Introduce characters utilities in rtl/character.hxx

Signed-off-by: Stephan Bergmann <sbergman@redhat.com>, with slight modifications
to sal/inc/rtl/character.hxx:
* Replaced "#pragma once" with explicit include guard for now.
* Missing includes.
* Cosmetic clean-up.

Change-Id: I94d01cd4e766f92c70f941839a67101fa2c97654

4 files changed, 155 insertions, 29 deletions

diff --git a/sal/ZipPackage_sal_odk_headers.mk b/sal/ZipPackage_sal_odk_headers.mk
index 458dbe5f03ca..696bc60c899a 100644
--- a/sal/ZipPackage_sal_odk_headers.mk
+++ b/sal/ZipPackage_sal_odk_headers.mk
@@ -32,6 +32,7 @@ $(eval $(call gb_ZipPackage_add_files,sal_odk_headers,inc/rtl,include/rtl,\
 	rtl/bootstrap.hxx \
 	rtl/byteseq.h \
 	rtl/byteseq.hxx \
+	rtl/character.hxx \
 	rtl/cipher.h \
 	rtl/crc.h \
 	rtl/digest.h \
diff --git a/sal/inc/rtl/character.hxx b/sal/inc/rtl/character.hxx
new file mode 100644
index 000000000000..5e3a8ff4234e
--- /dev/null
+++ b/sal/inc/rtl/character.hxx
@@ -0,0 +1,144 @@
+/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
+/*
+ * This file is part of the LibreOffice project.
+ *
+ * This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at http://mozilla.org/MPL/2.0/.
+ *
+ * This file incorporates work covered by the following license notice:
+ *
+ *   Licensed to the Apache Software Foundation (ASF) under one or more
+ *   contributor license agreements. See the NOTICE file distributed
+ *   with this work for additional information regarding copyright
+ *   ownership. The ASF licenses this file to you under the Apache
+ *   License, Version 2.0 (the "License"); you may not use this file
+ *   except in compliance with the License. You may obtain a copy of
+ *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
+ */
+
+#ifndef INCLUDED_RTL_CHARACTER_HXX
+#define INCLUDED_RTL_CHARACTER_HXX
+
+#include "sal/config.h"
+
+#include "sal/types.h"
+
+namespace rtl
+{
+/** Check for ASCII character.
+
+    @param nUtf32  Some UCS-4 character.
+
+    @return  True if nChar is a ASCII character (0x00--0x7F).
+
+    @since LibreOffice 4.1
+ */
+inline bool isAscii(sal_uInt32 nUtf32)
+{
+    return nUtf32 <= 0x7F;
+}
+
+/** Check for ASCII lower case character.
+
+    @param nUtf32  Some UCS-4 character.
+
+    @return  True if nChar is a US-ASCII lower case alphabetic character
+    (ASCII 'a'--'z').
+
+    @since LibreOffice 4.1
+ */
+inline bool isAsciiLowerCase(sal_uInt32 nUtf32)
+{
+    return nUtf32 >= 'a' && nUtf32 <= 'z';
+}
+
+/** Check for US-ASCII upper case character.
+
+    @param nUtf32  Some UCS-4 character.
+
+    @return  True if nChar is a US-ASCII upper case alphabetic character
+    (US-ASCII 'A'--'Z').
+
+    @since LibreOffice 4.1
+ */
+inline bool isAsciiUpperCase(sal_uInt32 nUtf32)
+{
+    return nUtf32 >= 'A' && nUtf32 <= 'Z';
+}
+
+/** Check for ASCII alphanumeric character.
+
+    @param nUtf32  Some UCS-4 character.
+
+    @return  True if nUtf32 is a US-ASCII alphanumeric character
+    (ASCII '0'--'9', 'A'--'Z' or 'a'--'z').
+
+    @since LibreOffice 4.1
+ */
+inline bool isAsciiAlpha(sal_uInt32 nUtf32)
+{
+    return isAsciiLowerCase(nUtf32) || isAsciiUpperCase(nUtf32);
+}
+
+/** Check for ASCII digit character.
+
+    @param nUtf32  Some UCS-4 character.
+
+    @return  True if nChar is a ASCII (decimal) digit character
+    (ASCII '0'--'9').
+
+    @since LibreOffice 4.1
+ */
+inline bool isAsciiDigit(sal_uInt32 nUtf32)
+{
+    return nUtf32 >= '0' && nUtf32 <= '9';
+}
+
+/** Check for US-ASCII alphanumeric character.
+
+    @param nUtf32  Some UCS-4 character.
+
+    @return  True if nChar is a US-ASCII alphanumeric character (US-ASCII
+    '0'--'9', 'A'--'Z' or 'a'--'z').
+
+    @since LibreOffice 4.1
+ */
+inline bool isAsciiAlphanumeric(sal_uInt32 nUtf32)
+{
+    return isAsciiDigit(nUtf32) || isAsciiAlpha(nUtf32);
+}
+
+/** Check for US-ASCII canonic hexadecimal digit character.
+
+    @param nUtf32  Some UCS-4 character.
+
+    @return  True if nChar is a US-ASCII canonic (i.e., upper case)
+    hexadecimal digit character (US-ASCII '0'--'9' or 'A'--'F').
+
+    @since LibreOffice 4.1
+ */
+inline bool isAsciiCanonicHexDigit(sal_uInt32 nUtf32)
+{
+    return isAsciiDigit(nUtf32) || (nUtf32 >= 'A' && nUtf32 <= 'F');
+}
+
+/** Check for US-ASCII hexadecimal digit character.
+
+    @param nUtf32  Some UCS-4 character.
+
+    @return  True if nChar is a US-ASCII hexadecimal digit character (US-
+    ASCII '0'--'9', 'A'--'F', 'a'--'f').
+
+    @since LibreOffice 4.1
+ */
+inline bool isAsciiHexDigit(sal_uInt32 nUtf32)
+{
+    return isAsciiCanonicHexDigit(nUtf32) || (nUtf32 >= 'a' && nUtf32 <= 'f');
+}
+
+}//rtl namespace
+
+#endif
+
+/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
diff --git a/sal/rtl/math.cxx b/sal/rtl/math.cxx
index c0d93b6a85af..590ea0ee46b0 100644
--- a/sal/rtl/math.cxx
+++ b/sal/rtl/math.cxx
@@ -22,6 +22,7 @@
 
 #include "osl/diagnose.h"
 #include "rtl/alloc.h"
+#include "rtl/character.hxx"
 #include "rtl/math.hxx"
 #include "rtl/strbuf.h"
 #include "rtl/string.h"
@@ -689,13 +690,6 @@ inline bool long10Overflow( long& nExp, int nAdd )
     return false;
 }
 
-// We are only concerned about ASCII arabic numerical digits here
-template< typename CharT >
-inline bool isDigit( CharT c )
-{
-    return 0x30 <= c && c <= 0x39;
-}
-
 template< typename CharT >
 inline double stringToDouble(CharT const * pBegin, CharT const * pEnd,
                              CharT cDecSeparator, CharT cGroupSeparator,
@@ -755,7 +749,7 @@ inline double stringToDouble(CharT const * pBegin, CharT const * pEnd,
         for (; p != pEnd; ++p)
         {
             CharT c = *p;
-            if (isDigit(c))
+            if (rtl::isAsciiDigit(c))
             {
                 fVal = fVal * 10.0 + static_cast< double >( c - CharT('0') );
                 ++nValExp;
@@ -783,7 +777,7 @@ inline double stringToDouble(CharT const * pBegin, CharT const * pEnd,
             for (; p != pEnd; ++p)
             {
                 CharT c = *p;
-                if (!isDigit(c))
+                if (!rtl::isAsciiDigit(c))
                     break;
                 if ( nDigs < nSigs )
                 {   // further digits (more than nSigs) don't have any
@@ -821,7 +815,7 @@ inline double stringToDouble(CharT const * pBegin, CharT const * pEnd,
             if ( fVal == 0.0 )
             {   // no matter what follows, zero stays zero, but carry on the
                 // offset
-                while (p != pEnd && isDigit(*p))
+                while (p != pEnd && rtl::isAsciiDigit(*p))
                     ++p;
             }
             else
@@ -831,7 +825,7 @@ inline double stringToDouble(CharT const * pBegin, CharT const * pEnd,
                 for (; p != pEnd; ++p)
                 {
                     CharT c = *p;
-                    if (!isDigit(c))
+                    if (!rtl::isAsciiDigit(c))
                         break;
                     int i = c - CharT('0');
                     if ( long10Overflow( nExp, i ) )
@@ -876,7 +870,7 @@ inline double stringToDouble(CharT const * pBegin, CharT const * pEnd,
                 fVal = HUGE_VAL;
                 eStatus = rtl_math_ConversionStatus_OutOfRange;
                 // Eat any further digits:
-                while (p != pEnd && isDigit(*p))
+                while (p != pEnd && rtl::isAsciiDigit(*p))
                     ++p;
             }
             else if (pEnd - p >= 4 && p[1] == CharT('N') && p[2] == CharT('A')
@@ -897,7 +891,7 @@ inline double stringToDouble(CharT const * pBegin, CharT const * pEnd,
                     bSign = false; // don't negate again
                 }
                 // Eat any further digits:
-                while (p != pEnd && isDigit(*p))
+                while (p != pEnd && rtl::isAsciiDigit(*p))
                     ++p;
             }
         }
diff --git a/sal/rtl/uri.cxx b/sal/rtl/uri.cxx
index 9edd087ca335..7e4402f785c8 100644
--- a/sal/rtl/uri.cxx
+++ b/sal/rtl/uri.cxx
@@ -20,6 +20,7 @@
 #include "surrogates.hxx"
 
 #include "osl/diagnose.h"
+#include "rtl/character.hxx"
 #include "rtl/strbuf.hxx"
 #include "rtl/textenc.h"
 #include "rtl/textcvt.h"
@@ -39,20 +40,6 @@ std::size_t const nCharClassSize = 128;
 
 sal_Unicode const cEscapePrefix = 0x25; // '%'
 
-inline bool isDigit(sal_uInt32 nUtf32)
-{
-    return nUtf32 >= 0x30 && nUtf32 <= 0x39; // '0'--'9'
-}
-
-inline bool isAlpha(sal_uInt32 nUtf32)
-{
-    // 'A'--'Z', 'a'--'z'
-    return (
-            (nUtf32 >= 0x41 && nUtf32 <= 0x5A) ||
-            (nUtf32 >= 0x61 && nUtf32 <= 0x7A)
-           );
-}
-
 inline bool isHighSurrogate(sal_uInt32 nUtf16)
 {
     return SAL_RTL_IS_HIGH_SURROGATE(nUtf16);
@@ -376,7 +363,7 @@ void parseUriRef(rtl_uString const * pUriRef, Components * pComponents)
     sal_Unicode const * pEnd = pBegin + pUriRef->length;
     sal_Unicode const * pPos = pBegin;
 
-    if (pPos != pEnd && isAlpha(*pPos))
+    if (pPos != pEnd && rtl::isAsciiAlpha(*pPos))
     {
         for (sal_Unicode const * p = pPos + 1; p != pEnd; ++p)
         {
@@ -387,7 +374,7 @@ void parseUriRef(rtl_uString const * pUriRef, Components * pComponents)
                 pPos = p;
                 break;
             }
-            else if (!isAlpha(*p) && !isDigit(*p) && *p != '+' && *p != '-'
+            else if (!rtl::isAsciiAlphanumeric(*p) && *p != '+' && *p != '-'
                      && *p != '.')
             {
                 break;