X-Git-Url: http://git.tdb.fi/?p=libs%2Fcore.git;a=blobdiff_plain;f=source%2Futils.cpp;h=c2bafb9b85527c9b570dea735588397900f4f6ca;hp=244055b232cb81df6347f9200d315683d372aff3;hb=cc69674bf670178a8076f2ce9740c2e60c4644c9;hpb=43a991406be78b2905ef1f13f4e7589f6f6c1ba3

diff --git a/source/utils.cpp b/source/utils.cpp
index 244055b..c2bafb9 100644
--- a/source/utils.cpp
+++ b/source/utils.cpp
@@ -1,11 +1,13 @@
 /* $Id$
 
 This file is part of libmspstrings
-Copyright © 2006-2007 Mikko Rasa
+Copyright © 2006-2008 Mikko Rasa
 Distributed under the LGPL
 */
 
+#include <algorithm>
 #include <list>
+#include <msp/core/except.h>
 #include "utils.h"
 
 using namespace std;
@@ -44,18 +46,18 @@ vector<string> do_split(const string &str, const string &sep, int max_split)
 	return result;
 }
 
+bool check_str(const std::string &str, int (*pred)(int))
+{
+	for(string::const_iterator i=str.begin(); i!=str.end(); ++i)
+		if(!pred(*i))
+			return false;
+	return true;
 }
 
-namespace Msp {
-
-/**
-Compares two strings, ignoring case.
+}
 
-@param   s1  First string
-@param   s2  Second string
+namespace Msp {
 
-@return  -1 if s1<s2, 0 if s1==s2, 1 if s1>s2
-*/
 int strcasecmp(const string &s1, const string &s2)
 {
 	string::const_iterator i1=s1.begin();
@@ -71,9 +73,6 @@ int strcasecmp(const string &s1, const string &s2)
 	return 0;
 }
 
-/**
-Returns a lowercase copy of the given string.
-*/
 string tolower(const string &str)
 {
 	string result(str);
@@ -81,9 +80,6 @@ string tolower(const string &str)
 	return result;
 }
 
-/**
-Returns an uppercase copy of the given string.
-*/
 string toupper(const string &str)
 {
 	string result(str);
@@ -91,19 +87,19 @@ string toupper(const string &str)
 	return result;
 }
 
-vector<string> split_fields(const string &str, const string &sep, int max_split)
+bool isnumrc(const string &str)
 {
-	return do_split<true, true>(str, sep, max_split);
+	return check_str(str, isdigit);
 }
 
-vector<string> split_fields(const string &str, char sep, int max_split)
+bool isalpha(const string &str)
 {
-	return split_fields(str, string(1, sep), max_split);
+	return check_str(str, isalpha);
 }
 
-vector<string> split_long(const string &str, const string &sep, int max_split)
+bool isalnum(const string &str)
 {
-	return do_split<true, false>(str, sep, max_split);
+	return check_str(str, isalnum);
 }
 
 vector<string> split(const string &str, const string &sep, int max_split)
@@ -116,6 +112,21 @@ vector<string> split(const string &str, char sep, int max_split)
 	return split(str, string(1, sep), max_split);
 }
 
+vector<string> split_long(const string &str, const string &sep, int max_split)
+{
+	return do_split<true, false>(str, sep, max_split);
+}
+
+vector<string> split_fields(const string &str, const string &sep, int max_split)
+{
+	return do_split<true, true>(str, sep, max_split);
+}
+
+vector<string> split_fields(const string &str, char sep, int max_split)
+{
+	return split_fields(str, string(1, sep), max_split);
+}
+
 /**
 Splits a string to parts.
 
@@ -158,38 +169,145 @@ vector<string> split(const string &str, char sep, bool allow_empty)
 	return split(str, string(1, sep), allow_empty);
 }
 
-/**
-Builds a single string from the strings in the given sequence by concatenating
-them.
+string strip(const string &s)
+{
+	string result=s;
+	if(!result.erase(0, result.find_first_not_of(" \t\r\n")).empty())
+		result.erase(result.find_last_not_of(" \t\r\n")+1);
+	return result;
+}
 
-@param  seq  A sequence of strings
-@param  sep  Separator to be inserted between strings
-*/
-template<typename T>
-string join(const T &seq, const string &sep)
+string c_unescape(const std::string &str)
 {
+	bool escape=false;
+	unsigned numeric_type=0;
+	unsigned numeric_pos=0;
+	unsigned numeric_value=0;
 	string result;
-	for(typename T::const_iterator i=seq.begin(); i!=seq.end(); ++i)
+	for(string::const_iterator i=str.begin(); i!=str.end(); ++i)
 	{
-		if(i!=seq.begin())
-			result+=sep;
-		result+=*i;
+		if(numeric_type==16)
+		{
+			unsigned digit=0;
+			if(*i>='0' && *i<='9')
+				digit=*i-'0';
+			else if(*i>='a' && *i<='f')
+				digit=*i-'a'+10;
+			else if(*i>='A' && *i<='F')
+				digit=*i-'A'+10;
+			else
+				throw InvalidParameterValue("Invalid hexadecimal digit");
+
+			numeric_value=(numeric_value<<4 | digit);
+			++numeric_pos;
+			if(numeric_pos==2)
+			{
+				result+=numeric_value;
+				numeric_type=0;
+			}
+		}
+		else if(numeric_type==8)
+		{
+			unsigned digit=0;
+			if(*i>='0' && *i<='7')
+				digit=*i-'0';
+			else
+				throw InvalidParameterValue("Invalid octal digit");
+
+			numeric_value=(numeric_value<<3 | digit);
+			++numeric_pos;
+			if(numeric_pos==3)
+			{
+				result+=numeric_value;
+				numeric_type=0;
+			}
+		}
+		else if(escape)
+		{
+			if(*i=='x')
+			{
+				numeric_type=16;
+				numeric_pos=0;
+				numeric_value=0;
+			}
+			else if(*i>='0' && *i<='3')
+			{
+				numeric_type=8;
+				numeric_pos=1;
+				numeric_value=*i-'0';
+			}
+			else if(*i=='n')
+				result+='\n';
+			else if(*i=='t')
+				result+='\t';
+			else if(*i=='r')
+				result+='\r';
+			else if(*i=='b')
+				result+='\b';
+			else if(*i=='v')
+				result+='\v';
+			else if(*i=='a')
+				result+='\a';
+			else if(*i=='f')
+				result+='\f';
+			else if(*i=='\"')
+				result+='\"';
+			else if(*i=='\'')
+				result+='\'';
+			else if(*i=='\\')
+				result+='\\';
+			else
+				throw InvalidParameterValue("Invalid escape sequence");
+
+			escape=false;
+		}
+		else if(*i=='\\')
+			escape=true;
+		else
+			result+=*i;
 	}
 
+	if(escape)	
+		throw InvalidParameterValue("Stray backslash at end of string");
+
 	return result;
 }
-template string join<list<string> >(const list<string> &, const string &);
-template string join<vector<string> >(const vector<string> &, const string &);
 
-/**
-Returns a copy of the given string with leading and trailing whitespace
-removed.
-*/
-string strip(const string &s)
+string c_escape(const string &str, bool escape_8bit)
 {
-	string result=s;
-	if(!result.erase(0, result.find_first_not_of(" \t\r\n")).empty())
-		result.erase(result.find_last_not_of(" \t\r\n")+1);
+	string result;
+
+	for(string::const_iterator i=str.begin(); i!=str.end(); ++i)
+	{
+		if(*i=='\n')
+			result+="\\n";
+		else if(*i=='\t')
+			result+="\\t";
+		else if(*i=='\r')
+			result+="\\r";
+		else if(*i=='\b')
+			result+="\\b";
+		else if(*i=='\v')
+			result+="\\v";
+		else if(*i=='\a')
+			result+="\\a";
+		else if(*i=='\f')
+			result+="\\f";
+		else if(*i=='\"')
+			result+="\\\"";
+		else if(*i=='\'')
+			result+="\\\'";
+		else if(*i=='\\')
+			result+="\\\\";
+		else if(static_cast<unsigned char>(*i)<' ' || (escape_8bit && (*i&0x80)))
+		{
+			char buf[4]={'\\', '0'+((*i>>6)&3), '0'+((*i>>3)&7), '0'+(*i&7)};
+			result.append(buf, 4);
+		}
+		else
+			result+=*i;
+	}
+
 	return result;
 }