bookloupe-testing: bookloupe/bookloupe.c@7522c36859d0

     1 /*************************************************************************/

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

     3 /*                                                                       */

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

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

     6 /*                                                                       */

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

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

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

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

    11 /*                                                                       */

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

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

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

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

    16 /*                                                                       */

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

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

    19 /*************************************************************************/

    21 #include <stdio.h>

    22 #include <stdlib.h>

    23 #include <string.h>

    24 #include <ctype.h>

    26 #define MAXWORDLEN    80    /* max length of one word             */

    27 #define LINEBUFSIZE 2048    /* buffer size for an input line      */

    29 #define MAX_USER_TYPOS 1000

    30 #define USERTYPO_FILE "gutcheck.typ"

    32 #ifndef MAX_PATH

    33 #define MAX_PATH 16384

    34 #endif

    36 char aline[LINEBUFSIZE];

    37 char prevline[LINEBUFSIZE];

    39 /* Common typos. */

    40 char *typo[] = {

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

    70     "se", ""

    71 };

    73 char *usertypo[MAX_USER_TYPOS];

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

    76 char *okword[] = {

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

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

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

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

    81 };

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

    84 char *abbrev[] = {

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

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

    87 };

    89 /*

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

    91  * but are common scannos or otherwise common letter combinations.

    92  */

    93 char *nostart[] = {

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

    95 };

    97 /*

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

    99  * but are common scannos or otherwise common letter combinations.

   100  */

   101 char *noend[] = {

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

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

   104 };

   106 char *markup[] = {

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

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

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

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

   111 };

   113 char *DPmarkup[] = {

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

   115 };

   117 char *nocomma[] = {

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

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

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

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

   122 };

   124 char *noperiod[] = {

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

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

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

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

   129 };

   131 char vowels[] = "aeiouàáâãäæèéêëìíîïòóôõöùúûü";

   133 struct {

   134     char *htmlent;

   135     char *htmlnum;

   136     char *textent;

   137 } entities[] = {

   138     "&amp;",	"&#38;",     "&",

   139     "&lt;",	"&#60;",     "<",

   140     "&gt;",	"&#62;",     ">",

   141     "&deg;",	"&#176;",    " degrees",

   142     "&pound;",	"&#163;",    "L",

   143     "&quot;",	"&#34;",     "\"", /* quotation mark = APL quote */

   144     "&OElig;",	"&#338;",    "OE", /* latin capital ligature OE */

   145     "&oelig;",	"&#339;",    "oe", /* latin small ligature oe */

   146     "&Scaron;",	"&#352;",    "S", /* latin capital letter S with caron */

   147     "&scaron;",	"&#353;",    "s", /* latin small letter s with caron */

   148     "&Yuml;",	"&#376;",    "Y", /* latin capital letter Y with diaeresis */

   149     "&circ;",	"&#710;",    "",  /* modifier letter circumflex accent */

   150     "&tilde;",	"&#732;",    "~", /* small tilde, U+02DC ISOdia */

   151     "&ensp;",	"&#8194;",   " ", /* en space, U+2002 ISOpub */

   152     "&emsp;",	"&#8195;",   " ", /* em space, U+2003 ISOpub */

   153     "&thinsp;",	"&#8201;",   " ", /* thin space, U+2009 ISOpub */

   154     "&ndash;",	"&#8211;",   "-", /* en dash, U+2013 ISOpub */

   155     "&mdash;",	"&#8212;",   "--", /* em dash, U+2014 ISOpub */

   156     "&rsquo;",	"&#8217;",   "'", /* right single quotation mark */

   157     "&sbquo;",	"&#8218;",   "'", /* single low-9 quotation mark */

   158     "&ldquo;",	"&#8220;",   "\"", /* left double quotation mark */

   159     "&rdquo;",	"&#8221;",   "\"", /* right double quotation mark */

   160     "&bdquo;",	"&#8222;",   "\"", /* double low-9 quotation mark */

   161     "&lsaquo;",	"&#8249;",   "\"", /* single left-pointing angle quotation mark */

   162     "&rsaquo;",	"&#8250;",   "\"", /* single right-pointing angle quotation mark */

   163     "&nbsp;",	"&#160;",    " ", /* no-break space = non-breaking space, */

   164     "&iexcl;",	"&#161;",    "!", /* inverted exclamation mark */

   165     "&cent;",	"&#162;",    "c", /* cent sign */

   166     "&pound;",	"&#163;",    "L", /* pound sign */

   167     "&curren;",	"&#164;",    "$", /* currency sign */

   168     "&yen;",	"&#165;",    "Y", /* yen sign = yuan sign */

   169     "&sect;",	"&#167;",    "--", /* section sign */

   170     "&uml;",	"&#168;",    " ", /* diaeresis = spacing diaeresis */

   171     "&copy;",	"&#169;",    "(C) ", /* copyright sign */

   172     "&ordf;",	"&#170;",    " ", /* feminine ordinal indicator */

   173     "&laquo;",	"&#171;",    "\"", /* left-pointing double angle quotation mark */

   174     "&shy;",	"&#173;",    "-", /* soft hyphen = discretionary hyphen */

   175     "&reg;",	"&#174;",    "(R) ", /* registered sign = registered trade mark sign */

   176     "&macr;",	"&#175;",    " ", /* macron = spacing macron = overline */

   177     "&deg;",	"&#176;",    " degrees", /* degree sign */

   178     "&plusmn;",	"&#177;",    "+-", /* plus-minus sign = plus-or-minus sign */

   179     "&sup2;",	"&#178;",    "2", /* superscript two = superscript digit two */

   180     "&sup3;",	"&#179;",    "3", /* superscript three = superscript digit three */

   181     "&acute;",	"&#180;",    " ", /* acute accent = spacing acute */

   182     "&micro;",	"&#181;",    "m", /* micro sign */

   183     "&para;",	"&#182;",    "--", /* pilcrow sign = paragraph sign */

   184     "&cedil;",	"&#184;",    " ", /* cedilla = spacing cedilla */

   185     "&sup1;",	"&#185;",    "1", /* superscript one = superscript digit one */

   186     "&ordm;",	"&#186;",    " ", /* masculine ordinal indicator */

   187     "&raquo;",	"&#187;",    "\"", /* right-pointing double angle quotation mark */

   188     "&frac14;",	"&#188;",    "1/4", /* vulgar fraction one quarter */

   189     "&frac12;",	"&#189;",    "1/2", /* vulgar fraction one half */

   190     "&frac34;",	"&#190;",    "3/4", /* vulgar fraction three quarters */

   191     "&iquest;",	"&#191;",    "?", /* inverted question mark */

   192     "&Agrave;",	"&#192;",    "A", /* latin capital letter A with grave */

   193     "&Aacute;",	"&#193;",    "A", /* latin capital letter A with acute */

   194     "&Acirc;",	"&#194;",    "A", /* latin capital letter A with circumflex */

   195     "&Atilde;",	"&#195;",    "A", /* latin capital letter A with tilde */

   196     "&Auml;",	"&#196;",    "A", /* latin capital letter A with diaeresis */

   197     "&Aring;",	"&#197;",    "A", /* latin capital letter A with ring above */

   198     "&AElig;",	"&#198;",    "AE", /* latin capital letter AE */

   199     "&Ccedil;",	"&#199;",    "C", /* latin capital letter C with cedilla */

   200     "&Egrave;",	"&#200;",    "E", /* latin capital letter E with grave */

   201     "&Eacute;",	"&#201;",    "E", /* latin capital letter E with acute */

   202     "&Ecirc;",	"&#202;",    "E", /* latin capital letter E with circumflex */

   203     "&Euml;",	"&#203;",    "E", /* latin capital letter E with diaeresis */

   204     "&Igrave;",	"&#204;",    "I", /* latin capital letter I with grave */

   205     "&Iacute;",	"&#205;",    "I", /* latin capital letter I with acute */

   206     "&Icirc;",	"&#206;",    "I", /* latin capital letter I with circumflex */

   207     "&Iuml;",	"&#207;",    "I", /* latin capital letter I with diaeresis */

   208     "&ETH;",	"&#208;",    "E", /* latin capital letter ETH */

   209     "&Ntilde;",	"&#209;",    "N", /* latin capital letter N with tilde */

   210     "&Ograve;",	"&#210;",    "O", /* latin capital letter O with grave */

   211     "&Oacute;",	"&#211;",    "O", /* latin capital letter O with acute */

   212     "&Ocirc;",	"&#212;",    "O", /* latin capital letter O with circumflex */

   213     "&Otilde;",	"&#213;",    "O", /* latin capital letter O with tilde */

   214     "&Ouml;",	"&#214;",    "O", /* latin capital letter O with diaeresis */

   215     "&times;",	"&#215;",    "*", /* multiplication sign */

   216     "&Oslash;",	"&#216;",    "O", /* latin capital letter O with stroke */

   217     "&Ugrave;",	"&#217;",    "U", /* latin capital letter U with grave */

   218     "&Uacute;",	"&#218;",    "U", /* latin capital letter U with acute */

   219     "&Ucirc;",	"&#219;",    "U", /* latin capital letter U with circumflex */

   220     "&Uuml;",	"&#220;",    "U", /* latin capital letter U with diaeresis */

   221     "&Yacute;",	"&#221;",    "Y", /* latin capital letter Y with acute */

   222     "&THORN;",	"&#222;",    "TH", /* latin capital letter THORN */

   223     "&szlig;",	"&#223;",    "sz", /* latin small letter sharp s = ess-zed */

   224     "&agrave;",	"&#224;",    "a", /* latin small letter a with grave */

   225     "&aacute;",	"&#225;",    "a", /* latin small letter a with acute */

   226     "&acirc;",	"&#226;",    "a", /* latin small letter a with circumflex */

   227     "&atilde;",	"&#227;",    "a", /* latin small letter a with tilde */

   228     "&auml;",	"&#228;",    "a", /* latin small letter a with diaeresis */

   229     "&aring;",	"&#229;",    "a", /* latin small letter a with ring above */

   230     "&aelig;",	"&#230;",    "ae", /* latin small letter ae */

   231     "&ccedil;",	"&#231;",    "c", /* latin small letter c with cedilla */

   232     "&egrave;",	"&#232;",    "e", /* latin small letter e with grave */

   233     "&eacute;",	"&#233;",    "e", /* latin small letter e with acute */

   234     "&ecirc;",	"&#234;",    "e", /* latin small letter e with circumflex */

   235     "&euml;",	"&#235;",    "e", /* latin small letter e with diaeresis */

   236     "&igrave;",	"&#236;",    "i", /* latin small letter i with grave */

   237     "&iacute;",	"&#237;",    "i", /* latin small letter i with acute */

   238     "&icirc;",	"&#238;",    "i", /* latin small letter i with circumflex */

   239     "&iuml;",	"&#239;",    "i", /* latin small letter i with diaeresis */

   240     "&eth;",	"&#240;",    "eth", /* latin small letter eth */

   241     "&ntilde;",	"&#241;",    "n", /* latin small letter n with tilde */

   242     "&ograve;",	"&#242;",    "o", /* latin small letter o with grave */

   243     "&oacute;",	"&#243;",    "o", /* latin small letter o with acute */

   244     "&ocirc;",	"&#244;",    "o", /* latin small letter o with circumflex */

   245     "&otilde;",	"&#245;",    "o", /* latin small letter o with tilde */

   246     "&ouml;",	"&#246;",    "o", /* latin small letter o with diaeresis */

   247     "&divide;",	"&#247;",    "/", /* division sign */

   248     "&oslash;",	"&#248;",    "o", /* latin small letter o with stroke */

   249     "&ugrave;",	"&#249;",    "u", /* latin small letter u with grave */

   250     "&uacute;",	"&#250;",    "u", /* latin small letter u with acute */

   251     "&ucirc;",	"&#251;",    "u", /* latin small letter u with circumflex */

   252     "&uuml;",	"&#252;",    "u", /* latin small letter u with diaeresis */

   253     "&yacute;",	"&#253;",    "y", /* latin small letter y with acute */

   254     "&thorn;",	"&#254;",    "th", /* latin small letter thorn */

   255     "&yuml;",	"&#255;",    "y", /* latin small letter y with diaeresis */

   256     "", ""

   257 };

   259 /* special characters */

   260 #define CHAR_SPACE        32

   261 #define CHAR_TAB           9

   262 #define CHAR_LF           10

   263 #define CHAR_CR           13

   264 #define CHAR_DQUOTE       34

   265 #define CHAR_SQUOTE       39

   266 #define CHAR_OPEN_SQUOTE  96

   267 #define CHAR_TILDE       126

   268 #define CHAR_ASTERISK     42

   269 #define CHAR_FORESLASH    47

   270 #define CHAR_CARAT        94

   272 #define CHAR_UNDERSCORE    '_'

   273 #define CHAR_OPEN_CBRACK   '{'

   274 #define CHAR_CLOSE_CBRACK  '}'

   275 #define CHAR_OPEN_RBRACK   '('

   276 #define CHAR_CLOSE_RBRACK  ')'

   277 #define CHAR_OPEN_SBRACK   '['

   278 #define CHAR_CLOSE_SBRACK  ']'

   280 /* longest and shortest normal PG line lengths */

   281 #define LONGEST_PG_LINE   75

   282 #define WAY_TOO_LONG      80

   283 #define SHORTEST_PG_LINE  55

   285 #define SWITCHES "ESTPXLOYHWVMUD" /* switches:-                            */

   286                                   /*     D - ignore DP-specific markup     */

   287                                   /*     E - echo queried line             */

   288                                   /*     S - check single quotes           */

   289                                   /*     T - check common typos            */

   290                                   /*     P - require closure of quotes on  */

   291                                   /*         every paragraph               */

   292                                   /*     X - "Trust no one" :-) Paranoid!  */

   293                                   /*         Queries everything            */

   294                                   /*     L - line end checking defaults on */

   295                                   /*         -L turns it off               */

   296                                   /*     O - overview. Just shows counts.  */

   297                                   /*     Y - puts errors to stdout         */

   298                                   /*         instead of stderr             */

   299                                   /*     H - Echoes header fields          */

   300                                   /*     M - Ignore markup in < >          */

   301                                   /*     U - Use file of User-defined Typos*/

   302                                   /*     W - Defaults for use on Web upload*/

   303                                   /*     V - Verbose - list EVERYTHING!    */

   304 #define SWITNO 14                 /* max number of switch parms            */

   305                                   /*        - used for defining array-size */

   306 #define MINARGS   1               /* minimum no of args excl switches      */

   307 #define MAXARGS   1               /* maximum no of args excl switches      */

   309 int pswit[SWITNO];                /* program switches set by SWITCHES      */

   311 #define ECHO_SWITCH      0

   312 #define SQUOTE_SWITCH    1

   313 #define TYPO_SWITCH      2

   314 #define QPARA_SWITCH     3

   315 #define PARANOID_SWITCH  4

   316 #define LINE_END_SWITCH  5

   317 #define OVERVIEW_SWITCH  6

   318 #define STDOUT_SWITCH    7

   319 #define HEADER_SWITCH    8

   320 #define WEB_SWITCH       9

   321 #define VERBOSE_SWITCH   10

   322 #define MARKUP_SWITCH    11

   323 #define USERTYPO_SWITCH  12

   324 #define DP_SWITCH        13

   326 long cnt_dquot;       /* for overview mode, count of doublequote queries */

   327 long cnt_squot;       /* for overview mode, count of singlequote queries */

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

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

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

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

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

   333 long cnt_punct;       /* for overview mode, count of punctuation and spacing queries */

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

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

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

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

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

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

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

   342 void proghelp(void);

   343 void procfile(char *);

   345 #define LOW_THRESHOLD    0

   346 #define HIGH_THRESHOLD   1

   348 #define START 0

   349 #define END 1

   350 #define PREV 0

   351 #define NEXT 1

   352 #define FIRST_OF_PAIR 0

   353 #define SECOND_OF_PAIR 1

   355 #define MAX_WORDPAIR 1000

   357 char running_from[MAX_PATH];

   359 int mixdigit(char *);

   360 char *getaword(char *,char *);

   361 int matchword(char *,char *);

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

   363 void lowerit(char *);

   364 int gcisalpha(unsigned char);

   365 int gcisdigit(unsigned char);

   366 int gcisletter(unsigned char);

   367 char *gcstrchr(char *s,char c);

   368 void postprocess_for_HTML(char *);

   369 char *linehasmarkup(char *);

   370 char *losemarkup(char *);

   371 int tagcomp(char *,char *);

   372 char *loseentities(char *);

   373 int isroman(char *);

   374 int usertypo_count;

   375 void postprocess_for_DP(char *);

   377 char wrk[LINEBUFSIZE];

   379 #define MAX_QWORD 50

   380 #define MAX_QWORD_LENGTH 40

   381 char qword[MAX_QWORD][MAX_QWORD_LENGTH];

   382 char qperiod[MAX_QWORD][MAX_QWORD_LENGTH];

   383 signed int dupcnt[MAX_QWORD];

   385 int main(int argc,char **argv)

   386 {

   387     char *argsw,*s;

   388     int i,switno,invarg;

   389     char usertypo_file[MAX_PATH];

   390     FILE *usertypofile;

   391     if (strlen(argv[0])<sizeof(running_from))

   392 	/* save the path to the executable */

   393         strcpy(running_from,argv[0]);

   394     /* find out what directory we're running from */

   395     s=running_from+strlen(running_from);

   396     for (;*s!='/' && *s!='\\' && s>=running_from;s--)

   397         *s=0;

   398     switno=strlen(SWITCHES);

   399     for (i=switno;--i>0;)

   400         pswit[i]=0;           /* initialise switches */

   401     /*

   402      * Standard loop to extract switches.

   403      * When we come out of this loop, the arguments will be

   404      * in argv[0] upwards and the switches used will be

   405      * represented by their equivalent elements in pswit[]

   406      */

   407     while (--argc>0 && **++argv=='-')

   408         for (argsw=argv[0]+1;*argsw!='\0';argsw++)

   409             for (i=switno,invarg=1;(--i>=0) && invarg==1;)

   410                 if ((toupper(*argsw))==SWITCHES[i])

   411 		{

   412                     invarg=0;

   413                     pswit[i]=1;

   414 		}

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

   416     pswit[PARANOID_SWITCH]^=1;

   417     if (pswit[PARANOID_SWITCH])

   418 	/* if running in paranoid mode force typo checks as well   */

   419         pswit[TYPO_SWITCH]=pswit[TYPO_SWITCH]^1;

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

   421     pswit[LINE_END_SWITCH]^=1;

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

   423     pswit[ECHO_SWITCH]^=1;

   424     if (pswit[OVERVIEW_SWITCH])

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

   426         pswit[ECHO_SWITCH]=0;

   427     /*

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

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

   430      */

   431     if (pswit[WEB_SWITCH])

   432     {

   433 	/* specific override for web uploads */

   434         pswit[ECHO_SWITCH]=1;

   435         pswit[SQUOTE_SWITCH]=0;

   436         pswit[TYPO_SWITCH]=1;

   437         pswit[QPARA_SWITCH]=0;

   438         pswit[PARANOID_SWITCH]=1;

   439         pswit[LINE_END_SWITCH]=0;

   440         pswit[OVERVIEW_SWITCH]=0;

   441         pswit[STDOUT_SWITCH]=0;

   442         pswit[HEADER_SWITCH]=1;

   443         pswit[VERBOSE_SWITCH]=0;

   444         pswit[MARKUP_SWITCH]=0;

   445         pswit[USERTYPO_SWITCH]=0;

   446         pswit[DP_SWITCH]=0;

   447     }

   448     if (argc<MINARGS || argc>MAXARGS)

   449     {

   450 	/* check number of args */

   451         proghelp();

   452         return 1;

   453     }

   454     /* read in the user-defined stealth scanno list */

   455     if (pswit[USERTYPO_SWITCH])

   456     {

   457 	/* ... we were told we had one! */

   458         usertypofile=fopen(USERTYPO_FILE,"rb");

   459         if (!usertypofile)

   460 	{

   461 	    /* not in cwd. try excuteable directory. */

   462             strcpy(usertypo_file,running_from);

   463             strcat(usertypo_file,USERTYPO_FILE);

   464             usertypofile=fopen(usertypo_file,"rb");

   465             if (!usertypofile) {

   466 		/* we ain't got no user typo file! */

   467                 printf("   --> I couldn't find gutcheck.typ "

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

   469 	    }

   470 	}

   471         usertypo_count=0;

   472         if (usertypofile)

   473 	{

   474 	    /* we managed to open a User Typo File! */

   475             if (pswit[USERTYPO_SWITCH])

   476 	    {

   477                 while (flgets(aline,LINEBUFSIZE-1,usertypofile,

   478 		  (long)usertypo_count))

   479 		{

   480                     if (strlen(aline)>1)

   481 		    {

   482                         if ((int)*aline>33)

   483 			{

   484                             s=malloc(strlen(aline)+1);

   485                             if (!s)

   486 			    {

   487                                 fprintf(stderr,"bookloupe: cannot get enough "

   488 				  "memory for user typo file!\n");

   489                                 exit(1);

   490 			    }

   491                             strcpy(s,aline);

   492                             usertypo[usertypo_count]=s;

   493                             usertypo_count++;

   494                             if (usertypo_count>=MAX_USER_TYPOS)

   495 			    {

   496                                 printf("   --> Only %d user-defined typos "

   497 				  "allowed: ignoring the rest\n",

   498 				  MAX_USER_TYPOS);

   499                                 break;

   500 			    }

   501 			}

   502 		    }

   503 		}

   504 	    }

   505             fclose(usertypofile);

   506 	}

   507     }

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

   509     cnt_dquot=cnt_squot=cnt_brack=cnt_bin=cnt_odd=cnt_long=

   510     cnt_short=cnt_punct=cnt_dash=cnt_word=cnt_html=cnt_lineend=

   511     cnt_spacend=0;

   512     procfile(argv[0]);

   513     if (pswit[OVERVIEW_SWITCH])

   514     {

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

   516 	  checked_linecnt,linecnt,linecnt-checked_linecnt);

   517         printf("    --------------- Queries found --------------\n");

   518         if (cnt_long)

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

   520         if (cnt_short)

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

   522         if (cnt_lineend)

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

   524         if (cnt_word)

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

   526         if (cnt_dquot)

   527 	    printf("    Unmatched quotes:              %14ld\n",cnt_dquot);

   528         if (cnt_squot)

   529 	    printf("    Unmatched SingleQuotes:        %14ld\n",cnt_squot);

   530         if (cnt_brack)

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

   532         if (cnt_bin)

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

   534         if (cnt_odd)

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

   536         if (cnt_punct)

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

   538         if (cnt_dash)

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

   540         if (cnt_html)

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

   542         printf("\n");

   543         printf("    TOTAL QUERIES                  %14ld\n",

   544           cnt_dquot+cnt_squot+cnt_brack+cnt_bin+cnt_odd+cnt_long+

   545           cnt_short+cnt_punct+cnt_dash+cnt_word+cnt_html+cnt_lineend);

   546     }

   547     return 0;

   548 }

   550 struct first_pass_results {

   551     long firstline,astline;

   552     long footerline,totlen,binlen,alphalen,endquote_count,shortline,dotcomma;

   553     long fslashline,hyphens,longline,verylongline,htmcount,standalone_digit;

   554     long spacedash,emdash,space_emdash,non_PG_space_emdash,PG_space_emdash;

   555     signed int Dutchcount,Frenchcount;

   556 };

   558 /*

   559  * first_pass:

   560  *

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

   562  * file, decide whether to report some things that

   563  * occur many times in the text like long or short

   564  * lines, non-standard dashes, etc.

   565  */

   566 struct first_pass_results *first_pass(FILE *infile)

   567 {

   568     char laststart=CHAR_SPACE,*s;

   569     signed int i,llen;

   570     unsigned int lastlen=0,lastblen=0;

   571     long spline=0,nspline=0;

   572     static struct first_pass_results results={0};

   573     char inword[MAXWORDLEN]="";

   574     while (fgets(aline,LINEBUFSIZE-1,infile))

   575     {

   576         while (aline[strlen(aline)-1]==10 || aline[strlen(aline)-1]==13)

   577 	    aline[strlen(aline)-1]=0;

   578         linecnt++;

   579         if (strstr(aline,"*END") && strstr(aline,"SMALL PRINT") &&

   580 	  (strstr(aline,"PUBLIC DOMAIN") || strstr(aline,"COPYRIGHT")))

   581 	{

   582             if (spline)

   583                 printf("   --> Duplicate header?\n");

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

   585 	}

   586         if (!strncmp(aline,"*** START",9) && strstr(aline,"PROJECT GUTENBERG"))

   587 	{

   588             if (nspline)

   589                 printf("   --> Duplicate header?\n");

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

   591 	}

   592         if (spline || nspline)

   593 	{

   594             lowerit(aline);

   595             if (strstr(aline,"end") && strstr(aline,"project gutenberg"))

   596 	    {

   597                 if (strstr(aline,"end")<strstr(aline,"project gutenberg"))

   598 		{

   599                     if (results.footerline)

   600 		    {

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

   602                         if (!nspline)

   603                             printf("   --> Duplicate footer?\n");

   604 		    }

   605                     else

   606                         results.footerline=linecnt;

   607 		}

   608 	    }

   609 	}

   610         if (spline)

   611 	    results.firstline=spline;

   612         if (nspline)

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

   614         if (results.footerline)

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

   616         llen=strlen(aline);

   617         results.totlen+=llen;

   618         for (i=0;i<llen;i++)

   619 	{

   620             if ((unsigned char)aline[i]>127)

   621 		results.binlen++;

   622             if (gcisalpha(aline[i]))

   623 		results.alphalen++;

   624             if (i>0 && aline[i]==CHAR_DQUOTE && isalpha(aline[i-1]))

   625 		results.endquote_count++;

   626 	}

   627         if (strlen(aline)>2 && lastlen>2 && lastlen<SHORTEST_PG_LINE &&

   628 	  lastblen>2 && lastblen>SHORTEST_PG_LINE && laststart!=CHAR_SPACE)

   629 	    results.shortline++;

   630         if (*aline && (unsigned char)aline[strlen(aline)-1]<=CHAR_SPACE)

   631 	    cnt_spacend++;

   632         if (strstr(aline,".,"))

   633 	    results.dotcomma++;

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

   635         /* locase text on the line */

   636         if (strstr(aline,"*"))

   637 	{

   638             for (s=aline;*s;s++)

   639                 if (*s>='a' && *s<='z')

   640                     break;

   641              if (*s)

   642 		results.astline++;

   643 	}

   644         if (strstr(aline,"/"))

   645             results.fslashline++;

   646         for (i=llen-1;i>0 && (unsigned char)aline[i]<=CHAR_SPACE;i--)

   647 	    ;

   648         if (aline[i]=='-' && aline[i-1]!='-')

   649 	    results.hyphens++;

   650         if (llen>LONGEST_PG_LINE)

   651 	    results.longline++;

   652         if (llen>WAY_TOO_LONG)

   653 	    results.verylongline++;

   654         if (strstr(aline,"<") && strstr(aline,">"))

   655 	{

   656             i=(signed int)(strstr(aline,">")-strstr(aline,"<")+1);

   657             if (i>0)

   658                 results.htmcount++;

   659             if (strstr(aline,"<i>"))

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

   661 	}

   662         /* Check for spaced em-dashes */

   663         if (strstr(aline,"--"))

   664 	{

   665             results.emdash++;

   666             if (*(strstr(aline,"--")-1)==CHAR_SPACE ||

   667                (*(strstr(aline,"--")+2)==CHAR_SPACE))

   668 		results.space_emdash++;

   669             if (*(strstr(aline,"--")-1)==CHAR_SPACE &&

   670                (*(strstr(aline,"--")+2)==CHAR_SPACE))

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

   672 		results.non_PG_space_emdash++;

   673             if (*(strstr(aline,"--")-1)!=CHAR_SPACE &&

   674                (*(strstr(aline,"--")+2)!=CHAR_SPACE))

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

   676 		results.PG_space_emdash++;

   677 	}

   678         for (s=aline;*s;)

   679 	{

   680             s=getaword(s,inword);

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

   682                 results.Dutchcount++;

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

   684                 results.Frenchcount++;

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

   686                 results.standalone_digit++;

   687 	}

   688         /* Check for spaced dashes */

   689         if (strstr(aline," -") && *(strstr(aline," -")+2)!='-')

   690 	    results.spacedash++;

   691         lastblen=lastlen;

   692         lastlen=strlen(aline);

   693         laststart=aline[0];

   694     }

   695     return &results;

   696 }

   698 struct warnings {

   699     signed int shortline,longline,bin,dash,dotcomma,ast,fslash,digit,hyphen;

   700     signed int endquote,isDutch,isFrench;

   701 };

   703 /*

   704  * report_first_pass:

   705  *

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

   707  */

   708 struct warnings *report_first_pass(struct first_pass_results *results)

   709 {

   710     static struct warnings warnings={0};

   711     if (cnt_spacend>0)

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

   713 	  cnt_spacend);

   714     warnings.dotcomma=1;

   715     if (results->dotcomma>5)

   716     {

   717         warnings.dotcomma=0;

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

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

   720     }

   721     /*

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

   723      * don't bother reporting them.

   724      */

   725     warnings.shortline=1;

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

   727     {

   728         warnings.shortline=0;

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

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

   731     }

   732     /*

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

   734      * don't bother reporting them.

   735      */

   736     warnings.longline=1;

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

   738     {

   739         warnings.longline=0;

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

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

   742     }

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

   744     warnings.ast=1;

   745     if (results->astline>10)

   746     {

   747         warnings.ast=0;

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

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

   750     }

   751     /*

   752      * If more than 10 lines contain forward slashes,

   753      * don't bother reporting them.

   754      */

   755     warnings.fslash=1;

   756     if (results->fslashline>10)

   757     {

   758         warnings.fslash=0;

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

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

   761     }

   762     /*

   763      * If more than 20 lines contain unpunctuated endquotes,

   764      * don't bother reporting them.

   765      */

   766     warnings.endquote=1;

   767     if (results->endquote_count>20)

   768     {

   769         warnings.endquote=0;

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

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

   772     }

   773     /*

   774      * If more than 15 lines contain standalone digits,

   775      * don't bother reporting them.

   776      */

   777     warnings.digit=1;

   778     if (results->standalone_digit>10)

   779     {

   780         warnings.digit=0;

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

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

   783     }

   784     /*

   785      * If more than 20 lines contain hyphens at end,

   786      * don't bother reporting them.

   787      */

   788     warnings.hyphen=1;

   789     if (results->hyphens>20)

   790     {

   791         warnings.hyphen=0;

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

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

   794     }

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

   796     {

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

   798         pswit[MARKUP_SWITCH]=1;

   799     }

   800     if (results->verylongline>0)

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

   802 	  results->verylongline);

   803     /*

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

   805      * assume it's deliberate.

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

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

   808      */

   809     warnings.dash=1;

   810     if (results->spacedash+results->non_PG_space_emdash>

   811       results->PG_space_emdash)

   812     {

   813         warnings.dash=0;

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

   815 	  "Not reporting them.\n",

   816 	  results->spacedash+results->non_PG_space_emdash);

   817     }

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

   819     warnings.bin=1;

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

   821     {

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

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

   824         exit(1);

   825     }

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

   827     {

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

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

   830         exit(1);

   831     }

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

   833     {

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

   835 	  "Not reporting them.\n");

   836         warnings.bin=0;

   837     }

   838     warnings.isDutch=0;

   839     if (results->Dutchcount>50)

   840     {

   841         warnings.isDutch=1;

   842         printf("   --> This looks like Dutch - "

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

   844     }

   845     warnings.isFrench=0;

   846     if (results->Frenchcount>50)

   847     {

   848         warnings.isFrench=1;

   849         printf("   --> This looks like French - "

   850 	  "switching off some doublepunct.\n");

   851     }

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

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

   854     else

   855     {

   856         if (results->firstline)

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

   858         if (results->footerline)

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

   860     }

   861     printf("\n");

   862     if (pswit[VERBOSE_SWITCH])

   863     {

   864         warnings.bin=1;

   865         warnings.shortline=1;

   866         warnings.dotcomma=1;

   867         warnings.longline=1;

   868         warnings.dash=1;

   869         warnings.digit=1;

   870         warnings.ast=1;

   871         warnings.fslash=1;

   872         warnings.hyphen=1;

   873         warnings.endquote=1;

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

   875     }

   876     if (warnings.isDutch)

   877         warnings.dash=0;

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

   879       results->footerline>results->firstline &&

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

   881     {

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

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

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

   885 	  "as well.\n");

   886         results->firstline=0;

   887     }

   888     return &warnings;

   889 }

   891 struct counters {

   892     long quot;

   893     signed int c_unders,c_brack,s_brack,r_brack;

   894     signed int open_single_quote,close_single_quote;

   895 };

   897 /*

   898  * analyse_quotes:

   899  *

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

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

   902  * but spaces.

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

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

   905  * separate sections are common.

   906  *

   907  * Returns: Non-zero if the line is empty.

   908  */

   909 int analyse_quotes(const char *s,struct counters *counters)

   910 {

   911     signed int guessquote=0;

   912     int isemptyline=1;    /* assume the line is empty until proven otherwise */

   913     while (*s)

   914     {

   915 	if (*s==CHAR_DQUOTE)

   916 	    counters->quot++;

   917 	if (*s==CHAR_SQUOTE || *s==CHAR_OPEN_SQUOTE)

   918 	{

   919 	    if (s==aline)

   920 	    {

   921 		/*

   922 		 * At start of line, it can only be an openquote.

   923 		 * Hardcode a very common exception!

   924 		 */

   925 		if (strncmp(s+2,"tis",3) && strncmp(s+2,"Tis",3))

   926 		    counters->open_single_quote++;

   927 	    }

   928 	    else if (gcisalpha(s[-1]) && gcisalpha(s[1]))

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

   930 		;

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

   932 	    else if (*s==CHAR_OPEN_SQUOTE || gcisalpha(s[1]))

   933 	    {

   934 		/* it damwell better BE an openquote */

   935 		if (strncmp(s+1,"tis",3) && strncmp(s+1,"Tis",3))

   936 		    /* hardcode a very common exception! */

   937 		    counters->open_single_quote++;

   938 	    }

   939 	    else

   940 	    {

   941 		/* now - is it a closequote? */

   942 		guessquote=0;   /* accumulate clues */

   943 		if (gcisalpha(s[-1]))

   944 		{

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

   946 		    guessquote++;

   947 		    if (s[-1]=='s')

   948 		    {

   949 			/* looks like a plural apostrophe */

   950 			guessquote-=3;

   951 			if (s[1]==CHAR_SPACE)  /* bonus marks! */

   952 			    guessquote-=2;

   953 		    }

   954 		}

   955 		/* it doesn't have a letter either side */

   956 		else if (strchr(".?!,;:",s[-1]) && strchr(".?!,;: ",s[1]))

   957 		    guessquote+=8; /* looks like a closequote */

   958 		else

   959 		    guessquote++;

   960 		if (counters->open_single_quote>counters->close_single_quote)

   961 		    /*

   962 		     * Give it the benefit of some doubt,

   963 		     * if a squote is already open.

   964 		     */

   965 		    guessquote++;

   966 		else

   967 		    guessquote--;

   968 		if (guessquote>=0)

   969 		    counters->close_single_quote++;

   970 	    }

   971 	}

   972 	if (*s!=CHAR_SPACE && *s!='-' && *s!='.' && *s!=CHAR_ASTERISK &&

   973 	  *s!=13 && *s!=10)

   974 	    isemptyline=0;  /* ignore lines like  *  *  *  as spacers */

   975 	if (*s==CHAR_UNDERSCORE)

   976 	    counters->c_unders++;

   977 	if (*s==CHAR_OPEN_CBRACK)

   978 	    counters->c_brack++;

   979 	if (*s==CHAR_CLOSE_CBRACK)

   980 	    counters->c_brack--;

   981 	if (*s==CHAR_OPEN_RBRACK)

   982 	    counters->r_brack++;

   983 	if (*s==CHAR_CLOSE_RBRACK)

   984 	    counters->r_brack--;

   985 	if (*s==CHAR_OPEN_SBRACK)

   986 	    counters->s_brack++;

   987 	if (*s==CHAR_CLOSE_SBRACK)

   988 	    counters->s_brack--;

   989 	s++;

   990     }

   991     return isemptyline;

   992 }

   994 /*

   995  * check_for_odd_characters:

   996  *

   997  * Check for binary and other odd characters.

   998  */

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

  1000   int isemptyline)

  1001 {

  1002     /* Don't repeat multiple warnings on one line. */

  1003     signed int eNon_A=0,eTab=0,eTilde=0,eCarat=0,eFSlash=0,eAst=0;

  1004     const char *s;

  1005     unsigned char c;

  1006     for (s=aline;*s;s++)

  1007     {

  1008 	c=*(unsigned char *)s;

  1009 	if (!eNon_A && (*s<CHAR_SPACE && *s!=9 && *s!='\n' || c>127))

  1010 	{

  1011 	    if (pswit[ECHO_SWITCH])

  1012 		printf("\n%s\n",aline);

  1013 	    if (!pswit[OVERVIEW_SWITCH])

  1014 		if (c>127 && c<160)

  1015 		    printf("    Line %ld column %d - "

  1016 		      "Non-ISO-8859 character %d\n",linecnt,(int)(s-aline)+1,c);

  1017 		else

  1018 		    printf("    Line %ld column %d - Non-ASCII character %d\n",

  1019 		      linecnt,(int)(s-aline)+1,c);

  1020 	    else

  1021 		cnt_bin++;

  1022 	    eNon_A=1;

  1023 	}

  1024 	if (!eTab && *s==CHAR_TAB)

  1025 	{

  1026 	    if (pswit[ECHO_SWITCH])

  1027 		printf("\n%s\n",aline);

  1028 	    if (!pswit[OVERVIEW_SWITCH])

  1029 		printf("    Line %ld column %d - Tab character?\n",

  1030 		  linecnt,(int)(s-aline)+1);

  1031 	    else

  1032 		cnt_odd++;

  1033 	    eTab=1;

  1034 	}

  1035 	if (!eTilde && *s==CHAR_TILDE)

  1036 	{

  1037 	    /*

  1038 	     * Often used by OCR software to indicate an

  1039 	     * unrecognizable character.

  1040 	     */

  1041 	    if (pswit[ECHO_SWITCH])

  1042 		printf("\n%s\n",aline);

  1043 	    if (!pswit[OVERVIEW_SWITCH])

  1044 		printf("    Line %ld column %d - Tilde character?\n",

  1045 		  linecnt,(int)(s-aline)+1);

  1046 	    else

  1047 		cnt_odd++;

  1048 	    eTilde=1;

  1049 	}

  1050 	if (!eCarat && *s==CHAR_CARAT)

  1051 	{

  1052 	    if (pswit[ECHO_SWITCH])

  1053 		printf("\n%s\n",aline);

  1054 	    if (!pswit[OVERVIEW_SWITCH])

  1055 		printf("    Line %ld column %d - Carat character?\n",

  1056 		  linecnt,(int)(s-aline)+1);

  1057 	    else

  1058 		cnt_odd++;

  1059 	    eCarat=1;

  1060 	}

  1061 	if (!eFSlash && *s==CHAR_FORESLASH && warnings->fslash)

  1062 	{

  1063 	    if (pswit[ECHO_SWITCH])

  1064 		printf("\n%s\n",aline);

  1065 	    if (!pswit[OVERVIEW_SWITCH])

  1066 		printf("    Line %ld column %d - Forward slash?\n",

  1067 		  linecnt,(int)(s-aline)+1);

  1068 	    else

  1069 		cnt_odd++;

  1070 	    eFSlash=1;

  1071 	}

  1072 	/*

  1073 	 * Report asterisks only in paranoid mode,

  1074 	 * since they're often deliberate.

  1075 	 */

  1076 	if (!eAst && pswit[PARANOID_SWITCH] && warnings->ast && !isemptyline &&

  1077 	  *s==CHAR_ASTERISK)

  1078 	{

  1079 	    if (pswit[ECHO_SWITCH])

  1080 		printf("\n%s\n",aline);

  1081 	    if (!pswit[OVERVIEW_SWITCH])

  1082 		printf("    Line %ld column %d - Asterisk?\n",

  1083 		  linecnt,(int)(s-aline)+1);

  1084 	    else

  1085 		cnt_odd++;

  1086 	    eAst=1;

  1087 	}

  1088     }

  1089 }

  1091 /*

  1092  * check_for_long_line:

  1093  *

  1094  * Check for line too long.

  1095  */

  1096 void check_for_long_line(const char *aline)

  1097 {

  1098     if (strlen(aline)>LONGEST_PG_LINE)

  1099     {

  1100 	if (pswit[ECHO_SWITCH])

  1101 	    printf("\n%s\n",aline);

  1102 	if (!pswit[OVERVIEW_SWITCH])

  1103 	    printf("    Line %ld column %d - Long line %d\n",

  1104 	      linecnt,strlen(aline),strlen(aline));

  1105 	else

  1106 	    cnt_long++;

  1107     }

  1108 }

  1110 struct line_properties {

  1111     unsigned int len,blen;

  1112     char start;

  1113 };

  1115 /*

  1116  * check_for_short_line:

  1117  *

  1118  * Check for line too short.

  1119  *

  1120  * This one is a bit trickier to implement: we don't want to

  1121  * flag the last line of a paragraph for being short, so we

  1122  * have to wait until we know that our current line is a

  1123  * "normal" line, then report the _previous_ line if it was too

  1124  * short. We also don't want to report indented lines like

  1125  * chapter heads or formatted quotations. We therefore keep

  1126  * last->len as the length of the last line examined, and

  1127  * last->blen as the length of the last but one, and try to

  1128  * suppress unnecessary warnings by checking that both were of

  1129  * "normal" length. We keep the first character of the last

  1130  * line in last->start, and if it was a space, we assume that

  1131  * the formatting is deliberate. I can't figure out a way to

  1132  * distinguish something like a quoted verse left-aligned or

  1133  * the header or footer of a letter from a paragraph of short

  1134  * lines - maybe if I examined the whole paragraph, and if the

  1135  * para has less than, say, 8 lines and if all lines are short,

  1136  * then just assume it's OK? Need to look at some texts to see

  1137  * how often a formula like this would get the right result.

  1138  */

  1139 void check_for_short_line(const char *aline,const struct line_properties *last)

  1140 {

  1141     if (strlen(aline)>1 && last->len>1 && last->len<SHORTEST_PG_LINE &&

  1142       last->blen>1 && last->blen>SHORTEST_PG_LINE && last->start!=CHAR_SPACE)

  1143     {

  1144 	if (pswit[ECHO_SWITCH])

  1145 	    printf("\n%s\n",prevline);

  1146 	if (!pswit[OVERVIEW_SWITCH])

  1147 	    printf("    Line %ld column %d - Short line %d?\n",

  1148 	      linecnt-1,strlen(prevline),strlen(prevline));

  1149 	else

  1150 	    cnt_short++;

  1151     }

  1152 }

  1154 /*

  1155  * check_for_starting_punctuation:

  1156  *

  1157  * Look for punctuation other than full ellipses at start of line.

  1158  */

  1159 void check_for_starting_punctuation(const char *aline)

  1160 {

  1161     if (*aline && strchr(".?!,;:",aline[0]) && strncmp(". . .",aline,5))

  1162     {

  1163 	if (pswit[ECHO_SWITCH])

  1164 	    printf("\n%s\n",aline);

  1165 	if (!pswit[OVERVIEW_SWITCH])

  1166 	    printf("    Line %ld column 1 - Begins with punctuation?\n",

  1167 	      linecnt);

  1168 	else

  1169 	    cnt_punct++;

  1170     }

  1171 }

  1173 /*

  1174  * check_for_spaced_emdash:

  1175  *

  1176  * Check for spaced em-dashes.

  1177  *

  1178  * We must check _all_ occurrences of "--" on the line

  1179  * hence the loop - even if the first double-dash is OK

  1180  * there may be another that's wrong later on.

  1181  */

  1182 void check_for_spaced_emdash(const char *aline)

  1183 {

  1184     const char *s,*t;

  1185     s=aline;

  1186     while ((t=strstr(s,"--")))

  1187     {

  1188 	if (t>aline && t[-1]==CHAR_SPACE || t[2]==CHAR_SPACE)

  1189 	{

  1190 	    if (pswit[ECHO_SWITCH])

  1191 		printf("\n%s\n",aline);

  1192 	    if (!pswit[OVERVIEW_SWITCH])

  1193 		printf("    Line %ld column %d - Spaced em-dash?\n",

  1194 		  linecnt,(int)(t-aline)+1);

  1195 	    else

  1196 		cnt_dash++;

  1197 	}

  1198 	s=t+2;

  1199     }

  1200 }

  1202 /*

  1203  * check_for_spaced_dash:

  1204  *

  1205  * Check for spaced dashes.

  1206  */

  1207 void check_for_spaced_dash(const char *aline)

  1208 {

  1209     const char *s;

  1210     if ((s=strstr(aline," -")))

  1211     {

  1212 	if (s[2]!='-')

  1213 	{

  1214 	    if (pswit[ECHO_SWITCH])

  1215 		printf("\n%s\n",aline);

  1216 	    if (!pswit[OVERVIEW_SWITCH])

  1217 		printf("    Line %ld column %d - Spaced dash?\n",

  1218 		  linecnt,(int)(s-aline)+1);

  1219 	    else

  1220 		cnt_dash++;

  1221 	}

  1222     }

  1223     else if ((s=strstr(aline,"- ")))

  1224     {

  1225 	if (s==aline || s[-1]!='-')

  1226 	{

  1227 	    if (pswit[ECHO_SWITCH])

  1228 		printf("\n%s\n",aline);

  1229 	    if (!pswit[OVERVIEW_SWITCH])

  1230 		printf("    Line %ld column %d - Spaced dash?\n",

  1231 		  linecnt,(int)(s-aline)+1);

  1232 	    else

  1233 		cnt_dash++;

  1234 	}

  1235     }

  1236 }

  1238 /*

  1239  * procfile:

  1240  *

  1241  * Process one file.

  1242  */

  1243 void procfile(char *filename)

  1244 {

  1245     char *s,*t,*s1,*wordstart;

  1246     char inword[MAXWORDLEN],testword[MAXWORDLEN];

  1247     char parastart[81];     /* first line of current para */

  1248     FILE *infile;

  1249     struct first_pass_results *first_pass_results;

  1250     struct warnings *warnings;

  1251     struct counters counters={0};

  1252     struct line_properties last={0};

  1253     int isemptyline;

  1254     long squot,start_para_line;

  1255     signed int i,j,llen,isacro,isellipsis,istypo,alower;

  1256     signed int dquotepar,squotepar;

  1257     signed int isnewpara,vowel,consonant;

  1258     char dquote_err[80],squote_err[80],rbrack_err[80],sbrack_err[80],

  1259       cbrack_err[80],unders_err[80];

  1260     signed int qword_index,qperiod_index,isdup;

  1261     signed int enddash;

  1262     last.start=CHAR_SPACE;

  1263     *dquote_err=*squote_err=*rbrack_err=*cbrack_err=*sbrack_err=

  1264       *unders_err=*prevline=0;

  1265     linecnt=checked_linecnt=start_para_line=0;

  1266     squot=0;

  1267     i=llen=isacro=isellipsis=istypo=0;

  1268     isnewpara=vowel=consonant=enddash=0;

  1269     qword_index=qperiod_index=isdup=0;

  1270     *inword=*testword=0;

  1271     dquotepar=squotepar=0;

  1272     for (j=0;j<MAX_QWORD;j++)

  1273     {

  1274         dupcnt[j]=0;

  1275         for (i=0;i<MAX_QWORD_LENGTH;i++)

  1276 	{

  1277             qword[i][j]=0;

  1278             qperiod[i][j]=0;

  1279 	}

  1280     }

  1281     infile=fopen(filename,"rb");

  1282     if (!infile)

  1283     {

  1284         if (pswit[STDOUT_SWITCH])

  1285             fprintf(stdout,"bookloupe: cannot open %s\n",filename);

  1286         else

  1287             fprintf(stderr,"bookloupe: cannot open %s\n",filename);

  1288 	exit(1);

  1289     }

  1290     fprintf(stdout,"\n\nFile: %s\n\n",filename);

  1291     first_pass_results=first_pass(infile);

  1292     warnings=report_first_pass(first_pass_results);

  1293     rewind(infile);

  1294     /*

  1295      * Here we go with the main pass. Hold onto yer hat!

  1296      * Re-init some variables we've dirtied.

  1297      */

  1298     squot=linecnt=0;

  1299     while (flgets(aline,LINEBUFSIZE-1,infile,linecnt+1))

  1300     {

  1301         linecnt++;

  1302         if (linecnt==1)

  1303 	    isnewpara=1;

  1304         if (pswit[DP_SWITCH] && !strncmp(aline,"-----File: ",11))

  1305 	    continue;    // skip DP page separators completely

  1306         if (linecnt<first_pass_results->firstline ||

  1307 	  (first_pass_results->footerline>0 &&

  1308 	  linecnt>first_pass_results->footerline))

  1309 	{

  1310             if (pswit[HEADER_SWITCH])

  1311 	    {

  1312                 if (!strncmp(aline,"Title:",6))

  1313                     printf("    %s\n",aline);

  1314                 if (!strncmp(aline,"Author:",7))

  1315                     printf("    %s\n",aline);

  1316                 if (!strncmp(aline,"Release Date:",13))

  1317                     printf("    %s\n",aline);

  1318                 if (!strncmp(aline,"Edition:",8))

  1319                     printf("    %s\n\n",aline);

  1320 	    }

  1321             continue;                /* skip through the header */

  1322 	}

  1323         checked_linecnt++;

  1324         s=aline;

  1325         /*

  1326 	 * If we are in a state of unbalanced quotes, and this line

  1327          * doesn't begin with a quote, output the stored error message.

  1328          * If the -P switch was used, print the warning even if the

  1329          * new para starts with quotes.

  1330 	 */

  1331         t=s;

  1332         while (*t==' ')

  1333 	    t++;

  1334         if (*dquote_err)

  1335             if (*t!=CHAR_DQUOTE || pswit[QPARA_SWITCH])

  1336 	    {

  1337                 if (!pswit[OVERVIEW_SWITCH])

  1338 		{

  1339                     if (pswit[ECHO_SWITCH])

  1340 			printf("\n%s\n",parastart);

  1341                     printf(dquote_err);

  1342 		}

  1343                 else

  1344                     cnt_dquot++;

  1345             }

  1346         if (*squote_err)

  1347 	{

  1348             if (*t!=CHAR_SQUOTE && *t!=CHAR_OPEN_SQUOTE ||

  1349 	      pswit[QPARA_SWITCH] || squot)

  1350 	    {

  1351                 if (!pswit[OVERVIEW_SWITCH])

  1352 		{

  1353                     if (pswit[ECHO_SWITCH])

  1354 			printf("\n%s\n",parastart);

  1355                     printf(squote_err);

  1356 		}

  1357                 else

  1358                     cnt_squot++;

  1359 	    }

  1360             squot=0;

  1361 	}

  1362         if (*rbrack_err)

  1363 	{

  1364             if (!pswit[OVERVIEW_SWITCH])

  1365 	    {

  1366                 if (pswit[ECHO_SWITCH])

  1367 		    printf("\n%s\n",parastart);

  1368                 printf(rbrack_err);

  1369 	    }

  1370             else

  1371                 cnt_brack++;

  1372 	}

  1373         if (*sbrack_err)

  1374 	{

  1375             if (!pswit[OVERVIEW_SWITCH])

  1376 	    {

  1377                 if (pswit[ECHO_SWITCH])

  1378 		    printf("\n%s\n",parastart);

  1379                 printf(sbrack_err);

  1380 	    }

  1381             else

  1382                 cnt_brack++;

  1383 	}

  1384         if (*cbrack_err)

  1385 	{

  1386             if (!pswit[OVERVIEW_SWITCH])

  1387 	    {

  1388                 if (pswit[ECHO_SWITCH])

  1389 		    printf("\n%s\n",parastart);

  1390                 printf(cbrack_err);

  1391 	    }

  1392             else

  1393                 cnt_brack++;

  1394 	}

  1395         if (*unders_err)

  1396 	{

  1397             if (!pswit[OVERVIEW_SWITCH])

  1398 	    {

  1399                 if (pswit[ECHO_SWITCH])

  1400 		    printf("\n%s\n",parastart);

  1401                 printf(unders_err);

  1402 	    }

  1403             else

  1404                 cnt_brack++;

  1405 	}

  1406         *dquote_err=*squote_err=*rbrack_err=*cbrack_err=

  1407 	  *sbrack_err=*unders_err=0;

  1408 	isemptyline=analyse_quotes(aline,&counters);

  1409         if (isnewpara && !isemptyline)

  1410 	{

  1411 	    /* This line is the start of a new paragraph. */

  1412             start_para_line=linecnt;

  1413 	    /* Capture its first line in case we want to report it later. */

  1414             strncpy(parastart,aline,80);

  1415             parastart[79]=0;

  1416             dquotepar=squotepar=0; /* restart the quote count */

  1417             s=aline;

  1418             while (!gcisalpha(*s) && !gcisdigit(*s) && *s)

  1419 		s++;

  1420             if (*s>='a' && *s<='z')

  1421 	    {

  1422 		/* and its first letter is lowercase */

  1423                 if (pswit[ECHO_SWITCH])

  1424 		    printf("\n%s\n",aline);

  1425                 if (!pswit[OVERVIEW_SWITCH])

  1426                     printf("    Line %ld column %d - "

  1427 		      "Paragraph starts with lower-case\n",

  1428 		      linecnt,(int)(s-aline)+1);

  1429                 else

  1430                     cnt_punct++;

  1431 	    }

  1432             isnewpara=0; /* Signal the end of new para processing. */

  1433 	}

  1434         /* Check for an em-dash broken at line end. */

  1435         if (enddash && *aline=='-')

  1436 	{

  1437             if (pswit[ECHO_SWITCH])

  1438 		printf("\n%s\n",aline);

  1439             if (!pswit[OVERVIEW_SWITCH])

  1440                 printf("    Line %ld column 1 - Broken em-dash?\n",linecnt);

  1441             else

  1442                 cnt_punct++;

  1443 	}

  1444         enddash=0;

  1445         for (s=aline+strlen(aline)-1;*s==' ' && s>aline;s--)

  1446 	    ;

  1447         if (s>=aline && *s=='-')

  1448             enddash=1;

  1449 	/*

  1450          * Check for invalid or questionable characters in the line

  1451          * Anything above 127 is invalid for plain ASCII, and

  1452          * non-printable control characters should also be flagged.

  1453          * Tabs should generally not be there.

  1454 	 */

  1455         for (s=aline;*s;s++)

  1456 	{

  1457             i=(unsigned char)*s;

  1458             if (i<CHAR_SPACE && i!=CHAR_LF && i!=CHAR_CR && i!=CHAR_TAB)

  1459 	    {

  1460                 if (pswit[ECHO_SWITCH])

  1461 		    printf("\n%s\n",aline);

  1462                 if (!pswit[OVERVIEW_SWITCH])

  1463                     printf("    Line %ld column %d - Control character %d\n",

  1464 		      linecnt,(int)(s-aline)+1,i);

  1465                 else

  1466                     cnt_bin++;

  1467 	    }

  1468 	}

  1469         if (warnings->bin)

  1470 	    check_for_odd_characters(aline,warnings,isemptyline);

  1471         if (warnings->longline)

  1472 	    check_for_long_line(aline);

  1473         if (warnings->shortline)

  1474 	    check_for_short_line(aline,&last);

  1475         last.blen=last.len;

  1476         last.len=strlen(aline);

  1477         last.start=aline[0];

  1478 	check_for_starting_punctuation(aline);

  1479         if (warnings->dash)

  1480 	{

  1481 	    check_for_spaced_emdash(aline);

  1482 	    check_for_spaced_dash(aline);

  1483 	}

  1484         /*

  1485 	 * Check for unmarked paragraphs indicated by separate speakers.

  1486          * May well be false positive:

  1487          * "Bravo!" "Wonderful!" called the crowd.

  1488          * but useful all the same.

  1489 	 */

  1490         s=wrk;

  1491         *s=0;

  1492         if (strstr(aline,"\" \""))

  1493 	    s=strstr(aline,"\" \"");

  1494         if (strstr(aline,"\"  \""))

  1495 	    s=strstr(aline,"\"  \"");

  1496         if (*s)

  1497 	{

  1498             if (pswit[ECHO_SWITCH])

  1499 		printf("\n%s\n",aline);

  1500             if (!pswit[OVERVIEW_SWITCH])

  1501                 printf("    Line %ld column %d - "

  1502 		  "Query missing paragraph break?\n",

  1503 		  linecnt,(int)(s-aline)+1);

  1504             else

  1505                 cnt_punct++;

  1506 	}

  1507         /*

  1508 	 * Check for "to he" and other easy he/be errors.

  1509          * This is a very inadequate effort on the he/be problem,

  1510          * but the phrase "to he" is always an error, whereas "to

  1511          * be" is quite common.

  1512          * Similarly, '"Quiet!", be said.' is a non-be error

  1513          * "to he" is _not_ always an error!:

  1514          *       "Where they went to he couldn't say."

  1515          * Another false positive:

  1516          *       What would "Cinderella" be without the . . .

  1517          * and another: "If he wants to he can see for himself."

  1518 	 */

  1519         s=wrk;

  1520         *s=0;

  1521         if (strstr(aline," to he "))

  1522 	    s=strstr(aline," to he ");

  1523         if (strstr(aline,"\" be "))

  1524 	    s=strstr(aline,"\" be ");

  1525         if (strstr(aline,"\", be "))

  1526 	    s=strstr(aline,"\", be ");

  1527         if (strstr(aline," is be "))

  1528 	    s=strstr(aline," is be ");

  1529         if (strstr(aline," be is "))

  1530 	    s=strstr(aline," be is ");

  1531         if (strstr(aline," was be "))

  1532 	    s=strstr(aline," was be ");

  1533         if (strstr(aline," be would "))

  1534 	    s=strstr(aline," be would ");

  1535         if (strstr(aline," be could "))

  1536 	    s=strstr(aline," be could ");

  1537         if (*s)

  1538 	{

  1539             if (pswit[ECHO_SWITCH])

  1540 		printf("\n%s\n",aline);

  1541             if (!pswit[OVERVIEW_SWITCH])

  1542                 printf("    Line %ld column %d - Query he/be error?\n",

  1543 		  linecnt,(int)(s-aline)+1);

  1544             else

  1545                 cnt_word++;

  1546 	}

  1547         s=wrk;

  1548         *s=0;

  1549         if (strstr(aline," i bad "))

  1550 	    s=strstr(aline," i bad ");

  1551         if (strstr(aline," you bad "))

  1552 	    s=strstr(aline," you bad ");

  1553         if (strstr(aline," he bad "))

  1554 	    s=strstr(aline," he bad ");

  1555         if (strstr(aline," she bad "))

  1556 	    s=strstr(aline," she bad ");

  1557         if (strstr(aline," they bad "))

  1558 	    s=strstr(aline," they bad ");

  1559         if (strstr(aline," a had "))

  1560 	    s=strstr(aline," a had ");

  1561         if (strstr(aline," the had "))

  1562 	    s=strstr(aline," the had ");

  1563         if (*s)

  1564 	{

  1565             if (pswit[ECHO_SWITCH])

  1566 		printf("\n%s\n",aline);

  1567             if (!pswit[OVERVIEW_SWITCH])

  1568                 printf("    Line %ld column %d - Query had/bad error?\n",

  1569 		  linecnt,(int)(s-aline)+1);

  1570             else

  1571                 cnt_word++;

  1572 	}

  1573         s=wrk;

  1574         *s=0;

  1575         if (strstr(aline,", hut "))

  1576 	    s=strstr(aline,", hut ");

  1577         if (strstr(aline,"; hut "))

  1578 	    s=strstr(aline,"; hut ");

  1579         if (*s)

  1580 	{

  1581             if (pswit[ECHO_SWITCH])

  1582 		printf("\n%s\n",aline);

  1583             if (!pswit[OVERVIEW_SWITCH])

  1584                 printf("    Line %ld column %d - Query hut/but error?\n",

  1585 		  linecnt,(int)(s-aline)+1);

  1586             else

  1587                 cnt_word++;

  1588 	}

  1589         /*

  1590 	 * Special case - angled bracket in front of "From" placed there by an

  1591 	 * MTA when sending an e-mail.

  1592 	 */

  1593         if (strstr(aline,">From"))

  1594 	{

  1595             if (pswit[ECHO_SWITCH])

  1596 		printf("\n%s\n",aline);

  1597             if (!pswit[OVERVIEW_SWITCH])

  1598                 printf("    Line %ld column %d - "

  1599 		  "Query angled bracket with From\n",

  1600 		  linecnt,(int)(strstr(aline,">From")-aline)+1);

  1601             else

  1602                 cnt_punct++;

  1603 	}

  1604         /*

  1605 	 * Check for a single character line -

  1606 	 * often an overflow from bad wrapping.

  1607 	 */

  1608         if (*aline && !aline[1])

  1609 	{

  1610             if (*aline=='I' || *aline=='V' || *aline=='X' || *aline=='L' ||

  1611 	      gcisdigit(*aline))

  1612                 ; /* Nothing - ignore numerals alone on a line. */

  1613             else

  1614 	    {

  1615                 if (pswit[ECHO_SWITCH])

  1616 		    printf("\n%s\n",aline);

  1617                 if (!pswit[OVERVIEW_SWITCH])

  1618                     printf("    Line %ld column 1 - "

  1619 		      "Query single character line\n",linecnt);

  1620                 else

  1621                     cnt_punct++;

  1622 	    }

  1623 	}

  1624         /* Check for I" - often should be ! */

  1625         if (strstr(aline," I\""))

  1626 	{

  1627             if (pswit[ECHO_SWITCH])

  1628 		printf("\n%s\n",aline);

  1629             if (!pswit[OVERVIEW_SWITCH])

  1630                 printf("    Line %ld column %ld - Query I=exclamation mark?\n",

  1631 		  linecnt,strstr(aline," I\"")-aline);

  1632             else

  1633                 cnt_punct++;

  1634 	}

  1635         /*

  1636 	 * Check for period without a capital letter. Cut-down from gutspell.

  1637          * Only works when it happens on a single line.

  1638 	 */

  1639         if (pswit[PARANOID_SWITCH])

  1640 	{

  1641             for (t=s=aline;strstr(t,". ");)

  1642 	    {

  1643                 t=strstr(t,". ");

  1644                 if (t==s)

  1645 		{

  1646                     t++;

  1647 		    /* start of line punctuation is handled elsewhere */

  1648                     continue;

  1649 		}

  1650                 if (!gcisalpha(t[-1]))

  1651 		{

  1652                     t++;

  1653                     continue;

  1654 		}

  1655                 if (warnings->isDutch)

  1656 		{

  1657 		    /* For Frank & Jeroen -- 's Middags case */

  1658                     if (t[2]==CHAR_SQUOTE && t[3]>='a' && t[3]<='z' &&

  1659 		      t[4]==CHAR_SPACE && t[5]>='A' && t[5]<='Z')

  1660 		    {

  1661                         t++;

  1662                         continue;

  1663 		    }

  1664 		}

  1665                 s1=t+2;

  1666                 while (*s1 && !gcisalpha(*s1) && !isdigit(*s1))

  1667                     s1++;

  1668                 if (*s1>='a' && *s1<='z')

  1669 		{

  1670 		    /* we have something to investigate */

  1671                     istypo=1;

  1672 		    /* so let's go back and find out */

  1673                     for (s1=t-1;s1>=s &&

  1674 		      (gcisalpha(*s1) || gcisdigit(*s1) || *s1==CHAR_SQUOTE &&

  1675 		      gcisalpha(s1[1]) && gcisalpha(s1[-1]));s1--)

  1676 			;

  1677                     s1++;

  1678                     for (i=0;*s1 && *s1!='.';s1++,i++)

  1679                         testword[i]=*s1;

  1680                     testword[i]=0;

  1681                     for (i=0;*abbrev[i];i++)

  1682                         if (!strcmp(testword,abbrev[i]))

  1683                             istypo=0;

  1684                     if (gcisdigit(*testword))

  1685 			istypo=0;

  1686                     if (!testword[1])

  1687 			istypo=0;

  1688                     if (isroman(testword))

  1689 			istypo=0;

  1690                     if (istypo)

  1691 		    {

  1692                         istypo=0;

  1693                         for (i=0;testword[i];i++)

  1694                             if (strchr(vowels,testword[i]))

  1695                                 istypo=1;

  1696 		    }

  1697                     if (istypo)

  1698 		    {

  1699                         isdup=0;

  1700                         if (strlen(testword)<MAX_QWORD_LENGTH &&

  1701 			  !pswit[VERBOSE_SWITCH])

  1702                             for (i=0;i<qperiod_index;i++)

  1703                                 if (!strcmp(testword,qperiod[i]))

  1704                                     isdup=1;

  1705                         if (!isdup)

  1706 			{

  1707                             if (qperiod_index<MAX_QWORD &&

  1708 			      strlen(testword)<MAX_QWORD_LENGTH)

  1709 			    {

  1710                                 strcpy(qperiod[qperiod_index],testword);

  1711                                 qperiod_index++;

  1712 			    }

  1713                             if (pswit[ECHO_SWITCH])

  1714 				printf("\n%s\n",aline);

  1715                             if (!pswit[OVERVIEW_SWITCH])

  1716                                 printf("    Line %ld column %d - "

  1717 				  "Extra period?\n",linecnt,(int)(t-aline)+1);

  1718                             else

  1719                                 cnt_punct++;

  1720 			}

  1721 		    }

  1722 		}

  1723 	    t++;

  1724 	    }

  1725 	}

  1726         if (pswit[TYPO_SWITCH])

  1727 	{

  1728             /* Check for words usually not followed by punctuation. */

  1729             for (s=aline;*s;)

  1730 	    {

  1731                 wordstart=s;

  1732                 s=getaword(s,inword);

  1733                 if (!*inword)

  1734 		    continue;

  1735                 lowerit(inword);

  1736                 for (i=0;*nocomma[i];i++)

  1737                     if (!strcmp(inword,nocomma[i]))

  1738 		    {

  1739                         if (*s==',' || *s==';' || *s==':')

  1740 			{

  1741                             if (pswit[ECHO_SWITCH])

  1742 				printf("\n%s\n",aline);

  1743                             if (!pswit[OVERVIEW_SWITCH])

  1744                                 printf("    Line %ld column %d - "

  1745 				  "Query punctuation after %s?\n",

  1746 				  linecnt,(int)(s-aline)+1,inword);

  1747                             else

  1748                                 cnt_punct++;

  1749 			}

  1750 		    }

  1751 		for (i=0;*noperiod[i];i++)

  1752                     if (!strcmp(inword,noperiod[i]))

  1753 		    {

  1754                         if (*s=='.' || *s=='!')

  1755 			{

  1756                             if (pswit[ECHO_SWITCH])

  1757 				printf("\n%s\n",aline);

  1758                             if (!pswit[OVERVIEW_SWITCH])

  1759                                 printf("    Line %ld column %d - "

  1760 				  "Query punctuation after %s?\n",

  1761 				  linecnt,(int)(s-aline)+1,inword);

  1762                             else

  1763                                 cnt_punct++;

  1764 			}

  1765 		    }

  1766 	    }

  1767 	}

  1768         /*

  1769 	 * Check for commonly mistyped words,

  1770 	 * and digits like 0 for O in a word.

  1771 	 */

  1772         for (s=aline;*s;)

  1773 	{

  1774             wordstart=s;

  1775             s=getaword(s,inword);

  1776             if (!*inword)

  1777 		continue; /* don't bother with empty lines */

  1778             if (mixdigit(inword))

  1779 	    {

  1780                 if (pswit[ECHO_SWITCH])

  1781 		    printf("\n%s\n",aline);

  1782                 if (!pswit[OVERVIEW_SWITCH])

  1783                     printf("    Line %ld column %d - Query digit in %s\n",

  1784 		      linecnt,(int)(wordstart-aline)+1,inword);

  1785                 else

  1786                     cnt_word++;

  1787 	    }

  1788             /*

  1789 	     * Put the word through a series of tests for likely typos and OCR

  1790 	     * errors.

  1791 	     */

  1792             if (pswit[TYPO_SWITCH])

  1793 	    {

  1794                 istypo=0;

  1795                 strcpy(testword,inword);

  1796                 alower=0;

  1797                 for (i=0;i<(signed int)strlen(testword);i++)

  1798 		{

  1799 		    /* lowercase for testing */

  1800                     if (testword[i]>='a' && testword[i]<='z')

  1801 			alower=1;

  1802                     if (alower && testword[i]>='A' && testword[i]<='Z')

  1803 		    {

  1804                         /*

  1805 			 * We have an uppercase mid-word. However, there are

  1806 			 * common cases:

  1807                          *   Mac and Mc like McGill

  1808                          *   French contractions like l'Abbe

  1809 			 */

  1810                         if (i==2 && testword[0]=='m' && testword[1]=='c' ||

  1811                           i==3 && testword[0]=='m' && testword[1]=='a' &&

  1812 			  testword[2]=='c' || i>0 && testword[i-1]==CHAR_SQUOTE)

  1813 			    ; /* do nothing! */

  1814                         else

  1815                             istypo=1;

  1816 		    }

  1817                     testword[i]=(char)tolower(testword[i]);

  1818 		}

  1819                 /*

  1820 		 * Check for certain unlikely two-letter combinations at word

  1821 		 * start and end.

  1822 		 */

  1823                 if (strlen(testword)>1)

  1824 		{

  1825                     for (i=0;*nostart[i];i++)

  1826                         if (!strncmp(testword,nostart[i],2))

  1827                             istypo=1;

  1828                     for (i=0;*noend[i];i++)

  1829                         if (!strncmp(testword+strlen(testword)-2,noend[i],2))

  1830                             istypo=1;

  1831 		}

  1832                 /* ght is common, gbt never. Like that. */

  1833                 if (strstr(testword,"cb"))

  1834 		    istypo=1;

  1835                 if (strstr(testword,"gbt"))

  1836 		    istypo=1;

  1837                 if (strstr(testword,"pbt"))

  1838 		    istypo=1;

  1839                 if (strstr(testword,"tbs"))

  1840 		    istypo=1;

  1841                 if (strstr(testword,"mrn"))

  1842 		    istypo=1;

  1843                 if (strstr(testword,"ahle"))

  1844 		    istypo=1;

  1845                 if (strstr(testword,"ihle"))

  1846 		    istypo=1;

  1847                 /*

  1848 		 * "TBE" does happen - like HEARTBEAT - but uncommon.

  1849                  * Also "TBI" - frostbite, outbid - but uncommon.

  1850                  * Similarly "ii" like Hawaii, or Pompeii, and in Roman

  1851 		 * numerals, but "ii" is a common scanno.

  1852 		 */

  1853                 if (strstr(testword,"tbi"))

  1854 		    istypo=1;

  1855                 if (strstr(testword,"tbe"))

  1856 		    istypo=1;

  1857                 if (strstr(testword,"ii"))

  1858 		    istypo=1;

  1859                 /*

  1860 		 * Check for no vowels or no consonants.

  1861                  * If none, flag a typo.

  1862 		 */

  1863                 if (!istypo && strlen(testword)>1)

  1864 		{

  1865                     vowel=consonant=0;

  1866                     for (i=0;testword[i];i++)

  1867 		    {

  1868                         if (testword[i]=='y' || gcisdigit(testword[i]))

  1869 			{

  1870 			    /* Yah, this is loose. */

  1871                             vowel++;

  1872                             consonant++;

  1873 			}

  1874                         else if (strchr(vowels,testword[i]))

  1875 			    vowel++;

  1876 			else

  1877 			    consonant++;

  1878 		    }

  1879                     if (!vowel || !consonant)

  1880                         istypo=1;

  1881 		}

  1882                 /*

  1883 		 * Now exclude the word from being reported if it's in

  1884                  * the okword list.

  1885 		 */

  1886                 for (i=0;*okword[i];i++)

  1887                     if (!strcmp(testword,okword[i]))

  1888                         istypo=0;

  1889                 /*

  1890 		 * What looks like a typo may be a Roman numeral.

  1891 		 * Exclude these.

  1892 		 */

  1893                 if (istypo && isroman(testword))

  1894 		    istypo=0;

  1895                 /* Check the manual list of typos. */

  1896                 if (!istypo)

  1897                     for (i=0;*typo[i];i++)

  1898                         if (!strcmp(testword,typo[i]))

  1899                             istypo=1;

  1900                 /*

  1901 		 * Check lowercase s, l, i and m - special cases.

  1902                  *   "j" - often a semi-colon gone wrong.

  1903                  *   "d" for a missing apostrophe - he d

  1904                  *   "n" for "in"

  1905 		 */

  1906                 if (!istypo && strlen(testword)==1 && strchr("slmijdn",*inword))

  1907 		    istypo=1;

  1908                 if (istypo)

  1909 		{

  1910                     isdup=0;

  1911                     if (strlen(testword)<MAX_QWORD_LENGTH &&

  1912 		      !pswit[VERBOSE_SWITCH])

  1913                         for (i=0;i<qword_index;i++)

  1914                             if (!strcmp(testword,qword[i]))

  1915 			    {

  1916                                 isdup=1;

  1917                                 ++dupcnt[i];

  1918 			    }

  1919                     if (!isdup)

  1920 		    {

  1921                         if (qword_index<MAX_QWORD &&

  1922 			  strlen(testword)<MAX_QWORD_LENGTH)

  1923 			{

  1924                             strcpy(qword[qword_index],testword);

  1925                             qword_index++;

  1926 			}

  1927                         if (pswit[ECHO_SWITCH])

  1928 			    printf("\n%s\n",aline);

  1929                         if (!pswit[OVERVIEW_SWITCH])

  1930 			{

  1931                             printf("    Line %ld column %d - Query word %s",

  1932 			      linecnt,(int)(wordstart-aline)+1,inword);

  1933                             if (strlen(testword)<MAX_QWORD_LENGTH &&

  1934 			      !pswit[VERBOSE_SWITCH])

  1935                                 printf(" - not reporting duplicates");

  1936                             printf("\n");

  1937 			}

  1938                         else

  1939                             cnt_word++;

  1940 		    }

  1941 		}

  1942 	    }

  1943 	    /* check the user's list of typos */

  1944 	    if (!istypo && usertypo_count)

  1945 		for (i=0;i<usertypo_count;i++)

  1946 		    if (!strcmp(testword,usertypo[i]))

  1947 		    {

  1948 			if (pswit[ECHO_SWITCH])

  1949 			    printf("\n%s\n",aline);

  1950 			if (!pswit[OVERVIEW_SWITCH])

  1951 			    printf("    Line %ld column %d - "

  1952 			      "Query possible scanno %s\n",

  1953 			      linecnt,(int)(wordstart-aline)+2,inword);

  1954 		    }

  1955             if (pswit[PARANOID_SWITCH] && warnings->digit)

  1956 	    {

  1957 		/* In paranoid mode, query all 0 and 1 standing alone. */

  1958                 if (!strcmp(inword,"0") || !strcmp(inword,"1"))

  1959 		{

  1960                     if (pswit[ECHO_SWITCH])

  1961 			printf("\n%s\n",aline);

  1962                     if (!pswit[OVERVIEW_SWITCH])

  1963                         printf("    Line %ld column %d - Query standalone %s\n",

  1964 			  linecnt,(int)(wordstart-aline)+2,inword);

  1965                     else

  1966                         cnt_word++;

  1967 		}

  1968 	    }

  1969 	}

  1970 	/*

  1971          * Look for added or missing spaces around punctuation and quotes.

  1972          * If there is a punctuation character like ! with no space on

  1973          * either side, suspect a missing!space. If there are spaces on

  1974          * both sides , assume a typo. If we see a double quote with no

  1975          * space or punctuation on either side of it, assume unspaced

  1976          * quotes "like"this.

  1977 	 */

  1978         llen=strlen(aline);

  1979         for (i=1;i<llen;i++)

  1980 	{

  1981 	    /* For each character in the line after the first. */

  1982             if (strchr(".?!,;:_",aline[i]))  /* if it's punctuation */

  1983 	    {

  1984 		/* we need to suppress warnings for acronyms like M.D. */

  1985                 isacro=0;

  1986 		/* we need to suppress warnings for ellipsis . . . */

  1987                 isellipsis=0;

  1988 		/* if there are letters on both sides of it or ... */

  1989                 if (gcisalpha(aline[i-1]) && gcisalpha(aline[i+1]) ||

  1990                    gcisalpha(aline[i+1]) && strchr("?!,;:",aline[i]))

  1991 		{

  1992 		    /* ...if it's strict punctuation followed by an alpha */

  1993                     if (aline[i]=='.')

  1994 		    {

  1995                         if (i>2 && aline[i-2]=='.')

  1996 			    isacro=1;

  1997                         if (i+2<llen && aline[i+2]=='.')

  1998 			    isacro=1;

  1999 		    }

  2000                     if (!isacro)

  2001 		    {

  2002                         if (pswit[ECHO_SWITCH])

  2003 			    printf("\n%s\n",aline);

  2004                         if (!pswit[OVERVIEW_SWITCH])

  2005                             printf("    Line %ld column %d - Missing space?\n",

  2006 			      linecnt,i+1);

  2007                         else

  2008                             cnt_punct++;

  2009 		    }

  2010 		}

  2011                 if (aline[i-1]==CHAR_SPACE &&

  2012 		  (aline[i+1]==CHAR_SPACE || aline[i+1]==0))

  2013 		{

  2014 		    /*

  2015 		     * If there are spaces on both sides,

  2016 		     * or space before and end of line.

  2017 		     */

  2018                     if (aline[i]=='.')

  2019 		    {

  2020                         if (i>2 && aline[i-2]=='.')

  2021 			    isellipsis=1;

  2022                         if (i+2<llen && aline[i+2]=='.')

  2023 			    isellipsis=1;

  2024 		    }

  2025                     if (!isemptyline && !isellipsis)

  2026 		    {

  2027                         if (pswit[ECHO_SWITCH])

  2028 			    printf("\n%s\n",aline);

  2029                         if (!pswit[OVERVIEW_SWITCH])

  2030                             printf("    Line %ld column %d - "

  2031 			      "Spaced punctuation?\n",linecnt,i+1);

  2032                         else

  2033                             cnt_punct++;

  2034 		    }

  2035 		}

  2036 	    }

  2037 	}

  2038         /* Split out the characters that CANNOT be preceded by space. */

  2039         llen=strlen(aline);

  2040         for (i=1;i<llen;i++)

  2041 	{

  2042 	    /* for each character in the line after the first */

  2043             if (strchr("?!,;:",aline[i]))

  2044 	    {

  2045 		/* if it's punctuation that _cannot_ have a space before it */

  2046                 if (aline[i-1]==CHAR_SPACE && !isemptyline &&

  2047 		  aline[i+1]!=CHAR_SPACE)

  2048 		{

  2049 		    /*

  2050 		     * If aline[i+1) DOES == space,

  2051 		     * it was already reported just above.

  2052 		     */

  2053                     if (pswit[ECHO_SWITCH])

  2054 			printf("\n%s\n",aline);

  2055                     if (!pswit[OVERVIEW_SWITCH])

  2056                         printf("    Line %ld column %d - Spaced punctuation?\n",

  2057 			  linecnt,i+1);

  2058                     else

  2059                         cnt_punct++;

  2060 		}

  2061 	    }

  2062 	}

  2063         /*

  2064 	 * Special case " .X" where X is any alpha.

  2065          * This plugs a hole in the acronym code above.

  2066 	 * Inelegant, but maintainable.

  2067 	 */

  2068         llen=strlen(aline);

  2069         for (i=1;i<llen;i++)

  2070 	{

  2071 	    /* for each character in the line after the first */

  2072             if (aline[i]=='.')

  2073 	    {

  2074 		/* if it's a period */

  2075                 if (aline[i-1]==CHAR_SPACE && gcisalpha(aline[i+1]))

  2076 		{

  2077 		    /*

  2078 		     * If the period follows a space and

  2079 		     * is followed by a letter.

  2080 		     */

  2081                     if (pswit[ECHO_SWITCH])

  2082 			printf("\n%s\n",aline);

  2083                     if (!pswit[OVERVIEW_SWITCH])

  2084                         printf("    Line %ld column %d - Spaced punctuation?\n",

  2085 			  linecnt,i+1);

  2086                     else

  2087                         cnt_punct++;

  2088 		}

  2089 	    }

  2090 	}

  2091         for (i=1;i<llen;i++)

  2092 	{

  2093 	    /* for each character in the line after the first */

  2094             if (aline[i]==CHAR_DQUOTE)

  2095 	    {

  2096                 if (!strchr(" _-.'`,;:!/([{?}])",aline[i-1]) &&

  2097 		  !strchr(" _-.'`,;:!/([{?}])",aline[i+1]) && aline[i+1] ||

  2098 		  !strchr(" _-([{'`",aline[i-1]) && gcisalpha(aline[i+1]))

  2099 		{

  2100 		    if (pswit[ECHO_SWITCH])

  2101 			printf("\n%s\n",aline);

  2102 		    if (!pswit[OVERVIEW_SWITCH])

  2103 			printf("    Line %ld column %d - Unspaced quotes?\n",

  2104 			  linecnt,i+1);

  2105 		    else

  2106 			cnt_punct++;

  2107 		}

  2108 	    }

  2109 	}

  2110         /* Check parity of quotes. */

  2111         for (s=aline;*s;s++)

  2112 	{

  2113             if (*s==CHAR_DQUOTE)

  2114 	    {

  2115                 if (!(dquotepar=!dquotepar))

  2116 		{

  2117 		    /* parity even */

  2118                     if (!strchr("_-.'`/,;:!?)]} ",s[1]))

  2119 		    {

  2120                         if (pswit[ECHO_SWITCH])

  2121 			    printf("\n%s\n",aline);

  2122                         if (!pswit[OVERVIEW_SWITCH])

  2123                             printf("    Line %ld column %d - "

  2124 			      "Wrongspaced quotes?\n",linecnt,(int)(s-aline)+1);

  2125                         else

  2126                             cnt_punct++;

  2127 		    }

  2128 		}

  2129                 else

  2130 		{

  2131 		    /* parity odd */

  2132                     if (!gcisalpha(s[1]) && !isdigit(s[1]) &&

  2133 		      !strchr("_-/.'`([{$",s[1]) || !s[1])

  2134 		    {

  2135                         if (pswit[ECHO_SWITCH])

  2136 			    printf("\n%s\n",aline);

  2137                         if (!pswit[OVERVIEW_SWITCH])

  2138                             printf("    Line %ld column %d - "

  2139 			      "Wrongspaced quotes?\n",linecnt,(int)(s-aline)+1);

  2140                         else

  2141                             cnt_punct++;

  2142 		    }

  2143 		}

  2144 	    }

  2145 	}

  2146 	if (*aline==CHAR_DQUOTE)

  2147 	{

  2148 	    if (strchr(",;:!?)]} ",aline[1]))

  2149 	    {

  2150 		if (pswit[ECHO_SWITCH])

  2151 		    printf("\n%s\n",aline);

  2152 		if (!pswit[OVERVIEW_SWITCH])

  2153 		    printf("    Line %ld column 1 - Wrongspaced quotes?\n",

  2154 		      linecnt);

  2155 		else

  2156 		    cnt_punct++;

  2157 	    }

  2158 	}

  2159         if (pswit[SQUOTE_SWITCH])

  2160 	{

  2161             for (s=aline;*s;s++)

  2162 	    {

  2163                 if ((*s==CHAR_SQUOTE || *s==CHAR_OPEN_SQUOTE) &&

  2164 		  (s==aline || s>aline && !gcisalpha(s[-1]) ||

  2165 		  !gcisalpha(s[1])))

  2166 		{

  2167                     if (!(squotepar=!squotepar))

  2168 		    {

  2169 			/* parity even */

  2170                         if (!strchr("_-.'`/\",;:!?)]} ",s[1]))

  2171 			{

  2172                             if (pswit[ECHO_SWITCH])

  2173 				printf("\n%s\n",aline);

  2174                             if (!pswit[OVERVIEW_SWITCH])

  2175                                 printf("    Line %ld column %d - "

  2176 				  "Wrongspaced singlequotes?\n",

  2177 				  linecnt,(int)(s-aline)+1);

  2178                             else

  2179                                 cnt_punct++;

  2180 			}

  2181 		    }

  2182                     else

  2183 		    {

  2184 			/* parity odd */

  2185                         if (!gcisalpha(s[1]) && !isdigit(s[1]) &&

  2186 			  !strchr("_-/\".'`",s[1]) || !s[1])

  2187 			{

  2188                             if (pswit[ECHO_SWITCH])

  2189 				printf("\n%s\n",aline);

  2190                             if (!pswit[OVERVIEW_SWITCH])

  2191                                 printf("    Line %ld column %d - "

  2192 				  "Wrongspaced singlequotes?\n",

  2193 				  linecnt,(int)(s-aline)+1);

  2194                             else

  2195                                 cnt_punct++;

  2196 			}

  2197 		    }

  2198 		}

  2199 	    }

  2200 	}

  2201         /*

  2202 	 * Look for double punctuation like ,. or ,,

  2203          * Thanks to DW for the suggestion!

  2204          * In books with references, ".," and ".;" are common

  2205          * e.g. "etc., etc.," and vol. 1.; vol 3.;

  2206          * OTOH, from my initial tests, there are also fairly

  2207          * common errors. What to do? Make these cases paranoid?

  2208          * ".," is the most common, so warnings->dotcomma is used

  2209          * to suppress detailed reporting if it occurs often.

  2210 	 */

  2211         llen=strlen(aline);

  2212         for (i=0;i<llen;i++)

  2213 	{

  2214 	    /* for each punctuation character in the line */

  2215             if (strchr(".?!,;:",aline[i]) && (strchr(".?!,;:",aline[i+1])) &&

  2216 	      aline[i] && aline[i+1])

  2217 	    {

  2218 		/* followed by punctuation, it's a query, unless . . . */

  2219                 if (aline[i]==aline[i+1] && (aline[i]=='.' || aline[i]=='?' ||

  2220 		  aline[i]=='!') ||

  2221 		  !warnings->dotcomma && aline[i]=='.' && aline[i+1]==',' ||

  2222 		  warnings->isFrench && !strncmp(aline+i,",...",4) ||

  2223 		  warnings->isFrench && !strncmp(aline+i,"...,",4) ||

  2224 		  warnings->isFrench && !strncmp(aline+i,";...",4) ||

  2225 		  warnings->isFrench && !strncmp(aline+i,"...;",4) ||

  2226 		  warnings->isFrench && !strncmp(aline+i,":...",4) ||

  2227 		  warnings->isFrench && !strncmp(aline+i,"...:",4) ||

  2228 		  warnings->isFrench && !strncmp(aline+i,"!...",4) ||

  2229 		  warnings->isFrench && !strncmp(aline+i,"...!",4) ||

  2230 		  warnings->isFrench && !strncmp(aline+i,"?...",4) ||

  2231 		  warnings->isFrench && !strncmp(aline+i,"...?",4))

  2232 		{

  2233 		    if (warnings->isFrench && !strncmp(aline+i,",...",4) ||

  2234 		      warnings->isFrench && !strncmp(aline+i,"...,",4) ||

  2235 		      warnings->isFrench && !strncmp(aline+i,";...",4) ||

  2236 		      warnings->isFrench && !strncmp(aline+i,"...;",4) ||

  2237 		      warnings->isFrench && !strncmp(aline+i,":...",4) ||

  2238 		      warnings->isFrench && !strncmp(aline+i,"...:",4) ||

  2239 		      warnings->isFrench && !strncmp(aline+i,"!...",4) ||

  2240 		      warnings->isFrench && !strncmp(aline+i,"...!",4) ||

  2241 		      warnings->isFrench && !strncmp(aline+i,"?...",4) ||

  2242 		      warnings->isFrench && !strncmp(aline+i,"...?",4))

  2243 			i+=4;

  2244 		    ; /* do nothing for .. !! and ?? which can be legit */

  2245 		}

  2246                 else

  2247 		{

  2248                     if (pswit[ECHO_SWITCH])

  2249 			printf("\n%s\n",aline);

  2250                     if (!pswit[OVERVIEW_SWITCH])

  2251                         printf("    Line %ld column %d - Double punctuation?\n",

  2252 			  linecnt,i+1);

  2253                     else

  2254                         cnt_punct++;

  2255 		}

  2256 	    }

  2257 	}

  2258         s=aline;

  2259         while (strstr(s," \" "))

  2260 	{

  2261             if (pswit[ECHO_SWITCH])

  2262 		printf("\n%s\n",aline);

  2263             if (!pswit[OVERVIEW_SWITCH])

  2264                 printf("    Line %ld column %d - Spaced doublequote?\n",

  2265 		  linecnt,(int)(strstr(s," \" ")-aline+1));

  2266             else

  2267                 cnt_punct++;

  2268             s=strstr(s," \" ")+2;

  2269 	}

  2270         s=aline;

  2271         while (strstr(s," ' "))

  2272 	{

  2273             if (pswit[ECHO_SWITCH])

  2274 		printf("\n%s\n",aline);

  2275             if (!pswit[OVERVIEW_SWITCH])

  2276                 printf("    Line %ld column %d - Spaced singlequote?\n",

  2277 		  linecnt,(int)(strstr(s," ' ")-aline+1));

  2278             else

  2279                 cnt_punct++;

  2280             s=strstr(s," ' ")+2;

  2281 	}

  2282         s=aline;

  2283         while (strstr(s," ` "))

  2284 	{

  2285             if (pswit[ECHO_SWITCH])

  2286 		printf("\n%s\n",aline);

  2287             if (!pswit[OVERVIEW_SWITCH])

  2288                 printf("    Line %ld column %d - Spaced singlequote?\n",

  2289 		  linecnt,(int)(strstr(s," ` ")-aline+1));

  2290             else

  2291                 cnt_punct++;

  2292             s=strstr(s," ` ")+2;

  2293 	}

  2294         /* check special case of 'S instead of 's at end of word */

  2295         s=aline+1;

  2296         while (*s)

  2297 	{

  2298             if (*s==CHAR_SQUOTE && s[1]=='S' && s[-1]>='a' && s[-1]<='z')

  2299 	    {

  2300                 if (pswit[ECHO_SWITCH])

  2301 		    printf("\n%s\n",aline);

  2302                 if (!pswit[OVERVIEW_SWITCH])

  2303                     printf("    Line %ld column %d - Capital \"S\"?\n",

  2304 		      linecnt,(int)(s-aline+2));

  2305                 else

  2306                     cnt_punct++;

  2307 	    }

  2308             s++;

  2309 	}

  2310         /*

  2311 	 * Now check special cases - start and end of line -

  2312          * for single and double quotes. Start is sometimes [sic]

  2313          * but better to query it anyway.

  2314          * While we're here, check for dash at end of line.

  2315 	 */

  2316         llen=strlen(aline);

  2317         if (llen>1)

  2318 	{

  2319             if (aline[llen-1]==CHAR_DQUOTE || aline[llen-1]==CHAR_SQUOTE ||

  2320 	      aline[llen-1]==CHAR_OPEN_SQUOTE)

  2321                 if (aline[llen-2]==CHAR_SPACE)

  2322 		{

  2323                     if (pswit[ECHO_SWITCH])

  2324 			printf("\n%s\n",aline);

  2325                     if (!pswit[OVERVIEW_SWITCH])

  2326                         printf("    Line %ld column %d - Spaced quote?\n",

  2327 			  linecnt,llen);

  2328                     else

  2329                         cnt_punct++;

  2330 		}

  2331             if ((aline[0]==CHAR_SQUOTE || aline[0]==CHAR_OPEN_SQUOTE) &&

  2332 	      aline[1]==CHAR_SPACE)

  2333 	    {

  2334 		if (pswit[ECHO_SWITCH])

  2335 		    printf("\n%s\n",aline);

  2336 		if (!pswit[OVERVIEW_SWITCH])

  2337 		    printf("    Line %ld column 1 - Spaced quote?\n",linecnt);

  2338 		else

  2339 		    cnt_punct++;

  2340 	    }

  2341             /*

  2342 	     * Dash at end of line may well be legit - paranoid mode only

  2343              * and don't report em-dash at line-end.

  2344 	     */

  2345             if (pswit[PARANOID_SWITCH] && warnings->hyphen)

  2346 	    {

  2347                 for (i=llen-1;i>0 && (unsigned char)aline[i]<=CHAR_SPACE;i--)

  2348 		    ;

  2349                 if (aline[i]=='-' && aline[i-1]!='-')

  2350 		{

  2351                     if (pswit[ECHO_SWITCH])

  2352 			printf("\n%s\n",aline);

  2353                     if (!pswit[OVERVIEW_SWITCH])

  2354                         printf("    Line %ld column %d - "

  2355 			  "Hyphen at end of line?\n",linecnt,i);

  2356 		}

  2357 	    }

  2358 	}

  2359         /*

  2360 	 * Brackets are often unspaced, but shouldn't be surrounded by alpha.

  2361          * If so, suspect a scanno like "a]most".

  2362 	 */

  2363         llen=strlen(aline);

  2364         for (i=1;i<llen-1;i++)

  2365 	{

  2366 	    /* for each bracket character in the line except 1st & last */

  2367             if (strchr("{[()]}",aline[i]) && gcisalpha(aline[i-1]) &&

  2368 	      gcisalpha(aline[i+1]))

  2369 	    {

  2370                 if (pswit[ECHO_SWITCH])

  2371 		    printf("\n%s\n",aline);

  2372                 if (!pswit[OVERVIEW_SWITCH])

  2373                     printf("    Line %ld column %d - Unspaced bracket?\n",

  2374 		      linecnt,i);

  2375                 else

  2376                     cnt_punct++;

  2377 	    }

  2378 	}

  2379         llen=strlen(aline);

  2380         if (warnings->endquote)

  2381 	{

  2382             for (i=1;i<llen;i++)

  2383 	    {

  2384 		/* for each character in the line except 1st */

  2385                 if (aline[i]==CHAR_DQUOTE && isalpha(aline[i-1]))

  2386 		{

  2387 		    if (pswit[ECHO_SWITCH])

  2388 			printf("\n%s\n",aline);

  2389 		    if (!pswit[OVERVIEW_SWITCH])

  2390 			printf("    Line %ld column %d - "

  2391 			  "endquote missing punctuation?\n",linecnt,i);

  2392 		    else

  2393 			cnt_punct++;

  2394 		}

  2395 	    }

  2396 	}

  2397 	/*

  2398          * Check for <HTML TAG>.

  2399          * If there is a < in the line, followed at some point

  2400          * by a > then we suspect HTML.

  2401 	 */

  2402         if (strstr(aline,"<") && strstr(aline,">"))

  2403 	{

  2404             i=(signed int)(strstr(aline,">")-strstr(aline,"<")+1);

  2405             if (i>0)

  2406 	    {

  2407                 strncpy(wrk,strstr(aline,"<"),i);

  2408                 wrk[i]=0;

  2409                 if (pswit[ECHO_SWITCH])

  2410 		    printf("\n%s\n",aline);

  2411                 if (!pswit[OVERVIEW_SWITCH])

  2412                     printf("    Line %ld column %d - HTML Tag? %s \n",

  2413 		      linecnt,(int)(strstr(aline,"<")-aline)+1,wrk);

  2414                 else

  2415                     cnt_html++;

  2416 	    }

  2417 	}

  2418         /*

  2419 	 * Check for &symbol; HTML.

  2420          * If there is a & in the line, followed at

  2421          * some point by a ; then we suspect HTML.

  2422 	 */

  2423         if (strstr(aline,"&") && strstr(aline,";"))

  2424 	{

  2425             i=(int)(strstr(aline,";")-strstr(aline,"&")+1);

  2426             for (s=strstr(aline,"&");s<strstr(aline,";");s++)

  2427                 if (*s==CHAR_SPACE)

  2428 		    i=0;                /* Don't report "Jones & Son;" */

  2429             if (i>0)

  2430 	    {

  2431                 strncpy(wrk,strstr(aline,"&"),i);

  2432                 wrk[i]=0;

  2433                 if (pswit[ECHO_SWITCH])

  2434 		    printf("\n%s\n",aline);

  2435                 if (!pswit[OVERVIEW_SWITCH])

  2436                     printf("    Line %ld column %d - HTML symbol? %s \n",

  2437 		      linecnt,(int)(strstr(aline,"&")-aline)+1,wrk);

  2438                 else

  2439                     cnt_html++;

  2440 	    }

  2441 	}

  2442         /*

  2443 	 * At end of paragraph, check for mismatched quotes.

  2444          * We don't want to report an error immediately, since it is a

  2445          * common convention to omit the quotes at end of paragraph if

  2446          * the next paragraph is a continuation of the same speaker.

  2447          * Where this is the case, the next para should begin with a

  2448          * quote, so we store the warning message and only display it

  2449          * at the top of the next iteration if the new para doesn't

  2450          * start with a quote.

  2451          * The -p switch overrides this default, and warns of unclosed

  2452          * quotes on _every_ paragraph, whether the next begins with a

  2453          * quote or not.

  2454 	 */

  2455         if (isemptyline)

  2456 	{

  2457 	    /* end of para - add up the totals */

  2458             if (counters.quot%2)

  2459                 sprintf(dquote_err,"    Line %ld - Mismatched quotes\n",

  2460 		  linecnt);

  2461             if (pswit[SQUOTE_SWITCH] && counters.open_single_quote &&

  2462 	      counters.open_single_quote!=counters.close_single_quote)

  2463                 sprintf(squote_err,"    Line %ld - Mismatched singlequotes?\n",

  2464 		  linecnt);

  2465             if (pswit[SQUOTE_SWITCH] && counters.open_single_quote &&

  2466 	      counters.open_single_quote!=counters.close_single_quote &&

  2467 	      counters.open_single_quote!=counters.close_single_quote+1)

  2468 		/*

  2469 		 * Flag it to be noted regardless of the

  2470 		 * first char of the next para.

  2471 		 */

  2472                 squot=1;

  2473             if (counters.r_brack)

  2474                 sprintf(rbrack_err,"    Line %ld - "

  2475 		  "Mismatched round brackets?\n",linecnt);

  2476             if (counters.s_brack)

  2477                 sprintf(sbrack_err,"    Line %ld - "

  2478 		  "Mismatched square brackets?\n",linecnt);

  2479             if (counters.c_brack)

  2480                 sprintf(cbrack_err,"    Line %ld - "

  2481 		  "Mismatched curly brackets?\n",linecnt);

  2482             if (counters.c_unders%2)

  2483                 sprintf(unders_err,"    Line %ld - Mismatched underscores?\n",

  2484 		  linecnt);

  2485 	    memset(&counters,0,sizeof(counters));

  2486 	    /* let the next iteration know that it's starting a new para */

  2487             isnewpara=1;

  2488 	}

  2489         /*

  2490 	 * Check for omitted punctuation at end of paragraph by working back

  2491 	 * through prevline. DW.

  2492          * Need to check this only for "normal" paras.

  2493          * So what is a "normal" para?

  2494          *    Not normal if one-liner (chapter headings, etc.)

  2495          *    Not normal if doesn't contain at least one locase letter

  2496          *    Not normal if starts with space

  2497 	 */

  2498         if (isemptyline)

  2499 	{

  2500 	    /* end of para */

  2501             for (s=prevline,i=0;*s && !i;s++)

  2502                 if (gcisletter(*s))

  2503 		    /* use i to indicate the presence of a letter on the line */

  2504                     i=1;

  2505             /*

  2506 	     * This next "if" is a problem.

  2507              * If we say "start_para_line <= linecnt - 1", that includes

  2508 	     * one-line "paragraphs" like chapter heads. Lotsa false positives.

  2509              * If we say "start_para_line < linecnt - 1" it doesn't, but then it

  2510              * misses genuine one-line paragraphs.

  2511 	     */

  2512             if (i && last.blen>2 && start_para_line<linecnt-1 &&

  2513 	      *prevline>CHAR_SPACE)

  2514 	    {

  2515                 for (i=strlen(prevline)-1;

  2516 		  (prevline[i]==CHAR_DQUOTE || prevline[i]==CHAR_SQUOTE) &&

  2517 		  prevline[i]>CHAR_SPACE && i>0;

  2518 		  i--)

  2519 		    ;

  2520                 for (;i>0;i--)

  2521 		{

  2522                     if (gcisalpha(prevline[i]))

  2523 		    {

  2524                         if (pswit[ECHO_SWITCH])

  2525 			    printf("\n%s\n",prevline);

  2526                         if (!pswit[OVERVIEW_SWITCH])

  2527                             printf("    Line %ld column %d - "

  2528 			      "No punctuation at para end?\n",

  2529 			      linecnt-1,strlen(prevline));

  2530                         else

  2531                             cnt_punct++;

  2532                         break;

  2533 		    }

  2534                     if (strchr("-.:!([{?}])",prevline[i]))

  2535                         break;

  2536 		}

  2537 	    }

  2538 	}

  2539         strcpy(prevline,aline);

  2540     }

  2541     fclose(infile);

  2542     if (!pswit[OVERVIEW_SWITCH])

  2543         for (i=0;i<MAX_QWORD;i++)

  2544             if (dupcnt[i])

  2545                 printf("\nNote: Queried word %s was duplicated %d time%s\n",

  2546 		  qword[i],dupcnt[i],"s");

  2547 }

  2549 /*

  2550  * flgets:

  2551  *

  2552  * Get one line from the input stream, checking for

  2553  * the existence of exactly one CR/LF line-end per line.

  2554  *

  2555  * Returns: a pointer to the line.

  2556  */

  2557 char *flgets(char *theline,int maxlen,FILE *thefile,long lcnt)

  2558 {

  2559     char c;

  2560     int len,isCR,cint;

  2561     *theline=0;

  2562     len=isCR=0;

  2563     c=cint=fgetc(thefile);

  2564     do

  2565     {

  2566         if (cint==EOF)

  2567             return NULL;

  2568 	/* either way, it's end of line */

  2569         if (c==10)

  2570 	{

  2571             if (isCR)

  2572                 break;

  2573             else

  2574 	    {

  2575 		/* Error - a LF without a preceding CR */

  2576                 if (pswit[LINE_END_SWITCH])

  2577 		{

  2578                     if (pswit[ECHO_SWITCH])

  2579 			printf("\n%s\n",theline);

  2580                     if (!pswit[OVERVIEW_SWITCH])

  2581                         printf("    Line %ld - No CR?\n",lcnt);

  2582                     else

  2583                         cnt_lineend++;

  2584 		}

  2585                 break;

  2586 	    }

  2587 	}

  2588         if (c==13)

  2589 	{

  2590             if (isCR)

  2591 	    {

  2592 		/* Error - two successive CRs */

  2593                 if (pswit[LINE_END_SWITCH])

  2594 		{

  2595                     if (pswit[ECHO_SWITCH])

  2596 			printf("\n%s\n",theline);

  2597                     if (!pswit[OVERVIEW_SWITCH])

  2598                         printf("    Line %ld - Two successive CRs?\n",lcnt);

  2599                     else

  2600                         cnt_lineend++;

  2601 		}

  2602 	    }

  2603             isCR=1;

  2604 	}

  2605         else

  2606 	{

  2607             if (pswit[LINE_END_SWITCH] && isCR)

  2608 	    {

  2609                 if (pswit[ECHO_SWITCH])

  2610 		    printf("\n%s\n",theline);

  2611                 if (!pswit[OVERVIEW_SWITCH])

  2612                     printf("    Line %ld column %d - CR without LF?\n",

  2613 		      lcnt,len+1);

  2614                 else

  2615                     cnt_lineend++;

  2616 	    }

  2617             theline[len]=c;

  2618             len++;

  2619             theline[len]=0;

  2620             isCR=0;

  2621 	}

  2622         c=cint=fgetc(thefile);

  2623     } while(len<maxlen);

  2624     if (pswit[MARKUP_SWITCH])

  2625         postprocess_for_HTML(theline);

  2626     if (pswit[DP_SWITCH])

  2627         postprocess_for_DP(theline);

  2628     return theline;

  2629 }

  2631 /*

  2632  * mixdigit:

  2633  *

  2634  * Takes a "word" as a parameter, and checks whether it

  2635  * contains a mixture of alpha and digits. Generally, this is an

  2636  * error, but may not be for cases like 4th or L5 12s. 3d.

  2637  *

  2638  * Returns: 0 if no error found, 1 if error.

  2639  */

  2640 int mixdigit(char *checkword)

  2641 {

  2642     int wehaveadigit,wehavealetter,firstdigits,query,wl;

  2643     char *s;

  2644     wehaveadigit=wehavealetter=query=0;

  2645     for (s=checkword;*s;s++)

  2646         if (gcisalpha(*s))

  2647             wehavealetter=1;

  2648         else

  2649             if (gcisdigit(*s))

  2650                 wehaveadigit=1;

  2651     if (wehaveadigit && wehavealetter)

  2652     {

  2653 	/* Now exclude common legit cases, like "21st" and "12l. 3s. 11d." */

  2654         query=1;

  2655         wl=strlen(checkword);

  2656         for (firstdigits=0;gcisdigit(checkword[firstdigits]);firstdigits++)

  2657             ;

  2658         /* digits, ending in st, rd, nd, th of either case */

  2659         if (firstdigits+2==wl && (matchword(checkword+wl-2,"st") ||

  2660 	  matchword(checkword+wl-2,"rd") || matchword(checkword+wl-2,"nd") ||

  2661 	  matchword(checkword+wl-2,"th")))

  2662 	    query=0;

  2663         if (firstdigits+3==wl && (matchword(checkword+wl-3,"sts") ||

  2664 	  matchword(checkword+wl-3,"rds") || matchword(checkword+wl-3,"nds") ||

  2665 	  matchword(checkword+wl-3,"ths")))

  2666 	    query=0;

  2667         if (firstdigits+3==wl && (matchword(checkword+wl-4,"stly") ||

  2668 	  matchword(checkword+wl-4,"rdly") ||

  2669 	  matchword(checkword+wl-4,"ndly") || matchword(checkword+wl-4,"thly")))

  2670 	    query=0;

  2671         /* digits, ending in l, L, s or d */

  2672         if (firstdigits+1==wl && (checkword[wl-1]=='l' ||

  2673 	  checkword[wl-1]=='L' || checkword[wl-1]=='s' || checkword[wl-1]=='d'))

  2674 	    query=0;

  2675         /*

  2676 	 * L at the start of a number, representing Britsh pounds, like L500.

  2677          * This is cute. We know the current word is mixeddigit. If the first

  2678          * letter is L, there must be at least one digit following. If both

  2679          * digits and letters follow, we have a genuine error, else we have a

  2680          * capital L followed by digits, and we accept that as a non-error.

  2681 	 */

  2682         if (checkword[0]=='L' && !mixdigit(checkword+1))

  2683 	    query=0;

  2684     }

  2685     return query;

  2686 }

  2688 /*

  2689  * getaword:

  2690  *

  2691  * Extracts the first/next "word" from the line, and puts

  2692  * it into "thisword". A word is defined as one English word unit--or

  2693  * at least that's the aim.

  2694  *

  2695  * Returns: a pointer to the position in the line where we will start

  2696  *          looking for the next word.

  2697  */

  2698 char *getaword(char *fromline,char *thisword)

  2699 {

  2700     int i,wordlen;

  2701     char *s;

  2702     wordlen=0;

  2703     for (;!gcisdigit(*fromline) && !gcisalpha(*fromline) && *fromline;

  2704       fromline++)

  2705 	;

  2706     /*

  2707      * Use a look-ahead to handle exceptions for numbers like 1,000 and 1.35.

  2708      * Especially yucky is the case of L1,000

  2709      * This section looks for a pattern of characters including a digit

  2710      * followed by a comma or period followed by one or more digits.

  2711      * If found, it returns this whole pattern as a word; otherwise we discard

  2712      * the results and resume our normal programming.

  2713      */

  2714     s=fromline;

  2715     for (;(gcisdigit(*s) || gcisalpha(*s) || *s==',' || *s=='.') &&

  2716       wordlen<MAXWORDLEN;s++)

  2717     {

  2718 	thisword[wordlen]=*s;

  2719         wordlen++;

  2720     }

  2721     thisword[wordlen]=0;

  2722     for (i=1;i<wordlen-1;i++)

  2723     {

  2724         if (thisword[i]=='.' || thisword[i]==',')

  2725 	{

  2726             if (gcisdigit(thisword[i-1]) && gcisdigit(thisword[i-1]))

  2727 	    {

  2728                 fromline=s;

  2729                 return fromline;

  2730 	    }

  2731 	}

  2732     }

  2733     /* we didn't find a punctuated number - do the regular getword thing */

  2734     wordlen=0;

  2735     for (;(gcisdigit(*fromline) || gcisalpha(*fromline) || *fromline=='\'') &&

  2736       wordlen<MAXWORDLEN;fromline++)

  2737     {

  2738         thisword[wordlen]=*fromline;

  2739         wordlen++;

  2740     }

  2741     thisword[wordlen]=0;

  2742     return fromline;

  2743 }

  2745 /*

  2746  * matchword:

  2747  *

  2748  * A case-insensitive string matcher.

  2749  */

  2750 int matchword(char *checkfor,char *thisword)

  2751 {

  2752     unsigned int ismatch,i;

  2753     if (strlen(checkfor)!=strlen(thisword))

  2754 	return 0;

  2755     ismatch=1;     /* assume a match until we find a difference */

  2756     for (i=0;i<strlen(checkfor);i++)

  2757         if (toupper(checkfor[i])!=toupper(thisword[i]))

  2758             ismatch=0;

  2759     return ismatch;

  2760 }

  2762 /*

  2763  * lowerit:

  2764  *

  2765  * Lowercase the line.

  2766  */

  2768 void lowerit(char *theline)

  2769 {

  2770     for (;*theline;theline++)

  2771         if (*theline>='A' && *theline<='Z')

  2772             *theline+=32;

  2773 }

  2775 /*

  2776  * isroman:

  2777  *

  2778  * Is this word a Roman Numeral?

  2779  *

  2780  * It doesn't actually validate that the number is a valid Roman Numeral--for

  2781  * example it will pass MXXXXXXXXXX as a valid Roman Numeral, but that's not

  2782  * what we're here to do. If it passes this, it LOOKS like a Roman numeral.

  2783  * Anyway, the actual Romans were pretty tolerant of bad arithmetic, or

  2784  * expressions thereof, except when it came to taxes. Allow any number of M,

  2785  * an optional D, an optional CM or CD, any number of optional Cs, an optional

  2786  * XL or an optional XC, an optional IX or IV, an optional V and any number

  2787  * of optional Is.

  2788  */

  2789 int isroman(char *t)

  2790 {

  2791     char *s;

  2792     if (!t || !*t)

  2793 	return 0;

  2794     s=t;

  2795     while (*t=='m' && *t)

  2796 	t++;

  2797     if (*t=='d')

  2798 	t++;

  2799     if (*t=='c' && t[1]=='m')

  2800 	t+=2;

  2801     if (*t=='c' && t[1]=='d')

  2802 	t+=2;

  2803     while (*t=='c' && *t)

  2804 	t++;

  2805     if (*t=='x' && t[1]=='l')

  2806 	t+=2;

  2807     if (*t=='x' && t[1]=='c')

  2808 	t+=2;

  2809     if (*t=='l')

  2810 	t++;

  2811     while (*t=='x' && *t)

  2812 	t++;

  2813     if (*t=='i' && t[1]=='x')

  2814 	t+=2;

  2815     if (*t=='i' && t[1]=='v')

  2816 	t+=2;

  2817     if (*t=='v')

  2818 	t++;

  2819     while (*t=='i' && *t)

  2820 	t++;

  2821     return !*t;

  2822 }

  2824 /*

  2825  * gcisalpha:

  2826  *

  2827  * A version of isalpha() that is somewhat lenient on 8-bit texts.

  2828  * If we use the standard function, 8-bit accented characters break

  2829  * words, so that tete with accented characters appears to be two words, "t"

  2830  * and "t", with 8-bit characters between them. This causes over-reporting of

  2831  * errors. gcisalpha() recognizes accented letters from the CP1252 (Windows)

  2832  * and ISO-8859-1 character sets, which are the most common PG 8-bit types.

  2833  */

  2834 int gcisalpha(unsigned char c)

  2835 {

  2836     if (c>='a' && c<='z')

  2837 	return 1;

  2838     if (c>='A' && c<='Z')

  2839 	return 1;

  2840     if (c<140)

  2841 	return 0;

  2842     if (c>=192 && c!=208 && c!=215 && c!=222 && c!=240 && c!=247 && c!=254)

  2843 	return 1;

  2844     if (c==140 || c==142 || c==156 || c==158 || c==159)

  2845 	return 1;

  2846     return 0;

  2847 }

  2849 /*

  2850  * gcisdigit:

  2851  *

  2852  * A version of isdigit() that doesn't get confused in 8-bit texts.

  2853  */

  2854 int gcisdigit(unsigned char c)

  2855 {

  2856     return c>='0' && c<='9';

  2857 }

  2859 /*

  2860  * gcisletter:

  2861  *

  2862  * A version of isletter() that doesn't get confused in 8-bit texts.

  2863  * NB: this is ISO-8891-1-specific.

  2864  */

  2865 int gcisletter(unsigned char c)

  2866 {

  2867     return c>='A' && c<='Z' || c>='a' && c<='z' || c>=192;

  2868 }

  2870 /*

  2871  * gcstrchr:

  2872  *

  2873  * Wraps strchr to return NULL if the character being searched for is zero.

  2874  */

  2875 char *gcstrchr(char *s,char c)

  2876 {

  2877     if (!c)

  2878 	return NULL;

  2879     return strchr(s,c);

  2880 }

  2882 /*

  2883  * postprocess_for_DP:

  2884  *

  2885  * Invoked with the -d switch from flgets().

  2886  * It simply "removes" from the line a hard-coded set of common

  2887  * DP-specific tags, so that the line passed to the main routine has

  2888  * been pre-cleaned of DP markup.

  2889  */

  2890 void postprocess_for_DP(char *theline)

  2891 {

  2892     char *s,*t;

  2893     int i;

  2894     if (!*theline)

  2895         return;

  2896     for (i=0;*DPmarkup[i];i++)

  2897     {

  2898         s=strstr(theline,DPmarkup[i]);

  2899         while (s)

  2900 	{

  2901             t=s+strlen(DPmarkup[i]);

  2902             while (*t)

  2903 	    {

  2904                 *s=*t;

  2905                 t++;

  2906 		s++;

  2907 	    }

  2908             *s=0;

  2909             s=strstr(theline,DPmarkup[i]);

  2910 	}

  2911     }

  2912 }

  2914 /*

  2915  * postprocess_for_HTML:

  2916  *

  2917  * Invoked with the -m switch from flgets().

  2918  * It simply "removes" from the line a hard-coded set of common

  2919  * HTML tags and "replaces" a hard-coded set of common HTML

  2920  * entities, so that the line passed to the main routine has

  2921  * been pre-cleaned of HTML.

  2922  */

  2923 void postprocess_for_HTML(char *theline)

  2924 {

  2925     if (strstr(theline,"<") && strstr(theline,">"))

  2926         while (losemarkup(theline))

  2927             ;

  2928     while (loseentities(theline))

  2929         ;

  2930 }

  2932 char *losemarkup(char *theline)

  2933 {

  2934     char *s,*t;

  2935     int i;

  2936     if (!*theline)

  2937         return NULL;

  2938     s=strstr(theline,"<");

  2939     t=strstr(theline,">");

  2940     if (!s || !t)

  2941 	return NULL;

  2942     for (i=0;*markup[i];i++)

  2943         if (!tagcomp(s+1,markup[i]))

  2944 	{

  2945             if (!t[1])

  2946 	    {

  2947                 *s=0;

  2948                 return s;

  2949 	    }

  2950             else if (t>s)

  2951 	    {

  2952 		strcpy(s,t+1);

  2953 		return s;

  2954 	    }

  2955         }

  2956     /* It's an unrecognized <xxx>. */

  2957     return NULL;

  2958 }

  2960 char *loseentities(char *theline)

  2961 {

  2962     int i;

  2963     char *s,*t;

  2964     if (!*theline)

  2965         return NULL;

  2966     for (i=0;*entities[i].htmlent;i++)

  2967     {

  2968         s=strstr(theline,entities[i].htmlent);

  2969         if (s)

  2970 	{

  2971             t=malloc((size_t)strlen(s));

  2972             if (!t)

  2973 		return NULL;

  2974             strcpy(t,s+strlen(entities[i].htmlent));

  2975             strcpy(s,entities[i].textent);

  2976             strcat(s,t);

  2977             free(t);

  2978             return theline;

  2979 	}

  2980     }

  2981     for (i=0;*entities[i].htmlnum;i++)

  2982     {

  2983         s=strstr(theline,entities[i].htmlnum);

  2984         if (s)

  2985 	{

  2986             t=malloc((size_t)strlen(s));

  2987             if (!t)

  2988 		return NULL;

  2989             strcpy(t,s+strlen(entities[i].htmlnum));

  2990             strcpy(s,entities[i].textent);

  2991             strcat(s,t);

  2992             free(t);

  2993             return theline;

  2994 	}

  2995     }

  2996     return NULL;

  2997 }

  2999 int tagcomp(char *strin,char *basetag)

  3000 {

  3001     char *s,*t;

  3002     s=basetag;

  3003     t=strin;

  3004     if (*t=='/')

  3005 	t++; /* ignore a slash */

  3006     while (*s && *t)

  3007     {

  3008         if (tolower(*s)!=tolower(*t))

  3009 	    return 1;

  3010         s++;

  3011 	t++;

  3012     }

  3013     return 0;

  3014 }

  3016 void proghelp()

  3017 {

  3018     fputs("Bookloupe version " PACKAGE_VERSION ".\n",stderr);

  3019     fputs("Copyright 2000-2005 Jim Tinsley <jtinsley@pobox.com>.\n",stderr);

  3020     fputs("Copyright 2012- J. Ali Harlow <ali@juiblex.co.uk>.\n",stderr);

  3021     fputs("Bookloupe comes wih ABSOLUTELY NO WARRANTY. "

  3022       "For details, read the file COPYING.\n",stderr);

  3023     fputs("This is Free Software; "

  3024       "you may redistribute it under certain conditions (GPL);\n",stderr);

  3025     fputs("read the file COPYING for details.\n\n",stderr);

  3026     fputs("Usage is: bookloupe [-setpxloyhud] filename\n",stderr);

  3027     fputs("  where -s checks single quotes, -e suppresses echoing lines, "

  3028       "-t checks typos\n",stderr);

  3029     fputs("  -x (paranoid) switches OFF -t and extra checks, "

  3030       "-l turns OFF line-end checks\n",stderr);

  3031     fputs("  -o just displays overview without detail, "

  3032       "-h echoes header fields\n",stderr);

  3033     fputs("  -v (verbose) unsuppresses duplicate reporting, "

  3034       "-m suppresses markup\n",stderr);

  3035     fputs("  -d ignores DP-specific markup,\n",stderr);

  3036     fputs("  -u uses a file gutcheck.typ to query user-defined "

  3037       "possible typos\n",stderr);

  3038     fputs("Sample usage: bookloupe warpeace.txt \n",stderr);

  3039     fputs("\n",stderr);

  3040     fputs("Bookloupe looks for errors in Project Gutenberg(TM) etexts.\n",

  3041       stderr);

  3042     fputs("Bookloupe queries anything it thinks shouldn't be in a PG text; "

  3043       "non-ASCII\n",stderr);

  3044     fputs("characters like accented letters, "

  3045       "lines longer than 75 or shorter than 55,\n",stderr);

  3046     fputs("unbalanced quotes or brackets, "

  3047       "a variety of badly formatted punctuation, \n",stderr);

  3048     fputs("HTML tags, some likely typos. "

  3049       "It is NOT a substitute for human judgement.\n",stderr);

  3050     fputs("\n",stderr);

  3051 }

author	ali <ali@juiblex.co.uk>
	Sat May 25 20:29:36 2013 +0100 (2013-05-25)
changeset 47	7522c36859d0
parent 46	aa45307a6328
child 48	274aa1e09099
permissions	-rw-r--r--