spell.c File Reference

#include "postgres.h"
#include "catalog/pg_collation.h"
#include "miscadmin.h"
#include "tsearch/dicts/spell.h"
#include "tsearch/ts_locale.h"
#include "utils/formatting.h"
#include "utils/memutils.h"

Include dependency graph for spell.c:

Go to the source code of this file.

Data Structures
struct	SplitVar

Macros
#define	tmpalloc(sz)   MemoryContextAlloc(Conf->buildCxt, (sz))
#define	tmpalloc0(sz)   MemoryContextAllocZero(Conf->buildCxt, (sz))
#define	COMPACT_ALLOC_CHUNK   8192 /* amount to get from palloc at once */
#define	COMPACT_MAX_REQ   1024 /* must be < COMPACT_ALLOC_CHUNK */
#define	cpalloc(size)   compact_palloc0(Conf, size)
#define	cpalloc0(size)   compact_palloc0(Conf, size)
#define	MAX_NORM   1024
#define	MAXNORMLEN   256
#define	STRNCMP(s, p)   strncmp( (s), (p), strlen(p) )
#define	GETWCHAR(W, L, N, T)   ( ((const uint8*)(W))[ ((T)==FF_PREFIX) ? (N) : ( (L) - 1 - (N) ) ] )
#define	GETCHAR(A, N, T)   GETWCHAR( (A)->repl, (A)->replen, N, T )
#define	PAE_WAIT_MASK   0
#define	PAE_INMASK   1
#define	PAE_WAIT_FIND   2
#define	PAE_INFIND   3
#define	PAE_WAIT_REPL   4
#define	PAE_INREPL   5
#define	PAE_WAIT_TYPE   6
#define	PAE_WAIT_FLAG   7

Typedefs
typedef struct SplitVar	SplitVar

Functions
void	NIStartBuild (IspellDict *Conf)
void	NIFinishBuild (IspellDict *Conf)
static void *	compact_palloc0 (IspellDict *Conf, size_t size)
static char *	cpstrdup (IspellDict *Conf, const char *str)
static char *	lowerstr_ctx (IspellDict *Conf, const char *src)
static int	cmpspell (const void *s1, const void *s2)
static int	cmpspellaffix (const void *s1, const void *s2)
static int	cmpcmdflag (const void *f1, const void *f2)
static char *	findchar (char *str, int c)
static char *	findchar2 (char *str, int c1, int c2)
static int	strbcmp (const unsigned char *s1, const unsigned char *s2)
static int	strbncmp (const unsigned char *s1, const unsigned char *s2, size_t count)
static int	cmpaffix (const void *s1, const void *s2)
static void	getNextFlagFromString (IspellDict *Conf, const char **sflagset, char *sflag)
static bool	IsAffixFlagInUse (IspellDict *Conf, int affix, const char *affixflag)
static void	NIAddSpell (IspellDict *Conf, const char *word, const char *flag)
void	NIImportDictionary (IspellDict *Conf, const char *filename)
static int	FindWord (IspellDict *Conf, const char *word, const char *affixflag, int flag)
static void	NIAddAffix (IspellDict *Conf, const char *flag, char flagflags, const char *mask, const char *find, const char *repl, int type)
static bool	get_nextfield (char **str, char *next)
static int	parse_ooaffentry (char *str, char *type, char *flag, char *find, char *repl, char *mask)
static bool	parse_affentry (char *str, char *mask, char *find, char *repl)
static void	setCompoundAffixFlagValue (IspellDict *Conf, CompoundAffixFlag *entry, char *s, uint32 val)
static void	addCompoundAffixFlagValue (IspellDict *Conf, char *s, uint32 val)
static int	getCompoundAffixFlagValue (IspellDict *Conf, const char *s)
static const char *	getAffixFlagSet (IspellDict *Conf, char *s)
static void	NIImportOOAffixes (IspellDict *Conf, const char *filename)
void	NIImportAffixes (IspellDict *Conf, const char *filename)
static int	MergeAffix (IspellDict *Conf, int a1, int a2)
static uint32	makeCompoundFlags (IspellDict *Conf, int affix)
static SPNode *	mkSPNode (IspellDict *Conf, int low, int high, int level)
void	NISortDictionary (IspellDict *Conf)
static AffixNode *	mkANode (IspellDict *Conf, int low, int high, int level, int type)
static void	mkVoidAffix (IspellDict *Conf, bool issuffix, int startsuffix)
static bool	isAffixInUse (IspellDict *Conf, const char *affixflag)
void	NISortAffixes (IspellDict *Conf)
static AffixNodeData *	FindAffixes (AffixNode *node, const char *word, int wrdlen, int *level, int type)
static char *	CheckAffix (const char *word, size_t len, AFFIX *Affix, int flagflags, char *newword, int *baselen)
static int	addToResult (char **forms, char **cur, char *word)
static char **	NormalizeSubWord (IspellDict *Conf, const char *word, int flag)
static int	CheckCompoundAffixes (CMPDAffix **ptr, const char *word, int len, bool CheckInPlace)
static SplitVar *	CopyVar (SplitVar *s, int makedup)
static void	AddStem (SplitVar *v, char *word)
static SplitVar *	SplitToVariants (IspellDict *Conf, SPNode *snode, SplitVar *orig, const char *word, int wordlen, int startpos, int minpos)
static void	addNorm (TSLexeme **lres, TSLexeme **lcur, char *word, int flags, uint16 NVariant)
TSLexeme *	NINormalizeWord (IspellDict *Conf, const char *word)

Variables
static const char *	VoidString = ""

Macro Definition Documentation

COMPACT_ALLOC_CHUNK

#define COMPACT_ALLOC_CHUNK   8192 /* amount to get from palloc at once */

Definition at line 127 of file spell.c.

COMPACT_MAX_REQ

#define COMPACT_MAX_REQ   1024 /* must be < COMPACT_ALLOC_CHUNK */

Definition at line 128 of file spell.c.

cpalloc

#define cpalloc

(

size

)

compact_palloc0(Conf, size)

Definition at line 159 of file spell.c.

cpalloc0

#define cpalloc0

(

size

)

compact_palloc0(Conf, size)

Definition at line 160 of file spell.c.

GETCHAR

#define GETCHAR	(		A,
			N,
			T
	)		GETWCHAR( (A)->repl, (A)->replen, N, T )

Definition at line 193 of file spell.c.

GETWCHAR

#define GETWCHAR	(		W,
			L,
			N,
			T
	)		( ((const uint8*)(W))[ ((T)==FF_PREFIX) ? (N) : ( (L) - 1 - (N) ) ] )

Definition at line 192 of file spell.c.

MAX_NORM

#define MAX_NORM   1024

Definition at line 188 of file spell.c.

MAXNORMLEN

#define MAXNORMLEN   256

Definition at line 189 of file spell.c.

PAE_INFIND

#define PAE_INFIND   3

Definition at line 776 of file spell.c.

PAE_INMASK

#define PAE_INMASK   1

Definition at line 774 of file spell.c.

PAE_INREPL

#define PAE_INREPL   5

Definition at line 778 of file spell.c.

PAE_WAIT_FIND

#define PAE_WAIT_FIND   2

Definition at line 775 of file spell.c.

PAE_WAIT_FLAG

#define PAE_WAIT_FLAG   7

Definition at line 780 of file spell.c.

PAE_WAIT_MASK

#define PAE_WAIT_MASK   0

Definition at line 773 of file spell.c.

PAE_WAIT_REPL

#define PAE_WAIT_REPL   4

Definition at line 777 of file spell.c.

PAE_WAIT_TYPE

#define PAE_WAIT_TYPE   6

Definition at line 779 of file spell.c.

STRNCMP

#define STRNCMP	(		s,
			p
	)		strncmp( (s), (p), strlen(p) )

Definition at line 191 of file spell.c.

tmpalloc

#define tmpalloc

(

sz

)

MemoryContextAlloc(Conf->buildCxt, (sz))

Definition at line 80 of file spell.c.

tmpalloc0

#define tmpalloc0

(

sz

)

MemoryContextAllocZero(Conf->buildCxt, (sz))

Definition at line 81 of file spell.c.

Typedef Documentation

SplitVar

typedef struct SplitVar SplitVar

Function Documentation

addCompoundAffixFlagValue()

static void addCompoundAffixFlagValue ( IspellDict *  Conf, char *  s, uint32  val  )

static

Definition at line 1069 of file spell.c.

{
    CompoundAffixFlag *newValue;
    char        sbuf[BUFSIZ];
    char       *sflag;
    int         clen;
 
    while (*s && isspace((unsigned char) *s))
        s += pg_mblen(s);
 
    if (!*s)
        ereport(ERROR,
                (errcode(ERRCODE_CONFIG_FILE_ERROR),
                 errmsg("syntax error")));
 
    /* Get flag without \n */
    sflag = sbuf;
    while (*s && !isspace((unsigned char) *s) && *s != '\n')
    {
        clen = pg_mblen(s);
        COPYCHAR(sflag, s);
        sflag += clen;
        s += clen;
    }
    *sflag = '\0';
 
    /* Resize array or allocate memory for array CompoundAffixFlag */
    if (Conf->nCompoundAffixFlag >= Conf->mCompoundAffixFlag)
    {
        if (Conf->mCompoundAffixFlag)
        {
            Conf->mCompoundAffixFlag *= 2;
            Conf->CompoundAffixFlags = (CompoundAffixFlag *)
                repalloc(Conf->CompoundAffixFlags,
                         Conf->mCompoundAffixFlag * sizeof(CompoundAffixFlag));
        }
        else
        {
            Conf->mCompoundAffixFlag = 10;
            Conf->CompoundAffixFlags = (CompoundAffixFlag *)
                tmpalloc(Conf->mCompoundAffixFlag * sizeof(CompoundAffixFlag));
        }
    }
 
    newValue = Conf->CompoundAffixFlags + Conf->nCompoundAffixFlag;
 
    setCompoundAffixFlagValue(Conf, newValue, sbuf, val);
 
    Conf->usecompound = true;
    Conf->nCompoundAffixFlag++;
}

References COPYCHAR, ereport, errcode(), errmsg(), ERROR, fb(), pg_mblen(), repalloc(), setCompoundAffixFlagValue(), tmpalloc, and val.

Referenced by NIImportAffixes(), and NIImportOOAffixes().

addNorm()

static void addNorm ( TSLexeme **  lres, TSLexeme **  lcur, char *  word, int  flags, uint16  NVariant  )

static

Definition at line 2529 of file spell.c.

{
    if (*lres == NULL)
        *lcur = *lres = palloc_array(TSLexeme, MAX_NORM);
 
    if (*lcur - *lres < MAX_NORM - 1)
    {
        (*lcur)->lexeme = word;
        (*lcur)->flags = flags;
        (*lcur)->nvariant = NVariant;
        (*lcur)++;
        (*lcur)->lexeme = NULL;
    }
}

References fb(), MAX_NORM, palloc_array, and word().

Referenced by NINormalizeWord().

AddStem()

static void AddStem ( SplitVar *  v, char *  word  )

static

Definition at line 2366 of file spell.c.

{
    if (v->nstem >= v->lenstem)
    {
        v->lenstem *= 2;
        v->stem = (char **) repalloc(v->stem, sizeof(char *) * v->lenstem);
    }
 
    v->stem[v->nstem] = word;
    v->nstem++;
}

References SplitVar::lenstem, SplitVar::nstem, repalloc(), SplitVar::stem, and word().

Referenced by SplitToVariants().

addToResult()

static int addToResult ( char **  forms, char **  cur, char *  word  )

static

Definition at line 2166 of file spell.c.

{
    if (cur - forms >= MAX_NORM - 1)
        return 0;
    if (forms == cur || strcmp(word, *(cur - 1)) != 0)
    {
        *cur = pstrdup(word);
        *(cur + 1) = NULL;
        return 1;
    }
 
    return 0;
}

References cur, fb(), MAX_NORM, pstrdup(), and word().

Referenced by NormalizeSubWord().

CheckAffix()

static char * CheckAffix ( const char *  word, size_t  len, AFFIX *  Affix, int  flagflags, char *  newword, int *  baselen  )

static

Definition at line 2076 of file spell.c.

{
    /*
     * Check compound allow flags
     */
 
    if (flagflags == 0)
    {
        if (Affix->flagflags & FF_COMPOUNDONLY)
            return NULL;
    }
    else if (flagflags & FF_COMPOUNDBEGIN)
    {
        if (Affix->flagflags & FF_COMPOUNDFORBIDFLAG)
            return NULL;
        if ((Affix->flagflags & FF_COMPOUNDBEGIN) == 0)
            if (Affix->type == FF_SUFFIX)
                return NULL;
    }
    else if (flagflags & FF_COMPOUNDMIDDLE)
    {
        if ((Affix->flagflags & FF_COMPOUNDMIDDLE) == 0 ||
            (Affix->flagflags & FF_COMPOUNDFORBIDFLAG))
            return NULL;
    }
    else if (flagflags & FF_COMPOUNDLAST)
    {
        if (Affix->flagflags & FF_COMPOUNDFORBIDFLAG)
            return NULL;
        if ((Affix->flagflags & FF_COMPOUNDLAST) == 0)
            if (Affix->type == FF_PREFIX)
                return NULL;
    }
 
    /*
     * make replace pattern of affix
     */
    if (Affix->type == FF_SUFFIX)
    {
        strcpy(newword, word);
        strcpy(newword + len - Affix->replen, Affix->find);
        if (baselen)            /* store length of non-changed part of word */
            *baselen = len - Affix->replen;
    }
    else
    {
        /*
         * if prefix is an all non-changed part's length then all word
         * contains only prefix and suffix, so out
         */
        if (baselen && *baselen + strlen(Affix->find) <= Affix->replen)
            return NULL;
        strcpy(newword, Affix->find);
        strcat(newword, word + Affix->replen);
    }
 
    /*
     * check resulting word
     */
    if (Affix->issimple)
        return newword;
    else if (Affix->isregis)
    {
        if (RS_execute(&(Affix->reg.regis), newword))
            return newword;
    }
    else
    {
        pg_wchar   *data;
        size_t      data_len;
        int         newword_len;
 
        /* Convert data string to wide characters */
        newword_len = strlen(newword);
        data = palloc_array(pg_wchar, newword_len + 1);
        data_len = pg_mb2wchar_with_len(newword, data, newword_len);
 
        if (pg_regexec(Affix->reg.pregex, data, data_len,
                       0, NULL, 0, NULL, 0) == REG_OKAY)
        {
            pfree(data);
            return newword;
        }
        pfree(data);
    }
 
    return NULL;
}

References data, fb(), FF_COMPOUNDBEGIN, FF_COMPOUNDFORBIDFLAG, FF_COMPOUNDLAST, FF_COMPOUNDMIDDLE, FF_COMPOUNDONLY, FF_PREFIX, FF_SUFFIX, aff_struct::find, aff_struct::flagflags, aff_struct::isregis, aff_struct::issimple, len, palloc_array, pfree(), pg_mb2wchar_with_len(), pg_regexec(), aff_struct::pregex, aff_struct::reg, REG_OKAY, aff_struct::regis, aff_struct::replen, RS_execute(), aff_struct::type, and word().

Referenced by NormalizeSubWord().

CheckCompoundAffixes()

static int CheckCompoundAffixes ( CMPDAffix **  ptr, const char *  word, int  len, bool  CheckInPlace  )

static

Definition at line 2299 of file spell.c.

{
    bool        issuffix;
 
    /* in case CompoundAffix is null: */
    if (*ptr == NULL)
        return -1;
 
    if (CheckInPlace)
    {
        while ((*ptr)->affix)
        {
            if (len > (*ptr)->len && strncmp((*ptr)->affix, word, (*ptr)->len) == 0)
            {
                len = (*ptr)->len;
                issuffix = (*ptr)->issuffix;
                (*ptr)++;
                return (issuffix) ? len : 0;
            }
            (*ptr)++;
        }
    }
    else
    {
        const char *affbegin;
 
        while ((*ptr)->affix)
        {
            if (len > (*ptr)->len && (affbegin = strstr(word, (*ptr)->affix)) != NULL)
            {
                len = (*ptr)->len + (affbegin - word);
                issuffix = (*ptr)->issuffix;
                (*ptr)++;
                return (issuffix) ? len : 0;
            }
            (*ptr)++;
        }
    }
    return -1;
}

References fb(), len, and word().

Referenced by SplitToVariants().

cmpaffix()

static int cmpaffix ( const void *  s1, const void *  s2  )

static

Definition at line 312 of file spell.c.

{
    const AFFIX *a1 = (const AFFIX *) s1;
    const AFFIX *a2 = (const AFFIX *) s2;
 
    if (a1->type < a2->type)
        return -1;
    if (a1->type > a2->type)
        return 1;
    if (a1->type == FF_PREFIX)
        return strcmp(a1->repl, a2->repl);
    else
        return strbcmp((const unsigned char *) a1->repl,
                       (const unsigned char *) a2->repl);
}

References a1, a2, fb(), FF_PREFIX, s1, s2, and strbcmp().

Referenced by NISortAffixes().

cmpcmdflag()

static int cmpcmdflag ( const void *  f1, const void *  f2  )

static

Definition at line 211 of file spell.c.

{
    const CompoundAffixFlag *fv1 = f1;
    const CompoundAffixFlag *fv2 = f2;
 
    Assert(fv1->flagMode == fv2->flagMode);
 
    if (fv1->flagMode == FM_NUM)
    {
        if (fv1->flag.i == fv2->flag.i)
            return 0;
 
        return (fv1->flag.i > fv2->flag.i) ? 1 : -1;
    }
 
    return strcmp(fv1->flag.s, fv2->flag.s);
}

References Assert, f1, f2, fb(), and FM_NUM.

Referenced by getCompoundAffixFlagValue(), and NIImportOOAffixes().

cmpspell()

static int cmpspell ( const void *  s1, const void *  s2  )

static

Definition at line 198 of file spell.c.

{
    return strcmp((*(SPELL *const *) s1)->word, (*(SPELL *const *) s2)->word);
}

References fb(), s1, s2, and word().

Referenced by NISortDictionary().

cmpspellaffix()

static int cmpspellaffix ( const void *  s1, const void *  s2  )

static

Definition at line 204 of file spell.c.

{
    return strcmp((*(SPELL *const *) s1)->p.flag,
                  (*(SPELL *const *) s2)->p.flag);
}

References fb(), s1, and s2.

Referenced by NISortDictionary().

compact_palloc0()

static void * compact_palloc0 ( IspellDict *  Conf, size_t  size  )

static

Definition at line 131 of file spell.c.

{
    void       *result;
 
    /* Should only be called during init */
    Assert(Conf->buildCxt != NULL);
 
    /* No point in this for large chunks */
    if (size > COMPACT_MAX_REQ)
        return palloc0(size);
 
    /* Keep everything maxaligned */
    size = MAXALIGN(size);
 
    /* Need more space? */
    if (size > Conf->avail)
    {
        Conf->firstfree = palloc0(COMPACT_ALLOC_CHUNK);
        Conf->avail = COMPACT_ALLOC_CHUNK;
    }
 
    result = Conf->firstfree;
    Conf->firstfree += size;
    Conf->avail -= size;
 
    return result;
}

References Assert, COMPACT_ALLOC_CHUNK, COMPACT_MAX_REQ, fb(), MAXALIGN, and palloc0().

CopyVar()

static SplitVar * CopyVar ( SplitVar *  s, int  makedup  )

static

Definition at line 2341 of file spell.c.

{
    SplitVar   *v = palloc_object(SplitVar);
 
    v->next = NULL;
    if (s)
    {
        int         i;
 
        v->lenstem = s->lenstem;
        v->stem = palloc_array(char *, v->lenstem);
        v->nstem = s->nstem;
        for (i = 0; i < s->nstem; i++)
            v->stem[i] = (makedup) ? pstrdup(s->stem[i]) : s->stem[i];
    }
    else
    {
        v->lenstem = 16;
        v->stem = palloc_array(char *, v->lenstem);
        v->nstem = 0;
    }
    return v;
}

References fb(), i, SplitVar::lenstem, SplitVar::next, SplitVar::nstem, palloc_array, palloc_object, pstrdup(), and SplitVar::stem.

Referenced by SplitToVariants().

cpstrdup()

static char * cpstrdup ( IspellDict *  Conf, const char *  str  )

static

Definition at line 163 of file spell.c.

{
    char       *res = cpalloc(strlen(str) + 1);
 
    strcpy(res, str);
    return res;
}

References cpalloc, fb(), and str.

Referenced by NIAddAffix(), NIAddSpell(), NIImportOOAffixes(), NISortDictionary(), and setCompoundAffixFlagValue().

FindAffixes()

static AffixNodeData * FindAffixes ( AffixNode *  node, const char *  word, int  wrdlen, int *  level, int  type  )

static

Definition at line 2033 of file spell.c.

{
    AffixNodeData *StopLow,
               *StopHigh,
               *StopMiddle;
    uint8 symbol;
 
    if (node->isvoid)
    {                           /* search void affixes */
        if (node->data->naff)
            return node->data;
        node = node->data->node;
    }
 
    while (node && *level < wrdlen)
    {
        StopLow = node->data;
        StopHigh = node->data + node->length;
        while (StopLow < StopHigh)
        {
            StopMiddle = StopLow + ((StopHigh - StopLow) >> 1);
            symbol = GETWCHAR(word, wrdlen, *level, type);
 
            if (StopMiddle->val == symbol)
            {
                (*level)++;
                if (StopMiddle->naff)
                    return StopMiddle;
                node = StopMiddle->node;
                break;
            }
            else if (StopMiddle->val < symbol)
                StopLow = StopMiddle + 1;
            else
                StopHigh = StopMiddle;
        }
        if (StopLow >= StopHigh)
            break;
    }
    return NULL;
}

References AffixNode::data, fb(), GETWCHAR, AffixNode::isvoid, AffixNode::length, AffixNodeData::naff, AffixNodeData::node, type, and word().

Referenced by NormalizeSubWord().

findchar()

static char * findchar ( char *  str, int  c  )

static

Definition at line 230 of file spell.c.

{
    while (*str)
    {
        if (t_iseq(str, c))
            return str;
        str += pg_mblen(str);
    }
 
    return NULL;
}

References fb(), pg_mblen(), str, and t_iseq.

Referenced by NIImportDictionary().

findchar2()

static char * findchar2 ( char *  str, int  c1, int  c2  )

static

Definition at line 243 of file spell.c.

{
    while (*str)
    {
        if (t_iseq(str, c1) || t_iseq(str, c2))
            return str;
        str += pg_mblen(str);
    }
 
    return NULL;
}

References fb(), pg_mblen(), str, and t_iseq.

Referenced by NIImportAffixes().

FindWord()

static int FindWord ( IspellDict *  Conf, const char *  word, const char *  affixflag, int  flag  )

static

Definition at line 604 of file spell.c.

{
    SPNode     *node = Conf->Dictionary;
    SPNodeData *StopLow,
               *StopHigh,
               *StopMiddle;
    const uint8 *ptr = (const uint8 *) word;
 
    flag &= FF_COMPOUNDFLAGMASK;
 
    while (node && *ptr)
    {
        StopLow = node->data;
        StopHigh = node->data + node->length;
        while (StopLow < StopHigh)
        {
            StopMiddle = StopLow + ((StopHigh - StopLow) >> 1);
            if (StopMiddle->val == *ptr)
            {
                if (*(ptr + 1) == '\0' && StopMiddle->isword)
                {
                    if (flag == 0)
                    {
                        /*
                         * The word can be formed only with another word. And
                         * in the flag parameter there is not a sign that we
                         * search compound words.
                         */
                        if (StopMiddle->compoundflag & FF_COMPOUNDONLY)
                            return 0;
                    }
                    else if ((flag & StopMiddle->compoundflag) == 0)
                        return 0;
 
                    /*
                     * Check if this affix rule is presented in the affix set
                     * with index StopMiddle->affix.
                     */
                    if (IsAffixFlagInUse(Conf, StopMiddle->affix, affixflag))
                        return 1;
                }
                node = StopMiddle->node;
                ptr++;
                break;
            }
            else if (StopMiddle->val < *ptr)
                StopLow = StopMiddle + 1;
            else
                StopHigh = StopMiddle;
        }
        if (StopLow >= StopHigh)
            break;
    }
    return 0;
}

References SPNode::data, fb(), FF_COMPOUNDFLAGMASK, FF_COMPOUNDONLY, flag(), IsAffixFlagInUse(), SPNode::length, and word().

Referenced by NormalizeSubWord().

get_nextfield()

static bool get_nextfield ( char **  str, char *  next  )

static

Definition at line 793 of file spell.c.

{
    int         state = PAE_WAIT_MASK;
    int         avail = BUFSIZ;
 
    while (**str)
    {
        if (state == PAE_WAIT_MASK)
        {
            if (t_iseq(*str, '#'))
                return false;
            else if (!isspace((unsigned char) **str))
            {
                int         clen = pg_mblen(*str);
 
                if (clen < avail)
                {
                    COPYCHAR(next, *str);
                    next += clen;
                    avail -= clen;
                }
                state = PAE_INMASK;
            }
        }
        else                    /* state == PAE_INMASK */
        {
            if (isspace((unsigned char) **str))
            {
                *next = '\0';
                return true;
            }
            else
            {
                int         clen = pg_mblen(*str);
 
                if (clen < avail)
                {
                    COPYCHAR(next, *str);
                    next += clen;
                    avail -= clen;
                }
            }
        }
        *str += pg_mblen(*str);
    }
 
    *next = '\0';
 
    return (state == PAE_INMASK);   /* OK if we got a nonempty field */
}

References COPYCHAR, fb(), next, PAE_INMASK, PAE_WAIT_MASK, pg_mblen(), str, and t_iseq.

Referenced by parse_ooaffentry().

getAffixFlagSet()

static const char * getAffixFlagSet ( IspellDict *  Conf, char *  s  )

static

Definition at line 1162 of file spell.c.

{
    if (Conf->useFlagAliases && *s != '\0')
    {
        int         curaffix;
        char       *end;
 
        errno = 0;
        curaffix = strtol(s, &end, 10);
        if (s == end || errno == ERANGE)
            ereport(ERROR,
                    (errcode(ERRCODE_CONFIG_FILE_ERROR),
                     errmsg("invalid affix alias \"%s\"", s)));
 
        if (curaffix > 0 && curaffix < Conf->nAffixData)
 
            /*
             * Do not subtract 1 from curaffix because empty string was added
             * in NIImportOOAffixes
             */
            return Conf->AffixData[curaffix];
        else if (curaffix > Conf->nAffixData)
            ereport(ERROR,
                    (errcode(ERRCODE_CONFIG_FILE_ERROR),
                     errmsg("invalid affix alias \"%s\"", s)));
        return VoidString;
    }
    else
        return s;
}

References ereport, errcode(), errmsg(), ERROR, fb(), and VoidString.

Referenced by NIImportOOAffixes().

getCompoundAffixFlagValue()

static int getCompoundAffixFlagValue ( IspellDict *  Conf, const char *  s  )

static

Definition at line 1126 of file spell.c.

{
    uint32      flag = 0;
    CompoundAffixFlag *found,
                key;
    char        sflag[BUFSIZ];
    const char *flagcur;
 
    if (Conf->nCompoundAffixFlag == 0)
        return 0;
 
    flagcur = s;
    while (*flagcur)
    {
        getNextFlagFromString(Conf, &flagcur, sflag);
        setCompoundAffixFlagValue(Conf, &key, sflag, 0);
 
        found = (CompoundAffixFlag *)
            bsearch(&key, Conf->CompoundAffixFlags,
                    Conf->nCompoundAffixFlag, sizeof(CompoundAffixFlag),
                    cmpcmdflag);
        if (found != NULL)
            flag |= found->value;
    }
 
    return flag;
}

References cmpcmdflag(), fb(), flag(), getNextFlagFromString(), setCompoundAffixFlagValue(), and CompoundAffixFlag::value.

Referenced by makeCompoundFlags(), and NIImportOOAffixes().

getNextFlagFromString()

static void getNextFlagFromString ( IspellDict *  Conf, const char **  sflagset, char *  sflag  )

static

Definition at line 350 of file spell.c.

{
    int32       s;
    char       *next;
    const char *sbuf = *sflagset;
    int         maxstep;
    bool        stop = false;
    bool        met_comma = false;
 
    maxstep = (Conf->flagMode == FM_LONG) ? 2 : 1;
 
    while (**sflagset)
    {
        switch (Conf->flagMode)
        {
            case FM_LONG:
            case FM_CHAR:
                COPYCHAR(sflag, *sflagset);
                sflag += pg_mblen(*sflagset);
 
                /* Go to start of the next flag */
                *sflagset += pg_mblen(*sflagset);
 
                /* Check if we get all characters of flag */
                maxstep--;
                stop = (maxstep == 0);
                break;
            case FM_NUM:
                errno = 0;
                s = strtol(*sflagset, &next, 10);
                if (*sflagset == next || errno == ERANGE)
                    ereport(ERROR,
                            (errcode(ERRCODE_CONFIG_FILE_ERROR),
                             errmsg("invalid affix flag \"%s\"", *sflagset)));
                if (s < 0 || s > FLAGNUM_MAXSIZE)
                    ereport(ERROR,
                            (errcode(ERRCODE_CONFIG_FILE_ERROR),
                             errmsg("affix flag \"%s\" is out of range",
                                    *sflagset)));
                sflag += sprintf(sflag, "%0d", s);
 
                /* Go to start of the next flag */
                *sflagset = next;
                while (**sflagset)
                {
                    if (isdigit((unsigned char) **sflagset))
                    {
                        if (!met_comma)
                            ereport(ERROR,
                                    (errcode(ERRCODE_CONFIG_FILE_ERROR),
                                     errmsg("invalid affix flag \"%s\"",
                                            *sflagset)));
                        break;
                    }
                    else if (t_iseq(*sflagset, ','))
                    {
                        if (met_comma)
                            ereport(ERROR,
                                    (errcode(ERRCODE_CONFIG_FILE_ERROR),
                                     errmsg("invalid affix flag \"%s\"",
                                            *sflagset)));
                        met_comma = true;
                    }
                    else if (!isspace((unsigned char) **sflagset))
                    {
                        ereport(ERROR,
                                (errcode(ERRCODE_CONFIG_FILE_ERROR),
                                 errmsg("invalid character in affix flag \"%s\"",
                                        *sflagset)));
                    }
 
                    *sflagset += pg_mblen(*sflagset);
                }
                stop = true;
                break;
            default:
                elog(ERROR, "unrecognized type of Conf->flagMode: %d",
                     Conf->flagMode);
        }
 
        if (stop)
            break;
    }
 
    if (Conf->flagMode == FM_LONG && maxstep > 0)
        ereport(ERROR,
                (errcode(ERRCODE_CONFIG_FILE_ERROR),
                 errmsg("invalid affix flag \"%s\" with \"long\" flag value",
                        sbuf)));
 
    *sflag = '\0';
}

References COPYCHAR, elog, ereport, errcode(), errmsg(), ERROR, fb(), FLAGNUM_MAXSIZE, FM_CHAR, FM_LONG, FM_NUM, next, pg_mblen(), sprintf, and t_iseq.

Referenced by getCompoundAffixFlagValue(), and IsAffixFlagInUse().

IsAffixFlagInUse()

static bool IsAffixFlagInUse ( IspellDict *  Conf, int  affix, const char *  affixflag  )

static

Definition at line 456 of file spell.c.

{
    const char *flagcur;
    char        flag[BUFSIZ];
 
    if (*affixflag == 0)
        return true;
 
    Assert(affix < Conf->nAffixData);
 
    flagcur = Conf->AffixData[affix];
 
    while (*flagcur)
    {
        getNextFlagFromString(Conf, &flagcur, flag);
        /* Compare first affix flag in flagcur with affixflag */
        if (strcmp(flag, affixflag) == 0)
            return true;
    }
 
    /* Could not find affixflag */
    return false;
}

References Assert, fb(), flag(), and getNextFlagFromString().

Referenced by FindWord(), and isAffixInUse().

isAffixInUse()

static bool isAffixInUse ( IspellDict *  Conf, const char *  affixflag  )

static

Definition at line 1966 of file spell.c.

{
    int         i;
 
    for (i = 0; i < Conf->nAffixData; i++)
        if (IsAffixFlagInUse(Conf, i, affixflag))
            return true;
 
    return false;
}

References fb(), i, and IsAffixFlagInUse().

Referenced by NISortAffixes().

lowerstr_ctx()

static char * lowerstr_ctx ( IspellDict *  Conf, const char *  src  )

static

Definition at line 176 of file spell.c.

{
    MemoryContext saveCtx;
    char       *dst;
 
    saveCtx = MemoryContextSwitchTo(Conf->buildCxt);
    dst = str_tolower(src, strlen(src), DEFAULT_COLLATION_OID);
    MemoryContextSwitchTo(saveCtx);
 
    return dst;
}

References fb(), MemoryContextSwitchTo(), and str_tolower().

Referenced by NIImportDictionary(), and NIImportOOAffixes().

makeCompoundFlags()

static uint32 makeCompoundFlags ( IspellDict *  Conf, int  affix  )

static

Definition at line 1627 of file spell.c.

{
    Assert(affix < Conf->nAffixData);
 
    return (getCompoundAffixFlagValue(Conf, Conf->AffixData[affix]) &
            FF_COMPOUNDFLAGMASK);
}

References Assert, fb(), FF_COMPOUNDFLAGMASK, and getCompoundAffixFlagValue().

Referenced by mkSPNode().

MergeAffix()

static int MergeAffix ( IspellDict *  Conf, int  a1, int  a2  )

static

Definition at line 1576 of file spell.c.

{
    const char **ptr;
 
    Assert(a1 < Conf->nAffixData && a2 < Conf->nAffixData);
 
    /* Do not merge affix flags if one of affix flags is empty */
    if (*Conf->AffixData[a1] == '\0')
        return a2;
    else if (*Conf->AffixData[a2] == '\0')
        return a1;
 
    /* Double the size of AffixData if there's not enough space */
    if (Conf->nAffixData + 1 >= Conf->lenAffixData)
    {
        Conf->lenAffixData *= 2;
        Conf->AffixData = (const char **) repalloc(Conf->AffixData,
                                                   sizeof(char *) * Conf->lenAffixData);
    }
 
    ptr = Conf->AffixData + Conf->nAffixData;
    if (Conf->flagMode == FM_NUM)
    {
        char       *p = cpalloc(strlen(Conf->AffixData[a1]) +
                                strlen(Conf->AffixData[a2]) +
                                1 /* comma */ + 1 /* \0 */ );
 
        sprintf(p, "%s,%s", Conf->AffixData[a1], Conf->AffixData[a2]);
        *ptr = p;
    }
    else
    {
        char       *p = cpalloc(strlen(Conf->AffixData[a1]) +
                                strlen(Conf->AffixData[a2]) +
                                1 /* \0 */ );
 
        sprintf(p, "%s%s", Conf->AffixData[a1], Conf->AffixData[a2]);
        *ptr = p;
    }
    ptr++;
    *ptr = NULL;
    Conf->nAffixData++;
 
    return Conf->nAffixData - 1;
}

References a1, a2, Assert, cpalloc, fb(), FM_NUM, repalloc(), and sprintf.

Referenced by mkSPNode().

mkANode()

static AffixNode * mkANode ( IspellDict *  Conf, int  low, int  high, int  level, int  type  )

static

Definition at line 1835 of file spell.c.

{
    int         i;
    int         nchar = 0;
    uint8       lastchar = '\0';
    AffixNode  *rs;
    AffixNodeData *data;
    int         lownew = low;
    int         naff;
    AFFIX     **aff;
 
    for (i = low; i < high; i++)
        if (Conf->Affix[i].replen > level && lastchar != GETCHAR(Conf->Affix + i, level, type))
        {
            nchar++;
            lastchar = GETCHAR(Conf->Affix + i, level, type);
        }
 
    if (!nchar)
        return NULL;
 
    aff = (AFFIX **) tmpalloc(sizeof(AFFIX *) * (high - low + 1));
    naff = 0;
 
    rs = (AffixNode *) cpalloc0(ANHRDSZ + nchar * sizeof(AffixNodeData));
    rs->length = nchar;
    data = rs->data;
 
    lastchar = '\0';
    for (i = low; i < high; i++)
        if (Conf->Affix[i].replen > level)
        {
            if (lastchar != GETCHAR(Conf->Affix + i, level, type))
            {
                if (lastchar)
                {
                    /* Next level of the prefix tree */
                    data->node = mkANode(Conf, lownew, i, level + 1, type);
                    if (naff)
                    {
                        data->naff = naff;
                        data->aff = (AFFIX **) cpalloc(sizeof(AFFIX *) * naff);
                        memcpy(data->aff, aff, sizeof(AFFIX *) * naff);
                        naff = 0;
                    }
                    data++;
                    lownew = i;
                }
                lastchar = GETCHAR(Conf->Affix + i, level, type);
            }
            data->val = GETCHAR(Conf->Affix + i, level, type);
            if (Conf->Affix[i].replen == level + 1)
            {                   /* affix stopped */
                aff[naff++] = Conf->Affix + i;
            }
        }
 
    /* Next level of the prefix tree */
    data->node = mkANode(Conf, lownew, high, level + 1, type);
    if (naff)
    {
        data->naff = naff;
        data->aff = (AFFIX **) cpalloc(sizeof(AFFIX *) * naff);
        memcpy(data->aff, aff, sizeof(AFFIX *) * naff);
        naff = 0;
    }
 
    pfree(aff);
 
    return rs;
}

References ANHRDSZ, cpalloc, cpalloc0, AffixNode::data, data, fb(), GETCHAR, i, AffixNode::length, mkANode(), AffixNodeData::node, pfree(), tmpalloc, and type.

Referenced by mkANode(), and NISortAffixes().

mkSPNode()

static SPNode * mkSPNode ( IspellDict *  Conf, int  low, int  high, int  level  )

static

Definition at line 1644 of file spell.c.

{
    int         i;
    int         nchar = 0;
    char        lastchar = '\0';
    SPNode     *rs;
    SPNodeData *data;
    int         lownew = low;
 
    for (i = low; i < high; i++)
        if (Conf->Spell[i]->p.d.len > level && lastchar != Conf->Spell[i]->word[level])
        {
            nchar++;
            lastchar = Conf->Spell[i]->word[level];
        }
 
    if (!nchar)
        return NULL;
 
    rs = (SPNode *) cpalloc0(SPNHDRSZ + nchar * sizeof(SPNodeData));
    rs->length = nchar;
    data = rs->data;
 
    lastchar = '\0';
    for (i = low; i < high; i++)
        if (Conf->Spell[i]->p.d.len > level)
        {
            if (lastchar != Conf->Spell[i]->word[level])
            {
                if (lastchar)
                {
                    /* Next level of the prefix tree */
                    data->node = mkSPNode(Conf, lownew, i, level + 1);
                    lownew = i;
                    data++;
                }
                lastchar = Conf->Spell[i]->word[level];
            }
            data->val = ((uint8 *) (Conf->Spell[i]->word))[level];
            if (Conf->Spell[i]->p.d.len == level + 1)
            {
                bool        clearCompoundOnly = false;
 
                if (data->isword && data->affix != Conf->Spell[i]->p.d.affix)
                {
                    /*
                     * MergeAffix called a few times. If one of word is
                     * allowed to be in compound word and another isn't, then
                     * clear FF_COMPOUNDONLY flag.
                     */
 
                    clearCompoundOnly = (FF_COMPOUNDONLY & data->compoundflag
                                         & makeCompoundFlags(Conf, Conf->Spell[i]->p.d.affix))
                        ? false : true;
                    data->affix = MergeAffix(Conf, data->affix, Conf->Spell[i]->p.d.affix);
                }
                else
                    data->affix = Conf->Spell[i]->p.d.affix;
                data->isword = 1;
 
                data->compoundflag = makeCompoundFlags(Conf, data->affix);
 
                if ((data->compoundflag & FF_COMPOUNDONLY) &&
                    (data->compoundflag & FF_COMPOUNDFLAG) == 0)
                    data->compoundflag |= FF_COMPOUNDFLAG;
 
                if (clearCompoundOnly)
                    data->compoundflag &= ~FF_COMPOUNDONLY;
            }
        }
 
    /* Next level of the prefix tree */
    data->node = mkSPNode(Conf, lownew, high, level + 1);
 
    return rs;
}

References cpalloc0, SPNode::data, data, fb(), FF_COMPOUNDFLAG, FF_COMPOUNDONLY, i, SPNode::length, makeCompoundFlags(), MergeAffix(), mkSPNode(), SPNodeData::node, and SPNHDRSZ.

Referenced by mkSPNode(), and NISortDictionary().

mkVoidAffix()

static void mkVoidAffix ( IspellDict *  Conf, bool  issuffix, int  startsuffix  )

static

Definition at line 1912 of file spell.c.

{
    int         i,
                cnt = 0;
    int         start = (issuffix) ? startsuffix : 0;
    int         end = (issuffix) ? Conf->naffixes : startsuffix;
    AffixNode  *Affix = (AffixNode *) palloc0(ANHRDSZ + sizeof(AffixNodeData));
 
    Affix->length = 1;
    Affix->isvoid = 1;
 
    if (issuffix)
    {
        Affix->data->node = Conf->Suffix;
        Conf->Suffix = Affix;
    }
    else
    {
        Affix->data->node = Conf->Prefix;
        Conf->Prefix = Affix;
    }
 
    /* Count affixes with empty replace string */
    for (i = start; i < end; i++)
        if (Conf->Affix[i].replen == 0)
            cnt++;
 
    /* There is not affixes with empty replace string */
    if (cnt == 0)
        return;
 
    Affix->data->aff = (AFFIX **) cpalloc(sizeof(AFFIX *) * cnt);
    Affix->data->naff = (uint32) cnt;
 
    cnt = 0;
    for (i = start; i < end; i++)
        if (Conf->Affix[i].replen == 0)
        {
            Affix->data->aff[cnt] = Conf->Affix + i;
            cnt++;
        }
}

References AffixNodeData::aff, ANHRDSZ, cpalloc, AffixNode::data, fb(), i, AffixNode::isvoid, AffixNode::length, AffixNodeData::naff, AffixNodeData::node, palloc0(), and start.

Referenced by NISortAffixes().

NIAddAffix()

static void NIAddAffix ( IspellDict *  Conf, const char *  flag, char  flagflags, const char *  mask, const char *  find, const char *  repl, int  type  )

static

Definition at line 679 of file spell.c.

{
    AFFIX      *Affix;
 
    if (Conf->naffixes >= Conf->maffixes)
    {
        if (Conf->maffixes)
        {
            Conf->maffixes *= 2;
            Conf->Affix = (AFFIX *) repalloc(Conf->Affix, Conf->maffixes * sizeof(AFFIX));
        }
        else
        {
            Conf->maffixes = 16;
            Conf->Affix = palloc_array(AFFIX, Conf->maffixes);
        }
    }
 
    Affix = Conf->Affix + Conf->naffixes;
 
    /* This affix rule can be applied for words with any ending */
    if (strcmp(mask, ".") == 0 || *mask == '\0')
    {
        Affix->issimple = 1;
        Affix->isregis = 0;
    }
    /* This affix rule will use regis to search word ending */
    else if (RS_isRegis(mask))
    {
        Affix->issimple = 0;
        Affix->isregis = 1;
        RS_compile(&(Affix->reg.regis), (type == FF_SUFFIX),
                   *mask ? mask : VoidString);
    }
    /* This affix rule will use regex_t to search word ending */
    else
    {
        int         masklen;
        int         wmasklen;
        int         err;
        pg_wchar   *wmask;
        char       *tmask;
 
        Affix->issimple = 0;
        Affix->isregis = 0;
        tmask = (char *) tmpalloc(strlen(mask) + 3);
        if (type == FF_SUFFIX)
            sprintf(tmask, "%s$", mask);
        else
            sprintf(tmask, "^%s", mask);
 
        masklen = strlen(tmask);
        wmask = (pg_wchar *) tmpalloc((masklen + 1) * sizeof(pg_wchar));
        wmasklen = pg_mb2wchar_with_len(tmask, wmask, masklen);
 
        /*
         * The regex and all internal state created by pg_regcomp are
         * allocated in the dictionary's memory context, and will be freed
         * automatically when it is destroyed.
         */
        Affix->reg.pregex = palloc_object(regex_t);
        err = pg_regcomp(Affix->reg.pregex, wmask, wmasklen,
                         REG_ADVANCED | REG_NOSUB,
                         DEFAULT_COLLATION_OID);
        if (err)
        {
            char        errstr[100];
 
            pg_regerror(err, Affix->reg.pregex, errstr, sizeof(errstr));
            ereport(ERROR,
                    (errcode(ERRCODE_INVALID_REGULAR_EXPRESSION),
                     errmsg("invalid regular expression: %s", errstr)));
        }
    }
 
    Affix->flagflags = flagflags;
    if ((Affix->flagflags & FF_COMPOUNDONLY) || (Affix->flagflags & FF_COMPOUNDPERMITFLAG))
    {
        if ((Affix->flagflags & FF_COMPOUNDFLAG) == 0)
            Affix->flagflags |= FF_COMPOUNDFLAG;
    }
    Affix->flag = cpstrdup(Conf, flag);
    Affix->type = type;
 
    Affix->find = (find && *find) ? cpstrdup(Conf, find) : VoidString;
    if ((Affix->replen = strlen(repl)) > 0)
        Affix->repl = cpstrdup(Conf, repl);
    else
        Affix->repl = VoidString;
    Conf->naffixes++;
}

References cpstrdup(), ereport, err(), errcode(), errmsg(), ERROR, fb(), FF_COMPOUNDFLAG, FF_COMPOUNDONLY, FF_COMPOUNDPERMITFLAG, FF_SUFFIX, find(), aff_struct::find, aff_struct::flag, flag(), aff_struct::flagflags, aff_struct::isregis, aff_struct::issimple, palloc_array, palloc_object, pg_mb2wchar_with_len(), pg_regcomp(), pg_regerror(), aff_struct::pregex, aff_struct::reg, REG_ADVANCED, REG_NOSUB, regex_t, aff_struct::regis, repalloc(), aff_struct::repl, aff_struct::replen, RS_compile(), RS_isRegis(), sprintf, tmpalloc, type, aff_struct::type, and VoidString.

Referenced by NIImportAffixes(), and NIImportOOAffixes().

NIAddSpell()

static void NIAddSpell ( IspellDict *  Conf, const char *  word, const char *  flag  )

static

Definition at line 488 of file spell.c.

{
    if (Conf->nspell >= Conf->mspell)
    {
        if (Conf->mspell)
        {
            Conf->mspell *= 2;
            Conf->Spell = (SPELL **) repalloc(Conf->Spell, Conf->mspell * sizeof(SPELL *));
        }
        else
        {
            Conf->mspell = 1024 * 20;
            Conf->Spell = (SPELL **) tmpalloc(Conf->mspell * sizeof(SPELL *));
        }
    }
    Conf->Spell[Conf->nspell] = (SPELL *) tmpalloc(SPELLHDRSZ + strlen(word) + 1);
    strcpy(Conf->Spell[Conf->nspell]->word, word);
    Conf->Spell[Conf->nspell]->p.flag = (*flag != '\0')
        ? cpstrdup(Conf, flag) : VoidString;
    Conf->nspell++;
}

References cpstrdup(), fb(), flag(), repalloc(), SPELLHDRSZ, tmpalloc, VoidString, and word().

Referenced by NIImportDictionary().

NIFinishBuild()

void NIFinishBuild

(

IspellDict *

Conf

)

Definition at line 104 of file spell.c.

{
    /* Release no-longer-needed temp memory */
    MemoryContextDelete(Conf->buildCxt);
    /* Just for cleanliness, zero the now-dangling pointers */
    Conf->buildCxt = NULL;
    Conf->Spell = NULL;
    Conf->firstfree = NULL;
    Conf->CompoundAffixFlags = NULL;
}

References fb(), and MemoryContextDelete().

Referenced by dispell_init().

NIImportAffixes()

void NIImportAffixes	(	IspellDict *	Conf,
		const char *	filename
	)

Definition at line 1429 of file spell.c.

{
    char       *pstr = NULL;
    char        flag[BUFSIZ];
    char        mask[BUFSIZ];
    char        find[BUFSIZ];
    char        repl[BUFSIZ];
    char       *s;
    bool        suffixes = false;
    bool        prefixes = false;
    char        flagflags = 0;
    tsearch_readline_state trst;
    bool        oldformat = false;
    char       *recoded = NULL;
 
    if (!tsearch_readline_begin(&trst, filename))
        ereport(ERROR,
                (errcode(ERRCODE_CONFIG_FILE_ERROR),
                 errmsg("could not open affix file \"%s\": %m",
                        filename)));
 
    Conf->usecompound = false;
    Conf->useFlagAliases = false;
    Conf->flagMode = FM_CHAR;
 
    while ((recoded = tsearch_readline(&trst)) != NULL)
    {
        pstr = str_tolower(recoded, strlen(recoded), DEFAULT_COLLATION_OID);
 
        /* Skip comments and empty lines */
        if (*pstr == '#' || *pstr == '\n')
            goto nextline;
 
        if (STRNCMP(pstr, "compoundwords") == 0)
        {
            /* Find case-insensitive L flag in non-lowercased string */
            s = findchar2(recoded, 'l', 'L');
            if (s)
            {
                while (*s && !isspace((unsigned char) *s))
                    s += pg_mblen(s);
                while (*s && isspace((unsigned char) *s))
                    s += pg_mblen(s);
 
                if (*s && pg_mblen(s) == 1)
                {
                    addCompoundAffixFlagValue(Conf, s, FF_COMPOUNDFLAG);
                    Conf->usecompound = true;
                }
                oldformat = true;
                goto nextline;
            }
        }
        if (STRNCMP(pstr, "suffixes") == 0)
        {
            suffixes = true;
            prefixes = false;
            oldformat = true;
            goto nextline;
        }
        if (STRNCMP(pstr, "prefixes") == 0)
        {
            suffixes = false;
            prefixes = true;
            oldformat = true;
            goto nextline;
        }
        if (STRNCMP(pstr, "flag") == 0)
        {
            s = recoded + 4;    /* we need non-lowercased string */
            flagflags = 0;
 
            while (*s && isspace((unsigned char) *s))
                s += pg_mblen(s);
 
            if (*s == '*')
            {
                flagflags |= FF_CROSSPRODUCT;
                s++;
            }
            else if (*s == '~')
            {
                flagflags |= FF_COMPOUNDONLY;
                s++;
            }
 
            if (*s == '\\')
                s++;
 
            /*
             * An old-format flag is a single ASCII character; we expect it to
             * be followed by EOL, whitespace, or ':'.  Otherwise this is a
             * new-format flag command.
             */
            if (*s && pg_mblen(s) == 1)
            {
                COPYCHAR(flag, s);
                flag[1] = '\0';
 
                s++;
                if (*s == '\0' || *s == '#' || *s == '\n' || *s == ':' ||
                    isspace((unsigned char) *s))
                {
                    oldformat = true;
                    goto nextline;
                }
            }
            goto isnewformat;
        }
        if (STRNCMP(recoded, "COMPOUNDFLAG") == 0 ||
            STRNCMP(recoded, "COMPOUNDMIN") == 0 ||
            STRNCMP(recoded, "PFX") == 0 ||
            STRNCMP(recoded, "SFX") == 0)
            goto isnewformat;
 
        if ((!suffixes) && (!prefixes))
            goto nextline;
 
        if (!parse_affentry(pstr, mask, find, repl))
            goto nextline;
 
        NIAddAffix(Conf, flag, flagflags, mask, find, repl, suffixes ? FF_SUFFIX : FF_PREFIX);
 
nextline:
        pfree(recoded);
        pfree(pstr);
    }
    tsearch_readline_end(&trst);
    return;
 
isnewformat:
    if (oldformat)
        ereport(ERROR,
                (errcode(ERRCODE_CONFIG_FILE_ERROR),
                 errmsg("affix file contains both old-style and new-style commands")));
    tsearch_readline_end(&trst);
 
    NIImportOOAffixes(Conf, filename);
}

References addCompoundAffixFlagValue(), COPYCHAR, ereport, errcode(), errmsg(), ERROR, fb(), FF_COMPOUNDFLAG, FF_COMPOUNDONLY, FF_CROSSPRODUCT, FF_PREFIX, FF_SUFFIX, filename, find(), findchar2(), flag(), FM_CHAR, NIAddAffix(), NIImportOOAffixes(), parse_affentry(), pfree(), pg_mblen(), prefixes(), str_tolower(), STRNCMP, tsearch_readline(), tsearch_readline_begin(), and tsearch_readline_end().

Referenced by dispell_init().

NIImportDictionary()

void NIImportDictionary	(	IspellDict *	Conf,
		const char *	filename
	)

Definition at line 519 of file spell.c.

{
    tsearch_readline_state trst;
    char       *line;
 
    if (!tsearch_readline_begin(&trst, filename))
        ereport(ERROR,
                (errcode(ERRCODE_CONFIG_FILE_ERROR),
                 errmsg("could not open dictionary file \"%s\": %m",
                        filename)));
 
    while ((line = tsearch_readline(&trst)) != NULL)
    {
        char       *s,
                   *pstr;
 
        /* Set of affix flags */
        const char *flag;
 
        /* Extract flag from the line */
        flag = NULL;
        if ((s = findchar(line, '/')))
        {
            *s++ = '\0';
            flag = s;
            while (*s)
            {
                /* we allow only single encoded flags for faster works */
                if (pg_mblen(s) == 1 && isprint((unsigned char) *s) && !isspace((unsigned char) *s))
                    s++;
                else
                {
                    *s = '\0';
                    break;
                }
            }
        }
        else
            flag = "";
 
        /* Remove trailing spaces */
        s = line;
        while (*s)
        {
            if (isspace((unsigned char) *s))
            {
                *s = '\0';
                break;
            }
            s += pg_mblen(s);
        }
        pstr = lowerstr_ctx(Conf, line);
 
        NIAddSpell(Conf, pstr, flag);
        pfree(pstr);
 
        pfree(line);
    }
    tsearch_readline_end(&trst);
}

References ereport, errcode(), errmsg(), ERROR, fb(), filename, findchar(), flag(), lowerstr_ctx(), NIAddSpell(), pfree(), pg_mblen(), tsearch_readline(), tsearch_readline_begin(), and tsearch_readline_end().

Referenced by dispell_init().

NIImportOOAffixes()

static void NIImportOOAffixes ( IspellDict *  Conf, const char *  filename  )

static

Definition at line 1200 of file spell.c.

{
    char        type[BUFSIZ],
               *ptype = NULL;
    char        sflag[BUFSIZ];
    char        mask[BUFSIZ],
               *pmask;
    char        find[BUFSIZ],
               *pfind;
    char        repl[BUFSIZ],
               *prepl;
    bool        isSuffix = false;
    int         naffix = 0,
                curaffix = 0;
    int         sflaglen = 0;
    char        flagflags = 0;
    tsearch_readline_state trst;
    char       *recoded;
 
    /* read file to find any flag */
    Conf->usecompound = false;
    Conf->useFlagAliases = false;
    Conf->flagMode = FM_CHAR;
 
    if (!tsearch_readline_begin(&trst, filename))
        ereport(ERROR,
                (errcode(ERRCODE_CONFIG_FILE_ERROR),
                 errmsg("could not open affix file \"%s\": %m",
                        filename)));
 
    while ((recoded = tsearch_readline(&trst)) != NULL)
    {
        if (*recoded == '\0' || isspace((unsigned char) *recoded) || t_iseq(recoded, '#'))
        {
            pfree(recoded);
            continue;
        }
 
        if (STRNCMP(recoded, "COMPOUNDFLAG") == 0)
            addCompoundAffixFlagValue(Conf, recoded + strlen("COMPOUNDFLAG"),
                                      FF_COMPOUNDFLAG);
        else if (STRNCMP(recoded, "COMPOUNDBEGIN") == 0)
            addCompoundAffixFlagValue(Conf, recoded + strlen("COMPOUNDBEGIN"),
                                      FF_COMPOUNDBEGIN);
        else if (STRNCMP(recoded, "COMPOUNDLAST") == 0)
            addCompoundAffixFlagValue(Conf, recoded + strlen("COMPOUNDLAST"),
                                      FF_COMPOUNDLAST);
        /* COMPOUNDLAST and COMPOUNDEND are synonyms */
        else if (STRNCMP(recoded, "COMPOUNDEND") == 0)
            addCompoundAffixFlagValue(Conf, recoded + strlen("COMPOUNDEND"),
                                      FF_COMPOUNDLAST);
        else if (STRNCMP(recoded, "COMPOUNDMIDDLE") == 0)
            addCompoundAffixFlagValue(Conf, recoded + strlen("COMPOUNDMIDDLE"),
                                      FF_COMPOUNDMIDDLE);
        else if (STRNCMP(recoded, "ONLYINCOMPOUND") == 0)
            addCompoundAffixFlagValue(Conf, recoded + strlen("ONLYINCOMPOUND"),
                                      FF_COMPOUNDONLY);
        else if (STRNCMP(recoded, "COMPOUNDPERMITFLAG") == 0)
            addCompoundAffixFlagValue(Conf,
                                      recoded + strlen("COMPOUNDPERMITFLAG"),
                                      FF_COMPOUNDPERMITFLAG);
        else if (STRNCMP(recoded, "COMPOUNDFORBIDFLAG") == 0)
            addCompoundAffixFlagValue(Conf,
                                      recoded + strlen("COMPOUNDFORBIDFLAG"),
                                      FF_COMPOUNDFORBIDFLAG);
        else if (STRNCMP(recoded, "FLAG") == 0)
        {
            char       *s = recoded + strlen("FLAG");
 
            while (*s && isspace((unsigned char) *s))
                s += pg_mblen(s);
 
            if (*s)
            {
                if (STRNCMP(s, "long") == 0)
                    Conf->flagMode = FM_LONG;
                else if (STRNCMP(s, "num") == 0)
                    Conf->flagMode = FM_NUM;
                else if (STRNCMP(s, "default") != 0)
                    ereport(ERROR,
                            (errcode(ERRCODE_CONFIG_FILE_ERROR),
                             errmsg("Ispell dictionary supports only "
                                    "\"default\", \"long\", "
                                    "and \"num\" flag values")));
            }
        }
 
        pfree(recoded);
    }
    tsearch_readline_end(&trst);
 
    if (Conf->nCompoundAffixFlag > 1)
        qsort(Conf->CompoundAffixFlags, Conf->nCompoundAffixFlag,
              sizeof(CompoundAffixFlag), cmpcmdflag);
 
    if (!tsearch_readline_begin(&trst, filename))
        ereport(ERROR,
                (errcode(ERRCODE_CONFIG_FILE_ERROR),
                 errmsg("could not open affix file \"%s\": %m",
                        filename)));
 
    while ((recoded = tsearch_readline(&trst)) != NULL)
    {
        int         fields_read;
 
        if (*recoded == '\0' || isspace((unsigned char) *recoded) || t_iseq(recoded, '#'))
            goto nextline;
 
        fields_read = parse_ooaffentry(recoded, type, sflag, find, repl, mask);
 
        if (ptype)
            pfree(ptype);
        ptype = lowerstr_ctx(Conf, type);
 
        /* First try to parse AF parameter (alias compression) */
        if (STRNCMP(ptype, "af") == 0)
        {
            /* First line is the number of aliases */
            if (!Conf->useFlagAliases)
            {
                Conf->useFlagAliases = true;
                naffix = atoi(sflag);
                if (naffix <= 0)
                    ereport(ERROR,
                            (errcode(ERRCODE_CONFIG_FILE_ERROR),
                             errmsg("invalid number of flag vector aliases")));
 
                /* Also reserve place for empty flag set */
                naffix++;
 
                Conf->AffixData = palloc0_array(const char *, naffix);
                Conf->lenAffixData = Conf->nAffixData = naffix;
 
                /* Add empty flag set into AffixData */
                Conf->AffixData[curaffix] = VoidString;
                curaffix++;
            }
            /* Other lines are aliases */
            else
            {
                if (curaffix < naffix)
                {
                    Conf->AffixData[curaffix] = cpstrdup(Conf, sflag);
                    curaffix++;
                }
                else
                    ereport(ERROR,
                            (errcode(ERRCODE_CONFIG_FILE_ERROR),
                             errmsg("number of aliases exceeds specified number %d",
                                    naffix - 1)));
            }
            goto nextline;
        }
        /* Else try to parse prefixes and suffixes */
        if (fields_read < 4 ||
            (STRNCMP(ptype, "sfx") != 0 && STRNCMP(ptype, "pfx") != 0))
            goto nextline;
 
        sflaglen = strlen(sflag);
        if (sflaglen == 0
            || (sflaglen > 1 && Conf->flagMode == FM_CHAR)
            || (sflaglen > 2 && Conf->flagMode == FM_LONG))
            goto nextline;
 
        /*--------
         * Affix header. For example:
         * SFX \ N 1
         *--------
         */
        if (fields_read == 4)
        {
            isSuffix = (STRNCMP(ptype, "sfx") == 0);
            if (t_iseq(find, 'y') || t_iseq(find, 'Y'))
                flagflags = FF_CROSSPRODUCT;
            else
                flagflags = 0;
        }
        /*--------
         * Affix fields. For example:
         * SFX \   0    Y/L [^Y]
         *--------
         */
        else
        {
            char       *ptr;
            int         aflg = 0;
 
            /* Get flags after '/' (flags are case sensitive) */
            if ((ptr = strchr(repl, '/')) != NULL)
                aflg |= getCompoundAffixFlagValue(Conf,
                                                  getAffixFlagSet(Conf,
                                                                  ptr + 1));
            /* Get lowercased version of string before '/' */
            prepl = lowerstr_ctx(Conf, repl);
            if ((ptr = strchr(prepl, '/')) != NULL)
                *ptr = '\0';
            pfind = lowerstr_ctx(Conf, find);
            pmask = lowerstr_ctx(Conf, mask);
            if (t_iseq(find, '0'))
                *pfind = '\0';
            if (t_iseq(repl, '0'))
                *prepl = '\0';
 
            NIAddAffix(Conf, sflag, flagflags | aflg, pmask, pfind, prepl,
                       isSuffix ? FF_SUFFIX : FF_PREFIX);
            pfree(prepl);
            pfree(pfind);
            pfree(pmask);
        }
 
nextline:
        pfree(recoded);
    }
 
    tsearch_readline_end(&trst);
    if (ptype)
        pfree(ptype);
}

References addCompoundAffixFlagValue(), cmpcmdflag(), cpstrdup(), ereport, errcode(), errmsg(), ERROR, fb(), FF_COMPOUNDBEGIN, FF_COMPOUNDFLAG, FF_COMPOUNDFORBIDFLAG, FF_COMPOUNDLAST, FF_COMPOUNDMIDDLE, FF_COMPOUNDONLY, FF_COMPOUNDPERMITFLAG, FF_CROSSPRODUCT, FF_PREFIX, FF_SUFFIX, filename, find(), FM_CHAR, FM_LONG, FM_NUM, getAffixFlagSet(), getCompoundAffixFlagValue(), lowerstr_ctx(), NIAddAffix(), palloc0_array, parse_ooaffentry(), pfree(), pg_mblen(), qsort, STRNCMP, t_iseq, tsearch_readline(), tsearch_readline_begin(), tsearch_readline_end(), type, and VoidString.

Referenced by NIImportAffixes().

NINormalizeWord()

TSLexeme * NINormalizeWord	(	IspellDict *	Conf,
		const char *	word
	)

Definition at line 2545 of file spell.c.

{
    char      **res;
    TSLexeme   *lcur = NULL,
               *lres = NULL;
    uint16      NVariant = 1;
 
    res = NormalizeSubWord(Conf, word, 0);
 
    if (res)
    {
        char      **ptr = res;
 
        while (*ptr && (lcur - lres) < MAX_NORM)
        {
            addNorm(&lres, &lcur, *ptr, 0, NVariant++);
            ptr++;
        }
        pfree(res);
    }
 
    if (Conf->usecompound)
    {
        int         wordlen = strlen(word);
        SplitVar   *ptr,
                   *var = SplitToVariants(Conf, NULL, NULL, word, wordlen, 0, -1);
        int         i;
 
        while (var)
        {
            if (var->nstem > 1)
            {
                char      **subres = NormalizeSubWord(Conf, var->stem[var->nstem - 1], FF_COMPOUNDLAST);
 
                if (subres)
                {
                    char      **subptr = subres;
 
                    while (*subptr)
                    {
                        for (i = 0; i < var->nstem - 1; i++)
                        {
                            addNorm(&lres, &lcur, (subptr == subres) ? var->stem[i] : pstrdup(var->stem[i]), 0, NVariant);
                        }
 
                        addNorm(&lres, &lcur, *subptr, 0, NVariant);
                        subptr++;
                        NVariant++;
                    }
 
                    pfree(subres);
                    var->stem[0] = NULL;
                    pfree(var->stem[var->nstem - 1]);
                }
            }
 
            for (i = 0; i < var->nstem && var->stem[i]; i++)
                pfree(var->stem[i]);
            ptr = var->next;
            pfree(var->stem);
            pfree(var);
            var = ptr;
        }
    }
 
    return lres;
}

References addNorm(), fb(), FF_COMPOUNDLAST, i, MAX_NORM, SplitVar::next, NormalizeSubWord(), SplitVar::nstem, pfree(), pstrdup(), SplitToVariants(), SplitVar::stem, and word().

Referenced by dispell_lexize().

NISortAffixes()

void NISortAffixes

(

IspellDict *

Conf

)

Definition at line 1981 of file spell.c.

{
    AFFIX      *Affix;
    size_t      i;
    CMPDAffix  *ptr;
    int         firstsuffix = Conf->naffixes;
 
    if (Conf->naffixes == 0)
        return;
 
    /* Store compound affixes in the Conf->CompoundAffix array */
    if (Conf->naffixes > 1)
        qsort(Conf->Affix, Conf->naffixes, sizeof(AFFIX), cmpaffix);
    Conf->CompoundAffix = ptr = palloc_array(CMPDAffix, Conf->naffixes);
    ptr->affix = NULL;
 
    for (i = 0; i < Conf->naffixes; i++)
    {
        Affix = &(((AFFIX *) Conf->Affix)[i]);
        if (Affix->type == FF_SUFFIX && i < firstsuffix)
            firstsuffix = i;
 
        if ((Affix->flagflags & FF_COMPOUNDFLAG) && Affix->replen > 0 &&
            isAffixInUse(Conf, Affix->flag))
        {
            bool        issuffix = (Affix->type == FF_SUFFIX);
 
            if (ptr == Conf->CompoundAffix ||
                issuffix != (ptr - 1)->issuffix ||
                strbncmp((const unsigned char *) (ptr - 1)->affix,
                         (const unsigned char *) Affix->repl,
                         (ptr - 1)->len))
            {
                /* leave only unique and minimal suffixes */
                ptr->affix = Affix->repl;
                ptr->len = Affix->replen;
                ptr->issuffix = issuffix;
                ptr++;
            }
        }
    }
    ptr->affix = NULL;
    Conf->CompoundAffix = (CMPDAffix *) repalloc(Conf->CompoundAffix, sizeof(CMPDAffix) * (ptr - Conf->CompoundAffix + 1));
 
    /* Start build a prefix tree */
    Conf->Prefix = mkANode(Conf, 0, firstsuffix, 0, FF_PREFIX);
    Conf->Suffix = mkANode(Conf, firstsuffix, Conf->naffixes, 0, FF_SUFFIX);
    mkVoidAffix(Conf, true, firstsuffix);
    mkVoidAffix(Conf, false, firstsuffix);
}

References CMPDAffix::affix, cmpaffix(), fb(), FF_COMPOUNDFLAG, FF_PREFIX, FF_SUFFIX, aff_struct::flag, aff_struct::flagflags, i, isAffixInUse(), CMPDAffix::issuffix, CMPDAffix::len, mkANode(), mkVoidAffix(), palloc_array, qsort, repalloc(), aff_struct::repl, aff_struct::replen, strbncmp(), and aff_struct::type.

Referenced by dispell_init().

NISortDictionary()

void NISortDictionary

(

IspellDict *

Conf

)

Definition at line 1726 of file spell.c.

{
    int         i;
    int         naffix;
    int         curaffix;
 
    /* compress affixes */
 
    /*
     * If we use flag aliases then we need to use Conf->AffixData filled in
     * the NIImportOOAffixes().
     */
    if (Conf->useFlagAliases)
    {
        for (i = 0; i < Conf->nspell; i++)
        {
            char       *end;
 
            if (*Conf->Spell[i]->p.flag != '\0')
            {
                errno = 0;
                curaffix = strtol(Conf->Spell[i]->p.flag, &end, 10);
                if (Conf->Spell[i]->p.flag == end || errno == ERANGE)
                    ereport(ERROR,
                            (errcode(ERRCODE_CONFIG_FILE_ERROR),
                             errmsg("invalid affix alias \"%s\"",
                                    Conf->Spell[i]->p.flag)));
                if (curaffix < 0 || curaffix >= Conf->nAffixData)
                    ereport(ERROR,
                            (errcode(ERRCODE_CONFIG_FILE_ERROR),
                             errmsg("invalid affix alias \"%s\"",
                                    Conf->Spell[i]->p.flag)));
                if (*end != '\0' && !isdigit((unsigned char) *end) && !isspace((unsigned char) *end))
                    ereport(ERROR,
                            (errcode(ERRCODE_CONFIG_FILE_ERROR),
                             errmsg("invalid affix alias \"%s\"",
                                    Conf->Spell[i]->p.flag)));
            }
            else
            {
                /*
                 * If Conf->Spell[i]->p.flag is empty, then get empty value of
                 * Conf->AffixData (0 index).
                 */
                curaffix = 0;
            }
 
            Conf->Spell[i]->p.d.affix = curaffix;
            Conf->Spell[i]->p.d.len = strlen(Conf->Spell[i]->word);
        }
    }
    /* Otherwise fill Conf->AffixData here */
    else
    {
        /* Count the number of different flags used in the dictionary */
        qsort(Conf->Spell, Conf->nspell, sizeof(SPELL *),
              cmpspellaffix);
 
        naffix = 0;
        for (i = 0; i < Conf->nspell; i++)
        {
            if (i == 0 ||
                strcmp(Conf->Spell[i]->p.flag, Conf->Spell[i - 1]->p.flag) != 0)
                naffix++;
        }
 
        /*
         * Fill in Conf->AffixData with the affixes that were used in the
         * dictionary. Replace textual flag-field of Conf->Spell entries with
         * indexes into Conf->AffixData array.
         */
        Conf->AffixData = palloc0_array(const char *, naffix);
 
        curaffix = -1;
        for (i = 0; i < Conf->nspell; i++)
        {
            if (i == 0 ||
                strcmp(Conf->Spell[i]->p.flag, Conf->AffixData[curaffix]) != 0)
            {
                curaffix++;
                Assert(curaffix < naffix);
                Conf->AffixData[curaffix] = cpstrdup(Conf,
                                                     Conf->Spell[i]->p.flag);
            }
 
            Conf->Spell[i]->p.d.affix = curaffix;
            Conf->Spell[i]->p.d.len = strlen(Conf->Spell[i]->word);
        }
 
        Conf->lenAffixData = Conf->nAffixData = naffix;
    }
 
    /* Start build a prefix tree */
    qsort(Conf->Spell, Conf->nspell, sizeof(SPELL *), cmpspell);
    Conf->Dictionary = mkSPNode(Conf, 0, Conf->nspell, 0);
}

References Assert, cmpspell(), cmpspellaffix(), cpstrdup(), ereport, errcode(), errmsg(), ERROR, fb(), i, mkSPNode(), palloc0_array, and qsort.

Referenced by dispell_init().

NIStartBuild()

void NIStartBuild

(

IspellDict *

Conf

)

Definition at line 89 of file spell.c.

{
    /*
     * The temp context is a child of CurTransactionContext, so that it will
     * go away automatically on error.
     */
    Conf->buildCxt = AllocSetContextCreate(CurTransactionContext,
                                           "Ispell dictionary init context",
                                           ALLOCSET_DEFAULT_SIZES);
}

References ALLOCSET_DEFAULT_SIZES, AllocSetContextCreate, CurTransactionContext, and fb().

Referenced by dispell_init().

NormalizeSubWord()

static char ** NormalizeSubWord ( IspellDict *  Conf, const char *  word, int  flag  )

static

Definition at line 2181 of file spell.c.

{
    AffixNodeData *suffix = NULL,
               *prefix = NULL;
    int         slevel = 0,
                plevel = 0;
    int         wrdlen = strlen(word),
                swrdlen;
    char      **forms;
    char      **cur;
    char        newword[2 * MAXNORMLEN] = "";
    char        pnewword[2 * MAXNORMLEN] = "";
    AffixNode  *snode = Conf->Suffix,
               *pnode;
    int         i,
                j;
 
    if (wrdlen > MAXNORMLEN)
        return NULL;
    cur = forms = palloc_array(char *, MAX_NORM);
    *cur = NULL;
 
 
    /* Check that the word itself is normal form */
    if (FindWord(Conf, word, VoidString, flag))
    {
        *cur = pstrdup(word);
        cur++;
        *cur = NULL;
    }
 
    /* Find all other NORMAL forms of the 'word' (check only prefix) */
    pnode = Conf->Prefix;
    plevel = 0;
    while (pnode)
    {
        prefix = FindAffixes(pnode, word, wrdlen, &plevel, FF_PREFIX);
        if (!prefix)
            break;
        for (j = 0; j < prefix->naff; j++)
        {
            if (CheckAffix(word, wrdlen, prefix->aff[j], flag, newword, NULL))
            {
                /* prefix success */
                if (FindWord(Conf, newword, prefix->aff[j]->flag, flag))
                    cur += addToResult(forms, cur, newword);
            }
        }
        pnode = prefix->node;
    }
 
    /*
     * Find all other NORMAL forms of the 'word' (check suffix and then
     * prefix)
     */
    while (snode)
    {
        int         baselen = 0;
 
        /* find possible suffix */
        suffix = FindAffixes(snode, word, wrdlen, &slevel, FF_SUFFIX);
        if (!suffix)
            break;
        /* foreach suffix check affix */
        for (i = 0; i < suffix->naff; i++)
        {
            if (CheckAffix(word, wrdlen, suffix->aff[i], flag, newword, &baselen))
            {
                /* suffix success */
                if (FindWord(Conf, newword, suffix->aff[i]->flag, flag))
                    cur += addToResult(forms, cur, newword);
 
                /* now we will look changed word with prefixes */
                pnode = Conf->Prefix;
                plevel = 0;
                swrdlen = strlen(newword);
                while (pnode)
                {
                    prefix = FindAffixes(pnode, newword, swrdlen, &plevel, FF_PREFIX);
                    if (!prefix)
                        break;
                    for (j = 0; j < prefix->naff; j++)
                    {
                        if (CheckAffix(newword, swrdlen, prefix->aff[j], flag, pnewword, &baselen))
                        {
                            /* prefix success */
                            const char *ff = (prefix->aff[j]->flagflags & suffix->aff[i]->flagflags & FF_CROSSPRODUCT) ?
                                VoidString : prefix->aff[j]->flag;
 
                            if (FindWord(Conf, pnewword, ff, flag))
                                cur += addToResult(forms, cur, pnewword);
                        }
                    }
                    pnode = prefix->node;
                }
            }
        }
 
        snode = suffix->node;
    }
 
    if (cur == forms)
    {
        pfree(forms);
        return NULL;
    }
    return forms;
}

References addToResult(), AffixNodeData::aff, CheckAffix(), cur, fb(), FF_CROSSPRODUCT, FF_PREFIX, FF_SUFFIX, FindAffixes(), FindWord(), aff_struct::flag, flag(), aff_struct::flagflags, i, j, MAX_NORM, MAXNORMLEN, AffixNodeData::naff, AffixNodeData::node, palloc_array, pfree(), pstrdup(), VoidString, and word().

Referenced by NINormalizeWord(), and SplitToVariants().

parse_affentry()

static bool parse_affentry ( char *  str, char *  mask, char *  find, char *  repl  )

static

Definition at line 915 of file spell.c.

{
    int         state = PAE_WAIT_MASK;
    char       *pmask = mask,
               *pfind = find,
               *prepl = repl;
 
    *mask = *find = *repl = '\0';
 
    while (*str)
    {
        if (state == PAE_WAIT_MASK)
        {
            if (t_iseq(str, '#'))
                return false;
            else if (!isspace((unsigned char) *str))
            {
                COPYCHAR(pmask, str);
                pmask += pg_mblen(str);
                state = PAE_INMASK;
            }
        }
        else if (state == PAE_INMASK)
        {
            if (t_iseq(str, '>'))
            {
                *pmask = '\0';
                state = PAE_WAIT_FIND;
            }
            else if (!isspace((unsigned char) *str))
            {
                COPYCHAR(pmask, str);
                pmask += pg_mblen(str);
            }
        }
        else if (state == PAE_WAIT_FIND)
        {
            if (t_iseq(str, '-'))
            {
                state = PAE_INFIND;
            }
            else if (t_isalpha(str) || t_iseq(str, '\'') /* english 's */ )
            {
                COPYCHAR(prepl, str);
                prepl += pg_mblen(str);
                state = PAE_INREPL;
            }
            else if (!isspace((unsigned char) *str))
                ereport(ERROR,
                        (errcode(ERRCODE_CONFIG_FILE_ERROR),
                         errmsg("syntax error")));
        }
        else if (state == PAE_INFIND)
        {
            if (t_iseq(str, ','))
            {
                *pfind = '\0';
                state = PAE_WAIT_REPL;
            }
            else if (t_isalpha(str))
            {
                COPYCHAR(pfind, str);
                pfind += pg_mblen(str);
            }
            else if (!isspace((unsigned char) *str))
                ereport(ERROR,
                        (errcode(ERRCODE_CONFIG_FILE_ERROR),
                         errmsg("syntax error")));
        }
        else if (state == PAE_WAIT_REPL)
        {
            if (t_iseq(str, '-'))
            {
                break;          /* void repl */
            }
            else if (t_isalpha(str))
            {
                COPYCHAR(prepl, str);
                prepl += pg_mblen(str);
                state = PAE_INREPL;
            }
            else if (!isspace((unsigned char) *str))
                ereport(ERROR,
                        (errcode(ERRCODE_CONFIG_FILE_ERROR),
                         errmsg("syntax error")));
        }
        else if (state == PAE_INREPL)
        {
            if (t_iseq(str, '#'))
            {
                *prepl = '\0';
                break;
            }
            else if (t_isalpha(str))
            {
                COPYCHAR(prepl, str);
                prepl += pg_mblen(str);
            }
            else if (!isspace((unsigned char) *str))
                ereport(ERROR,
                        (errcode(ERRCODE_CONFIG_FILE_ERROR),
                         errmsg("syntax error")));
        }
        else
            elog(ERROR, "unrecognized state in parse_affentry: %d", state);
 
        str += pg_mblen(str);
    }
 
    *pmask = *pfind = *prepl = '\0';
 
    return (*mask && (*find || *repl));
}

References COPYCHAR, elog, ereport, errcode(), errmsg(), ERROR, fb(), find(), PAE_INFIND, PAE_INMASK, PAE_INREPL, PAE_WAIT_FIND, PAE_WAIT_MASK, PAE_WAIT_REPL, pg_mblen(), str, t_isalpha(), and t_iseq.

Referenced by NIImportAffixes().

parse_ooaffentry()

static int parse_ooaffentry ( char *  str, char *  type, char *  flag, char *  find, char *  repl, char *  mask  )

static

Definition at line 859 of file spell.c.

{
    int         state = PAE_WAIT_TYPE;
    int         fields_read = 0;
    bool        valid = false;
 
    *type = *flag = *find = *repl = *mask = '\0';
 
    while (*str)
    {
        switch (state)
        {
            case PAE_WAIT_TYPE:
                valid = get_nextfield(&str, type);
                state = PAE_WAIT_FLAG;
                break;
            case PAE_WAIT_FLAG:
                valid = get_nextfield(&str, flag);
                state = PAE_WAIT_FIND;
                break;
            case PAE_WAIT_FIND:
                valid = get_nextfield(&str, find);
                state = PAE_WAIT_REPL;
                break;
            case PAE_WAIT_REPL:
                valid = get_nextfield(&str, repl);
                state = PAE_WAIT_MASK;
                break;
            case PAE_WAIT_MASK:
                valid = get_nextfield(&str, mask);
                state = -1;     /* force loop exit */
                break;
            default:
                elog(ERROR, "unrecognized state in parse_ooaffentry: %d",
                     state);
                break;
        }
        if (valid)
            fields_read++;
        else
            break;              /* early EOL */
        if (state < 0)
            break;              /* got all fields */
    }
 
    return fields_read;
}

References elog, ERROR, fb(), find(), flag(), get_nextfield(), PAE_WAIT_FIND, PAE_WAIT_FLAG, PAE_WAIT_MASK, PAE_WAIT_REPL, PAE_WAIT_TYPE, str, and type.

Referenced by NIImportOOAffixes().

setCompoundAffixFlagValue()

static void setCompoundAffixFlagValue ( IspellDict *  Conf, CompoundAffixFlag *  entry, char *  s, uint32  val  )

static

Definition at line 1033 of file spell.c.

{
    if (Conf->flagMode == FM_NUM)
    {
        char       *next;
        int         i;
 
        errno = 0;
        i = strtol(s, &next, 10);
        if (s == next || errno == ERANGE)
            ereport(ERROR,
                    (errcode(ERRCODE_CONFIG_FILE_ERROR),
                     errmsg("invalid affix flag \"%s\"", s)));
        if (i < 0 || i > FLAGNUM_MAXSIZE)
            ereport(ERROR,
                    (errcode(ERRCODE_CONFIG_FILE_ERROR),
                     errmsg("affix flag \"%s\" is out of range", s)));
 
        entry->flag.i = i;
    }
    else
        entry->flag.s = cpstrdup(Conf, s);
 
    entry->flagMode = Conf->flagMode;
    entry->value = val;
}

References cpstrdup(), ereport, errcode(), errmsg(), ERROR, fb(), CompoundAffixFlag::flag, CompoundAffixFlag::flagMode, FLAGNUM_MAXSIZE, FM_NUM, i, CompoundAffixFlag::i, next, CompoundAffixFlag::s, val, and CompoundAffixFlag::value.

Referenced by addCompoundAffixFlagValue(), and getCompoundAffixFlagValue().

SplitToVariants()

static SplitVar * SplitToVariants ( IspellDict *  Conf, SPNode *  snode, SplitVar *  orig, const char *  word, int  wordlen, int  startpos, int  minpos  )

static

Definition at line 2379 of file spell.c.

{
    SplitVar   *var = NULL;
    SPNodeData *StopLow,
               *StopHigh,
               *StopMiddle = NULL;
    SPNode     *node = (snode) ? snode : Conf->Dictionary;
    int         level = (snode) ? minpos : startpos;    /* recursive
                                                         * minpos==level */
    int         lenaff;
    CMPDAffix  *caff;
    char       *notprobed;
    int         compoundflag = 0;
 
    /* since this function recurses, it could be driven to stack overflow */
    check_stack_depth();
 
    notprobed = (char *) palloc(wordlen);
    memset(notprobed, 1, wordlen);
    var = CopyVar(orig, 1);
 
    while (level < wordlen)
    {
        /* find word with epenthetic or/and compound affix */
        caff = Conf->CompoundAffix;
        while (level > startpos && (lenaff = CheckCompoundAffixes(&caff, word + level, wordlen - level, (node) ? true : false)) >= 0)
        {
            /*
             * there is one of compound affixes, so check word for existings
             */
            char        buf[MAXNORMLEN];
            char      **subres;
 
            lenaff = level - startpos + lenaff;
 
            if (!notprobed[startpos + lenaff - 1])
                continue;
 
            if (level + lenaff - 1 <= minpos)
                continue;
 
            if (lenaff >= MAXNORMLEN)
                continue;       /* skip too big value */
            if (lenaff > 0)
                memcpy(buf, word + startpos, lenaff);
            buf[lenaff] = '\0';
 
            if (level == 0)
                compoundflag = FF_COMPOUNDBEGIN;
            else if (level == wordlen - 1)
                compoundflag = FF_COMPOUNDLAST;
            else
                compoundflag = FF_COMPOUNDMIDDLE;
            subres = NormalizeSubWord(Conf, buf, compoundflag);
            if (subres)
            {
                /* Yes, it was a word from dictionary */
                SplitVar   *new = CopyVar(var, 0);
                SplitVar   *ptr = var;
                char      **sptr = subres;
 
                notprobed[startpos + lenaff - 1] = 0;
 
                while (*sptr)
                {
                    AddStem(new, *sptr);
                    sptr++;
                }
                pfree(subres);
 
                while (ptr->next)
                    ptr = ptr->next;
                ptr->next = SplitToVariants(Conf, NULL, new, word, wordlen, startpos + lenaff, startpos + lenaff);
 
                pfree(new->stem);
                pfree(new);
            }
        }
 
        if (!node)
            break;
 
        StopLow = node->data;
        StopHigh = node->data + node->length;
        while (StopLow < StopHigh)
        {
            StopMiddle = StopLow + ((StopHigh - StopLow) >> 1);
            if (StopMiddle->val == ((const uint8 *) (word))[level])
                break;
            else if (StopMiddle->val < ((const uint8 *) (word))[level])
                StopLow = StopMiddle + 1;
            else
                StopHigh = StopMiddle;
        }
 
        if (StopLow < StopHigh)
        {
            if (startpos == 0)
                compoundflag = FF_COMPOUNDBEGIN;
            else if (level == wordlen - 1)
                compoundflag = FF_COMPOUNDLAST;
            else
                compoundflag = FF_COMPOUNDMIDDLE;
 
            /* find infinitive */
            if (StopMiddle->isword &&
                (StopMiddle->compoundflag & compoundflag) &&
                notprobed[level])
            {
                /* ok, we found full compoundallowed word */
                if (level > minpos)
                {
                    /* and its length more than minimal */
                    if (wordlen == level + 1)
                    {
                        /* well, it was last word */
                        AddStem(var, pnstrdup(word + startpos, wordlen - startpos));
                        pfree(notprobed);
                        return var;
                    }
                    else
                    {
                        /* then we will search more big word at the same point */
                        SplitVar   *ptr = var;
 
                        while (ptr->next)
                            ptr = ptr->next;
                        ptr->next = SplitToVariants(Conf, node, var, word, wordlen, startpos, level);
                        /* we can find next word */
                        level++;
                        AddStem(var, pnstrdup(word + startpos, level - startpos));
                        node = Conf->Dictionary;
                        startpos = level;
                        continue;
                    }
                }
            }
            node = StopMiddle->node;
        }
        else
            node = NULL;
        level++;
    }
 
    AddStem(var, pnstrdup(word + startpos, wordlen - startpos));
    pfree(notprobed);
    return var;
}

References AddStem(), buf, check_stack_depth(), CheckCompoundAffixes(), CopyVar(), SPNode::data, fb(), FF_COMPOUNDBEGIN, FF_COMPOUNDLAST, FF_COMPOUNDMIDDLE, SPNode::length, MAXNORMLEN, SplitVar::next, NormalizeSubWord(), palloc(), pfree(), pnstrdup(), SplitToVariants(), startpos, and word().

Referenced by NINormalizeWord(), and SplitToVariants().

strbcmp()

static int strbcmp ( const unsigned char *  s1, const unsigned char *  s2  )

static

Definition at line 258 of file spell.c.

{
    int         l1 = strlen((const char *) s1) - 1,
                l2 = strlen((const char *) s2) - 1;
 
    while (l1 >= 0 && l2 >= 0)
    {
        if (s1[l1] < s2[l2])
            return -1;
        if (s1[l1] > s2[l2])
            return 1;
        l1--;
        l2--;
    }
    if (l1 < l2)
        return -1;
    if (l1 > l2)
        return 1;
 
    return 0;
}

References fb(), s1, and s2.

Referenced by cmpaffix().

strbncmp()

static int strbncmp ( const unsigned char *  s1, const unsigned char *  s2, size_t  count  )

static

Definition at line 281 of file spell.c.

{
    int         l1 = strlen((const char *) s1) - 1,
                l2 = strlen((const char *) s2) - 1,
                l = count;
 
    while (l1 >= 0 && l2 >= 0 && l > 0)
    {
        if (s1[l1] < s2[l2])
            return -1;
        if (s1[l1] > s2[l2])
            return 1;
        l1--;
        l2--;
        l--;
    }
    if (l == 0)
        return 0;
    if (l1 < l2)
        return -1;
    if (l1 > l2)
        return 1;
    return 0;
}

References fb(), s1, and s2.

Referenced by NISortAffixes().

Variable Documentation

VoidString

const char* VoidString = ""

static

Definition at line 195 of file spell.c.

Referenced by getAffixFlagSet(), NIAddAffix(), NIAddSpell(), NIImportOOAffixes(), and NormalizeSubWord().