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

98.88% Statements 356/360
98.01% Branches 99/101
100% Functions 17/17
98.88% Lines 356/360

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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 3611x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 317x 317x     317x 325x 148x 325x 177x 177x 325x 317x 1x 1x 1x 36075x 36075x 2462x 2462x 36075x 174x 320x 320x 174x 33613x 33613x 1x 1x 1x 1x 1x 1x 1x 1x 10970x 10970x 10970x 10970x 1x 1x 31943x 31943x 31943x 31943x 1x 1x 10970x 10970x 1x 1x 31943x 31943x 31943x 3604x 3604x 31943x 213796x 213796x 31943x 31943x 31943x 1x 1x 1x 1x 1x 1x 1x 1x 1x 10950x 10950x 1x 1x 10950x 10950x 10950x 10950x 1x 1x 1x 10950x 10950x 10950x 10950x 10950x 10970x 10970x 10970x 10970x 10950x 10950x 10970x 10970x 10970x 1155x 1155x 10970x 10970x 10950x 10950x 10950x 1x 1x 1x 764x 764x 764x 5458x 5458x 764x 764x 764x 1x 1x 1x 1x 1x 1155x 1155x 1155x 3017x 3017x 1205x 1205x 1205x 1205x 1205x 1205x 1205x 1092x 358x 3x 3x 355x 358x 44x 44x 44x 44x 358x 1092x 1205x 3017x 3017x 1155x 1155x 1155x 1x 1x 1092x 1092x 1092x 1092x 1092x 12208x 361x 361x 12208x 12208x 12208x 1092x 1092x 1092x 1x 1x 10970x 10970x 10970x 32359x 32359x 32359x 32359x 21x 32359x 43x 43x 10x 43x 33x 33x 33x 3x 3x 3x 3x 3x 3x 3x 3x 33x 43x 32359x 10970x 1x 1x 1x 10970x 10970x 10970x 32247x 32247x 10970x 10970x 1x 1x 32602x 32602x 32602x 32602x 32602x 32602x 32602x 16x 16x 32602x 32282x 32152x 32282x     130x 3x 3x 3x 3x 32282x 32602x 32602x 32602x 1x 1x 32152x 32152x 32152x 189692x 68x 68x 68x 189692x 32152x 32084x 32084x 68x 68x 68x 32152x 598x 598x 69x 598x 529x 529x 598x 68x 68x 68x 1x 1x 32152x 32152x 32152x 3x 3x 32149x 32152x 69858x 69858x 28223x 28223x 28223x 28223x 28223x 69858x 3926x 32152x 7353x 7353x 2441x 2441x 2441x 2441x 2441x 7353x 1485x 1485x 1485x 1x 1x 1x 1x 1x 10970x 10970x 10970x 10970x 10970x 216539x 304x 304x 304x 216235x 216235x 216235x 216235x 216539x 71502x 29548x 29548x 29548x 29548x 71502x 2265x 2265x 41954x 1171x 1171x 1171x 1171x 39689x 2x 2x 71502x 216539x 10970x 10970x 130x 130x 10970x 1x  
import * as Statements from "./statements";
import * as Expressions from "./expressions";
import * as Tokens from "../1_lexer/tokens";
import {ABAPRelease, LanguageVersion} from "../../version";
import {IFile} from "../../files/_ifile";
import {TokenNode, StatementNode} from "../nodes";
import {ArtifactsABAP} from "../artifacts";
import {Combi} from "./combi";
import {Unknown, Empty, Comment, NativeSQL, IStatement} from "./statements/_statement";
import {IStatementResult} from "./statement_result";
import {AbstractToken} from "../1_lexer/tokens/abstract_token";
import {IABAPLexerResult} from "../1_lexer/lexer_result";
import {ExpandMacros} from "./expand_macros";
import {Pragma} from "../1_lexer/tokens";
import {IRegistry} from "../../_iregistry";
import {IStatementRunnable} from "./statement_runnable";
import {reclassifySelect} from "./_select_reclassify";
 
export const STATEMENT_MAX_TOKENS = 20000;
 
const EMPTY_PRAGMAS: readonly AbstractToken[] = Object.freeze([]);
 
class StatementMap {
  // this also serves as container for statement matcher singletons,
  private readonly map: {[index: string]: {statement: IStatement, matcher?: IStatementRunnable}[]};
 
  public constructor() {
    this.map = {};
 
    for (const stat of ArtifactsABAP.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({statement: stat});
        } else {
          this.map[first] = [{statement: stat}];
        }
      }
    }
  }
 
  public lookup(str: string): readonly {statement: IStatement, matcher: IStatementRunnable}[] {
    const res = this.map[str.toUpperCase()];
    if (res === undefined) {
      return [];
    }
    if (res[0].matcher === undefined) {
      for (const r of res) {
        r.matcher = r.statement.getMatcher();
      }
    }
    return res as readonly {statement: IStatement, matcher: IStatementRunnable}[];
  }
}
 
class WorkArea {
  public readonly file: IFile;
  public readonly tokens: readonly AbstractToken[];
  public statements: StatementNode[];
 
  public constructor(file: IFile, tokens: readonly AbstractToken[]) {
    this.file = file;
    this.tokens = tokens;
    this.statements = [];
  }
 
  public addUnknown(pre: AbstractToken[], post: AbstractToken[], colon: AbstractToken | 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: AbstractToken[], tokens2: AbstractToken[]): 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 release: ABAPRelease;
  private readonly languageVersion: LanguageVersion;
  private readonly reg?: IRegistry;
 
  public constructor(release: ABAPRelease, reg?: IRegistry,
                     languageVersion: LanguageVersion = LanguageVersion.Normal) {
    if (!StatementParser.map) {
      StatementParser.map = new StatementMap();
    }
    this.release = release;
    this.languageVersion = languageVersion;
    this.reg = reg;
  }
 
  /** input is one full object */
  public run(input: readonly IABAPLexerResult[], globalMacros: readonly string[]): IStatementResult[] {
    const macros = new ExpandMacros(globalMacros, this.release, this.reg, this.languageVersion);
 
    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, w.file);
    }
 
    for (const w of wa) {
      const res = macros.handleMacros(w.statements, w.file);
      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 AbstractToken[]): 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) {
      // dont use CALL METHOD, when executing lazy, it easily gives a Move for the last statement if lazy logic is evaluated
      if (statement.get() instanceof Unknown) {
        const concat = statement.concatTokens().toUpperCase();
        if (concat.startsWith("CALL METHOD ") === false
            && concat.startsWith("RAISE EXCEPTION TYPE ") === false
            && concat.startsWith("READ TABLE ") === false
            && concat.startsWith("LOOP AT ") === false
            && concat.startsWith("SELECT SINGLE ") === false
            && concat.startsWith("CALL FUNCTION ") === false) {
          for (const {first, second} of this.buildSplits(statement.getTokens())) {
            if (second.length === 1) {
              continue; // probably punctuation
            }
            const s = this.categorizeStatement(new StatementNode(new Unknown()).setChildren(this.tokensToNodes(second)));
            if (!(s.get() instanceof Unknown) && !(s.get() instanceof Empty)) {
              result.push(new StatementNode(new Unknown()).setChildren(this.tokensToNodes(first)));
              statement = s;
              break;
            }
          }
        }
      }
      result.push(statement);
    }
 
    wa.statements = result;
  }
 
  private buildSplits(tokens: readonly AbstractToken[]): {first: AbstractToken[], second: AbstractToken[]}[] {
    const res: {first: AbstractToken[], second: AbstractToken[]}[] = [];
    const before: AbstractToken[] = [];
    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 {
          wa.statements[i] = new StatementNode(new NativeSQL()).setChildren(this.tokensToNodes(statement.getTokens()));
 
          if (statement.concatTokens().toUpperCase().endsWith("ENDMETHOD.")) {
            const tokens = statement.getTokens();
            const startTokens = this.tokensToNodes(tokens.slice(tokens.length - 2, tokens.length));
            const endTokens = this.tokensToNodes(tokens.slice(0, tokens.length - 2));
            wa.statements[i] = new StatementNode(new NativeSQL()).setChildren(endTokens);
            const item = new StatementNode(new Statements.EndMethod()).setChildren(startTokens);
            wa.statements.splice(i + 1, 0, item);
            sql = false;
          }
        }
      }
    }
  }
 
// 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) {
        // special case, everything crashes if StatementNodes doesnt have children
        statement = new StatementNode(new Empty(), undefined, [lastToken])
          .setChildren(this.tokensToNodes([lastToken]));
      }
    }
 
    return statement;
  }
 
  private removePragma(tokens: readonly AbstractToken[]): {tokens: AbstractToken[], pragmas: readonly AbstractToken[]} {
    let hasPragma = false;
    // skip the last token as it is the punctuation
    for (let i = 0; i < tokens.length - 1; i++) {
      if (tokens[i] instanceof Tokens.Pragma) {
        hasPragma = true;
        break;
      }
    }
    if (hasPragma === false) {
      return {tokens: tokens.slice(0, tokens.length - 1), pragmas: EMPTY_PRAGMAS};
    }
 
    const result: AbstractToken[] = [];
    const pragmas: AbstractToken[] = [];
    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.matcher, filtered, this.release, this.languageVersion);
      if (match) {
        const last = tokens[tokens.length - 1];
        match.push(new TokenNode(last));
        const matched = new StatementNode(st.statement, statement.getColon(), pragmas).setChildren(match);
        return reclassifySelect(matched, st.statement, pragmas);
      }
    }
    // next try the statements without specific keywords
    for (const st of StatementParser.map.lookup("")) {
      const match = Combi.run(st.matcher, filtered, this.release, this.languageVersion);
      if (match) {
        const last = tokens[tokens.length - 1];
        match.push(new TokenNode(last));
        const matched = new StatementNode(st.statement, statement.getColon(), pragmas).setChildren(match);
        return reclassifySelect(matched, st.statement, pragmas);
      }
    }
 
    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: AbstractToken[] = [];
    let pre: AbstractToken[] = [];
    let colon: AbstractToken | 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.length === 1) {
        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);
    }
  }
}