bookloupe-testing: bookloupe/bookloupe.c@8ade5460e220

     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 const char *getaword(const 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 signed int dupcnt[MAX_QWORD];

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

   385 {

   386     char *argsw,*s;

   387     int i,switno,invarg;

   388     char usertypo_file[MAX_PATH];

   389     FILE *usertypofile;

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

   391 	/* save the path to the executable */

   392         strcpy(running_from,argv[0]);

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

   394     s=running_from+strlen(running_from);

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

   396         *s=0;

   397     switno=strlen(SWITCHES);

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

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

   400     /*

   401      * Standard loop to extract switches.

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

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

   404      * represented by their equivalent elements in pswit[]

   405      */

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

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

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

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

   410 		{

   411                     invarg=0;

   412                     pswit[i]=1;

   413 		}

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

   415     pswit[PARANOID_SWITCH]^=1;

   416     if (pswit[PARANOID_SWITCH])

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

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

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

   420     pswit[LINE_END_SWITCH]^=1;

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

   422     pswit[ECHO_SWITCH]^=1;

   423     if (pswit[OVERVIEW_SWITCH])

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

   425         pswit[ECHO_SWITCH]=0;

   426     /*

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

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

   429      */

   430     if (pswit[WEB_SWITCH])

   431     {

   432 	/* specific override for web uploads */

   433         pswit[ECHO_SWITCH]=1;

   434         pswit[SQUOTE_SWITCH]=0;

   435         pswit[TYPO_SWITCH]=1;

   436         pswit[QPARA_SWITCH]=0;

   437         pswit[PARANOID_SWITCH]=1;

   438         pswit[LINE_END_SWITCH]=0;

   439         pswit[OVERVIEW_SWITCH]=0;

   440         pswit[STDOUT_SWITCH]=0;

   441         pswit[HEADER_SWITCH]=1;

   442         pswit[VERBOSE_SWITCH]=0;

   443         pswit[MARKUP_SWITCH]=0;

   444         pswit[USERTYPO_SWITCH]=0;

   445         pswit[DP_SWITCH]=0;

   446     }

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

   448     {

   449 	/* check number of args */

   450         proghelp();

   451         return 1;

   452     }

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

   454     if (pswit[USERTYPO_SWITCH])

   455     {

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

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

   458         if (!usertypofile)

   459 	{

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

   461             strcpy(usertypo_file,running_from);

   462             strcat(usertypo_file,USERTYPO_FILE);

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

   464             if (!usertypofile) {

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

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

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

   468 	    }

   469 	}

   470         usertypo_count=0;

   471         if (usertypofile)

   472 	{

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

   474             if (pswit[USERTYPO_SWITCH])

   475 	    {

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

   477 		  (long)usertypo_count))

   478 		{

   479                     if (strlen(aline)>1)

   480 		    {

   481                         if ((int)*aline>33)

   482 			{

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

   484                             if (!s)

   485 			    {

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

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

   488                                 exit(1);

   489 			    }

   490                             strcpy(s,aline);

   491                             usertypo[usertypo_count]=s;

   492                             usertypo_count++;

   493                             if (usertypo_count>=MAX_USER_TYPOS)

   494 			    {

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

   496 				  "allowed: ignoring the rest\n",

   497 				  MAX_USER_TYPOS);

   498                                 break;

   499 			    }

   500 			}

   501 		    }

   502 		}

   503 	    }

   504             fclose(usertypofile);

   505 	}

   506     }

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

   508     cnt_dquot=cnt_squot=cnt_brack=cnt_bin=cnt_odd=cnt_long=

   509     cnt_short=cnt_punct=cnt_dash=cnt_word=cnt_html=cnt_lineend=

   510     cnt_spacend=0;

   511     procfile(argv[0]);

   512     if (pswit[OVERVIEW_SWITCH])

   513     {

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

   515 	  checked_linecnt,linecnt,linecnt-checked_linecnt);

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

   517         if (cnt_long)

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

   519         if (cnt_short)

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

   521         if (cnt_lineend)

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

   523         if (cnt_word)

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

   525         if (cnt_dquot)

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

   527         if (cnt_squot)

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

   529         if (cnt_brack)

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

   531         if (cnt_bin)

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

   533         if (cnt_odd)

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

   535         if (cnt_punct)

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

   537         if (cnt_dash)

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

   539         if (cnt_html)

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

   541         printf("\n");

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

   543           cnt_dquot+cnt_squot+cnt_brack+cnt_bin+cnt_odd+cnt_long+

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

   545     }

   546     return 0;

   547 }

   549 struct first_pass_results {

   550     long firstline,astline;

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

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

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

   554     signed int Dutchcount,Frenchcount;

   555 };

   557 /*

   558  * first_pass:

   559  *

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

   561  * file, decide whether to report some things that

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

   563  * lines, non-standard dashes, etc.

   564  */

   565 struct first_pass_results *first_pass(FILE *infile)

   566 {

   567     char laststart=CHAR_SPACE;

   568     const char *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  * check_for_jeebies:

  1267  *

  1268  * Check for "to he" and other easy h/b errors.

  1269  *

  1270  * This is a very inadequate effort on the h/b problem,

  1271  * but the phrase "to he" is always an error, whereas "to

  1272  * be" is quite common.

  1273  * Similarly, '"Quiet!", be said.' is a non-be error

  1274  * "to he" is _not_ always an error!:

  1275  *       "Where they went to he couldn't say."

  1276  * Another false positive:

  1277  *       What would "Cinderella" be without the . . .

  1278  * and another: "If he wants to he can see for himself."

  1279  */

  1280 void check_for_jeebies(const char *aline)

  1281 {

  1282     const char *s;

  1283     s=strstr(aline," be could ");

  1284     if (!s)

  1285 	s=strstr(aline," be would ");

  1286     if (!s)

  1287 	s=strstr(aline," was be ");

  1288     if (!s)

  1289 	s=strstr(aline," be is ");

  1290     if (!s)

  1291 	s=strstr(aline," is be ");

  1292     if (!s)

  1293 	s=strstr(aline,"\", be ");

  1294     if (!s)

  1295 	s=strstr(aline,"\" be ");

  1296     if (!s)

  1297 	s=strstr(aline,"\" be ");

  1298     if (!s)

  1299 	s=strstr(aline," to he ");

  1300     if (s)

  1301     {

  1302 	if (pswit[ECHO_SWITCH])

  1303 	    printf("\n%s\n",aline);

  1304 	if (!pswit[OVERVIEW_SWITCH])

  1305 	    printf("    Line %ld column %d - Query he/be error?\n",

  1306 	      linecnt,(int)(s-aline)+1);

  1307 	else

  1308 	    cnt_word++;

  1309     }

  1310     s=strstr(aline," the had ");

  1311     if (!s)

  1312 	s=strstr(aline," a had ");

  1313     if (!s)

  1314 	s=strstr(aline," they bad ");

  1315     if (!s)

  1316 	s=strstr(aline," she bad ");

  1317     if (!s)

  1318 	s=strstr(aline," he bad ");

  1319     if (!s)

  1320 	s=strstr(aline," you bad ");

  1321     if (!s)

  1322 	s=strstr(aline," i bad ");

  1323     if (s)

  1324     {

  1325 	if (pswit[ECHO_SWITCH])

  1326 	    printf("\n%s\n",aline);

  1327 	if (!pswit[OVERVIEW_SWITCH])

  1328 	    printf("    Line %ld column %d - Query had/bad error?\n",

  1329 	      linecnt,(int)(s-aline)+1);

  1330 	else

  1331 	    cnt_word++;

  1332     }

  1333     s=strstr(aline,"; hut ");

  1334     if (!s)

  1335 	s=strstr(aline,", hut ");

  1336     if (s)

  1337     {

  1338 	if (pswit[ECHO_SWITCH])

  1339 	    printf("\n%s\n",aline);

  1340 	if (!pswit[OVERVIEW_SWITCH])

  1341 	    printf("    Line %ld column %d - Query hut/but error?\n",

  1342 	      linecnt,(int)(s-aline)+1);

  1343 	else

  1344 	    cnt_word++;

  1345     }

  1346 }

  1348 /*

  1349  * check_for_mta_from:

  1350  *

  1351  * Special case - angled bracket in front of "From" placed there by an

  1352  * MTA when sending an e-mail.

  1353  */

  1354 void check_for_mta_from(const char *aline)

  1355 {

  1356     const char *s;

  1357     s=strstr(aline,">From");

  1358     if (s)

  1359     {

  1360 	if (pswit[ECHO_SWITCH])

  1361 	    printf("\n%s\n",aline);

  1362 	if (!pswit[OVERVIEW_SWITCH])

  1363 	    printf("    Line %ld column %d - Query angled bracket with From\n",

  1364 	      linecnt,(int)(s-aline)+1);

  1365 	else

  1366 	    cnt_punct++;

  1367     }

  1368 }

  1370 /*

  1371  * check_for_orphan_character:

  1372  *

  1373  * Check for a single character line -

  1374  * often an overflow from bad wrapping.

  1375  */

  1376 void check_for_orphan_character(const char *aline)

  1377 {

  1378     if (*aline && !aline[1])

  1379     {

  1380 	if (*aline=='I' || *aline=='V' || *aline=='X' || *aline=='L' ||

  1381 	  gcisdigit(*aline))

  1382 	    ; /* Nothing - ignore numerals alone on a line. */

  1383 	else

  1384 	{

  1385 	    if (pswit[ECHO_SWITCH])

  1386 		printf("\n%s\n",aline);

  1387 	    if (!pswit[OVERVIEW_SWITCH])

  1388 		printf("    Line %ld column 1 - Query single character line\n",

  1389 		  linecnt);

  1390 	    else

  1391 		cnt_punct++;

  1392 	}

  1393     }

  1394 }

  1396 /*

  1397  * check_for_pling_scanno:

  1398  *

  1399  * Check for I" - often should be !

  1400  */

  1401 void check_for_pling_scanno(const char *aline)

  1402 {

  1403     const char *s;

  1404     s=strstr(aline," I\"");

  1405     if (s)

  1406     {

  1407 	if (pswit[ECHO_SWITCH])

  1408 	    printf("\n%s\n",aline);

  1409 	if (!pswit[OVERVIEW_SWITCH])

  1410 	    printf("    Line %ld column %ld - Query I=exclamation mark?\n",

  1411 	      linecnt,s-aline);

  1412 	else

  1413 	    cnt_punct++;

  1414     }

  1415 }

  1417 /*

  1418  * check_for_extra_period:

  1419  *

  1420  * Check for period without a capital letter. Cut-down from gutspell.

  1421  * Only works when it happens on a single line.

  1422  */

  1423 void check_for_extra_period(const char *aline,const struct warnings *warnings)

  1424 {

  1425     const char *s,*t,*s1;

  1426     signed int i,istypo,isdup;

  1427     static char qperiod[MAX_QWORD][MAX_QWORD_LENGTH];

  1428     static int qperiod_index=0;

  1429     char testword[MAXWORDLEN]="";

  1430     if (pswit[PARANOID_SWITCH])

  1431     {

  1432 	for (t=s=aline;strstr(t,". ");)

  1433 	{

  1434 	    t=strstr(t,". ");

  1435 	    if (t==s)

  1436 	    {

  1437 		t++;

  1438 		/* start of line punctuation is handled elsewhere */

  1439 		continue;

  1440 	    }

  1441 	    if (!gcisalpha(t[-1]))

  1442 	    {

  1443 		t++;

  1444 		continue;

  1445 	    }

  1446 	    if (warnings->isDutch)

  1447 	    {

  1448 		/* For Frank & Jeroen -- 's Middags case */

  1449 		if (t[2]==CHAR_SQUOTE && t[3]>='a' && t[3]<='z' &&

  1450 		  t[4]==CHAR_SPACE && t[5]>='A' && t[5]<='Z')

  1451 		{

  1452 		    t++;

  1453 		    continue;

  1454 		}

  1455 	    }

  1456 	    s1=t+2;

  1457 	    while (*s1 && !gcisalpha(*s1) && !isdigit(*s1))

  1458 		s1++;

  1459 	    if (*s1>='a' && *s1<='z')

  1460 	    {

  1461 		/* we have something to investigate */

  1462 		istypo=1;

  1463 		/* so let's go back and find out */

  1464 		for (s1=t-1;s1>=s &&

  1465 		  (gcisalpha(*s1) || gcisdigit(*s1) || *s1==CHAR_SQUOTE &&

  1466 		  gcisalpha(s1[1]) && gcisalpha(s1[-1]));s1--)

  1467 		    ;

  1468 		s1++;

  1469 		for (i=0;*s1 && *s1!='.';s1++,i++)

  1470 		    testword[i]=*s1;

  1471 		testword[i]=0;

  1472 		for (i=0;*abbrev[i];i++)

  1473 		    if (!strcmp(testword,abbrev[i]))

  1474 			istypo=0;

  1475 		if (gcisdigit(*testword))

  1476 		    istypo=0;

  1477 		if (!testword[1])

  1478 		    istypo=0;

  1479 		if (isroman(testword))

  1480 		    istypo=0;

  1481 		if (istypo)

  1482 		{

  1483 		    istypo=0;

  1484 		    for (i=0;testword[i];i++)

  1485 			if (strchr(vowels,testword[i]))

  1486 			    istypo=1;

  1487 		}

  1488 		if (istypo)

  1489 		{

  1490 		    isdup=0;

  1491 		    if (strlen(testword)<MAX_QWORD_LENGTH &&

  1492 		      !pswit[VERBOSE_SWITCH])

  1493 			for (i=0;i<qperiod_index;i++)

  1494 			    if (!strcmp(testword,qperiod[i]))

  1495 				isdup=1;

  1496 		    if (!isdup)

  1497 		    {

  1498 			if (qperiod_index<MAX_QWORD &&

  1499 			  strlen(testword)<MAX_QWORD_LENGTH)

  1500 			{

  1501 			    strcpy(qperiod[qperiod_index],testword);

  1502 			    qperiod_index++;

  1503 			}

  1504 			if (pswit[ECHO_SWITCH])

  1505 			    printf("\n%s\n",aline);

  1506 			if (!pswit[OVERVIEW_SWITCH])

  1507 			    printf("    Line %ld column %d - Extra period?\n",

  1508 			      linecnt,(int)(t-aline)+1);

  1509 			else

  1510 			    cnt_punct++;

  1511 		    }

  1512 		}

  1513 	    }

  1514 	    t++;

  1515 	}

  1516     }

  1517 }

  1519 /*

  1520  * check_for_following_punctuation:

  1521  *

  1522  * Check for words usually not followed by punctuation.

  1523  */

  1524 void check_for_following_punctuation(const char *aline)

  1525 {

  1526     int i;

  1527     const char *s,*wordstart;

  1528     char inword[MAXWORDLEN];

  1529     if (pswit[TYPO_SWITCH])

  1530     {

  1531 	for (s=aline;*s;)

  1532 	{

  1533 	    wordstart=s;

  1534 	    s=getaword(s,inword);

  1535 	    if (!*inword)

  1536 		continue;

  1537 	    lowerit(inword);

  1538 	    for (i=0;*nocomma[i];i++)

  1539 		if (!strcmp(inword,nocomma[i]))

  1540 		{

  1541 		    if (*s==',' || *s==';' || *s==':')

  1542 		    {

  1543 			if (pswit[ECHO_SWITCH])

  1544 			    printf("\n%s\n",aline);

  1545 			if (!pswit[OVERVIEW_SWITCH])

  1546 			    printf("    Line %ld column %d - "

  1547 			      "Query punctuation after %s?\n",

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

  1549 			else

  1550 			    cnt_punct++;

  1551 		    }

  1552 		}

  1553 	    for (i=0;*noperiod[i];i++)

  1554 		if (!strcmp(inword,noperiod[i]))

  1555 		{

  1556 		    if (*s=='.' || *s=='!')

  1557 		    {

  1558 			if (pswit[ECHO_SWITCH])

  1559 			    printf("\n%s\n",aline);

  1560 			if (!pswit[OVERVIEW_SWITCH])

  1561 			    printf("    Line %ld column %d - "

  1562 			      "Query punctuation after %s?\n",

  1563 			      linecnt,(int)(s-aline)+1,inword);

  1564 			else

  1565 			    cnt_punct++;

  1566 		    }

  1567 		}

  1568 	}

  1569     }

  1570 }

  1572 /*

  1573  * check_for_typos:

  1574  *

  1575  * Check for commonly mistyped words,

  1576  * and digits like 0 for O in a word.

  1577  */

  1578 void check_for_typos(const char *aline,struct warnings *warnings)

  1579 {

  1580     const char *s,*wordstart;

  1581     char inword[MAXWORDLEN],testword[MAXWORDLEN];

  1582     int i,istypo,isdup,alower,vowel,consonant;

  1583     static int qword_index=0;

  1584     for (s=aline;*s;)

  1585     {

  1586 	wordstart=s;

  1587 	s=getaword(s,inword);

  1588 	if (!*inword)

  1589 	    continue; /* don't bother with empty lines */

  1590 	if (mixdigit(inword))

  1591 	{

  1592 	    if (pswit[ECHO_SWITCH])

  1593 		printf("\n%s\n",aline);

  1594 	    if (!pswit[OVERVIEW_SWITCH])

  1595 		printf("    Line %ld column %d - Query digit in %s\n",

  1596 		  linecnt,(int)(wordstart-aline)+1,inword);

  1597 	    else

  1598 		cnt_word++;

  1599 	}

  1600 	/*

  1601 	 * Put the word through a series of tests for likely typos and OCR

  1602 	 * errors.

  1603 	 */

  1604 	if (pswit[TYPO_SWITCH])

  1605 	{

  1606 	    istypo=0;

  1607 	    strcpy(testword,inword);

  1608 	    alower=0;

  1609 	    for (i=0;i<(signed int)strlen(testword);i++)

  1610 	    {

  1611 		/* lowercase for testing */

  1612 		if (testword[i]>='a' && testword[i]<='z')

  1613 		    alower=1;

  1614 		if (alower && testword[i]>='A' && testword[i]<='Z')

  1615 		{

  1616 		    /*

  1617 		     * We have an uppercase mid-word. However, there are

  1618 		     * common cases:

  1619 		     *   Mac and Mc like McGill

  1620 		     *   French contractions like l'Abbe

  1621 		     */

  1622 		    if (i==2 && testword[0]=='m' && testword[1]=='c' ||

  1623 		      i==3 && testword[0]=='m' && testword[1]=='a' &&

  1624 		      testword[2]=='c' || i>0 && testword[i-1]==CHAR_SQUOTE)

  1625 			; /* do nothing! */

  1626 		    else

  1627 			istypo=1;

  1628 		}

  1629 		testword[i]=(char)tolower(testword[i]);

  1630 	    }

  1631 	    /*

  1632 	     * Check for certain unlikely two-letter combinations at word

  1633 	     * start and end.

  1634 	     */

  1635 	    if (strlen(testword)>1)

  1636 	    {

  1637 		for (i=0;*nostart[i];i++)

  1638 		    if (!strncmp(testword,nostart[i],2))

  1639 			istypo=1;

  1640 		for (i=0;*noend[i];i++)

  1641 		    if (!strncmp(testword+strlen(testword)-2,noend[i],2))

  1642 			istypo=1;

  1643 	    }

  1644 	    /* ght is common, gbt never. Like that. */

  1645 	    if (strstr(testword,"cb"))

  1646 		istypo=1;

  1647 	    if (strstr(testword,"gbt"))

  1648 		istypo=1;

  1649 	    if (strstr(testword,"pbt"))

  1650 		istypo=1;

  1651 	    if (strstr(testword,"tbs"))

  1652 		istypo=1;

  1653 	    if (strstr(testword,"mrn"))

  1654 		istypo=1;

  1655 	    if (strstr(testword,"ahle"))

  1656 		istypo=1;

  1657 	    if (strstr(testword,"ihle"))

  1658 		istypo=1;

  1659 	    /*

  1660 	     * "TBE" does happen - like HEARTBEAT - but uncommon.

  1661 	     * Also "TBI" - frostbite, outbid - but uncommon.

  1662 	     * Similarly "ii" like Hawaii, or Pompeii, and in Roman

  1663 	     * numerals, but "ii" is a common scanno.

  1664 	     */

  1665 	    if (strstr(testword,"tbi"))

  1666 		istypo=1;

  1667 	    if (strstr(testword,"tbe"))

  1668 		istypo=1;

  1669 	    if (strstr(testword,"ii"))

  1670 		istypo=1;

  1671 	    /*

  1672 	     * Check for no vowels or no consonants.

  1673 	     * If none, flag a typo.

  1674 	     */

  1675 	    if (!istypo && strlen(testword)>1)

  1676 	    {

  1677 		vowel=consonant=0;

  1678 		for (i=0;testword[i];i++)

  1679 		{

  1680 		    if (testword[i]=='y' || gcisdigit(testword[i]))

  1681 		    {

  1682 			/* Yah, this is loose. */

  1683 			vowel++;

  1684 			consonant++;

  1685 		    }

  1686 		    else if (strchr(vowels,testword[i]))

  1687 			vowel++;

  1688 		    else

  1689 			consonant++;

  1690 		}

  1691 		if (!vowel || !consonant)

  1692 		    istypo=1;

  1693 	    }

  1694 	    /*

  1695 	     * Now exclude the word from being reported if it's in

  1696 	     * the okword list.

  1697 	     */

  1698 	    for (i=0;*okword[i];i++)

  1699 		if (!strcmp(testword,okword[i]))

  1700 		    istypo=0;

  1701 	    /*

  1702 	     * What looks like a typo may be a Roman numeral.

  1703 	     * Exclude these.

  1704 	     */

  1705 	    if (istypo && isroman(testword))

  1706 		istypo=0;

  1707 	    /* Check the manual list of typos. */

  1708 	    if (!istypo)

  1709 		for (i=0;*typo[i];i++)

  1710 		    if (!strcmp(testword,typo[i]))

  1711 			istypo=1;

  1712 	    /*

  1713 	     * Check lowercase s, l, i and m - special cases.

  1714 	     *   "j" - often a semi-colon gone wrong.

  1715 	     *   "d" for a missing apostrophe - he d

  1716 	     *   "n" for "in"

  1717 	     */

  1718 	    if (!istypo && strlen(testword)==1 && strchr("slmijdn",*inword))

  1719 		istypo=1;

  1720 	    if (istypo)

  1721 	    {

  1722 		isdup=0;

  1723 		if (strlen(testword)<MAX_QWORD_LENGTH &&

  1724 		  !pswit[VERBOSE_SWITCH])

  1725 		    for (i=0;i<qword_index;i++)

  1726 			if (!strcmp(testword,qword[i]))

  1727 			{

  1728 			    isdup=1;

  1729 			    ++dupcnt[i];

  1730 			}

  1731 		if (!isdup)

  1732 		{

  1733 		    if (qword_index<MAX_QWORD &&

  1734 		      strlen(testword)<MAX_QWORD_LENGTH)

  1735 		    {

  1736 			strcpy(qword[qword_index],testword);

  1737 			qword_index++;

  1738 		    }

  1739 		    if (pswit[ECHO_SWITCH])

  1740 			printf("\n%s\n",aline);

  1741 		    if (!pswit[OVERVIEW_SWITCH])

  1742 		    {

  1743 			printf("    Line %ld column %d - Query word %s",

  1744 			  linecnt,(int)(wordstart-aline)+1,inword);

  1745 			if (strlen(testword)<MAX_QWORD_LENGTH &&

  1746 			  !pswit[VERBOSE_SWITCH])

  1747 			    printf(" - not reporting duplicates");

  1748 			printf("\n");

  1749 		    }

  1750 		    else

  1751 			cnt_word++;

  1752 		}

  1753 	    }

  1754 	}

  1755 	/* check the user's list of typos */

  1756 	if (!istypo && usertypo_count)

  1757 	    for (i=0;i<usertypo_count;i++)

  1758 		if (!strcmp(testword,usertypo[i]))

  1759 		{

  1760 		    if (pswit[ECHO_SWITCH])

  1761 			printf("\n%s\n",aline);

  1762 		    if (!pswit[OVERVIEW_SWITCH])

  1763 			printf("    Line %ld column %d - "

  1764 			  "Query possible scanno %s\n",

  1765 			  linecnt,(int)(wordstart-aline)+2,inword);

  1766 		}

  1767 	if (pswit[PARANOID_SWITCH] && warnings->digit)

  1768 	{

  1769 	    /* In paranoid mode, query all 0 and 1 standing alone. */

  1770 	    if (!strcmp(inword,"0") || !strcmp(inword,"1"))

  1771 	    {

  1772 		if (pswit[ECHO_SWITCH])

  1773 		    printf("\n%s\n",aline);

  1774 		if (!pswit[OVERVIEW_SWITCH])

  1775 		    printf("    Line %ld column %d - Query standalone %s\n",

  1776 		      linecnt,(int)(wordstart-aline)+2,inword);

  1777 		else

  1778 		    cnt_word++;

  1779 	    }

  1780 	}

  1781     }

  1782 }

  1784 struct parities {

  1785     int dquote,squote;

  1786 };

  1788 /*

  1789  * check_for_misspaced_punctuation:

  1790  *

  1791  * Look for added or missing spaces around punctuation and quotes.

  1792  * If there is a punctuation character like ! with no space on

  1793  * either side, suspect a missing!space. If there are spaces on

  1794  * both sides , assume a typo. If we see a double quote with no

  1795  * space or punctuation on either side of it, assume unspaced

  1796  * quotes "like"this.

  1797  */

  1798 void check_for_misspaced_punctuation(const char *aline,

  1799   struct parities *parities,int isemptyline)

  1800 {

  1801     int i,llen,isacro,isellipsis;

  1802     const char *s;

  1803     llen=strlen(aline);

  1804     for (i=1;i<llen;i++)

  1805     {

  1806 	/* For each character in the line after the first. */

  1807 	if (strchr(".?!,;:_",aline[i]))  /* if it's punctuation */

  1808 	{

  1809 	    /* we need to suppress warnings for acronyms like M.D. */

  1810 	    isacro=0;

  1811 	    /* we need to suppress warnings for ellipsis . . . */

  1812 	    isellipsis=0;

  1813 	    /* if there are letters on both sides of it or ... */

  1814 	    if (gcisalpha(aline[i-1]) && gcisalpha(aline[i+1]) ||

  1815 	       gcisalpha(aline[i+1]) && strchr("?!,;:",aline[i]))

  1816 	    {

  1817 		/* ...if it's strict punctuation followed by an alpha */

  1818 		if (aline[i]=='.')

  1819 		{

  1820 		    if (i>2 && aline[i-2]=='.')

  1821 			isacro=1;

  1822 		    if (i+2<llen && aline[i+2]=='.')

  1823 			isacro=1;

  1824 		}

  1825 		if (!isacro)

  1826 		{

  1827 		    if (pswit[ECHO_SWITCH])

  1828 			printf("\n%s\n",aline);

  1829 		    if (!pswit[OVERVIEW_SWITCH])

  1830 			printf("    Line %ld column %d - Missing space?\n",

  1831 			  linecnt,i+1);

  1832 		    else

  1833 			cnt_punct++;

  1834 		}

  1835 	    }

  1836 	    if (aline[i-1]==CHAR_SPACE &&

  1837 	      (aline[i+1]==CHAR_SPACE || aline[i+1]==0))

  1838 	    {

  1839 		/*

  1840 		 * If there are spaces on both sides,

  1841 		 * or space before and end of line.

  1842 		 */

  1843 		if (aline[i]=='.')

  1844 		{

  1845 		    if (i>2 && aline[i-2]=='.')

  1846 			isellipsis=1;

  1847 		    if (i+2<llen && aline[i+2]=='.')

  1848 			isellipsis=1;

  1849 		}

  1850 		if (!isemptyline && !isellipsis)

  1851 		{

  1852 		    if (pswit[ECHO_SWITCH])

  1853 			printf("\n%s\n",aline);

  1854 		    if (!pswit[OVERVIEW_SWITCH])

  1855 			printf("    Line %ld column %d - "

  1856 			  "Spaced punctuation?\n",linecnt,i+1);

  1857 		    else

  1858 			cnt_punct++;

  1859 		}

  1860 	    }

  1861 	}

  1862     }

  1863     /* Split out the characters that CANNOT be preceded by space. */

  1864     llen=strlen(aline);

  1865     for (i=1;i<llen;i++)

  1866     {

  1867 	/* for each character in the line after the first */

  1868 	if (strchr("?!,;:",aline[i]))

  1869 	{

  1870 	    /* if it's punctuation that _cannot_ have a space before it */

  1871 	    if (aline[i-1]==CHAR_SPACE && !isemptyline &&

  1872 	      aline[i+1]!=CHAR_SPACE)

  1873 	    {

  1874 		/*

  1875 		 * If aline[i+1) DOES == space,

  1876 		 * it was already reported just above.

  1877 		 */

  1878 		if (pswit[ECHO_SWITCH])

  1879 		    printf("\n%s\n",aline);

  1880 		if (!pswit[OVERVIEW_SWITCH])

  1881 		    printf("    Line %ld column %d - Spaced punctuation?\n",

  1882 		      linecnt,i+1);

  1883 		else

  1884 		    cnt_punct++;

  1885 	    }

  1886 	}

  1887     }

  1888     /*

  1889      * Special case " .X" where X is any alpha.

  1890      * This plugs a hole in the acronym code above.

  1891      * Inelegant, but maintainable.

  1892      */

  1893     llen=strlen(aline);

  1894     for (i=1;i<llen;i++)

  1895     {

  1896 	/* for each character in the line after the first */

  1897 	if (aline[i]=='.')

  1898 	{

  1899 	    /* if it's a period */

  1900 	    if (aline[i-1]==CHAR_SPACE && gcisalpha(aline[i+1]))

  1901 	    {

  1902 		/*

  1903 		 * If the period follows a space and

  1904 		 * is followed by a letter.

  1905 		 */

  1906 		if (pswit[ECHO_SWITCH])

  1907 		    printf("\n%s\n",aline);

  1908 		if (!pswit[OVERVIEW_SWITCH])

  1909 		    printf("    Line %ld column %d - Spaced punctuation?\n",

  1910 		      linecnt,i+1);

  1911 		else

  1912 		    cnt_punct++;

  1913 	    }

  1914 	}

  1915     }

  1916     for (i=1;i<llen;i++)

  1917     {

  1918 	/* for each character in the line after the first */

  1919 	if (aline[i]==CHAR_DQUOTE)

  1920 	{

  1921 	    if (!strchr(" _-.'`,;:!/([{?}])",aline[i-1]) &&

  1922 	      !strchr(" _-.'`,;:!/([{?}])",aline[i+1]) && aline[i+1] ||

  1923 	      !strchr(" _-([{'`",aline[i-1]) && gcisalpha(aline[i+1]))

  1924 	    {

  1925 		if (pswit[ECHO_SWITCH])

  1926 		    printf("\n%s\n",aline);

  1927 		if (!pswit[OVERVIEW_SWITCH])

  1928 		    printf("    Line %ld column %d - Unspaced quotes?\n",

  1929 		      linecnt,i+1);

  1930 		else

  1931 		    cnt_punct++;

  1932 	    }

  1933 	}

  1934     }

  1935     /* Check parity of quotes. */

  1936     for (s=aline;*s;s++)

  1937     {

  1938 	if (*s==CHAR_DQUOTE)

  1939 	{

  1940 	    parities->dquote=!parities->dquote;

  1941 	    if (!parities->dquote)

  1942 	    {

  1943 		/* parity even */

  1944 		if (!strchr("_-.'`/,;:!?)]} ",s[1]))

  1945 		{

  1946 		    if (pswit[ECHO_SWITCH])

  1947 			printf("\n%s\n",aline);

  1948 		    if (!pswit[OVERVIEW_SWITCH])

  1949 			printf("    Line %ld column %d - "

  1950 			  "Wrongspaced quotes?\n",linecnt,(int)(s-aline)+1);

  1951 		    else

  1952 			cnt_punct++;

  1953 		}

  1954 	    }

  1955 	    else

  1956 	    {

  1957 		/* parity odd */

  1958 		if (!gcisalpha(s[1]) && !isdigit(s[1]) &&

  1959 		  !strchr("_-/.'`([{$",s[1]) || !s[1])

  1960 		{

  1961 		    if (pswit[ECHO_SWITCH])

  1962 			printf("\n%s\n",aline);

  1963 		    if (!pswit[OVERVIEW_SWITCH])

  1964 			printf("    Line %ld column %d - "

  1965 			  "Wrongspaced quotes?\n",linecnt,(int)(s-aline)+1);

  1966 		    else

  1967 			cnt_punct++;

  1968 		}

  1969 	    }

  1970 	}

  1971     }

  1972     if (*aline==CHAR_DQUOTE)

  1973     {

  1974 	if (strchr(",;:!?)]} ",aline[1]))

  1975 	{

  1976 	    if (pswit[ECHO_SWITCH])

  1977 		printf("\n%s\n",aline);

  1978 	    if (!pswit[OVERVIEW_SWITCH])

  1979 		printf("    Line %ld column 1 - Wrongspaced quotes?\n",

  1980 		  linecnt);

  1981 	    else

  1982 		cnt_punct++;

  1983 	}

  1984     }

  1985     if (pswit[SQUOTE_SWITCH])

  1986     {

  1987 	for (s=aline;*s;s++)

  1988 	{

  1989 	    if ((*s==CHAR_SQUOTE || *s==CHAR_OPEN_SQUOTE) &&

  1990 	      (s==aline || s>aline && !gcisalpha(s[-1]) ||

  1991 	      !gcisalpha(s[1])))

  1992 	    {

  1993 		parities->squote=!parities->squote;

  1994 		if (!parities->squote)

  1995 		{

  1996 		    /* parity even */

  1997 		    if (!strchr("_-.'`/\",;:!?)]} ",s[1]))

  1998 		    {

  1999 			if (pswit[ECHO_SWITCH])

  2000 			    printf("\n%s\n",aline);

  2001 			if (!pswit[OVERVIEW_SWITCH])

  2002 			    printf("    Line %ld column %d - "

  2003 			      "Wrongspaced singlequotes?\n",

  2004 			      linecnt,(int)(s-aline)+1);

  2005 			else

  2006 			    cnt_punct++;

  2007 		    }

  2008 		}

  2009 		else

  2010 		{

  2011 		    /* parity odd */

  2012 		    if (!gcisalpha(s[1]) && !isdigit(s[1]) &&

  2013 		      !strchr("_-/\".'`",s[1]) || !s[1])

  2014 		    {

  2015 			if (pswit[ECHO_SWITCH])

  2016 			    printf("\n%s\n",aline);

  2017 			if (!pswit[OVERVIEW_SWITCH])

  2018 			    printf("    Line %ld column %d - "

  2019 			      "Wrongspaced singlequotes?\n",

  2020 			      linecnt,(int)(s-aline)+1);

  2021 			else

  2022 			    cnt_punct++;

  2023 		    }

  2024 		}

  2025 	    }

  2026 	}

  2027     }

  2028 }

  2030 /*

  2031  * procfile:

  2032  *

  2033  * Process one file.

  2034  */

  2035 void procfile(char *filename)

  2036 {

  2037     const char *s,*t;

  2038     char parastart[81];     /* first line of current para */

  2039     FILE *infile;

  2040     struct first_pass_results *first_pass_results;

  2041     struct warnings *warnings;

  2042     struct counters counters={0};

  2043     struct line_properties last={0};

  2044     struct parities parities={0};

  2045     int isemptyline;

  2046     long squot,start_para_line;

  2047     signed int i,llen,isacro,isellipsis;

  2048     signed int isnewpara;

  2049     char dquote_err[80],squote_err[80],rbrack_err[80],sbrack_err[80],

  2050       cbrack_err[80],unders_err[80];

  2051     signed int enddash;

  2052     last.start=CHAR_SPACE;

  2053     *dquote_err=*squote_err=*rbrack_err=*cbrack_err=*sbrack_err=

  2054       *unders_err=*prevline=0;

  2055     linecnt=checked_linecnt=start_para_line=0;

  2056     squot=0;

  2057     i=llen=isacro=isellipsis=0;

  2058     isnewpara=enddash=0;

  2059     infile=fopen(filename,"rb");

  2060     if (!infile)

  2061     {

  2062         if (pswit[STDOUT_SWITCH])

  2063             fprintf(stdout,"bookloupe: cannot open %s\n",filename);

  2064         else

  2065             fprintf(stderr,"bookloupe: cannot open %s\n",filename);

  2066 	exit(1);

  2067     }

  2068     fprintf(stdout,"\n\nFile: %s\n\n",filename);

  2069     first_pass_results=first_pass(infile);

  2070     warnings=report_first_pass(first_pass_results);

  2071     rewind(infile);

  2072     /*

  2073      * Here we go with the main pass. Hold onto yer hat!

  2074      * Re-init some variables we've dirtied.

  2075      */

  2076     squot=linecnt=0;

  2077     while (flgets(aline,LINEBUFSIZE-1,infile,linecnt+1))

  2078     {

  2079         linecnt++;

  2080         if (linecnt==1)

  2081 	    isnewpara=1;

  2082         if (pswit[DP_SWITCH] && !strncmp(aline,"-----File: ",11))

  2083 	    continue;    // skip DP page separators completely

  2084         if (linecnt<first_pass_results->firstline ||

  2085 	  (first_pass_results->footerline>0 &&

  2086 	  linecnt>first_pass_results->footerline))

  2087 	{

  2088             if (pswit[HEADER_SWITCH])

  2089 	    {

  2090                 if (!strncmp(aline,"Title:",6))

  2091                     printf("    %s\n",aline);

  2092                 if (!strncmp(aline,"Author:",7))

  2093                     printf("    %s\n",aline);

  2094                 if (!strncmp(aline,"Release Date:",13))

  2095                     printf("    %s\n",aline);

  2096                 if (!strncmp(aline,"Edition:",8))

  2097                     printf("    %s\n\n",aline);

  2098 	    }

  2099             continue;                /* skip through the header */

  2100 	}

  2101         checked_linecnt++;

  2102         s=aline;

  2103         /*

  2104 	 * If we are in a state of unbalanced quotes, and this line

  2105          * doesn't begin with a quote, output the stored error message.

  2106          * If the -P switch was used, print the warning even if the

  2107          * new para starts with quotes.

  2108 	 */

  2109         t=s;

  2110         while (*t==' ')

  2111 	    t++;

  2112         if (*dquote_err)

  2113             if (*t!=CHAR_DQUOTE || pswit[QPARA_SWITCH])

  2114 	    {

  2115                 if (!pswit[OVERVIEW_SWITCH])

  2116 		{

  2117                     if (pswit[ECHO_SWITCH])

  2118 			printf("\n%s\n",parastart);

  2119                     printf(dquote_err);

  2120 		}

  2121                 else

  2122                     cnt_dquot++;

  2123             }

  2124         if (*squote_err)

  2125 	{

  2126             if (*t!=CHAR_SQUOTE && *t!=CHAR_OPEN_SQUOTE ||

  2127 	      pswit[QPARA_SWITCH] || squot)

  2128 	    {

  2129                 if (!pswit[OVERVIEW_SWITCH])

  2130 		{

  2131                     if (pswit[ECHO_SWITCH])

  2132 			printf("\n%s\n",parastart);

  2133                     printf(squote_err);

  2134 		}

  2135                 else

  2136                     cnt_squot++;

  2137 	    }

  2138             squot=0;

  2139 	}

  2140         if (*rbrack_err)

  2141 	{

  2142             if (!pswit[OVERVIEW_SWITCH])

  2143 	    {

  2144                 if (pswit[ECHO_SWITCH])

  2145 		    printf("\n%s\n",parastart);

  2146                 printf(rbrack_err);

  2147 	    }

  2148             else

  2149                 cnt_brack++;

  2150 	}

  2151         if (*sbrack_err)

  2152 	{

  2153             if (!pswit[OVERVIEW_SWITCH])

  2154 	    {

  2155                 if (pswit[ECHO_SWITCH])

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

  2157                 printf(sbrack_err);

  2158 	    }

  2159             else

  2160                 cnt_brack++;

  2161 	}

  2162         if (*cbrack_err)

  2163 	{

  2164             if (!pswit[OVERVIEW_SWITCH])

  2165 	    {

  2166                 if (pswit[ECHO_SWITCH])

  2167 		    printf("\n%s\n",parastart);

  2168                 printf(cbrack_err);

  2169 	    }

  2170             else

  2171                 cnt_brack++;

  2172 	}

  2173         if (*unders_err)

  2174 	{

  2175             if (!pswit[OVERVIEW_SWITCH])

  2176 	    {

  2177                 if (pswit[ECHO_SWITCH])

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

  2179                 printf(unders_err);

  2180 	    }

  2181             else

  2182                 cnt_brack++;

  2183 	}

  2184         *dquote_err=*squote_err=*rbrack_err=*cbrack_err=

  2185 	  *sbrack_err=*unders_err=0;

  2186 	isemptyline=analyse_quotes(aline,&counters);

  2187         if (isnewpara && !isemptyline)

  2188 	{

  2189 	    /* This line is the start of a new paragraph. */

  2190             start_para_line=linecnt;

  2191 	    /* Capture its first line in case we want to report it later. */

  2192             strncpy(parastart,aline,80);

  2193             parastart[79]=0;

  2194 	    memset(&parities,0,sizeof(parities));  /* restart the quote count */

  2195             s=aline;

  2196             while (!gcisalpha(*s) && !gcisdigit(*s) && *s)

  2197 		s++;

  2198             if (*s>='a' && *s<='z')

  2199 	    {

  2200 		/* and its first letter is lowercase */

  2201                 if (pswit[ECHO_SWITCH])

  2202 		    printf("\n%s\n",aline);

  2203                 if (!pswit[OVERVIEW_SWITCH])

  2204                     printf("    Line %ld column %d - "

  2205 		      "Paragraph starts with lower-case\n",

  2206 		      linecnt,(int)(s-aline)+1);

  2207                 else

  2208                     cnt_punct++;

  2209 	    }

  2210             isnewpara=0; /* Signal the end of new para processing. */

  2211 	}

  2212         /* Check for an em-dash broken at line end. */

  2213         if (enddash && *aline=='-')

  2214 	{

  2215             if (pswit[ECHO_SWITCH])

  2216 		printf("\n%s\n",aline);

  2217             if (!pswit[OVERVIEW_SWITCH])

  2218                 printf("    Line %ld column 1 - Broken em-dash?\n",linecnt);

  2219             else

  2220                 cnt_punct++;

  2221 	}

  2222         enddash=0;

  2223         for (s=aline+strlen(aline)-1;*s==' ' && s>aline;s--)

  2224 	    ;

  2225         if (s>=aline && *s=='-')

  2226             enddash=1;

  2227 	/*

  2228          * Check for invalid or questionable characters in the line

  2229          * Anything above 127 is invalid for plain ASCII, and

  2230          * non-printable control characters should also be flagged.

  2231          * Tabs should generally not be there.

  2232 	 */

  2233         for (s=aline;*s;s++)

  2234 	{

  2235             i=(unsigned char)*s;

  2236             if (i<CHAR_SPACE && i!=CHAR_LF && i!=CHAR_CR && i!=CHAR_TAB)

  2237 	    {

  2238                 if (pswit[ECHO_SWITCH])

  2239 		    printf("\n%s\n",aline);

  2240                 if (!pswit[OVERVIEW_SWITCH])

  2241                     printf("    Line %ld column %d - Control character %d\n",

  2242 		      linecnt,(int)(s-aline)+1,i);

  2243                 else

  2244                     cnt_bin++;

  2245 	    }

  2246 	}

  2247         if (warnings->bin)

  2248 	    check_for_odd_characters(aline,warnings,isemptyline);

  2249         if (warnings->longline)

  2250 	    check_for_long_line(aline);

  2251         if (warnings->shortline)

  2252 	    check_for_short_line(aline,&last);

  2253         last.blen=last.len;

  2254         last.len=strlen(aline);

  2255         last.start=aline[0];

  2256 	check_for_starting_punctuation(aline);

  2257         if (warnings->dash)

  2258 	{

  2259 	    check_for_spaced_emdash(aline);

  2260 	    check_for_spaced_dash(aline);

  2261 	}

  2262 	check_for_unmarked_paragraphs(aline);

  2263 	check_for_jeebies(aline);

  2264 	check_for_mta_from(aline);

  2265 	check_for_orphan_character(aline);

  2266 	check_for_pling_scanno(aline);

  2267 	check_for_extra_period(aline,warnings);

  2268 	check_for_following_punctuation(aline);

  2269 	check_for_typos(aline,warnings);

  2270 	check_for_misspaced_punctuation(aline,&parities,isemptyline);

  2271         /*

  2272 	 * Look for double punctuation like ,. or ,,

  2273          * Thanks to DW for the suggestion!

  2274          * In books with references, ".," and ".;" are common

  2275          * e.g. "etc., etc.," and vol. 1.; vol 3.;

  2276          * OTOH, from my initial tests, there are also fairly

  2277          * common errors. What to do? Make these cases paranoid?

  2278          * ".," is the most common, so warnings->dotcomma is used

  2279          * to suppress detailed reporting if it occurs often.

  2280 	 */

  2281         llen=strlen(aline);

  2282         for (i=0;i<llen;i++)

  2283 	{

  2284 	    /* for each punctuation character in the line */

  2285             if (strchr(".?!,;:",aline[i]) && (strchr(".?!,;:",aline[i+1])) &&

  2286 	      aline[i] && aline[i+1])

  2287 	    {

  2288 		/* followed by punctuation, it's a query, unless . . . */

  2289                 if (aline[i]==aline[i+1] && (aline[i]=='.' || aline[i]=='?' ||

  2290 		  aline[i]=='!') ||

  2291 		  !warnings->dotcomma && aline[i]=='.' && aline[i+1]==',' ||

  2292 		  warnings->isFrench && !strncmp(aline+i,",...",4) ||

  2293 		  warnings->isFrench && !strncmp(aline+i,"...,",4) ||

  2294 		  warnings->isFrench && !strncmp(aline+i,";...",4) ||

  2295 		  warnings->isFrench && !strncmp(aline+i,"...;",4) ||

  2296 		  warnings->isFrench && !strncmp(aline+i,":...",4) ||

  2297 		  warnings->isFrench && !strncmp(aline+i,"...:",4) ||

  2298 		  warnings->isFrench && !strncmp(aline+i,"!...",4) ||

  2299 		  warnings->isFrench && !strncmp(aline+i,"...!",4) ||

  2300 		  warnings->isFrench && !strncmp(aline+i,"?...",4) ||

  2301 		  warnings->isFrench && !strncmp(aline+i,"...?",4))

  2302 		{

  2303 		    if (warnings->isFrench && !strncmp(aline+i,",...",4) ||

  2304 		      warnings->isFrench && !strncmp(aline+i,"...,",4) ||

  2305 		      warnings->isFrench && !strncmp(aline+i,";...",4) ||

  2306 		      warnings->isFrench && !strncmp(aline+i,"...;",4) ||

  2307 		      warnings->isFrench && !strncmp(aline+i,":...",4) ||

  2308 		      warnings->isFrench && !strncmp(aline+i,"...:",4) ||

  2309 		      warnings->isFrench && !strncmp(aline+i,"!...",4) ||

  2310 		      warnings->isFrench && !strncmp(aline+i,"...!",4) ||

  2311 		      warnings->isFrench && !strncmp(aline+i,"?...",4) ||

  2312 		      warnings->isFrench && !strncmp(aline+i,"...?",4))

  2313 			i+=4;

  2314 		    ; /* do nothing for .. !! and ?? which can be legit */

  2315 		}

  2316                 else

  2317 		{

  2318                     if (pswit[ECHO_SWITCH])

  2319 			printf("\n%s\n",aline);

  2320                     if (!pswit[OVERVIEW_SWITCH])

  2321                         printf("    Line %ld column %d - Double punctuation?\n",

  2322 			  linecnt,i+1);

  2323                     else

  2324                         cnt_punct++;

  2325 		}

  2326 	    }

  2327 	}

  2328         s=aline;

  2329         while (strstr(s," \" "))

  2330 	{

  2331             if (pswit[ECHO_SWITCH])

  2332 		printf("\n%s\n",aline);

  2333             if (!pswit[OVERVIEW_SWITCH])

  2334                 printf("    Line %ld column %d - Spaced doublequote?\n",

  2335 		  linecnt,(int)(strstr(s," \" ")-aline+1));

  2336             else

  2337                 cnt_punct++;

  2338             s=strstr(s," \" ")+2;

  2339 	}

  2340         s=aline;

  2341         while (strstr(s," ' "))

  2342 	{

  2343             if (pswit[ECHO_SWITCH])

  2344 		printf("\n%s\n",aline);

  2345             if (!pswit[OVERVIEW_SWITCH])

  2346                 printf("    Line %ld column %d - Spaced singlequote?\n",

  2347 		  linecnt,(int)(strstr(s," ' ")-aline+1));

  2348             else

  2349                 cnt_punct++;

  2350             s=strstr(s," ' ")+2;

  2351 	}

  2352         s=aline;

  2353         while (strstr(s," ` "))

  2354 	{

  2355             if (pswit[ECHO_SWITCH])

  2356 		printf("\n%s\n",aline);

  2357             if (!pswit[OVERVIEW_SWITCH])

  2358                 printf("    Line %ld column %d - Spaced singlequote?\n",

  2359 		  linecnt,(int)(strstr(s," ` ")-aline+1));

  2360             else

  2361                 cnt_punct++;

  2362             s=strstr(s," ` ")+2;

  2363 	}

  2364         /* check special case of 'S instead of 's at end of word */

  2365         s=aline+1;

  2366         while (*s)

  2367 	{

  2368             if (*s==CHAR_SQUOTE && s[1]=='S' && s[-1]>='a' && s[-1]<='z')

  2369 	    {

  2370                 if (pswit[ECHO_SWITCH])

  2371 		    printf("\n%s\n",aline);

  2372                 if (!pswit[OVERVIEW_SWITCH])

  2373                     printf("    Line %ld column %d - Capital \"S\"?\n",

  2374 		      linecnt,(int)(s-aline+2));

  2375                 else

  2376                     cnt_punct++;

  2377 	    }

  2378             s++;

  2379 	}

  2380         /*

  2381 	 * Now check special cases - start and end of line -

  2382          * for single and double quotes. Start is sometimes [sic]

  2383          * but better to query it anyway.

  2384          * While we're here, check for dash at end of line.

  2385 	 */

  2386         llen=strlen(aline);

  2387         if (llen>1)

  2388 	{

  2389             if (aline[llen-1]==CHAR_DQUOTE || aline[llen-1]==CHAR_SQUOTE ||

  2390 	      aline[llen-1]==CHAR_OPEN_SQUOTE)

  2391                 if (aline[llen-2]==CHAR_SPACE)

  2392 		{

  2393                     if (pswit[ECHO_SWITCH])

  2394 			printf("\n%s\n",aline);

  2395                     if (!pswit[OVERVIEW_SWITCH])

  2396                         printf("    Line %ld column %d - Spaced quote?\n",

  2397 			  linecnt,llen);

  2398                     else

  2399                         cnt_punct++;

  2400 		}

  2401             if ((aline[0]==CHAR_SQUOTE || aline[0]==CHAR_OPEN_SQUOTE) &&

  2402 	      aline[1]==CHAR_SPACE)

  2403 	    {

  2404 		if (pswit[ECHO_SWITCH])

  2405 		    printf("\n%s\n",aline);

  2406 		if (!pswit[OVERVIEW_SWITCH])

  2407 		    printf("    Line %ld column 1 - Spaced quote?\n",linecnt);

  2408 		else

  2409 		    cnt_punct++;

  2410 	    }

  2411             /*

  2412 	     * Dash at end of line may well be legit - paranoid mode only

  2413              * and don't report em-dash at line-end.

  2414 	     */

  2415             if (pswit[PARANOID_SWITCH] && warnings->hyphen)

  2416 	    {

  2417                 for (i=llen-1;i>0 && (unsigned char)aline[i]<=CHAR_SPACE;i--)

  2418 		    ;

  2419                 if (aline[i]=='-' && aline[i-1]!='-')

  2420 		{

  2421                     if (pswit[ECHO_SWITCH])

  2422 			printf("\n%s\n",aline);

  2423                     if (!pswit[OVERVIEW_SWITCH])

  2424                         printf("    Line %ld column %d - "

  2425 			  "Hyphen at end of line?\n",linecnt,i);

  2426 		}

  2427 	    }

  2428 	}

  2429         /*

  2430 	 * Brackets are often unspaced, but shouldn't be surrounded by alpha.

  2431          * If so, suspect a scanno like "a]most".

  2432 	 */

  2433         llen=strlen(aline);

  2434         for (i=1;i<llen-1;i++)

  2435 	{

  2436 	    /* for each bracket character in the line except 1st & last */

  2437             if (strchr("{[()]}",aline[i]) && gcisalpha(aline[i-1]) &&

  2438 	      gcisalpha(aline[i+1]))

  2439 	    {

  2440                 if (pswit[ECHO_SWITCH])

  2441 		    printf("\n%s\n",aline);

  2442                 if (!pswit[OVERVIEW_SWITCH])

  2443                     printf("    Line %ld column %d - Unspaced bracket?\n",

  2444 		      linecnt,i);

  2445                 else

  2446                     cnt_punct++;

  2447 	    }

  2448 	}

  2449         llen=strlen(aline);

  2450         if (warnings->endquote)

  2451 	{

  2452             for (i=1;i<llen;i++)

  2453 	    {

  2454 		/* for each character in the line except 1st */

  2455                 if (aline[i]==CHAR_DQUOTE && isalpha(aline[i-1]))

  2456 		{

  2457 		    if (pswit[ECHO_SWITCH])

  2458 			printf("\n%s\n",aline);

  2459 		    if (!pswit[OVERVIEW_SWITCH])

  2460 			printf("    Line %ld column %d - "

  2461 			  "endquote missing punctuation?\n",linecnt,i);

  2462 		    else

  2463 			cnt_punct++;

  2464 		}

  2465 	    }

  2466 	}

  2467 	/*

  2468          * Check for <HTML TAG>.

  2469          * If there is a < in the line, followed at some point

  2470          * by a > then we suspect HTML.

  2471 	 */

  2472         if (strstr(aline,"<") && strstr(aline,">"))

  2473 	{

  2474             i=(signed int)(strstr(aline,">")-strstr(aline,"<")+1);

  2475             if (i>0)

  2476 	    {

  2477                 strncpy(wrk,strstr(aline,"<"),i);

  2478                 wrk[i]=0;

  2479                 if (pswit[ECHO_SWITCH])

  2480 		    printf("\n%s\n",aline);

  2481                 if (!pswit[OVERVIEW_SWITCH])

  2482                     printf("    Line %ld column %d - HTML Tag? %s \n",

  2483 		      linecnt,(int)(strstr(aline,"<")-aline)+1,wrk);

  2484                 else

  2485                     cnt_html++;

  2486 	    }

  2487 	}

  2488         /*

  2489 	 * Check for &symbol; HTML.

  2490          * If there is a & in the line, followed at

  2491          * some point by a ; then we suspect HTML.

  2492 	 */

  2493         if (strstr(aline,"&") && strstr(aline,";"))

  2494 	{

  2495             i=(int)(strstr(aline,";")-strstr(aline,"&")+1);

  2496             for (s=strstr(aline,"&");s<strstr(aline,";");s++)

  2497                 if (*s==CHAR_SPACE)

  2498 		    i=0;                /* Don't report "Jones & Son;" */

  2499             if (i>0)

  2500 	    {

  2501                 strncpy(wrk,strstr(aline,"&"),i);

  2502                 wrk[i]=0;

  2503                 if (pswit[ECHO_SWITCH])

  2504 		    printf("\n%s\n",aline);

  2505                 if (!pswit[OVERVIEW_SWITCH])

  2506                     printf("    Line %ld column %d - HTML symbol? %s \n",

  2507 		      linecnt,(int)(strstr(aline,"&")-aline)+1,wrk);

  2508                 else

  2509                     cnt_html++;

  2510 	    }

  2511 	}

  2512         /*

  2513 	 * At end of paragraph, check for mismatched quotes.

  2514          * We don't want to report an error immediately, since it is a

  2515          * common convention to omit the quotes at end of paragraph if

  2516          * the next paragraph is a continuation of the same speaker.

  2517          * Where this is the case, the next para should begin with a

  2518          * quote, so we store the warning message and only display it

  2519          * at the top of the next iteration if the new para doesn't

  2520          * start with a quote.

  2521          * The -p switch overrides this default, and warns of unclosed

  2522          * quotes on _every_ paragraph, whether the next begins with a

  2523          * quote or not.

  2524 	 */

  2525         if (isemptyline)

  2526 	{

  2527 	    /* end of para - add up the totals */

  2528             if (counters.quot%2)

  2529                 sprintf(dquote_err,"    Line %ld - Mismatched quotes\n",

  2530 		  linecnt);

  2531             if (pswit[SQUOTE_SWITCH] && counters.open_single_quote &&

  2532 	      counters.open_single_quote!=counters.close_single_quote)

  2533                 sprintf(squote_err,"    Line %ld - Mismatched singlequotes?\n",

  2534 		  linecnt);

  2535             if (pswit[SQUOTE_SWITCH] && counters.open_single_quote &&

  2536 	      counters.open_single_quote!=counters.close_single_quote &&

  2537 	      counters.open_single_quote!=counters.close_single_quote+1)

  2538 		/*

  2539 		 * Flag it to be noted regardless of the

  2540 		 * first char of the next para.

  2541 		 */

  2542                 squot=1;

  2543             if (counters.r_brack)

  2544                 sprintf(rbrack_err,"    Line %ld - "

  2545 		  "Mismatched round brackets?\n",linecnt);

  2546             if (counters.s_brack)

  2547                 sprintf(sbrack_err,"    Line %ld - "

  2548 		  "Mismatched square brackets?\n",linecnt);

  2549             if (counters.c_brack)

  2550                 sprintf(cbrack_err,"    Line %ld - "

  2551 		  "Mismatched curly brackets?\n",linecnt);

  2552             if (counters.c_unders%2)

  2553                 sprintf(unders_err,"    Line %ld - Mismatched underscores?\n",

  2554 		  linecnt);

  2555 	    memset(&counters,0,sizeof(counters));

  2556 	    /* let the next iteration know that it's starting a new para */

  2557             isnewpara=1;

  2558 	}

  2559         /*

  2560 	 * Check for omitted punctuation at end of paragraph by working back

  2561 	 * through prevline. DW.

  2562          * Need to check this only for "normal" paras.

  2563          * So what is a "normal" para?

  2564          *    Not normal if one-liner (chapter headings, etc.)

  2565          *    Not normal if doesn't contain at least one locase letter

  2566          *    Not normal if starts with space

  2567 	 */

  2568         if (isemptyline)

  2569 	{

  2570 	    /* end of para */

  2571             for (s=prevline,i=0;*s && !i;s++)

  2572                 if (gcisletter(*s))

  2573 		    /* use i to indicate the presence of a letter on the line */

  2574                     i=1;

  2575             /*

  2576 	     * This next "if" is a problem.

  2577              * If we say "start_para_line <= linecnt - 1", that includes

  2578 	     * one-line "paragraphs" like chapter heads. Lotsa false positives.

  2579              * If we say "start_para_line < linecnt - 1" it doesn't, but then it

  2580              * misses genuine one-line paragraphs.

  2581 	     */

  2582             if (i && last.blen>2 && start_para_line<linecnt-1 &&

  2583 	      *prevline>CHAR_SPACE)

  2584 	    {

  2585                 for (i=strlen(prevline)-1;

  2586 		  (prevline[i]==CHAR_DQUOTE || prevline[i]==CHAR_SQUOTE) &&

  2587 		  prevline[i]>CHAR_SPACE && i>0;

  2588 		  i--)

  2589 		    ;

  2590                 for (;i>0;i--)

  2591 		{

  2592                     if (gcisalpha(prevline[i]))

  2593 		    {

  2594                         if (pswit[ECHO_SWITCH])

  2595 			    printf("\n%s\n",prevline);

  2596                         if (!pswit[OVERVIEW_SWITCH])

  2597                             printf("    Line %ld column %d - "

  2598 			      "No punctuation at para end?\n",

  2599 			      linecnt-1,strlen(prevline));

  2600                         else

  2601                             cnt_punct++;

  2602                         break;

  2603 		    }

  2604                     if (strchr("-.:!([{?}])",prevline[i]))

  2605                         break;

  2606 		}

  2607 	    }

  2608 	}

  2609         strcpy(prevline,aline);

  2610     }

  2611     fclose(infile);

  2612     if (!pswit[OVERVIEW_SWITCH])

  2613         for (i=0;i<MAX_QWORD;i++)

  2614             if (dupcnt[i])

  2615                 printf("\nNote: Queried word %s was duplicated %d time%s\n",

  2616 		  qword[i],dupcnt[i],"s");

  2617 }

  2619 /*

  2620  * flgets:

  2621  *

  2622  * Get one line from the input stream, checking for

  2623  * the existence of exactly one CR/LF line-end per line.

  2624  *

  2625  * Returns: a pointer to the line.

  2626  */

  2627 char *flgets(char *theline,int maxlen,FILE *thefile,long lcnt)

  2628 {

  2629     char c;

  2630     int len,isCR,cint;

  2631     *theline=0;

  2632     len=isCR=0;

  2633     c=cint=fgetc(thefile);

  2634     do

  2635     {

  2636         if (cint==EOF)

  2637             return NULL;

  2638 	/* either way, it's end of line */

  2639         if (c==10)

  2640 	{

  2641             if (isCR)

  2642                 break;

  2643             else

  2644 	    {

  2645 		/* Error - a LF without a preceding CR */

  2646                 if (pswit[LINE_END_SWITCH])

  2647 		{

  2648                     if (pswit[ECHO_SWITCH])

  2649 			printf("\n%s\n",theline);

  2650                     if (!pswit[OVERVIEW_SWITCH])

  2651                         printf("    Line %ld - No CR?\n",lcnt);

  2652                     else

  2653                         cnt_lineend++;

  2654 		}

  2655                 break;

  2656 	    }

  2657 	}

  2658         if (c==13)

  2659 	{

  2660             if (isCR)

  2661 	    {

  2662 		/* Error - two successive CRs */

  2663                 if (pswit[LINE_END_SWITCH])

  2664 		{

  2665                     if (pswit[ECHO_SWITCH])

  2666 			printf("\n%s\n",theline);

  2667                     if (!pswit[OVERVIEW_SWITCH])

  2668                         printf("    Line %ld - Two successive CRs?\n",lcnt);

  2669                     else

  2670                         cnt_lineend++;

  2671 		}

  2672 	    }

  2673             isCR=1;

  2674 	}

  2675         else

  2676 	{

  2677             if (pswit[LINE_END_SWITCH] && isCR)

  2678 	    {

  2679                 if (pswit[ECHO_SWITCH])

  2680 		    printf("\n%s\n",theline);

  2681                 if (!pswit[OVERVIEW_SWITCH])

  2682                     printf("    Line %ld column %d - CR without LF?\n",

  2683 		      lcnt,len+1);

  2684                 else

  2685                     cnt_lineend++;

  2686 	    }

  2687             theline[len]=c;

  2688             len++;

  2689             theline[len]=0;

  2690             isCR=0;

  2691 	}

  2692         c=cint=fgetc(thefile);

  2693     } while(len<maxlen);

  2694     if (pswit[MARKUP_SWITCH])

  2695         postprocess_for_HTML(theline);

  2696     if (pswit[DP_SWITCH])

  2697         postprocess_for_DP(theline);

  2698     return theline;

  2699 }

  2701 /*

  2702  * mixdigit:

  2703  *

  2704  * Takes a "word" as a parameter, and checks whether it

  2705  * contains a mixture of alpha and digits. Generally, this is an

  2706  * error, but may not be for cases like 4th or L5 12s. 3d.

  2707  *

  2708  * Returns: 0 if no error found, 1 if error.

  2709  */

  2710 int mixdigit(char *checkword)

  2711 {

  2712     int wehaveadigit,wehavealetter,firstdigits,query,wl;

  2713     char *s;

  2714     wehaveadigit=wehavealetter=query=0;

  2715     for (s=checkword;*s;s++)

  2716         if (gcisalpha(*s))

  2717             wehavealetter=1;

  2718         else

  2719             if (gcisdigit(*s))

  2720                 wehaveadigit=1;

  2721     if (wehaveadigit && wehavealetter)

  2722     {

  2723 	/* Now exclude common legit cases, like "21st" and "12l. 3s. 11d." */

  2724         query=1;

  2725         wl=strlen(checkword);

  2726         for (firstdigits=0;gcisdigit(checkword[firstdigits]);firstdigits++)

  2727             ;

  2728         /* digits, ending in st, rd, nd, th of either case */

  2729         if (firstdigits+2==wl && (matchword(checkword+wl-2,"st") ||

  2730 	  matchword(checkword+wl-2,"rd") || matchword(checkword+wl-2,"nd") ||

  2731 	  matchword(checkword+wl-2,"th")))

  2732 	    query=0;

  2733         if (firstdigits+3==wl && (matchword(checkword+wl-3,"sts") ||

  2734 	  matchword(checkword+wl-3,"rds") || matchword(checkword+wl-3,"nds") ||

  2735 	  matchword(checkword+wl-3,"ths")))

  2736 	    query=0;

  2737         if (firstdigits+3==wl && (matchword(checkword+wl-4,"stly") ||

  2738 	  matchword(checkword+wl-4,"rdly") ||

  2739 	  matchword(checkword+wl-4,"ndly") || matchword(checkword+wl-4,"thly")))

  2740 	    query=0;

  2741         /* digits, ending in l, L, s or d */

  2742         if (firstdigits+1==wl && (checkword[wl-1]=='l' ||

  2743 	  checkword[wl-1]=='L' || checkword[wl-1]=='s' || checkword[wl-1]=='d'))

  2744 	    query=0;

  2745         /*

  2746 	 * L at the start of a number, representing Britsh pounds, like L500.

  2747          * This is cute. We know the current word is mixeddigit. If the first

  2748          * letter is L, there must be at least one digit following. If both

  2749          * digits and letters follow, we have a genuine error, else we have a

  2750          * capital L followed by digits, and we accept that as a non-error.

  2751 	 */

  2752         if (checkword[0]=='L' && !mixdigit(checkword+1))

  2753 	    query=0;

  2754     }

  2755     return query;

  2756 }

  2758 /*

  2759  * getaword:

  2760  *

  2761  * Extracts the first/next "word" from the line, and puts

  2762  * it into "thisword". A word is defined as one English word unit--or

  2763  * at least that's the aim.

  2764  *

  2765  * Returns: a pointer to the position in the line where we will start

  2766  *          looking for the next word.

  2767  */

  2768 const char *getaword(const char *fromline,char *thisword)

  2769 {

  2770     int i,wordlen;

  2771     const char *s;

  2772     wordlen=0;

  2773     for (;!gcisdigit(*fromline) && !gcisalpha(*fromline) && *fromline;

  2774       fromline++)

  2775 	;

  2776     /*

  2777      * Use a look-ahead to handle exceptions for numbers like 1,000 and 1.35.

  2778      * Especially yucky is the case of L1,000

  2779      * This section looks for a pattern of characters including a digit

  2780      * followed by a comma or period followed by one or more digits.

  2781      * If found, it returns this whole pattern as a word; otherwise we discard

  2782      * the results and resume our normal programming.

  2783      */

  2784     s=fromline;

  2785     for (;(gcisdigit(*s) || gcisalpha(*s) || *s==',' || *s=='.') &&

  2786       wordlen<MAXWORDLEN;s++)

  2787     {

  2788 	thisword[wordlen]=*s;

  2789         wordlen++;

  2790     }

  2791     thisword[wordlen]=0;

  2792     for (i=1;i<wordlen-1;i++)

  2793     {

  2794         if (thisword[i]=='.' || thisword[i]==',')

  2795 	{

  2796             if (gcisdigit(thisword[i-1]) && gcisdigit(thisword[i-1]))

  2797 	    {

  2798                 fromline=s;

  2799                 return fromline;

  2800 	    }

  2801 	}

  2802     }

  2803     /* we didn't find a punctuated number - do the regular getword thing */

  2804     wordlen=0;

  2805     for (;(gcisdigit(*fromline) || gcisalpha(*fromline) || *fromline=='\'') &&

  2806       wordlen<MAXWORDLEN;fromline++)

  2807     {

  2808         thisword[wordlen]=*fromline;

  2809         wordlen++;

  2810     }

  2811     thisword[wordlen]=0;

  2812     return fromline;

  2813 }

  2815 /*

  2816  * matchword:

  2817  *

  2818  * A case-insensitive string matcher.

  2819  */

  2820 int matchword(char *checkfor,char *thisword)

  2821 {

  2822     unsigned int ismatch,i;

  2823     if (strlen(checkfor)!=strlen(thisword))

  2824 	return 0;

  2825     ismatch=1;     /* assume a match until we find a difference */

  2826     for (i=0;i<strlen(checkfor);i++)

  2827         if (toupper(checkfor[i])!=toupper(thisword[i]))

  2828             ismatch=0;

  2829     return ismatch;

  2830 }

  2832 /*

  2833  * lowerit:

  2834  *

  2835  * Lowercase the line.

  2836  */

  2838 void lowerit(char *theline)

  2839 {

  2840     for (;*theline;theline++)

  2841         if (*theline>='A' && *theline<='Z')

  2842             *theline+=32;

  2843 }

  2845 /*

  2846  * isroman:

  2847  *

  2848  * Is this word a Roman Numeral?

  2849  *

  2850  * It doesn't actually validate that the number is a valid Roman Numeral--for

  2851  * example it will pass MXXXXXXXXXX as a valid Roman Numeral, but that's not

  2852  * what we're here to do. If it passes this, it LOOKS like a Roman numeral.

  2853  * Anyway, the actual Romans were pretty tolerant of bad arithmetic, or

  2854  * expressions thereof, except when it came to taxes. Allow any number of M,

  2855  * an optional D, an optional CM or CD, any number of optional Cs, an optional

  2856  * XL or an optional XC, an optional IX or IV, an optional V and any number

  2857  * of optional Is.

  2858  */

  2859 int isroman(char *t)

  2860 {

  2861     char *s;

  2862     if (!t || !*t)

  2863 	return 0;

  2864     s=t;

  2865     while (*t=='m' && *t)

  2866 	t++;

  2867     if (*t=='d')

  2868 	t++;

  2869     if (*t=='c' && t[1]=='m')

  2870 	t+=2;

  2871     if (*t=='c' && t[1]=='d')

  2872 	t+=2;

  2873     while (*t=='c' && *t)

  2874 	t++;

  2875     if (*t=='x' && t[1]=='l')

  2876 	t+=2;

  2877     if (*t=='x' && t[1]=='c')

  2878 	t+=2;

  2879     if (*t=='l')

  2880 	t++;

  2881     while (*t=='x' && *t)

  2882 	t++;

  2883     if (*t=='i' && t[1]=='x')

  2884 	t+=2;

  2885     if (*t=='i' && t[1]=='v')

  2886 	t+=2;

  2887     if (*t=='v')

  2888 	t++;

  2889     while (*t=='i' && *t)

  2890 	t++;

  2891     return !*t;

  2892 }

  2894 /*

  2895  * gcisalpha:

  2896  *

  2897  * A version of isalpha() that is somewhat lenient on 8-bit texts.

  2898  * If we use the standard function, 8-bit accented characters break

  2899  * words, so that tete with accented characters appears to be two words, "t"

  2900  * and "t", with 8-bit characters between them. This causes over-reporting of

  2901  * errors. gcisalpha() recognizes accented letters from the CP1252 (Windows)

  2902  * and ISO-8859-1 character sets, which are the most common PG 8-bit types.

  2903  */

  2904 int gcisalpha(unsigned char c)

  2905 {

  2906     if (c>='a' && c<='z')

  2907 	return 1;

  2908     if (c>='A' && c<='Z')

  2909 	return 1;

  2910     if (c<140)

  2911 	return 0;

  2912     if (c>=192 && c!=208 && c!=215 && c!=222 && c!=240 && c!=247 && c!=254)

  2913 	return 1;

  2914     if (c==140 || c==142 || c==156 || c==158 || c==159)

  2915 	return 1;

  2916     return 0;

  2917 }

  2919 /*

  2920  * gcisdigit:

  2921  *

  2922  * A version of isdigit() that doesn't get confused in 8-bit texts.

  2923  */

  2924 int gcisdigit(unsigned char c)

  2925 {

  2926     return c>='0' && c<='9';

  2927 }

  2929 /*

  2930  * gcisletter:

  2931  *

  2932  * A version of isletter() that doesn't get confused in 8-bit texts.

  2933  * NB: this is ISO-8891-1-specific.

  2934  */

  2935 int gcisletter(unsigned char c)

  2936 {

  2937     return c>='A' && c<='Z' || c>='a' && c<='z' || c>=192;

  2938 }

  2940 /*

  2941  * gcstrchr:

  2942  *

  2943  * Wraps strchr to return NULL if the character being searched for is zero.

  2944  */

  2945 char *gcstrchr(char *s,char c)

  2946 {

  2947     if (!c)

  2948 	return NULL;

  2949     return strchr(s,c);

  2950 }

  2952 /*

  2953  * postprocess_for_DP:

  2954  *

  2955  * Invoked with the -d switch from flgets().

  2956  * It simply "removes" from the line a hard-coded set of common

  2957  * DP-specific tags, so that the line passed to the main routine has

  2958  * been pre-cleaned of DP markup.

  2959  */

  2960 void postprocess_for_DP(char *theline)

  2961 {

  2962     char *s,*t;

  2963     int i;

  2964     if (!*theline)

  2965         return;

  2966     for (i=0;*DPmarkup[i];i++)

  2967     {

  2968         s=strstr(theline,DPmarkup[i]);

  2969         while (s)

  2970 	{

  2971             t=s+strlen(DPmarkup[i]);

  2972             while (*t)

  2973 	    {

  2974                 *s=*t;

  2975                 t++;

  2976 		s++;

  2977 	    }

  2978             *s=0;

  2979             s=strstr(theline,DPmarkup[i]);

  2980 	}

  2981     }

  2982 }

  2984 /*

  2985  * postprocess_for_HTML:

  2986  *

  2987  * Invoked with the -m switch from flgets().

  2988  * It simply "removes" from the line a hard-coded set of common

  2989  * HTML tags and "replaces" a hard-coded set of common HTML

  2990  * entities, so that the line passed to the main routine has

  2991  * been pre-cleaned of HTML.

  2992  */

  2993 void postprocess_for_HTML(char *theline)

  2994 {

  2995     if (strstr(theline,"<") && strstr(theline,">"))

  2996         while (losemarkup(theline))

  2997             ;

  2998     while (loseentities(theline))

  2999         ;

  3000 }

  3002 char *losemarkup(char *theline)

  3003 {

  3004     char *s,*t;

  3005     int i;

  3006     if (!*theline)

  3007         return NULL;

  3008     s=strstr(theline,"<");

  3009     t=strstr(theline,">");

  3010     if (!s || !t)

  3011 	return NULL;

  3012     for (i=0;*markup[i];i++)

  3013         if (!tagcomp(s+1,markup[i]))

  3014 	{

  3015             if (!t[1])

  3016 	    {

  3017                 *s=0;

  3018                 return s;

  3019 	    }

  3020             else if (t>s)

  3021 	    {

  3022 		strcpy(s,t+1);

  3023 		return s;

  3024 	    }

  3025         }

  3026     /* It's an unrecognized <xxx>. */

  3027     return NULL;

  3028 }

  3030 char *loseentities(char *theline)

  3031 {

  3032     int i;

  3033     char *s,*t;

  3034     if (!*theline)

  3035         return NULL;

  3036     for (i=0;*entities[i].htmlent;i++)

  3037     {

  3038         s=strstr(theline,entities[i].htmlent);

  3039         if (s)

  3040 	{

  3041             t=malloc((size_t)strlen(s));

  3042             if (!t)

  3043 		return NULL;

  3044             strcpy(t,s+strlen(entities[i].htmlent));

  3045             strcpy(s,entities[i].textent);

  3046             strcat(s,t);

  3047             free(t);

  3048             return theline;

  3049 	}

  3050     }

  3051     for (i=0;*entities[i].htmlnum;i++)

  3052     {

  3053         s=strstr(theline,entities[i].htmlnum);

  3054         if (s)

  3055 	{

  3056             t=malloc((size_t)strlen(s));

  3057             if (!t)

  3058 		return NULL;

  3059             strcpy(t,s+strlen(entities[i].htmlnum));

  3060             strcpy(s,entities[i].textent);

  3061             strcat(s,t);

  3062             free(t);

  3063             return theline;

  3064 	}

  3065     }

  3066     return NULL;

  3067 }

  3069 int tagcomp(char *strin,char *basetag)

  3070 {

  3071     char *s,*t;

  3072     s=basetag;

  3073     t=strin;

  3074     if (*t=='/')

  3075 	t++; /* ignore a slash */

  3076     while (*s && *t)

  3077     {

  3078         if (tolower(*s)!=tolower(*t))

  3079 	    return 1;

  3080         s++;

  3081 	t++;

  3082     }

  3083     return 0;

  3084 }

  3086 void proghelp()

  3087 {

  3088     fputs("Bookloupe version " PACKAGE_VERSION ".\n",stderr);

  3089     fputs("Copyright 2000-2005 Jim Tinsley <jtinsley@pobox.com>.\n",stderr);

  3090     fputs("Copyright 2012- J. Ali Harlow <ali@juiblex.co.uk>.\n",stderr);

  3091     fputs("Bookloupe comes wih ABSOLUTELY NO WARRANTY. "

  3092       "For details, read the file COPYING.\n",stderr);

  3093     fputs("This is Free Software; "

  3094       "you may redistribute it under certain conditions (GPL);\n",stderr);

  3095     fputs("read the file COPYING for details.\n\n",stderr);

  3096     fputs("Usage is: bookloupe [-setpxloyhud] filename\n",stderr);

  3097     fputs("  where -s checks single quotes, -e suppresses echoing lines, "

  3098       "-t checks typos\n",stderr);

  3099     fputs("  -x (paranoid) switches OFF -t and extra checks, "

  3100       "-l turns OFF line-end checks\n",stderr);

  3101     fputs("  -o just displays overview without detail, "

  3102       "-h echoes header fields\n",stderr);

  3103     fputs("  -v (verbose) unsuppresses duplicate reporting, "

  3104       "-m suppresses markup\n",stderr);

  3105     fputs("  -d ignores DP-specific markup,\n",stderr);

  3106     fputs("  -u uses a file gutcheck.typ to query user-defined "

  3107       "possible typos\n",stderr);

  3108     fputs("Sample usage: bookloupe warpeace.txt \n",stderr);

  3109     fputs("\n",stderr);

  3110     fputs("Bookloupe looks for errors in Project Gutenberg(TM) etexts.\n",

  3111       stderr);

  3112     fputs("Bookloupe queries anything it thinks shouldn't be in a PG text; "

  3113       "non-ASCII\n",stderr);

  3114     fputs("characters like accented letters, "

  3115       "lines longer than 75 or shorter than 55,\n",stderr);

  3116     fputs("unbalanced quotes or brackets, "

  3117       "a variety of badly formatted punctuation, \n",stderr);

  3118     fputs("HTML tags, some likely typos. "

  3119       "It is NOT a substitute for human judgement.\n",stderr);

  3120     fputs("\n",stderr);

  3121 }

author	ali <ali@juiblex.co.uk>
	Sun May 26 17:23:48 2013 +0100 (2013-05-26)
changeset 56	8ade5460e220
parent 55	6b786cc05b3c
child 57	1e89f47e56df
permissions	-rw-r--r--