
ali@47	1470	{
ali@47	1471	const char *s;
ali@47	1472	if ((s=strstr(aline," -")))
ali@47	1473	{
ali@70	1474	if (g_utf8_get_char(g_utf8_next_char(g_utf8_next_char(s)))!='-')
ali@47	1475	{
ali@47	1476	if (pswit[ECHO_SWITCH])
ali@70	1477	g_print("\n%s\n",aline);
ali@47	1478	if (!pswit[OVERVIEW_SWITCH])
ali@70	1479	g_print(" Line %ld column %ld - Spaced dash?\n",
ali@70	1480	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@47	1481	else
ali@47	1482	cnt_dash++;
ali@47	1483	}
ali@47	1484	}
ali@47	1485	else if ((s=strstr(aline,"- ")))
ali@47	1486	{
ali@70	1487	if (s==aline \|\| g_utf8_get_char(g_utf8_prev_char(s))!='-')
ali@47	1488	{
ali@47	1489	if (pswit[ECHO_SWITCH])
ali@70	1490	g_print("\n%s\n",aline);
ali@47	1491	if (!pswit[OVERVIEW_SWITCH])
ali@70	1492	g_print(" Line %ld column %ld - Spaced dash?\n",
ali@70	1493	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@47	1494	else
ali@47	1495	cnt_dash++;
ali@47	1496	}
ali@47	1497	}
ali@47	1498	}
ali@47	1499
ali@47	1500	/*
ali@48	1501	* check_for_unmarked_paragraphs:
ali@48	1502	*
ali@48	1503	* Check for unmarked paragraphs indicated by separate speakers.
ali@48	1504	*
ali@48	1505	* May well be false positive:
ali@48	1506	* "Bravo!" "Wonderful!" called the crowd.
ali@48	1507	* but useful all the same.
ali@48	1508	*/
ali@48	1509	void check_for_unmarked_paragraphs(const char *aline)
ali@48	1510	{
ali@48	1511	const char *s;
ali@48	1512	s=strstr(aline,"\" \"");
ali@48	1513	if (!s)
ali@48	1514	s=strstr(aline,"\" \"");
ali@48	1515	if (s)
ali@48	1516	{
ali@48	1517	if (pswit[ECHO_SWITCH])
ali@70	1518	g_print("\n%s\n",aline);
ali@48	1519	if (!pswit[OVERVIEW_SWITCH])
ali@70	1520	g_print(" Line %ld column %ld - "
ali@70	1521	"Query missing paragraph break?\n",
ali@70	1522	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@48	1523	else
ali@48	1524	cnt_punct++;
ali@48	1525	}
ali@48	1526	}
ali@48	1527
ali@48	1528	/*
ali@49	1529	* check_for_jeebies:
ali@49	1530	*
ali@49	1531	* Check for "to he" and other easy h/b errors.
ali@49	1532	*
ali@49	1533	* This is a very inadequate effort on the h/b problem,
ali@49	1534	* but the phrase "to he" is always an error, whereas "to
ali@49	1535	* be" is quite common.
ali@49	1536	* Similarly, '"Quiet!", be said.' is a non-be error
ali@49	1537	* "to he" is _not_ always an error!:
ali@49	1538	* "Where they went to he couldn't say."
ali@49	1539	* Another false positive:
ali@49	1540	* What would "Cinderella" be without the . . .
ali@49	1541	* and another: "If he wants to he can see for himself."
ali@49	1542	*/
ali@49	1543	void check_for_jeebies(const char *aline)
ali@49	1544	{
ali@49	1545	const char *s;
ali@49	1546	s=strstr(aline," be could ");
ali@49	1547	if (!s)
ali@49	1548	s=strstr(aline," be would ");
ali@49	1549	if (!s)
ali@49	1550	s=strstr(aline," was be ");
ali@49	1551	if (!s)
ali@49	1552	s=strstr(aline," be is ");
ali@49	1553	if (!s)
ali@49	1554	s=strstr(aline," is be ");
ali@49	1555	if (!s)
ali@49	1556	s=strstr(aline,"\", be ");
ali@49	1557	if (!s)
ali@49	1558	s=strstr(aline,"\" be ");
ali@49	1559	if (!s)
ali@49	1560	s=strstr(aline,"\" be ");
ali@49	1561	if (!s)
ali@49	1562	s=strstr(aline," to he ");
ali@49	1563	if (s)
ali@49	1564	{
ali@49	1565	if (pswit[ECHO_SWITCH])
ali@70	1566	g_print("\n%s\n",aline);
ali@49	1567	if (!pswit[OVERVIEW_SWITCH])
ali@70	1568	g_print(" Line %ld column %ld - Query he/be error?\n",
ali@70	1569	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@49	1570	else
ali@49	1571	cnt_word++;
ali@49	1572	}
ali@49	1573	s=strstr(aline," the had ");
ali@49	1574	if (!s)
ali@49	1575	s=strstr(aline," a had ");
ali@49	1576	if (!s)
ali@49	1577	s=strstr(aline," they bad ");
ali@49	1578	if (!s)
ali@49	1579	s=strstr(aline," she bad ");
ali@49	1580	if (!s)
ali@49	1581	s=strstr(aline," he bad ");
ali@49	1582	if (!s)
ali@49	1583	s=strstr(aline," you bad ");
ali@49	1584	if (!s)
ali@49	1585	s=strstr(aline," i bad ");
ali@49	1586	if (s)
ali@49	1587	{
ali@49	1588	if (pswit[ECHO_SWITCH])
ali@70	1589	g_print("\n%s\n",aline);
ali@49	1590	if (!pswit[OVERVIEW_SWITCH])
ali@70	1591	g_print(" Line %ld column %ld - Query had/bad error?\n",
ali@70	1592	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@49	1593	else
ali@49	1594	cnt_word++;
ali@49	1595	}
ali@49	1596	s=strstr(aline,"; hut ");
ali@49	1597	if (!s)
ali@49	1598	s=strstr(aline,", hut ");
ali@49	1599	if (s)
ali@49	1600	{
ali@49	1601	if (pswit[ECHO_SWITCH])
ali@70	1602	g_print("\n%s\n",aline);
ali@49	1603	if (!pswit[OVERVIEW_SWITCH])
ali@70	1604	g_print(" Line %ld column %ld - Query hut/but error?\n",
ali@70	1605	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@49	1606	else
ali@49	1607	cnt_word++;
ali@49	1608	}
ali@49	1609	}
ali@49	1610
ali@49	1611	/*
ali@50	1612	* check_for_mta_from:
ali@50	1613	*
ali@50	1614	* Special case - angled bracket in front of "From" placed there by an
ali@50	1615	* MTA when sending an e-mail.
ali@50	1616	*/
ali@50	1617	void check_for_mta_from(const char *aline)
ali@50	1618	{
ali@50	1619	const char *s;
ali@50	1620	s=strstr(aline,">From");
ali@50	1621	if (s)
ali@50	1622	{
ali@50	1623	if (pswit[ECHO_SWITCH])
ali@70	1624	g_print("\n%s\n",aline);
ali@50	1625	if (!pswit[OVERVIEW_SWITCH])
ali@70	1626	g_print(" Line %ld column %ld - "
ali@70	1627	"Query angled bracket with From\n",
ali@70	1628	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@50	1629	else
ali@50	1630	cnt_punct++;
ali@50	1631	}
ali@50	1632	}
ali@50	1633
ali@50	1634	/*
ali@51	1635	* check_for_orphan_character:
ali@51	1636	*
ali@51	1637	* Check for a single character line -
ali@51	1638	* often an overflow from bad wrapping.
ali@51	1639	*/
ali@51	1640	void check_for_orphan_character(const char *aline)
ali@51	1641	{
ali@70	1642	gunichar c;
ali@70	1643	c=g_utf8_get_char(aline);
ali@70	1644	if (c && !*g_utf8_next_char(aline))
ali@51	1645	{
ali@70	1646	if (c=='I' \|\| c=='V' \|\| c=='X' \|\| c=='L' \|\| g_unichar_isdigit(c))
ali@51	1647	; /* Nothing - ignore numerals alone on a line. */
ali@51	1648	else
ali@51	1649	{
ali@51	1650	if (pswit[ECHO_SWITCH])
ali@70	1651	g_print("\n%s\n",aline);
ali@51	1652	if (!pswit[OVERVIEW_SWITCH])
ali@70	1653	g_print(" Line %ld column 1 - Query single character line\n",
ali@51	1654	linecnt);
ali@51	1655	else
ali@51	1656	cnt_punct++;
ali@51	1657	}
ali@51	1658	}
ali@51	1659	}
ali@51	1660
ali@51	1661	/*
ali@52	1662	* check_for_pling_scanno:
ali@52	1663	*
ali@52	1664	* Check for I" - often should be !
ali@52	1665	*/
ali@52	1666	void check_for_pling_scanno(const char *aline)
ali@52	1667	{
ali@52	1668	const char *s;
ali@52	1669	s=strstr(aline," I\"");
ali@52	1670	if (s)
ali@52	1671	{
ali@52	1672	if (pswit[ECHO_SWITCH])
ali@70	1673	g_print("\n%s\n",aline);
ali@52	1674	if (!pswit[OVERVIEW_SWITCH])
ali@70	1675	g_print(" Line %ld column %ld - Query I=exclamation mark?\n",
ali@70	1676	linecnt,g_utf8_pointer_to_offset(aline,s));
ali@52	1677	else
ali@52	1678	cnt_punct++;
ali@52	1679	}
ali@52	1680	}
ali@52	1681
ali@52	1682	/*
ali@53	1683	* check_for_extra_period:
ali@53	1684	*
ali@53	1685	* Check for period without a capital letter. Cut-down from gutspell.
ali@53	1686	* Only works when it happens on a single line.
ali@53	1687	*/
ali@53	1688	void check_for_extra_period(const char aline,const struct warnings warnings)
ali@53	1689	{
ali@99	1690	const char s,t,s1,sprev;
ali@69	1691	int i;
ali@70	1692	gsize len;
ali@69	1693	gboolean istypo;
ali@69	1694	gchar *testword;
ali@99	1695	gunichar c,nc,pc,*decomposition;
ali@53	1696	if (pswit[PARANOID_SWITCH])
ali@53	1697	{
ali@70	1698	for (t=aline;t=strstr(t,". ");)
ali@53	1699	{
ali@69	1700	if (t==aline)
ali@53	1701	{
ali@70	1702	t=g_utf8_next_char(t);
ali@53	1703	/* start of line punctuation is handled elsewhere */
ali@53	1704	continue;
ali@53	1705	}
ali@70	1706	if (!g_unichar_isalpha(g_utf8_get_char(g_utf8_prev_char(t))))
ali@53	1707	{
ali@70	1708	t=g_utf8_next_char(t);
ali@53	1709	continue;
ali@53	1710	}
ali@53	1711	if (warnings->isDutch)
ali@53	1712	{
ali@53	1713	/* For Frank & Jeroen -- 's Middags case */
ali@70	1714	gunichar c2,c3,c4,c5;
ali@70	1715	c2=g_utf8_get_char(g_utf8_offset_to_pointer(t,2));
ali@70	1716	c3=g_utf8_get_char(g_utf8_offset_to_pointer(t,3));
ali@70	1717	c4=g_utf8_get_char(g_utf8_offset_to_pointer(t,4));
ali@70	1718	c5=g_utf8_get_char(g_utf8_offset_to_pointer(t,5));
ali@99	1719	if (CHAR_IS_APOSTROPHE(c2) &&
ali@99	1720	g_unichar_islower(c3) && c4==CHAR_SPACE &&
ali@99	1721	g_unichar_isupper(c5))
ali@53	1722	{
ali@70	1723	t=g_utf8_next_char(t);
ali@53	1724	continue;
ali@53	1725	}
ali@53	1726	}
ali@70	1727	s1=g_utf8_next_char(g_utf8_next_char(t));
ali@70	1728	while (*s1 && !g_unichar_isalpha(g_utf8_get_char(s1)) &&
ali@173	1729	!g_unichar_isdigit(g_utf8_get_char(s1)))
ali@70	1730	s1=g_utf8_next_char(s1);
ali@70	1731	if (g_unichar_islower(g_utf8_get_char(s1)))
ali@53	1732	{
ali@53	1733	/* we have something to investigate */
ali@69	1734	istypo=TRUE;
ali@53	1735	/* so let's go back and find out */
ali@99	1736	nc=g_utf8_get_char(t);
ali@99	1737	s1=g_utf8_prev_char(t);
ali@99	1738	c=g_utf8_get_char(s1);
ali@99	1739	sprev=g_utf8_prev_char(s1);
ali@99	1740	pc=g_utf8_get_char(sprev);
ali@99	1741	while (s1>=aline &&
ali@99	1742	(g_unichar_isalpha(c) \|\| g_unichar_isdigit(c) \|\|
ali@99	1743	g_unichar_isalpha(pc) && CHAR_IS_APOSTROPHE(c) &&
ali@99	1744	g_unichar_isalpha(nc)))
ali@99	1745	{
ali@99	1746	nc=c;
ali@99	1747	s1=sprev;
ali@99	1748	c=pc;
ali@99	1749	sprev=g_utf8_prev_char(s1);
ali@99	1750	pc=g_utf8_get_char(sprev);
ali@99	1751	}
ali@70	1752	s1=g_utf8_next_char(s1);
ali@69	1753	s=strchr(s1,'.');
ali@69	1754	if (s)
ali@69	1755	testword=g_strndup(s1,s-s1);
ali@69	1756	else
ali@69	1757	testword=g_strdup(s1);
ali@53	1758	for (i=0;*abbrev[i];i++)
ali@53	1759	if (!strcmp(testword,abbrev[i]))
ali@69	1760	istypo=FALSE;
ali@70	1761	if (g_unichar_isdigit(g_utf8_get_char(testword)))
ali@69	1762	istypo=FALSE;
ali@70	1763	if (!*g_utf8_next_char(testword))
ali@69	1764	istypo=FALSE;
ali@53	1765	if (isroman(testword))
ali@69	1766	istypo=FALSE;
ali@53	1767	if (istypo)
ali@53	1768	{
ali@69	1769	istypo=FALSE;
ali@70	1770	for (s=testword;*s;s=g_utf8_next_char(s))
ali@70	1771	{
ali@70	1772	decomposition=g_unicode_canonical_decomposition(
ali@70	1773	g_utf8_get_char(s),&len);
ali@70	1774	if (g_utf8_strchr("aeiou",-1,decomposition[0]))
ali@69	1775	istypo=TRUE;
ali@70	1776	g_free(decomposition);
ali@70	1777	}
ali@53	1778	}
ali@69	1779	if (istypo &&
ali@69	1780	(pswit[VERBOSE_SWITCH] \|\| !g_tree_lookup(qperiod,testword)))
ali@53	1781	{
ali@69	1782	g_tree_insert(qperiod,g_strdup(testword),
ali@69	1783	GINT_TO_POINTER(1));
ali@69	1784	if (pswit[ECHO_SWITCH])
ali@70	1785	g_print("\n%s\n",aline);
ali@69	1786	if (!pswit[OVERVIEW_SWITCH])
ali@70	1787	g_print(" Line %ld column %ld - Extra period?\n",
ali@70	1788	linecnt,g_utf8_pointer_to_offset(aline,t)+1);
ali@69	1789	else
ali@69	1790	cnt_punct++;
ali@53	1791	}
ali@69	1792	g_free(testword);
ali@53	1793	}
ali@70	1794	t=g_utf8_next_char(t);
ali@53	1795	}
ali@53	1796	}
ali@53	1797	}
ali@53	1798
ali@53	1799	/*
ali@54	1800	* check_for_following_punctuation:
ali@54	1801	*
ali@54	1802	* Check for words usually not followed by punctuation.
ali@54	1803	*/
ali@54	1804	void check_for_following_punctuation(const char *aline)
ali@54	1805	{
ali@54	1806	int i;
ali@54	1807	const char s,wordstart;
ali@70	1808	gunichar c;
ali@69	1809	gchar inword,t;
ali@54	1810	if (pswit[TYPO_SWITCH])
ali@54	1811	{
ali@54	1812	for (s=aline;*s;)
ali@54	1813	{
ali@54	1814	wordstart=s;
ali@69	1815	t=getaword(&s);
ali@69	1816	if (!*t)
ali@69	1817	{
ali@69	1818	g_free(t);
ali@54	1819	continue;
ali@69	1820	}
ali@70	1821	inword=g_utf8_strdown(t,-1);
ali@69	1822	g_free(t);
ali@54	1823	for (i=0;*nocomma[i];i++)
ali@54	1824	if (!strcmp(inword,nocomma[i]))
ali@54	1825	{
ali@70	1826	c=g_utf8_get_char(s);
ali@70	1827	if (c==',' \|\| c==';' \|\| c==':')
ali@54	1828	{
ali@54	1829	if (pswit[ECHO_SWITCH])
ali@70	1830	g_print("\n%s\n",aline);
ali@54	1831	if (!pswit[OVERVIEW_SWITCH])
ali@70	1832	g_print(" Line %ld column %ld - "
ali@54	1833	"Query punctuation after %s?\n",
ali@70	1834	linecnt,g_utf8_pointer_to_offset(aline,s)+1,
ali@70	1835	inword);
ali@54	1836	else
ali@54	1837	cnt_punct++;
ali@54	1838	}
ali@54	1839	}
ali@54	1840	for (i=0;*noperiod[i];i++)
ali@54	1841	if (!strcmp(inword,noperiod[i]))
ali@54	1842	{
ali@70	1843	c=g_utf8_get_char(s);
ali@70	1844	if (c=='.' \|\| c=='!')
ali@54	1845	{
ali@54	1846	if (pswit[ECHO_SWITCH])
ali@70	1847	g_print("\n%s\n",aline);
ali@54	1848	if (!pswit[OVERVIEW_SWITCH])
ali@70	1849	g_print(" Line %ld column %ld - "
ali@54	1850	"Query punctuation after %s?\n",
ali@70	1851	linecnt,g_utf8_pointer_to_offset(aline,s)+1,
ali@70	1852	inword);
ali@54	1853	else
ali@54	1854	cnt_punct++;
ali@54	1855	}
ali@54	1856	}
ali@69	1857	g_free(inword);
ali@54	1858	}
ali@54	1859	}
ali@54	1860	}
ali@54	1861
ali@54	1862	/*
ali@55	1863	* check_for_typos:
ali@55	1864	*
ali@55	1865	* Check for commonly mistyped words,
ali@55	1866	* and digits like 0 for O in a word.
ali@55	1867	*/
ali@55	1868	void check_for_typos(const char aline,struct warnings warnings)
ali@55	1869	{
ali@70	1870	const char s,t,nt,wordstart;
ali@70	1871	gchar *inword;
ali@70	1872	gunichar *decomposition;
ali@70	1873	gchar *testword;
ali@70	1874	int i,vowel,consonant,*dupcnt;
ali@70	1875	gboolean isdup,istypo,alower;
ali@99	1876	gunichar c,pc;
ali@70	1877	long offset,len;
ali@70	1878	gsize decomposition_len;
ali@55	1879	for (s=aline;*s;)
ali@55	1880	{
ali@55	1881	wordstart=s;
ali@69	1882	inword=getaword(&s);
ali@55	1883	if (!*inword)
ali@69	1884	{
ali@69	1885	g_free(inword);
ali@55	1886	continue; /* don't bother with empty lines */
ali@69	1887	}
ali@55	1888	if (mixdigit(inword))
ali@55	1889	{
ali@55	1890	if (pswit[ECHO_SWITCH])
ali@70	1891	g_print("\n%s\n",aline);
ali@55	1892	if (!pswit[OVERVIEW_SWITCH])
ali@70	1893	g_print(" Line %ld column %ld - Query digit in %s\n",
ali@70	1894	linecnt,g_utf8_pointer_to_offset(aline,wordstart)+1,inword);
ali@55	1895	else
ali@55	1896	cnt_word++;
ali@55	1897	}
ali@55	1898	/*
ali@55	1899	* Put the word through a series of tests for likely typos and OCR
ali@55	1900	* errors.
ali@55	1901	*/
ali@69	1902	if (pswit[TYPO_SWITCH] \|\| pswit[USERTYPO_SWITCH])
ali@55	1903	{
ali@69	1904	istypo=FALSE;
ali@70	1905	alower=FALSE;
ali@70	1906	for (t=inword;*t;t=g_utf8_next_char(t))
ali@55	1907	{
ali@70	1908	c=g_utf8_get_char(t);
ali@70	1909	nt=g_utf8_next_char(t);
ali@55	1910	/* lowercase for testing */
ali@70	1911	if (g_unichar_islower(c))
ali@70	1912	alower=TRUE;
ali@70	1913	if (alower && (g_unichar_isupper(c) \|\| g_unichar_istitle(c)))
ali@55	1914	{
ali@55	1915	/*
ali@55	1916	* We have an uppercase mid-word. However, there are
ali@55	1917	* common cases:
ali@55	1918	* Mac and Mc like McGill
ali@55	1919	* French contractions like l'Abbe
ali@55	1920	*/
ali@70	1921	offset=g_utf8_pointer_to_offset(inword,t);
ali@99	1922	if (offset>0)
ali@99	1923	pc=g_utf8_get_char(g_utf8_prev_char(t));
ali@99	1924	else
ali@99	1925	pc='\0';
ali@70	1926	if (offset==2 && c=='m' && g_utf8_get_char(nt)=='c' \|\|
ali@70	1927	offset==3 && c=='m' && g_utf8_get_char(nt)=='a' &&
ali@70	1928	g_utf8_get_char(g_utf8_next_char(nt))=='c' \|\|
ali@99	1929	CHAR_IS_APOSTROPHE(pc))
ali@55	1930	; /* do nothing! */
ali@55	1931	else
ali@69	1932	istypo=TRUE;
ali@55	1933	}
ali@55	1934	}
ali@70	1935	testword=g_utf8_casefold(inword,-1);
ali@69	1936	}
ali@69	1937	if (pswit[TYPO_SWITCH])
ali@69	1938	{
ali@55	1939	/*
ali@55	1940	* Check for certain unlikely two-letter combinations at word
ali@55	1941	* start and end.
ali@55	1942	*/
ali@70	1943	len=g_utf8_strlen(testword,-1);
ali@70	1944	if (len>1)
ali@55	1945	{
ali@55	1946	for (i=0;*nostart[i];i++)
ali@70	1947	if (g_str_has_prefix(testword,nostart[i]))
ali@69	1948	istypo=TRUE;
ali@55	1949	for (i=0;*noend[i];i++)
ali@70	1950	if (g_str_has_suffix(testword,noend[i]))
ali@69	1951	istypo=TRUE;
ali@55	1952	}
ali@55	1953	/* ght is common, gbt never. Like that. */
ali@55	1954	if (strstr(testword,"cb"))
ali@69	1955	istypo=TRUE;
ali@55	1956	if (strstr(testword,"gbt"))
ali@69	1957	istypo=TRUE;
ali@55	1958	if (strstr(testword,"pbt"))
ali@69	1959	istypo=TRUE;
ali@55	1960	if (strstr(testword,"tbs"))
ali@69	1961	istypo=TRUE;
ali@55	1962	if (strstr(testword,"mrn"))
ali@69	1963	istypo=TRUE;
ali@55	1964	if (strstr(testword,"ahle"))
ali@69	1965	istypo=TRUE;
ali@55	1966	if (strstr(testword,"ihle"))
ali@69	1967	istypo=TRUE;
ali@55	1968	/*
ali@55	1969	* "TBE" does happen - like HEARTBEAT - but uncommon.
ali@55	1970	* Also "TBI" - frostbite, outbid - but uncommon.
ali@55	1971	* Similarly "ii" like Hawaii, or Pompeii, and in Roman
ali@55	1972	* numerals, but "ii" is a common scanno.
ali@55	1973	*/
ali@55	1974	if (strstr(testword,"tbi"))
ali@69	1975	istypo=TRUE;
ali@55	1976	if (strstr(testword,"tbe"))
ali@69	1977	istypo=TRUE;
ali@55	1978	if (strstr(testword,"ii"))
ali@69	1979	istypo=TRUE;
ali@55	1980	/*
ali@55	1981	* Check for no vowels or no consonants.
ali@55	1982	* If none, flag a typo.
ali@55	1983	*/
ali@70	1984	if (!istypo && len>1)
ali@55	1985	{
ali@55	1986	vowel=consonant=0;
ali@70	1987	for (t=testword;*t;t=g_utf8_next_char(t))
ali@55	1988	{
ali@70	1989	c=g_utf8_get_char(t);
ali@70	1990	decomposition=
ali@70	1991	g_unicode_canonical_decomposition(c,&decomposition_len);
ali@70	1992	if (c=='y' \|\| g_unichar_isdigit(c))
ali@55	1993	{
ali@55	1994	/* Yah, this is loose. */
ali@55	1995	vowel++;
ali@55	1996	consonant++;
ali@55	1997	}
ali@70	1998	else if (g_utf8_strchr("aeiou",-1,decomposition[0]))
ali@55	1999	vowel++;
ali@55	2000	else
ali@55	2001	consonant++;
ali@70	2002	g_free(decomposition);
ali@55	2003	}
ali@55	2004	if (!vowel \|\| !consonant)
ali@69	2005	istypo=TRUE;
ali@55	2006	}
ali@55	2007	/*
ali@55	2008	* Now exclude the word from being reported if it's in
ali@55	2009	* the okword list.
ali@55	2010	*/
ali@55	2011	for (i=0;*okword[i];i++)
ali@55	2012	if (!strcmp(testword,okword[i]))
ali@69	2013	istypo=FALSE;
ali@55	2014	/*
ali@55	2015	* What looks like a typo may be a Roman numeral.
ali@55	2016	* Exclude these.
ali@55	2017	*/
ali@55	2018	if (istypo && isroman(testword))
ali@69	2019	istypo=FALSE;
ali@55	2020	/* Check the manual list of typos. */
ali@55	2021	if (!istypo)
ali@55	2022	for (i=0;*typo[i];i++)
ali@55	2023	if (!strcmp(testword,typo[i]))
ali@69	2024	istypo=TRUE;
ali@55	2025	/*
ali@55	2026	* Check lowercase s, l, i and m - special cases.
ali@55	2027	* "j" - often a semi-colon gone wrong.
ali@55	2028	* "d" for a missing apostrophe - he d
ali@55	2029	* "n" for "in"
ali@55	2030	*/
ali@70	2031	if (!istypo && len==1 &&
ali@70	2032	g_utf8_strchr("slmijdn",-1,g_utf8_get_char(inword)))
ali@69	2033	istypo=TRUE;
ali@55	2034	if (istypo)
ali@55	2035	{
ali@69	2036	dupcnt=g_tree_lookup(qword,testword);
ali@69	2037	if (dupcnt)
ali@69	2038	{
ali@69	2039	(*dupcnt)++;
ali@69	2040	isdup=!pswit[VERBOSE_SWITCH];
ali@69	2041	}
ali@69	2042	else
ali@69	2043	{
ali@69	2044	dupcnt=g_new0(int,1);
ali@69	2045	g_tree_insert(qword,g_strdup(testword),dupcnt);
ali@69	2046	isdup=FALSE;
ali@69	2047	}
ali@55	2048	if (!isdup)
ali@55	2049	{
ali@55	2050	if (pswit[ECHO_SWITCH])
ali@70	2051	g_print("\n%s\n",aline);
ali@55	2052	if (!pswit[OVERVIEW_SWITCH])
ali@55	2053	{
ali@70	2054	g_print(" Line %ld column %ld - Query word %s",
ali@70	2055	linecnt,g_utf8_pointer_to_offset(aline,wordstart)+1,
ali@70	2056	inword);
ali@69	2057	if (!pswit[VERBOSE_SWITCH])
ali@70	2058	g_print(" - not reporting duplicates");
ali@70	2059	g_print("\n");
ali@55	2060	}
ali@55	2061	else
ali@55	2062	cnt_word++;
ali@55	2063	}
ali@55	2064	}
ali@55	2065	}
ali@55	2066	/* check the user's list of typos */
ali@69	2067	if (!istypo && usertypo && g_tree_lookup(usertypo,testword))
ali@69	2068	{
ali@69	2069	if (pswit[ECHO_SWITCH])
ali@70	2070	g_print("\n%s\n",aline);
ali@69	2071	if (!pswit[OVERVIEW_SWITCH])
ali@70	2072	g_print(" Line %ld column %ld - Query possible scanno %s\n",
ali@70	2073	linecnt,g_utf8_pointer_to_offset(aline,wordstart)+2,inword);
ali@69	2074	}
ali@69	2075	if (pswit[TYPO_SWITCH] \|\| pswit[USERTYPO_SWITCH])
ali@69	2076	g_free(testword);
ali@55	2077	if (pswit[PARANOID_SWITCH] && warnings->digit)
ali@55	2078	{
ali@55	2079	/* In paranoid mode, query all 0 and 1 standing alone. */
ali@55	2080	if (!strcmp(inword,"0") \|\| !strcmp(inword,"1"))
ali@55	2081	{
ali@55	2082	if (pswit[ECHO_SWITCH])
ali@70	2083	g_print("\n%s\n",aline);
ali@55	2084	if (!pswit[OVERVIEW_SWITCH])
ali@70	2085	g_print(" Line %ld column %ld - Query standalone %s\n",
ali@70	2086	linecnt,g_utf8_pointer_to_offset(aline,wordstart)+2,
ali@70	2087	inword);
ali@55	2088	else
ali@55	2089	cnt_word++;
ali@55	2090	}
ali@55	2091	}
ali@69	2092	g_free(inword);
ali@55	2093	}
ali@55	2094	}
ali@55	2095
ali@56	2096	/*
ali@56	2097	* check_for_misspaced_punctuation:
ali@56	2098	*
ali@56	2099	* Look for added or missing spaces around punctuation and quotes.
ali@56	2100	* If there is a punctuation character like ! with no space on
ali@56	2101	* either side, suspect a missing!space. If there are spaces on
ali@56	2102	* both sides , assume a typo. If we see a double quote with no
ali@56	2103	* space or punctuation on either side of it, assume unspaced
ali@56	2104	* quotes "like"this.
ali@56	2105	*/
ali@56	2106	void check_for_misspaced_punctuation(const char *aline,
ali@69	2107	struct parities *parities,gboolean isemptyline)
ali@56	2108	{
ali@69	2109	gboolean isacro,isellipsis;
ali@56	2110	const char *s;
ali@70	2111	gunichar c,nc,pc,n2c;
ali@142	2112	int parity;
ali@70	2113	c=g_utf8_get_char(aline);
ali@70	2114	nc=c?g_utf8_get_char(g_utf8_next_char(aline)):0;
ali@70	2115	for (s=g_utf8_next_char(aline);nc;s=g_utf8_next_char(s))
ali@56	2116	{
ali@70	2117	pc=c;
ali@70	2118	c=nc;
ali@70	2119	nc=g_utf8_get_char(g_utf8_next_char(s));
ali@56	2120	/* For each character in the line after the first. */
ali@70	2121	if (g_utf8_strchr(".?!,;:_",-1,c)) /* if it's punctuation */
ali@56	2122	{
ali@56	2123	/* we need to suppress warnings for acronyms like M.D. */
ali@69	2124	isacro=FALSE;
ali@56	2125	/* we need to suppress warnings for ellipsis . . . */
ali@69	2126	isellipsis=FALSE;
ali@70	2127	/*
ali@70	2128	* If there are letters on both sides of it or
ali@70	2129	* if it's strict punctuation followed by an alpha.
ali@70	2130	*/
ali@70	2131	if (g_unichar_isalpha(nc) && (g_unichar_isalpha(pc) \|\|
ali@70	2132	g_utf8_strchr("?!,;:",-1,c)))
ali@56	2133	{
ali@70	2134	if (c=='.')
ali@56	2135	{
ali@70	2136	if (g_utf8_pointer_to_offset(aline,s)>2 &&
ali@70	2137	g_utf8_get_char(g_utf8_offset_to_pointer(s,-2))=='.')
ali@69	2138	isacro=TRUE;
ali@70	2139	n2c=g_utf8_get_char(g_utf8_next_char(g_utf8_next_char(s)));
ali@70	2140	if (nc && n2c=='.')
ali@69	2141	isacro=TRUE;
ali@56	2142	}
ali@56	2143	if (!isacro)
ali@56	2144	{
ali@56	2145	if (pswit[ECHO_SWITCH])
ali@70	2146	g_print("\n%s\n",aline);
ali@56	2147	if (!pswit[OVERVIEW_SWITCH])
ali@70	2148	g_print(" Line %ld column %ld - Missing space?\n",
ali@70	2149	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@56	2150	else
ali@56	2151	cnt_punct++;
ali@56	2152	}
ali@56	2153	}
ali@70	2154	if (pc==CHAR_SPACE && (nc==CHAR_SPACE \|\| !nc))
ali@56	2155	{
ali@56	2156	/*
ali@56	2157	* If there are spaces on both sides,
ali@56	2158	* or space before and end of line.
ali@56	2159	*/
ali@70	2160	if (c=='.')
ali@56	2161	{
ali@70	2162	if (g_utf8_pointer_to_offset(aline,s)>2 &&
ali@70	2163	g_utf8_get_char(g_utf8_offset_to_pointer(s,-2))=='.')
ali@69	2164	isellipsis=TRUE;
ali@70	2165	n2c=g_utf8_get_char(g_utf8_next_char(g_utf8_next_char(s)));
ali@70	2166	if (nc && n2c=='.')
ali@69	2167	isellipsis=TRUE;
ali@56	2168	}
ali@56	2169	if (!isemptyline && !isellipsis)
ali@56	2170	{
ali@56	2171	if (pswit[ECHO_SWITCH])
ali@70	2172	g_print("\n%s\n",aline);
ali@56	2173	if (!pswit[OVERVIEW_SWITCH])
ali@70	2174	g_print(" Line %ld column %ld - "
ali@70	2175	"Spaced punctuation?\n",linecnt,
ali@70	2176	g_utf8_pointer_to_offset(aline,s)+1);
ali@56	2177	else
ali@56	2178	cnt_punct++;
ali@56	2179	}
ali@56	2180	}
ali@56	2181	}
ali@56	2182	}
ali@56	2183	/* Split out the characters that CANNOT be preceded by space. */
ali@70	2184	c=g_utf8_get_char(aline);
ali@70	2185	nc=c?g_utf8_get_char(g_utf8_next_char(aline)):0;
ali@70	2186	for (s=g_utf8_next_char(aline);nc;s=g_utf8_next_char(s))
ali@56	2187	{
ali@70	2188	pc=c;
ali@70	2189	c=nc;
ali@70	2190	nc=g_utf8_get_char(g_utf8_next_char(s));
ali@56	2191	/* for each character in the line after the first */
ali@70	2192	if (g_utf8_strchr("?!,;:",-1,c))
ali@56	2193	{
ali@56	2194	/* if it's punctuation that _cannot_ have a space before it */
ali@70	2195	if (pc==CHAR_SPACE && !isemptyline && nc!=CHAR_SPACE)
ali@56	2196	{
ali@56	2197	/*
ali@70	2198	* If nc DOES == space,
ali@56	2199	* it was already reported just above.
ali@56	2200	*/
ali@56	2201	if (pswit[ECHO_SWITCH])
ali@70	2202	g_print("\n%s\n",aline);
ali@56	2203	if (!pswit[OVERVIEW_SWITCH])
ali@70	2204	g_print(" Line %ld column %ld - Spaced punctuation?\n",
ali@70	2205	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@56	2206	else
ali@56	2207	cnt_punct++;
ali@56	2208	}
ali@56	2209	}
ali@56	2210	}
ali@56	2211	/*
ali@56	2212	* Special case " .X" where X is any alpha.
ali@56	2213	* This plugs a hole in the acronym code above.
ali@56	2214	* Inelegant, but maintainable.
ali@56	2215	*/
ali@70	2216	c=g_utf8_get_char(aline);
ali@70	2217	nc=c?g_utf8_get_char(g_utf8_next_char(aline)):0;
ali@70	2218	for (s=g_utf8_next_char(aline);nc;s=g_utf8_next_char(s))
ali@56	2219	{
ali@70	2220	pc=c;
ali@70	2221	c=nc;
ali@70	2222	nc=g_utf8_get_char(g_utf8_next_char(s));
ali@56	2223	/* for each character in the line after the first */
ali@70	2224	if (c=='.')
ali@56	2225	{
ali@56	2226	/* if it's a period */
ali@70	2227	if (pc==CHAR_SPACE && g_unichar_isalpha(nc))
ali@56	2228	{
ali@56	2229	/*
ali@56	2230	* If the period follows a space and
ali@56	2231	* is followed by a letter.
ali@56	2232	*/
ali@56	2233	if (pswit[ECHO_SWITCH])
ali@70	2234	g_print("\n%s\n",aline);
ali@56	2235	if (!pswit[OVERVIEW_SWITCH])
ali@70	2236	g_print(" Line %ld column %ld - Spaced punctuation?\n",
ali@70	2237	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@56	2238	else
ali@56	2239	cnt_punct++;
ali@56	2240	}
ali@56	2241	}
ali@56	2242	}
ali@70	2243	c=g_utf8_get_char(aline);
ali@70	2244	nc=c?g_utf8_get_char(g_utf8_next_char(aline)):0;
ali@70	2245	for (s=g_utf8_next_char(aline);nc;s=g_utf8_next_char(s))
ali@56	2246	{
ali@70	2247	pc=c;
ali@70	2248	c=nc;
ali@70	2249	nc=g_utf8_get_char(g_utf8_next_char(s));
ali@56	2250	/* for each character in the line after the first */
ali@142	2251	if (CHAR_IS_DQUOTE(c))
ali@56	2252	{
ali@70	2253	if (!g_utf8_strchr(" _-.'`,;:!/([{?}])",-1,pc) &&
ali@70	2254	!g_utf8_strchr(" _-.'`,;:!/([{?}])",-1,nc) && nc \|\|
ali@70	2255	!g_utf8_strchr(" _-([{'`",-1,pc) && g_unichar_isalpha(nc))
ali@56	2256	{
ali@56	2257	if (pswit[ECHO_SWITCH])
ali@70	2258	g_print("\n%s\n",aline);
ali@56	2259	if (!pswit[OVERVIEW_SWITCH])
ali@70	2260	g_print(" Line %ld column %ld - Unspaced quotes?\n",
ali@70	2261	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@56	2262	else
ali@56	2263	cnt_punct++;
ali@56	2264	}
ali@56	2265	}
ali@56	2266	}
ali@56	2267	/* Check parity of quotes. */
ali@70	2268	nc=g_utf8_get_char(aline);
ali@70	2269	for (s=aline;*s;s=g_utf8_next_char(s))
ali@56	2270	{
ali@70	2271	c=nc;
ali@70	2272	nc=g_utf8_get_char(g_utf8_next_char(s));
ali@142	2273	if (CHAR_IS_DQUOTE(c))
ali@56	2274	{
ali@142	2275	if (c==CHAR_DQUOTE)
ali@142	2276	{
ali@142	2277	parities->dquote=!parities->dquote;
ali@142	2278	parity=parities->dquote;
ali@142	2279	}
ali@142	2280	else if (c==CHAR_LD_QUOTE)
ali@142	2281	parity=1;
ali@142	2282	else
ali@142	2283	parity=0;
ali@142	2284	if (!parity)
ali@56	2285	{
ali@56	2286	/* parity even */
ali@173	2287	if (!g_utf8_strchr("_-.'`‘’/,;:!?)]} ",-1,nc))
ali@56	2288	{
ali@56	2289	if (pswit[ECHO_SWITCH])
ali@70	2290	g_print("\n%s\n",aline);
ali@56	2291	if (!pswit[OVERVIEW_SWITCH])
ali@70	2292	g_print(" Line %ld column %ld - "
ali@70	2293	"Wrongspaced quotes?\n",
ali@70	2294	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@56	2295	else
ali@56	2296	cnt_punct++;
ali@56	2297	}
ali@56	2298	}
ali@56	2299	else
ali@56	2300	{
ali@56	2301	/* parity odd */
ali@173	2302	if (!g_unichar_isalpha(nc) && !g_unichar_isdigit(nc) &&
ali@173	2303	!g_utf8_strchr("_-/.'`‘’([{$",-1,nc) \|\| !nc)
ali@56	2304	{
ali@56	2305	if (pswit[ECHO_SWITCH])
ali@70	2306	g_print("\n%s\n",aline);
ali@56	2307	if (!pswit[OVERVIEW_SWITCH])
ali@70	2308	g_print(" Line %ld column %ld - "
ali@70	2309	"Wrongspaced quotes?\n",
ali@70	2310	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@56	2311	else
ali@56	2312	cnt_punct++;
ali@56	2313	}
ali@56	2314	}
ali@56	2315	}
ali@56	2316	}
ali@142	2317	c=g_utf8_get_char(aline);
ali@142	2318	if (CHAR_IS_DQUOTE(c))
ali@56	2319	{
ali@70	2320	if (g_utf8_strchr(",;:!?)]} ",-1,
ali@70	2321	g_utf8_get_char(g_utf8_next_char(aline))))
ali@56	2322	{
ali@56	2323	if (pswit[ECHO_SWITCH])
ali@70	2324	g_print("\n%s\n",aline);
ali@56	2325	if (!pswit[OVERVIEW_SWITCH])
ali@70	2326	g_print(" Line %ld column 1 - Wrongspaced quotes?\n",
ali@56	2327	linecnt);
ali@56	2328	else
ali@56	2329	cnt_punct++;
ali@56	2330	}
ali@56	2331	}
ali@56	2332	if (pswit[SQUOTE_SWITCH])
ali@56	2333	{
ali@70	2334	nc=g_utf8_get_char(aline);
ali@70	2335	for (s=aline;*s;s=g_utf8_next_char(s))
ali@56	2336	{
ali@70	2337	c=nc;
ali@70	2338	nc=g_utf8_get_char(g_utf8_next_char(s));
ali@99	2339	if (CHAR_IS_SQUOTE(c) && (s==aline \|\| s>aline &&
ali@70	2340	!g_unichar_isalpha(g_utf8_get_char(g_utf8_prev_char(s))) \|\|
ali@70	2341	!g_unichar_isalpha(nc)))
ali@56	2342	{
ali@56	2343	parities->squote=!parities->squote;
ali@56	2344	if (!parities->squote)
ali@56	2345	{
ali@56	2346	/* parity even */
ali@70	2347	if (!g_utf8_strchr("_-.'`/\",;:!?)]} ",-1,nc))
ali@56	2348	{
ali@56	2349	if (pswit[ECHO_SWITCH])
ali@70	2350	g_print("\n%s\n",aline);
ali@56	2351	if (!pswit[OVERVIEW_SWITCH])
ali@70	2352	g_print(" Line %ld column %ld - "
ali@56	2353	"Wrongspaced singlequotes?\n",
ali@70	2354	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@56	2355	else
ali@56	2356	cnt_punct++;
ali@56	2357	}
ali@56	2358	}
ali@56	2359	else
ali@56	2360	{
ali@56	2361	/* parity odd */
ali@173	2362	if (!g_unichar_isalpha(nc) && !g_unichar_isdigit(nc) &&
ali@70	2363	!g_utf8_strchr("_-/\".'`",-1,nc) \|\| !nc)
ali@56	2364	{
ali@56	2365	if (pswit[ECHO_SWITCH])
ali@70	2366	g_print("\n%s\n",aline);
ali@56	2367	if (!pswit[OVERVIEW_SWITCH])
ali@70	2368	g_print(" Line %ld column %ld - "
ali@56	2369	"Wrongspaced singlequotes?\n",
ali@70	2370	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@56	2371	else
ali@56	2372	cnt_punct++;
ali@56	2373	}
ali@56	2374	}
ali@56	2375	}
ali@56	2376	}
ali@56	2377	}
ali@56	2378	}
ali@56	2379
ali@55	2380	/*
ali@57	2381	* check_for_double_punctuation:
ali@57	2382	*
ali@57	2383	* Look for double punctuation like ,. or ,,
ali@57	2384	* Thanks to DW for the suggestion!
ali@57	2385	* In books with references, ".," and ".;" are common
ali@57	2386	* e.g. "etc., etc.," and vol. 1.; vol 3.;
ali@57	2387	* OTOH, from my initial tests, there are also fairly
ali@57	2388	* common errors. What to do? Make these cases paranoid?
ali@57	2389	* ".," is the most common, so warnings->dotcomma is used
ali@57	2390	* to suppress detailed reporting if it occurs often.
ali@57	2391	*/
ali@57	2392	void check_for_double_punctuation(const char aline,struct warnings warnings)
ali@57	2393	{
ali@70	2394	const char *s;
ali@70	2395	gunichar c,nc;
ali@70	2396	nc=g_utf8_get_char(aline);
ali@70	2397	for (s=aline;*s;s=g_utf8_next_char(s))
ali@57	2398	{
ali@70	2399	c=nc;
ali@70	2400	nc=g_utf8_get_char(g_utf8_next_char(s));
ali@57	2401	/* for each punctuation character in the line */
ali@70	2402	if (c && nc && g_utf8_strchr(".?!,;:",-1,c) &&
ali@70	2403	g_utf8_strchr(".?!,;:",-1,nc))
ali@57	2404	{
ali@57	2405	/* followed by punctuation, it's a query, unless . . . */
ali@70	2406	if (c==nc && (c=='.' \|\| c=='?' \|\| c=='!') \|\|
ali@70	2407	!warnings->dotcomma && c=='.' && nc==',' \|\|
ali@70	2408	warnings->isFrench && g_str_has_prefix(s,",...") \|\|
ali@70	2409	warnings->isFrench && g_str_has_prefix(s,"...,") \|\|
ali@70	2410	warnings->isFrench && g_str_has_prefix(s,";...") \|\|
ali@70	2411	warnings->isFrench && g_str_has_prefix(s,"...;") \|\|
ali@70	2412	warnings->isFrench && g_str_has_prefix(s,":...") \|\|
ali@70	2413	warnings->isFrench && g_str_has_prefix(s,"...:") \|\|
ali@70	2414	warnings->isFrench && g_str_has_prefix(s,"!...") \|\|
ali@70	2415	warnings->isFrench && g_str_has_prefix(s,"...!") \|\|
ali@70	2416	warnings->isFrench && g_str_has_prefix(s,"?...") \|\|
ali@70	2417	warnings->isFrench && g_str_has_prefix(s,"...?"))
ali@57	2418	{
ali@70	2419	if (warnings->isFrench && g_str_has_prefix(s,",...") \|\|
ali@70	2420	warnings->isFrench && g_str_has_prefix(s,"...,") \|\|
ali@70	2421	warnings->isFrench && g_str_has_prefix(s,";...") \|\|
ali@70	2422	warnings->isFrench && g_str_has_prefix(s,"...;") \|\|
ali@70	2423	warnings->isFrench && g_str_has_prefix(s,":...") \|\|
ali@70	2424	warnings->isFrench && g_str_has_prefix(s,"...:") \|\|
ali@70	2425	warnings->isFrench && g_str_has_prefix(s,"!...") \|\|
ali@70	2426	warnings->isFrench && g_str_has_prefix(s,"...!") \|\|
ali@70	2427	warnings->isFrench && g_str_has_prefix(s,"?...") \|\|
ali@70	2428	warnings->isFrench && g_str_has_prefix(s,"...?"))
ali@70	2429	{
ali@70	2430	s+=4;
ali@70	2431	nc=g_utf8_get_char(g_utf8_next_char(s));
ali@70	2432	}
ali@57	2433	; /* do nothing for .. !! and ?? which can be legit */
ali@57	2434	}
ali@57	2435	else
ali@57	2436	{
ali@57	2437	if (pswit[ECHO_SWITCH])
ali@70	2438	g_print("\n%s\n",aline);
ali@57	2439	if (!pswit[OVERVIEW_SWITCH])
ali@70	2440	g_print(" Line %ld column %ld - Double punctuation?\n",
ali@70	2441	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@57	2442	else
ali@57	2443	cnt_punct++;
ali@57	2444	}
ali@57	2445	}
ali@57	2446	}
ali@57	2447	}
ali@57	2448
ali@57	2449	/*
ali@58	2450	* check_for_spaced_quotes:
ali@58	2451	*/
ali@58	2452	void check_for_spaced_quotes(const char *aline)
ali@58	2453	{
ali@99	2454	int i;
ali@58	2455	const char s,t;
ali@99	2456	const gunichar single_quotes[]={CHAR_SQUOTE,CHAR_OPEN_SQUOTE,CHAR_LS_QUOTE,
ali@99	2457	CHAR_RS_QUOTE};
ali@99	2458	GString *pattern;
ali@58	2459	s=aline;
ali@58	2460	while ((t=strstr(s," \" ")))
ali@58	2461	{
ali@58	2462	if (pswit[ECHO_SWITCH])
ali@70	2463	g_print("\n%s\n",aline);
ali@58	2464	if (!pswit[OVERVIEW_SWITCH])
ali@70	2465	g_print(" Line %ld column %ld - Spaced doublequote?\n",
ali@70	2466	linecnt,g_utf8_pointer_to_offset(aline,t)+1);
ali@58	2467	else
ali@58	2468	cnt_punct++;
ali@70	2469	s=g_utf8_next_char(g_utf8_next_char(t));
ali@58	2470	}
ali@99	2471	pattern=g_string_new(NULL);
ali@99	2472	for(i=0;i<G_N_ELEMENTS(single_quotes);i++)
ali@58	2473	{
ali@99	2474	g_string_assign(pattern," ");
ali@99	2475	g_string_append_unichar(pattern,single_quotes[i]);
ali@99	2476	g_string_append_c(pattern,' ');
ali@99	2477	s=aline;
ali@99	2478	while ((t=strstr(s,pattern->str)))
ali@99	2479	{
ali@99	2480	if (pswit[ECHO_SWITCH])
ali@99	2481	g_print("\n%s\n",aline);
ali@99	2482	if (!pswit[OVERVIEW_SWITCH])
ali@99	2483	g_print(" Line %ld column %ld - Spaced singlequote?\n",
ali@99	2484	linecnt,g_utf8_pointer_to_offset(aline,t)+1);
ali@99	2485	else
ali@99	2486	cnt_punct++;
ali@99	2487	s=g_utf8_next_char(g_utf8_next_char(t));
ali@99	2488	}
ali@58	2489	}
ali@99	2490	g_string_free(pattern,TRUE);
ali@58	2491	}
ali@58	2492
ali@58	2493	/*
ali@59	2494	* check_for_miscased_genative:
ali@59	2495	*
ali@59	2496	* Check special case of 'S instead of 's at end of word.
ali@59	2497	*/
ali@59	2498	void check_for_miscased_genative(const char *aline)
ali@59	2499	{
ali@59	2500	const char *s;
ali@70	2501	gunichar c,nc,pc;
ali@69	2502	if (!*aline)
ali@69	2503	return;
ali@70	2504	c=g_utf8_get_char(aline);
ali@70	2505	nc=c?g_utf8_get_char(g_utf8_next_char(aline)):0;
ali@70	2506	for (s=g_utf8_next_char(aline);nc;s=g_utf8_next_char(s))
ali@59	2507	{
ali@70	2508	pc=c;
ali@70	2509	c=nc;
ali@70	2510	nc=g_utf8_get_char(g_utf8_next_char(s));
ali@99	2511	if (CHAR_IS_APOSTROPHE(c) && nc=='S' && g_unichar_islower(pc))
ali@59	2512	{
ali@59	2513	if (pswit[ECHO_SWITCH])
ali@70	2514	g_print("\n%s\n",aline);
ali@59	2515	if (!pswit[OVERVIEW_SWITCH])
ali@70	2516	g_print(" Line %ld column %ld - Capital \"S\"?\n",
ali@70	2517	linecnt,g_utf8_pointer_to_offset(aline,s)+2);
ali@59	2518	else
ali@59	2519	cnt_punct++;
ali@59	2520	}
ali@59	2521	}
ali@59	2522	}
ali@59	2523
ali@59	2524	/*
ali@60	2525	* check_end_of_line:
ali@60	2526	*
ali@60	2527	* Now check special cases - start and end of line -
ali@60	2528	* for single and double quotes. Start is sometimes [sic]
ali@60	2529	* but better to query it anyway.
ali@60	2530	* While we're here, check for dash at end of line.
ali@60	2531	*/
ali@60	2532	void check_end_of_line(const char aline,struct warnings warnings)
ali@60	2533	{
ali@70	2534	int lbytes;
ali@70	2535	const char *s;
ali@70	2536	gunichar c1,c2;
ali@70	2537	lbytes=strlen(aline);
ali@70	2538	if (g_utf8_strlen(aline,lbytes)>1)
ali@60	2539	{
ali@70	2540	s=g_utf8_prev_char(aline+lbytes);
ali@70	2541	c1=g_utf8_get_char(s);
ali@70	2542	c2=g_utf8_get_char(g_utf8_prev_char(s));
ali@142	2543	if ((CHAR_IS_DQUOTE(c1) \|\| CHAR_IS_SQUOTE(c1)) && c2==CHAR_SPACE)
ali@60	2544	{
ali@60	2545	if (pswit[ECHO_SWITCH])
ali@70	2546	g_print("\n%s\n",aline);
ali@60	2547	if (!pswit[OVERVIEW_SWITCH])
ali@70	2548	g_print(" Line %ld column %ld - Spaced quote?\n",linecnt,
ali@70	2549	g_utf8_strlen(aline,lbytes));
ali@70	2550	else
ali@70	2551	cnt_punct++;
ali@70	2552	}
ali@70	2553	c1=g_utf8_get_char(aline);
ali@70	2554	c2=g_utf8_get_char(g_utf8_next_char(aline));
ali@99	2555	if (CHAR_IS_SQUOTE(c1) && c2==CHAR_SPACE)
ali@70	2556	{
ali@70	2557	if (pswit[ECHO_SWITCH])
ali@70	2558	g_print("\n%s\n",aline);
ali@70	2559	if (!pswit[OVERVIEW_SWITCH])
ali@70	2560	g_print(" Line %ld column 1 - Spaced quote?\n",linecnt);
ali@60	2561	else
ali@60	2562	cnt_punct++;
ali@60	2563	}
ali@60	2564	/*
ali@60	2565	* Dash at end of line may well be legit - paranoid mode only
ali@60	2566	* and don't report em-dash at line-end.
ali@60	2567	*/
ali@60	2568	if (pswit[PARANOID_SWITCH] && warnings->hyphen)
ali@60	2569	{
ali@70	2570	for (s=g_utf8_prev_char(aline+lbytes);
ali@70	2571	s>aline && g_utf8_get_char(s)<=CHAR_SPACE;s=g_utf8_prev_char(s))
ali@60	2572	;
ali@70	2573	if (g_utf8_get_char(s)=='-' &&
ali@70	2574	g_utf8_get_char(g_utf8_prev_char(s))!='-')
ali@60	2575	{
ali@60	2576	if (pswit[ECHO_SWITCH])
ali@70	2577	g_print("\n%s\n",aline);
ali@60	2578	if (!pswit[OVERVIEW_SWITCH])
ali@70	2579	g_print(" Line %ld column %ld - "
ali@70	2580	"Hyphen at end of line?\n",
ali@70	2581	linecnt,g_utf8_pointer_to_offset(aline,s));
ali@60	2582	}
ali@60	2583	}
ali@60	2584	}
ali@60	2585	}
ali@60	2586
ali@60	2587	/*
ali@61	2588	* check_for_unspaced_bracket:
ali@61	2589	*
ali@61	2590	* Brackets are often unspaced, but shouldn't be surrounded by alpha.
ali@61	2591	* If so, suspect a scanno like "a]most".
ali@61	2592	*/
ali@61	2593	void check_for_unspaced_bracket(const char *aline)
ali@61	2594	{
ali@70	2595	const char *s;
ali@70	2596	gunichar c,nc,pc;
ali@70	2597	c=g_utf8_get_char(aline);
ali@70	2598	nc=c?g_utf8_get_char(g_utf8_next_char(aline)):0;
ali@70	2599	for (s=g_utf8_next_char(aline);nc;s=g_utf8_next_char(s))
ali@61	2600	{
ali@70	2601	pc=c;
ali@70	2602	c=nc;
ali@70	2603	nc=g_utf8_get_char(g_utf8_next_char(s));
ali@70	2604	if (!nc)
ali@70	2605	break;
ali@61	2606	/* for each bracket character in the line except 1st & last */
ali@70	2607	if (g_utf8_strchr("{[()]}",-1,c) &&
ali@70	2608	g_unichar_isalpha(pc) && g_unichar_isalpha(nc))
ali@61	2609	{
ali@61	2610	if (pswit[ECHO_SWITCH])
ali@70	2611	g_print("\n%s\n",aline);
ali@61	2612	if (!pswit[OVERVIEW_SWITCH])
ali@70	2613	g_print(" Line %ld column %ld - Unspaced bracket?\n",
ali@70	2614	linecnt,g_utf8_pointer_to_offset(aline,s));
ali@61	2615	else
ali@61	2616	cnt_punct++;
ali@61	2617	}
ali@61	2618	}
ali@61	2619	}
ali@61	2620
ali@61	2621	/*
ali@62	2622	* check_for_unpunctuated_endquote:
ali@62	2623	*/
ali@62	2624	void check_for_unpunctuated_endquote(const char *aline)
ali@62	2625	{
ali@70	2626	const char *s;
ali@70	2627	gunichar c,nc,pc;
ali@142	2628	QuoteClass qc;
ali@70	2629	c=g_utf8_get_char(aline);
ali@70	2630	nc=c?g_utf8_get_char(g_utf8_next_char(aline)):0;
ali@70	2631	for (s=g_utf8_next_char(aline);nc;s=g_utf8_next_char(s))
ali@62	2632	{
ali@70	2633	pc=c;
ali@70	2634	c=nc;
ali@142	2635	qc=CHAR_IS_DQUOTE(c)?QUOTE_CLASS(c):INVALID_QUOTE;
ali@70	2636	nc=g_utf8_get_char(g_utf8_next_char(s));
ali@62	2637	/* for each character in the line except 1st */
ali@147	2638	if ((qc==CLOSING_QUOTE \|\| qc==NEUTRAL_QUOTE) && g_unichar_isalpha(pc))
ali@62	2639	{
ali@62	2640	if (pswit[ECHO_SWITCH])
ali@70	2641	g_print("\n%s\n",aline);
ali@62	2642	if (!pswit[OVERVIEW_SWITCH])
ali@70	2643	g_print(" Line %ld column %ld - "
ali@70	2644	"endquote missing punctuation?\n",
ali@70	2645	linecnt,g_utf8_pointer_to_offset(aline,s));
ali@62	2646	else
ali@62	2647	cnt_punct++;
ali@62	2648	}
ali@62	2649	}
ali@62	2650	}
ali@62	2651
ali@62	2652	/*
ali@63	2653	* check_for_html_tag:
ali@63	2654	*
ali@63	2655	* Check for <HTML TAG>.
ali@63	2656	*
ali@63	2657	* If there is a < in the line, followed at some point
ali@63	2658	* by a > then we suspect HTML.
ali@63	2659	*/
ali@63	2660	void check_for_html_tag(const char *aline)
ali@63	2661	{
ali@63	2662	const char open,close;
ali@70	2663	gchar *tag;
ali@70	2664	open=strchr(aline,'<');
ali@63	2665	if (open)
ali@63	2666	{
ali@70	2667	close=strchr(g_utf8_next_char(open),'>');
ali@63	2668	if (close)
ali@63	2669	{
ali@70	2670	if (pswit[ECHO_SWITCH])
ali@70	2671	g_print("\n%s\n",aline);
ali@70	2672	if (!pswit[OVERVIEW_SWITCH])
ali@63	2673	{
ali@70	2674	tag=g_strndup(open,close-open+1);
ali@70	2675	g_print(" Line %ld column %ld - HTML Tag? %s \n",
ali@70	2676	linecnt,g_utf8_pointer_to_offset(aline,open)+1,tag);
ali@70	2677	g_free(tag);
ali@63	2678	}
ali@70	2679	else
ali@70	2680	cnt_html++;
ali@63	2681	}
ali@63	2682	}
ali@63	2683	}
ali@63	2684
ali@63	2685	/*
ali@64	2686	* check_for_html_entity:
ali@64	2687	*
ali@64	2688	* Check for &symbol; HTML.
ali@64	2689	*
ali@64	2690	* If there is a & in the line, followed at
ali@64	2691	* some point by a ; then we suspect HTML.
ali@64	2692	*/
ali@64	2693	void check_for_html_entity(const char *aline)
ali@64	2694	{
ali@64	2695	const char s,amp,*scolon;
ali@70	2696	gchar *entity;
ali@70	2697	amp=strchr(aline,'&');
ali@64	2698	if (amp)
ali@64	2699	{
ali@70	2700	scolon=strchr(amp,';');
ali@64	2701	if (scolon)
ali@64	2702	{
ali@70	2703	for (s=amp;s<scolon;s=g_utf8_next_char(s))
ali@70	2704	if (g_utf8_get_char(s)==CHAR_SPACE)
ali@70	2705	break; /* Don't report "Jones & Son;" */
ali@70	2706	if (s>=scolon)
ali@64	2707	{
ali@64	2708	if (pswit[ECHO_SWITCH])
ali@70	2709	g_print("\n%s\n",aline);
ali@64	2710	if (!pswit[OVERVIEW_SWITCH])
ali@70	2711	{
ali@70	2712	entity=g_strndup(amp,scolon-amp+1);
ali@70	2713	g_print(" Line %ld column %d - HTML symbol? %s \n",
ali@70	2714	linecnt,(int)(amp-aline)+1,entity);
ali@70	2715	g_free(entity);
ali@70	2716	}
ali@64	2717	else
ali@64	2718	cnt_html++;
ali@64	2719	}
ali@64	2720	}
ali@64	2721	}
ali@64	2722	}
ali@64	2723
ali@65	2724	/*
ali@66	2725	* check_for_omitted_punctuation:
ali@66	2726	*
ali@66	2727	* Check for omitted punctuation at end of paragraph by working back
ali@66	2728	* through prevline. DW.
ali@66	2729	* Need to check this only for "normal" paras.
ali@66	2730	* So what is a "normal" para?
ali@66	2731	* Not normal if one-liner (chapter headings, etc.)
ali@66	2732	* Not normal if doesn't contain at least one locase letter
ali@66	2733	* Not normal if starts with space
ali@66	2734	*/
ali@66	2735	void check_for_omitted_punctuation(const char *prevline,
ali@66	2736	struct line_properties *last,int start_para_line)
ali@66	2737	{
ali@70	2738	gboolean letter_on_line=FALSE;
ali@66	2739	const char *s;
ali@99	2740	gunichar c;
ali@142	2741	gboolean closing_quote;
ali@70	2742	for (s=prevline;*s;s=g_utf8_next_char(s))
ali@70	2743	if (g_unichar_isalpha(g_utf8_get_char(s)))
ali@70	2744	{
ali@70	2745	letter_on_line=TRUE;
ali@70	2746	break;
ali@70	2747	}
ali@66	2748	/*
ali@66	2749	* This next "if" is a problem.
ali@66	2750	* If we say "start_para_line <= linecnt - 1", that includes
ali@66	2751	* one-line "paragraphs" like chapter heads. Lotsa false positives.
ali@66	2752	* If we say "start_para_line < linecnt - 1" it doesn't, but then it
ali@66	2753	* misses genuine one-line paragraphs.
ali@66	2754	*/
ali@70	2755	if (letter_on_line && last->blen>2 && start_para_line<linecnt-1 &&
ali@70	2756	g_utf8_get_char(prevline)>CHAR_SPACE)
ali@66	2757	{
ali@99	2758	s=prevline+strlen(prevline);
ali@99	2759	do
ali@99	2760	{
ali@99	2761	s=g_utf8_prev_char(s);
ali@99	2762	c=g_utf8_get_char(s);
ali@142	2763	if (QUOTE_CLASS(c)==CLOSING_QUOTE \|\| QUOTE_CLASS(c)==NEUTRAL_QUOTE)
ali@142	2764	closing_quote=TRUE;
ali@142	2765	else
ali@142	2766	closing_quote=FALSE;
ali@142	2767	} while (closing_quote && s>prevline);
ali@70	2768	for (;s>prevline;s=g_utf8_prev_char(s))
ali@66	2769	{
ali@70	2770	if (g_unichar_isalpha(g_utf8_get_char(s)))
ali@66	2771	{
ali@66	2772	if (pswit[ECHO_SWITCH])
ali@70	2773	g_print("\n%s\n",prevline);
ali@66	2774	if (!pswit[OVERVIEW_SWITCH])
ali@70	2775	g_print(" Line %ld column %ld - "
ali@66	2776	"No punctuation at para end?\n",
ali@70	2777	linecnt-1,g_utf8_strlen(prevline,-1));
ali@66	2778	else
ali@66	2779	cnt_punct++;
ali@66	2780	break;
ali@66	2781	}
ali@147	2782	if (g_utf8_strchr("—-.:!([{?}])",-1,g_utf8_get_char(s)))
ali@66	2783	break;
ali@66	2784	}
ali@66	2785	}
ali@66	2786	}
ali@66	2787
ali@69	2788	gboolean report_duplicate_queries(gpointer key,gpointer value,gpointer data)
ali@69	2789	{
ali@69	2790	const char *word=key;
ali@69	2791	int *dupcnt=value;
ali@69	2792	if (*dupcnt)
ali@70	2793	g_print("\nNote: Queried word %s was duplicated %d times\n",
ali@69	2794	word,*dupcnt);
ali@69	2795	return FALSE;
ali@69	2796	}
ali@69	2797
ali@70	2798	void print_as_windows_1252(const char *string)
ali@70	2799	{
ali@70	2800	gsize inbytes,outbytes;
ali@70	2801	gchar buf,bp;
ali@86	2802	static GIConv converter=(GIConv)-1;
ali@70	2803	if (!string)
ali@70	2804	{
ali@70	2805	if (converter!=(GIConv)-1)
ali@70	2806	g_iconv_close(converter);
ali@70	2807	converter=(GIConv)-1;
ali@70	2808	return;
ali@70	2809	}
ali@86	2810	if (converter==(GIConv)-1)
ali@70	2811	converter=g_iconv_open("WINDOWS-1252","UTF-8");
ali@70	2812	if (converter!=(GIConv)-1)
ali@70	2813	{
ali@70	2814	inbytes=outbytes=strlen(string);
ali@70	2815	bp=buf=g_malloc(outbytes+1);
ali@70	2816	g_iconv(converter,(char **)&string,&inbytes,&bp,&outbytes);
ali@70	2817	*bp='\0';
ali@70	2818	fputs(buf,stdout);
ali@70	2819	g_free(buf);
ali@70	2820	}
ali@70	2821	else
ali@70	2822	fputs(string,stdout);
ali@70	2823	}
ali@70	2824
ali@72	2825	void print_as_utf_8(const char *string)
ali@72	2826	{
ali@72	2827	fputs(string,stdout);
ali@72	2828	}
ali@72	2829
ali@66	2830	/*
ali@41	2831	* procfile:
ali@41	2832	*
ali@41	2833	* Process one file.
ali@41	2834	*/
ali@69	2835	void procfile(const char *filename)
ali@41	2836	{
ali@65	2837	const char *s;
ali@69	2838	gchar parastart=NULL; / first line of current para */
ali@69	2839	gchar etext,aline;
ali@69	2840	gchar *etext_ptr;
ali@69	2841	GError *err=NULL;
ali@41	2842	struct first_pass_results *first_pass_results;
ali@42	2843	struct warnings *warnings;
ali@43	2844	struct counters counters={0};
ali@45	2845	struct line_properties last={0};
ali@56	2846	struct parities parities={0};
ali@69	2847	struct pending pending={0};
ali@69	2848	gboolean isemptyline;
ali@68	2849	long start_para_line=0;
ali@69	2850	gboolean isnewpara=FALSE,enddash=FALSE;
ali@45	2851	last.start=CHAR_SPACE;
ali@68	2852	linecnt=checked_linecnt=0;
ali@69	2853	etext=read_etext(filename,&err);
ali@69	2854	if (!etext)
ali@41	2855	{
ali@68	2856	if (pswit[STDOUT_SWITCH])
ali@69	2857	fprintf(stdout,"bookloupe: %s: %s\n",filename,err->message);
ali@68	2858	else
ali@69	2859	fprintf(stderr,"bookloupe: %s: %s\n",filename,err->message);
ali@41	2860	exit(1);
ali@41	2861	}
ali@70	2862	g_print("\n\nFile: %s\n\n",filename);
ali@69	2863	first_pass_results=first_pass(etext);
ali@42	2864	warnings=report_first_pass(first_pass_results);
ali@69	2865	qword=g_tree_new_full((GCompareDataFunc)strcmp,NULL,g_free,g_free);
ali@69	2866	qperiod=g_tree_new_full((GCompareDataFunc)strcmp,NULL,g_free,NULL);
ali@40	2867	/*
ali@40	2868	* Here we go with the main pass. Hold onto yer hat!
ali@40	2869	*/
ali@65	2870	linecnt=0;
ali@69	2871	etext_ptr=etext;
ali@69	2872	while ((aline=flgets(&etext_ptr,linecnt+1)))
ali@40	2873	{
ali@68	2874	linecnt++;
ali@68	2875	if (linecnt==1)
ali@69	2876	isnewpara=TRUE;
ali@70	2877	if (pswit[DP_SWITCH] && g_str_has_prefix(aline,"-----File: "))
ali@40	2878	continue; // skip DP page separators completely
ali@68	2879	if (linecnt<first_pass_results->firstline \|\|
ali@41	2880	(first_pass_results->footerline>0 &&
ali@41	2881	linecnt>first_pass_results->footerline))
ali@40	2882	{
ali@68	2883	if (pswit[HEADER_SWITCH])
ali@40	2884	{
ali@70	2885	if (g_str_has_prefix(aline,"Title:"))
ali@70	2886	g_print(" %s\n",aline);
ali@70	2887	if (g_str_has_prefix(aline,"Author:"))
ali@70	2888	g_print(" %s\n",aline);
ali@70	2889	if (g_str_has_prefix(aline,"Release Date:"))
ali@70	2890	g_print(" %s\n",aline);
ali@70	2891	if (g_str_has_prefix(aline,"Edition:"))
ali@70	2892	g_print(" %s\n\n",aline);
ali@40	2893	}
ali@68	2894	continue; /* skip through the header */
ali@40	2895	}
ali@68	2896	checked_linecnt++;
ali@65	2897	print_pending(aline,parastart,&pending);
ali@164	2898	isemptyline=analyse_quotes(aline,&counters);
ali@68	2899	if (isnewpara && !isemptyline)
ali@40	2900	{
ali@40	2901	/* This line is the start of a new paragraph. */
ali@68	2902	start_para_line=linecnt;
ali@40	2903	/* Capture its first line in case we want to report it later. */
ali@69	2904	g_free(parastart);
ali@69	2905	parastart=g_strdup(aline);
ali@56	2906	memset(&parities,0,sizeof(parities)); /* restart the quote count */
ali@68	2907	s=aline;
ali@70	2908	while (*s && !g_unichar_isalpha(g_utf8_get_char(s)) &&
ali@70	2909	!g_unichar_isdigit(g_utf8_get_char(s)))
ali@70	2910	s=g_utf8_next_char(s);
ali@70	2911	if (g_unichar_islower(g_utf8_get_char(s)))
ali@40	2912	{
ali@40	2913	/* and its first letter is lowercase */
ali@68	2914	if (pswit[ECHO_SWITCH])
ali@70	2915	g_print("\n%s\n",aline);
ali@68	2916	if (!pswit[OVERVIEW_SWITCH])
ali@70	2917	g_print(" Line %ld column %ld - "
ali@40	2918	"Paragraph starts with lower-case\n",
ali@70	2919	linecnt,g_utf8_pointer_to_offset(aline,s)+1);
ali@68	2920	else
ali@68	2921	cnt_punct++;
ali@40	2922	}
ali@69	2923	isnewpara=FALSE; /* Signal the end of new para processing. */
ali@40	2924	}
ali@68	2925	/* Check for an em-dash broken at line end. */
ali@70	2926	if (enddash && g_utf8_get_char(aline)=='-')
ali@40	2927	{
ali@68	2928	if (pswit[ECHO_SWITCH])
ali@70	2929	g_print("\n%s\n",aline);
ali@68	2930	if (!pswit[OVERVIEW_SWITCH])
ali@70	2931	g_print(" Line %ld column 1 - Broken em-dash?\n",linecnt);
ali@68	2932	else
ali@68	2933	cnt_punct++;
ali@40	2934	}
ali@69	2935	enddash=FALSE;
ali@70	2936	for (s=g_utf8_prev_char(aline+strlen(aline));
ali@70	2937	g_utf8_get_char(s)==' ' && s>aline;s=g_utf8_prev_char(s))
ali@40	2938	;
ali@70	2939	if (s>=aline && g_utf8_get_char(s)=='-')
ali@69	2940	enddash=TRUE;
ali@67	2941	check_for_control_characters(aline);
ali@68	2942	if (warnings->bin)
ali@44	2943	check_for_odd_characters(aline,warnings,isemptyline);
ali@68	2944	if (warnings->longline)
ali@45	2945	check_for_long_line(aline);
ali@68	2946	if (warnings->shortline)
ali@45	2947	check_for_short_line(aline,&last);
ali@68	2948	last.blen=last.len;
ali@70	2949	last.len=g_utf8_strlen(aline,-1);
ali@70	2950	last.start=g_utf8_get_char(aline);
ali@46	2951	check_for_starting_punctuation(aline);
ali@68	2952	if (warnings->dash)
ali@40	2953	{
ali@47	2954	check_for_spaced_emdash(aline);
ali@47	2955	check_for_spaced_dash(aline);
ali@40	2956	}
ali@48	2957	check_for_unmarked_paragraphs(aline);
ali@49	2958	check_for_jeebies(aline);
ali@50	2959	check_for_mta_from(aline);
ali@51	2960	check_for_orphan_character(aline);
ali@52	2961	check_for_pling_scanno(aline);
ali@53	2962	check_for_extra_period(aline,warnings);
ali@54	2963	check_for_following_punctuation(aline);
ali@55	2964	check_for_typos(aline,warnings);
ali@56	2965	check_for_misspaced_punctuation(aline,&parities,isemptyline);
ali@57	2966	check_for_double_punctuation(aline,warnings);
ali@58	2967	check_for_spaced_quotes(aline);
ali@59	2968	check_for_miscased_genative(aline);
ali@60	2969	check_end_of_line(aline,warnings);
ali@61	2970	check_for_unspaced_bracket(aline);
ali@68	2971	if (warnings->endquote)
ali@62	2972	check_for_unpunctuated_endquote(aline);
ali@63	2973	check_for_html_tag(aline);
ali@64	2974	check_for_html_entity(aline);
ali@68	2975	if (isemptyline)
ali@40	2976	{
ali@65	2977	check_for_mismatched_quotes(&counters,&pending);
ali@103	2978	counters_reset(&counters);
ali@40	2979	/* let the next iteration know that it's starting a new para */
ali@69	2980	isnewpara=TRUE;
ali@69	2981	if (prevline)
ali@69	2982	check_for_omitted_punctuation(prevline,&last,start_para_line);
ali@40	2983	}
ali@69	2984	g_free(prevline);
ali@69	2985	prevline=g_strdup(aline);
ali@0	2986	}
ali@103	2987	linecnt++;
ali@103	2988	check_for_mismatched_quotes(&counters,&pending);
ali@103	2989	print_pending(NULL,parastart,&pending);
ali@103	2990	reset_pending(&pending);
ali@69	2991	if (prevline)
ali@69	2992	{
ali@69	2993	g_free(prevline);
ali@69	2994	prevline=NULL;
ali@69	2995	}
ali@69	2996	g_free(parastart);
ali@69	2997	g_free(prevline);
ali@69	2998	g_free(etext);
ali@79	2999	if (!pswit[OVERVIEW_SWITCH] && !pswit[VERBOSE_SWITCH])
ali@69	3000	g_tree_foreach(qword,report_duplicate_queries,NULL);
ali@69	3001	g_tree_unref(qword);
ali@69	3002	g_tree_unref(qperiod);
ali@99	3003	counters_destroy(&counters);
ali@70	3004	g_set_print_handler(NULL);
ali@70	3005	print_as_windows_1252(NULL);
ali@71	3006	if (pswit[MARKUP_SWITCH])
ali@71	3007	loseentities(NULL);
ali@0	3008	}
ali@0	3009
ali@40	3010	/*
ali@40	3011	* flgets:
ali@40	3012	*
ali@69	3013	* Get one line from the input text, checking for
ali@40	3014	* the existence of exactly one CR/LF line-end per line.
ali@40	3015	*
ali@40	3016	* Returns: a pointer to the line.
ali@40	3017	*/
ali@69	3018	char flgets(char *etext,long lcnt)
ali@0	3019	{
ali@70	3020	gunichar c;
ali@69	3021	gboolean isCR=FALSE;
ali@69	3022	char theline=etext;
ali@70	3023	char *eos=theline;
ali@70	3024	gchar *s;
ali@70	3025	for (;;)
ali@40	3026	{
ali@70	3027	c=g_utf8_get_char(*etext);
ali@173	3028	if (!c)
ali@173	3029	{
ali@173	3030	if (*etext==theline)
ali@173	3031	return NULL;
ali@173	3032	else if (pswit[LINE_END_SWITCH])
ali@173	3033	{
ali@173	3034	if (pswit[ECHO_SWITCH])
ali@173	3035	{
ali@173	3036	s=g_strndup(theline,eos-theline);
ali@173	3037	g_print("\n%s\n",s);
ali@173	3038	g_free(s);
ali@173	3039	}
ali@173	3040	if (!pswit[OVERVIEW_SWITCH])
ali@173	3041	/* There may, or may not, have been a CR */
ali@173	3042	g_print(" Line %ld - No LF?\n",lcnt);
ali@173	3043	else
ali@173	3044	cnt_lineend++;
ali@173	3045	}
ali@173	3046	break;
ali@173	3047	}
ali@70	3048	etext=g_utf8_next_char(etext);
ali@40	3049	/* either way, it's end of line */
ali@69	3050	if (c=='\n')
ali@40	3051	{
ali@68	3052	if (isCR)
ali@68	3053	break;
ali@68	3054	else
ali@40	3055	{
ali@40	3056	/* Error - a LF without a preceding CR */
ali@68	3057	if (pswit[LINE_END_SWITCH])
ali@40	3058	{
ali@68	3059	if (pswit[ECHO_SWITCH])
ali@70	3060	{
ali@70	3061	s=g_strndup(theline,eos-theline);
ali@70	3062	g_print("\n%s\n",s);
ali@70	3063	g_free(s);
ali@70	3064	}
ali@68	3065	if (!pswit[OVERVIEW_SWITCH])
ali@70	3066	g_print(" Line %ld - No CR?\n",lcnt);
ali@68	3067	else
ali@68	3068	cnt_lineend++;
ali@40	3069	}
ali@68	3070	break;
ali@40	3071	}
ali@40	3072	}
ali@69	3073	if (c=='\r')
ali@40	3074	{
ali@68	3075	if (isCR)
ali@40	3076	{
ali@40	3077	/* Error - two successive CRs */
ali@68	3078	if (pswit[LINE_END_SWITCH])
ali@40	3079	{
ali@68	3080	if (pswit[ECHO_SWITCH])
ali@70	3081	{
ali@70	3082	s=g_strndup(theline,eos-theline);
ali@70	3083	g_print("\n%s\n",s);
ali@70	3084	g_free(s);
ali@70	3085	}
ali@68	3086	if (!pswit[OVERVIEW_SWITCH])
ali@70	3087	g_print(" Line %ld - Two successive CRs?\n",lcnt);
ali@68	3088	else
ali@68	3089	cnt_lineend++;
ali@40	3090	}
ali@40	3091	}
ali@69	3092	isCR=TRUE;
ali@40	3093	}
ali@68	3094	else
ali@40	3095	{
ali@68	3096	if (pswit[LINE_END_SWITCH] && isCR)
ali@40	3097	{
ali@68	3098	if (pswit[ECHO_SWITCH])
ali@70	3099	{
ali@70	3100	s=g_strndup(theline,eos-theline);
ali@70	3101	g_print("\n%s\n",s);
ali@70	3102	g_free(s);
ali@70	3103	}
ali@68	3104	if (!pswit[OVERVIEW_SWITCH])
ali@70	3105	g_print(" Line %ld column %ld - CR without LF?\n",
ali@70	3106	lcnt,g_utf8_pointer_to_offset(theline,eos)+1);
ali@68	3107	else
ali@68	3108	cnt_lineend++;
ali@70	3109	*eos=' ';
ali@40	3110	}
ali@69	3111	isCR=FALSE;
ali@70	3112	eos=g_utf8_next_char(eos);
ali@40	3113	}
ali@69	3114	}
ali@70	3115	*eos='\0';
ali@0	3116	if (pswit[MARKUP_SWITCH])
ali@68	3117	postprocess_for_HTML(theline);
ali@0	3118	if (pswit[DP_SWITCH])
ali@68	3119	postprocess_for_DP(theline);
ali@40	3120	return theline;
ali@0	3121	}
ali@0	3122
ali@40	3123	/*
ali@40	3124	* mixdigit:
ali@40	3125	*
ali@40	3126	* Takes a "word" as a parameter, and checks whether it
ali@40	3127	* contains a mixture of alpha and digits. Generally, this is an
ali@40	3128	* error, but may not be for cases like 4th or L5 12s. 3d.
ali@40	3129	*
ali@70	3130	* Returns: TRUE iff an is error found.
ali@40	3131	*/
ali@70	3132	gboolean mixdigit(const char *checkword)
ali@0	3133	{
ali@70	3134	gboolean wehaveadigit,wehavealetter,query;
ali@70	3135	const char s,nondigit;
ali@70	3136	wehaveadigit=wehavealetter=query=FALSE;
ali@70	3137	for (s=checkword;*s;s=g_utf8_next_char(s))
ali@70	3138	if (g_unichar_isalpha(g_utf8_get_char(s)))
ali@70	3139	wehavealetter=TRUE;
ali@70	3140	else if (g_unichar_isdigit(g_utf8_get_char(s)))
ali@70	3141	wehaveadigit=TRUE;
ali@40	3142	if (wehaveadigit && wehavealetter)
ali@40	3143	{
ali@40	3144	/* Now exclude common legit cases, like "21st" and "12l. 3s. 11d." */
ali@70	3145	query=TRUE;
ali@70	3146	for (nondigit=checkword;g_unichar_isdigit(g_utf8_get_char(nondigit));
ali@70	3147	nondigit=g_utf8_next_char(nondigit))
ali@68	3148	;
ali@68	3149	/* digits, ending in st, rd, nd, th of either case */
ali@70	3150	if (!g_ascii_strcasecmp(nondigit,"st") \|\|
ali@70	3151	!g_ascii_strcasecmp(nondigit,"rd") \|\|
ali@70	3152	!g_ascii_strcasecmp(nondigit,"nd") \|\|
ali@70	3153	!g_ascii_strcasecmp(nondigit,"th"))
ali@70	3154	query=FALSE;
ali@70	3155	if (!g_ascii_strcasecmp(nondigit,"sts") \|\|
ali@70	3156	!g_ascii_strcasecmp(nondigit,"rds") \|\|
ali@70	3157	!g_ascii_strcasecmp(nondigit,"nds") \|\|
ali@70	3158	!g_ascii_strcasecmp(nondigit,"ths"))
ali@70	3159	query=FALSE;
ali@70	3160	if (!g_ascii_strcasecmp(nondigit,"stly") \|\|
ali@70	3161	!g_ascii_strcasecmp(nondigit,"rdly") \|\|
ali@70	3162	!g_ascii_strcasecmp(nondigit,"ndly") \|\|
ali@70	3163	!g_ascii_strcasecmp(nondigit,"thly"))
ali@70	3164	query=FALSE;
ali@68	3165	/* digits, ending in l, L, s or d */
ali@70	3166	if (!g_ascii_strcasecmp(nondigit,"l") \|\| !strcmp(nondigit,"s") \|\|
ali@70	3167	!strcmp(nondigit,"d"))
ali@70	3168	query=FALSE;
ali@68	3169	/*
ali@40	3170	* L at the start of a number, representing Britsh pounds, like L500.
ali@70	3171	* This is cute. We know the current word is mixed digit. If the first
ali@68	3172	* letter is L, there must be at least one digit following. If both
ali@68	3173	* digits and letters follow, we have a genuine error, else we have a
ali@68	3174	* capital L followed by digits, and we accept that as a non-error.
ali@40	3175	*/
ali@70	3176	if (g_utf8_get_char(checkword)=='L' &&
ali@70	3177	!mixdigit(g_utf8_next_char(checkword)))
ali@70	3178	query=FALSE;
ali@40	3179	}
ali@40	3180	return query;
ali@0	3181	}
ali@0	3182
ali@40	3183	/*
ali@40	3184	* getaword:
ali@40	3185	*
ali@69	3186	* Extracts the first/next "word" from the line, and returns it.
ali@69	3187	* A word is defined as one English word unit--or at least that's the aim.
ali@69	3188	* "ptr" is advanced to the position in the line where we will start
ali@69	3189	* looking for the next word.
ali@40	3190	*
ali@69	3191	* Returns: A newly-allocated string.
ali@40	3192	*/
ali@69	3193	gchar getaword(const char *ptr)
ali@0	3194	{
ali@70	3195	const char s,t;
ali@69	3196	GString *word;
ali@70	3197	gunichar c,pc;
ali@69	3198	word=g_string_new(NULL);
ali@70	3199	for (;!g_unichar_isdigit(g_utf8_get_char(*ptr)) &&
ali@70	3200	!g_unichar_isalpha(g_utf8_get_char(*ptr)) &&
ali@70	3201	*ptr;ptr=g_utf8_next_char(*ptr))
ali@40	3202	;
ali@40	3203	/*
ali@40	3204	* Use a look-ahead to handle exceptions for numbers like 1,000 and 1.35.
ali@40	3205	* Especially yucky is the case of L1,000
ali@40	3206	* This section looks for a pattern of characters including a digit
ali@40	3207	* followed by a comma or period followed by one or more digits.
ali@40	3208	* If found, it returns this whole pattern as a word; otherwise we discard
ali@40	3209	* the results and resume our normal programming.
ali@40	3210	*/
ali@69	3211	s=*ptr;
ali@70	3212	for (;g_unichar_isdigit(g_utf8_get_char(s)) \|\|
ali@70	3213	g_unichar_isalpha(g_utf8_get_char(s)) \|\|
ali@70	3214	g_utf8_get_char(s)==',' \|\| g_utf8_get_char(s)=='.';s=g_utf8_next_char(s))
ali@70	3215	g_string_append_unichar(word,g_utf8_get_char(s));
ali@82	3216	if (word->len)
ali@40	3217	{
ali@82	3218	for (t=g_utf8_next_char(word->str);*t;t=g_utf8_next_char(t))
ali@40	3219	{
ali@82	3220	c=g_utf8_get_char(t);
ali@82	3221	pc=g_utf8_get_char(g_utf8_prev_char(t));
ali@82	3222	if ((c=='.' \|\| c==',') && g_unichar_isdigit(pc))
ali@82	3223	{
ali@82	3224	*ptr=s;
ali@82	3225	return g_string_free(word,FALSE);
ali@82	3226	}
ali@40	3227	}
ali@40	3228	}
ali@0	3229	/* we didn't find a punctuated number - do the regular getword thing */
ali@69	3230	g_string_truncate(word,0);
ali@99	3231	c=g_utf8_get_char(*ptr);
ali@99	3232	for (;g_unichar_isdigit(c) \|\| g_unichar_isalpha(c) \|\| CHAR_IS_APOSTROPHE(c);
ali@99	3233	ptr=g_utf8_next_char(ptr),c=g_utf8_get_char(*ptr))
ali@99	3234	g_string_append_unichar(word,c);
ali@69	3235	return g_string_free(word,FALSE);
ali@0	3236	}
ali@0	3237
ali@40	3238	/*
ali@40	3239	* isroman:
ali@40	3240	*
ali@40	3241	* Is this word a Roman Numeral?
ali@40	3242	*
ali@40	3243	* It doesn't actually validate that the number is a valid Roman Numeral--for
ali@40	3244	* example it will pass MXXXXXXXXXX as a valid Roman Numeral, but that's not
ali@40	3245	* what we're here to do. If it passes this, it LOOKS like a Roman numeral.
ali@40	3246	* Anyway, the actual Romans were pretty tolerant of bad arithmetic, or
ali@40	3247	* expressions thereof, except when it came to taxes. Allow any number of M,
ali@40	3248	* an optional D, an optional CM or CD, any number of optional Cs, an optional
ali@40	3249	* XL or an optional XC, an optional IX or IV, an optional V and any number
ali@40	3250	* of optional Is.
ali@40	3251	*/
ali@69	3252	gboolean isroman(const char *t)
ali@0	3253	{
ali@69	3254	const char *s;
ali@40	3255	if (!t \|\| !*t)
ali@69	3256	return FALSE;
ali@40	3257	s=t;
ali@70	3258	while (g_utf8_get_char(t)=='m' && *t)
ali@40	3259	t++;
ali@70	3260	if (g_utf8_get_char(t)=='d')
ali@40	3261	t++;
ali@70	3262	if (g_str_has_prefix(t,"cm"))
ali@40	3263	t+=2;
ali@70	3264	if (g_str_has_prefix(t,"cd"))
ali@40	3265	t+=2;
ali@70	3266	while (g_utf8_get_char(t)=='c' && *t)
ali@40	3267	t++;
ali@70	3268	if (g_str_has_prefix(t,"xl"))
ali@40	3269	t+=2;
ali@70	3270	if (g_str_has_prefix(t,"xc"))
ali@40	3271	t+=2;
ali@70	3272	if (g_utf8_get_char(t)=='l')
ali@40	3273	t++;
ali@70	3274	while (g_utf8_get_char(t)=='x' && *t)
ali@40	3275	t++;
ali@70	3276	if (g_str_has_prefix(t,"ix"))
ali@40	3277	t+=2;
ali@70	3278	if (g_str_has_prefix(t,"iv"))
ali@40	3279	t+=2;
ali@70	3280	if (g_utf8_get_char(t)=='v')
ali@40	3281	t++;
ali@70	3282	while (g_utf8_get_char(t)=='i' && *t)
ali@40	3283	t++;
ali@40	3284	return !*t;
ali@0	3285	}
ali@0	3286
ali@40	3287	/*
ali@40	3288	* postprocess_for_DP:
ali@40	3289	*
ali@40	3290	* Invoked with the -d switch from flgets().
ali@40	3291	* It simply "removes" from the line a hard-coded set of common
ali@40	3292	* DP-specific tags, so that the line passed to the main routine has
ali@40	3293	* been pre-cleaned of DP markup.
ali@40	3294	*/
ali@0	3295	void postprocess_for_DP(char *theline)
ali@0	3296	{
ali@40	3297	char s,t;
ali@0	3298	int i;
ali@0	3299	if (!*theline)
ali@68	3300	return;
ali@40	3301	for (i=0;*DPmarkup[i];i++)
ali@70	3302	while ((s=strstr(theline,DPmarkup[i])))
ali@40	3303	{
ali@68	3304	t=s+strlen(DPmarkup[i]);
ali@70	3305	memmove(s,t,strlen(t)+1);
ali@40	3306	}
ali@0	3307	}
ali@0	3308
ali@40	3309	/*
ali@40	3310	* postprocess_for_HTML:
ali@40	3311	*
ali@40	3312	* Invoked with the -m switch from flgets().
ali@40	3313	* It simply "removes" from the line a hard-coded set of common
ali@40	3314	* HTML tags and "replaces" a hard-coded set of common HTML
ali@40	3315	* entities, so that the line passed to the main routine has
ali@40	3316	* been pre-cleaned of HTML.
ali@40	3317	*/
ali@0	3318	void postprocess_for_HTML(char *theline)
ali@0	3319	{
ali@70	3320	while (losemarkup(theline))
ali@70	3321	;
ali@71	3322	loseentities(theline);
ali@0	3323	}
ali@0	3324
ali@0	3325	char losemarkup(char theline)
ali@0	3326	{
ali@40	3327	char s,t;
ali@0	3328	int i;
ali@70	3329	s=strchr(theline,'<');
ali@70	3330	t=s?strchr(s,'>'):NULL;
ali@40	3331	if (!s \|\| !t)
ali@40	3332	return NULL;
ali@40	3333	for (i=0;*markup[i];i++)
ali@70	3334	if (tagcomp(g_utf8_next_char(s),markup[i]))
ali@40	3335	{
ali@70	3336	t=g_utf8_next_char(t);
ali@70	3337	memmove(s,t,strlen(t)+1);
ali@70	3338	return s;
ali@68	3339	}
ali@40	3340	/* It's an unrecognized <xxx>. */
ali@40	3341	return NULL;
ali@0	3342	}
ali@0	3343
ali@71	3344	void loseentities(char *theline)
ali@0	3345	{
ali@0	3346	int i;
ali@71	3347	gsize nb;
ali@71	3348	char amp,scolon;
ali@71	3349	gchar s,t;
ali@71	3350	gunichar c;
ali@71	3351	GTree *entities=NULL;
ali@86	3352	static GIConv translit=(GIConv)-1,to_utf8=(GIConv)-1;
ali@71	3353	if (!theline)
ali@40	3354	{
ali@71	3355	if (entities)
ali@71	3356	g_tree_destroy(entities);
ali@71	3357	entities=NULL;
ali@86	3358	if (translit!=(GIConv)-1)
ali@71	3359	g_iconv_close(translit);
ali@71	3360	translit=(GIConv)-1;
ali@86	3361	if (to_utf8!=(GIConv)-1)
ali@71	3362	g_iconv_close(to_utf8);
ali@71	3363	to_utf8=(GIConv)-1;
ali@71	3364	return;
ali@71	3365	}
ali@71	3366	if (!*theline)
ali@71	3367	return;
ali@71	3368	if (!entities)
ali@71	3369	{
ali@71	3370	entities=g_tree_new((GCompareFunc)strcmp);
ali@71	3371	for(i=0;i<G_N_ELEMENTS(HTMLentities);i++)
ali@71	3372	g_tree_insert(entities,HTMLentities[i].name,
ali@71	3373	GUINT_TO_POINTER(HTMLentities[i].c));
ali@71	3374	}
ali@71	3375	if (translit==(GIConv)-1)
ali@71	3376	translit=g_iconv_open("ISO_8859-1//TRANSLIT","UTF-8");
ali@71	3377	if (to_utf8==(GIConv)-1)
ali@71	3378	to_utf8=g_iconv_open("UTF-8","ISO_8859-1");
ali@71	3379	while((amp=strchr(theline,'&')))
ali@71	3380	{
ali@71	3381	scolon=strchr(amp,';');
ali@71	3382	if (scolon)
ali@40	3383	{
ali@71	3384	if (amp[1]=='#')
ali@71	3385	{
ali@71	3386	if (amp+2+strspn(amp+2,"0123456789")==scolon)
ali@71	3387	c=strtol(amp+2,NULL,10);
ali@71	3388	else if (amp[2]=='x' &&
ali@71	3389	amp+3+strspn(amp+3,"0123456789abcdefABCDEF")==scolon)
ali@71	3390	c=strtol(amp+3,NULL,16);
ali@71	3391	}
ali@71	3392	else
ali@71	3393	{
ali@71	3394	s=g_strndup(amp+1,scolon-(amp+1));
ali@71	3395	c=GPOINTER_TO_UINT(g_tree_lookup(entities,s));
ali@71	3396	g_free(s);
ali@71	3397	}
ali@40	3398	}
ali@71	3399	else
ali@71	3400	c=0;
ali@71	3401	if (c)
ali@71	3402	{
ali@71	3403	theline=amp;
ali@71	3404	if (c<128 \|\| c>=192 && c<=255) /* An ISO-8859-1 character */
ali@71	3405	theline+=g_unichar_to_utf8(c,theline);
ali@71	3406	else
ali@71	3407	{
ali@71	3408	s=g_malloc(6);
ali@71	3409	nb=g_unichar_to_utf8(c,s);
ali@71	3410	t=g_convert_with_iconv(s,nb,translit,NULL,&nb,NULL);
ali@71	3411	g_free(s);
ali@71	3412	s=g_convert_with_iconv(t,nb,to_utf8,NULL,&nb,NULL);
ali@71	3413	g_free(t);
ali@71	3414	memcpy(theline,s,nb);
ali@71	3415	g_free(s);
ali@71	3416	theline+=nb;
ali@71	3417	}
ali@71	3418	memmove(theline,g_utf8_next_char(scolon),
ali@71	3419	strlen(g_utf8_next_char(scolon))+1);
ali@71	3420	}
ali@71	3421	else
ali@71	3422	theline=g_utf8_next_char(amp);
ali@40	3423	}
ali@0	3424	}
ali@0	3425
ali@70	3426	gboolean tagcomp(const char strin,const char basetag)
ali@0	3427	{
ali@70	3428	gboolean retval;
ali@70	3429	gchar s,t;
ali@70	3430	if (g_utf8_get_char(strin)=='/')
ali@70	3431	t=g_utf8_casefold(g_utf8_next_char(strin),-1); /* ignore a slash */
ali@70	3432	else
ali@70	3433	t=g_utf8_casefold(strin,-1);
ali@70	3434	s=g_utf8_casefold(basetag,-1);
ali@70	3435	retval=g_str_has_prefix(t,s);
ali@70	3436	g_free(s);
ali@70	3437	g_free(t);
ali@70	3438	return retval;
ali@0	3439	}
ali@0	3440
ali@69	3441	void proghelp(GOptionContext *context)
ali@0	3442	{
ali@69	3443	gchar *help;
ali@40	3444	fputs("Bookloupe version " PACKAGE_VERSION ".\n",stderr);
ali@40	3445	fputs("Copyright 2000-2005 Jim Tinsley <jtinsley@pobox.com>.\n",stderr);
ali@40	3446	fputs("Copyright 2012- J. Ali Harlow <ali@juiblex.co.uk>.\n",stderr);
ali@40	3447	fputs("Bookloupe comes wih ABSOLUTELY NO WARRANTY. "
ali@40	3448	"For details, read the file COPYING.\n",stderr);
ali@40	3449	fputs("This is Free Software; "
ali@40	3450	"you may redistribute it under certain conditions (GPL);\n",stderr);
ali@40	3451	fputs("read the file COPYING for details.\n\n",stderr);
ali@69	3452	help=g_option_context_get_help(context,TRUE,NULL);
ali@69	3453	fputs(help,stderr);
ali@69	3454	g_free(help);
ali@69	3455	fputs("Sample usage: bookloupe warpeace.txt\n\n",stderr);
ali@40	3456	fputs("Bookloupe queries anything it thinks shouldn't be in a PG text; "
ali@40	3457	"non-ASCII\n",stderr);
ali@40	3458	fputs("characters like accented letters, "
ali@40	3459	"lines longer than 75 or shorter than 55,\n",stderr);
ali@40	3460	fputs("unbalanced quotes or brackets, "
ali@40	3461	"a variety of badly formatted punctuation, \n",stderr);
ali@40	3462	fputs("HTML tags, some likely typos. "
ali@40	3463	"It is NOT a substitute for human judgement.\n",stderr);
ali@0	3464	fputs("\n",stderr);
ali@0	3465	}

author	ali <ali@juiblex.co.uk>
	Sun Sep 29 22:51:27 2013 +0100 (2013-09-29)
changeset 177	f2b7980222cf
parent 173	783eff3047bc
child 178	db7b24d83bed
permissions	-rw-r--r--