[thirdparty/git.git] / grep.h

#ifndef GREP_H
#define GREP_H
#include "color.h"
#ifdef USE_LIBPCRE2
#define PCRE2_CODE_UNIT_WIDTH 8
#include <pcre2.h>
#if (PCRE2_MAJOR >= 10 && PCRE2_MINOR >= 36) || PCRE2_MAJOR >= 11
#define GIT_PCRE2_VERSION_10_36_OR_HIGHER
#endif
#if (PCRE2_MAJOR >= 10 && PCRE2_MINOR >= 34) || PCRE2_MAJOR >= 11
#define GIT_PCRE2_VERSION_10_34_OR_HIGHER
#endif
#else
typedef int pcre2_code;
typedef int pcre2_match_data;
typedef int pcre2_compile_context;
typedef int pcre2_general_context;
#endif
#ifndef PCRE2_MATCH_INVALID_UTF
/* PCRE2_MATCH_* dummy also with !USE_LIBPCRE2, for test-pcre2-config.c */
#define PCRE2_MATCH_INVALID_UTF 0
#endif
#include "thread-utils.h"
#include "userdiff.h"

struct repository;

enum grep_pat_token {
	GREP_PATTERN,
	GREP_PATTERN_HEAD,
	GREP_PATTERN_BODY,
	GREP_AND,
	GREP_OPEN_PAREN,
	GREP_CLOSE_PAREN,
	GREP_NOT,
	GREP_OR
};

enum grep_context {
	GREP_CONTEXT_HEAD,
	GREP_CONTEXT_BODY
};

enum grep_header_field {
	GREP_HEADER_FIELD_MIN = 0,
	GREP_HEADER_AUTHOR = GREP_HEADER_FIELD_MIN,
	GREP_HEADER_COMMITTER,
	GREP_HEADER_REFLOG,

	/* Must be at the end of the enum */
	GREP_HEADER_FIELD_MAX
};

enum grep_color {
	GREP_COLOR_CONTEXT,
	GREP_COLOR_FILENAME,
	GREP_COLOR_FUNCTION,
	GREP_COLOR_LINENO,
	GREP_COLOR_COLUMNNO,
	GREP_COLOR_MATCH_CONTEXT,
	GREP_COLOR_MATCH_SELECTED,
	GREP_COLOR_SELECTED,
	GREP_COLOR_SEP,
	NR_GREP_COLORS
};

struct grep_pat {
	struct grep_pat *next;
	const char *origin;
	int no;
	enum grep_pat_token token;
	char *pattern;
	size_t patternlen;
	enum grep_header_field field;
	regex_t regexp;
	pcre2_code *pcre2_pattern;
	pcre2_match_data *pcre2_match_data;
	pcre2_compile_context *pcre2_compile_context;
	pcre2_general_context *pcre2_general_context;
	const uint8_t *pcre2_tables;
	uint32_t pcre2_jit_on;
	unsigned fixed:1;
	unsigned is_fixed:1;
	unsigned ignore_case:1;
	unsigned word_regexp:1;
};

enum grep_expr_node {
	GREP_NODE_ATOM,
	GREP_NODE_NOT,
	GREP_NODE_AND,
	GREP_NODE_TRUE,
	GREP_NODE_OR
};

enum grep_pattern_type {
	GREP_PATTERN_TYPE_UNSPECIFIED = 0,
	GREP_PATTERN_TYPE_BRE,
	GREP_PATTERN_TYPE_ERE,
	GREP_PATTERN_TYPE_FIXED,
	GREP_PATTERN_TYPE_PCRE
};

struct grep_expr {
	enum grep_expr_node node;
	unsigned hit;
	union {
		struct grep_pat *atom;
		struct grep_expr *unary;
		struct {
			struct grep_expr *left;
			struct grep_expr *right;
		} binary;
	} u;
};

struct grep_opt {
	struct grep_pat *pattern_list;
	struct grep_pat **pattern_tail;
	struct grep_pat *header_list;
	struct grep_pat **header_tail;
	struct grep_expr *pattern_expression;

	/*
	 * NEEDSWORK: See if we can remove this field, because the repository
	 * should probably be per-source. That is, grep.c functions using this
	 * field should probably start using "repo" in "struct grep_source"
	 * instead.
	 *
	 * This is potentially the cause of at least one bug - "git grep"
	 * ignoring the textconv attributes from submodules. See [1] for more
	 * information.
	 * [1] https://lore.kernel.org/git/CAHd-oW5iEQarYVxEXoTG-ua2zdoybTrSjCBKtO0YT292fm0NQQ@mail.gmail.com/
	 */
	struct repository *repo;

	const char *prefix;
	int prefix_length;
	regex_t regexp;
	int linenum;
	int columnnum;
	int invert;
	int ignore_case;
	int status_only;
	int name_only;
	int unmatch_name_only;
	int count;
	int word_regexp;
	int fixed;
	int all_match;
#define GREP_BINARY_DEFAULT	0
#define GREP_BINARY_NOMATCH	1
#define GREP_BINARY_TEXT	2
	int binary;
	int allow_textconv;
	int extended;
	int use_reflog_filter;
	int pcre2;
	int relative;
	int pathname;
	int null_following_name;
	int only_matching;
	int color;
	int max_depth;
	int funcname;
	int funcbody;
	int extended_regexp_option;
	int pattern_type_option;
	int ignore_locale;
	char colors[NR_GREP_COLORS][COLOR_MAXLEN];
	unsigned pre_context;
	unsigned post_context;
	unsigned last_shown;
	int show_hunk_mark;
	int file_break;
	int heading;
	void *priv;

	void (*output)(struct grep_opt *opt, const void *data, size_t size);
	void *output_priv;
};

int grep_config(const char *var, const char *value, void *);
void grep_init(struct grep_opt *, struct repository *repo, const char *prefix);
void grep_commit_pattern_type(enum grep_pattern_type, struct grep_opt *opt);

void append_grep_pat(struct grep_opt *opt, const char *pat, size_t patlen, const char *origin, int no, enum grep_pat_token t);
void append_grep_pattern(struct grep_opt *opt, const char *pat, const char *origin, int no, enum grep_pat_token t);
void append_header_grep_pattern(struct grep_opt *, enum grep_header_field, const char *);
void compile_grep_patterns(struct grep_opt *opt);
void free_grep_patterns(struct grep_opt *opt);
int grep_buffer(struct grep_opt *opt, char *buf, unsigned long size);

struct grep_source {
	char *name;

	enum grep_source_type {
		GREP_SOURCE_OID,
		GREP_SOURCE_FILE,
		GREP_SOURCE_BUF,
	} type;
	void *identifier;
	struct repository *repo; /* if GREP_SOURCE_OID */

	char *buf;
	unsigned long size;

	char *path; /* for attribute lookups */
	struct userdiff_driver *driver;
};

void grep_source_init_file(struct grep_source *gs, const char *name,
			   const char *path);
void grep_source_init_oid(struct grep_source *gs, const char *name,
			  const char *path, const struct object_id *oid,
			  struct repository *repo);
void grep_source_clear_data(struct grep_source *gs);
void grep_source_clear(struct grep_source *gs);
void grep_source_load_driver(struct grep_source *gs,
			     struct index_state *istate);


int grep_source(struct grep_opt *opt, struct grep_source *gs);

struct grep_opt *grep_opt_dup(const struct grep_opt *opt);
int grep_threads_ok(const struct grep_opt *opt);

/*
 * Mutex used around access to the attributes machinery if
 * opt->use_threads.  Must be initialized/destroyed by callers!
 */
extern int grep_use_locks;
extern pthread_mutex_t grep_attr_mutex;

#endif
Commit	Line	Data
	1	#ifndef GREP_H
	2	#define GREP_H
	3	#include "color.h"
	4	#ifdef USE_LIBPCRE2
	5	#define PCRE2_CODE_UNIT_WIDTH 8
	6	#include <pcre2.h>
	7	#if (PCRE2_MAJOR >= 10 && PCRE2_MINOR >= 36) \|\| PCRE2_MAJOR >= 11
	8	#define GIT_PCRE2_VERSION_10_36_OR_HIGHER
	9	#endif
	10	#if (PCRE2_MAJOR >= 10 && PCRE2_MINOR >= 34) \|\| PCRE2_MAJOR >= 11
	11	#define GIT_PCRE2_VERSION_10_34_OR_HIGHER
	12	#endif
	13	#else
	14	typedef int pcre2_code;
	15	typedef int pcre2_match_data;
	16	typedef int pcre2_compile_context;
	17	typedef int pcre2_general_context;
	18	#endif
	19	#ifndef PCRE2_MATCH_INVALID_UTF
	20	/* PCRE2_MATCH_* dummy also with !USE_LIBPCRE2, for test-pcre2-config.c */
	21	#define PCRE2_MATCH_INVALID_UTF 0
	22	#endif
	23	#include "thread-utils.h"
	24	#include "userdiff.h"
	25
	26	struct repository;
	27
	28	enum grep_pat_token {
	29	GREP_PATTERN,
	30	GREP_PATTERN_HEAD,
	31	GREP_PATTERN_BODY,
	32	GREP_AND,
	33	GREP_OPEN_PAREN,
	34	GREP_CLOSE_PAREN,
	35	GREP_NOT,
	36	GREP_OR
	37	};
	38
	39	enum grep_context {
	40	GREP_CONTEXT_HEAD,
	41	GREP_CONTEXT_BODY
	42	};
	43
	44	enum grep_header_field {
	45	GREP_HEADER_FIELD_MIN = 0,
	46	GREP_HEADER_AUTHOR = GREP_HEADER_FIELD_MIN,
	47	GREP_HEADER_COMMITTER,
	48	GREP_HEADER_REFLOG,
	49
	50	/* Must be at the end of the enum */
	51	GREP_HEADER_FIELD_MAX
	52	};
	53
	54	enum grep_color {
	55	GREP_COLOR_CONTEXT,
	56	GREP_COLOR_FILENAME,
	57	GREP_COLOR_FUNCTION,
	58	GREP_COLOR_LINENO,
	59	GREP_COLOR_COLUMNNO,
	60	GREP_COLOR_MATCH_CONTEXT,
	61	GREP_COLOR_MATCH_SELECTED,
	62	GREP_COLOR_SELECTED,
	63	GREP_COLOR_SEP,
	64	NR_GREP_COLORS
	65	};
	66
	67	struct grep_pat {
	68	struct grep_pat *next;
	69	const char *origin;
	70	int no;
	71	enum grep_pat_token token;
	72	char *pattern;
	73	size_t patternlen;
	74	enum grep_header_field field;
	75	regex_t regexp;
	76	pcre2_code *pcre2_pattern;
	77	pcre2_match_data *pcre2_match_data;
	78	pcre2_compile_context *pcre2_compile_context;
	79	pcre2_general_context *pcre2_general_context;
	80	const uint8_t *pcre2_tables;
	81	uint32_t pcre2_jit_on;
	82	unsigned fixed:1;
	83	unsigned is_fixed:1;
	84	unsigned ignore_case:1;
	85	unsigned word_regexp:1;
	86	};
	87
	88	enum grep_expr_node {
	89	GREP_NODE_ATOM,
	90	GREP_NODE_NOT,
	91	GREP_NODE_AND,
	92	GREP_NODE_TRUE,
	93	GREP_NODE_OR
	94	};
	95
	96	enum grep_pattern_type {
	97	GREP_PATTERN_TYPE_UNSPECIFIED = 0,
	98	GREP_PATTERN_TYPE_BRE,
	99	GREP_PATTERN_TYPE_ERE,
	100	GREP_PATTERN_TYPE_FIXED,
	101	GREP_PATTERN_TYPE_PCRE
	102	};
	103
	104	struct grep_expr {
	105	enum grep_expr_node node;
	106	unsigned hit;
	107	union {
	108	struct grep_pat *atom;
	109	struct grep_expr *unary;
	110	struct {
	111	struct grep_expr *left;
	112	struct grep_expr *right;
	113	} binary;
	114	} u;
	115	};
	116
	117	struct grep_opt {
	118	struct grep_pat *pattern_list;
	119	struct grep_pat **pattern_tail;
	120	struct grep_pat *header_list;
	121	struct grep_pat **header_tail;
	122	struct grep_expr *pattern_expression;
	123
	124	/*
	125	* NEEDSWORK: See if we can remove this field, because the repository
	126	* should probably be per-source. That is, grep.c functions using this
	127	* field should probably start using "repo" in "struct grep_source"
	128	* instead.
	129	*
	130	* This is potentially the cause of at least one bug - "git grep"
	131	* ignoring the textconv attributes from submodules. See [1] for more
	132	* information.
	133	* [1] https://lore.kernel.org/git/CAHd-oW5iEQarYVxEXoTG-ua2zdoybTrSjCBKtO0YT292fm0NQQ@mail.gmail.com/
	134	*/
	135	struct repository *repo;
	136
	137	const char *prefix;
	138	int prefix_length;
	139	regex_t regexp;
	140	int linenum;
	141	int columnnum;
	142	int invert;
	143	int ignore_case;
	144	int status_only;
	145	int name_only;
	146	int unmatch_name_only;
	147	int count;
	148	int word_regexp;
	149	int fixed;
	150	int all_match;
	151	#define GREP_BINARY_DEFAULT 0
	152	#define GREP_BINARY_NOMATCH 1
	153	#define GREP_BINARY_TEXT 2
	154	int binary;
	155	int allow_textconv;
	156	int extended;
	157	int use_reflog_filter;
	158	int pcre2;
	159	int relative;
	160	int pathname;
	161	int null_following_name;
	162	int only_matching;
	163	int color;
	164	int max_depth;
	165	int funcname;
	166	int funcbody;
	167	int extended_regexp_option;
	168	int pattern_type_option;
	169	int ignore_locale;
	170	char colors[NR_GREP_COLORS][COLOR_MAXLEN];
	171	unsigned pre_context;
	172	unsigned post_context;
	173	unsigned last_shown;
	174	int show_hunk_mark;
	175	int file_break;
	176	int heading;
	177	void *priv;
	178
	179	void (output)(struct grep_opt opt, const void *data, size_t size);
	180	void *output_priv;
	181	};
	182
	183	int grep_config(const char var, const char value, void *);
	184	void grep_init(struct grep_opt , struct repository repo, const char *prefix);
	185	void grep_commit_pattern_type(enum grep_pattern_type, struct grep_opt *opt);
	186
	187	void append_grep_pat(struct grep_opt opt, const char pat, size_t patlen, const char *origin, int no, enum grep_pat_token t);
	188	void append_grep_pattern(struct grep_opt opt, const char pat, const char *origin, int no, enum grep_pat_token t);
	189	void append_header_grep_pattern(struct grep_opt , enum grep_header_field, const char );
	190	void compile_grep_patterns(struct grep_opt *opt);
	191	void free_grep_patterns(struct grep_opt *opt);
	192	int grep_buffer(struct grep_opt opt, char buf, unsigned long size);
	193
	194	struct grep_source {
	195	char *name;
	196
	197	enum grep_source_type {
	198	GREP_SOURCE_OID,
	199	GREP_SOURCE_FILE,
	200	GREP_SOURCE_BUF,
	201	} type;
	202	void *identifier;
	203	struct repository repo; / if GREP_SOURCE_OID */
	204
	205	char *buf;
	206	unsigned long size;
	207
	208	char path; / for attribute lookups */
	209	struct userdiff_driver *driver;
	210	};
	211
	212	void grep_source_init_file(struct grep_source gs, const char name,
	213	const char *path);
	214	void grep_source_init_oid(struct grep_source gs, const char name,
	215	const char path, const struct object_id oid,
	216	struct repository *repo);
	217	void grep_source_clear_data(struct grep_source *gs);
	218	void grep_source_clear(struct grep_source *gs);
	219	void grep_source_load_driver(struct grep_source *gs,
	220	struct index_state *istate);
	221
	222
	223	int grep_source(struct grep_opt opt, struct grep_source gs);
	224
	225	struct grep_opt grep_opt_dup(const struct grep_opt opt);
	226	int grep_threads_ok(const struct grep_opt *opt);
	227
	228	/*
	229	* Mutex used around access to the attributes machinery if
	230	* opt->use_threads. Must be initialized/destroyed by callers!
	231	*/
	232	extern int grep_use_locks;
	233	extern pthread_mutex_t grep_attr_mutex;
	234
	235	#endif