bookloupe-testing: bookloupe/bookloupe.c@274aa1e09099

     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  * check_for_unmarked_paragraphs:

  1240  *

  1241  * Check for unmarked paragraphs indicated by separate speakers.

  1242  *

  1243  * May well be false positive:

  1244  * "Bravo!" "Wonderful!" called the crowd.

  1245  * but useful all the same.

  1246  */

  1247 void check_for_unmarked_paragraphs(const char *aline)

  1248 {

  1249     const char *s;

  1250     s=strstr(aline,"\"  \"");

  1251     if (!s)

  1252 	s=strstr(aline,"\" \"");

  1253     if (s)

  1254     {

  1255 	if (pswit[ECHO_SWITCH])

  1256 	    printf("\n%s\n",aline);

  1257 	if (!pswit[OVERVIEW_SWITCH])

  1258 	    printf("    Line %ld column %d - Query missing paragraph break?\n",

  1259 	      linecnt,(int)(s-aline)+1);

  1260 	else

  1261 	    cnt_punct++;

  1262     }

  1263 }

  1265 /*

  1266  * procfile:

  1267  *

  1268  * Process one file.

  1269  */

  1270 void procfile(char *filename)

  1271 {

  1272     char *s,*t,*s1,*wordstart;

  1273     char inword[MAXWORDLEN],testword[MAXWORDLEN];

  1274     char parastart[81];     /* first line of current para */

  1275     FILE *infile;

  1276     struct first_pass_results *first_pass_results;

  1277     struct warnings *warnings;

  1278     struct counters counters={0};

  1279     struct line_properties last={0};

  1280     int isemptyline;

  1281     long squot,start_para_line;

  1282     signed int i,j,llen,isacro,isellipsis,istypo,alower;

  1283     signed int dquotepar,squotepar;

  1284     signed int isnewpara,vowel,consonant;

  1285     char dquote_err[80],squote_err[80],rbrack_err[80],sbrack_err[80],

  1286       cbrack_err[80],unders_err[80];

  1287     signed int qword_index,qperiod_index,isdup;

  1288     signed int enddash;

  1289     last.start=CHAR_SPACE;

  1290     *dquote_err=*squote_err=*rbrack_err=*cbrack_err=*sbrack_err=

  1291       *unders_err=*prevline=0;

  1292     linecnt=checked_linecnt=start_para_line=0;

  1293     squot=0;

  1294     i=llen=isacro=isellipsis=istypo=0;

  1295     isnewpara=vowel=consonant=enddash=0;

  1296     qword_index=qperiod_index=isdup=0;

  1297     *inword=*testword=0;

  1298     dquotepar=squotepar=0;

  1299     for (j=0;j<MAX_QWORD;j++)

  1300     {

  1301         dupcnt[j]=0;

  1302         for (i=0;i<MAX_QWORD_LENGTH;i++)

  1303 	{

  1304             qword[i][j]=0;

  1305             qperiod[i][j]=0;

  1306 	}

  1307     }

  1308     infile=fopen(filename,"rb");

  1309     if (!infile)

  1310     {

  1311         if (pswit[STDOUT_SWITCH])

  1312             fprintf(stdout,"bookloupe: cannot open %s\n",filename);

  1313         else

  1314             fprintf(stderr,"bookloupe: cannot open %s\n",filename);

  1315 	exit(1);

  1316     }

  1317     fprintf(stdout,"\n\nFile: %s\n\n",filename);

  1318     first_pass_results=first_pass(infile);

  1319     warnings=report_first_pass(first_pass_results);

  1320     rewind(infile);

  1321     /*

  1322      * Here we go with the main pass. Hold onto yer hat!

  1323      * Re-init some variables we've dirtied.

  1324      */

  1325     squot=linecnt=0;

  1326     while (flgets(aline,LINEBUFSIZE-1,infile,linecnt+1))

  1327     {

  1328         linecnt++;

  1329         if (linecnt==1)

  1330 	    isnewpara=1;

  1331         if (pswit[DP_SWITCH] && !strncmp(aline,"-----File: ",11))

  1332 	    continue;    // skip DP page separators completely

  1333         if (linecnt<first_pass_results->firstline ||

  1334 	  (first_pass_results->footerline>0 &&

  1335 	  linecnt>first_pass_results->footerline))

  1336 	{

  1337             if (pswit[HEADER_SWITCH])

  1338 	    {

  1339                 if (!strncmp(aline,"Title:",6))

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

  1341                 if (!strncmp(aline,"Author:",7))

  1342                     printf("    %s\n",aline);

  1343                 if (!strncmp(aline,"Release Date:",13))

  1344                     printf("    %s\n",aline);

  1345                 if (!strncmp(aline,"Edition:",8))

  1346                     printf("    %s\n\n",aline);

  1347 	    }

  1348             continue;                /* skip through the header */

  1349 	}

  1350         checked_linecnt++;

  1351         s=aline;

  1352         /*

  1353 	 * If we are in a state of unbalanced quotes, and this line

  1354          * doesn't begin with a quote, output the stored error message.

  1355          * If the -P switch was used, print the warning even if the

  1356          * new para starts with quotes.

  1357 	 */

  1358         t=s;

  1359         while (*t==' ')

  1360 	    t++;

  1361         if (*dquote_err)

  1362             if (*t!=CHAR_DQUOTE || pswit[QPARA_SWITCH])

  1363 	    {

  1364                 if (!pswit[OVERVIEW_SWITCH])

  1365 		{

  1366                     if (pswit[ECHO_SWITCH])

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

  1368                     printf(dquote_err);

  1369 		}

  1370                 else

  1371                     cnt_dquot++;

  1372             }

  1373         if (*squote_err)

  1374 	{

  1375             if (*t!=CHAR_SQUOTE && *t!=CHAR_OPEN_SQUOTE ||

  1376 	      pswit[QPARA_SWITCH] || squot)

  1377 	    {

  1378                 if (!pswit[OVERVIEW_SWITCH])

  1379 		{

  1380                     if (pswit[ECHO_SWITCH])

  1381 			printf("\n%s\n",parastart);

  1382                     printf(squote_err);

  1383 		}

  1384                 else

  1385                     cnt_squot++;

  1386 	    }

  1387             squot=0;

  1388 	}

  1389         if (*rbrack_err)

  1390 	{

  1391             if (!pswit[OVERVIEW_SWITCH])

  1392 	    {

  1393                 if (pswit[ECHO_SWITCH])

  1394 		    printf("\n%s\n",parastart);

  1395                 printf(rbrack_err);

  1396 	    }

  1397             else

  1398                 cnt_brack++;

  1399 	}

  1400         if (*sbrack_err)

  1401 	{

  1402             if (!pswit[OVERVIEW_SWITCH])

  1403 	    {

  1404                 if (pswit[ECHO_SWITCH])

  1405 		    printf("\n%s\n",parastart);

  1406                 printf(sbrack_err);

  1407 	    }

  1408             else

  1409                 cnt_brack++;

  1410 	}

  1411         if (*cbrack_err)

  1412 	{

  1413             if (!pswit[OVERVIEW_SWITCH])

  1414 	    {

  1415                 if (pswit[ECHO_SWITCH])

  1416 		    printf("\n%s\n",parastart);

  1417                 printf(cbrack_err);

  1418 	    }

  1419             else

  1420                 cnt_brack++;

  1421 	}

  1422         if (*unders_err)

  1423 	{

  1424             if (!pswit[OVERVIEW_SWITCH])

  1425 	    {

  1426                 if (pswit[ECHO_SWITCH])

  1427 		    printf("\n%s\n",parastart);

  1428                 printf(unders_err);

  1429 	    }

  1430             else

  1431                 cnt_brack++;

  1432 	}

  1433         *dquote_err=*squote_err=*rbrack_err=*cbrack_err=

  1434 	  *sbrack_err=*unders_err=0;

  1435 	isemptyline=analyse_quotes(aline,&counters);

  1436         if (isnewpara && !isemptyline)

  1437 	{

  1438 	    /* This line is the start of a new paragraph. */

  1439             start_para_line=linecnt;

  1440 	    /* Capture its first line in case we want to report it later. */

  1441             strncpy(parastart,aline,80);

  1442             parastart[79]=0;

  1443             dquotepar=squotepar=0; /* restart the quote count */

  1444             s=aline;

  1445             while (!gcisalpha(*s) && !gcisdigit(*s) && *s)

  1446 		s++;

  1447             if (*s>='a' && *s<='z')

  1448 	    {

  1449 		/* and its first letter is lowercase */

  1450                 if (pswit[ECHO_SWITCH])

  1451 		    printf("\n%s\n",aline);

  1452                 if (!pswit[OVERVIEW_SWITCH])

  1453                     printf("    Line %ld column %d - "

  1454 		      "Paragraph starts with lower-case\n",

  1455 		      linecnt,(int)(s-aline)+1);

  1456                 else

  1457                     cnt_punct++;

  1458 	    }

  1459             isnewpara=0; /* Signal the end of new para processing. */

  1460 	}

  1461         /* Check for an em-dash broken at line end. */

  1462         if (enddash && *aline=='-')

  1463 	{

  1464             if (pswit[ECHO_SWITCH])

  1465 		printf("\n%s\n",aline);

  1466             if (!pswit[OVERVIEW_SWITCH])

  1467                 printf("    Line %ld column 1 - Broken em-dash?\n",linecnt);

  1468             else

  1469                 cnt_punct++;

  1470 	}

  1471         enddash=0;

  1472         for (s=aline+strlen(aline)-1;*s==' ' && s>aline;s--)

  1473 	    ;

  1474         if (s>=aline && *s=='-')

  1475             enddash=1;

  1476 	/*

  1477          * Check for invalid or questionable characters in the line

  1478          * Anything above 127 is invalid for plain ASCII, and

  1479          * non-printable control characters should also be flagged.

  1480          * Tabs should generally not be there.

  1481 	 */

  1482         for (s=aline;*s;s++)

  1483 	{

  1484             i=(unsigned char)*s;

  1485             if (i<CHAR_SPACE && i!=CHAR_LF && i!=CHAR_CR && i!=CHAR_TAB)

  1486 	    {

  1487                 if (pswit[ECHO_SWITCH])

  1488 		    printf("\n%s\n",aline);

  1489                 if (!pswit[OVERVIEW_SWITCH])

  1490                     printf("    Line %ld column %d - Control character %d\n",

  1491 		      linecnt,(int)(s-aline)+1,i);

  1492                 else

  1493                     cnt_bin++;

  1494 	    }

  1495 	}

  1496         if (warnings->bin)

  1497 	    check_for_odd_characters(aline,warnings,isemptyline);

  1498         if (warnings->longline)

  1499 	    check_for_long_line(aline);

  1500         if (warnings->shortline)

  1501 	    check_for_short_line(aline,&last);

  1502         last.blen=last.len;

  1503         last.len=strlen(aline);

  1504         last.start=aline[0];

  1505 	check_for_starting_punctuation(aline);

  1506         if (warnings->dash)

  1507 	{

  1508 	    check_for_spaced_emdash(aline);

  1509 	    check_for_spaced_dash(aline);

  1510 	}

  1511 	check_for_unmarked_paragraphs(aline);

  1512         /*

  1513 	 * Check for "to he" and other easy he/be errors.

  1514          * This is a very inadequate effort on the he/be problem,

  1515          * but the phrase "to he" is always an error, whereas "to

  1516          * be" is quite common.

  1517          * Similarly, '"Quiet!", be said.' is a non-be error

  1518          * "to he" is _not_ always an error!:

  1519          *       "Where they went to he couldn't say."

  1520          * Another false positive:

  1521          *       What would "Cinderella" be without the . . .

  1522          * and another: "If he wants to he can see for himself."

  1523 	 */

  1524         s=wrk;

  1525         *s=0;

  1526         if (strstr(aline," to he "))

  1527 	    s=strstr(aline," to he ");

  1528         if (strstr(aline,"\" be "))

  1529 	    s=strstr(aline,"\" be ");

  1530         if (strstr(aline,"\", be "))

  1531 	    s=strstr(aline,"\", be ");

  1532         if (strstr(aline," is be "))

  1533 	    s=strstr(aline," is be ");

  1534         if (strstr(aline," be is "))

  1535 	    s=strstr(aline," be is ");

  1536         if (strstr(aline," was be "))

  1537 	    s=strstr(aline," was be ");

  1538         if (strstr(aline," be would "))

  1539 	    s=strstr(aline," be would ");

  1540         if (strstr(aline," be could "))

  1541 	    s=strstr(aline," be could ");

  1542         if (*s)

  1543 	{

  1544             if (pswit[ECHO_SWITCH])

  1545 		printf("\n%s\n",aline);

  1546             if (!pswit[OVERVIEW_SWITCH])

  1547                 printf("    Line %ld column %d - Query he/be error?\n",

  1548 		  linecnt,(int)(s-aline)+1);

  1549             else

  1550                 cnt_word++;

  1551 	}

  1552         s=wrk;

  1553         *s=0;

  1554         if (strstr(aline," i bad "))

  1555 	    s=strstr(aline," i bad ");

  1556         if (strstr(aline," you bad "))

  1557 	    s=strstr(aline," you bad ");

  1558         if (strstr(aline," he bad "))

  1559 	    s=strstr(aline," he bad ");

  1560         if (strstr(aline," she bad "))

  1561 	    s=strstr(aline," she bad ");

  1562         if (strstr(aline," they bad "))

  1563 	    s=strstr(aline," they bad ");

  1564         if (strstr(aline," a had "))

  1565 	    s=strstr(aline," a had ");

  1566         if (strstr(aline," the had "))

  1567 	    s=strstr(aline," the had ");

  1568         if (*s)

  1569 	{

  1570             if (pswit[ECHO_SWITCH])

  1571 		printf("\n%s\n",aline);

  1572             if (!pswit[OVERVIEW_SWITCH])

  1573                 printf("    Line %ld column %d - Query had/bad error?\n",

  1574 		  linecnt,(int)(s-aline)+1);

  1575             else

  1576                 cnt_word++;

  1577 	}

  1578         s=wrk;

  1579         *s=0;

  1580         if (strstr(aline,", hut "))

  1581 	    s=strstr(aline,", hut ");

  1582         if (strstr(aline,"; hut "))

  1583 	    s=strstr(aline,"; hut ");

  1584         if (*s)

  1585 	{

  1586             if (pswit[ECHO_SWITCH])

  1587 		printf("\n%s\n",aline);

  1588             if (!pswit[OVERVIEW_SWITCH])

  1589                 printf("    Line %ld column %d - Query hut/but error?\n",

  1590 		  linecnt,(int)(s-aline)+1);

  1591             else

  1592                 cnt_word++;

  1593 	}

  1594         /*

  1595 	 * Special case - angled bracket in front of "From" placed there by an

  1596 	 * MTA when sending an e-mail.

  1597 	 */

  1598         if (strstr(aline,">From"))

  1599 	{

  1600             if (pswit[ECHO_SWITCH])

  1601 		printf("\n%s\n",aline);

  1602             if (!pswit[OVERVIEW_SWITCH])

  1603                 printf("    Line %ld column %d - "

  1604 		  "Query angled bracket with From\n",

  1605 		  linecnt,(int)(strstr(aline,">From")-aline)+1);

  1606             else

  1607                 cnt_punct++;

  1608 	}

  1609         /*

  1610 	 * Check for a single character line -

  1611 	 * often an overflow from bad wrapping.

  1612 	 */

  1613         if (*aline && !aline[1])

  1614 	{

  1615             if (*aline=='I' || *aline=='V' || *aline=='X' || *aline=='L' ||

  1616 	      gcisdigit(*aline))

  1617                 ; /* Nothing - ignore numerals alone on a line. */

  1618             else

  1619 	    {

  1620                 if (pswit[ECHO_SWITCH])

  1621 		    printf("\n%s\n",aline);

  1622                 if (!pswit[OVERVIEW_SWITCH])

  1623                     printf("    Line %ld column 1 - "

  1624 		      "Query single character line\n",linecnt);

  1625                 else

  1626                     cnt_punct++;

  1627 	    }

  1628 	}

  1629         /* Check for I" - often should be ! */

  1630         if (strstr(aline," I\""))

  1631 	{

  1632             if (pswit[ECHO_SWITCH])

  1633 		printf("\n%s\n",aline);

  1634             if (!pswit[OVERVIEW_SWITCH])

  1635                 printf("    Line %ld column %ld - Query I=exclamation mark?\n",

  1636 		  linecnt,strstr(aline," I\"")-aline);

  1637             else

  1638                 cnt_punct++;

  1639 	}

  1640         /*

  1641 	 * Check for period without a capital letter. Cut-down from gutspell.

  1642          * Only works when it happens on a single line.

  1643 	 */

  1644         if (pswit[PARANOID_SWITCH])

  1645 	{

  1646             for (t=s=aline;strstr(t,". ");)

  1647 	    {

  1648                 t=strstr(t,". ");

  1649                 if (t==s)

  1650 		{

  1651                     t++;

  1652 		    /* start of line punctuation is handled elsewhere */

  1653                     continue;

  1654 		}

  1655                 if (!gcisalpha(t[-1]))

  1656 		{

  1657                     t++;

  1658                     continue;

  1659 		}

  1660                 if (warnings->isDutch)

  1661 		{

  1662 		    /* For Frank & Jeroen -- 's Middags case */

  1663                     if (t[2]==CHAR_SQUOTE && t[3]>='a' && t[3]<='z' &&

  1664 		      t[4]==CHAR_SPACE && t[5]>='A' && t[5]<='Z')

  1665 		    {

  1666                         t++;

  1667                         continue;

  1668 		    }

  1669 		}

  1670                 s1=t+2;

  1671                 while (*s1 && !gcisalpha(*s1) && !isdigit(*s1))

  1672                     s1++;

  1673                 if (*s1>='a' && *s1<='z')

  1674 		{

  1675 		    /* we have something to investigate */

  1676                     istypo=1;

  1677 		    /* so let's go back and find out */

  1678                     for (s1=t-1;s1>=s &&

  1679 		      (gcisalpha(*s1) || gcisdigit(*s1) || *s1==CHAR_SQUOTE &&

  1680 		      gcisalpha(s1[1]) && gcisalpha(s1[-1]));s1--)

  1681 			;

  1682                     s1++;

  1683                     for (i=0;*s1 && *s1!='.';s1++,i++)

  1684                         testword[i]=*s1;

  1685                     testword[i]=0;

  1686                     for (i=0;*abbrev[i];i++)

  1687                         if (!strcmp(testword,abbrev[i]))

  1688                             istypo=0;

  1689                     if (gcisdigit(*testword))

  1690 			istypo=0;

  1691                     if (!testword[1])

  1692 			istypo=0;

  1693                     if (isroman(testword))

  1694 			istypo=0;

  1695                     if (istypo)

  1696 		    {

  1697                         istypo=0;

  1698                         for (i=0;testword[i];i++)

  1699                             if (strchr(vowels,testword[i]))

  1700                                 istypo=1;

  1701 		    }

  1702                     if (istypo)

  1703 		    {

  1704                         isdup=0;

  1705                         if (strlen(testword)<MAX_QWORD_LENGTH &&

  1706 			  !pswit[VERBOSE_SWITCH])

  1707                             for (i=0;i<qperiod_index;i++)

  1708                                 if (!strcmp(testword,qperiod[i]))

  1709                                     isdup=1;

  1710                         if (!isdup)

  1711 			{

  1712                             if (qperiod_index<MAX_QWORD &&

  1713 			      strlen(testword)<MAX_QWORD_LENGTH)

  1714 			    {

  1715                                 strcpy(qperiod[qperiod_index],testword);

  1716                                 qperiod_index++;

  1717 			    }

  1718                             if (pswit[ECHO_SWITCH])

  1719 				printf("\n%s\n",aline);

  1720                             if (!pswit[OVERVIEW_SWITCH])

  1721                                 printf("    Line %ld column %d - "

  1722 				  "Extra period?\n",linecnt,(int)(t-aline)+1);

  1723                             else

  1724                                 cnt_punct++;

  1725 			}

  1726 		    }

  1727 		}

  1728 	    t++;

  1729 	    }

  1730 	}

  1731         if (pswit[TYPO_SWITCH])

  1732 	{

  1733             /* Check for words usually not followed by punctuation. */

  1734             for (s=aline;*s;)

  1735 	    {

  1736                 wordstart=s;

  1737                 s=getaword(s,inword);

  1738                 if (!*inword)

  1739 		    continue;

  1740                 lowerit(inword);

  1741                 for (i=0;*nocomma[i];i++)

  1742                     if (!strcmp(inword,nocomma[i]))

  1743 		    {

  1744                         if (*s==',' || *s==';' || *s==':')

  1745 			{

  1746                             if (pswit[ECHO_SWITCH])

  1747 				printf("\n%s\n",aline);

  1748                             if (!pswit[OVERVIEW_SWITCH])

  1749                                 printf("    Line %ld column %d - "

  1750 				  "Query punctuation after %s?\n",

  1751 				  linecnt,(int)(s-aline)+1,inword);

  1752                             else

  1753                                 cnt_punct++;

  1754 			}

  1755 		    }

  1756 		for (i=0;*noperiod[i];i++)

  1757                     if (!strcmp(inword,noperiod[i]))

  1758 		    {

  1759                         if (*s=='.' || *s=='!')

  1760 			{

  1761                             if (pswit[ECHO_SWITCH])

  1762 				printf("\n%s\n",aline);

  1763                             if (!pswit[OVERVIEW_SWITCH])

  1764                                 printf("    Line %ld column %d - "

  1765 				  "Query punctuation after %s?\n",

  1766 				  linecnt,(int)(s-aline)+1,inword);

  1767                             else

  1768                                 cnt_punct++;

  1769 			}

  1770 		    }

  1771 	    }

  1772 	}

  1773         /*

  1774 	 * Check for commonly mistyped words,

  1775 	 * and digits like 0 for O in a word.

  1776 	 */

  1777         for (s=aline;*s;)

  1778 	{

  1779             wordstart=s;

  1780             s=getaword(s,inword);

  1781             if (!*inword)

  1782 		continue; /* don't bother with empty lines */

  1783             if (mixdigit(inword))

  1784 	    {

  1785                 if (pswit[ECHO_SWITCH])

  1786 		    printf("\n%s\n",aline);

  1787                 if (!pswit[OVERVIEW_SWITCH])

  1788                     printf("    Line %ld column %d - Query digit in %s\n",

  1789 		      linecnt,(int)(wordstart-aline)+1,inword);

  1790                 else

  1791                     cnt_word++;

  1792 	    }

  1793             /*

  1794 	     * Put the word through a series of tests for likely typos and OCR

  1795 	     * errors.

  1796 	     */

  1797             if (pswit[TYPO_SWITCH])

  1798 	    {

  1799                 istypo=0;

  1800                 strcpy(testword,inword);

  1801                 alower=0;

  1802                 for (i=0;i<(signed int)strlen(testword);i++)

  1803 		{

  1804 		    /* lowercase for testing */

  1805                     if (testword[i]>='a' && testword[i]<='z')

  1806 			alower=1;

  1807                     if (alower && testword[i]>='A' && testword[i]<='Z')

  1808 		    {

  1809                         /*

  1810 			 * We have an uppercase mid-word. However, there are

  1811 			 * common cases:

  1812                          *   Mac and Mc like McGill

  1813                          *   French contractions like l'Abbe

  1814 			 */

  1815                         if (i==2 && testword[0]=='m' && testword[1]=='c' ||

  1816                           i==3 && testword[0]=='m' && testword[1]=='a' &&

  1817 			  testword[2]=='c' || i>0 && testword[i-1]==CHAR_SQUOTE)

  1818 			    ; /* do nothing! */

  1819                         else

  1820                             istypo=1;

  1821 		    }

  1822                     testword[i]=(char)tolower(testword[i]);

  1823 		}

  1824                 /*

  1825 		 * Check for certain unlikely two-letter combinations at word

  1826 		 * start and end.

  1827 		 */

  1828                 if (strlen(testword)>1)

  1829 		{

  1830                     for (i=0;*nostart[i];i++)

  1831                         if (!strncmp(testword,nostart[i],2))

  1832                             istypo=1;

  1833                     for (i=0;*noend[i];i++)

  1834                         if (!strncmp(testword+strlen(testword)-2,noend[i],2))

  1835                             istypo=1;

  1836 		}

  1837                 /* ght is common, gbt never. Like that. */

  1838                 if (strstr(testword,"cb"))

  1839 		    istypo=1;

  1840                 if (strstr(testword,"gbt"))

  1841 		    istypo=1;

  1842                 if (strstr(testword,"pbt"))

  1843 		    istypo=1;

  1844                 if (strstr(testword,"tbs"))

  1845 		    istypo=1;

  1846                 if (strstr(testword,"mrn"))

  1847 		    istypo=1;

  1848                 if (strstr(testword,"ahle"))

  1849 		    istypo=1;

  1850                 if (strstr(testword,"ihle"))

  1851 		    istypo=1;

  1852                 /*

  1853 		 * "TBE" does happen - like HEARTBEAT - but uncommon.

  1854                  * Also "TBI" - frostbite, outbid - but uncommon.

  1855                  * Similarly "ii" like Hawaii, or Pompeii, and in Roman

  1856 		 * numerals, but "ii" is a common scanno.

  1857 		 */

  1858                 if (strstr(testword,"tbi"))

  1859 		    istypo=1;

  1860                 if (strstr(testword,"tbe"))

  1861 		    istypo=1;

  1862                 if (strstr(testword,"ii"))

  1863 		    istypo=1;

  1864                 /*

  1865 		 * Check for no vowels or no consonants.

  1866                  * If none, flag a typo.

  1867 		 */

  1868                 if (!istypo && strlen(testword)>1)

  1869 		{

  1870                     vowel=consonant=0;

  1871                     for (i=0;testword[i];i++)

  1872 		    {

  1873                         if (testword[i]=='y' || gcisdigit(testword[i]))

  1874 			{

  1875 			    /* Yah, this is loose. */

  1876                             vowel++;

  1877                             consonant++;

  1878 			}

  1879                         else if (strchr(vowels,testword[i]))

  1880 			    vowel++;

  1881 			else

  1882 			    consonant++;

  1883 		    }

  1884                     if (!vowel || !consonant)

  1885                         istypo=1;

  1886 		}

  1887                 /*

  1888 		 * Now exclude the word from being reported if it's in

  1889                  * the okword list.

  1890 		 */

  1891                 for (i=0;*okword[i];i++)

  1892                     if (!strcmp(testword,okword[i]))

  1893                         istypo=0;

  1894                 /*

  1895 		 * What looks like a typo may be a Roman numeral.

  1896 		 * Exclude these.

  1897 		 */

  1898                 if (istypo && isroman(testword))

  1899 		    istypo=0;

  1900                 /* Check the manual list of typos. */

  1901                 if (!istypo)

  1902                     for (i=0;*typo[i];i++)

  1903                         if (!strcmp(testword,typo[i]))

  1904                             istypo=1;

  1905                 /*

  1906 		 * Check lowercase s, l, i and m - special cases.

  1907                  *   "j" - often a semi-colon gone wrong.

  1908                  *   "d" for a missing apostrophe - he d

  1909                  *   "n" for "in"

  1910 		 */

  1911                 if (!istypo && strlen(testword)==1 && strchr("slmijdn",*inword))

  1912 		    istypo=1;

  1913                 if (istypo)

  1914 		{

  1915                     isdup=0;

  1916                     if (strlen(testword)<MAX_QWORD_LENGTH &&

  1917 		      !pswit[VERBOSE_SWITCH])

  1918                         for (i=0;i<qword_index;i++)

  1919                             if (!strcmp(testword,qword[i]))

  1920 			    {

  1921                                 isdup=1;

  1922                                 ++dupcnt[i];

  1923 			    }

  1924                     if (!isdup)

  1925 		    {

  1926                         if (qword_index<MAX_QWORD &&

  1927 			  strlen(testword)<MAX_QWORD_LENGTH)

  1928 			{

  1929                             strcpy(qword[qword_index],testword);

  1930                             qword_index++;

  1931 			}

  1932                         if (pswit[ECHO_SWITCH])

  1933 			    printf("\n%s\n",aline);

  1934                         if (!pswit[OVERVIEW_SWITCH])

  1935 			{

  1936                             printf("    Line %ld column %d - Query word %s",

  1937 			      linecnt,(int)(wordstart-aline)+1,inword);

  1938                             if (strlen(testword)<MAX_QWORD_LENGTH &&

  1939 			      !pswit[VERBOSE_SWITCH])

  1940                                 printf(" - not reporting duplicates");

  1941                             printf("\n");

  1942 			}

  1943                         else

  1944                             cnt_word++;

  1945 		    }

  1946 		}

  1947 	    }

  1948 	    /* check the user's list of typos */

  1949 	    if (!istypo && usertypo_count)

  1950 		for (i=0;i<usertypo_count;i++)

  1951 		    if (!strcmp(testword,usertypo[i]))

  1952 		    {

  1953 			if (pswit[ECHO_SWITCH])

  1954 			    printf("\n%s\n",aline);

  1955 			if (!pswit[OVERVIEW_SWITCH])

  1956 			    printf("    Line %ld column %d - "

  1957 			      "Query possible scanno %s\n",

  1958 			      linecnt,(int)(wordstart-aline)+2,inword);

  1959 		    }

  1960             if (pswit[PARANOID_SWITCH] && warnings->digit)

  1961 	    {

  1962 		/* In paranoid mode, query all 0 and 1 standing alone. */

  1963                 if (!strcmp(inword,"0") || !strcmp(inword,"1"))

  1964 		{

  1965                     if (pswit[ECHO_SWITCH])

  1966 			printf("\n%s\n",aline);

  1967                     if (!pswit[OVERVIEW_SWITCH])

  1968                         printf("    Line %ld column %d - Query standalone %s\n",

  1969 			  linecnt,(int)(wordstart-aline)+2,inword);

  1970                     else

  1971                         cnt_word++;

  1972 		}

  1973 	    }

  1974 	}

  1975 	/*

  1976          * Look for added or missing spaces around punctuation and quotes.

  1977          * If there is a punctuation character like ! with no space on

  1978          * either side, suspect a missing!space. If there are spaces on

  1979          * both sides , assume a typo. If we see a double quote with no

  1980          * space or punctuation on either side of it, assume unspaced

  1981          * quotes "like"this.

  1982 	 */

  1983         llen=strlen(aline);

  1984         for (i=1;i<llen;i++)

  1985 	{

  1986 	    /* For each character in the line after the first. */

  1987             if (strchr(".?!,;:_",aline[i]))  /* if it's punctuation */

  1988 	    {

  1989 		/* we need to suppress warnings for acronyms like M.D. */

  1990                 isacro=0;

  1991 		/* we need to suppress warnings for ellipsis . . . */

  1992                 isellipsis=0;

  1993 		/* if there are letters on both sides of it or ... */

  1994                 if (gcisalpha(aline[i-1]) && gcisalpha(aline[i+1]) ||

  1995                    gcisalpha(aline[i+1]) && strchr("?!,;:",aline[i]))

  1996 		{

  1997 		    /* ...if it's strict punctuation followed by an alpha */

  1998                     if (aline[i]=='.')

  1999 		    {

  2000                         if (i>2 && aline[i-2]=='.')

  2001 			    isacro=1;

  2002                         if (i+2<llen && aline[i+2]=='.')

  2003 			    isacro=1;

  2004 		    }

  2005                     if (!isacro)

  2006 		    {

  2007                         if (pswit[ECHO_SWITCH])

  2008 			    printf("\n%s\n",aline);

  2009                         if (!pswit[OVERVIEW_SWITCH])

  2010                             printf("    Line %ld column %d - Missing space?\n",

  2011 			      linecnt,i+1);

  2012                         else

  2013                             cnt_punct++;

  2014 		    }

  2015 		}

  2016                 if (aline[i-1]==CHAR_SPACE &&

  2017 		  (aline[i+1]==CHAR_SPACE || aline[i+1]==0))

  2018 		{

  2019 		    /*

  2020 		     * If there are spaces on both sides,

  2021 		     * or space before and end of line.

  2022 		     */

  2023                     if (aline[i]=='.')

  2024 		    {

  2025                         if (i>2 && aline[i-2]=='.')

  2026 			    isellipsis=1;

  2027                         if (i+2<llen && aline[i+2]=='.')

  2028 			    isellipsis=1;

  2029 		    }

  2030                     if (!isemptyline && !isellipsis)

  2031 		    {

  2032                         if (pswit[ECHO_SWITCH])

  2033 			    printf("\n%s\n",aline);

  2034                         if (!pswit[OVERVIEW_SWITCH])

  2035                             printf("    Line %ld column %d - "

  2036 			      "Spaced punctuation?\n",linecnt,i+1);

  2037                         else

  2038                             cnt_punct++;

  2039 		    }

  2040 		}

  2041 	    }

  2042 	}

  2043         /* Split out the characters that CANNOT be preceded by space. */

  2044         llen=strlen(aline);

  2045         for (i=1;i<llen;i++)

  2046 	{

  2047 	    /* for each character in the line after the first */

  2048             if (strchr("?!,;:",aline[i]))

  2049 	    {

  2050 		/* if it's punctuation that _cannot_ have a space before it */

  2051                 if (aline[i-1]==CHAR_SPACE && !isemptyline &&

  2052 		  aline[i+1]!=CHAR_SPACE)

  2053 		{

  2054 		    /*

  2055 		     * If aline[i+1) DOES == space,

  2056 		     * it was already reported just above.

  2057 		     */

  2058                     if (pswit[ECHO_SWITCH])

  2059 			printf("\n%s\n",aline);

  2060                     if (!pswit[OVERVIEW_SWITCH])

  2061                         printf("    Line %ld column %d - Spaced punctuation?\n",

  2062 			  linecnt,i+1);

  2063                     else

  2064                         cnt_punct++;

  2065 		}

  2066 	    }

  2067 	}

  2068         /*

  2069 	 * Special case " .X" where X is any alpha.

  2070          * This plugs a hole in the acronym code above.

  2071 	 * Inelegant, but maintainable.

  2072 	 */

  2073         llen=strlen(aline);

  2074         for (i=1;i<llen;i++)

  2075 	{

  2076 	    /* for each character in the line after the first */

  2077             if (aline[i]=='.')

  2078 	    {

  2079 		/* if it's a period */

  2080                 if (aline[i-1]==CHAR_SPACE && gcisalpha(aline[i+1]))

  2081 		{

  2082 		    /*

  2083 		     * If the period follows a space and

  2084 		     * is followed by a letter.

  2085 		     */

  2086                     if (pswit[ECHO_SWITCH])

  2087 			printf("\n%s\n",aline);

  2088                     if (!pswit[OVERVIEW_SWITCH])

  2089                         printf("    Line %ld column %d - Spaced punctuation?\n",

  2090 			  linecnt,i+1);

  2091                     else

  2092                         cnt_punct++;

  2093 		}

  2094 	    }

  2095 	}

  2096         for (i=1;i<llen;i++)

  2097 	{

  2098 	    /* for each character in the line after the first */

  2099             if (aline[i]==CHAR_DQUOTE)

  2100 	    {

  2101                 if (!strchr(" _-.'`,;:!/([{?}])",aline[i-1]) &&

  2102 		  !strchr(" _-.'`,;:!/([{?}])",aline[i+1]) && aline[i+1] ||

  2103 		  !strchr(" _-([{'`",aline[i-1]) && gcisalpha(aline[i+1]))

  2104 		{

  2105 		    if (pswit[ECHO_SWITCH])

  2106 			printf("\n%s\n",aline);

  2107 		    if (!pswit[OVERVIEW_SWITCH])

  2108 			printf("    Line %ld column %d - Unspaced quotes?\n",

  2109 			  linecnt,i+1);

  2110 		    else

  2111 			cnt_punct++;

  2112 		}

  2113 	    }

  2114 	}

  2115         /* Check parity of quotes. */

  2116         for (s=aline;*s;s++)

  2117 	{

  2118             if (*s==CHAR_DQUOTE)

  2119 	    {

  2120                 if (!(dquotepar=!dquotepar))

  2121 		{

  2122 		    /* parity even */

  2123                     if (!strchr("_-.'`/,;:!?)]} ",s[1]))

  2124 		    {

  2125                         if (pswit[ECHO_SWITCH])

  2126 			    printf("\n%s\n",aline);

  2127                         if (!pswit[OVERVIEW_SWITCH])

  2128                             printf("    Line %ld column %d - "

  2129 			      "Wrongspaced quotes?\n",linecnt,(int)(s-aline)+1);

  2130                         else

  2131                             cnt_punct++;

  2132 		    }

  2133 		}

  2134                 else

  2135 		{

  2136 		    /* parity odd */

  2137                     if (!gcisalpha(s[1]) && !isdigit(s[1]) &&

  2138 		      !strchr("_-/.'`([{$",s[1]) || !s[1])

  2139 		    {

  2140                         if (pswit[ECHO_SWITCH])

  2141 			    printf("\n%s\n",aline);

  2142                         if (!pswit[OVERVIEW_SWITCH])

  2143                             printf("    Line %ld column %d - "

  2144 			      "Wrongspaced quotes?\n",linecnt,(int)(s-aline)+1);

  2145                         else

  2146                             cnt_punct++;

  2147 		    }

  2148 		}

  2149 	    }

  2150 	}

  2151 	if (*aline==CHAR_DQUOTE)

  2152 	{

  2153 	    if (strchr(",;:!?)]} ",aline[1]))

  2154 	    {

  2155 		if (pswit[ECHO_SWITCH])

  2156 		    printf("\n%s\n",aline);

  2157 		if (!pswit[OVERVIEW_SWITCH])

  2158 		    printf("    Line %ld column 1 - Wrongspaced quotes?\n",

  2159 		      linecnt);

  2160 		else

  2161 		    cnt_punct++;

  2162 	    }

  2163 	}

  2164         if (pswit[SQUOTE_SWITCH])

  2165 	{

  2166             for (s=aline;*s;s++)

  2167 	    {

  2168                 if ((*s==CHAR_SQUOTE || *s==CHAR_OPEN_SQUOTE) &&

  2169 		  (s==aline || s>aline && !gcisalpha(s[-1]) ||

  2170 		  !gcisalpha(s[1])))

  2171 		{

  2172                     if (!(squotepar=!squotepar))

  2173 		    {

  2174 			/* parity even */

  2175                         if (!strchr("_-.'`/\",;:!?)]} ",s[1]))

  2176 			{

  2177                             if (pswit[ECHO_SWITCH])

  2178 				printf("\n%s\n",aline);

  2179                             if (!pswit[OVERVIEW_SWITCH])

  2180                                 printf("    Line %ld column %d - "

  2181 				  "Wrongspaced singlequotes?\n",

  2182 				  linecnt,(int)(s-aline)+1);

  2183                             else

  2184                                 cnt_punct++;

  2185 			}

  2186 		    }

  2187                     else

  2188 		    {

  2189 			/* parity odd */

  2190                         if (!gcisalpha(s[1]) && !isdigit(s[1]) &&

  2191 			  !strchr("_-/\".'`",s[1]) || !s[1])

  2192 			{

  2193                             if (pswit[ECHO_SWITCH])

  2194 				printf("\n%s\n",aline);

  2195                             if (!pswit[OVERVIEW_SWITCH])

  2196                                 printf("    Line %ld column %d - "

  2197 				  "Wrongspaced singlequotes?\n",

  2198 				  linecnt,(int)(s-aline)+1);

  2199                             else

  2200                                 cnt_punct++;

  2201 			}

  2202 		    }

  2203 		}

  2204 	    }

  2205 	}

  2206         /*

  2207 	 * Look for double punctuation like ,. or ,,

  2208          * Thanks to DW for the suggestion!

  2209          * In books with references, ".," and ".;" are common

  2210          * e.g. "etc., etc.," and vol. 1.; vol 3.;

  2211          * OTOH, from my initial tests, there are also fairly

  2212          * common errors. What to do? Make these cases paranoid?

  2213          * ".," is the most common, so warnings->dotcomma is used

  2214          * to suppress detailed reporting if it occurs often.

  2215 	 */

  2216         llen=strlen(aline);

  2217         for (i=0;i<llen;i++)

  2218 	{

  2219 	    /* for each punctuation character in the line */

  2220             if (strchr(".?!,;:",aline[i]) && (strchr(".?!,;:",aline[i+1])) &&

  2221 	      aline[i] && aline[i+1])

  2222 	    {

  2223 		/* followed by punctuation, it's a query, unless . . . */

  2224                 if (aline[i]==aline[i+1] && (aline[i]=='.' || aline[i]=='?' ||

  2225 		  aline[i]=='!') ||

  2226 		  !warnings->dotcomma && aline[i]=='.' && aline[i+1]==',' ||

  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 		  warnings->isFrench && !strncmp(aline+i,"...:",4) ||

  2233 		  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 		{

  2238 		    if (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 		      warnings->isFrench && !strncmp(aline+i,"...:",4) ||

  2244 		      warnings->isFrench && !strncmp(aline+i,"!...",4) ||

  2245 		      warnings->isFrench && !strncmp(aline+i,"...!",4) ||

  2246 		      warnings->isFrench && !strncmp(aline+i,"?...",4) ||

  2247 		      warnings->isFrench && !strncmp(aline+i,"...?",4))

  2248 			i+=4;

  2249 		    ; /* do nothing for .. !! and ?? which can be legit */

  2250 		}

  2251                 else

  2252 		{

  2253                     if (pswit[ECHO_SWITCH])

  2254 			printf("\n%s\n",aline);

  2255                     if (!pswit[OVERVIEW_SWITCH])

  2256                         printf("    Line %ld column %d - Double punctuation?\n",

  2257 			  linecnt,i+1);

  2258                     else

  2259                         cnt_punct++;

  2260 		}

  2261 	    }

  2262 	}

  2263         s=aline;

  2264         while (strstr(s," \" "))

  2265 	{

  2266             if (pswit[ECHO_SWITCH])

  2267 		printf("\n%s\n",aline);

  2268             if (!pswit[OVERVIEW_SWITCH])

  2269                 printf("    Line %ld column %d - Spaced doublequote?\n",

  2270 		  linecnt,(int)(strstr(s," \" ")-aline+1));

  2271             else

  2272                 cnt_punct++;

  2273             s=strstr(s," \" ")+2;

  2274 	}

  2275         s=aline;

  2276         while (strstr(s," ' "))

  2277 	{

  2278             if (pswit[ECHO_SWITCH])

  2279 		printf("\n%s\n",aline);

  2280             if (!pswit[OVERVIEW_SWITCH])

  2281                 printf("    Line %ld column %d - Spaced singlequote?\n",

  2282 		  linecnt,(int)(strstr(s," ' ")-aline+1));

  2283             else

  2284                 cnt_punct++;

  2285             s=strstr(s," ' ")+2;

  2286 	}

  2287         s=aline;

  2288         while (strstr(s," ` "))

  2289 	{

  2290             if (pswit[ECHO_SWITCH])

  2291 		printf("\n%s\n",aline);

  2292             if (!pswit[OVERVIEW_SWITCH])

  2293                 printf("    Line %ld column %d - Spaced singlequote?\n",

  2294 		  linecnt,(int)(strstr(s," ` ")-aline+1));

  2295             else

  2296                 cnt_punct++;

  2297             s=strstr(s," ` ")+2;

  2298 	}

  2299         /* check special case of 'S instead of 's at end of word */

  2300         s=aline+1;

  2301         while (*s)

  2302 	{

  2303             if (*s==CHAR_SQUOTE && s[1]=='S' && s[-1]>='a' && s[-1]<='z')

  2304 	    {

  2305                 if (pswit[ECHO_SWITCH])

  2306 		    printf("\n%s\n",aline);

  2307                 if (!pswit[OVERVIEW_SWITCH])

  2308                     printf("    Line %ld column %d - Capital \"S\"?\n",

  2309 		      linecnt,(int)(s-aline+2));

  2310                 else

  2311                     cnt_punct++;

  2312 	    }

  2313             s++;

  2314 	}

  2315         /*

  2316 	 * Now check special cases - start and end of line -

  2317          * for single and double quotes. Start is sometimes [sic]

  2318          * but better to query it anyway.

  2319          * While we're here, check for dash at end of line.

  2320 	 */

  2321         llen=strlen(aline);

  2322         if (llen>1)

  2323 	{

  2324             if (aline[llen-1]==CHAR_DQUOTE || aline[llen-1]==CHAR_SQUOTE ||

  2325 	      aline[llen-1]==CHAR_OPEN_SQUOTE)

  2326                 if (aline[llen-2]==CHAR_SPACE)

  2327 		{

  2328                     if (pswit[ECHO_SWITCH])

  2329 			printf("\n%s\n",aline);

  2330                     if (!pswit[OVERVIEW_SWITCH])

  2331                         printf("    Line %ld column %d - Spaced quote?\n",

  2332 			  linecnt,llen);

  2333                     else

  2334                         cnt_punct++;

  2335 		}

  2336             if ((aline[0]==CHAR_SQUOTE || aline[0]==CHAR_OPEN_SQUOTE) &&

  2337 	      aline[1]==CHAR_SPACE)

  2338 	    {

  2339 		if (pswit[ECHO_SWITCH])

  2340 		    printf("\n%s\n",aline);

  2341 		if (!pswit[OVERVIEW_SWITCH])

  2342 		    printf("    Line %ld column 1 - Spaced quote?\n",linecnt);

  2343 		else

  2344 		    cnt_punct++;

  2345 	    }

  2346             /*

  2347 	     * Dash at end of line may well be legit - paranoid mode only

  2348              * and don't report em-dash at line-end.

  2349 	     */

  2350             if (pswit[PARANOID_SWITCH] && warnings->hyphen)

  2351 	    {

  2352                 for (i=llen-1;i>0 && (unsigned char)aline[i]<=CHAR_SPACE;i--)

  2353 		    ;

  2354                 if (aline[i]=='-' && aline[i-1]!='-')

  2355 		{

  2356                     if (pswit[ECHO_SWITCH])

  2357 			printf("\n%s\n",aline);

  2358                     if (!pswit[OVERVIEW_SWITCH])

  2359                         printf("    Line %ld column %d - "

  2360 			  "Hyphen at end of line?\n",linecnt,i);

  2361 		}

  2362 	    }

  2363 	}

  2364         /*

  2365 	 * Brackets are often unspaced, but shouldn't be surrounded by alpha.

  2366          * If so, suspect a scanno like "a]most".

  2367 	 */

  2368         llen=strlen(aline);

  2369         for (i=1;i<llen-1;i++)

  2370 	{

  2371 	    /* for each bracket character in the line except 1st & last */

  2372             if (strchr("{[()]}",aline[i]) && gcisalpha(aline[i-1]) &&

  2373 	      gcisalpha(aline[i+1]))

  2374 	    {

  2375                 if (pswit[ECHO_SWITCH])

  2376 		    printf("\n%s\n",aline);

  2377                 if (!pswit[OVERVIEW_SWITCH])

  2378                     printf("    Line %ld column %d - Unspaced bracket?\n",

  2379 		      linecnt,i);

  2380                 else

  2381                     cnt_punct++;

  2382 	    }

  2383 	}

  2384         llen=strlen(aline);

  2385         if (warnings->endquote)

  2386 	{

  2387             for (i=1;i<llen;i++)

  2388 	    {

  2389 		/* for each character in the line except 1st */

  2390                 if (aline[i]==CHAR_DQUOTE && isalpha(aline[i-1]))

  2391 		{

  2392 		    if (pswit[ECHO_SWITCH])

  2393 			printf("\n%s\n",aline);

  2394 		    if (!pswit[OVERVIEW_SWITCH])

  2395 			printf("    Line %ld column %d - "

  2396 			  "endquote missing punctuation?\n",linecnt,i);

  2397 		    else

  2398 			cnt_punct++;

  2399 		}

  2400 	    }

  2401 	}

  2402 	/*

  2403          * Check for <HTML TAG>.

  2404          * If there is a < in the line, followed at some point

  2405          * by a > then we suspect HTML.

  2406 	 */

  2407         if (strstr(aline,"<") && strstr(aline,">"))

  2408 	{

  2409             i=(signed int)(strstr(aline,">")-strstr(aline,"<")+1);

  2410             if (i>0)

  2411 	    {

  2412                 strncpy(wrk,strstr(aline,"<"),i);

  2413                 wrk[i]=0;

  2414                 if (pswit[ECHO_SWITCH])

  2415 		    printf("\n%s\n",aline);

  2416                 if (!pswit[OVERVIEW_SWITCH])

  2417                     printf("    Line %ld column %d - HTML Tag? %s \n",

  2418 		      linecnt,(int)(strstr(aline,"<")-aline)+1,wrk);

  2419                 else

  2420                     cnt_html++;

  2421 	    }

  2422 	}

  2423         /*

  2424 	 * Check for &symbol; HTML.

  2425          * If there is a & in the line, followed at

  2426          * some point by a ; then we suspect HTML.

  2427 	 */

  2428         if (strstr(aline,"&") && strstr(aline,";"))

  2429 	{

  2430             i=(int)(strstr(aline,";")-strstr(aline,"&")+1);

  2431             for (s=strstr(aline,"&");s<strstr(aline,";");s++)

  2432                 if (*s==CHAR_SPACE)

  2433 		    i=0;                /* Don't report "Jones & Son;" */

  2434             if (i>0)

  2435 	    {

  2436                 strncpy(wrk,strstr(aline,"&"),i);

  2437                 wrk[i]=0;

  2438                 if (pswit[ECHO_SWITCH])

  2439 		    printf("\n%s\n",aline);

  2440                 if (!pswit[OVERVIEW_SWITCH])

  2441                     printf("    Line %ld column %d - HTML symbol? %s \n",

  2442 		      linecnt,(int)(strstr(aline,"&")-aline)+1,wrk);

  2443                 else

  2444                     cnt_html++;

  2445 	    }

  2446 	}

  2447         /*

  2448 	 * At end of paragraph, check for mismatched quotes.

  2449          * We don't want to report an error immediately, since it is a

  2450          * common convention to omit the quotes at end of paragraph if

  2451          * the next paragraph is a continuation of the same speaker.

  2452          * Where this is the case, the next para should begin with a

  2453          * quote, so we store the warning message and only display it

  2454          * at the top of the next iteration if the new para doesn't

  2455          * start with a quote.

  2456          * The -p switch overrides this default, and warns of unclosed

  2457          * quotes on _every_ paragraph, whether the next begins with a

  2458          * quote or not.

  2459 	 */

  2460         if (isemptyline)

  2461 	{

  2462 	    /* end of para - add up the totals */

  2463             if (counters.quot%2)

  2464                 sprintf(dquote_err,"    Line %ld - Mismatched quotes\n",

  2465 		  linecnt);

  2466             if (pswit[SQUOTE_SWITCH] && counters.open_single_quote &&

  2467 	      counters.open_single_quote!=counters.close_single_quote)

  2468                 sprintf(squote_err,"    Line %ld - Mismatched singlequotes?\n",

  2469 		  linecnt);

  2470             if (pswit[SQUOTE_SWITCH] && counters.open_single_quote &&

  2471 	      counters.open_single_quote!=counters.close_single_quote &&

  2472 	      counters.open_single_quote!=counters.close_single_quote+1)

  2473 		/*

  2474 		 * Flag it to be noted regardless of the

  2475 		 * first char of the next para.

  2476 		 */

  2477                 squot=1;

  2478             if (counters.r_brack)

  2479                 sprintf(rbrack_err,"    Line %ld - "

  2480 		  "Mismatched round brackets?\n",linecnt);

  2481             if (counters.s_brack)

  2482                 sprintf(sbrack_err,"    Line %ld - "

  2483 		  "Mismatched square brackets?\n",linecnt);

  2484             if (counters.c_brack)

  2485                 sprintf(cbrack_err,"    Line %ld - "

  2486 		  "Mismatched curly brackets?\n",linecnt);

  2487             if (counters.c_unders%2)

  2488                 sprintf(unders_err,"    Line %ld - Mismatched underscores?\n",

  2489 		  linecnt);

  2490 	    memset(&counters,0,sizeof(counters));

  2491 	    /* let the next iteration know that it's starting a new para */

  2492             isnewpara=1;

  2493 	}

  2494         /*

  2495 	 * Check for omitted punctuation at end of paragraph by working back

  2496 	 * through prevline. DW.

  2497          * Need to check this only for "normal" paras.

  2498          * So what is a "normal" para?

  2499          *    Not normal if one-liner (chapter headings, etc.)

  2500          *    Not normal if doesn't contain at least one locase letter

  2501          *    Not normal if starts with space

  2502 	 */

  2503         if (isemptyline)

  2504 	{

  2505 	    /* end of para */

  2506             for (s=prevline,i=0;*s && !i;s++)

  2507                 if (gcisletter(*s))

  2508 		    /* use i to indicate the presence of a letter on the line */

  2509                     i=1;

  2510             /*

  2511 	     * This next "if" is a problem.

  2512              * If we say "start_para_line <= linecnt - 1", that includes

  2513 	     * one-line "paragraphs" like chapter heads. Lotsa false positives.

  2514              * If we say "start_para_line < linecnt - 1" it doesn't, but then it

  2515              * misses genuine one-line paragraphs.

  2516 	     */

  2517             if (i && last.blen>2 && start_para_line<linecnt-1 &&

  2518 	      *prevline>CHAR_SPACE)

  2519 	    {

  2520                 for (i=strlen(prevline)-1;

  2521 		  (prevline[i]==CHAR_DQUOTE || prevline[i]==CHAR_SQUOTE) &&

  2522 		  prevline[i]>CHAR_SPACE && i>0;

  2523 		  i--)

  2524 		    ;

  2525                 for (;i>0;i--)

  2526 		{

  2527                     if (gcisalpha(prevline[i]))

  2528 		    {

  2529                         if (pswit[ECHO_SWITCH])

  2530 			    printf("\n%s\n",prevline);

  2531                         if (!pswit[OVERVIEW_SWITCH])

  2532                             printf("    Line %ld column %d - "

  2533 			      "No punctuation at para end?\n",

  2534 			      linecnt-1,strlen(prevline));

  2535                         else

  2536                             cnt_punct++;

  2537                         break;

  2538 		    }

  2539                     if (strchr("-.:!([{?}])",prevline[i]))

  2540                         break;

  2541 		}

  2542 	    }

  2543 	}

  2544         strcpy(prevline,aline);

  2545     }

  2546     fclose(infile);

  2547     if (!pswit[OVERVIEW_SWITCH])

  2548         for (i=0;i<MAX_QWORD;i++)

  2549             if (dupcnt[i])

  2550                 printf("\nNote: Queried word %s was duplicated %d time%s\n",

  2551 		  qword[i],dupcnt[i],"s");

  2552 }

  2554 /*

  2555  * flgets:

  2556  *

  2557  * Get one line from the input stream, checking for

  2558  * the existence of exactly one CR/LF line-end per line.

  2559  *

  2560  * Returns: a pointer to the line.

  2561  */

  2562 char *flgets(char *theline,int maxlen,FILE *thefile,long lcnt)

  2563 {

  2564     char c;

  2565     int len,isCR,cint;

  2566     *theline=0;

  2567     len=isCR=0;

  2568     c=cint=fgetc(thefile);

  2569     do

  2570     {

  2571         if (cint==EOF)

  2572             return NULL;

  2573 	/* either way, it's end of line */

  2574         if (c==10)

  2575 	{

  2576             if (isCR)

  2577                 break;

  2578             else

  2579 	    {

  2580 		/* Error - a LF without a preceding CR */

  2581                 if (pswit[LINE_END_SWITCH])

  2582 		{

  2583                     if (pswit[ECHO_SWITCH])

  2584 			printf("\n%s\n",theline);

  2585                     if (!pswit[OVERVIEW_SWITCH])

  2586                         printf("    Line %ld - No CR?\n",lcnt);

  2587                     else

  2588                         cnt_lineend++;

  2589 		}

  2590                 break;

  2591 	    }

  2592 	}

  2593         if (c==13)

  2594 	{

  2595             if (isCR)

  2596 	    {

  2597 		/* Error - two successive CRs */

  2598                 if (pswit[LINE_END_SWITCH])

  2599 		{

  2600                     if (pswit[ECHO_SWITCH])

  2601 			printf("\n%s\n",theline);

  2602                     if (!pswit[OVERVIEW_SWITCH])

  2603                         printf("    Line %ld - Two successive CRs?\n",lcnt);

  2604                     else

  2605                         cnt_lineend++;

  2606 		}

  2607 	    }

  2608             isCR=1;

  2609 	}

  2610         else

  2611 	{

  2612             if (pswit[LINE_END_SWITCH] && isCR)

  2613 	    {

  2614                 if (pswit[ECHO_SWITCH])

  2615 		    printf("\n%s\n",theline);

  2616                 if (!pswit[OVERVIEW_SWITCH])

  2617                     printf("    Line %ld column %d - CR without LF?\n",

  2618 		      lcnt,len+1);

  2619                 else

  2620                     cnt_lineend++;

  2621 	    }

  2622             theline[len]=c;

  2623             len++;

  2624             theline[len]=0;

  2625             isCR=0;

  2626 	}

  2627         c=cint=fgetc(thefile);

  2628     } while(len<maxlen);

  2629     if (pswit[MARKUP_SWITCH])

  2630         postprocess_for_HTML(theline);

  2631     if (pswit[DP_SWITCH])

  2632         postprocess_for_DP(theline);

  2633     return theline;

  2634 }

  2636 /*

  2637  * mixdigit:

  2638  *

  2639  * Takes a "word" as a parameter, and checks whether it

  2640  * contains a mixture of alpha and digits. Generally, this is an

  2641  * error, but may not be for cases like 4th or L5 12s. 3d.

  2642  *

  2643  * Returns: 0 if no error found, 1 if error.

  2644  */

  2645 int mixdigit(char *checkword)

  2646 {

  2647     int wehaveadigit,wehavealetter,firstdigits,query,wl;

  2648     char *s;

  2649     wehaveadigit=wehavealetter=query=0;

  2650     for (s=checkword;*s;s++)

  2651         if (gcisalpha(*s))

  2652             wehavealetter=1;

  2653         else

  2654             if (gcisdigit(*s))

  2655                 wehaveadigit=1;

  2656     if (wehaveadigit && wehavealetter)

  2657     {

  2658 	/* Now exclude common legit cases, like "21st" and "12l. 3s. 11d." */

  2659         query=1;

  2660         wl=strlen(checkword);

  2661         for (firstdigits=0;gcisdigit(checkword[firstdigits]);firstdigits++)

  2662             ;

  2663         /* digits, ending in st, rd, nd, th of either case */

  2664         if (firstdigits+2==wl && (matchword(checkword+wl-2,"st") ||

  2665 	  matchword(checkword+wl-2,"rd") || matchword(checkword+wl-2,"nd") ||

  2666 	  matchword(checkword+wl-2,"th")))

  2667 	    query=0;

  2668         if (firstdigits+3==wl && (matchword(checkword+wl-3,"sts") ||

  2669 	  matchword(checkword+wl-3,"rds") || matchword(checkword+wl-3,"nds") ||

  2670 	  matchword(checkword+wl-3,"ths")))

  2671 	    query=0;

  2672         if (firstdigits+3==wl && (matchword(checkword+wl-4,"stly") ||

  2673 	  matchword(checkword+wl-4,"rdly") ||

  2674 	  matchword(checkword+wl-4,"ndly") || matchword(checkword+wl-4,"thly")))

  2675 	    query=0;

  2676         /* digits, ending in l, L, s or d */

  2677         if (firstdigits+1==wl && (checkword[wl-1]=='l' ||

  2678 	  checkword[wl-1]=='L' || checkword[wl-1]=='s' || checkword[wl-1]=='d'))

  2679 	    query=0;

  2680         /*

  2681 	 * L at the start of a number, representing Britsh pounds, like L500.

  2682          * This is cute. We know the current word is mixeddigit. If the first

  2683          * letter is L, there must be at least one digit following. If both

  2684          * digits and letters follow, we have a genuine error, else we have a

  2685          * capital L followed by digits, and we accept that as a non-error.

  2686 	 */

  2687         if (checkword[0]=='L' && !mixdigit(checkword+1))

  2688 	    query=0;

  2689     }

  2690     return query;

  2691 }

  2693 /*

  2694  * getaword:

  2695  *

  2696  * Extracts the first/next "word" from the line, and puts

  2697  * it into "thisword". A word is defined as one English word unit--or

  2698  * at least that's the aim.

  2699  *

  2700  * Returns: a pointer to the position in the line where we will start

  2701  *          looking for the next word.

  2702  */

  2703 char *getaword(char *fromline,char *thisword)

  2704 {

  2705     int i,wordlen;

  2706     char *s;

  2707     wordlen=0;

  2708     for (;!gcisdigit(*fromline) && !gcisalpha(*fromline) && *fromline;

  2709       fromline++)

  2710 	;

  2711     /*

  2712      * Use a look-ahead to handle exceptions for numbers like 1,000 and 1.35.

  2713      * Especially yucky is the case of L1,000

  2714      * This section looks for a pattern of characters including a digit

  2715      * followed by a comma or period followed by one or more digits.

  2716      * If found, it returns this whole pattern as a word; otherwise we discard

  2717      * the results and resume our normal programming.

  2718      */

  2719     s=fromline;

  2720     for (;(gcisdigit(*s) || gcisalpha(*s) || *s==',' || *s=='.') &&

  2721       wordlen<MAXWORDLEN;s++)

  2722     {

  2723 	thisword[wordlen]=*s;

  2724         wordlen++;

  2725     }

  2726     thisword[wordlen]=0;

  2727     for (i=1;i<wordlen-1;i++)

  2728     {

  2729         if (thisword[i]=='.' || thisword[i]==',')

  2730 	{

  2731             if (gcisdigit(thisword[i-1]) && gcisdigit(thisword[i-1]))

  2732 	    {

  2733                 fromline=s;

  2734                 return fromline;

  2735 	    }

  2736 	}

  2737     }

  2738     /* we didn't find a punctuated number - do the regular getword thing */

  2739     wordlen=0;

  2740     for (;(gcisdigit(*fromline) || gcisalpha(*fromline) || *fromline=='\'') &&

  2741       wordlen<MAXWORDLEN;fromline++)

  2742     {

  2743         thisword[wordlen]=*fromline;

  2744         wordlen++;

  2745     }

  2746     thisword[wordlen]=0;

  2747     return fromline;

  2748 }

  2750 /*

  2751  * matchword:

  2752  *

  2753  * A case-insensitive string matcher.

  2754  */

  2755 int matchword(char *checkfor,char *thisword)

  2756 {

  2757     unsigned int ismatch,i;

  2758     if (strlen(checkfor)!=strlen(thisword))

  2759 	return 0;

  2760     ismatch=1;     /* assume a match until we find a difference */

  2761     for (i=0;i<strlen(checkfor);i++)

  2762         if (toupper(checkfor[i])!=toupper(thisword[i]))

  2763             ismatch=0;

  2764     return ismatch;

  2765 }

  2767 /*

  2768  * lowerit:

  2769  *

  2770  * Lowercase the line.

  2771  */

  2773 void lowerit(char *theline)

  2774 {

  2775     for (;*theline;theline++)

  2776         if (*theline>='A' && *theline<='Z')

  2777             *theline+=32;

  2778 }

  2780 /*

  2781  * isroman:

  2782  *

  2783  * Is this word a Roman Numeral?

  2784  *

  2785  * It doesn't actually validate that the number is a valid Roman Numeral--for

  2786  * example it will pass MXXXXXXXXXX as a valid Roman Numeral, but that's not

  2787  * what we're here to do. If it passes this, it LOOKS like a Roman numeral.

  2788  * Anyway, the actual Romans were pretty tolerant of bad arithmetic, or

  2789  * expressions thereof, except when it came to taxes. Allow any number of M,

  2790  * an optional D, an optional CM or CD, any number of optional Cs, an optional

  2791  * XL or an optional XC, an optional IX or IV, an optional V and any number

  2792  * of optional Is.

  2793  */

  2794 int isroman(char *t)

  2795 {

  2796     char *s;

  2797     if (!t || !*t)

  2798 	return 0;

  2799     s=t;

  2800     while (*t=='m' && *t)

  2801 	t++;

  2802     if (*t=='d')

  2803 	t++;

  2804     if (*t=='c' && t[1]=='m')

  2805 	t+=2;

  2806     if (*t=='c' && t[1]=='d')

  2807 	t+=2;

  2808     while (*t=='c' && *t)

  2809 	t++;

  2810     if (*t=='x' && t[1]=='l')

  2811 	t+=2;

  2812     if (*t=='x' && t[1]=='c')

  2813 	t+=2;

  2814     if (*t=='l')

  2815 	t++;

  2816     while (*t=='x' && *t)

  2817 	t++;

  2818     if (*t=='i' && t[1]=='x')

  2819 	t+=2;

  2820     if (*t=='i' && t[1]=='v')

  2821 	t+=2;

  2822     if (*t=='v')

  2823 	t++;

  2824     while (*t=='i' && *t)

  2825 	t++;

  2826     return !*t;

  2827 }

  2829 /*

  2830  * gcisalpha:

  2831  *

  2832  * A version of isalpha() that is somewhat lenient on 8-bit texts.

  2833  * If we use the standard function, 8-bit accented characters break

  2834  * words, so that tete with accented characters appears to be two words, "t"

  2835  * and "t", with 8-bit characters between them. This causes over-reporting of

  2836  * errors. gcisalpha() recognizes accented letters from the CP1252 (Windows)

  2837  * and ISO-8859-1 character sets, which are the most common PG 8-bit types.

  2838  */

  2839 int gcisalpha(unsigned char c)

  2840 {

  2841     if (c>='a' && c<='z')

  2842 	return 1;

  2843     if (c>='A' && c<='Z')

  2844 	return 1;

  2845     if (c<140)

  2846 	return 0;

  2847     if (c>=192 && c!=208 && c!=215 && c!=222 && c!=240 && c!=247 && c!=254)

  2848 	return 1;

  2849     if (c==140 || c==142 || c==156 || c==158 || c==159)

  2850 	return 1;

  2851     return 0;

  2852 }

  2854 /*

  2855  * gcisdigit:

  2856  *

  2857  * A version of isdigit() that doesn't get confused in 8-bit texts.

  2858  */

  2859 int gcisdigit(unsigned char c)

  2860 {

  2861     return c>='0' && c<='9';

  2862 }

  2864 /*

  2865  * gcisletter:

  2866  *

  2867  * A version of isletter() that doesn't get confused in 8-bit texts.

  2868  * NB: this is ISO-8891-1-specific.

  2869  */

  2870 int gcisletter(unsigned char c)

  2871 {

  2872     return c>='A' && c<='Z' || c>='a' && c<='z' || c>=192;

  2873 }

  2875 /*

  2876  * gcstrchr:

  2877  *

  2878  * Wraps strchr to return NULL if the character being searched for is zero.

  2879  */

  2880 char *gcstrchr(char *s,char c)

  2881 {

  2882     if (!c)

  2883 	return NULL;

  2884     return strchr(s,c);

  2885 }

  2887 /*

  2888  * postprocess_for_DP:

  2889  *

  2890  * Invoked with the -d switch from flgets().

  2891  * It simply "removes" from the line a hard-coded set of common

  2892  * DP-specific tags, so that the line passed to the main routine has

  2893  * been pre-cleaned of DP markup.

  2894  */

  2895 void postprocess_for_DP(char *theline)

  2896 {

  2897     char *s,*t;

  2898     int i;

  2899     if (!*theline)

  2900         return;

  2901     for (i=0;*DPmarkup[i];i++)

  2902     {

  2903         s=strstr(theline,DPmarkup[i]);

  2904         while (s)

  2905 	{

  2906             t=s+strlen(DPmarkup[i]);

  2907             while (*t)

  2908 	    {

  2909                 *s=*t;

  2910                 t++;

  2911 		s++;

  2912 	    }

  2913             *s=0;

  2914             s=strstr(theline,DPmarkup[i]);

  2915 	}

  2916     }

  2917 }

  2919 /*

  2920  * postprocess_for_HTML:

  2921  *

  2922  * Invoked with the -m switch from flgets().

  2923  * It simply "removes" from the line a hard-coded set of common

  2924  * HTML tags and "replaces" a hard-coded set of common HTML

  2925  * entities, so that the line passed to the main routine has

  2926  * been pre-cleaned of HTML.

  2927  */

  2928 void postprocess_for_HTML(char *theline)

  2929 {

  2930     if (strstr(theline,"<") && strstr(theline,">"))

  2931         while (losemarkup(theline))

  2932             ;

  2933     while (loseentities(theline))

  2934         ;

  2935 }

  2937 char *losemarkup(char *theline)

  2938 {

  2939     char *s,*t;

  2940     int i;

  2941     if (!*theline)

  2942         return NULL;

  2943     s=strstr(theline,"<");

  2944     t=strstr(theline,">");

  2945     if (!s || !t)

  2946 	return NULL;

  2947     for (i=0;*markup[i];i++)

  2948         if (!tagcomp(s+1,markup[i]))

  2949 	{

  2950             if (!t[1])

  2951 	    {

  2952                 *s=0;

  2953                 return s;

  2954 	    }

  2955             else if (t>s)

  2956 	    {

  2957 		strcpy(s,t+1);

  2958 		return s;

  2959 	    }

  2960         }

  2961     /* It's an unrecognized <xxx>. */

  2962     return NULL;

  2963 }

  2965 char *loseentities(char *theline)

  2966 {

  2967     int i;

  2968     char *s,*t;

  2969     if (!*theline)

  2970         return NULL;

  2971     for (i=0;*entities[i].htmlent;i++)

  2972     {

  2973         s=strstr(theline,entities[i].htmlent);

  2974         if (s)

  2975 	{

  2976             t=malloc((size_t)strlen(s));

  2977             if (!t)

  2978 		return NULL;

  2979             strcpy(t,s+strlen(entities[i].htmlent));

  2980             strcpy(s,entities[i].textent);

  2981             strcat(s,t);

  2982             free(t);

  2983             return theline;

  2984 	}

  2985     }

  2986     for (i=0;*entities[i].htmlnum;i++)

  2987     {

  2988         s=strstr(theline,entities[i].htmlnum);

  2989         if (s)

  2990 	{

  2991             t=malloc((size_t)strlen(s));

  2992             if (!t)

  2993 		return NULL;

  2994             strcpy(t,s+strlen(entities[i].htmlnum));

  2995             strcpy(s,entities[i].textent);

  2996             strcat(s,t);

  2997             free(t);

  2998             return theline;

  2999 	}

  3000     }

  3001     return NULL;

  3002 }

  3004 int tagcomp(char *strin,char *basetag)

  3005 {

  3006     char *s,*t;

  3007     s=basetag;

  3008     t=strin;

  3009     if (*t=='/')

  3010 	t++; /* ignore a slash */

  3011     while (*s && *t)

  3012     {

  3013         if (tolower(*s)!=tolower(*t))

  3014 	    return 1;

  3015         s++;

  3016 	t++;

  3017     }

  3018     return 0;

  3019 }

  3021 void proghelp()

  3022 {

  3023     fputs("Bookloupe version " PACKAGE_VERSION ".\n",stderr);

  3024     fputs("Copyright 2000-2005 Jim Tinsley <jtinsley@pobox.com>.\n",stderr);

  3025     fputs("Copyright 2012- J. Ali Harlow <ali@juiblex.co.uk>.\n",stderr);

  3026     fputs("Bookloupe comes wih ABSOLUTELY NO WARRANTY. "

  3027       "For details, read the file COPYING.\n",stderr);

  3028     fputs("This is Free Software; "

  3029       "you may redistribute it under certain conditions (GPL);\n",stderr);

  3030     fputs("read the file COPYING for details.\n\n",stderr);

  3031     fputs("Usage is: bookloupe [-setpxloyhud] filename\n",stderr);

  3032     fputs("  where -s checks single quotes, -e suppresses echoing lines, "

  3033       "-t checks typos\n",stderr);

  3034     fputs("  -x (paranoid) switches OFF -t and extra checks, "

  3035       "-l turns OFF line-end checks\n",stderr);

  3036     fputs("  -o just displays overview without detail, "

  3037       "-h echoes header fields\n",stderr);

  3038     fputs("  -v (verbose) unsuppresses duplicate reporting, "

  3039       "-m suppresses markup\n",stderr);

  3040     fputs("  -d ignores DP-specific markup,\n",stderr);

  3041     fputs("  -u uses a file gutcheck.typ to query user-defined "

  3042       "possible typos\n",stderr);

  3043     fputs("Sample usage: bookloupe warpeace.txt \n",stderr);

  3044     fputs("\n",stderr);

  3045     fputs("Bookloupe looks for errors in Project Gutenberg(TM) etexts.\n",

  3046       stderr);

  3047     fputs("Bookloupe queries anything it thinks shouldn't be in a PG text; "

  3048       "non-ASCII\n",stderr);

  3049     fputs("characters like accented letters, "

  3050       "lines longer than 75 or shorter than 55,\n",stderr);

  3051     fputs("unbalanced quotes or brackets, "

  3052       "a variety of badly formatted punctuation, \n",stderr);

  3053     fputs("HTML tags, some likely typos. "

  3054       "It is NOT a substitute for human judgement.\n",stderr);

  3055     fputs("\n",stderr);

  3056 }

author	ali <ali@juiblex.co.uk>
	Sat May 25 20:38:17 2013 +0100 (2013-05-25)
changeset 48	274aa1e09099
parent 47	7522c36859d0
child 49	8a53979c0d65
permissions	-rw-r--r--