All files / src/abap/2_statements statement_parser.ts

98.7% Statements 303/307
97.56% Branches 80/82
100% Functions 17/17
98.7% Lines 303/307

Press n or j to go to the next uncovered block, b, p or k for the previous block.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 3071x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 293x 293x     293x 300x 129x 300x 171x 171x 300x 293x 1x 1x 1x 14805x 14805x 770x 770x 14035x 14035x 1x 1x 1x 1x 1x 1x 1x 1x 5374x 5374x 5374x 5374x 1x 1x 13657x 13657x 13657x 13657x 1x 1x 5374x 5374x 1x 1x 13657x 13657x 13657x 1356x 1356x 13657x 78976x 78976x 13657x 13657x 13657x 1x 1x 1x 1x 1x 1x 1x 5356x 1x 1x 5356x 5356x 1x 1x 1x 5356x 5356x 5356x 5356x 5356x 5374x 5374x 5374x 5374x 5356x 5356x 5374x 5374x 5374x 311x 311x 5374x 5374x 5356x 5356x 5356x 1x 1x 1x 294x 294x 294x 1492x 1492x 294x 294x 294x 1x 1x 1x 1x 1x 311x 311x 311x 614x 317x 104x 104x 25x 25x 25x 25x 104x 317x 614x 614x 311x 311x 311x 1x 1x 317x 317x 317x 317x 317x 2797x 104x 104x 2797x 2797x 2797x 317x 317x 317x 1x 1x 5374x 5374x 5374x 13861x 13861x 13861x 13861x 13x 13861x 14x 14x 5x 14x 9x 9x 14x 13861x 5374x 5374x 1x 1x 1x 5374x 5374x 5374x 13797x 13797x 5374x 5374x 1x 1x 13901x 13901x 13901x 13901x 13901x 13901x 13901x 14x 14x 13901x 13747x 13642x 13747x     105x 1x 1x 13747x 13901x 13901x 13901x 1x 1x 13642x 13642x 13642x 13642x 13642x 67565x 67565x 32x 67565x 67533x 67533x 67565x 13642x 13642x 13642x 1x 1x 13642x 13642x 13642x 2x 2x 13640x 13642x 27422x 27422x 12475x 12475x 12475x 12475x 27422x 1165x 13642x 2085x 2085x 828x 828x 828x 828x 2085x 337x 337x 337x 1x 1x 1x 1x 1x 5374x 5374x 5374x 5374x 5374x 80328x 140x 140x 140x 80188x 80188x 80188x 80188x 80328x 12876x 12876x 12876x 12876x 80328x 679x 679x 67312x 569x 569x 569x 569x 66633x 2x 2x 80328x 5374x 5374x 102x 102x 5374x 1x
import * as Statements from "./statements";
import * as Expressions from "./expressions";
import * as Tokens from "../1_lexer/tokens";
import {Version} from "../../version";
import {IFile} from "../../files/_ifile";
import {TokenNode, StatementNode} from "../nodes";
import {Artifacts} from "../artifacts";
import {Combi} from "./combi";
import {Unknown, Empty, Comment, NativeSQL, IStatement} from "./statements/_statement";
import {IStatementResult} from "./statement_result";
import {Token} from "../1_lexer/tokens/_token";
import {IABAPLexerResult} from "../1_lexer/lexer_result";
import {ExpandMacros} from "./expand_macros";
import {Pragma} from "../1_lexer/tokens";
 
export const STATEMENT_MAX_TOKENS = 1000;
 
class StatementMap {
  private readonly map: {[index: string]: IStatement[]};
 
  public constructor() {
    this.map = {};
 
    for (const stat of Artifacts.getStatements()) {
      const f = stat.getMatcher().first();
      if (f.length === 0) {
        throw new Error("StatementMap, first must have contents");
      }
      for (const first of f) {
        if (this.map[first]) {
          this.map[first].push(stat);
        } else {
          this.map[first] = [stat];
        }
      }
    }
  }
 
  public lookup(str: string): readonly IStatement[] {
    const res = this.map[str.toUpperCase()];
    if (res === undefined) {
      return [];
    }
    return res;
  }
}
 
class WorkArea {
  private readonly file: IFile;
  public readonly tokens: readonly Token[];
  public statements: StatementNode[];
 
  public constructor(file: IFile, tokens: readonly Token[]) {
    this.file = file;
    this.tokens = tokens;
    this.statements = [];
  }
 
  public addUnknown(pre: Token[], post: Token[], colon: Token | undefined) {
    const st = new StatementNode(new Unknown(), colon);
    st.setChildren(this.tokensToNodes(pre, post));
    this.statements.push(st);
  }
 
  public toResult(): IStatementResult {
    return {file: this.file, tokens: this.tokens, statements: this.statements};
  }
 
  private tokensToNodes(tokens1: Token[], tokens2: Token[]): TokenNode[] {
    const ret: TokenNode[] = [];
 
    for (const t of tokens1) {
      ret.push(new TokenNode(t));
    }
    for (const t of tokens2) {
      ret.push(new TokenNode(t));
    }
 
    return ret;
  }
}
 
export class StatementParser {
  private static map: StatementMap;
  private readonly version: Version;
 
  public constructor(version: Version) {
    if (!StatementParser.map) {
      StatementParser.map = new StatementMap();
    }
    this.version = version;
  }
 
  /** input is one full object */
  public run(input: readonly IABAPLexerResult[], globalMacros: readonly string[]): IStatementResult[] {
    const macros = new ExpandMacros(globalMacros, this.version);
 
    const wa = input.map(i => new WorkArea(i.file, i.tokens));
 
    for (const w of wa) {
      this.process(w);
      this.categorize(w);
      macros.find(w.statements);
    }
 
    for (const w of wa) {
      const res = macros.handleMacros(w.statements);
      w.statements = res.statements;
      if (res.containsUnknown === true) {
        this.lazyUnknown(w);
      }
      this.nativeSQL(w);
    }
 
    return wa.map(w => w.toResult());
  }
 
  // todo, refactor, remove method here and only have in WorkArea class
  private tokensToNodes(tokens: readonly Token[]): TokenNode[] {
    const ret: TokenNode[] = [];
 
    for (const t of tokens) {
      ret.push(new TokenNode(t));
    }
 
    return ret;
  }
 
// tries to split Unknown statements by newlines, when adding/writing a new statement
// in an editor, adding the statement terminator is typically the last thing to do
// note: this will not work if the second statement is a macro call, guess this is okay
  private lazyUnknown(wa: WorkArea) {
    const result: StatementNode[] = [];
 
    for (let statement of wa.statements) {
      if (statement.get() instanceof Unknown) {
        for (const {first, second} of this.buildSplits(statement.getTokens())) {
          const s = this.categorizeStatement(new StatementNode(new Unknown()).setChildren(this.tokensToNodes(second)));
          if (!(s.get() instanceof Unknown)) {
            result.push(new StatementNode(new Unknown()).setChildren(this.tokensToNodes(first)));
            statement = s;
            break;
          }
        }
      }
      result.push(statement);
    }
 
    wa.statements = result;
  }
 
  private buildSplits(tokens: readonly Token[]): {first: Token[], second: Token[]}[] {
    const res: {first: Token[], second: Token[]}[] = [];
    const before: Token[] = [];
    let prevRow = tokens[0].getRow();
 
    for (let i = 0; i < tokens.length; i++) {
      if (tokens[i].getRow() !== prevRow) {
        res.push({first: [...before], second: [...tokens].splice(i)});
      }
      prevRow = tokens[i].getRow();
      before.push(tokens[i]);
    }
 
    return res;
  }
 
  private nativeSQL(wa: WorkArea) {
    let sql = false;
 
    for (let i = 0; i < wa.statements.length; i++) {
      const statement = wa.statements[i];
      const type = statement.get();
      if (type instanceof Statements.ExecSQL
          || (type instanceof Statements.MethodImplementation && statement.findDirectExpression(Expressions.Language))) {
        sql = true;
      } else if (sql === true) {
        if (type instanceof Statements.EndExec
            || type instanceof Statements.EndMethod) {
          sql = false;
        } else if (!(type instanceof Comment)) {
          wa.statements[i] = new StatementNode(new NativeSQL()).setChildren(this.tokensToNodes(statement.getTokens()));
        }
      }
    }
 
  }
 
// for each statement, run statement matchers to figure out which kind of statement it is
  private categorize(wa: WorkArea) {
    const result: StatementNode[] = [];
 
    for (const statement of wa.statements) {
      result.push(this.categorizeStatement(statement));
    }
    wa.statements = result;
  }
 
  private categorizeStatement(input: StatementNode) {
    let statement = input;
 
    const length = input.getChildren().length;
    const lastToken = input.getLastToken();
    const isPunctuation = lastToken instanceof Tokens.Punctuation;
 
    if (length === 1 && isPunctuation) {
      const tokens = statement.getTokens();
      statement = new StatementNode(new Empty()).setChildren(this.tokensToNodes(tokens));
    } else if (statement.get() instanceof Unknown) {
      if (isPunctuation) {
        statement = this.match(statement);
      } else if (length > STATEMENT_MAX_TOKENS) {
        // if the statement contains more than STATEMENT_MAX_TOKENS tokens, just give up
        statement = input;
      } else if (length === 1 && lastToken instanceof Pragma) {
        statement = new StatementNode(new Empty(), undefined, [lastToken]);
      }
    }
 
    return statement;
  }
 
  private removePragma(tokens: readonly Token[]): {tokens: Token[], pragmas: Token[]} {
    const result: Token[] = [];
    const pragmas: Token[] = [];
 
    // skip the last token as it is the punctuation
    for (let i = 0; i < tokens.length - 1; i++) {
      const t = tokens[i];
      if (t instanceof Tokens.Pragma) {
        pragmas.push(t);
      } else {
        result.push(t);
      }
    }
 
    return {tokens: result, pragmas: pragmas};
  }
 
  private match(statement: StatementNode): StatementNode {
    const tokens = statement.getTokens();
    const {tokens: filtered, pragmas} = this.removePragma(tokens);
    if (filtered.length === 0) {
      return new StatementNode(new Empty()).setChildren(this.tokensToNodes(tokens));
    }
 
    for (const st of StatementParser.map.lookup(filtered[0].getStr())) {
      const match = Combi.run(st.getMatcher(), filtered, this.version);
      if (match) {
        const last = tokens[tokens.length - 1];
        match.push(new TokenNode(last));
        return new StatementNode(st, statement.getColon(), pragmas).setChildren(match);
      }
    }
    // next try the statements without specific keywords
    for (const st of StatementParser.map.lookup("")) {
      const match = Combi.run(st.getMatcher(), filtered, this.version);
      if (match) {
        const last = tokens[tokens.length - 1];
        match.push(new TokenNode(last));
        return new StatementNode(st, statement.getColon(), pragmas).setChildren(match);
      }
    }
 
    return statement;
  }
 
// takes care of splitting tokens into statements, also handles chained statements
// statements are split by "," or "."
// additional colons/chaining after the first colon are ignored
  private process(wa: WorkArea) {
    let add: Token[] = [];
    let pre: Token[] = [];
    let colon: Token | undefined = undefined;
 
    for (const token of wa.tokens) {
      if (token instanceof Tokens.Comment) {
        wa.statements.push(new StatementNode(new Comment()).setChildren(this.tokensToNodes([token])));
        continue;
      }
 
      add.push(token);
 
      const str = token.getStr();
      if (str === ".") {
        wa.addUnknown(pre, add, colon);
        add = [];
        pre = [];
        colon = undefined;
      } else if (str === "," && pre.length > 0) {
        wa.addUnknown(pre, add, colon);
        add = [];
      } else if (str === ":" && colon === undefined) {
        colon = token;
        add.pop(); // do not add colon token to statement
        pre.push(...add);
        add = [];
      } else if (str === ":") {
        add.pop(); // do not add colon token to statement
      }
    }
 
    if (add.length > 0) {
      wa.addUnknown(pre, add, colon);
    }
  }
}