yb_ddl_parser 0.1.0

data/ext/yb_ddl_parser/vendor/yugabyte-postgres/src/include/tsearch/dicts/spell.h

@@ -0,0 +1,247 @@
+/*-------------------------------------------------------------------------
+ *
+ * spell.h
+ *
+ * Declarations for ISpell dictionary
+ *
+ * Portions Copyright (c) 1996-2022, PostgreSQL Global Development Group
+ *
+ * src/include/tsearch/dicts/spell.h
+ *
+ *-------------------------------------------------------------------------
+ */
+
+#ifndef __SPELL_H__
+#define __SPELL_H__
+
+#include "regex/regex.h"
+#include "tsearch/dicts/regis.h"
+#include "tsearch/ts_public.h"
+
+/*
+ * SPNode and SPNodeData are used to represent prefix tree (Trie) to store
+ * a words list.
+ */
+struct SPNode;
+
+typedef struct
+{
+	uint32		val:8,
+				isword:1,
+	/* Stores compound flags listed below */
+				compoundflag:4,
+	/* Reference to an entry of the AffixData field */
+				affix:19;
+	struct SPNode *node;
+} SPNodeData;
+
+/*
+ * Names of FF_ are correlated with Hunspell options in affix file
+ * http://hunspell.sourceforge.net/
+ */
+#define FF_COMPOUNDONLY		0x01
+#define FF_COMPOUNDBEGIN	0x02
+#define FF_COMPOUNDMIDDLE	0x04
+#define FF_COMPOUNDLAST		0x08
+#define FF_COMPOUNDFLAG		( FF_COMPOUNDBEGIN | FF_COMPOUNDMIDDLE | \
+							FF_COMPOUNDLAST )
+#define FF_COMPOUNDFLAGMASK		0x0f
+
+typedef struct SPNode
+{
+	uint32		length;
+	SPNodeData	data[FLEXIBLE_ARRAY_MEMBER];
+} SPNode;
+
+#define SPNHDRSZ	(offsetof(SPNode,data))
+
+/*
+ * Represents an entry in a words list.
+ */
+typedef struct spell_struct
+{
+	union
+	{
+		/*
+		 * flag is filled in by NIImportDictionary(). After
+		 * NISortDictionary(), d is used instead of flag.
+		 */
+		char	   *flag;
+		/* d is used in mkSPNode() */
+		struct
+		{
+			/* Reference to an entry of the AffixData field */
+			int			affix;
+			/* Length of the word */
+			int			len;
+		}			d;
+	}			p;
+	char		word[FLEXIBLE_ARRAY_MEMBER];
+} SPELL;
+
+#define SPELLHDRSZ	(offsetof(SPELL, word))
+
+/*
+ * If an affix uses a regex, we have to store that separately in a struct
+ * that won't move around when arrays of affixes are enlarged or sorted.
+ * This is so that it can be found to be cleaned up at context destruction.
+ */
+typedef struct aff_regex_struct
+{
+	regex_t		regex;
+	MemoryContextCallback mcallback;
+} aff_regex_struct;
+
+/*
+ * Represents an entry in an affix list.
+ */
+typedef struct aff_struct
+{
+	char	   *flag;
+	/* FF_SUFFIX or FF_PREFIX */
+	uint32		type:1,
+				flagflags:7,
+				issimple:1,
+				isregis:1,
+				replen:14;
+	char	   *find;
+	char	   *repl;
+	union
+	{
+		aff_regex_struct *pregex;
+		Regis		regis;
+	}			reg;
+} AFFIX;
+
+/*
+ * affixes use dictionary flags too
+ */
+#define FF_COMPOUNDPERMITFLAG	0x10
+#define FF_COMPOUNDFORBIDFLAG	0x20
+#define FF_CROSSPRODUCT			0x40
+
+/*
+ * Don't change the order of these. Initialization sorts by these,
+ * and expects prefixes to come first after sorting.
+ */
+#define FF_SUFFIX				1
+#define FF_PREFIX				0
+
+/*
+ * AffixNode and AffixNodeData are used to represent prefix tree (Trie) to store
+ * an affix list.
+ */
+struct AffixNode;
+
+typedef struct
+{
+	uint32		val:8,
+				naff:24;
+	AFFIX	  **aff;
+	struct AffixNode *node;
+} AffixNodeData;
+
+typedef struct AffixNode
+{
+	uint32		isvoid:1,
+				length:31;
+	AffixNodeData data[FLEXIBLE_ARRAY_MEMBER];
+} AffixNode;
+
+#define ANHRDSZ		   (offsetof(AffixNode, data))
+
+typedef struct
+{
+	char	   *affix;
+	int			len;
+	bool		issuffix;
+} CMPDAffix;
+
+/*
+ * Type of encoding affix flags in Hunspell dictionaries
+ */
+typedef enum
+{
+	FM_CHAR,					/* one character (like ispell) */
+	FM_LONG,					/* two characters */
+	FM_NUM						/* number, >= 0 and < 65536 */
+} FlagMode;
+
+/*
+ * Structure to store Hunspell options. Flag representation depends on flag
+ * type. These flags are about support of compound words.
+ */
+typedef struct CompoundAffixFlag
+{
+	union
+	{
+		/* Flag name if flagMode is FM_CHAR or FM_LONG */
+		char	   *s;
+		/* Flag name if flagMode is FM_NUM */
+		uint32		i;
+	}			flag;
+	/* we don't have a bsearch_arg version, so, copy FlagMode */
+	FlagMode	flagMode;
+	uint32		value;
+} CompoundAffixFlag;
+
+#define FLAGNUM_MAXSIZE		(1 << 16)
+
+typedef struct
+{
+	int			maffixes;
+	int			naffixes;
+	AFFIX	   *Affix;
+
+	AffixNode  *Suffix;
+	AffixNode  *Prefix;
+
+	SPNode	   *Dictionary;
+	/* Array of sets of affixes */
+	char	  **AffixData;
+	int			lenAffixData;
+	int			nAffixData;
+	bool		useFlagAliases;
+
+	CMPDAffix  *CompoundAffix;
+
+	bool		usecompound;
+	FlagMode	flagMode;
+
+	/*
+	 * All follow fields are actually needed only for initialization
+	 */
+
+	/* Array of Hunspell options in affix file */
+	CompoundAffixFlag *CompoundAffixFlags;
+	/* number of entries in CompoundAffixFlags array */
+	int			nCompoundAffixFlag;
+	/* allocated length of CompoundAffixFlags array */
+	int			mCompoundAffixFlag;
+
+	/*
+	 * Remaining fields are only used during dictionary construction; they are
+	 * set up by NIStartBuild and cleared by NIFinishBuild.
+	 */
+	MemoryContext buildCxt;		/* temp context for construction */
+
+	/* Temporary array of all words in the dict file */
+	SPELL	  **Spell;
+	int			nspell;			/* number of valid entries in Spell array */
+	int			mspell;			/* allocated length of Spell array */
+
+	/* These are used to allocate "compact" data without palloc overhead */
+	char	   *firstfree;		/* first free address (always maxaligned) */
+	size_t		avail;			/* free space remaining at firstfree */
+} IspellDict;
+
+extern TSLexeme *NINormalizeWord(IspellDict *Conf, char *word);
+
+extern void NIStartBuild(IspellDict *Conf);
+extern void NIImportAffixes(IspellDict *Conf, const char *filename);
+extern void NIImportDictionary(IspellDict *Conf, const char *filename);
+extern void NISortDictionary(IspellDict *Conf);
+extern void NISortAffixes(IspellDict *Conf);
+extern void NIFinishBuild(IspellDict *Conf);
+
+#endif

data/ext/yb_ddl_parser/vendor/yugabyte-postgres/src/include/tsearch/ts_cache.h

@@ -0,0 +1,98 @@
+/*-------------------------------------------------------------------------
+ *
+ * ts_cache.h
+ *	  Tsearch related object caches.
+ *
+ * Portions Copyright (c) 1996-2022, PostgreSQL Global Development Group
+ * Portions Copyright (c) 1994, Regents of the University of California
+ *
+ * src/include/tsearch/ts_cache.h
+ *
+ *-------------------------------------------------------------------------
+ */
+#ifndef TS_CACHE_H
+#define TS_CACHE_H
+
+#include "utils/guc.h"
+
+
+/*
+ * All TS*CacheEntry structs must share this common header
+ * (see InvalidateTSCacheCallBack)
+ */
+typedef struct TSAnyCacheEntry
+{
+	Oid			objId;
+	bool		isvalid;
+} TSAnyCacheEntry;
+
+
+typedef struct TSParserCacheEntry
+{
+	/* prsId is the hash lookup key and MUST BE FIRST */
+	Oid			prsId;			/* OID of the parser */
+	bool		isvalid;
+
+	Oid			startOid;
+	Oid			tokenOid;
+	Oid			endOid;
+	Oid			headlineOid;
+	Oid			lextypeOid;
+
+	/*
+	 * Pre-set-up fmgr call of most needed parser's methods
+	 */
+	FmgrInfo	prsstart;
+	FmgrInfo	prstoken;
+	FmgrInfo	prsend;
+	FmgrInfo	prsheadline;
+} TSParserCacheEntry;
+
+typedef struct TSDictionaryCacheEntry
+{
+	/* dictId is the hash lookup key and MUST BE FIRST */
+	Oid			dictId;
+	bool		isvalid;
+
+	/* most frequent fmgr call */
+	Oid			lexizeOid;
+	FmgrInfo	lexize;
+
+	MemoryContext dictCtx;		/* memory context to store private data */
+	void	   *dictData;
+} TSDictionaryCacheEntry;
+
+typedef struct
+{
+	int			len;
+	Oid		   *dictIds;
+} ListDictionary;
+
+typedef struct
+{
+	/* cfgId is the hash lookup key and MUST BE FIRST */
+	Oid			cfgId;
+	bool		isvalid;
+
+	Oid			prsId;
+
+	int			lenmap;
+	ListDictionary *map;
+} TSConfigCacheEntry;
+
+
+/*
+ * GUC variable for current configuration
+ */
+extern PGDLLIMPORT char *TSCurrentConfig;
+
+
+extern TSParserCacheEntry *lookup_ts_parser_cache(Oid prsId);
+extern TSDictionaryCacheEntry *lookup_ts_dictionary_cache(Oid dictId);
+extern TSConfigCacheEntry *lookup_ts_config_cache(Oid cfgId);
+
+extern Oid	getTSCurrentConfig(bool emitError);
+extern bool check_TSCurrentConfig(char **newval, void **extra, GucSource source);
+extern void assign_TSCurrentConfig(const char *newval, void *extra);
+
+#endif							/* TS_CACHE_H */

data/ext/yb_ddl_parser/vendor/yugabyte-postgres/src/include/tsearch/ts_locale.h

@@ -0,0 +1,63 @@
+/*-------------------------------------------------------------------------
+ *
+ * ts_locale.h
+ *		locale compatibility layer for tsearch
+ *
+ * Copyright (c) 1998-2022, PostgreSQL Global Development Group
+ *
+ * src/include/tsearch/ts_locale.h
+ *
+ *-------------------------------------------------------------------------
+ */
+#ifndef __TSLOCALE_H__
+#define __TSLOCALE_H__
+
+#include <ctype.h>
+#include <limits.h>
+
+#include "lib/stringinfo.h"
+#include "mb/pg_wchar.h"
+#include "utils/pg_locale.h"
+
+/*
+ * towlower() and friends should be in <wctype.h>, but some pre-C99 systems
+ * declare them in <wchar.h>, so include that too.
+ */
+#include <wchar.h>
+#ifdef HAVE_WCTYPE_H
+#include <wctype.h>
+#endif
+
+/* working state for tsearch_readline (should be a local var in caller) */
+typedef struct
+{
+	FILE	   *fp;
+	const char *filename;
+	int			lineno;
+	StringInfoData buf;			/* current input line, in UTF-8 */
+	char	   *curline;		/* current input line, in DB's encoding */
+	/* curline may be NULL, or equal to buf.data, or a palloc'd string */
+	ErrorContextCallback cb;
+} tsearch_readline_state;
+
+#define TOUCHAR(x)	(*((const unsigned char *) (x)))
+
+/* The second argument of t_iseq() must be a plain ASCII character */
+#define t_iseq(x,c)		(TOUCHAR(x) == (unsigned char) (c))
+
+#define COPYCHAR(d,s)	memcpy(d, s, pg_mblen(s))
+
+extern int	t_isdigit(const char *ptr);
+extern int	t_isspace(const char *ptr);
+extern int	t_isalpha(const char *ptr);
+extern int	t_isprint(const char *ptr);
+
+extern char *lowerstr(const char *str);
+extern char *lowerstr_with_len(const char *str, int len);
+
+extern bool tsearch_readline_begin(tsearch_readline_state *stp,
+								   const char *filename);
+extern char *tsearch_readline(tsearch_readline_state *stp);
+extern void tsearch_readline_end(tsearch_readline_state *stp);
+
+#endif							/* __TSLOCALE_H__ */

data/ext/yb_ddl_parser/vendor/yugabyte-postgres/src/include/tsearch/ts_public.h

@@ -0,0 +1,159 @@
+/*-------------------------------------------------------------------------
+ *
+ * ts_public.h
+ *	  Public interface to various tsearch modules, such as
+ *	  parsers and dictionaries.
+ *
+ * Copyright (c) 1998-2022, PostgreSQL Global Development Group
+ *
+ * src/include/tsearch/ts_public.h
+ *
+ *-------------------------------------------------------------------------
+ */
+#ifndef _PG_TS_PUBLIC_H_
+#define _PG_TS_PUBLIC_H_
+
+#include "tsearch/ts_type.h"
+
+/*
+ * Parser's framework
+ */
+
+/*
+ * returning type for prslextype method of parser
+ */
+typedef struct
+{
+	int			lexid;
+	char	   *alias;
+	char	   *descr;
+} LexDescr;
+
+/*
+ * Interface to headline generator (tsparser's prsheadline function)
+ *
+ * HeadlineParsedText describes the text that is to be highlighted.
+ * Some fields are passed from the core code to the prsheadline function,
+ * while others are output from the prsheadline function.
+ *
+ * The principal data is words[], an array of HeadlineWordEntry,
+ * one entry per token, of length curwords.
+ * The fields of HeadlineWordEntry are:
+ *
+ * in, selected, replace, skip: these flags are initially zero
+ * and may be set by the prsheadline function.  A consecutive group
+ * of tokens marked "in" form a "fragment" to be output.
+ * Such tokens may additionally be marked selected, replace, or skip
+ * to modify how they are shown.  (If you set more than one of those
+ * bits, you get an unspecified one of those behaviors.)
+ *
+ * type, len, pos, word: filled by core code to describe the token.
+ *
+ * item: if the token matches any operand of the tsquery of interest,
+ * a pointer to such an operand.  (If there are multiple matching
+ * operands, we generate extra copies of the HeadlineWordEntry to hold
+ * all the pointers.  The extras are marked with repeated = 1 and should
+ * be ignored except for checking the item pointer.)
+ */
+typedef struct
+{
+	uint32		selected:1,		/* token is to be highlighted */
+				in:1,			/* token is part of headline */
+				replace:1,		/* token is to be replaced with a space */
+				repeated:1,		/* duplicate entry to hold item pointer */
+				skip:1,			/* token is to be skipped (not output) */
+				unused:3,		/* available bits */
+				type:8,			/* parser's token category */
+				len:16;			/* length of token */
+	WordEntryPos pos;			/* position of token */
+	char	   *word;			/* text of token (not null-terminated) */
+	QueryOperand *item;			/* a matching query operand, or NULL if none */
+} HeadlineWordEntry;
+
+typedef struct
+{
+	/* Fields filled by core code before calling prsheadline function: */
+	HeadlineWordEntry *words;
+	int32		lenwords;		/* allocated length of words[] */
+	int32		curwords;		/* current number of valid entries */
+	int32		vectorpos;		/* used by ts_parse.c in filling pos fields */
+
+	/* The prsheadline function must fill these fields: */
+	/* Strings for marking selected tokens and separating fragments: */
+	char	   *startsel;		/* palloc'd strings */
+	char	   *stopsel;
+	char	   *fragdelim;
+	int16		startsellen;	/* lengths of strings */
+	int16		stopsellen;
+	int16		fragdelimlen;
+} HeadlineParsedText;
+
+/*
+ * Common useful things for tsearch subsystem
+ */
+extern char *get_tsearch_config_filename(const char *basename,
+										 const char *extension);
+
+/*
+ * Often useful stopword list management
+ */
+typedef struct
+{
+	int			len;
+	char	  **stop;
+} StopList;
+
+extern void readstoplist(const char *fname, StopList *s,
+						 char *(*wordop) (const char *));
+extern bool searchstoplist(StopList *s, char *key);
+
+/*
+ * Interface with dictionaries
+ */
+
+/* return struct for any lexize function */
+typedef struct
+{
+	/*----------
+	 * Number of current variant of split word.  For example the Norwegian
+	 * word 'fotballklubber' has two variants to split: ( fotball, klubb )
+	 * and ( fot, ball, klubb ). So, dictionary should return:
+	 *
+	 * nvariant    lexeme
+	 *	   1	   fotball
+	 *	   1	   klubb
+	 *	   2	   fot
+	 *	   2	   ball
+	 *	   2	   klubb
+	 *
+	 * In general, a TSLexeme will be considered to belong to the same split
+	 * variant as the previous one if they have the same nvariant value.
+	 * The exact values don't matter, only changes from one lexeme to next.
+	 *----------
+	 */
+	uint16		nvariant;
+
+	uint16		flags;			/* See flag bits below */
+
+	char	   *lexeme;			/* C string */
+} TSLexeme;
+
+/* Flag bits that can appear in TSLexeme.flags */
+#define TSL_ADDPOS		0x01
+#define TSL_PREFIX		0x02
+#define TSL_FILTER		0x04
+
+/*
+ * Struct for supporting complex dictionaries like thesaurus.
+ * 4th argument for dictlexize method is a pointer to this
+ */
+typedef struct
+{
+	bool		isend;			/* in: marks for lexize_info about text end is
+								 * reached */
+	bool		getnext;		/* out: dict wants next lexeme */
+	void	   *private_state;	/* internal dict state between calls with
+								 * getnext == true */
+} DictSubState;
+
+#endif							/* _PG_TS_PUBLIC_H_ */