diff options
| -rw-r--r-- | ChangeLog | 14 | ||||
| -rw-r--r-- | src/output.cc | 188 | ||||
| -rw-r--r-- | tests/Makefile.in | 19 | ||||
| -rw-r--r-- | tests/permutc2.exp | 2 | ||||
| -rw-r--r-- | tests/smtp.gperf | 206 |
5 files changed, 359 insertions, 70 deletions
@@ -1,3 +1,17 @@ +2003-01-01 Bruno Haible <bruno@clisp.org> + + * src/output.c (Output_Compare::output_firstchar_comparison): New + method. + (Output_Compare_Strcmp::output_comparison, + Output_Compare_Strncmp::output_comparison, + Output_Compare_Memcmp::output_comparison): Use it. + * tests/permutc2.exp: Update. + + * tests/smtp.gperf: New file, based on a contribution by Bruce Lilly. + * tests/Makefile.in (check-smtp): New rule. + (check): Depend on it. + (clean): Update. + 2002-12-12 Bruno Haible <bruno@clisp.org> * src/search.h (Search::init_selchars_tuple, diff --git a/src/output.cc b/src/output.cc index f1bd0d0..23363f4 100644 --- a/src/output.cc +++ b/src/output.cc @@ -1,5 +1,5 @@ /* Output routines. - Copyright (C) 1989-1998, 2000, 2002 Free Software Foundation, Inc. + Copyright (C) 1989-1998, 2000, 2002-2003 Free Software Foundation, Inc. Written by Douglas C. Schmidt <schmidt@ics.uci.edu> and Bruno Haible <bruno@clisp.org>. @@ -144,22 +144,22 @@ Output::num_hash_values () const struct Output_Constants { - virtual void output_start () = 0; - virtual void output_item (const char *name, int value) = 0; - virtual void output_end () = 0; - Output_Constants () {} - virtual ~Output_Constants () {} + virtual void output_start () = 0; + virtual void output_item (const char *name, int value) = 0; + virtual void output_end () = 0; + Output_Constants () {} + virtual ~Output_Constants () {} }; /* This class outputs an enumeration in #define syntax. */ struct Output_Defines : public Output_Constants { - virtual void output_start (); - virtual void output_item (const char *name, int value); - virtual void output_end (); - Output_Defines () {} - virtual ~Output_Defines () {} + virtual void output_start (); + virtual void output_item (const char *name, int value); + virtual void output_end (); + Output_Defines () {} + virtual ~Output_Defines () {} }; void Output_Defines::output_start () @@ -180,11 +180,12 @@ void Output_Defines::output_end () struct Output_Enum : public Output_Constants { - virtual void output_start (); - virtual void output_item (const char *name, int value); - virtual void output_end (); - Output_Enum (const char *indent) : _indentation (indent) {} - virtual ~Output_Enum () {} + virtual void output_start (); + virtual void output_item (const char *name, int value); + virtual void output_end (); + Output_Enum (const char *indent) + : _indentation (indent) {} + virtual ~Output_Enum () {} private: const char *_indentation; bool _pending_comma; @@ -407,18 +408,18 @@ output_const_type (const char *const_string, const char *type_string) struct Output_Expr { - virtual void output_expr () const = 0; - Output_Expr () {} - virtual ~Output_Expr () {} + virtual void output_expr () const = 0; + Output_Expr () {} + virtual ~Output_Expr () {} }; /* This class outputs an expression formed by a single string. */ struct Output_Expr1 : public Output_Expr { - virtual void output_expr () const; - Output_Expr1 (const char *piece1) : _p1 (piece1) {} - virtual ~Output_Expr1 () {} + virtual void output_expr () const; + Output_Expr1 (const char *piece1) : _p1 (piece1) {} + virtual ~Output_Expr1 () {} private: const char *_p1; }; @@ -435,10 +436,10 @@ void Output_Expr1::output_expr () const struct Output_Expr2 : public Output_Expr { - virtual void output_expr () const; - Output_Expr2 (const char *piece1, const char *piece2) - : _p1 (piece1), _p2 (piece2) {} - virtual ~Output_Expr2 () {} + virtual void output_expr () const; + Output_Expr2 (const char *piece1, const char *piece2) + : _p1 (piece1), _p2 (piece2) {} + virtual ~Output_Expr2 () {} private: const char *_p1; const char *_p2; @@ -462,37 +463,76 @@ struct Output_Compare the string being looked up. expr2 outputs a simple expression of type 'const char *' referring to the constant string stored in the gperf generated hash table. */ - virtual void output_comparison (const Output_Expr& expr1, - const Output_Expr& expr2) const = 0; - Output_Compare () {} - virtual ~Output_Compare () {} + virtual void output_comparison (const Output_Expr& expr1, + const Output_Expr& expr2) const = 0; + /* Outputs the comparison expression for the first byte. + Returns true if the this comparison is complete. */ + bool output_firstchar_comparison (const Output_Expr& expr1, + const Output_Expr& expr2) const; + Output_Compare () {} + virtual ~Output_Compare () {} }; +bool Output_Compare::output_firstchar_comparison (const Output_Expr& expr1, + const Output_Expr& expr2) const +{ + /* First, we emit a comparison of the first byte of the two strings. + This catches most cases where the string being looked up is not in the + hash table but happens to have the same hash code as an element of the + hash table. */ + if (option[UPPERLOWER]) + { + /* Incomplete comparison, just for speedup. */ + printf ("(((unsigned char)*"); + expr1.output_expr (); + printf (" ^ (unsigned char)*"); + expr2.output_expr (); + printf (") & ~32) == 0"); + return false; + } + else + { + /* Complete comparison. */ + printf ("*"); + expr1.output_expr (); + printf (" == *"); + expr2.output_expr (); + return true; + } +} + /* This class outputs a comparison using strcmp. */ struct Output_Compare_Strcmp : public Output_Compare { - virtual void output_comparison (const Output_Expr& expr1, - const Output_Expr& expr2) const; - Output_Compare_Strcmp () {} - virtual ~Output_Compare_Strcmp () {} + virtual void output_comparison (const Output_Expr& expr1, + const Output_Expr& expr2) const; + Output_Compare_Strcmp () {} + virtual ~Output_Compare_Strcmp () {} }; void Output_Compare_Strcmp::output_comparison (const Output_Expr& expr1, const Output_Expr& expr2) const { - printf ("*"); - expr1.output_expr (); - printf (" == *"); - expr2.output_expr (); + bool firstchar_done = output_firstchar_comparison (expr1, expr2); printf (" && !"); if (option[UPPERLOWER]) printf ("gperf_case_"); printf ("strcmp ("); - expr1.output_expr (); - printf (" + 1, "); - expr2.output_expr (); - printf (" + 1)"); + if (firstchar_done) + { + expr1.output_expr (); + printf (" + 1, "); + expr2.output_expr (); + printf (" + 1"); + } + else + { + expr1.output_expr (); + printf (", "); + expr2.output_expr (); + } + printf (")"); } /* This class outputs a comparison using strncmp. @@ -501,27 +541,35 @@ void Output_Compare_Strcmp::output_comparison (const Output_Expr& expr1, struct Output_Compare_Strncmp : public Output_Compare { - virtual void output_comparison (const Output_Expr& expr1, - const Output_Expr& expr2) const; - Output_Compare_Strncmp () {} - virtual ~Output_Compare_Strncmp () {} + virtual void output_comparison (const Output_Expr& expr1, + const Output_Expr& expr2) const; + Output_Compare_Strncmp () {} + virtual ~Output_Compare_Strncmp () {} }; void Output_Compare_Strncmp::output_comparison (const Output_Expr& expr1, const Output_Expr& expr2) const { - printf ("*"); - expr1.output_expr (); - printf (" == *"); - expr2.output_expr (); + bool firstchar_done = output_firstchar_comparison (expr1, expr2); printf (" && !"); if (option[UPPERLOWER]) printf ("gperf_case_"); printf ("strncmp ("); - expr1.output_expr (); - printf (" + 1, "); - expr2.output_expr (); - printf (" + 1, len - 1) && "); + if (firstchar_done) + { + expr1.output_expr (); + printf (" + 1, "); + expr2.output_expr (); + printf (" + 1, len - 1"); + } + else + { + expr1.output_expr (); + printf (", "); + expr2.output_expr (); + printf (", len"); + } + printf (") && "); expr2.output_expr (); printf ("[len] == '\\0'"); } @@ -533,27 +581,35 @@ void Output_Compare_Strncmp::output_comparison (const Output_Expr& expr1, struct Output_Compare_Memcmp : public Output_Compare { - virtual void output_comparison (const Output_Expr& expr1, - const Output_Expr& expr2) const; - Output_Compare_Memcmp () {} - virtual ~Output_Compare_Memcmp () {} + virtual void output_comparison (const Output_Expr& expr1, + const Output_Expr& expr2) const; + Output_Compare_Memcmp () {} + virtual ~Output_Compare_Memcmp () {} }; void Output_Compare_Memcmp::output_comparison (const Output_Expr& expr1, const Output_Expr& expr2) const { - printf ("*"); - expr1.output_expr (); - printf (" == *"); - expr2.output_expr (); + bool firstchar_done = output_firstchar_comparison (expr1, expr2); printf (" && !"); if (option[UPPERLOWER]) printf ("gperf_case_"); printf ("memcmp ("); - expr1.output_expr (); - printf (" + 1, "); - expr2.output_expr (); - printf (" + 1, len - 1)"); + if (firstchar_done) + { + expr1.output_expr (); + printf (" + 1, "); + expr2.output_expr (); + printf (" + 1, len - 1"); + } + else + { + expr1.output_expr (); + printf (", "); + expr2.output_expr (); + printf (", len"); + } + printf (")"); } /* ------------------------------------------------------------------------- */ diff --git a/tests/Makefile.in b/tests/Makefile.in index c7c3c37..0e670b5 100644 --- a/tests/Makefile.in +++ b/tests/Makefile.in @@ -1,6 +1,6 @@ # Makefile for gperf/tests -# Copyright (C) 1989, 1992, 1993, 1995, 1998, 2000, 2002 Free Software Foundation, Inc. +# Copyright (C) 1989, 1992, 1993, 1995, 1998, 2000, 2002, 2003 Free Software Foundation, Inc. # Written by Douglas C. Schmidt <schmidt@ics.uci.edu> # and Bruno Haible <bruno@clisp.org>. # @@ -58,7 +58,7 @@ installdirs : uninstall : -check : check-link-c check-link-c++ check-c check-ada check-modula3 check-pascal check-lang-utf8 check-lang-ucs2 check-test +check : check-link-c check-link-c++ check-c check-ada check-modula3 check-pascal check-lang-utf8 check-lang-ucs2 check-smtp check-test @true extracheck : @CHECK_LANG_SYNTAX@ @@ -121,6 +121,19 @@ check-lang-ucs2: ./lu2out -v < $(srcdir)/lang-ucs2.in > lang-ucs2.out diff $(srcdir)/lang-ucs2.exp lang-ucs2.out +# check case-insensitive lookup +check-smtp: + @echo "testing SMTP keywords, case-insensitive" + $(GPERF) --struct-type --readonly-table --enum --global -K field_name -N header_entry --ignore-case $(srcdir)/smtp.gperf > smtp.c + $(CC) $(CFLAGS) -o smtp smtp.c + ./smtp + $(GPERF) --struct-type --readonly-table --enum --global -K field_name -N header_entry --ignore-case --compare-strncmp $(srcdir)/smtp.gperf > smtp.c + $(CC) $(CFLAGS) -o smtp smtp.c + ./smtp + $(GPERF) --struct-type --readonly-table --enum --global -K field_name -N header_entry --ignore-case --compare-lengths $(srcdir)/smtp.gperf > smtp.c + $(CC) $(CFLAGS) -o smtp smtp.c + ./smtp + # these next 5 are demos that show off the generated code check-test: $(GPERF) -L C -F ', 0, 0' -p -j1 -i 1 -g -o -t -G -N is_reserved_word -k1,3,'$$' < $(srcdir)/c-parse.gperf > c-parse.out @@ -273,7 +286,7 @@ check-lang-syntax : force mostlyclean : clean clean : force - $(RM) *.o core *inset.c output.* *.out aout cout lu2out lu8out m3out pout preout tmp-* valitest* + $(RM) *.o core *inset.c output.* *.out aout cout lu2out lu8out m3out pout preout smtp.c smtp tmp-* valitest* distclean : clean $(RM) config.status config.log config.cache Makefile diff --git a/tests/permutc2.exp b/tests/permutc2.exp index 02632d5..30eb911 100644 --- a/tests/permutc2.exp +++ b/tests/permutc2.exp @@ -108,7 +108,7 @@ in_word_set (str, len) { register const char *s = wordlist[key]; - if (*str == *s && !gperf_case_strcmp (str + 1, s + 1)) + if ((((unsigned char)*str ^ (unsigned char)*s) & ~32) == 0 && !gperf_case_strcmp (str, s)) return s; } } diff --git a/tests/smtp.gperf b/tests/smtp.gperf new file mode 100644 index 0000000..33137b9 --- /dev/null +++ b/tests/smtp.gperf @@ -0,0 +1,206 @@ +%{ +/* gperf --struct-type --readonly-table --enum --global -K field_name -N header_entry --ignore-case */ +/* Contributed by Bruce Lilly + derived from http://users.erols.com/blilly/mailparse/fields.gperf */ + +#include <string.h> + +%} +struct header_state { const char *field_name; }; +%% +Accept-Language +Action +Alternate-Recipient +Approved +Archive +Arrival-Date +Autoforwarded +Autosubmitted +Bcc +Cc +Comments +Complaints-To +Content-alternative +Content-Base +Content-Description +Content-Disposition +Content-Duration +Content-Features +Content-ID +Content-Language +Content-Location +Content-MD5 +Content-Transfer-Encoding +Content-Type +Control +Conversion +Conversion-With-Loss +DL-Expansion-History +DSN-Gateway +Date +Deferred-Delivery +Delivery-Date +Diagnostic-Code +Discarded-X400-IPMS-Extensions +Discarded-X400-MTS-Extensions +Disclose-Recipients +Disposition +Disposition-Notification-Options +Disposition-Notification-To +Distribution +Encrypted +Error +Expires +Failure +Final-Log-ID +Final-Recipient +Followup-To +From +Generate-Delivery-Report +Importance +In-Reply-To +Incomplete-Copy +Injector-Info +Keywords +Last-Attempt-Date +Latest-Delivery-Time +Lines +List-Archive +List-Help +List-ID +List-Post +List-Owner +List-Subscribe +List-Unsubscribe +MDN-Gateway +Media-Accept-Features +MIME-Version +Mail-Copies-To +Message-ID +Message-Type +Newsgroups +Organization +Original-Encoded-Information-Types +Original-Envelope-ID +Original-Message-ID +Original-Recipient +Originator-Return-Address +Path +Posted-And-Mailed +Prevent-Nondelivery-Report +Priority +Received +Received-content-MIC +Received-From-MTA +References +Remote-MTA +Reply-By +Reply-To +Reporting-MTA +Reporting-UA +Return-Path +Sender +Sensitivity +Status +Subject +Summary +Supersedes +To +User-Agent +Warning +Will-Retry-Until +X400-Content-Identifier +X400-Content-Return +X400-Content-Type +X400-MTS-Identifier +X400-Originator +X400-Received +X400-Recipients +Xref +%% + +#include <stdio.h> +#include <stdlib.h> +#include <ctype.h> + +static int +my_case_strcmp (s1, s2) + register const char *s1; + register const char *s2; +{ + for (;;) + { + unsigned char c1 = *s1++; + unsigned char c2 = *s2++; + if (c1 >= 'A' && c1 <= 'Z') + c1 += 'a' - 'A'; + if (c2 >= 'A' && c2 <= 'Z') + c2 += 'a' - 'A'; + if (c1 != 0 && c1 == c2) + continue; + return (int)c1 - (int)c2; + } +} + +int +main (argc, argv) + int argc; + char *argv[]; +{ + int i, j, k, n, exitcode; + unsigned int len; + const struct header_state *hs; + + n = 1; + if (argc > 1) + n = atoi (argv[1]); + if (n < 1) + n = 1; + + exitcode = 0; + for (i = 0; i < n; i++) + { + for (j = 0; j <= MAX_HASH_VALUE; j++) + { + const char *s = wordlist[j].field_name; + len = strlen (s); + if (len) + { + hs = header_entry (s, len); + if (!(hs && strcmp (hs->field_name, s) == 0)) + { + fprintf (stderr, "%s != %s\n", s, hs ? hs->field_name : "(null)"); + exitcode = 1; + } + } + } + for (j = 0; j <= MAX_HASH_VALUE; j++) + { + char s[MAX_WORD_LENGTH+1]; + /* expensive copy with case conversion (for testing) */ + strcpy (s, wordlist[j].field_name); + len = strlen (s); + if (len) + { + for (k = 0; k < len; k++) + if (isupper (s[k])) + s[k] = tolower (s[k]); + else if (islower (s[k])) + s[k] = toupper (s[k]); + hs = header_entry (s, len); + if (!(hs && my_case_strcmp (hs->field_name, s) == 0)) + { + fprintf (stderr, "%s != %s\n", s, hs ? hs->field_name : "(null)"); + exitcode = 1; + } + } + } + hs = header_entry ("Dave", 4); + if (hs) + { + fprintf (stderr, "Dave == %s\n", hs->field_name); + exitcode = 1; + } + } + return exitcode; +} |