view anagram/agcore/p.cpp @ 24:a4899cdfc2d6 default tip

Obfuscate the regexps to strip off the IBM compiler's copyright banners. I don't want bots scanning github to think they're real copyright notices because that could cause real problems.
author David A. Holland
date Mon, 13 Jun 2022 00:40:23 -0400
parents 13d2b8934445
children
line wrap: on
line source

/*
 * AnaGram, A System for Syntax Directed Programming
 * Copyright 1993-2002 Parsifal Software. All Rights Reserved.
 * See the file COPYING for license and usage terms.
 *
 * p.cpp
 */

#include "agdict.h"
#include "assert.h"
#include "bpe3.h"
#include "binsort.h"
#include "config.h"
#include "cs.h"
#include "data.h"
#include "dict.h"
#include "error.h"
#include "lexeme.h"
#include "p.h"
#include "pgg24-defs.h"
#include "q1a.h"
#include "q1glbl.h"
#include "rule.h"
#include "stacks.h"
#include "tsd.h"
#include "token.h"

//#define INCLUDE_LOGGING
#include "log.h"


AgString            etr_file_name;
AgString            infile_name("");
int                 parse_abort_flag = 0;
AgString            simple_file_name;
int                 syntax_error_flag = 0;
AgBalancedTree<int> valueToken;

static void open_result_windows(void) {
  LOGSECTION("open_result_windows");
  LOGV(unres_con->nt) LCV(prr->nt) LCV(key_mess->nt);
  if (unres_con->nt) {
    log_error("Grammar is ambiguous. See Conflicts Window");
  }
  if (prr->nt) {
    log_error("Conflicts resolved by precedence rules");
  }
  if (key_mess->nt) {
    log_error("Keyword anomalies found");
  }
}

static void checkValueTokens(void) {
  LOGSECTION("checkValueTokens");
  int n = valueToken.size();
  LOGV(n);
  int k;
  AgBalancedTree<int> badRules;
  for (k = 0; k < n; k++) {
    Token token = valueToken[k];
    int nRules = token->ruleList.size();
    LOGV(token) LCV(nRules) LCV(token->value_type);
    int i;
    for (i = 0; i < nRules; i++) {
      Rule rule = token.rule(i);
      if (rule->proc_name || rule->length() == 0) {
	continue;
      }
      LOGV(rule) LCV(rule->proc_name) LCV(rule->length()) 
	LCV(rule->length() ? rule.token(0)->value_type : 0);
      LOGV(token->value_type) LCV(rule.token(0)->value_type);

      if (token->value_type == rule.token(0)->value_type) {
	continue;
      }
      LOGS("Bad rule") LV(rule);
      badRules.insert((int)rule);
    }
  }

  for (Each<Rule> r; r.loopNotFinished(); r.getNext()) {
    //if (r->proc_name) continue;
    n = r->elementList.size();
    int j;
    int variableCount = 0;
    for (j = 0; j < n; j++) {
      RuleElement element = r->elementList[j];
      if (element.cVariable) {
        variableCount++;
        if (r->proc_name == 0) {
	  continue;
	}
        if ((int)element.token->value_type == void_token_type) {
          ssprintf("Parameter %s has type void",
		   cVariableList[element.cVariable].pointer());
          log_error(r->line, r->col);
        }
      }
      if (element.token->immediate_action) {
	variableCount = 0;
      }
    }
    if (r->proc_name == 0 && variableCount) {
      badRules.insert((int) r);
    }
  }
  n = badRules.size();
  for (k = 0; k < n; k++) {
    char buf[100];
    Rule r = badRules[k];
    sprintf(buf, "Missing reduction procedure, R%03d", (int) r);
    errorList.push(Error(r->line, r->col, buf));
  }
}

void new_syntax_analyzer(void) {
  LOGSECTION("new_syntax_analyzer");
  q1();
  LOGS("Returned from q1");
  nstates = nits;
  build_parse_table();
  LOGS("Parse table built");
  checkValueTokens();
  LOGS("Tokens checked");
  open_result_windows();
  LOGS("Result windows open");
  syntax_state = syntax_analyzed;
}

void scan_input(void) {
  LOGSECTION("scan_input");
  unsigned i, *l;

  Rule ruleZero = Rule::create();
  ruleZero->prim_tkn = Token(0);

  prod_dict = null_list_dict();
  LOGS("ready to parse");
  parse();
  LOGS("parse complete");

  prod_dict = delete_list_dict(prod_dict);
  nprods = bnf_table->nt;
  if (!(syntax_error_flag || parse_abort_flag)) {
    if (nprods == 0) {
      log_error("No productions in syntax file");
      errorList.top().setFatal();
    }
  }
  if (syntax_error_flag ||
      parse_abort_flag ||
      nprods == 0) {
    syntax_state = syntax_error;
    return;
  }
  syntax_state = syntax_parsed;
  if (default_input_type == 0) {
    default_input_type = int_token_type;
  }
  for (Each<Token> token; token.loopNotFinished(); token.getNext()) {
    if (token->value_type) {
      continue;
    }
    token->value_type = (!token->non_terminal_flag || token->token_set_id) ?
      default_input_type : default_token_type;
  }
#ifdef INCLUDE_LOGGING
  if (174 < Token::count()) {
    Token token174 = Token(174);
    LOGV(token174->immediate_action) LCV(token174->ruleList.size());
  }
#endif
  LOGS("call set_universe");
  set_universe();
  LOGS("call check_key_reserve");
  check_key_reserve();
  LOGS("call build_sets");
  build_sets();
  LOGS("call set_lexemes");
  set_lexemes();
  if (grammar_token == 0 && !(syntax_error_flag || parse_abort_flag)) {
    log_error("No grammar token specified");
    errorList.top().setFatal();
  }
  nprods = bnf_table->nt;

  ibnf_table = spec_tsd(nprods,2);
  l = (unsigned *) bnf_table->sb;
  for (i = 0; i < nprods; i++,l += 2) {
    at(ibnf_table, l[1], l[0]);
  }

  LOGS("Sort bnf tables");
  LOGV(nprods);
  sort_tuples(bnf_table, 2);
  sort_tuples(ibnf_table, 1);

  AgStack<RuleElement> elementStack;
  if (disregard_token) {
    elementStack.push(RuleElement(disregard_token,0));
  }
  elementStack.push(RuleElement(grammar_token, 0));

  if ((int)map_token_number[grammar_token].value_type != void_token_type) {
    valueToken.insert(grammar_token);
  }

  LOGS("Set up ruleZero->elementList");
  ruleZero->elementList = AgArray<RuleElement>(elementStack);
  summarize_grammar();
  bnf_table = delete_tsd(bnf_table);
  no_assertions = 0;
}