generic-library/poco
1
0
Fork 0
mirror of https://github.com/pocoproject/poco.git synced 2025-10-28 11:31:53 +01:00
Code Issues Projects Releases Wiki Activity

feat(Foundation): Include utf8proc for enhanced Unicode support (#4710)

Browse Source
This commit is contained in:
Günter Obiltschnig
2024-09-27 12:46:13 +02:00
parent 6faf90773a
commit 92e06497f5
10 changed files with 18726 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
Foundation/include/Poco/UTF8String.h
View File

@@ -37,6 +37,15 @@ struct Foundation_API UTF8
/// removeBOM() removes the UTF-8 Byte Order Mark sequence (0xEF, 0xBB, 0xBF)
/// from the beginning of the given string, if it's there.
{
enum NormalizationForm
/// Normalization form for normalize().
{
NORMALIZATION_FORM_D, /// Canonical Decomposition
NORMALIZATION_FORM_C, /// Canonical Decomposition, followed by Canonical Composition
NORMALIZATION_FORM_KD, /// Compatibility Decomposition
NORMALIZATION_FORM_KC /// Compatibility Decomposition, followed by Canonical Composition
};
static int icompare(const std::string& str, std::string::size_type pos, std::string::size_type n, std::string::const_iterator it2, std::string::const_iterator end2);
static int icompare(const std::string& str1, const std::string& str2);
static int icompare(const std::string& str1, std::string::size_type n1, const std::string& str2, std::string::size_type n2);
@@ -74,6 +83,14 @@ struct Foundation_API UTF8
static std::string unescape(const std::string::const_iterator& begin, const std::string::const_iterator& end);
/// Creates an UTF8 string from a string that contains escaped characters.
static std::string normalize(const std::string& s, NormalizationForm form);
/// Normalizes the given UTF8 string according to the given normalization form.
/// Returns the normalized UTF8 string.
static std::string normalize(const std::string::const_iterator& begin, const std::string::const_iterator& end, NormalizationForm form);
/// Normalizes the given UTF8 string range according to the given normalization form.
/// Returns the normalized UTF8 string.
};