summaryrefslogtreecommitdiffstats
path: root/third-party/utf8cpp/samples
diff options
context:
space:
mode:
authorDaniel Baumann <daniel.baumann@progress-linux.org>2024-04-28 12:34:54 +0000
committerDaniel Baumann <daniel.baumann@progress-linux.org>2024-04-28 12:34:54 +0000
commit0915b3ef56dfac3113cce55a59a5765dc94976be (patch)
treea8fea11d50b4f083e1bf0f90025ece7f0824784a /third-party/utf8cpp/samples
parentInitial commit. (diff)
downloadicinga2-0915b3ef56dfac3113cce55a59a5765dc94976be.tar.xz
icinga2-0915b3ef56dfac3113cce55a59a5765dc94976be.zip
Adding upstream version 2.13.6.upstream/2.13.6upstream
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to '')
-rw-r--r--third-party/utf8cpp/samples/docsample.cpp52
1 files changed, 52 insertions, 0 deletions
diff --git a/third-party/utf8cpp/samples/docsample.cpp b/third-party/utf8cpp/samples/docsample.cpp
new file mode 100644
index 0000000..6f8953b
--- /dev/null
+++ b/third-party/utf8cpp/samples/docsample.cpp
@@ -0,0 +1,52 @@
+#include "../source/utf8.h"
+#include <iostream>
+#include <fstream>
+#include <string>
+#include <vector>
+
+
+using namespace std;
+
+int main(int argc, char** argv)
+{
+ if (argc != 2) {
+ cout << "\nUsage: docsample filename\n";
+ return 0;
+ }
+ const char* test_file_path = argv[1];
+ // Open the test file (must be UTF-8 encoded)
+ ifstream fs8(test_file_path);
+ if (!fs8.is_open()) {
+ cout << "Could not open " << test_file_path << endl;
+ return 0;
+ }
+
+ unsigned line_count = 1;
+ string line;
+ // Play with all the lines in the file
+ while (getline(fs8, line)) {
+ // check for invalid utf-8 (for a simple yes/no check, there is also utf8::is_valid function)
+ string::iterator end_it = utf8::find_invalid(line.begin(), line.end());
+ if (end_it != line.end()) {
+ cout << "Invalid UTF-8 encoding detected at line " << line_count << "\n";
+ cout << "This part is fine: " << string(line.begin(), end_it) << "\n";
+ }
+ // Get the line length (at least for the valid part)
+ int length = utf8::distance(line.begin(), end_it);
+ cout << "Length of line " << line_count << " is " << length << "\n";
+
+ // Convert it to utf-16
+ vector<unsigned short> utf16line;
+ utf8::utf8to16(line.begin(), end_it, back_inserter(utf16line));
+ // And back to utf-8;
+ string utf8line;
+ utf8::utf16to8(utf16line.begin(), utf16line.end(), back_inserter(utf8line));
+ // Confirm that the conversion went OK:
+ if (utf8line != string(line.begin(), end_it))
+ cout << "Error in UTF-16 conversion at line: " << line_count << "\n";
+
+ line_count++;
+ }
+
+ return 0;
+}