Index: head/usr.bin/indent/args.c =================================================================== --- head/usr.bin/indent/args.c (revision 334555) +++ head/usr.bin/indent/args.c (revision 334556) @@ -1,358 +1,356 @@ /*- * SPDX-License-Identifier: BSD-4-Clause * * Copyright (c) 1985 Sun Microsystems, Inc. * Copyright (c) 1980, 1993 * The Regents of the University of California. All rights reserved. * All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. All advertising materials mentioning features or use of this software * must display the following acknowledgement: * This product includes software developed by the University of * California, Berkeley and its contributors. * 4. Neither the name of the University nor the names of its contributors * may be used to endorse or promote products derived from this software * without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. */ #if 0 #ifndef lint static char sccsid[] = "@(#)args.c 8.1 (Berkeley) 6/6/93"; #endif /* not lint */ #endif #include __FBSDID("$FreeBSD$"); /* * Argument scanning and profile reading code. Default parameters are set * here as well. */ #include #include #include #include #include #include #include "indent_globs.h" #include "indent.h" /* profile types */ #define PRO_SPECIAL 1 /* special case */ #define PRO_BOOL 2 /* boolean */ #define PRO_INT 3 /* integer */ #define PRO_FONT 4 /* troff font */ /* profile specials for booleans */ #define ON 1 /* turn it on */ #define OFF 0 /* turn it off */ /* profile specials for specials */ #define IGN 1 /* ignore it */ #define CLI 2 /* case label indent (float) */ #define STDIN 3 /* use stdin */ #define KEY 4 /* type (keyword) */ static void scan_profile(FILE *); #define KEY_FILE 5 /* only used for args */ const char *option_source = "?"; void add_typedefs_from_file(const char *str); /* * N.B.: because of the way the table here is scanned, options whose names are * substrings of other options must occur later; that is, with -lp vs -l, -lp * must be first. Also, while (most) booleans occur more than once, the last * default value is the one actually assigned. */ struct pro { const char *p_name; /* name, e.g. -bl, -cli */ int p_type; /* type (int, bool, special) */ int p_default; /* the default value (if int) */ int p_special; /* depends on type */ int *p_obj; /* the associated variable */ } pro[] = { {"T", PRO_SPECIAL, 0, KEY, 0}, {"U", PRO_SPECIAL, 0, KEY_FILE, 0}, {"P", PRO_SPECIAL, 0, IGN, 0}, {"bacc", PRO_BOOL, false, ON, &blanklines_around_conditional_compilation}, {"badp", PRO_BOOL, false, ON, &blanklines_after_declarations_at_proctop}, {"bad", PRO_BOOL, false, ON, &blanklines_after_declarations}, {"bap", PRO_BOOL, false, ON, &blanklines_after_procs}, {"bbb", PRO_BOOL, false, ON, &blanklines_before_blockcomments}, {"bc", PRO_BOOL, true, OFF, &ps.leave_comma}, {"bl", PRO_BOOL, true, OFF, &btype_2}, {"br", PRO_BOOL, true, ON, &btype_2}, {"bs", PRO_BOOL, false, ON, &Bill_Shannon}, {"cdb", PRO_BOOL, true, ON, &comment_delimiter_on_blankline}, {"cd", PRO_INT, 0, 0, &ps.decl_com_ind}, {"ce", PRO_BOOL, true, ON, &cuddle_else}, {"ci", PRO_INT, 0, 0, &continuation_indent}, {"cli", PRO_SPECIAL, 0, CLI, 0}, {"c", PRO_INT, 33, 0, &ps.com_ind}, {"di", PRO_INT, 16, 0, &ps.decl_indent}, {"dj", PRO_BOOL, false, ON, &ps.ljust_decl}, {"d", PRO_INT, 0, 0, &ps.unindent_displace}, {"eei", PRO_BOOL, false, ON, &extra_expression_indent}, {"ei", PRO_BOOL, true, ON, &ps.else_if}, {"fbc", PRO_FONT, 0, 0, (int *) &blkcomf}, {"fbs", PRO_BOOL, true, ON, &function_brace_split}, {"fbx", PRO_FONT, 0, 0, (int *) &boxcomf}, {"fb", PRO_FONT, 0, 0, (int *) &bodyf}, {"fc1", PRO_BOOL, true, ON, &format_col1_comments}, {"fcb", PRO_BOOL, true, ON, &format_block_comments}, {"fc", PRO_FONT, 0, 0, (int *) &scomf}, {"fk", PRO_FONT, 0, 0, (int *) &keywordf}, {"fs", PRO_FONT, 0, 0, (int *) &stringf}, {"ip", PRO_BOOL, true, ON, &ps.indent_parameters}, {"i", PRO_INT, 8, 0, &ps.ind_size}, {"lc", PRO_INT, 0, 0, &block_comment_max_col}, {"ldi", PRO_INT, -1, 0, &ps.local_decl_indent}, {"lp", PRO_BOOL, true, ON, &lineup_to_parens}, {"l", PRO_INT, 78, 0, &max_col}, {"nbacc", PRO_BOOL, false, OFF, &blanklines_around_conditional_compilation}, {"nbadp", PRO_BOOL, false, OFF, &blanklines_after_declarations_at_proctop}, {"nbad", PRO_BOOL, false, OFF, &blanklines_after_declarations}, {"nbap", PRO_BOOL, false, OFF, &blanklines_after_procs}, {"nbbb", PRO_BOOL, false, OFF, &blanklines_before_blockcomments}, {"nbc", PRO_BOOL, true, ON, &ps.leave_comma}, {"nbs", PRO_BOOL, false, OFF, &Bill_Shannon}, {"ncdb", PRO_BOOL, true, OFF, &comment_delimiter_on_blankline}, {"nce", PRO_BOOL, true, OFF, &cuddle_else}, {"ndj", PRO_BOOL, false, OFF, &ps.ljust_decl}, {"neei", PRO_BOOL, false, OFF, &extra_expression_indent}, {"nei", PRO_BOOL, true, OFF, &ps.else_if}, {"nfbs", PRO_BOOL, true, OFF, &function_brace_split}, {"nfc1", PRO_BOOL, true, OFF, &format_col1_comments}, {"nfcb", PRO_BOOL, true, OFF, &format_block_comments}, {"nip", PRO_BOOL, true, OFF, &ps.indent_parameters}, {"nlp", PRO_BOOL, true, OFF, &lineup_to_parens}, {"npcs", PRO_BOOL, false, OFF, &proc_calls_space}, {"npro", PRO_SPECIAL, 0, IGN, 0}, {"npsl", PRO_BOOL, true, OFF, &procnames_start_line}, - {"nps", PRO_BOOL, false, OFF, &pointer_as_binop}, {"nsac", PRO_BOOL, false, OFF, &space_after_cast}, {"nsc", PRO_BOOL, true, OFF, &star_comment_cont}, {"nsob", PRO_BOOL, false, OFF, &swallow_optional_blanklines}, {"nut", PRO_BOOL, true, OFF, &use_tabs}, {"nv", PRO_BOOL, false, OFF, &verbose}, {"pcs", PRO_BOOL, false, ON, &proc_calls_space}, {"psl", PRO_BOOL, true, ON, &procnames_start_line}, - {"ps", PRO_BOOL, false, ON, &pointer_as_binop}, {"sac", PRO_BOOL, false, ON, &space_after_cast}, {"sc", PRO_BOOL, true, ON, &star_comment_cont}, {"sob", PRO_BOOL, false, ON, &swallow_optional_blanklines}, {"st", PRO_SPECIAL, 0, STDIN, 0}, {"ta", PRO_BOOL, false, ON, &auto_typedefs}, {"ts", PRO_INT, 8, 0, &tabsize}, {"troff", PRO_BOOL, false, ON, &troff}, {"ut", PRO_BOOL, true, ON, &use_tabs}, {"v", PRO_BOOL, false, ON, &verbose}, /* whew! */ {0, 0, 0, 0, 0} }; /* * set_profile reads $HOME/.indent.pro and ./.indent.pro and handles arguments * given in these files. */ void set_profile(const char *profile_name) { FILE *f; char fname[PATH_MAX]; static char prof[] = ".indent.pro"; if (profile_name == NULL) snprintf(fname, sizeof(fname), "%s/%s", getenv("HOME"), prof); else snprintf(fname, sizeof(fname), "%s", profile_name + 2); if ((f = fopen(option_source = fname, "r")) != NULL) { scan_profile(f); (void) fclose(f); } if ((f = fopen(option_source = prof, "r")) != NULL) { scan_profile(f); (void) fclose(f); } option_source = "Command line"; } static void scan_profile(FILE *f) { int comment, i; char *p; char buf[BUFSIZ]; while (1) { p = buf; comment = 0; while ((i = getc(f)) != EOF) { if (i == '*' && !comment && p > buf && p[-1] == '/') { comment = p - buf; *p++ = i; } else if (i == '/' && comment && p > buf && p[-1] == '*') { p = buf + comment - 1; comment = 0; } else if (isspace(i)) { if (p > buf && !comment) break; } else { *p++ = i; } } if (p != buf) { *p++ = 0; if (verbose) printf("profile: %s\n", buf); set_option(buf); } else if (i == EOF) return; } } static const char * eqin(const char *s1, const char *s2) { while (*s1) { if (*s1++ != *s2++) return (NULL); } return (s2); } /* * Set the defaults. */ void set_defaults(void) { struct pro *p; /* * Because ps.case_indent is a float, we can't initialize it from the * table: */ ps.case_indent = 0.0; /* -cli0.0 */ for (p = pro; p->p_name; p++) if (p->p_type != PRO_SPECIAL && p->p_type != PRO_FONT) *p->p_obj = p->p_default; } void set_option(char *arg) { struct pro *p; const char *param_start; arg++; /* ignore leading "-" */ for (p = pro; p->p_name; p++) if (*p->p_name == *arg && (param_start = eqin(p->p_name, arg)) != NULL) goto found; errx(1, "%s: unknown parameter \"%s\"", option_source, arg - 1); found: switch (p->p_type) { case PRO_SPECIAL: switch (p->p_special) { case IGN: break; case CLI: if (*param_start == 0) goto need_param; ps.case_indent = atof(param_start); break; case STDIN: if (input == NULL) input = stdin; if (output == NULL) output = stdout; break; case KEY: if (*param_start == 0) goto need_param; add_typename(param_start); break; case KEY_FILE: if (*param_start == 0) goto need_param; add_typedefs_from_file(param_start); break; default: errx(1, "set_option: internal error: p_special %d", p->p_special); } break; case PRO_BOOL: if (p->p_special == OFF) *p->p_obj = false; else *p->p_obj = true; break; case PRO_INT: if (!isdigit(*param_start)) { need_param: errx(1, "%s: ``%s'' requires a parameter", option_source, p->p_name); } *p->p_obj = atoi(param_start); break; case PRO_FONT: parsefont((struct fstate *) p->p_obj, param_start); break; default: errx(1, "set_option: internal error: p_type %d", p->p_type); } } void add_typedefs_from_file(const char *str) { FILE *file; char line[BUFSIZ]; if ((file = fopen(str, "r")) == NULL) { fprintf(stderr, "indent: cannot open file %s\n", str); exit(1); } while ((fgets(line, BUFSIZ, file)) != NULL) { /* Remove trailing whitespace */ line[strcspn(line, " \t\n\r")] = '\0'; add_typename(line); } fclose(file); } Index: head/usr.bin/indent/indent_globs.h =================================================================== --- head/usr.bin/indent/indent_globs.h (revision 334555) +++ head/usr.bin/indent/indent_globs.h (revision 334556) @@ -1,333 +1,332 @@ /*- * SPDX-License-Identifier: BSD-4-Clause * * Copyright (c) 1985 Sun Microsystems, Inc. * Copyright (c) 1980, 1993 * The Regents of the University of California. All rights reserved. * All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. All advertising materials mentioning features or use of this software * must display the following acknowledgement: * This product includes software developed by the University of * California, Berkeley and its contributors. * 4. Neither the name of the University nor the names of its contributors * may be used to endorse or promote products derived from this software * without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. * * @(#)indent_globs.h 8.1 (Berkeley) 6/6/93 * $FreeBSD$ */ #define BACKSLASH '\\' #define bufsize 200 /* size of internal buffers */ #define sc_size 5000 /* size of save_com buffer */ #define label_offset 2 /* number of levels a label is placed to left * of code */ #define false 0 #define true 1 FILE *input; /* the fid for the input file */ FILE *output; /* the output file */ #define CHECK_SIZE_CODE \ if (e_code >= l_code) { \ int nsize = l_code-s_code+400; \ int code_len = e_code-s_code; \ codebuf = (char *) realloc(codebuf, nsize); \ if (codebuf == NULL) \ err(1, NULL); \ e_code = codebuf + code_len + 1; \ l_code = codebuf + nsize - 5; \ s_code = codebuf + 1; \ } #define CHECK_SIZE_COM \ if (e_com >= l_com) { \ int nsize = l_com-s_com+400; \ int com_len = e_com - s_com; \ int blank_pos = last_bl - s_com; \ combuf = (char *) realloc(combuf, nsize); \ if (combuf == NULL) \ err(1, NULL); \ e_com = combuf + com_len + 1; \ last_bl = combuf + blank_pos + 1; \ l_com = combuf + nsize - 5; \ s_com = combuf + 1; \ } #define CHECK_SIZE_LAB \ if (e_lab >= l_lab) { \ int nsize = l_lab-s_lab+400; \ int label_len = e_lab - s_lab; \ labbuf = (char *) realloc(labbuf, nsize); \ if (labbuf == NULL) \ err(1, NULL); \ e_lab = labbuf + label_len + 1; \ l_lab = labbuf + nsize - 5; \ s_lab = labbuf + 1; \ } #define CHECK_SIZE_TOKEN \ if (e_token >= l_token) { \ int nsize = l_token-s_token+400; \ int token_len = e_token - s_token; \ tokenbuf = (char *) realloc(tokenbuf, nsize); \ if (tokenbuf == NULL) \ err(1, NULL); \ e_token = tokenbuf + token_len + 1; \ l_token = tokenbuf + nsize - 5; \ s_token = tokenbuf + 1; \ } char *labbuf; /* buffer for label */ char *s_lab; /* start ... */ char *e_lab; /* .. and end of stored label */ char *l_lab; /* limit of label buffer */ char *codebuf; /* buffer for code section */ char *s_code; /* start ... */ char *e_code; /* .. and end of stored code */ char *l_code; /* limit of code section */ char *combuf; /* buffer for comments */ char *s_com; /* start ... */ char *e_com; /* ... and end of stored comments */ char *l_com; /* limit of comment buffer */ #define token s_token char *tokenbuf; /* the last token scanned */ char *s_token; char *e_token; char *l_token; char *in_buffer; /* input buffer */ char *in_buffer_limit; /* the end of the input buffer */ char *buf_ptr; /* ptr to next character to be taken from * in_buffer */ char *buf_end; /* ptr to first after last char in in_buffer */ char save_com[sc_size]; /* input text is saved here when looking for * the brace after an if, while, etc */ char *sc_end; /* pointer into save_com buffer */ char *bp_save; /* saved value of buf_ptr when taking input * from save_com */ char *be_save; /* similarly saved value of buf_end */ int found_err; -int pointer_as_binop; int blanklines_after_declarations; int blanklines_before_blockcomments; int blanklines_after_procs; int blanklines_around_conditional_compilation; int swallow_optional_blanklines; int n_real_blanklines; int prefix_blankline_requested; int postfix_blankline_requested; int break_comma; /* when true and not in parens, break after a * comma */ int btype_2; /* when true, brace should be on same line as * if, while, etc */ float case_ind; /* indentation level to be used for a "case * n:" */ int code_lines; /* count of lines with code */ int had_eof; /* set to true when input is exhausted */ int line_no; /* the current line number. */ int max_col; /* the maximum allowable line length */ int verbose; /* when true, non-essential error messages are * printed */ int cuddle_else; /* true if else should cuddle up to '}' */ int star_comment_cont; /* true iff comment continuation lines should * have stars at the beginning of each line. */ int comment_delimiter_on_blankline; int troff; /* true iff were generating troff input */ int procnames_start_line; /* if true, the names of procedures * being defined get placed in column * 1 (ie. a newline is placed between * the type of the procedure and its * name) */ int proc_calls_space; /* If true, procedure calls look like: * foo (bar) rather than foo(bar) */ int format_block_comments; /* true if comments beginning with * `/ * \n' are to be reformatted */ int format_col1_comments; /* If comments which start in column 1 * are to be magically reformatted * (just like comments that begin in * later columns) */ int inhibit_formatting; /* true if INDENT OFF is in effect */ int suppress_blanklines;/* set iff following blanklines should be * suppressed */ int continuation_indent;/* set to the indentation between the edge of * code and continuation lines */ int lineup_to_parens; /* if true, continued code within parens will * be lined up to the open paren */ int Bill_Shannon; /* true iff a blank should always be inserted * after sizeof */ int blanklines_after_declarations_at_proctop; /* This is vaguely * similar to * blanklines_after_decla * rations except that * it only applies to * the first set of * declarations in a * procedure (just after * the first '{') and it * causes a blank line * to be generated even * if there are no * declarations */ int block_comment_max_col; int extra_expression_indent; /* true if continuation lines from the * expression part of "if(e)", * "while(e)", "for(e;e;e)" should be * indented an extra tab stop so that * they don't conflict with the code * that follows */ int function_brace_split; /* split function declaration and * brace onto separate lines */ int use_tabs; /* set true to use tabs for spacing, * false uses all spaces */ int auto_typedefs; /* set true to recognize identifiers * ending in "_t" like typedefs */ int space_after_cast; /* "b = (int) a" vs "b = (int)a" */ int tabsize; /* the size of a tab */ /* -troff font state information */ struct fstate { char font[4]; char size; int allcaps:1; } __aligned(sizeof(int)); char *chfont(struct fstate *, struct fstate *, char *); struct fstate keywordf, /* keyword font */ stringf, /* string font */ boxcomf, /* Box comment font */ blkcomf, /* Block comment font */ scomf, /* Same line comment font */ bodyf; /* major body font */ #define STACKSIZE 256 struct parser_state { int last_token; struct fstate cfont; /* Current font */ int p_stack[STACKSIZE]; /* this is the parsers stack */ int il[STACKSIZE]; /* this stack stores indentation levels */ float cstk[STACKSIZE];/* used to store case stmt indentation levels */ int box_com; /* set to true when we are in a "boxed" * comment. In that case, the first non-blank * char should be lined up with the / in / followed by * */ int comment_delta, n_comment_delta; int cast_mask; /* indicates which close parens potentially * close off casts */ int not_cast_mask; /* indicates which close parens definitely * close off something else than casts */ int block_init; /* true iff inside a block initialization */ int block_init_level; /* The level of brace nesting in an * initialization */ int last_nl; /* this is true if the last thing scanned was * a newline */ int in_or_st; /* Will be true iff there has been a * declarator (e.g. int or char) and no left * paren since the last semicolon. When true, * a '{' is starting a structure definition or * an initialization list */ int bl_line; /* set to 1 by dump_line if the line is blank */ int col_1; /* set to true if the last token started in * column 1 */ int com_col; /* this is the column in which the current * comment should start */ int com_ind; /* the column in which comments to the right * of code should start */ int com_lines; /* the number of lines with comments, set by * dump_line */ int dec_nest; /* current nesting level for structure or init */ int decl_com_ind; /* the column in which comments after * declarations should be put */ int decl_on_line; /* set to true if this line of code has part * of a declaration on it */ int i_l_follow; /* the level to which ind_level should be set * after the current line is printed */ int in_decl; /* set to true when we are in a declaration * stmt. The processing of braces is then * slightly different */ int in_stmt; /* set to 1 while in a stmt */ int ind_level; /* the current indentation level */ int ind_size; /* the size of one indentation level */ int ind_stmt; /* set to 1 if next line should have an extra * indentation level because we are in the * middle of a stmt */ int last_u_d; /* set to true after scanning a token which * forces a following operator to be unary */ int leave_comma; /* if true, never break declarations after * commas */ int ljust_decl; /* true if declarations should be left * justified */ int out_coms; /* the number of comments processed, set by * pr_comment */ int out_lines; /* the number of lines written, set by * dump_line */ int p_l_follow; /* used to remember how to indent following * statement */ int paren_level; /* parenthesization level. used to indent * within statements */ short paren_indents[20]; /* column positions of each paren */ int pcase; /* set to 1 if the current line label is a * case. It is printed differently from a * regular label */ int search_brace; /* set to true by parse when it is necessary * to buffer up all info up to the start of a * stmt after an if, while, etc */ int unindent_displace; /* comments not to the right of code * will be placed this many * indentation levels to the left of * code */ int use_ff; /* set to one if the current line should be * terminated with a form feed */ int want_blank; /* set to true when the following token should * be prefixed by a blank. (Said prefixing is * ignored in some cases.) */ int else_if; /* True iff else if pairs should be handled * specially */ int decl_indent; /* column to indent declared identifiers to */ int local_decl_indent; /* like decl_indent but for locals */ int keyword; /* the type of a keyword or 0 */ int dumped_decl_indent; float case_indent; /* The distance to indent case labels from the * switch statement */ int in_parameter_declaration; int indent_parameters; int tos; /* pointer to top of stack */ char procname[100]; /* The name of the current procedure */ int just_saw_decl; } ps; int ifdef_level; struct parser_state state_stack[5]; struct parser_state match_state[5]; Index: head/usr.bin/indent/lexi.c =================================================================== --- head/usr.bin/indent/lexi.c (revision 334555) +++ head/usr.bin/indent/lexi.c (revision 334556) @@ -1,675 +1,673 @@ /*- * SPDX-License-Identifier: BSD-4-Clause * * Copyright (c) 1985 Sun Microsystems, Inc. * Copyright (c) 1980, 1993 * The Regents of the University of California. All rights reserved. * All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. All advertising materials mentioning features or use of this software * must display the following acknowledgement: * This product includes software developed by the University of * California, Berkeley and its contributors. * 4. Neither the name of the University nor the names of its contributors * may be used to endorse or promote products derived from this software * without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. */ #if 0 #ifndef lint static char sccsid[] = "@(#)lexi.c 8.1 (Berkeley) 6/6/93"; #endif /* not lint */ #endif #include __FBSDID("$FreeBSD$"); /* * Here we have the token scanner for indent. It scans off one token and puts * it in the global variable "token". It returns a code, indicating the type * of token scanned. */ #include #include #include #include #include #include "indent_globs.h" #include "indent_codes.h" #include "indent.h" #define alphanum 1 #ifdef undef #define opchar 3 #endif struct templ { const char *rwd; int rwcode; }; /* * This table has to be sorted alphabetically, because it'll be used in binary * search. For the same reason, string must be the first thing in struct templ. */ struct templ specials[] = { {"auto", 10}, {"break", 9}, {"case", 8}, {"char", 4}, {"const", 4}, {"default", 8}, {"do", 6}, {"double", 4}, {"else", 6}, {"enum", 3}, {"extern", 10}, {"float", 4}, {"for", 5}, {"global", 4}, {"goto", 9}, {"if", 5}, {"int", 4}, {"long", 4}, {"offsetof", 1}, {"register", 10}, {"return", 9}, {"short", 4}, {"sizeof", 2}, {"static", 10}, {"struct", 3}, {"switch", 7}, {"typedef", 11}, {"union", 3}, {"unsigned", 4}, {"void", 4}, {"volatile", 4}, {"while", 5} }; const char **typenames; int typename_count; int typename_top = -1; char chartype[128] = { /* this is used to facilitate the decision of * what type (alphanumeric, operator) each * character is */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 0, 0, 1, 3, 3, 0, 0, 0, 3, 3, 0, 3, 0, 3, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 3, 3, 3, 3, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 3, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 3, 0, 3, 0 }; static int strcmp_type(const void *e1, const void *e2) { return (strcmp(e1, *(const char * const *)e2)); } int lexi(void) { int unary_delim; /* this is set to 1 if the current token * forces a following operator to be unary */ static int last_code; /* the last token type returned */ static int l_struct; /* set to 1 if the last token was 'struct' */ int code; /* internal code to be returned */ char qchar; /* the delimiter character for a string */ e_token = s_token; /* point to start of place to save token */ unary_delim = false; ps.col_1 = ps.last_nl; /* tell world that this token started in * column 1 iff the last thing scanned was nl */ ps.last_nl = false; while (*buf_ptr == ' ' || *buf_ptr == '\t') { /* get rid of blanks */ ps.col_1 = false; /* leading blanks imply token is not in column * 1 */ if (++buf_ptr >= buf_end) fill_buffer(); } /* Scan an alphanumeric token */ if (chartype[(int)*buf_ptr] == alphanum || (buf_ptr[0] == '.' && isdigit(buf_ptr[1]))) { /* * we have a character or number */ struct templ *p; if (isdigit(*buf_ptr) || (buf_ptr[0] == '.' && isdigit(buf_ptr[1]))) { enum base { BASE_2, BASE_8, BASE_10, BASE_16 }; int seendot = 0, seenexp = 0, seensfx = 0; enum base in_base = BASE_10; if (*buf_ptr == '0') { if (buf_ptr[1] == 'b' || buf_ptr[1] == 'B') in_base = BASE_2; else if (buf_ptr[1] == 'x' || buf_ptr[1] == 'X') in_base = BASE_16; else if (isdigit(buf_ptr[1])) in_base = BASE_8; } switch (in_base) { case BASE_2: *e_token++ = *buf_ptr++; *e_token++ = *buf_ptr++; while (*buf_ptr == '0' || *buf_ptr == '1') { CHECK_SIZE_TOKEN; *e_token++ = *buf_ptr++; } break; case BASE_8: *e_token++ = *buf_ptr++; while (*buf_ptr >= '0' && *buf_ptr <= '8') { CHECK_SIZE_TOKEN; *e_token++ = *buf_ptr++; } break; case BASE_16: *e_token++ = *buf_ptr++; *e_token++ = *buf_ptr++; while (isxdigit(*buf_ptr)) { CHECK_SIZE_TOKEN; *e_token++ = *buf_ptr++; } break; case BASE_10: while (1) { if (*buf_ptr == '.') { if (seendot) break; else seendot++; } CHECK_SIZE_TOKEN; *e_token++ = *buf_ptr++; if (!isdigit(*buf_ptr) && *buf_ptr != '.') { if ((*buf_ptr != 'E' && *buf_ptr != 'e') || seenexp) break; else { seenexp++; seendot++; CHECK_SIZE_TOKEN; *e_token++ = *buf_ptr++; if (*buf_ptr == '+' || *buf_ptr == '-') *e_token++ = *buf_ptr++; } } } break; } while (1) { if (!(seensfx & 1) && (*buf_ptr == 'U' || *buf_ptr == 'u')) { CHECK_SIZE_TOKEN; *e_token++ = *buf_ptr++; seensfx |= 1; continue; } if (!(seensfx & 2) && (strchr("fFlL", *buf_ptr) != NULL)) { CHECK_SIZE_TOKEN; if (buf_ptr[1] == buf_ptr[0]) *e_token++ = *buf_ptr++; *e_token++ = *buf_ptr++; seensfx |= 2; continue; } break; } } else while (chartype[(int)*buf_ptr] == alphanum || *buf_ptr == BACKSLASH) { /* fill_buffer() terminates buffer with newline */ if (*buf_ptr == BACKSLASH) { if (*(buf_ptr + 1) == '\n') { buf_ptr += 2; if (buf_ptr >= buf_end) fill_buffer(); } else break; } CHECK_SIZE_TOKEN; /* copy it over */ *e_token++ = *buf_ptr++; if (buf_ptr >= buf_end) fill_buffer(); } *e_token++ = '\0'; if (s_token[0] == 'L' && s_token[1] == '\0' && (*buf_ptr == '"' || *buf_ptr == '\'')) return (strpfx); while (*buf_ptr == ' ' || *buf_ptr == '\t') { /* get rid of blanks */ if (++buf_ptr >= buf_end) fill_buffer(); } ps.keyword = 0; if (l_struct && !ps.p_l_follow) { /* if last token was 'struct' and we're not * in parentheses, then this token * should be treated as a declaration */ l_struct = false; last_code = ident; ps.last_u_d = true; return (decl); } ps.last_u_d = l_struct; /* Operator after identifier is binary * unless last token was 'struct' */ l_struct = false; last_code = ident; /* Remember that this is the code we will * return */ p = bsearch(s_token, specials, sizeof(specials) / sizeof(specials[0]), sizeof(specials[0]), strcmp_type); if (p == NULL) { /* not a special keyword... */ char *u; /* ... so maybe a type_t or a typedef */ if ((auto_typedefs && ((u = strrchr(s_token, '_')) != NULL) && strcmp(u, "_t") == 0) || (typename_top >= 0 && bsearch(s_token, typenames, typename_top + 1, sizeof(typenames[0]), strcmp_type))) { ps.keyword = 4; /* a type name */ ps.last_u_d = true; goto found_typename; } } else { /* we have a keyword */ ps.keyword = p->rwcode; ps.last_u_d = true; switch (p->rwcode) { case 7: /* it is a switch */ return (swstmt); case 8: /* a case or default */ return (casestmt); case 3: /* a "struct" */ /* * Next time around, we will want to know that we have had a * 'struct' */ l_struct = true; /* FALLTHROUGH */ case 4: /* one of the declaration keywords */ found_typename: if (ps.p_l_follow) { /* inside parens: cast, param list, offsetof or sizeof */ ps.cast_mask |= (1 << ps.p_l_follow) & ~ps.not_cast_mask; break; } last_code = decl; return (decl); case 5: /* if, while, for */ return (sp_paren); case 6: /* do, else */ return (sp_nparen); case 10: /* storage class specifier */ return (storage); case 11: /* typedef */ return (type_def); default: /* all others are treated like any other * identifier */ return (ident); } /* end of switch */ } /* end of if (found_it) */ if (*buf_ptr == '(' && ps.tos <= 1 && ps.ind_level == 0 && ps.in_parameter_declaration == 0 && ps.block_init == 0) { char *tp = buf_ptr; while (tp < buf_end) if (*tp++ == ')' && (*tp == ';' || *tp == ',')) goto not_proc; strncpy(ps.procname, token, sizeof ps.procname - 1); if (ps.in_decl) ps.in_parameter_declaration = 1; return (last_code = funcname); not_proc:; } /* * The following hack attempts to guess whether or not the current * token is in fact a declaration keyword -- one that has been * typedefd */ if (((*buf_ptr == '*' && buf_ptr[1] != '=') || isalpha(*buf_ptr) || *buf_ptr == '_') && !ps.p_l_follow && !ps.block_init && (ps.last_token == rparen || ps.last_token == semicolon || ps.last_token == decl || ps.last_token == lbrace || ps.last_token == rbrace)) { ps.keyword = 4; /* a type name */ ps.last_u_d = true; last_code = decl; return decl; } if (last_code == decl) /* if this is a declared variable, then * following sign is unary */ ps.last_u_d = true; /* will make "int a -1" work */ last_code = ident; return (ident); /* the ident is not in the list */ } /* end of procesing for alpanum character */ /* Scan a non-alphanumeric token */ *e_token++ = *buf_ptr; /* if it is only a one-character token, it is * moved here */ *e_token = '\0'; if (++buf_ptr >= buf_end) fill_buffer(); switch (*token) { case '\n': unary_delim = ps.last_u_d; ps.last_nl = true; /* remember that we just had a newline */ code = (had_eof ? 0 : newline); /* * if data has been exhausted, the newline is a dummy, and we should * return code to stop */ break; case '\'': /* start of quoted character */ case '"': /* start of string */ qchar = *token; if (troff) { e_token[-1] = '`'; if (qchar == '"') *e_token++ = '`'; e_token = chfont(&bodyf, &stringf, e_token); } do { /* copy the string */ while (1) { /* move one character or [/] */ if (*buf_ptr == '\n') { diag2(1, "Unterminated literal"); goto stop_lit; } CHECK_SIZE_TOKEN; /* Only have to do this once in this loop, * since CHECK_SIZE guarantees that there * are at least 5 entries left */ *e_token = *buf_ptr++; if (buf_ptr >= buf_end) fill_buffer(); if (*e_token == BACKSLASH) { /* if escape, copy extra char */ if (*buf_ptr == '\n') /* check for escaped newline */ ++line_no; if (troff) { *++e_token = BACKSLASH; if (*buf_ptr == BACKSLASH) *++e_token = BACKSLASH; } *++e_token = *buf_ptr++; ++e_token; /* we must increment this again because we * copied two chars */ if (buf_ptr >= buf_end) fill_buffer(); } else break; /* we copied one character */ } /* end of while (1) */ } while (*e_token++ != qchar); if (troff) { e_token = chfont(&stringf, &bodyf, e_token - 1); if (qchar == '"') *e_token++ = '\''; } stop_lit: code = ident; break; case ('('): case ('['): unary_delim = true; code = lparen; break; case (')'): case (']'): code = rparen; break; case '#': unary_delim = ps.last_u_d; code = preesc; break; case '?': unary_delim = true; code = question; break; case (':'): code = colon; unary_delim = true; break; case (';'): unary_delim = true; code = semicolon; break; case ('{'): unary_delim = true; /* * if (ps.in_or_st) ps.block_init = 1; */ /* ? code = ps.block_init ? lparen : lbrace; */ code = lbrace; break; case ('}'): unary_delim = true; /* ? code = ps.block_init ? rparen : rbrace; */ code = rbrace; break; case 014: /* a form feed */ unary_delim = ps.last_u_d; ps.last_nl = true; /* remember this so we can set 'ps.col_1' * right */ code = form_feed; break; case (','): unary_delim = true; code = comma; break; case '.': unary_delim = false; code = period; break; case '-': case '+': /* check for -, +, --, ++ */ code = (ps.last_u_d ? unary_op : binary_op); unary_delim = true; if (*buf_ptr == token[0]) { /* check for doubled character */ *e_token++ = *buf_ptr++; /* buffer overflow will be checked at end of loop */ if (last_code == ident || last_code == rparen) { code = (ps.last_u_d ? unary_op : postop); /* check for following ++ or -- */ unary_delim = false; } } else if (*buf_ptr == '=') /* check for operator += */ *e_token++ = *buf_ptr++; else if (*buf_ptr == '>') { /* check for operator -> */ *e_token++ = *buf_ptr++; - if (!pointer_as_binop) { - unary_delim = false; - code = unary_op; - ps.want_blank = false; - } + unary_delim = false; + code = unary_op; + ps.want_blank = false; } break; /* buffer overflow will be checked at end of * switch */ case '=': if (ps.in_or_st) ps.block_init = 1; #ifdef undef if (chartype[*buf_ptr] == opchar) { /* we have two char assignment */ e_token[-1] = *buf_ptr++; if ((e_token[-1] == '<' || e_token[-1] == '>') && e_token[-1] == *buf_ptr) *e_token++ = *buf_ptr++; *e_token++ = '='; /* Flip =+ to += */ *e_token = 0; } #else if (*buf_ptr == '=') {/* == */ *e_token++ = '='; /* Flip =+ to += */ buf_ptr++; *e_token = 0; } #endif code = binary_op; unary_delim = true; break; /* can drop thru!!! */ case '>': case '<': case '!': /* ops like <, <<, <=, !=, etc */ if (*buf_ptr == '>' || *buf_ptr == '<' || *buf_ptr == '=') { *e_token++ = *buf_ptr; if (++buf_ptr >= buf_end) fill_buffer(); } if (*buf_ptr == '=') *e_token++ = *buf_ptr++; code = (ps.last_u_d ? unary_op : binary_op); unary_delim = true; break; default: if (token[0] == '/' && *buf_ptr == '*') { /* it is start of comment */ *e_token++ = '*'; if (++buf_ptr >= buf_end) fill_buffer(); code = comment; unary_delim = ps.last_u_d; break; } while (*(e_token - 1) == *buf_ptr || *buf_ptr == '=') { /* * handle ||, &&, etc, and also things as in int *****i */ *e_token++ = *buf_ptr; if (++buf_ptr >= buf_end) fill_buffer(); } code = (ps.last_u_d ? unary_op : binary_op); unary_delim = true; } /* end of switch */ if (code != newline) { l_struct = false; last_code = code; } if (buf_ptr >= buf_end) /* check for input buffer empty */ fill_buffer(); ps.last_u_d = unary_delim; *e_token = '\0'; /* null terminate the token */ return (code); } void alloc_typenames(void) { typenames = (const char **)malloc(sizeof(typenames[0]) * (typename_count = 16)); if (typenames == NULL) err(1, NULL); } void add_typename(const char *key) { int comparison; const char *copy; if (typename_top + 1 >= typename_count) { typenames = realloc((void *)typenames, sizeof(typenames[0]) * (typename_count *= 2)); if (typenames == NULL) err(1, NULL); } if (typename_top == -1) typenames[++typename_top] = copy = strdup(key); else if ((comparison = strcmp(key, typenames[typename_top])) >= 0) { /* take advantage of sorted input */ if (comparison == 0) /* remove duplicates */ return; typenames[++typename_top] = copy = strdup(key); } else { int p; for (p = 0; (comparison = strcmp(key, typenames[p])) > 0; p++) /* find place for the new key */; if (comparison == 0) /* remove duplicates */ return; memmove(&typenames[p + 1], &typenames[p], sizeof(typenames[0]) * (++typename_top - p)); typenames[p] = copy = strdup(key); } if (copy == NULL) err(1, NULL); }