VuoStringUtilities.cc

Go to the documentation of this file.

 
#include <sstream>
#include <CoreFoundation/CoreFoundation.h>
#include "VuoStringUtilities.hh"
 
extern "C" {
#include "mkdio.h"
}
 
bool VuoStringUtilities::beginsWith(string wholeString, string beginning)
{
    return wholeString.length() >= beginning.length() && wholeString.substr(0, beginning.length()) == beginning;
}
 
bool VuoStringUtilities::endsWith(string wholeString, string ending)
{
    if (wholeString.length() < ending.length())
        return false;
 
    return wholeString.compare(wholeString.length()-ending.length(), ending.length(), ending) == 0;
}
 
string VuoStringUtilities::substrAfter(string wholeString, string beginning)
{
    if (! beginsWith(wholeString, beginning))
        return "";
 
    return wholeString.substr(beginning.length());
}
 
string VuoStringUtilities::substrBefore(string wholeString, string ending)
{
    if (! endsWith(wholeString, ending))
        return "";
 
    return wholeString.substr(0, wholeString.length()-ending.length());
}
 
string VuoStringUtilities::replaceAll(string wholeString, char originalChar, char replacementChar)
{
    string outString = wholeString;
    size_t pos = 0;
    while ((pos = wholeString.find_first_of(originalChar, pos)) != string::npos)
    {
        outString[pos] = replacementChar;
        pos = pos + 1;
    }
    return outString;
}
 
size_t VuoStringUtilities::replaceAll(string &wholeString, string originalSubstring, string replacementSubstring)
{
    size_t replacementCount = 0;
    size_t startPos = 0;
    while ((startPos = wholeString.find(originalSubstring, startPos)) != string::npos)
    {
        wholeString.replace(startPos, originalSubstring.length(), replacementSubstring);
        startPos += replacementSubstring.length();
        ++replacementCount;
    }
    return replacementCount;
}
 
vector<string> VuoStringUtilities::split(const string &wholeString, char delimiter)
{
    vector<string> tokens;
    istringstream iss(wholeString);
    string token;
    while( getline(iss, token, delimiter) )
        tokens.push_back(token);
    return tokens;
}
 
string VuoStringUtilities::join(vector<string> partialStrings, char delimiter)
{
    string delimiterStr(1, delimiter);
    return join(partialStrings, delimiterStr);
}
 
string VuoStringUtilities::join(vector<string> partialStrings, string delimiter)
{
    string wholeString;
    for (vector<string>::iterator i = partialStrings.begin(); i != partialStrings.end(); )
    {
        wholeString += *i;
        if (++i != partialStrings.end())
            wholeString += delimiter;
    }
    return wholeString;
}
 
string VuoStringUtilities::join(set<string> partialStrings, string delimiter)
{
    string wholeString;
    for (set<string>::iterator i = partialStrings.begin(); i != partialStrings.end(); )
    {
        wholeString += *i;
        if (++i != partialStrings.end())
            wholeString += delimiter;
    }
    return wholeString;
}
 
string VuoStringUtilities::trim(string originalString)
{
    string whitespace = " \t\v\n\r\f";
 
    string::size_type begin = originalString.find_first_not_of(whitespace);
    if (begin == std::string::npos)
        return "";
 
    string::size_type end = originalString.find_last_not_of(whitespace);
 
    return originalString.substr(begin, end - begin + 1);
}
 
string VuoStringUtilities::buildCompositionIdentifier(const string &parentCompositionIdentifier, const string &nodeIdentifier)
{
    return parentCompositionIdentifier + "/" + nodeIdentifier;
}
 
string VuoStringUtilities::buildPortIdentifier(const string &nodeIdentifier, const string &portName)
{
    return nodeIdentifier + ":" + portName;
}
 
string VuoStringUtilities::prefixSymbolName(string symbolName, string moduleKey)
{
    return transcodeToIdentifier(moduleKey) + "__" + symbolName;
}
 
string VuoStringUtilities::transcodeToIdentifier(string str)
{
    CFMutableStringRef strCF = CFStringCreateMutable(NULL, 0);
    CFStringAppendCString(strCF, str.c_str(), kCFStringEncodingUTF8);
 
    CFStringNormalize(strCF, kCFStringNormalizationFormD);  // decomposes combining characters, so accents/diacritics are separated from their letters
 
    CFIndex strLength = CFStringGetLength(strCF);
    UniChar *strBuf = (UniChar *)malloc(strLength * sizeof(UniChar));
    if (!strBuf)
    {
        CFRelease(strCF);
        return string();
    }
    CFStringGetCharacters(strCF, CFRangeMake(0, strLength), strBuf);
 
    CFStringRef empty = CFStringCreateWithCString(NULL, "", kCFStringEncodingUTF8);
    CFStringRef underscore = CFStringCreateWithCString(NULL, "_", kCFStringEncodingUTF8);
    CFStringRef doubleUnderscore = CFStringCreateWithCString(NULL, "__", kCFStringEncodingUTF8);
    if (!empty || !underscore || !doubleUnderscore)
        return string();
 
    for (CFIndex i = strLength-1; i >= 0; --i)
    {
        UniChar c = strBuf[i];
 
        CFStringRef replacement = NULL;
        if (c > 127)  // non-ASCII
            replacement = empty;
        else if (c == '.' || isspace(c))
            replacement = underscore;
        else if (c == '/' || c == ':')
            replacement = doubleUnderscore;
        else if (! isValidCharInIdentifier(c))
            replacement = empty;
 
        if (replacement)
            CFStringReplace(strCF, CFRangeMake(i, 1), replacement);
    }
 
    // https://stackoverflow.com/questions/1609565/whats-the-cfstring-equiv-of-nsstrings-utf8string
 
    const char *useUTF8StringPtr = NULL;
    char *freeUTF8StringPtr = NULL;
 
    if ((useUTF8StringPtr = CFStringGetCStringPtr(strCF, kCFStringEncodingUTF8)) == NULL)
    {
        CFIndex maxBytes = 4 * strLength + 1;
        freeUTF8StringPtr = (char *)malloc(maxBytes);
        CFStringGetCString(strCF, freeUTF8StringPtr, maxBytes, kCFStringEncodingUTF8);
        useUTF8StringPtr = freeUTF8StringPtr;
    }
 
    string ret = useUTF8StringPtr;
 
    if (freeUTF8StringPtr != NULL)
        free(freeUTF8StringPtr);
 
    CFRelease(strCF);
    CFRelease(empty);
    CFRelease(underscore);
    CFRelease(doubleUnderscore);
    free(strBuf);
 
    return ret;
}
 
bool VuoStringUtilities::isValidCharInIdentifier(char ch)
{
    return isalnum(ch) || ch == '_';
}
 
string VuoStringUtilities::transcodeToGraphvizIdentifier(const string &originalString)
{
    string escapedString = originalString;
    for (string::size_type i = 0; (i = escapedString.find("\\", i)) != std::string::npos; i += 2)
        escapedString.replace(i, 1, "\\\\");
    for (string::size_type i = 0; (i = escapedString.find("\"", i)) != std::string::npos; i += 2)
        escapedString.replace(i, 1, "\\\"");
    for (string::size_type i = 0; (i = escapedString.find("{", i)) != std::string::npos; i += 2)
        escapedString.replace(i, 1, "\\{");
    for (string::size_type i = 0; (i = escapedString.find("}", i)) != std::string::npos; i += 2)
        escapedString.replace(i, 1, "\\}");
    for (string::size_type i = 0; (i = escapedString.find("<", i)) != std::string::npos; i += 2)
        escapedString.replace(i, 1, "\\<");
    for (string::size_type i = 0; (i = escapedString.find(">", i)) != std::string::npos; i += 2)
        escapedString.replace(i, 1, "\\>");
    for (string::size_type i = 0; (i = escapedString.find("|", i)) != std::string::npos; i += 2)
        escapedString.replace(i, 1, "\\|");
    for (string::size_type i = 0; (i = escapedString.find("  ", i)) != std::string::npos; i += 3)
        escapedString.replace(i, 2, " \\ ");
    return escapedString;
}
 
string VuoStringUtilities::transcodeFromGraphvizIdentifier(const string &graphvizIdentifier)
{
    string unescapedString;
    bool inEscape = false;
    for (string::const_iterator i = graphvizIdentifier.begin(); i != graphvizIdentifier.end(); ++i)
    {
        if (inEscape)
        {
            inEscape = false;
            unescapedString += *i;
            continue;
        }
 
        if (*i == '\\')
        {
            inEscape = true;
            continue;
        }
 
        unescapedString  += *i;
    }
    return unescapedString;
}
 
string VuoStringUtilities::formUniqueIdentifier(set<string> &takenIdentifiers,
                                                const string &preferredIdentifier, const string &identifierPrefix)
{
    auto isIdentifierAvailable = [&takenIdentifiers] (const string &identifier)
    {
        return takenIdentifiers.find(identifier) == takenIdentifiers.end();
    };
 
    string uniqueIdentifier = formUniqueIdentifier(isIdentifierAvailable, preferredIdentifier, identifierPrefix);
    takenIdentifiers.insert(uniqueIdentifier);
    return uniqueIdentifier;
}
 
string VuoStringUtilities::formUniqueIdentifier(std::function<bool(const string &)> isIdentifierAvailable,
                                                const string &preferredIdentifier, const string &identifierPrefix)
{
    string unique = preferredIdentifier;
    string prefix = (! identifierPrefix.empty() ? identifierPrefix : preferredIdentifier);
    int suffix = 2;
 
    while (! isIdentifierAvailable(unique))
    {
        ostringstream oss;
        oss << prefix << suffix++;
        unique = oss.str();
    }
 
    return unique;
}
 
string VuoStringUtilities::generateHtmlFromMarkdown(const string &markdownString)
{
    MMIOT *doc = mkd_string(markdownString.c_str(), markdownString.length(), MKD_NOPANTS);
    mkd_compile(doc, 0);
    char *html;
    mkd_document(doc, &html);
    string htmlString(html);
    mkd_cleanup(doc);
 
    // Remove the final linebreak from code blocks,
    // since Qt (unlike typical browser rendering engines) considers that whitespace significant.
    replaceAll(htmlString, "\n</code></pre>", "</code></pre>");
 
    return htmlString;
}
 
string VuoStringUtilities::generateHtmlFromMarkdownLine(const string &markdownString)
{
    size_t length = markdownString.length();
    if (!length)
        return "";
 
    char *html;
    mkd_line((char *)markdownString.c_str(), length, &html, MKD_NOPANTS);
    string htmlString(html);
    free(html);
    return htmlString;
}
 
string VuoStringUtilities::convertToCamelCase(const string &originalString,
                                              bool forceFirstLetterToUpper, bool forceFirstLetterToLower, bool forceInterveningLettersToLower,
                                              bool allowSeparatorDots)
{
    string camelCaseString;
    bool first = true;
    bool uppercaseNext = forceFirstLetterToUpper;
    bool lowercaseNext = forceFirstLetterToLower;
    bool previousWasDot = false;
    for (string::const_iterator i = originalString.begin(); i != originalString.end(); ++i)
    {
        if (first && !isalpha(*i))
            continue;
        first = false;
 
        bool isDot = *i == '.';
        if (allowSeparatorDots && isDot)
        {
            if (previousWasDot)
                continue;
            uppercaseNext = false;
        }
        else if (!isalnum(*i))
        {
            uppercaseNext = true;
            continue;
        }
 
        if (uppercaseNext)
            camelCaseString += toupper(*i);
        else if (lowercaseNext)
            camelCaseString += tolower(*i);
        else
            camelCaseString += *i;
 
        uppercaseNext = false;
        lowercaseNext = forceInterveningLettersToLower;
        previousWasDot = isDot;
    }
 
    // Trim trailing dots.
    if (allowSeparatorDots)
        while (endsWith(camelCaseString, "."))
            camelCaseString = substrBefore(camelCaseString, ".");
 
    return camelCaseString;
}
 
string VuoStringUtilities::expandCamelCase(string camelCaseString)
{
    // Only apply these transformations if the whole string matches,
    // since they may appear as substrings in contexts where they shouldn't be all-caps.
    if (camelCaseString == "x")
        return "X";
    else if (camelCaseString == "y")
        return "Y";
    else if (camelCaseString == "z")
        return "Z";
    else if (camelCaseString == "w")
        return "W";
    else if (camelCaseString == "xy")
        return "XY";
    else if (camelCaseString == "osc")
        return "OSC";
 
    string out;
    out += toupper(camelCaseString[0]);
 
    size_t length = camelCaseString.length();
    for (int i = 1; i < length; ++i)
    {
        char c = camelCaseString[i];
        if (isupper(c) || (isdigit(c) && !isdigit(camelCaseString[i-1])))
            out += " ";
        out += c;
    }
 
    string allCaps[]{
        "2d",
        "3d",
        "4d",
        "Xyzw",
        "Xyz",
        "Rgbaw",
        "Rgba",
        "Rgbw",
        "Rgb",
        "Wwcw",
        "Cmy",
        "Hsl",
        "Hdmi",
        "Sdi",
        "Ntsc",
//      "Pal",  // Appears in Leap Motion "Palm Velocity".
        "Url",
        "Midi",
        "Rss",
        "Csv",
        "Tsv",
        "Ascii",
        "Json",
        "Xml",
        "Dmx",
    };
    for (auto lower : allCaps)
    {
        string upper = lower;
        std::transform(upper.begin(), upper.end(), upper.begin(), ::toupper);
        VuoStringUtilities::replaceAll(out, lower, upper);
    }
 
    return out;
}
 
string VuoStringUtilities::makeRandomHash(int length)
{
    static const char alphanum[] =
        "0123456789"
        "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
        "abcdefghijklmnopqrstuvwxyz";
 
    string hash(length, 0);
    for (int i = 0; i < length; ++i)
        hash[i] = alphanum[arc4random_uniform(sizeof(alphanum)-1)];
 
    return hash;
}
 
const std::locale VuoStringUtilities::locale;
const std::collate<char> &VuoStringUtilities::collate = std::use_facet<std::collate<char> >(VuoStringUtilities::locale);