lexer.java.rl.erb

package gherkin.lexer;

import java.io.UnsupportedEncodingException;
import java.util.List;
import java.util.ArrayList;
import java.util.regex.Pattern;
import gherkin.lexer.Lexer;
import gherkin.lexer.Listener;
import gherkin.lexer.LexingError;

public class <%= @i18n.underscored_iso_code.capitalize %> implements Lexer {
  %%{
    machine lexer;
    alphtype byte;

    action begin_content {
      contentStart = p;
      currentLine = lineNumber;
      if(keyword != null) {
        startCol = p - lastNewline - (keyword.length() + 1);
      }
    }

    action start_docstring {
      currentLine = lineNumber;
      startCol = p - lastNewline;
    }
    
    action begin_docstring_content {
      contentStart = p;
    }

    action start_docstring_content_type {
      docstringContentTypeStart = p;
    }

    action end_docstring_content_type {
      docstringContentTypeEnd = p;
    }

    action store_docstring_content {
      String con = unindent(startCol, substring(data, contentStart, nextKeywordStart-1).replaceFirst("(\\r?\\n)?([\\t ])*\\Z", "").replace("\\\"\\\"\\\"", "\"\"\""));
      String conType = substring(data, docstringContentTypeStart, docstringContentTypeEnd).trim();
      listener.docString(conType, con, currentLine);
    }

    action store_feature_content {
      String[] nameDescription = nameAndUnindentedDescription(startCol, keywordContent(data, p, eof, nextKeywordStart, contentStart));
      listener.feature(keyword, nameDescription[0], nameDescription[1], currentLine);
      if(nextKeywordStart != -1) p = nextKeywordStart - 1;
      nextKeywordStart = -1;
    }

    action store_background_content {
      String[] nameDescription = nameAndUnindentedDescription(startCol, keywordContent(data, p, eof, nextKeywordStart, contentStart));
      listener.background(keyword, nameDescription[0], nameDescription[1], currentLine);
      if(nextKeywordStart != -1) p = nextKeywordStart - 1;
      nextKeywordStart = -1;
    }

    action store_scenario_content {
      String[] nameDescription = nameAndUnindentedDescription(startCol, keywordContent(data, p, eof, nextKeywordStart, contentStart));
      listener.scenario(keyword, nameDescription[0], nameDescription[1], currentLine);
      if(nextKeywordStart != -1) p = nextKeywordStart - 1;
      nextKeywordStart = -1;
    }

    action store_scenario_outline_content {
      String[] nameDescription = nameAndUnindentedDescription(startCol, keywordContent(data, p, eof, nextKeywordStart, contentStart));
      listener.scenarioOutline(keyword, nameDescription[0], nameDescription[1], currentLine);
      if(nextKeywordStart != -1) p = nextKeywordStart - 1;
      nextKeywordStart = -1;
    }

    action store_examples_content {
      String[] nameDescription = nameAndUnindentedDescription(startCol, keywordContent(data, p, eof, nextKeywordStart, contentStart));
      listener.examples(keyword, nameDescription[0], nameDescription[1], currentLine);
      if(nextKeywordStart != -1) p = nextKeywordStart - 1;
      nextKeywordStart = -1;
    }

    action store_step_content {
      listener.step(keyword, substring(data, contentStart, p).trim(), currentLine);
    }

    action store_comment_content {
      listener.comment(substring(data, contentStart, p).trim(), lineNumber);
      keywordStart = -1;
    }

    action store_tag_content {
      listener.tag(substring(data, contentStart, p).trim(), currentLine);
      keywordStart = -1;
    }

    action inc_line_number {
      lineNumber++;
    }

    action last_newline {
      lastNewline = p + 1;
    }

    action start_keyword {
      if(keywordStart == -1) keywordStart = p;
    }

    action end_keyword {
      keyword = substring(data, keywordStart, p).replaceFirst(":$","");
      keywordStart = -1;
    }

    action next_keyword_start {
      nextKeywordStart = p;
    }

    action start_row {
      p = p - 1;
      currentRow = new ArrayList<String>();
      currentLine = lineNumber;
    }

    action begin_cell_content {
      contentStart = p;
    }

    action store_cell_content {
      String con = substring(data, contentStart, p).trim();
      currentRow.add(con
        .replace("\\|", "|")
        .replace("\\n", "\n")
        .replace("\\\\", "\\")
      );
    }

    action store_row {
      listener.row(currentRow, currentLine);
    }

    action end_feature {
      if(cs < lexer_first_final) {
        String content = currentLineContent(data, lastNewline);
        throw new LexingError("Lexing error on line " + lineNumber + ": '" + content + "'. See http://wiki.github.com/cucumber/gherkin/lexingerror for more information.");
      } else {
        listener.eof();
      }
    }

    include lexer_common "lexer_common.<%= @i18n.underscored_iso_code %>.rl"; 
  }%%

  private final Listener listener;

  public <%= @i18n.underscored_iso_code.capitalize %>(Listener listener) {
    this.listener = listener;
  }

  %% write data noerror;

  public void scan(String source)  {
    String input = source + "\n%_FEATURE_END_%";
    byte[] data = null;
    try {
      data = input.getBytes("UTF-8");
    } catch(UnsupportedEncodingException e) {
      throw new RuntimeException(e);
    }
    int cs, p = 0, pe = data.length;
    int eof = pe;

    int lineNumber = 1;
    int lastNewline = 0;

    int contentStart = -1;
    int currentLine = -1;
    int docstringContentTypeStart = -1;
    int docstringContentTypeEnd = -1;
    int startCol = -1;
    int nextKeywordStart = -1;
    int keywordStart = -1;
    String keyword = null;
    List<String> currentRow = null;

    %% write init;
    %% write exec;
  }

  private String keywordContent(byte[] data, int p, int eof, int nextKeywordStart, int contentStart) {
    int endPoint = (nextKeywordStart == -1 || (p == eof)) ? p : nextKeywordStart;
    return substring(data, contentStart, endPoint);
  }

  private String[] nameAndUnindentedDescription(int startCol, String text) {
    String[] lines = text.split("\n");
    String name = lines.length > 0 ? lines[0].trim() : "";
    StringBuffer description = new StringBuffer();
    for(int i = 1; i < lines.length; i++) {
      description.append(lines[i]);
      description.append("\n");
    }
    return new String[]{name, unindent(startCol+2, description.toString()).replaceAll("\\s+$", "")};
  }

  private String unindent(int startCol, String text) {
    return Pattern.compile("^[\t ]{0," + startCol + "}", Pattern.MULTILINE).matcher(text).replaceAll("");
  }

  private String currentLineContent(byte[] data, int lastNewline) {
    return substring(data, lastNewline, data.length).trim();
  }

  private String substring(byte[] data, int start, int end) {
    try {
      return new String(data, start, end-start, "utf-8");
    } catch(java.io.UnsupportedEncodingException e) {
      throw new RuntimeException("Internal error", e);
    }
  }
}