Not logged in.  Login/Logout/Register | List snippets | | Create snippet | Upload image | Upload data

88
LINES

< > BotCompany Repo | #1013521 - javaTokWithExisting

JavaX fragment (include)

please include function javaTok.

static List<String> javaTokWithExisting(String s, L<S> existing) {
  ++javaTok_n;
  int nExisting = javaTok_opt && existing != null ? existing.size() : 0;
  ArrayList<String> tok = existing != null ? new ArrayList(nExisting) : new ArrayList();
  int l = s.length();
  
  int i = 0, n = 0;
  while (i < l) {
    int j = i;
    char c, d;
    
    // scan for whitespace
    while (j < l) {
      c = s.charAt(j);
      d = j+1 >= l ? '\0' : s.charAt(j+1);
      if (c == ' ' || c == '\t' || c == '\r' || c == '\n')
        ++j;
      else if (c == '/' && d == '*') {
        do ++j; while (j < l && !s.substring(j, Math.min(j+2, l)).equals("*/"));
        j = Math.min(j+2, l);
      } else if (c == '/' && d == '/') {
        do ++j; while (j < l && "\r\n".indexOf(s.charAt(j)) < 0);
      } else
        break;
    }
    
    if (n < nExisting && javaTokWithExisting_isCopyable(existing.get(n), s, i, j))
      tok.add(existing.get(n));
    else
      tok.add(javaTok_substringN(s, i, j));
    ++n;
    i = j;
    if (i >= l) break;
    c = s.charAt(i);
    d = i+1 >= l ? '\0' : s.charAt(i+1);

    // scan for non-whitespace
    
    // Special JavaX syntax: 'identifier
    if (c == '\'' && Character.isJavaIdentifierStart(d) && i+2 < l && "'\\".indexOf(s.charAt(i+2)) < 0) {
      j += 2;
      while (j < l && Character.isJavaIdentifierPart(s.charAt(j)))
        ++j;
    } else if (c == '\'' || c == '"') {
      char opener = c;
      ++j;
      while (j < l) {
        if (s.charAt(j) == opener /*|| s.charAt(j) == '\n'*/) { // allow multi-line strings
          ++j;
          break;
        } else if (s.charAt(j) == '\\' && j+1 < l)
          j += 2;
        else
          ++j;
      }
    } else if (Character.isJavaIdentifierStart(c))
      do ++j; while (j < l && (Character.isJavaIdentifierPart(s.charAt(j)) || "'".indexOf(s.charAt(j)) >= 0)); // for stuff like "don't"
    else if (Character.isDigit(c)) {
      do ++j; while (j < l && Character.isDigit(s.charAt(j)));
      if (j < l && s.charAt(j) == 'L') ++j; // Long constants like 1L
    } else if (c == '[' && d == '[') {
      do ++j; while (j+1 < l && !s.substring(j, j+2).equals("]]"));
      j = Math.min(j+2, l);
    } else if (c == '[' && d == '=' && i+2 < l && s.charAt(i+2) == '[') {
      do ++j; while (j+2 < l && !s.substring(j, j+3).equals("]=]"));
      j = Math.min(j+3, l);
    } else
      ++j;
      
    if (n < nExisting && javaTokWithExisting_isCopyable(existing.get(n), s, i, j))
      tok.add(existing.get(n));
    else
      tok.add(javaTok_substringC(s, i, j));
    ++n;
    i = j;
  }
  
  if ((tok.size() % 2) == 0) tok.add("");
  javaTok_elements += tok.size();
  return tok;
}

static bool javaTokWithExisting_isCopyable(S t, S s, int i, int j) {
  ret t.length() == j-i
    && s.regionMatches(i, t, 0, j-i); // << could be left out, but that's brave
}

Author comment

Began life as a copy of #1000688

download  show line numbers  debug dex   

Travelled to 9 computer(s): aoiabmzegqzx, cbybwowwnfue, cfunsshuasjs, gwrvuhgaqvyk, ishqpsrjomds, lpdgvwnxivlt, mqqgnosmbjvj, tslmcundralx, tvejysmllsmz

No comments. add comment

Snippet ID: #1013521
Snippet name: javaTokWithExisting
Eternal ID of this version: #1013521/4
Text MD5: aac89d3011d1086fa365589204239426
Author: stefan
Category:
Type: JavaX fragment (include)
Public (visible to everyone): Yes
Archived (hidden from active list): No
Created/modified: 2018-01-08 03:16:17
Source code size: 2975 bytes / 88 lines
Pitched / IR pitched: No / No
Views / Downloads: 196 / 215
Version history: 3 change(s)
Referenced in: [show references]

Formerly at http://tinybrain.de/1013521 & http://1013521.tinybrain.de