From 450c26c4713a2de440296260e8d0be8e6b283d75 Mon Sep 17 00:00:00 2001
From: Max Kellermann <max@duempel.org>
Date: Mon, 8 Apr 2013 23:51:39 +0200
Subject: tokenizer: convert to C++
---
src/AllCommands.cxx | 14 ++-
src/ConfigFile.cxx | 30 ++++---
src/tokenizer.c | 225 -------------------------------------------------
src/tokenizer.h | 83 ------------------
src/util/Tokenizer.cxx | 202 ++++++++++++++++++++++++++++++++++++++++++++
src/util/Tokenizer.hxx | 99 ++++++++++++++++++++++
6 files changed, 324 insertions(+), 329 deletions(-)
delete mode 100644 src/tokenizer.c
delete mode 100644 src/tokenizer.h
create mode 100644 src/util/Tokenizer.cxx
create mode 100644 src/util/Tokenizer.hxx
(limited to 'src')
diff --git a/src/AllCommands.cxx b/src/AllCommands.cxx
index 58dcf4dba..f3243915b 100644
--- a/src/AllCommands.cxx
+++ b/src/AllCommands.cxx
@@ -31,10 +31,7 @@
#include "tag.h"
#include "protocol/Result.hxx"
#include "Client.hxx"
-
-extern "C" {
-#include "tokenizer.h"
-}
+#include "util/Tokenizer.hxx"
#ifdef ENABLE_SQLITE
#include "StickerCommands.hxx"
@@ -329,10 +326,11 @@ command_process(Client *client, unsigned num, char *line)
/* get the command name (first word on the line) */
- argv[0] = tokenizer_next_word(&line, &error);
+ Tokenizer tokenizer(line);
+ argv[0] = tokenizer.NextWord(&error);
if (argv[0] == NULL) {
current_command = "";
- if (*line == 0)
+ if (tokenizer.IsEnd())
command_error(client, ACK_ERROR_UNKNOWN,
"No command given");
else {
@@ -351,7 +349,7 @@ command_process(Client *client, unsigned num, char *line)
while (argc < (int)G_N_ELEMENTS(argv) &&
(argv[argc] =
- tokenizer_next_param(&line, &error)) != NULL)
+ tokenizer.NextParam(&error)) != NULL)
++argc;
/* some error checks; we have to set current_command because
@@ -365,7 +363,7 @@ command_process(Client *client, unsigned num, char *line)
return COMMAND_RETURN_ERROR;
}
- if (*line != 0) {
+ if (!tokenizer.IsEnd()) {
command_error(client, ACK_ERROR_ARG,
"%s", error->message);
current_command = NULL;
diff --git a/src/ConfigFile.cxx b/src/ConfigFile.cxx
index e94f3f238..f7f525096 100644
--- a/src/ConfigFile.cxx
+++ b/src/ConfigFile.cxx
@@ -23,10 +23,10 @@
#include "ConfigData.hxx"
#include "ConfigTemplates.hxx"
#include "conf.h"
+#include "util/Tokenizer.hxx"
extern "C" {
#include "string_util.h"
-#include "tokenizer.h"
}
#include "fs/Path.hxx"
@@ -50,15 +50,17 @@ static bool
config_read_name_value(struct config_param *param, char *input, unsigned line,
GError **error_r)
{
- const char *name = tokenizer_next_word(&input, error_r);
+ Tokenizer tokenizer(input);
+
+ const char *name = tokenizer.NextWord(error_r);
if (name == NULL) {
- assert(*input != 0);
+ assert(!tokenizer.IsEnd());
return false;
}
- const char *value = tokenizer_next_string(&input, error_r);
+ const char *value = tokenizer.NextString(error_r);
if (value == NULL) {
- if (*input == 0) {
+ if (tokenizer.IsEnd()) {
assert(error_r == NULL || *error_r == NULL);
g_set_error(error_r, config_quark(), 0,
"Value missing");
@@ -69,7 +71,7 @@ config_read_name_value(struct config_param *param, char *input, unsigned line,
return false;
}
- if (*input != 0 && *input != CONF_COMMENT) {
+ if (!tokenizer.IsEnd() && tokenizer.CurrentChar() != CONF_COMMENT) {
g_set_error(error_r, config_quark(), 0,
"Unknown tokens after value");
return false;
@@ -173,9 +175,10 @@ ReadConfigFile(ConfigData &config_data, FILE *fp, GError **error_r)
/* the first token in each line is the name, followed
by either the value or '{' */
- name = tokenizer_next_word(&line, &error);
+ Tokenizer tokenizer(line);
+ name = tokenizer.NextWord(&error);
if (name == NULL) {
- assert(*line != 0);
+ assert(!tokenizer.IsEnd());
g_propagate_prefixed_error(error_r, error,
"line %i: ", count);
return false;
@@ -210,13 +213,13 @@ ReadConfigFile(ConfigData &config_data, FILE *fp, GError **error_r)
if (option.block) {
/* it's a block, call config_read_block() */
- if (*line != '{') {
+ if (tokenizer.CurrentChar() != '{') {
g_set_error(error_r, config_quark(), 0,
"line %i: '{' expected", count);
return false;
}
- line = strchug_fast(line + 1);
+ line = strchug_fast(tokenizer.Rest() + 1);
if (*line != 0 && *line != CONF_COMMENT) {
g_set_error(error_r, config_quark(), 0,
"line %i: Unknown tokens after '{'",
@@ -231,9 +234,9 @@ ReadConfigFile(ConfigData &config_data, FILE *fp, GError **error_r)
} else {
/* a string value */
- value = tokenizer_next_string(&line, &error);
+ value = tokenizer.NextString(&error);
if (value == NULL) {
- if (*line == 0)
+ if (tokenizer.IsEnd())
g_set_error(error_r, config_quark(), 0,
"line %i: Value missing",
count);
@@ -247,7 +250,8 @@ ReadConfigFile(ConfigData &config_data, FILE *fp, GError **error_r)
return false;
}
- if (*line != 0 && *line != CONF_COMMENT) {
+ if (!tokenizer.IsEnd() &&
+ tokenizer.CurrentChar() != CONF_COMMENT) {
g_set_error(error_r, config_quark(), 0,
"line %i: Unknown tokens after value",
count);
diff --git a/src/tokenizer.c b/src/tokenizer.c
deleted file mode 100644
index 4a98e882f..000000000
--- a/src/tokenizer.c
+++ /dev/null
@@ -1,225 +0,0 @@
-/*
- * Copyright (C) 2003-2011 The Music Player Daemon Project
- * http://www.musicpd.org
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
- */
-
-#include "config.h"
-#include "tokenizer.h"
-#include "string_util.h"
-
-#include <glib.h>
-
-#include <stdbool.h>
-#include <assert.h>
-#include <string.h>
-
-G_GNUC_CONST
-static GQuark
-tokenizer_quark(void)
-{
- return g_quark_from_static_string("tokenizer");
-}
-
-static inline bool
-valid_word_first_char(char ch)
-{
- return g_ascii_isalpha(ch);
-}
-
-static inline bool
-valid_word_char(char ch)
-{
- return g_ascii_isalnum(ch) || ch == '_';
-}
-
-char *
-tokenizer_next_word(char **input_p, GError **error_r)
-{
- char *word, *input;
-
- assert(input_p != NULL);
- assert(*input_p != NULL);
-
- word = input = *input_p;
-
- if (*input == 0)
- return NULL;
-
- /* check the first character */
-
- if (!valid_word_first_char(*input)) {
- g_set_error(error_r, tokenizer_quark(), 0,
- "Letter expected");
- return NULL;
- }
-
- /* now iterate over the other characters until we find a
- whitespace or end-of-string */
-
- while (*++input != 0) {
- if (g_ascii_isspace(*input)) {
- /* a whitespace: the word ends here */
- *input = 0;
- /* skip all following spaces, too */
- input = strchug_fast(input + 1);
- break;
- }
-
- if (!valid_word_char(*input)) {
- *input_p = input;
- g_set_error(error_r, tokenizer_quark(), 0,
- "Invalid word character");
- return NULL;
- }
- }
-
- /* end of string: the string is already null-terminated
- here */
-
- *input_p = input;
- return word;
-}
-
-static inline bool
-valid_unquoted_char(char ch)
-{
- return (unsigned char)ch > 0x20 && ch != '"' && ch != '\'';
-}
-
-char *
-tokenizer_next_unquoted(char **input_p, GError **error_r)
-{
- char *word, *input;
-
- assert(input_p != NULL);
- assert(*input_p != NULL);
-
- word = input = *input_p;
-
- if (*input == 0)
- return NULL;
-
- /* check the first character */
-
- if (!valid_unquoted_char(*input)) {
- g_set_error(error_r, tokenizer_quark(), 0,
- "Invalid unquoted character");
- return NULL;
- }
-
- /* now iterate over the other characters until we find a
- whitespace or end-of-string */
-
- while (*++input != 0) {
- if (g_ascii_isspace(*input)) {
- /* a whitespace: the word ends here */
- *input = 0;
- /* skip all following spaces, too */
- input = strchug_fast(input + 1);
- break;
- }
-
- if (!valid_unquoted_char(*input)) {
- *input_p = input;
- g_set_error(error_r, tokenizer_quark(), 0,
- "Invalid unquoted character");
- return NULL;
- }
- }
-
- /* end of string: the string is already null-terminated
- here */
-
- *input_p = input;
- return word;
-}
-
-char *
-tokenizer_next_string(char **input_p, GError **error_r)
-{
- char *word, *dest, *input;
-
- assert(input_p != NULL);
- assert(*input_p != NULL);
-
- word = dest = input = *input_p;
-
- if (*input == 0)
- /* end of line */
- return NULL;
-
- /* check for the opening " */
-
- if (*input != '"') {
- g_set_error(error_r, tokenizer_quark(), 0,
- "'\"' expected");
- return NULL;
- }
-
- ++input;
-
- /* copy all characters */
-
- while (*input != '"') {
- if (*input == '\\')
- /* the backslash escapes the following
- character */
- ++input;
-
- if (*input == 0) {
- /* return input-1 so the caller can see the
- difference between "end of line" and
- "error" */
- *input_p = input - 1;
- g_set_error(error_r, tokenizer_quark(), 0,
- "Missing closing '\"'");
- return NULL;
- }
-
- /* copy one character */
- *dest++ = *input++;
- }
-
- /* the following character must be a whitespace (or end of
- line) */
-
- ++input;
- if (*input != 0 && !g_ascii_isspace(*input)) {
- *input_p = input;
- g_set_error(error_r, tokenizer_quark(), 0,
- "Space expected after closing '\"'");
- return NULL;
- }
-
- /* finish the string and return it */
-
- *dest = 0;
- *input_p = strchug_fast(input);
- return word;
-}
-
-char *
-tokenizer_next_param(char **input_p, GError **error_r)
-{
- assert(input_p != NULL);
- assert(*input_p != NULL);
-
- if (**input_p == '"')
- return tokenizer_next_string(input_p, error_r);
- else
- return tokenizer_next_unquoted(input_p, error_r);
-}
diff --git a/src/tokenizer.h b/src/tokenizer.h
deleted file mode 100644
index 2026e5ad6..000000000
--- a/src/tokenizer.h
+++ /dev/null
@@ -1,83 +0,0 @@
-/*
- * Copyright (C) 2003-2011 The Music Player Daemon Project
- * http://www.musicpd.org
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
- */
-
-#ifndef MPD_TOKENIZER_H
-#define MPD_TOKENIZER_H
-
-#include "gerror.h"
-
-/**
- * Reads the next word from the input string. This function modifies
- * the input string.
- *
- * @param input_p the input string; this function returns a pointer to
- * the first non-whitespace character of the following token
- * @param error_r if this function returns NULL and **input_p!=0, it
- * optionally provides a GError object in this argument
- * @return a pointer to the null-terminated word, or NULL on error or
- * end of line
- */
-char *
-tokenizer_next_word(char **input_p, GError **error_r);
-
-/**
- * Reads the next unquoted word from the input string. This function
- * modifies the input string.
- *
- * @param input_p the input string; this function returns a pointer to
- * the first non-whitespace character of the following token
- * @param error_r if this function returns NULL and **input_p!=0, it
- * optionally provides a GError object in this argument
- * @return a pointer to the null-terminated word, or NULL on error or
- * end of line
- */
-char *
-tokenizer_next_unquoted(char **input_p, GError **error_r);
-
-/**
- * Reads the next quoted string from the input string. A backslash
- * escapes the following character. This function modifies the input
- * string.
- *
- * @param input_p the input string; this function returns a pointer to
- * the first non-whitespace character of the following token
- * @param error_r if this function returns NULL and **input_p!=0, it
- * optionally provides a GError object in this argument
- * @return a pointer to the null-terminated string, or NULL on error
- * or end of line
- */
-char *
-tokenizer_next_string(char **input_p, GError **error_r);
-
-/**
- * Reads the next unquoted word or quoted string from the input. This
- * is a wrapper for tokenizer_next_unquoted() and
- * tokenizer_next_string().
- *
- * @param input_p the input string; this function returns a pointer to
- * the first non-whitespace character of the following token
- * @param error_r if this function returns NULL and **input_p!=0, it
- * optionally provides a GError object in this argument
- * @return a pointer to the null-terminated string, or NULL on error
- * or end of line
- */
-char *
-tokenizer_next_param(char **input_p, GError **error_r);
-
-#endif
diff --git a/src/util/Tokenizer.cxx b/src/util/Tokenizer.cxx
new file mode 100644
index 000000000..9ade0d1b1
--- /dev/null
+++ b/src/util/Tokenizer.cxx
@@ -0,0 +1,202 @@
+/*
+ * Copyright (C) 2003-2013 The Music Player Daemon Project
+ * http://www.musicpd.org
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ */
+
+#include "config.h"
+#include "Tokenizer.hxx"
+#include "string_util.h"
+
+#include <glib.h>
+
+#include <stdbool.h>
+#include <assert.h>
+#include <string.h>
+
+G_GNUC_CONST
+static GQuark
+tokenizer_quark(void)
+{
+ return g_quark_from_static_string("tokenizer");
+}
+
+static inline bool
+valid_word_first_char(char ch)
+{
+ return g_ascii_isalpha(ch);
+}
+
+static inline bool
+valid_word_char(char ch)
+{
+ return g_ascii_isalnum(ch) || ch == '_';
+}
+
+char *
+Tokenizer::NextWord(GError **error_r)
+{
+ char *const word = input;
+
+ if (*input == 0)
+ return nullptr;
+
+ /* check the first character */
+
+ if (!valid_word_first_char(*input)) {
+ g_set_error(error_r, tokenizer_quark(), 0,
+ "Letter expected");
+ return nullptr;
+ }
+
+ /* now iterate over the other characters until we find a
+ whitespace or end-of-string */
+
+ while (*++input != 0) {
+ if (g_ascii_isspace(*input)) {
+ /* a whitespace: the word ends here */
+ *input = 0;
+ /* skip all following spaces, too */
+ input = strchug_fast(input + 1);
+ break;
+ }
+
+ if (!valid_word_char(*input)) {
+ g_set_error(error_r, tokenizer_quark(), 0,
+ "Invalid word character");
+ return nullptr;
+ }
+ }
+
+ /* end of string: the string is already null-terminated
+ here */
+
+ return word;
+}
+
+static inline bool
+valid_unquoted_char(char ch)
+{
+ return (unsigned char)ch > 0x20 && ch != '"' && ch != '\'';
+}
+
+char *
+Tokenizer::NextUnquoted(GError **error_r)
+{
+ char *const word = input;
+
+ if (*input == 0)
+ return nullptr;
+
+ /* check the first character */
+
+ if (!valid_unquoted_char(*input)) {
+ g_set_error(error_r, tokenizer_quark(), 0,
+ "Invalid unquoted character");
+ return nullptr;
+ }
+
+ /* now iterate over the other characters until we find a
+ whitespace or end-of-string */
+
+ while (*++input != 0) {
+ if (g_ascii_isspace(*input)) {
+ /* a whitespace: the word ends here */
+ *input = 0;
+ /* skip all following spaces, too */
+ input = strchug_fast(input + 1);
+ break;
+ }
+
+ if (!valid_unquoted_char(*input)) {
+ g_set_error(error_r, tokenizer_quark(), 0,
+ "Invalid unquoted character");
+ return nullptr;
+ }
+ }
+
+ /* end of string: the string is already null-terminated
+ here */
+
+ return word;
+}
+
+char *
+Tokenizer::NextString(GError **error_r)
+{
+ char *const word = input, *dest = input;
+
+ if (*input == 0)
+ /* end of line */
+ return nullptr;
+
+ /* check for the opening " */
+
+ if (*input != '"') {
+ g_set_error(error_r, tokenizer_quark(), 0,
+ "'\"' expected");
+ return nullptr;
+ }
+
+ ++input;
+
+ /* copy all characters */
+
+ while (*input != '"') {
+ if (*input == '\\')
+ /* the backslash escapes the following
+ character */
+ ++input;
+
+ if (*input == 0) {
+ /* return input-1 so the caller can see the
+ difference between "end of line" and
+ "error" */
+ --input;
+ g_set_error(error_r, tokenizer_quark(), 0,
+ "Missing closing '\"'");
+ return nullptr;
+ }
+
+ /* copy one character */
+ *dest++ = *input++;
+ }
+
+ /* the following character must be a whitespace (or end of
+ line) */
+
+ ++input;
+ if (*input != 0 && !g_ascii_isspace(*input)) {
+ g_set_error(error_r, tokenizer_quark(), 0,
+ "Space expected after closing '\"'");
+ return nullptr;
+ }
+
+ /* finish the string and return it */
+
+ *dest = 0;
+ input = strchug_fast(input);
+ return word;
+}
+
+char *
+Tokenizer::NextParam(GError **error_r)
+{
+ if (*input == '"')
+ return NextString(error_r);
+ else
+ return NextUnquoted(error_r);
+}
diff --git a/src/util/Tokenizer.hxx b/src/util/Tokenizer.hxx
new file mode 100644
index 000000000..da45348d4
--- /dev/null
+++ b/src/util/Tokenizer.hxx
@@ -0,0 +1,99 @@
+/*
+ * Copyright (C) 2003-2013 The Music Player Daemon Project
+ * http://www.musicpd.org
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ */
+
+#ifndef MPD_TOKENIZER_HXX
+#define MPD_TOKENIZER_HXX
+
+#include "gerror.h"
+
+class Tokenizer {
+ char *input;
+
+public:
+ /**
+ * @param _input the input string; the contents will be
+ * modified by this class
+ */
+ constexpr Tokenizer(char *_input):input(_input) {}
+
+ Tokenizer(const Tokenizer &) = delete;
+ Tokenizer &operator=(const Tokenizer &) = delete;
+
+ char *Rest() {
+ return input;
+ }
+
+ char CurrentChar() const {
+ return *input;
+ }
+
+ bool IsEnd() const {
+ return CurrentChar() == 0;
+ }
+
+ /**
+ * Reads the next word.
+ *
+ * @param error_r if this function returns nullptr and
+ * **input_p!=0, it optionally provides a GError object in
+ * this argument
+ * @return a pointer to the null-terminated word, or nullptr
+ * on error or end of line
+ */
+ char *NextWord(GError **error_r);
+
+ /**
+ * Reads the next unquoted word from the input string.
+ *
+ * @param error_r if this function returns nullptr and **input_p!=0, it
+ * optionally provides a GError object in this argument
+ * @return a pointer to the null-terminated word, or nullptr
+ * on error or end of line
+ */
+ char *NextUnquoted(GError **error_r);
+
+ /**
+ * Reads the next quoted string from the input string. A backslash
+ * escapes the following character. This function modifies the input
+ * string.
+ *
+ * @param input_p the input string; this function returns a pointer to
+ * the first non-whitespace character of the following token
+ * @param error_r if this function returns nullptr and **input_p!=0, it
+ * optionally provides a GError object in this argument
+ * @return a pointer to the null-terminated string, or nullptr on error
+ * or end of line
+ */
+ char *NextString(GError **error_r);
+
+ /**
+ * Reads the next unquoted word or quoted string from the
+ * input. This is a wrapper for NextUnquoted() and
+ * NextString().
+ *
+ * @param error_r if this function returns nullptr and
+ * **input_p!=0, it optionally provides a GError object in
+ * this argument
+ * @return a pointer to the null-terminated string, or nullptr
+ * on error or end of line
+ */
+ char *NextParam(GError **error_r);
+};
+
+#endif
--
cgit v1.2.3