fortran.c - ctags-5.8

Global variables defined

Data types defined

Functions defined

Macros defined

Source code

/*
*   $Id: fortran.c 660 2008-04-20 23:30:12Z elliotth $
*
*   Copyright (c) 1998-2003, Darren Hiebert
*
*   This source code is released for free distribution under the terms of the
*   GNU General Public License.
*
*   This module contains functions for generating tags for Fortran language
*   files.
*/

/*
*   INCLUDE FILES
*/
#include "general.h"  /* must always come first */

#include <string.h>
#include <limits.h>
#include <ctype.h>  /* to define tolower () */
#include <setjmp.h>

#include "debug.h"
#include "entry.h"
#include "keyword.h"
#include "options.h"
#include "parse.h"
#include "read.h"
#include "routines.h"
#include "vstring.h"

/*
*   MACROS
*/
‌#define isident(c)              (isalnum(c) || (c) == '_')
‌#define isBlank(c)              (boolean) (c == ' ' || c == '\t')
‌#define isType(token,t)         (boolean) ((token)->type == (t))
‌#define isKeyword(token,k)      (boolean) ((token)->keyword == (k))
‌#define isSecondaryKeyword(token,k)  (boolean) ((token)->secondary == NULL ? \
    FALSE : (token)->secondary->keyword == (k))

/*
*   DATA DECLARATIONS
*/

‌typedef enum eException {
    ExceptionNone, ExceptionEOF, ExceptionFixedFormat, ExceptionLoop
‌} exception_t;

/*  Used to designate type of line read in fixed source form.
 */
‌typedef enum eFortranLineType {
    LTYPE_UNDETERMINED,
    LTYPE_INVALID,
    LTYPE_COMMENT,
    LTYPE_CONTINUATION,
    LTYPE_EOF,
    LTYPE_INITIAL,
    LTYPE_SHORT
‌} lineType;

/*  Used to specify type of keyword.
 */
‌typedef enum eKeywordId {
    KEYWORD_NONE = -1,
    KEYWORD_allocatable,
    KEYWORD_assignment,
    KEYWORD_automatic,
    KEYWORD_block,
    KEYWORD_byte,
    KEYWORD_cexternal,
    KEYWORD_cglobal,
    KEYWORD_character,
    KEYWORD_common,
    KEYWORD_complex,
    KEYWORD_contains,
    KEYWORD_data,
    KEYWORD_dimension,
    KEYWORD_dllexport,
    KEYWORD_dllimport,
    KEYWORD_do,
    KEYWORD_double,
    KEYWORD_elemental,
    KEYWORD_end,
    KEYWORD_entry,
    KEYWORD_equivalence,
    KEYWORD_external,
    KEYWORD_format,
    KEYWORD_function,
    KEYWORD_if,
    KEYWORD_implicit,
    KEYWORD_include,
    KEYWORD_inline,
    KEYWORD_integer,
    KEYWORD_intent,
    KEYWORD_interface,
    KEYWORD_intrinsic,
    KEYWORD_logical,
    KEYWORD_map,
    KEYWORD_module,
    KEYWORD_namelist,
    KEYWORD_operator,
    KEYWORD_optional,
    KEYWORD_parameter,
    KEYWORD_pascal,
    KEYWORD_pexternal,
    KEYWORD_pglobal,
    KEYWORD_pointer,
    KEYWORD_precision,
    KEYWORD_private,
    KEYWORD_program,
    KEYWORD_public,
    KEYWORD_pure,
    KEYWORD_real,
    KEYWORD_record,
    KEYWORD_recursive,
    KEYWORD_save,
    KEYWORD_select,
    KEYWORD_sequence,
    KEYWORD_static,
    KEYWORD_stdcall,
    KEYWORD_structure,
    KEYWORD_subroutine,
    KEYWORD_target,
    KEYWORD_then,
    KEYWORD_type,
    KEYWORD_union,
    KEYWORD_use,
    KEYWORD_value,
    KEYWORD_virtual,
    KEYWORD_volatile,
    KEYWORD_where,
    KEYWORD_while
‌} keywordId;

/*  Used to determine whether keyword is valid for the token language and
 *  what its ID is.
 */
‌typedef struct sKeywordDesc {
    const char *name;
    keywordId id;
‌} keywordDesc;

‌typedef enum eTokenType {
    TOKEN_UNDEFINED,
    TOKEN_COMMA,
    TOKEN_DOUBLE_COLON,
    TOKEN_IDENTIFIER,
    TOKEN_KEYWORD,
    TOKEN_LABEL,
    TOKEN_NUMERIC,
    TOKEN_OPERATOR,
    TOKEN_PAREN_CLOSE,
    TOKEN_PAREN_OPEN,
    TOKEN_PERCENT,
    TOKEN_STATEMENT_END,
    TOKEN_STRING
‌} tokenType;

‌typedef enum eTagType {
    TAG_UNDEFINED = -1,
    TAG_BLOCK_DATA,
    TAG_COMMON_BLOCK,
    TAG_ENTRY_POINT,
    TAG_FUNCTION,
    TAG_INTERFACE,
    TAG_COMPONENT,
    TAG_LABEL,
    TAG_LOCAL,
    TAG_MODULE,
    TAG_NAMELIST,
    TAG_PROGRAM,
    TAG_SUBROUTINE,
    TAG_DERIVED_TYPE,
    TAG_VARIABLE,
    TAG_COUNT  /* must be last */
‌} tagType;

‌typedef struct sTokenInfo {
    tokenType type;
    keywordId keyword;
    tagType tag;
    vString* string;
    struct sTokenInfo *secondary;
    unsigned long lineNumber;
    fpos_t filePosition;
‌} tokenInfo;

/*
*   DATA DEFINITIONS
*/

‌static langType Lang_fortran;
‌static jmp_buf Exception;
‌static int Ungetc;
‌static unsigned int Column;
‌static boolean FreeSourceForm;
‌static boolean ParsingString;
‌static tokenInfo *Parent;

/* indexed by tagType */
‌static kindOption FortranKinds [] = {
    { TRUE,  'b', "block data", "block data"},
    { TRUE,  'c', "common",     "common blocks"},
    { TRUE,  'e', "entry",      "entry points"},
    { TRUE,  'f', "function",   "functions"},
    { FALSE, 'i', "interface",  "interface contents, generic names, and operators"},
    { TRUE,  'k', "component",  "type and structure components"},
    { TRUE,  'l', "label",      "labels"},
    { FALSE, 'L', "local",      "local, common block, and namelist variables"},
    { TRUE,  'm', "module",     "modules"},
    { TRUE,  'n', "namelist",   "namelists"},
    { TRUE,  'p', "program",    "programs"},
    { TRUE,  's', "subroutine", "subroutines"},
    { TRUE,  't', "type",       "derived types and structures"},
    { TRUE,  'v', "variable",   "program (global) and module variables"}
};

/* For efinitions of Fortran 77 with extensions:
 * http://www.fortran.com/fortran/F77_std/rjcnf0001.html
 * http://scienide.uwaterloo.ca/MIPSpro7/007-2362-004/sgi_html/index.html
 *
 * For the Compaq Fortran Reference Manual:
 * http://h18009.www1.hp.com/fortran/docs/lrm/dflrm.htm
 */

‌static const keywordDesc FortranKeywordTable [] = {
    /* keyword          keyword ID */
    { "allocatable",    KEYWORD_allocatable  },
    { "assignment",     KEYWORD_assignment   },
    { "automatic",      KEYWORD_automatic    },
    { "block",          KEYWORD_block        },
    { "byte",           KEYWORD_byte         },
    { "cexternal",      KEYWORD_cexternal    },
    { "cglobal",        KEYWORD_cglobal      },
    { "character",      KEYWORD_character    },
    { "common",         KEYWORD_common       },
    { "complex",        KEYWORD_complex      },
    { "contains",       KEYWORD_contains     },
    { "data",           KEYWORD_data         },
    { "dimension",      KEYWORD_dimension    },
    { "dll_export",     KEYWORD_dllexport    },
    { "dll_import",     KEYWORD_dllimport    },
    { "do",             KEYWORD_do           },
    { "double",         KEYWORD_double       },
    { "elemental",      KEYWORD_elemental    },
    { "end",            KEYWORD_end          },
    { "entry",          KEYWORD_entry        },
    { "equivalence",    KEYWORD_equivalence  },
    { "external",       KEYWORD_external     },
    { "format",         KEYWORD_format       },
    { "function",       KEYWORD_function     },
    { "if",             KEYWORD_if           },
    { "implicit",       KEYWORD_implicit     },
    { "include",        KEYWORD_include      },
    { "inline",         KEYWORD_inline       },
    { "integer",        KEYWORD_integer      },
    { "intent",         KEYWORD_intent       },
    { "interface",      KEYWORD_interface    },
    { "intrinsic",      KEYWORD_intrinsic    },
    { "logical",        KEYWORD_logical      },
    { "map",            KEYWORD_map          },
    { "module",         KEYWORD_module       },
    { "namelist",       KEYWORD_namelist     },
    { "operator",       KEYWORD_operator     },
    { "optional",       KEYWORD_optional     },
    { "parameter",      KEYWORD_parameter    },
    { "pascal",         KEYWORD_pascal       },
    { "pexternal",      KEYWORD_pexternal    },
    { "pglobal",        KEYWORD_pglobal      },
    { "pointer",        KEYWORD_pointer      },
    { "precision",      KEYWORD_precision    },
    { "private",        KEYWORD_private      },
    { "program",        KEYWORD_program      },
    { "public",         KEYWORD_public       },
    { "pure",           KEYWORD_pure         },
    { "real",           KEYWORD_real         },
    { "record",         KEYWORD_record       },
    { "recursive",      KEYWORD_recursive    },
    { "save",           KEYWORD_save         },
    { "select",         KEYWORD_select       },
    { "sequence",       KEYWORD_sequence     },
    { "static",         KEYWORD_static       },
    { "stdcall",        KEYWORD_stdcall      },
    { "structure",      KEYWORD_structure    },
    { "subroutine",     KEYWORD_subroutine   },
    { "target",         KEYWORD_target       },
    { "then",           KEYWORD_then         },
    { "type",           KEYWORD_type         },
    { "union",          KEYWORD_union        },
    { "use",            KEYWORD_use          },
    { "value",          KEYWORD_value        },
    { "virtual",        KEYWORD_virtual      },
    { "volatile",       KEYWORD_volatile     },
    { "where",          KEYWORD_where        },
    { "while",          KEYWORD_while        }
};

static struct {
    unsigned int count;
    unsigned int max;
    tokenInfo* list;
‌} Ancestors = { 0, 0, NULL };

/*
*   FUNCTION PROTOTYPES
*/
static void parseStructureStmt (tokenInfo *const token);
static void parseUnionStmt (tokenInfo *const token);
static void parseDerivedTypeDef (tokenInfo *const token);
static void parseFunctionSubprogram (tokenInfo *const token);
static void parseSubroutineSubprogram (tokenInfo *const token);

/*
*   FUNCTION DEFINITIONS
*/

‌static void ancestorPush (tokenInfo *const token)
{
    enum { incrementalIncrease = 10 };
    if (Ancestors.list == NULL)
    {
        Assert (Ancestors.max == 0);
        Ancestors.count = 0;
        Ancestors.max   = incrementalIncrease;
        Ancestors.list  = xMalloc (Ancestors.max, tokenInfo);
    }
    else if (Ancestors.count == Ancestors.max)
    {
        Ancestors.max += incrementalIncrease;
        Ancestors.list = xRealloc (Ancestors.list, Ancestors.max, tokenInfo);
    }
    Ancestors.list [Ancestors.count] = *token;
    Ancestors.list [Ancestors.count].string = vStringNewCopy (token->string);
    Ancestors.count++;
}

‌static void ancestorPop (void)
{
    Assert (Ancestors.count > 0);
    --Ancestors.count;
    vStringDelete (Ancestors.list [Ancestors.count].string);

    Ancestors.list [Ancestors.count].type       = TOKEN_UNDEFINED;
    Ancestors.list [Ancestors.count].keyword    = KEYWORD_NONE;
    Ancestors.list [Ancestors.count].secondary  = NULL;
    Ancestors.list [Ancestors.count].tag        = TAG_UNDEFINED;
    Ancestors.list [Ancestors.count].string     = NULL;
    Ancestors.list [Ancestors.count].lineNumber = 0L;
}

‌static const tokenInfo* ancestorScope (void)
{
    tokenInfo *result = NULL;
    unsigned int i;
    for (i = Ancestors.count  ;  i > 0  &&  result == NULL ;  --i)
    {
        tokenInfo *const token = Ancestors.list + i - 1;
        if (token->type == TOKEN_IDENTIFIER &&
            token->tag != TAG_UNDEFINED  && token->tag != TAG_INTERFACE)
            result = token;
    }
    return result;
}

‌static const tokenInfo* ancestorTop (void)
{
    Assert (Ancestors.count > 0);
    return &Ancestors.list [Ancestors.count - 1];
}

‌#define ancestorCount() (Ancestors.count)

‌static void ancestorClear (void)
{
    while (Ancestors.count > 0)
        ancestorPop ();
    if (Ancestors.list != NULL)
        eFree (Ancestors.list);
    Ancestors.list = NULL;
    Ancestors.count = 0;
    Ancestors.max = 0;
}

‌static boolean insideInterface (void)
{
    boolean result = FALSE;
    unsigned int i;
    for (i = 0  ;  i < Ancestors.count && !result ;  ++i)
    {
        if (Ancestors.list [i].tag == TAG_INTERFACE)
            result = TRUE;
    }
    return result;
}

‌static void buildFortranKeywordHash (void)
{
    const size_t count =
            sizeof (FortranKeywordTable) / sizeof (FortranKeywordTable [0]);
    size_t i;
    for (i = 0  ;  i < count  ;  ++i)
    {
        const keywordDesc* const p = &FortranKeywordTable [i];
        addKeyword (p->name, Lang_fortran, (int) p->id);
    }
}

/*
*   Tag generation functions
*/

‌static tokenInfo *newToken (void)
{
    tokenInfo *const token = xMalloc (1, tokenInfo);

    token->type         = TOKEN_UNDEFINED;
    token->keyword      = KEYWORD_NONE;
    token->tag          = TAG_UNDEFINED;
    token->string       = vStringNew ();
    token->secondary    = NULL;
    token->lineNumber   = getSourceLineNumber ();
    token->filePosition = getInputFilePosition ();

    return token;
}

‌static tokenInfo *newTokenFrom (tokenInfo *const token)
{
    tokenInfo *result = newToken ();
    *result = *token;
    result->string = vStringNewCopy (token->string);
    token->secondary = NULL;
    return result;
}

‌static void deleteToken (tokenInfo *const token)
{
    if (token != NULL)
    {
        vStringDelete (token->string);
        deleteToken (token->secondary);
        token->secondary = NULL;
        eFree (token);
    }
}

‌static boolean isFileScope (const tagType type)
{
    return (boolean) (type == TAG_LABEL || type == TAG_LOCAL);
}

‌static boolean includeTag (const tagType type)
{
    boolean include;
    Assert (type != TAG_UNDEFINED);
    include = FortranKinds [(int) type].enabled;
    if (include && isFileScope (type))
        include = Option.include.fileScope;
    return include;
}

‌static void makeFortranTag (tokenInfo *const token, tagType tag)
{
    token->tag = tag;
    if (includeTag (token->tag))
    {
        const char *const name = vStringValue (token->string);
        tagEntryInfo e;

        initTagEntry (&e, name);

        if (token->tag == TAG_COMMON_BLOCK)
            e.lineNumberEntry = (boolean) (Option.locate != EX_PATTERN);

        e.lineNumber    = token->lineNumber;
        e.filePosition    = token->filePosition;
        e.isFileScope    = isFileScope (token->tag);
        e.kindName        = FortranKinds [token->tag].name;
        e.kind            = FortranKinds [token->tag].letter;
        e.truncateLine    = (boolean) (token->tag != TAG_LABEL);

        if (ancestorCount () > 0)
        {
            const tokenInfo* const scope = ancestorScope ();
            if (scope != NULL)
            {
                e.extensionFields.scope [0] = FortranKinds [scope->tag].name;
                e.extensionFields.scope [1] = vStringValue (scope->string);
            }
        }
        if (! insideInterface () || includeTag (TAG_INTERFACE))
            makeTagEntry (&e);
    }
}

/*
*   Parsing functions
*/

‌static int skipLine (void)
{
    int c;

    do
        c = fileGetc ();
    while (c != EOF  &&  c != '\n');

    return c;
}

‌static void makeLabelTag (vString *const label)
{
    tokenInfo *token = newToken ();
    token->type  = TOKEN_LABEL;
    vStringCopy (token->string, label);
    makeFortranTag (token, TAG_LABEL);
    deleteToken (token);
}

‌static lineType getLineType (void)
{
    vString *label = vStringNew ();
    int column = 0;
    lineType type = LTYPE_UNDETERMINED;

    do  /* read in first 6 "margin" characters */
    {
        int c = fileGetc ();

        /* 3.2.1  Comment_Line.  A comment line is any line that contains
         * a C or an asterisk in column 1, or contains only blank characters
         * in  columns 1 through 72.  A comment line that contains a C or
         * an asterisk in column 1 may contain any character capable  of
         * representation in the processor in columns 2 through 72.
         */
        /*  EXCEPTION! Some compilers permit '!' as a commment character here.
         *
         *  Treat # and $ in column 1 as comment to permit preprocessor directives.
         *  Treat D and d in column 1 as comment for HP debug statements.
         */
        if (column == 0  &&  strchr ("*Cc!#$Dd", c) != NULL)
            type = LTYPE_COMMENT;
        else if (c == '\t')  /* EXCEPTION! Some compilers permit a tab here */
        {
            column = 8;
            type = LTYPE_INITIAL;
        }
        else if (column == 5)
        {
            /* 3.2.2  Initial_Line.  An initial line is any line that is not
             * a comment line and contains the character blank or the digit 0
             * in column 6.  Columns 1 through 5 may contain a statement label
             * (3.4), or each of the columns 1 through 5 must contain the
             * character blank.
             */
            if (c == ' '  ||  c == '0')
                type = LTYPE_INITIAL;

            /* 3.2.3  Continuation_Line.  A continuation line is any line that
             * contains any character of the FORTRAN character set other than
             * the character blank or the digit 0 in column 6 and contains
             * only blank characters in columns 1 through 5.
             */
            else if (vStringLength (label) == 0)
                type = LTYPE_CONTINUATION;
            else
                type = LTYPE_INVALID;
        }
        else if (c == ' ')
            ;
        else if (c == EOF)
            type = LTYPE_EOF;
        else if (c == '\n')
            type = LTYPE_SHORT;
        else if (isdigit (c))
            vStringPut (label, c);
        else
            type = LTYPE_INVALID;

        ++column;
    } while (column < 6  &&  type == LTYPE_UNDETERMINED);

    Assert (type != LTYPE_UNDETERMINED);

    if (vStringLength (label) > 0)
    {
        vStringTerminate (label);
        makeLabelTag (label);
    }
    vStringDelete (label);
    return type;
}

‌static int getFixedFormChar (void)
{
    boolean newline = FALSE;
    lineType type;
    int c = '\0';

    if (Column > 0)
    {
#ifdef STRICT_FIXED_FORM
        /*  EXCEPTION! Some compilers permit more than 72 characters per line.
         */
        if (Column > 71)
            c = skipLine ();
        else
#endif
        {
            c = fileGetc ();
            ++Column;
        }
        if (c == '\n')
        {
            newline = TRUE;  /* need to check for continuation line */
            Column = 0;
        }
        else if (c == '!'  &&  ! ParsingString)
        {
            c = skipLine ();
            newline = TRUE;  /* need to check for continuation line */
            Column = 0;
        }
        else if (c == '&')  /* check for free source form */
        {
            const int c2 = fileGetc ();
            if (c2 == '\n')
                longjmp (Exception, (int) ExceptionFixedFormat);
            else
                fileUngetc (c2);
        }
    }
    while (Column == 0)
    {
        type = getLineType ();
        switch (type)
        {
            case LTYPE_UNDETERMINED:
            case LTYPE_INVALID:
                longjmp (Exception, (int) ExceptionFixedFormat);
                break;

            case LTYPE_SHORT: break;
            case LTYPE_COMMENT: skipLine (); break;

            case LTYPE_EOF:
                Column = 6;
                if (newline)
                    c = '\n';
                else
                    c = EOF;
                break;

            case LTYPE_INITIAL:
                if (newline)
                {
                    c = '\n';
                    Column = 6;
                    break;
                }
                /* fall through to next case */
            case LTYPE_CONTINUATION:
                Column = 5;
                do
                {
                    c = fileGetc ();
                    ++Column;
                } while (isBlank (c));
                if (c == '\n')
                    Column = 0;
                else if (Column > 6)
                {
                    fileUngetc (c);
                    c = ' ';
                }
                break;

            default:
                Assert ("Unexpected line type" == NULL);
        }
    }
    return c;
}

‌static int skipToNextLine (void)
{
    int c = skipLine ();
    if (c != EOF)
        c = fileGetc ();
    return c;
}

‌static int getFreeFormChar (void)
{
    static boolean newline = TRUE;
    boolean advanceLine = FALSE;
    int c = fileGetc ();

    /* If the last nonblank, non-comment character of a FORTRAN 90
     * free-format text line is an ampersand then the next non-comment
     * line is a continuation line.
     */
    if (c == '&')
    {
        do
            c = fileGetc ();
        while (isspace (c)  &&  c != '\n');
        if (c == '\n')
        {
            newline = TRUE;
            advanceLine = TRUE;
        }
        else if (c == '!')
            advanceLine = TRUE;
        else
        {
            fileUngetc (c);
            c = '&';
        }
    }
    else if (newline && (c == '!' || c == '#'))
        advanceLine = TRUE;
    while (advanceLine)
    {
        while (isspace (c))
            c = fileGetc ();
        if (c == '!' || (newline && c == '#'))
        {
            c = skipToNextLine ();
            newline = TRUE;
            continue;
        }
        if (c == '&')
            c = fileGetc ();
        else
            advanceLine = FALSE;
    }
    newline = (boolean) (c == '\n');
    return c;
}

‌static int getChar (void)
{
    int c;

    if (Ungetc != '\0')
    {
        c = Ungetc;
        Ungetc = '\0';
    }
    else if (FreeSourceForm)
        c = getFreeFormChar ();
    else
        c = getFixedFormChar ();
    return c;
}

‌static void ungetChar (const int c)
{
    Ungetc = c;
}

/*  If a numeric is passed in 'c', this is used as the first digit of the
 *  numeric being parsed.
 */
‌static vString *parseInteger (int c)
{
    vString *string = vStringNew ();

    if (c == '-')
    {
        vStringPut (string, c);
        c = getChar ();
    }
    else if (! isdigit (c))
        c = getChar ();
    while (c != EOF  &&  isdigit (c))
    {
        vStringPut (string, c);
        c = getChar ();
    }
    vStringTerminate (string);

    if (c == '_')
    {
        do
            c = getChar ();
        while (c != EOF  &&  isalpha (c));
    }
    ungetChar (c);

    return string;
}

‌static vString *parseNumeric (int c)
{
    vString *string = vStringNew ();
    vString *integer = parseInteger (c);
    vStringCopy (string, integer);
    vStringDelete (integer);

    c = getChar ();
    if (c == '.')
    {
        integer = parseInteger ('\0');
        vStringPut (string, c);
        vStringCat (string, integer);
        vStringDelete (integer);
        c = getChar ();
    }
    if (tolower (c) == 'e')
    {
        integer = parseInteger ('\0');
        vStringPut (string, c);
        vStringCat (string, integer);
        vStringDelete (integer);
    }
    else
        ungetChar (c);

    vStringTerminate (string);

    return string;
}

‌static void parseString (vString *const string, const int delimiter)
{
    const unsigned long inputLineNumber = getInputLineNumber ();
    int c;
    ParsingString = TRUE;
    c = getChar ();
    while (c != delimiter  &&  c != '\n'  &&  c != EOF)
    {
        vStringPut (string, c);
        c = getChar ();
    }
    if (c == '\n'  ||  c == EOF)
    {
        verbose ("%s: unterminated character string at line %lu\n",
                getInputFileName (), inputLineNumber);
        if (c == EOF)
            longjmp (Exception, (int) ExceptionEOF);
        else if (! FreeSourceForm)
            longjmp (Exception, (int) ExceptionFixedFormat);
    }
    vStringTerminate (string);
    ParsingString = FALSE;
}

/*  Read a C identifier beginning with "firstChar" and places it into "name".
 */
‌static void parseIdentifier (vString *const string, const int firstChar)
{
    int c = firstChar;

    do
    {
        vStringPut (string, c);
        c = getChar ();
    } while (isident (c));

    vStringTerminate (string);
    ungetChar (c);  /* unget non-identifier character */
}

‌static void checkForLabel (void)
{
    tokenInfo* token = NULL;
    int length;
    int c;

    do
        c = getChar ();
    while (isBlank (c));

    for (length = 0  ;  isdigit (c)  &&  length < 5  ;  ++length)
    {
        if (token == NULL)
        {
            token = newToken ();
            token->type = TOKEN_LABEL;
        }
        vStringPut (token->string, c);
        c = getChar ();
    }
    if (length > 0  &&  token != NULL)
    {
        vStringTerminate (token->string);
        makeFortranTag (token, TAG_LABEL);
        deleteToken (token);
    }
    ungetChar (c);
}

‌static void readIdentifier (tokenInfo *const token, const int c)
{
    parseIdentifier (token->string, c);
    token->keyword = analyzeToken (token->string, Lang_fortran);
    if (! isKeyword (token, KEYWORD_NONE))
        token->type = TOKEN_KEYWORD;
    else
    {
        token->type = TOKEN_IDENTIFIER;
        if (strncmp (vStringValue (token->string), "end", 3) == 0)
        {
            vString *const sub = vStringNewInit (vStringValue (token->string) + 3);
            const keywordId kw = analyzeToken (sub, Lang_fortran);
            vStringDelete (sub);
            if (kw != KEYWORD_NONE)
            {
                token->secondary = newToken ();
                token->secondary->type = TOKEN_KEYWORD;
                token->secondary->keyword = kw;
                token->keyword = KEYWORD_end;
            }
        }
    }
}

‌static void readToken (tokenInfo *const token)
{
    int c;

    deleteToken (token->secondary);
    token->type        = TOKEN_UNDEFINED;
    token->tag         = TAG_UNDEFINED;
    token->keyword     = KEYWORD_NONE;
    token->secondary   = NULL;
    vStringClear (token->string);

getNextChar:
    c = getChar ();

    token->lineNumber    = getSourceLineNumber ();
    token->filePosition    = getInputFilePosition ();

    switch (c)
    {
        case EOF:  longjmp (Exception, (int) ExceptionEOF);  break;
        case ' ':  goto getNextChar;
        case '\t': goto getNextChar;
        case ',':  token->type = TOKEN_COMMA;       break;
        case '(':  token->type = TOKEN_PAREN_OPEN;  break;
        case ')':  token->type = TOKEN_PAREN_CLOSE; break;
        case '%':  token->type = TOKEN_PERCENT;     break;

        case '*':
        case '/':
        case '+':
        case '-':
        case '=':
        case '<':
        case '>':
        {
            const char *const operatorChars = "*/+=<>";
            do {
                vStringPut (token->string, c);
                c = getChar ();
            } while (strchr (operatorChars, c) != NULL);
            ungetChar (c);
            vStringTerminate (token->string);
            token->type = TOKEN_OPERATOR;
            break;
        }

        case '!':
            if (FreeSourceForm)
            {
                do
                   c = getChar ();
                while (c != '\n' && c != EOF);
            }
            else
            {
                skipLine ();
                Column = 0;
            }
            /* fall through to newline case */
        case '\n':
            token->type = TOKEN_STATEMENT_END;
            if (FreeSourceForm)
                checkForLabel ();
            break;

        case '.':
            parseIdentifier (token->string, c);
            c = getChar ();
            if (c == '.')
            {
                vStringPut (token->string, c);
                vStringTerminate (token->string);
                token->type = TOKEN_OPERATOR;
            }
            else
            {
                ungetChar (c);
                token->type = TOKEN_UNDEFINED;
            }
            break;

        case '"':
        case '\'':
            parseString (token->string, c);
            token->type = TOKEN_STRING;
            break;

        case ';':
            token->type = TOKEN_STATEMENT_END;
            break;

        case ':':
            c = getChar ();
            if (c == ':')
                token->type = TOKEN_DOUBLE_COLON;
            else
            {
                ungetChar (c);
                token->type = TOKEN_UNDEFINED;
            }
            break;

        default:
            if (isalpha (c))
                readIdentifier (token, c);
            else if (isdigit (c))
            {
                vString *numeric = parseNumeric (c);
                vStringCat (token->string, numeric);
                vStringDelete (numeric);
                token->type = TOKEN_NUMERIC;
            }
            else
                token->type = TOKEN_UNDEFINED;
            break;
    }
}

‌static void readSubToken (tokenInfo *const token)
{
    if (token->secondary == NULL)
    {
        token->secondary = newToken ();
        readToken (token->secondary);
    }
}

/*
*   Scanning functions
*/

‌static void skipToToken (tokenInfo *const token, tokenType type)
{
    while (! isType (token, type) && ! isType (token, TOKEN_STATEMENT_END) &&
            !(token->secondary != NULL && isType (token->secondary, TOKEN_STATEMENT_END)))
        readToken (token);
}

‌static void skipPast (tokenInfo *const token, tokenType type)
{
    skipToToken (token, type);
    if (! isType (token, TOKEN_STATEMENT_END))
        readToken (token);
}

‌static void skipToNextStatement (tokenInfo *const token)
{
    do
    {
        skipToToken (token, TOKEN_STATEMENT_END);
        readToken (token);
    } while (isType (token, TOKEN_STATEMENT_END));
}

/* skip over parenthesis enclosed contents starting at next token.
 * Token is left at the first token following closing parenthesis. If an
 * opening parenthesis is not found, `token' is moved to the end of the
 * statement.
 */
‌static void skipOverParens (tokenInfo *const token)
{
    int level = 0;
    do {
        if (isType (token, TOKEN_STATEMENT_END))
            break;
        else if (isType (token, TOKEN_PAREN_OPEN))
            ++level;
        else if (isType (token, TOKEN_PAREN_CLOSE))
            --level;
        readToken (token);
    } while (level > 0);
}

‌static boolean isTypeSpec (tokenInfo *const token)
{
    boolean result;
    switch (token->keyword)
    {
        case KEYWORD_byte:
        case KEYWORD_integer:
        case KEYWORD_real:
        case KEYWORD_double:
        case KEYWORD_complex:
        case KEYWORD_character:
        case KEYWORD_logical:
        case KEYWORD_record:
        case KEYWORD_type:
            result = TRUE;
            break;
        default:
            result = FALSE;
            break;
    }
    return result;
}

‌static boolean isSubprogramPrefix (tokenInfo *const token)
{
    boolean result;
    switch (token->keyword)
    {
        case KEYWORD_elemental:
        case KEYWORD_pure:
        case KEYWORD_recursive:
        case KEYWORD_stdcall:
            result = TRUE;
            break;
        default:
            result = FALSE;
            break;
    }
    return result;
}

/*  type-spec
 *      is INTEGER [kind-selector]
 *      or REAL [kind-selector] is ( etc. )
 *      or DOUBLE PRECISION
 *      or COMPLEX [kind-selector]
 *      or CHARACTER [kind-selector]
 *      or LOGICAL [kind-selector]
 *      or TYPE ( type-name )
 *
 *  Note that INTEGER and REAL may be followed by "*N" where "N" is an integer
 */
‌static void parseTypeSpec (tokenInfo *const token)
{
    /* parse type-spec, leaving `token' at first token following type-spec */
    Assert (isTypeSpec (token));
    switch (token->keyword)
    {
        case KEYWORD_character:
            /* skip char-selector */
            readToken (token);
            if (isType (token, TOKEN_OPERATOR) &&
                     strcmp (vStringValue (token->string), "*") == 0)
                readToken (token);
            if (isType (token, TOKEN_PAREN_OPEN))
                skipOverParens (token);
            else if (isType (token, TOKEN_NUMERIC))
                readToken (token);
            break;


        case KEYWORD_byte:
        case KEYWORD_complex:
        case KEYWORD_integer:
        case KEYWORD_logical:
        case KEYWORD_real:
            readToken (token);
            if (isType (token, TOKEN_PAREN_OPEN))
                skipOverParens (token);  /* skip kind-selector */
            if (isType (token, TOKEN_OPERATOR) &&
                strcmp (vStringValue (token->string), "*") == 0)
            {
                readToken (token);
                readToken (token);
            }
            break;

        case KEYWORD_double:
            readToken (token);
            if (isKeyword (token, KEYWORD_complex) ||
                isKeyword (token, KEYWORD_precision))
                    readToken (token);
            else
                skipToToken (token, TOKEN_STATEMENT_END);
            break;

        case KEYWORD_record:
            readToken (token);
            if (isType (token, TOKEN_OPERATOR) &&
                strcmp (vStringValue (token->string), "/") == 0)
            {
                readToken (token);  /* skip to structure name */
                readToken (token);  /* skip to '/' */
                readToken (token);  /* skip to variable name */
            }
            break;

        case KEYWORD_type:
            readToken (token);
            if (isType (token, TOKEN_PAREN_OPEN))
                skipOverParens (token);  /* skip type-name */
            else
                parseDerivedTypeDef (token);
            break;

        default:
            skipToToken (token, TOKEN_STATEMENT_END);
            break;
    }
}

‌static boolean skipStatementIfKeyword (tokenInfo *const token, keywordId keyword)
{
    boolean result = FALSE;
    if (isKeyword (token, keyword))
    {
        result = TRUE;
        skipToNextStatement (token);
    }
    return result;
}

/* parse a list of qualifying specifiers, leaving `token' at first token
 * following list. Examples of such specifiers are:
 *      [[, attr-spec] ::]
 *      [[, component-attr-spec-list] ::]
 *
 *  attr-spec
 *      is PARAMETER
 *      or access-spec (is PUBLIC or PRIVATE)
 *      or ALLOCATABLE
 *      or DIMENSION ( array-spec )
 *      or EXTERNAL
 *      or INTENT ( intent-spec )
 *      or INTRINSIC
 *      or OPTIONAL
 *      or POINTER
 *      or SAVE
 *      or TARGET
 *
 *  component-attr-spec
 *      is POINTER
 *      or DIMENSION ( component-array-spec )
 */
‌static void parseQualifierSpecList (tokenInfo *const token)
{
    do
    {
        readToken (token);  /* should be an attr-spec */
        switch (token->keyword)
        {
            case KEYWORD_parameter:
            case KEYWORD_allocatable:
            case KEYWORD_external:
            case KEYWORD_intrinsic:
            case KEYWORD_optional:
            case KEYWORD_private:
            case KEYWORD_pointer:
            case KEYWORD_public:
            case KEYWORD_save:
            case KEYWORD_target:
                readToken (token);
                break;

            case KEYWORD_dimension:
            case KEYWORD_intent:
                readToken (token);
                skipOverParens (token);
                break;

            default: skipToToken (token, TOKEN_STATEMENT_END); break;
        }
    } while (isType (token, TOKEN_COMMA));
    if (! isType (token, TOKEN_DOUBLE_COLON))
        skipToToken (token, TOKEN_STATEMENT_END);
}

‌static tagType variableTagType (void)
{
    tagType result = TAG_VARIABLE;
    if (ancestorCount () > 0)
    {
        const tokenInfo* const parent = ancestorTop ();
        switch (parent->tag)
        {
            case TAG_MODULE:       result = TAG_VARIABLE;  break;
            case TAG_DERIVED_TYPE: result = TAG_COMPONENT; break;
            case TAG_FUNCTION:     result = TAG_LOCAL;     break;
            case TAG_SUBROUTINE:   result = TAG_LOCAL;     break;
            default:               result = TAG_VARIABLE;  break;
        }
    }
    return result;
}

‌static void parseEntityDecl (tokenInfo *const token)
{
    Assert (isType (token, TOKEN_IDENTIFIER));
    makeFortranTag (token, variableTagType ());
    readToken (token);
    if (isType (token, TOKEN_PAREN_OPEN))
        skipOverParens (token);
    if (isType (token, TOKEN_OPERATOR) &&
            strcmp (vStringValue (token->string), "*") == 0)
    {
        readToken (token);  /* read char-length */
        if (isType (token, TOKEN_PAREN_OPEN))
            skipOverParens (token);
        else
            readToken (token);
    }
    if (isType (token, TOKEN_OPERATOR))
    {
        if (strcmp (vStringValue (token->string), "/") == 0)
        {  /* skip over initializations of structure field */
            readToken (token);
            skipPast (token, TOKEN_OPERATOR);
        }
        else if (strcmp (vStringValue (token->string), "=") == 0)
        {
            while (! isType (token, TOKEN_COMMA) &&
                    ! isType (token, TOKEN_STATEMENT_END))
            {
                readToken (token);
                if (isType (token, TOKEN_PAREN_OPEN))
                    skipOverParens (token);
            }
        }
    }
    /* token left at either comma or statement end */
}

‌static void parseEntityDeclList (tokenInfo *const token)
{
    if (isType (token, TOKEN_PERCENT))
        skipToNextStatement (token);
    else while (isType (token, TOKEN_IDENTIFIER) ||
                (isType (token, TOKEN_KEYWORD) &&
                 !isKeyword (token, KEYWORD_function) &&
                 !isKeyword (token, KEYWORD_subroutine)))
    {
        /* compilers accept keywoeds as identifiers */
        if (isType (token, TOKEN_KEYWORD))
            token->type = TOKEN_IDENTIFIER;
        parseEntityDecl (token);
        if (isType (token, TOKEN_COMMA))
            readToken (token);
        else if (isType (token, TOKEN_STATEMENT_END))
        {
            skipToNextStatement (token);
            break;
        }
    }
}

/*  type-declaration-stmt is
 *      type-spec [[, attr-spec] ... ::] entity-decl-list
 */
‌static void parseTypeDeclarationStmt (tokenInfo *const token)
{
    Assert (isTypeSpec (token));
    parseTypeSpec (token);
    if (!isType (token, TOKEN_STATEMENT_END))  /* if not end of derived type... */
    {
        if (isType (token, TOKEN_COMMA))
            parseQualifierSpecList (token);
        if (isType (token, TOKEN_DOUBLE_COLON))
            readToken (token);
        parseEntityDeclList (token);
    }
    if (isType (token, TOKEN_STATEMENT_END))
        skipToNextStatement (token);
}

/*  namelist-stmt is
 *      NAMELIST /namelist-group-name/ namelist-group-object-list
 *            [[,]/[namelist-group-name]/ namelist-block-object-list] ...
 *
 *  namelist-group-object is
 *      variable-name
 *
 *  common-stmt is
 *      COMMON [/[common-block-name]/] common-block-object-list
 *            [[,]/[common-block-name]/ common-block-object-list] ...
 *
 *  common-block-object is
 *      variable-name [ ( explicit-shape-spec-list ) ]
 */
‌static void parseCommonNamelistStmt (tokenInfo *const token, tagType type)
{
    Assert (isKeyword (token, KEYWORD_common) ||
            isKeyword (token, KEYWORD_namelist));
    readToken (token);
    do
    {
        if (isType (token, TOKEN_OPERATOR) &&
            strcmp (vStringValue (token->string), "/") == 0)
        {
            readToken (token);
            if (isType (token, TOKEN_IDENTIFIER))
            {
                makeFortranTag (token, type);
                readToken (token);
            }
            skipPast (token, TOKEN_OPERATOR);
        }
        if (isType (token, TOKEN_IDENTIFIER))
            makeFortranTag (token, TAG_LOCAL);
        readToken (token);
        if (isType (token, TOKEN_PAREN_OPEN))
            skipOverParens (token);  /* skip explicit-shape-spec-list */
        if (isType (token, TOKEN_COMMA))
            readToken (token);
    } while (! isType (token, TOKEN_STATEMENT_END));
    skipToNextStatement (token);
}

‌static void parseFieldDefinition (tokenInfo *const token)
{
    if (isTypeSpec (token))
        parseTypeDeclarationStmt (token);
    else if (isKeyword (token, KEYWORD_structure))
        parseStructureStmt (token);
    else if (isKeyword (token, KEYWORD_union))
        parseUnionStmt (token);
    else
        skipToNextStatement (token);
}

‌static void parseMap (tokenInfo *const token)
{
    Assert (isKeyword (token, KEYWORD_map));
    skipToNextStatement (token);
    while (! isKeyword (token, KEYWORD_end))
        parseFieldDefinition (token);
    readSubToken (token);
    /* should be at KEYWORD_map token */
    skipToNextStatement (token);
}

/* UNION
 *      MAP
 *          [field-definition] [field-definition] ...
 *      END MAP
 *      MAP
 *          [field-definition] [field-definition] ...
 *      END MAP
 *      [MAP
 *          [field-definition]
 *          [field-definition] ...
 *      END MAP] ...
 *  END UNION
 *      *
 *
 *  Typed data declarations (variables or arrays) in structure declarations
 *  have the form of normal Fortran typed data declarations. Data items with
 *  different types can be freely intermixed within a structure declaration.
 *
 *  Unnamed fields can be declared in a structure by specifying the pseudo
 *  name %FILL in place of an actual field name. You can use this mechanism to
 *  generate empty space in a record for purposes such as alignment.
 *
 *  All mapped field declarations that are made within a UNION declaration
 *  share a common location within the containing structure. When initializing
 *  the fields within a UNION, the final initialization value assigned
 *  overlays any value previously assigned to a field definition that shares
 *  that field.
 */
‌static void parseUnionStmt (tokenInfo *const token)
{
    Assert (isKeyword (token, KEYWORD_union));
    skipToNextStatement (token);
    while (isKeyword (token, KEYWORD_map))
        parseMap (token);
    /* should be at KEYWORD_end token */
    readSubToken (token);
    /* secondary token should be KEYWORD_end token */
    skipToNextStatement (token);
}

/*  STRUCTURE [/structure-name/] [field-names]
 *      [field-definition]
 *      [field-definition] ...
 *  END STRUCTURE
 *
 *  structure-name
 *        identifies the structure in a subsequent RECORD statement.
 *        Substructures can be established within a structure by means of either
 *        a nested STRUCTURE declaration or a RECORD statement.
 *
 *   field-names
 *        (for substructure declarations only) one or more names having the
 *        structure of the substructure being defined.
 *
 *   field-definition
 *        can be one or more of the following:
 *
 *            Typed data declarations, which can optionally include one or more
 *            data initialization values.
 *
 *            Substructure declarations (defined by either RECORD statements or
 *            subsequent STRUCTURE statements).
 *
 *            UNION declarations, which are mapped fields defined by a block of
 *            statements. The syntax of a UNION declaration is described below.
 *
 *            PARAMETER statements, which do not affect the form of the
 *            structure.
 */
‌static void parseStructureStmt (tokenInfo *const token)
{
    tokenInfo *name;
    Assert (isKeyword (token, KEYWORD_structure));
    readToken (token);
    if (isType (token, TOKEN_OPERATOR) &&
        strcmp (vStringValue (token->string), "/") == 0)
    {  /* read structure name */
        readToken (token);
        if (isType (token, TOKEN_IDENTIFIER))
            makeFortranTag (token, TAG_DERIVED_TYPE);
        name = newTokenFrom (token);
        skipPast (token, TOKEN_OPERATOR);
    }
    else
    {  /* fake out anonymous structure */
        name = newToken ();
        name->type = TOKEN_IDENTIFIER;
        name->tag = TAG_DERIVED_TYPE;
        vStringCopyS (name->string, "anonymous");
    }
    while (isType (token, TOKEN_IDENTIFIER))
    {  /* read field names */
        makeFortranTag (token, TAG_COMPONENT);
        readToken (token);
        if (isType (token, TOKEN_COMMA))
            readToken (token);
    }
    skipToNextStatement (token);
    ancestorPush (name);
    while (! isKeyword (token, KEYWORD_end))
        parseFieldDefinition (token);
    readSubToken (token);
    /* secondary token should be KEYWORD_structure token */
    skipToNextStatement (token);
    ancestorPop ();
    deleteToken (name);
}

/*  specification-stmt
 *      is access-stmt      (is access-spec [[::] access-id-list)
 *      or allocatable-stmt (is ALLOCATABLE [::] array-name etc.)
 *      or common-stmt      (is COMMON [ / [common-block-name] /] etc.)
 *      or data-stmt        (is DATA data-stmt-list [[,] data-stmt-set] ...)
 *      or dimension-stmt   (is DIMENSION [::] array-name etc.)
 *      or equivalence-stmt (is EQUIVALENCE equivalence-set-list)
 *      or external-stmt    (is EXTERNAL etc.)
 *      or intent-stmt      (is INTENT ( intent-spec ) [::] etc.)
 *      or instrinsic-stmt  (is INTRINSIC etc.)
 *      or namelist-stmt    (is NAMELIST / namelist-group-name / etc.)
 *      or optional-stmt    (is OPTIONAL [::] etc.)
 *      or pointer-stmt     (is POINTER [::] object-name etc.)
 *      or save-stmt        (is SAVE etc.)
 *      or target-stmt      (is TARGET [::] object-name etc.)
 *
 *  access-spec is PUBLIC or PRIVATE
 */
‌static boolean parseSpecificationStmt (tokenInfo *const token)
{
    boolean result = TRUE;
    switch (token->keyword)
    {
        case KEYWORD_common:
            parseCommonNamelistStmt (token, TAG_COMMON_BLOCK);
            break;

        case KEYWORD_namelist:
            parseCommonNamelistStmt (token, TAG_NAMELIST);
            break;

        case KEYWORD_structure:
            parseStructureStmt (token);
            break;

        case KEYWORD_allocatable:
        case KEYWORD_data:
        case KEYWORD_dimension:
        case KEYWORD_equivalence:
        case KEYWORD_external:
        case KEYWORD_intent:
        case KEYWORD_intrinsic:
        case KEYWORD_optional:
        case KEYWORD_pointer:
        case KEYWORD_private:
        case KEYWORD_public:
        case KEYWORD_save:
        case KEYWORD_target:
            skipToNextStatement (token);
            break;

        default:
            result = FALSE;
            break;
    }
    return result;
}

/*  component-def-stmt is
 *      type-spec [[, component-attr-spec-list] ::] component-decl-list
 *
 *  component-decl is
 *      component-name [ ( component-array-spec ) ] [ * char-length ]
 */
‌static void parseComponentDefStmt (tokenInfo *const token)
{
    Assert (isTypeSpec (token));
    parseTypeSpec (token);
    if (isType (token, TOKEN_COMMA))
        parseQualifierSpecList (token);
    if (isType (token, TOKEN_DOUBLE_COLON))
        readToken (token);
    parseEntityDeclList (token);
}

/*  derived-type-def is
 *      derived-type-stmt is (TYPE [[, access-spec] ::] type-name
 *          [private-sequence-stmt] ... (is PRIVATE or SEQUENCE)
 *          component-def-stmt
 *          [component-def-stmt] ...
 *          end-type-stmt
 */
‌static void parseDerivedTypeDef (tokenInfo *const token)
{
    if (isType (token, TOKEN_COMMA))
        parseQualifierSpecList (token);
    if (isType (token, TOKEN_DOUBLE_COLON))
        readToken (token);
    if (isType (token, TOKEN_IDENTIFIER))
        makeFortranTag (token, TAG_DERIVED_TYPE);
    ancestorPush (token);
    skipToNextStatement (token);
    if (isKeyword (token, KEYWORD_private) ||
        isKeyword (token, KEYWORD_sequence))
    {
        skipToNextStatement (token);
    }
    while (! isKeyword (token, KEYWORD_end))
    {
        if (isTypeSpec (token))
            parseComponentDefStmt (token);
        else
            skipToNextStatement (token);
    }
    readSubToken (token);
    /* secondary token should be KEYWORD_type token */
    skipToToken (token, TOKEN_STATEMENT_END);
    ancestorPop ();
}

/*  interface-block
 *      interface-stmt (is INTERFACE [generic-spec])
 *          [interface-body]
 *          [module-procedure-stmt] ...
 *          end-interface-stmt (is END INTERFACE)
 *
 *  generic-spec
 *      is generic-name
 *      or OPERATOR ( defined-operator )
 *      or ASSIGNMENT ( = )
 *
 *  interface-body
 *      is function-stmt
 *          [specification-part]
 *          end-function-stmt
 *      or subroutine-stmt
 *          [specification-part]
 *          end-subroutine-stmt
 *
 *  module-procedure-stmt is
 *      MODULE PROCEDURE procedure-name-list
 */
‌static void parseInterfaceBlock (tokenInfo *const token)
{
    tokenInfo *name = NULL;
    Assert (isKeyword (token, KEYWORD_interface));
    readToken (token);
    if (isType (token, TOKEN_IDENTIFIER))
    {
        makeFortranTag (token, TAG_INTERFACE);
        name = newTokenFrom (token);
    }
    else if (isKeyword (token, KEYWORD_assignment) ||
             isKeyword (token, KEYWORD_operator))
    {
        readToken (token);
        if (isType (token, TOKEN_PAREN_OPEN))
            readToken (token);
        if (isType (token, TOKEN_OPERATOR))
        {
            makeFortranTag (token, TAG_INTERFACE);
            name = newTokenFrom (token);
        }
    }
    if (name == NULL)
    {
        name = newToken ();
        name->type = TOKEN_IDENTIFIER;
        name->tag = TAG_INTERFACE;
    }
    ancestorPush (name);
    while (! isKeyword (token, KEYWORD_end))
    {
        switch (token->keyword)
        {
            case KEYWORD_function:   parseFunctionSubprogram (token);   break;
            case KEYWORD_subroutine: parseSubroutineSubprogram (token); break;

            default:
                if (isSubprogramPrefix (token))
                    readToken (token);
                else if (isTypeSpec (token))
                    parseTypeSpec (token);
                else
                    skipToNextStatement (token);
                break;
        }
    }
    readSubToken (token);
    /* secondary token should be KEYWORD_interface token */
    skipToNextStatement (token);
    ancestorPop ();
    deleteToken (name);
}

/*  entry-stmt is
 *      ENTRY entry-name [ ( dummy-arg-list ) ]
 */
‌static void parseEntryStmt (tokenInfo *const token)
{
    Assert (isKeyword (token, KEYWORD_entry));
    readToken (token);
    if (isType (token, TOKEN_IDENTIFIER))
        makeFortranTag (token, TAG_ENTRY_POINT);
    skipToNextStatement (token);
}

/*  stmt-function-stmt is
 *      function-name ([dummy-arg-name-list]) = scalar-expr
 */
‌static boolean parseStmtFunctionStmt (tokenInfo *const token)
{
    boolean result = FALSE;
    Assert (isType (token, TOKEN_IDENTIFIER));
#if 0  /* cannot reliably parse this yet */
    makeFortranTag (token, TAG_FUNCTION);
#endif
    readToken (token);
    if (isType (token, TOKEN_PAREN_OPEN))
    {
        skipOverParens (token);
        result = (boolean) (isType (token, TOKEN_OPERATOR) &&
            strcmp (vStringValue (token->string), "=") == 0);
    }
    skipToNextStatement (token);
    return result;
}

‌static boolean isIgnoredDeclaration (tokenInfo *const token)
{
    boolean result;
    switch (token->keyword)
    {
        case KEYWORD_cexternal:
        case KEYWORD_cglobal:
        case KEYWORD_dllexport:
        case KEYWORD_dllimport:
        case KEYWORD_external:
        case KEYWORD_format:
        case KEYWORD_include:
        case KEYWORD_inline:
        case KEYWORD_parameter:
        case KEYWORD_pascal:
        case KEYWORD_pexternal:
        case KEYWORD_pglobal:
        case KEYWORD_static:
        case KEYWORD_value:
        case KEYWORD_virtual:
        case KEYWORD_volatile:
            result = TRUE;
            break;

        default:
            result = FALSE;
            break;
    }
    return result;
}

/*  declaration-construct
 *      [derived-type-def]
 *      [interface-block]
 *      [type-declaration-stmt]
 *      [specification-stmt]
 *      [parameter-stmt] (is PARAMETER ( named-constant-def-list )
 *      [format-stmt]    (is FORMAT format-specification)
 *      [entry-stmt]
 *      [stmt-function-stmt]
 */
‌static boolean parseDeclarationConstruct (tokenInfo *const token)
{
    boolean result = TRUE;
    switch (token->keyword)
    {
        case KEYWORD_entry:        parseEntryStmt (token);      break;
        case KEYWORD_interface:    parseInterfaceBlock (token); break;
        case KEYWORD_stdcall:   readToken (token);           break;
        /* derived type handled by parseTypeDeclarationStmt(); */

        case KEYWORD_automatic:
            readToken (token);
            if (isTypeSpec (token))
                parseTypeDeclarationStmt (token);
            else
                skipToNextStatement (token);
            result = TRUE;
            break;

        default:
            if (isIgnoredDeclaration (token))
                skipToNextStatement (token);
            else if (isTypeSpec (token))
            {
                parseTypeDeclarationStmt (token);
                result = TRUE;
            }
            else if (isType (token, TOKEN_IDENTIFIER))
                result = parseStmtFunctionStmt (token);
            else
                result = parseSpecificationStmt (token);
            break;
    }
    return result;
}

/*  implicit-part-stmt
 *      is [implicit-stmt] (is IMPLICIT etc.)
 *      or [parameter-stmt] (is PARAMETER etc.)
 *      or [format-stmt] (is FORMAT etc.)
 *      or [entry-stmt] (is ENTRY entry-name etc.)
 */
‌static boolean parseImplicitPartStmt (tokenInfo *const token)
{
    boolean result = TRUE;
    switch (token->keyword)
    {
        case KEYWORD_entry: parseEntryStmt (token); break;

        case KEYWORD_implicit:
        case KEYWORD_include:
        case KEYWORD_parameter:
        case KEYWORD_format:
            skipToNextStatement (token);
            break;

        default: result = FALSE; break;
    }
    return result;
}

/*  specification-part is
 *      [use-stmt] ... (is USE module-name etc.)
 *      [implicit-part] (is [implicit-part-stmt] ... [implicit-stmt])
 *      [declaration-construct] ...
 */
‌static boolean parseSpecificationPart (tokenInfo *const token)
{
    boolean result = FALSE;
    while (skipStatementIfKeyword (token, KEYWORD_use))
        result = TRUE;
    while (parseImplicitPartStmt (token))
        result = TRUE;
    while (parseDeclarationConstruct (token))
        result = TRUE;
    return result;
}

/*  block-data is
 *      block-data-stmt (is BLOCK DATA [block-data-name]
 *          [specification-part]
 *          end-block-data-stmt (is END [BLOCK DATA [block-data-name]])
 */
‌static void parseBlockData (tokenInfo *const token)
{
    Assert (isKeyword (token, KEYWORD_block));
    readToken (token);
    if (isKeyword (token, KEYWORD_data))
    {
        readToken (token);
        if (isType (token, TOKEN_IDENTIFIER))
            makeFortranTag (token, TAG_BLOCK_DATA);
    }
    ancestorPush (token);
    skipToNextStatement (token);
    parseSpecificationPart (token);
    while (! isKeyword (token, KEYWORD_end))
        skipToNextStatement (token);
    readSubToken (token);
    /* secondary token should be KEYWORD_NONE or KEYWORD_block token */
    skipToNextStatement (token);
    ancestorPop ();
}

/*  internal-subprogram-part is
 *      contains-stmt (is CONTAINS)
 *          internal-subprogram
 *          [internal-subprogram] ...
 *
 *  internal-subprogram
 *      is function-subprogram
 *      or subroutine-subprogram
 */
‌static void parseInternalSubprogramPart (tokenInfo *const token)
{
    boolean done = FALSE;
    if (isKeyword (token, KEYWORD_contains))
        skipToNextStatement (token);
    do
    {
        switch (token->keyword)
        {
            case KEYWORD_function:   parseFunctionSubprogram (token);   break;
            case KEYWORD_subroutine: parseSubroutineSubprogram (token); break;
            case KEYWORD_end:        done = TRUE;                       break;

            default:
                if (isSubprogramPrefix (token))
                    readToken (token);
                else if (isTypeSpec (token))
                    parseTypeSpec (token);
                else
                    readToken (token);
                break;
        }
    } while (! done);
}

/*  module is
 *      module-stmt (is MODULE module-name)
 *          [specification-part]
 *          [module-subprogram-part]
 *          end-module-stmt (is END [MODULE [module-name]])
 *
 *  module-subprogram-part
 *      contains-stmt (is CONTAINS)
 *          module-subprogram
 *          [module-subprogram] ...
 *
 *  module-subprogram
 *      is function-subprogram
 *      or subroutine-subprogram
 */
‌static void parseModule (tokenInfo *const token)
{
    Assert (isKeyword (token, KEYWORD_module));
    readToken (token);
    if (isType (token, TOKEN_IDENTIFIER))
        makeFortranTag (token, TAG_MODULE);
    ancestorPush (token);
    skipToNextStatement (token);
    parseSpecificationPart (token);
    if (isKeyword (token, KEYWORD_contains))
        parseInternalSubprogramPart (token);
    while (! isKeyword (token, KEYWORD_end))
        skipToNextStatement (token);
    readSubToken (token);
    /* secondary token should be KEYWORD_NONE or KEYWORD_module token */
    skipToNextStatement (token);
    ancestorPop ();
}

/*  execution-part
 *      executable-construct
 *
 *  executable-contstruct is
 *      execution-part-construct [execution-part-construct]
 *
 *  execution-part-construct
 *      is executable-construct
 *      or format-stmt
 *      or data-stmt
 *      or entry-stmt
 */
‌static boolean parseExecutionPart (tokenInfo *const token)
{
    boolean result = FALSE;
    boolean done = FALSE;
    while (! done)
    {
        switch (token->keyword)
        {
            default:
                if (isSubprogramPrefix (token))
                    readToken (token);
                else
                    skipToNextStatement (token);
                result = TRUE;
                break;

            case KEYWORD_entry:
                parseEntryStmt (token);
                result = TRUE;
                break;

            case KEYWORD_contains:
            case KEYWORD_function:
            case KEYWORD_subroutine:
                done = TRUE;
                break;

            case KEYWORD_end:
                readSubToken (token);
                if (isSecondaryKeyword (token, KEYWORD_do) ||
                    isSecondaryKeyword (token, KEYWORD_if) ||
                    isSecondaryKeyword (token, KEYWORD_select) ||
                    isSecondaryKeyword (token, KEYWORD_where))
                {
                    skipToNextStatement (token);
                    result = TRUE;
                }
                else
                    done = TRUE;
                break;
        }
    }
    return result;
}

‌static void parseSubprogram (tokenInfo *const token, const tagType tag)
{
    Assert (isKeyword (token, KEYWORD_program) ||
            isKeyword (token, KEYWORD_function) ||
            isKeyword (token, KEYWORD_subroutine));
    readToken (token);
    if (isType (token, TOKEN_IDENTIFIER))
        makeFortranTag (token, tag);
    ancestorPush (token);
    skipToNextStatement (token);
    parseSpecificationPart (token);
    parseExecutionPart (token);
    if (isKeyword (token, KEYWORD_contains))
        parseInternalSubprogramPart (token);
    /* should be at KEYWORD_end token */
    readSubToken (token);
    /* secondary token should be one of KEYWORD_NONE, KEYWORD_program,
     * KEYWORD_function, KEYWORD_function
     */
    skipToNextStatement (token);
    ancestorPop ();
}


/*  function-subprogram is
 *      function-stmt (is [prefix] FUNCTION function-name etc.)
 *          [specification-part]
 *          [execution-part]
 *          [internal-subprogram-part]
 *          end-function-stmt (is END [FUNCTION [function-name]])
 *
 *  prefix
 *      is type-spec [RECURSIVE]
 *      or [RECURSIVE] type-spec
 */
‌static void parseFunctionSubprogram (tokenInfo *const token)
{
    parseSubprogram (token, TAG_FUNCTION);
}

/*  subroutine-subprogram is
 *      subroutine-stmt (is [RECURSIVE] SUBROUTINE subroutine-name etc.)
 *          [specification-part]
 *          [execution-part]
 *          [internal-subprogram-part]
 *          end-subroutine-stmt (is END [SUBROUTINE [function-name]])
 */
‌static void parseSubroutineSubprogram (tokenInfo *const token)
{
    parseSubprogram (token, TAG_SUBROUTINE);
}

/*  main-program is
 *      [program-stmt] (is PROGRAM program-name)
 *          [specification-part]
 *          [execution-part]
 *          [internal-subprogram-part ]
 *          end-program-stmt
 */
‌static void parseMainProgram (tokenInfo *const token)
{
    parseSubprogram (token, TAG_PROGRAM);
}

/*  program-unit
 *      is main-program
 *      or external-subprogram (is function-subprogram or subroutine-subprogram)
 *      or module
 *      or block-data
 */
‌static void parseProgramUnit (tokenInfo *const token)
{
    readToken (token);
    do
    {
        if (isType (token, TOKEN_STATEMENT_END))
            readToken (token);
        else switch (token->keyword)
        {
            case KEYWORD_block:      parseBlockData (token);            break;
            case KEYWORD_end:        skipToNextStatement (token);       break;
            case KEYWORD_function:   parseFunctionSubprogram (token);   break;
            case KEYWORD_module:     parseModule (token);               break;
            case KEYWORD_program:    parseMainProgram (token);          break;
            case KEYWORD_subroutine: parseSubroutineSubprogram (token); break;

            default:
                if (isSubprogramPrefix (token))
                    readToken (token);
                else
                {
                    boolean one = parseSpecificationPart (token);
                    boolean two = parseExecutionPart (token);
                    if (! (one || two))
                        readToken (token);
                }
                break;
        }
    } while (TRUE);
}

‌static boolean findFortranTags (const unsigned int passCount)
{
    tokenInfo *token;
    exception_t exception;
    boolean retry;

    Assert (passCount < 3);
    Parent = newToken ();
    token = newToken ();
    FreeSourceForm = (boolean) (passCount > 1);
    Column = 0;
    exception = (exception_t) setjmp (Exception);
    if (exception == ExceptionEOF)
        retry = FALSE;
    else if (exception == ExceptionFixedFormat  &&  ! FreeSourceForm)
    {
        verbose ("%s: not fixed source form; retry as free source form\n",
                getInputFileName ());
        retry = TRUE;
    }
    else
    {
        parseProgramUnit (token);
        retry = FALSE;
    }
    ancestorClear ();
    deleteToken (token);
    deleteToken (Parent);

    return retry;
}

‌static void initialize (const langType language)
{
    Lang_fortran = language;
    buildFortranKeywordHash ();
}

‌extern parserDefinition* FortranParser (void)
{
    static const char *const extensions [] = {
        "f", "for", "ftn", "f77", "f90", "f95",
#ifndef CASE_INSENSITIVE_FILENAMES
        "F", "FOR", "FTN", "F77", "F90", "F95",
#endif
        NULL
    };
    parserDefinition* def = parserNew ("Fortran");
    def->kinds      = FortranKinds;
    def->kindCount  = KIND_COUNT (FortranKinds);
    def->extensions = extensions;
    def->parser2    = findFortranTags;
    def->initialize = initialize;
    return def;
}

/* vi:set tabstop=4 shiftwidth=4: */