static void findAwkTags (void)
{
vString *name = vStringNew ();
const unsigned char *line;
while ((line = readLineFromInputFile ()) != NULL)
{
if (strncmp ((const char*) line, "function", (size_t) 8) == 0 &&
isspace ((int) line [8]))
{
const unsigned char *cp = line + 8;
while (isspace ((int) *cp))
++cp;
while (isalnum ((int) *cp) || *cp == '_')
{
vStringPut (name, (int) *cp);
++cp;
}
vStringTerminate (name);
while (isspace ((int) *cp))
++cp;
if (*cp == '(')
makeSimpleTag (name, AwkKinds, K_FUNCTION);
vStringClear (name);
if (*cp != '\0')
++cp;
}
}
vStringDelete (name);
}
/* in OCaml the file name is the module name used in the language
* with it first letter put in upper case */
static void computeModuleName ( void )
{
/* in Ocaml the file name define a module.
* so we define a module =)
*/
const char *filename = getInputFileName ();
int beginIndex = 0;
int endIndex = strlen (filename) - 1;
vString *moduleName = vStringNew ();
while (filename[endIndex] != '.' && endIndex > 0)
endIndex--;
/* avoid problem with path in front of filename */
beginIndex = endIndex;
while (beginIndex > 0)
{
if (filename[beginIndex] == '\\' || filename[beginIndex] == '/')
{
beginIndex++;
break;
}
beginIndex--;
}
vStringNCopyS (moduleName, &filename[beginIndex], endIndex - beginIndex);
vStringTerminate (moduleName);
if (isLowerAlpha (moduleName->buffer[0]))
moduleName->buffer[0] += ('A' - 'a');
addTag (moduleName, K_MODULE);
vStringDelete (moduleName);
}
static void makeClassTag (tokenInfo *const token, vString *const signature)
{
vString * fulltag;
if ( ! token->ignoreTag )
{
fulltag = vStringNew ();
if (vStringLength (token->scope) > 0)
{
vStringCopy(fulltag, token->scope);
vStringCatS (fulltag, ".");
vStringCatS (fulltag, vStringValue(token->string));
}
else
{
vStringCopy(fulltag, token->string);
}
vStringTerminate(fulltag);
if ( ! stringListHas(ClassNames, vStringValue (fulltag)) )
{
stringListAdd (ClassNames, vStringNewCopy (fulltag));
makeJsTag (token, JSTAG_CLASS, signature);
}
vStringDelete (fulltag);
}
}
static vString *parseNumeric (int c)
{
vString *string = vStringNew ();
vString *integer = parseInteger (c);
vStringCopy (string, integer);
vStringDelete (integer);
c = getcFromInputFile ();
if (c == '.')
{
integer = parseInteger ('\0');
vStringPut (string, c);
vStringCat (string, integer);
vStringDelete (integer);
c = getcFromInputFile ();
}
if (tolower (c) == 'e')
{
integer = parseInteger ('\0');
vStringPut (string, c);
vStringCat (string, integer);
vStringDelete (integer);
}
else if (!isspace (c))
ungetcToInputFile (c);
vStringTerminate (string);
return string;
}
static vString *parseNumeric (int c)
{
static vString *string = NULL;
if (string == NULL)
string = vStringNew ();
vStringCopy (string, parseInteger (c));
c = fileGetc ();
if (c == '.')
{
vStringPut (string, c);
vStringCat (string, parseInteger ('\0'));
c = fileGetc ();
}
if (tolower (c) == 'e')
{
vStringPut (string, c);
vStringCat (string, parseInteger ('\0'));
}
else if (!isspace (c))
fileUngetc (c);
vStringTerminate (string);
return string;
}
/* If a numeric is passed in 'c', this is used as the first digit of the
* numeric being parsed.
*/
static vString *parseInteger (int c)
{
static vString *string = NULL;
if (string == NULL)
string = vStringNew ();
vStringClear (string);
if (c == '\0')
c = fileGetc ();
if (c == '-')
{
vStringPut (string, c);
c = fileGetc ();
}
else if (! isdigit (c))
c = fileGetc ();
while (c != EOF && (isdigit (c) || c == '_'))
{
vStringPut (string, c);
c = fileGetc ();
}
vStringTerminate (string);
fileUngetc (c);
return string;
}
static const unsigned char *parseIdentifier (
const unsigned char *cp, vString *const identifier)
{
boolean stringLit = FALSE;
vStringClear (identifier);
while (*cp != '\0' && (!isIdentifier ((int) *cp) || stringLit))
{
int oneback = *cp;
cp++;
if (oneback == '(' && *cp == '*' && stringLit == FALSE)
{
CommentLevel++;
return ++cp;
}
if (*cp == '"' && oneback != '\\')
{
stringLit = TRUE;
continue;
}
if (stringLit && *cp == '"' && oneback != '\\')
stringLit = FALSE;
}
if (strcmp ((const char *) cp, "") == 0 || cp == NULL)
return cp;
while (isIdentifier ((int) *cp))
{
vStringPut (identifier, (int) *cp);
cp++;
}
vStringTerminate (identifier);
return cp;
}
/* Operator can be defined in OCaml as a function
* so we must be ample enough to parse them normally */
static ocamlKeyword eatOperator (lexingState * st)
{
int count = 0;
const unsigned char *root = st->cp;
vStringClear (st->name);
while (isOperator[st->cp[count]])
{
vStringPut (st->name, st->cp[count]);
count++;
}
vStringTerminate (st->name);
st->cp += count;
if (count <= 1)
{
switch (root[0])
{
case '|':
return Tok_Pipe;
case '=':
return Tok_EQ;
default:
return Tok_Op;
}
}
else if (count == 2 && root[0] == '-' && root[1] == '>')
return Tok_To;
else
return Tok_Op;
}
/*
* Emits a tag for the given 'name' of kind 'kind' at the current nesting.
*/
static void emitRubyTag (vString* name, rubyKind kind)
{
tagEntryInfo tag;
vString* scope;
if (!RubyKinds[kind].enabled) {
return;
}
vStringTerminate (name);
scope = stringListToScope (nesting);
initTagEntry (&tag, vStringValue (name));
if (vStringLength (scope) > 0) {
tag.extensionFields.scope [0] = "class";
tag.extensionFields.scope [1] = vStringValue (scope);
}
tag.kindName = RubyKinds [kind].name;
tag.kind = RubyKinds [kind].letter;
makeTagEntry (&tag);
stringListAdd (nesting, vStringNewCopy (name));
vStringClear (name);
vStringDelete (scope);
}
static void parseLet (const unsigned char *line, int infunction)
{
vString *name = vStringNew ();
const unsigned char *cp = line;
const unsigned char *np = line;
/* get the name */
if (isspace ((int) *cp))
{
while (*cp && isspace ((int) *cp))
++cp;
/*
* Ignore lets which set:
* & - local buffer vim settings
* @ - registers
* [ - Lists or Dictionaries
*/
if (!*cp || *cp == '&' || *cp == '@' || *cp == '[' )
goto cleanUp;
/*
* Ignore vim variables which are read only
* v: - Vim variables.
*/
np = cp;
++np;
if ((int) *cp == 'v' && (int) *np == ':' )
goto cleanUp;
/* Skip non-global vars in functions */
if (infunction && (int) *cp != 'g')
goto cleanUp;
/* deal with spaces, $, @ and & */
while (*cp && *cp != '$' && !isalnum ((int) *cp))
++cp;
if (!*cp)
goto cleanUp;
/* cp = skipPrefix (cp, &scope); */
do
{
if (!*cp)
break;
vStringPut (name, (int) *cp);
++cp;
} while (isalnum ((int) *cp) || *cp == '_' || *cp == '#' || *cp == ':' || *cp == '$');
vStringTerminate (name);
makeSimpleTag (name, VimKinds, K_VARIABLE);
vStringClear (name);
}
cleanUp:
vStringDelete (name);
}
static void addContext (tokenInfo* const parent, const tokenInfo* const child)
{
if (vStringLength (parent->string) > 0)
{
vStringCatS (parent->string, ".");
}
vStringCatS (parent->string, vStringValue(child->string));
vStringTerminate(parent->string);
}
static void addToScope (tokenInfo* const token, vString* const extra)
{
if (vStringLength (token->scope) > 0)
{
vStringCatS (token->scope, ".");
}
vStringCatS (token->scope, vStringValue(extra));
vStringTerminate(token->scope);
}
/* Determines whether or not "name" should be ignored, per the ignore list.
*/
extern boolean isIgnoreToken (const char *const name,
boolean *const pIgnoreParens,
const char **const replacement)
{
boolean result = FALSE;
if (c_tags_ignore != NULL)
{
const size_t nameLen = strlen (name);
unsigned int i;
guint len = g_strv_length (c_tags_ignore);
vString *token = vStringNew();
if (pIgnoreParens != NULL)
*pIgnoreParens = FALSE;
for (i = 0 ; i < len ; ++i)
{
size_t tokenLen;
vStringCopyS (token, c_tags_ignore[i]);
vStringTerminate (token);
tokenLen = vStringLength (token);
if (tokenLen >= 2 && vStringChar (token, tokenLen - 1) == '*' &&
strncmp (vStringValue (token), name, tokenLen - 1) == 0)
{
result = TRUE;
break;
}
if (strncmp (vStringValue (token), name, nameLen) == 0)
{
if (nameLen == tokenLen)
{
result = TRUE;
break;
}
else if (tokenLen == nameLen + 1 &&
vStringChar (token, tokenLen - 1) == '+')
{
result = TRUE;
if (pIgnoreParens != NULL)
*pIgnoreParens = TRUE;
break;
}
else if (vStringChar (token, nameLen) == '=')
{
if (replacement != NULL)
*replacement = vStringValue (token) + nameLen + 1;
break;
}
}
}
vStringDelete (token);
}
return result;
}
/* Algorithm adapted from from GNU etags.
* Scheme tag functions
* look for (def... xyzzy
* look for (def... (xyzzy
* look for (def ... ((... (xyzzy ....
* look for (set! xyzzy
*/
static void readIdentifier (vString *const name, const unsigned char *cp)
{
const unsigned char *p;
vStringClear (name);
/* Go till you get to white space or a syntactic break */
for (p = cp; *p != '\0' && *p != '(' && *p != ')' && !isspace (*p); p++)
vStringPut (name, (int) *p);
vStringTerminate (name);
}
static void add_tag(const char *token, haskellKind kind, vString *name)
{
int i;
for (i = 0; token[i] != '\0'; ++i)
vStringPut(name, token[i]);
vStringTerminate(name);
makeSimpleTag(name, HaskellKinds, kind);
vStringClear(name);
}
static void pushScope (tokenInfo *const token,
const tokenInfo *const parent,
const jsonKind parentKind)
{
if (vStringLength (token->scope) > 0)
vStringPut (token->scope, '.');
vStringCat (token->scope, parent->string);
vStringTerminate (token->scope);
token->scopeKind = parentKind;
}
/* Match the name of a tag (function, variable, type, ...) starting at pos. */
static char const *extract_name (char const *pos, vString * name)
{
while (isspace (*pos))
pos++;
vStringClear (name);
for (; *pos && !isspace (*pos) && *pos != '(' && *pos != ',' && *pos != '='; pos++)
vStringPut (name, *pos);
vStringTerminate (name);
return pos;
}
/*
* Emits a tag for the given 'name' of kind 'kind' at the current nesting.
*/
static void emitRubyTag (vString* name, rubyKind kind)
{
tagEntryInfo tag;
vString* scope;
tagEntryInfo *parent;
rubyKind parent_kind = K_UNDEFINED;
NestingLevel *lvl;
const char *unqualified_name;
const char *qualified_name;
int r;
if (!RubyKinds[kind].enabled) {
return;
}
vStringTerminate (name);
scope = nestingLevelsToScope (nesting);
lvl = nestingLevelsGetCurrent (nesting);
parent = getEntryOfNestingLevel (lvl);
if (parent)
parent_kind = parent->kind - RubyKinds;
qualified_name = vStringValue (name);
unqualified_name = strrchr (qualified_name, SCOPE_SEPARATOR);
if (unqualified_name && unqualified_name[1])
{
if (unqualified_name > qualified_name)
{
if (vStringLength (scope) > 0)
vStringPut (scope, SCOPE_SEPARATOR);
vStringNCatS (scope, qualified_name,
unqualified_name - qualified_name);
/* assume module parent type for a lack of a better option */
parent_kind = K_MODULE;
}
unqualified_name++;
}
else
unqualified_name = qualified_name;
initTagEntry (&tag, unqualified_name, &(RubyKinds [kind]));
if (vStringLength (scope) > 0) {
Assert (0 <= parent_kind &&
(size_t) parent_kind < (ARRAY_SIZE (RubyKinds)));
tag.extensionFields.scopeKind = &(RubyKinds [parent_kind]);
tag.extensionFields.scopeName = vStringValue (scope);
}
r = makeTagEntry (&tag);
nestingLevelsPush (nesting, r);
vStringClear (name);
vStringDelete (scope);
}
static void readAndEmitTag (const unsigned char** cp, rubyKind expected_kind)
{
if (isspace (**cp))
{
vString *name = vStringNew ();
vString *chunk = vStringNew ();
rubyKind actual_kind;
unsigned int i = 0;
/* parse the identifier, allowing scoping like "class Foo::Bar::Baz" */
while (1)
{
actual_kind = parseIdentifier (cp, chunk, expected_kind);
if (i++ > 0)
vStringPut (name, '.');
vStringCat (name, chunk);
vStringClear (chunk);
if (actual_kind != K_UNDEFINED && (*cp)[0] == ':' && (*cp)[1] == ':')
*cp += 2;
else
break;
}
vStringDelete (chunk);
vStringTerminate (name);
if (actual_kind == K_UNDEFINED || vStringLength (name) == 0)
{
/*
* What kind of tags should we create for code like this?
*
* %w(self.clfloor clfloor).each do |name|
* module_eval <<-"end;"
* def #{name}(x, y=1)
* q, r = x.divmod(y)
* q = q.to_i
* return q, r
* end
* end;
* end
*
* Or this?
*
* class << HTTP
*
* For now, we don't create any.
*/
}
else
{
emitRubyTag (name, actual_kind);
}
vStringDelete (name);
}
}
static void parseSelector (vString *const string, const int firstChar)
{
int c = firstChar;
do
{
vStringPut (string, (char) c);
c = fileGetc ();
} while (isSelectorChar (c));
fileUngetc (c);
vStringTerminate (string);
}
static void parseIdentifier (vString *const string, const int firstChar)
{
int c = firstChar;
do
{
vStringPut (string, c);
c = getcFromInputFile ();
} while (isIdentChar (c));
vStringTerminate (string);
ungetcToInputFile (c); /* always unget, LF might add a semicolon */
}
/*
* Copies the characters forming an identifier from *cp into
* name, leaving *cp pointing to the character after the identifier.
*/
static rubyKind parseIdentifier (
const unsigned char** cp, vString* name, rubyKind kind)
{
/* Method names are slightly different to class and variable names.
* A method name may optionally end with a question mark, exclamation
* point or equals sign. These are all part of the name.
* A method name may also contain a period if it's a singleton method.
*/
const char* also_ok = (kind == K_METHOD) ? "_.?!=" : "_";
skipWhitespace (cp);
/* Check for an anonymous (singleton) class such as "class << HTTP". */
if (kind == K_CLASS && **cp == '<' && *(*cp + 1) == '<')
{
return K_UNDEFINED;
}
/* Check for operators such as "def []=(key, val)". */
if (kind == K_METHOD || kind == K_SINGLETON)
{
if (parseRubyOperator (name, cp))
{
return kind;
}
}
/* Copy the identifier into 'name'. */
while (**cp != 0 && (isalnum (**cp) || charIsIn (**cp, also_ok)))
{
char last_char = **cp;
vStringPut (name, last_char);
++*cp;
if (kind == K_METHOD)
{
/* Recognize singleton methods. */
if (last_char == '.')
{
vStringTerminate (name);
vStringClear (name);
return parseIdentifier (cp, name, K_SINGLETON);
}
/* Recognize characters which mark the end of a method name. */
if (charIsIn (last_char, "?!="))
{
break;
}
}
}
return kind;
}
static void initPhpEntry (tagEntryInfo *const e, const tokenInfo *const token,
const phpKind kind, const accessType access)
{
int parentKind = -1;
const char *rootsep;
vStringClear (FullScope);
if (vStringLength (CurrentNamesapce) > 0)
{
parentKind = K_NAMESPACE;
rootsep = scopeSeparatorFor (PhpKinds + parentKind,
KIND_NULL);
vStringCatS (FullScope, rootsep);
vStringCat (FullScope, CurrentNamesapce);
}
initTagEntry (e, vStringValue (token->string), &(PhpKinds[kind]));
e->lineNumber = token->lineNumber;
e->filePosition = token->filePosition;
if (access != ACCESS_UNDEFINED)
e->extensionFields.access = accessToString (access);
if (vStringLength (token->scope) > 0)
{
parentKind = token->parentKind;
if (vStringLength (FullScope) > 0)
{
const char* sep;
sep = phpScopeSeparatorFor (parentKind,
K_NAMESPACE);
vStringCatS (FullScope, sep);
}
else
{
rootsep = scopeSeparatorFor (PhpKinds + parentKind,
KIND_NULL);
vStringCatS (FullScope, rootsep);
}
vStringCat (FullScope, token->scope);
}
if (vStringLength (FullScope) > 0)
{
Assert (parentKind >= 0);
vStringTerminate (FullScope);
e->extensionFields.scopeKind = &(PhpKinds[parentKind]);
e->extensionFields.scopeName = vStringValue (FullScope);
}
}
static void parseFunction (const unsigned char *line)
{
vString *name = vStringNew ();
/* boolean inFunction = FALSE; */
int scope;
const unsigned char *cp = line + 1;
if ((int) *++cp == 'n' && (int) *++cp == 'c' &&
(int) *++cp == 't' && (int) *++cp == 'i' &&
(int) *++cp == 'o' && (int) *++cp == 'n')
++cp;
if ((int) *cp == '!')
++cp;
if (isspace ((int) *cp))
{
while (*cp && isspace ((int) *cp))
++cp;
if (*cp)
{
cp = skipPrefix (cp, &scope);
if (isupper ((int) *cp) ||
scope == 's' || /* script scope */
scope == '<' || /* script scope */
scope == 'd' || /* dictionary */
scope == 'a') /* autoload */
{
do
{
vStringPut (name, (int) *cp);
++cp;
} while (isalnum ((int) *cp) || *cp == '_' || *cp == '.' || *cp == '#');
vStringTerminate (name);
makeSimpleTag (name, VimKinds, K_FUNCTION);
vStringClear (name);
}
}
}
/* TODO - update struct to indicate inside function */
while ((line = readVimLine ()) != NULL)
{
/*
* Vim7 added the for/endfo[r] construct, so we must first
* check for an "endfo", before a "endf"
*/
if ( (!strncmp ((const char*) line, "endfo", (size_t) 5) == 0) &&
(strncmp ((const char*) line, "endf", (size_t) 4) == 0) )
break;
/* TODO - call parseVimLine */
}
vStringDelete (name);
}
static void parseIdentifier (vString *const string, const int firstChar)
{
int c = firstChar;
do
{
vStringPut (string, (char) c);
c = getcFromInputFile ();
} while (isIdentChar (c));
ungetcToInputFile (c);
vStringTerminate (string);
}
static void readIdentifier (const int first, vString *const id)
{
int c = first;
vStringClear (id);
while (isIdentifier (c))
{
vStringPut (id, c);
c = nextChar ();
}
fileUngetc (c);
vStringTerminate (id);
}
/* Read a C identifier beginning with "firstChar" and places it into
* "name".
*/
static void parseIdentifier (vString *const string, const int firstChar)
{
int c = firstChar;
Assert (isIdentChar (c));
do
{
vStringPut (string, c);
c = fileGetc ();
} while (isIdentChar (c));
vStringTerminate (string);
fileUngetc (c); /* unget non-identifier character */
}
static const unsigned char *parseIdentifier (
const unsigned char *cp, vString *const identifier)
{
vStringClear (identifier);
while (isIdentifierCharacter ((int) *cp))
{
vStringPut (identifier, (int) *cp);
++cp;
}
vStringTerminate (identifier);
return cp;
}
static vString* determineEmacsModeAtFirstLine (const char* const line)
{
vString* mode = vStringNew ();
const char* p = strstr(line, "-*-");
if (p == NULL)
return mode;
p += strlen("-*-");
for ( ; isspace ((int) *p) ; ++p)
; /* no-op */
if (strncmp(p, "mode:", strlen("mode:")) == 0)
{
/* -*- mode: MODE; -*- */
p += strlen("mode:");
for ( ; isspace ((int) *p) ; ++p)
; /* no-op */
for ( ; *p != '\0' && isalnum ((int) *p) ; ++p)
vStringPut (mode, (int) *p);
vStringTerminate (mode);
}
else
{
/* -*- MODE -*- */
for ( ; *p != '\0' && isalnum ((int) *p) ; ++p)
vStringPut (mode, (int) *p);
vStringTerminate (mode);
for ( ; isspace ((int) *p) ; ++p)
; /* no-op */
if (strncmp(p, "-*-", strlen("-*-")) != 0)
vStringClear (mode);
}
return mode;
}
static void addToScope (tokenInfo *const token, const vString *const extra,
int kindOfUpperScope)
{
if (vStringLength (token->scope) > 0)
{
const char* sep;
sep = phpScopeSeparatorFor(token->parentKind,
kindOfUpperScope);
vStringCatS (token->scope, sep);
}
vStringCat (token->scope, extra);
vStringTerminate(token->scope);
}
