747 lines
24 KiB
C++
Executable File
747 lines
24 KiB
C++
Executable File
// Scintilla source code edit control
|
|
/** @file LexPython.cxx
|
|
** Lexer for Python.
|
|
**/
|
|
// Copyright 1998-2002 by Neil Hodgson <neilh@scintilla.org>
|
|
// The License.txt file describes the conditions under which this software may be distributed.
|
|
|
|
#include <stdlib.h>
|
|
#include <string.h>
|
|
#include <stdio.h>
|
|
#include <stdarg.h>
|
|
#include <assert.h>
|
|
#include <ctype.h>
|
|
|
|
#include <string>
|
|
#include <vector>
|
|
#include <map>
|
|
|
|
#include "ILexer.h"
|
|
#include "Scintilla.h"
|
|
#include "SciLexer.h"
|
|
|
|
#include "WordList.h"
|
|
#include "LexAccessor.h"
|
|
#include "Accessor.h"
|
|
#include "StyleContext.h"
|
|
#include "CharacterSet.h"
|
|
#include "LexerModule.h"
|
|
#include "OptionSet.h"
|
|
#include "SubStyles.h"
|
|
|
|
#ifdef SCI_NAMESPACE
|
|
using namespace Scintilla;
|
|
#endif
|
|
|
|
namespace {
|
|
// Use an unnamed namespace to protect the functions and classes from name conflicts
|
|
|
|
/* kwCDef, kwCTypeName only used for Cython */
|
|
enum kwType { kwOther, kwClass, kwDef, kwImport, kwCDef, kwCTypeName, kwCPDef };
|
|
|
|
enum literalsAllowed { litNone = 0, litU = 1, litB = 2 };
|
|
|
|
const int indicatorWhitespace = 1;
|
|
|
|
bool IsPyComment(Accessor &styler, Sci_Position pos, Sci_Position len) {
|
|
return len > 0 && styler[pos] == '#';
|
|
}
|
|
|
|
bool IsPyStringTypeChar(int ch, literalsAllowed allowed) {
|
|
return
|
|
((allowed & litB) && (ch == 'b' || ch == 'B')) ||
|
|
((allowed & litU) && (ch == 'u' || ch == 'U'));
|
|
}
|
|
|
|
bool IsPyStringStart(int ch, int chNext, int chNext2, literalsAllowed allowed) {
|
|
if (ch == '\'' || ch == '"')
|
|
return true;
|
|
if (IsPyStringTypeChar(ch, allowed)) {
|
|
if (chNext == '"' || chNext == '\'')
|
|
return true;
|
|
if ((chNext == 'r' || chNext == 'R') && (chNext2 == '"' || chNext2 == '\''))
|
|
return true;
|
|
}
|
|
if ((ch == 'r' || ch == 'R') && (chNext == '"' || chNext == '\''))
|
|
return true;
|
|
|
|
return false;
|
|
}
|
|
|
|
/* Return the state to use for the string starting at i; *nextIndex will be set to the first index following the quote(s) */
|
|
int GetPyStringState(Accessor &styler, Sci_Position i, Sci_PositionU *nextIndex, literalsAllowed allowed) {
|
|
char ch = styler.SafeGetCharAt(i);
|
|
char chNext = styler.SafeGetCharAt(i + 1);
|
|
|
|
// Advance beyond r, u, or ur prefix (or r, b, or br in Python 3.0), but bail if there are any unexpected chars
|
|
if (ch == 'r' || ch == 'R') {
|
|
i++;
|
|
ch = styler.SafeGetCharAt(i);
|
|
chNext = styler.SafeGetCharAt(i + 1);
|
|
} else if (IsPyStringTypeChar(ch, allowed)) {
|
|
if (chNext == 'r' || chNext == 'R')
|
|
i += 2;
|
|
else
|
|
i += 1;
|
|
ch = styler.SafeGetCharAt(i);
|
|
chNext = styler.SafeGetCharAt(i + 1);
|
|
}
|
|
|
|
if (ch != '"' && ch != '\'') {
|
|
*nextIndex = i + 1;
|
|
return SCE_P_DEFAULT;
|
|
}
|
|
|
|
if (ch == chNext && ch == styler.SafeGetCharAt(i + 2)) {
|
|
*nextIndex = i + 3;
|
|
|
|
if (ch == '"')
|
|
return SCE_P_TRIPLEDOUBLE;
|
|
else
|
|
return SCE_P_TRIPLE;
|
|
} else {
|
|
*nextIndex = i + 1;
|
|
|
|
if (ch == '"')
|
|
return SCE_P_STRING;
|
|
else
|
|
return SCE_P_CHARACTER;
|
|
}
|
|
}
|
|
|
|
inline bool IsAWordChar(int ch) {
|
|
return (ch < 0x80) && (isalnum(ch) || ch == '.' || ch == '_');
|
|
}
|
|
|
|
inline bool IsAWordStart(int ch) {
|
|
return (ch < 0x80) && (isalnum(ch) || ch == '_');
|
|
}
|
|
|
|
static bool IsFirstNonWhitespace(Sci_Position pos, Accessor &styler) {
|
|
Sci_Position line = styler.GetLine(pos);
|
|
Sci_Position start_pos = styler.LineStart(line);
|
|
for (Sci_Position i = start_pos; i < pos; i++) {
|
|
char ch = styler[i];
|
|
if (!(ch == ' ' || ch == '\t'))
|
|
return false;
|
|
}
|
|
return true;
|
|
}
|
|
|
|
// Options used for LexerPython
|
|
struct OptionsPython {
|
|
int whingeLevel;
|
|
bool base2or8Literals;
|
|
bool stringsU;
|
|
bool stringsB;
|
|
bool stringsOverNewline;
|
|
bool keywords2NoSubIdentifiers;
|
|
bool fold;
|
|
bool foldQuotes;
|
|
bool foldCompact;
|
|
|
|
OptionsPython() {
|
|
whingeLevel = 0;
|
|
base2or8Literals = true;
|
|
stringsU = true;
|
|
stringsB = true;
|
|
stringsOverNewline = false;
|
|
keywords2NoSubIdentifiers = false;
|
|
fold = false;
|
|
foldQuotes = false;
|
|
foldCompact = false;
|
|
}
|
|
|
|
literalsAllowed AllowedLiterals() const {
|
|
literalsAllowed allowedLiterals = stringsU ? litU : litNone;
|
|
if (stringsB)
|
|
allowedLiterals = static_cast<literalsAllowed>(allowedLiterals | litB);
|
|
return allowedLiterals;
|
|
}
|
|
};
|
|
|
|
static const char *const pythonWordListDesc[] = {
|
|
"Keywords",
|
|
"Highlighted identifiers",
|
|
0
|
|
};
|
|
|
|
struct OptionSetPython : public OptionSet<OptionsPython> {
|
|
OptionSetPython() {
|
|
DefineProperty("tab.timmy.whinge.level", &OptionsPython::whingeLevel,
|
|
"For Python code, checks whether indenting is consistent. "
|
|
"The default, 0 turns off indentation checking, "
|
|
"1 checks whether each line is potentially inconsistent with the previous line, "
|
|
"2 checks whether any space characters occur before a tab character in the indentation, "
|
|
"3 checks whether any spaces are in the indentation, and "
|
|
"4 checks for any tab characters in the indentation. "
|
|
"1 is a good level to use.");
|
|
|
|
DefineProperty("lexer.python.literals.binary", &OptionsPython::base2or8Literals,
|
|
"Set to 0 to not recognise Python 3 binary and octal literals: 0b1011 0o712.");
|
|
|
|
DefineProperty("lexer.python.strings.u", &OptionsPython::stringsU,
|
|
"Set to 0 to not recognise Python Unicode literals u\"x\" as used before Python 3.");
|
|
|
|
DefineProperty("lexer.python.strings.b", &OptionsPython::stringsB,
|
|
"Set to 0 to not recognise Python 3 bytes literals b\"x\".");
|
|
|
|
DefineProperty("lexer.python.strings.over.newline", &OptionsPython::stringsOverNewline,
|
|
"Set to 1 to allow strings to span newline characters.");
|
|
|
|
DefineProperty("lexer.python.keywords2.no.sub.identifiers", &OptionsPython::keywords2NoSubIdentifiers,
|
|
"When enabled, it will not style keywords2 items that are used as a sub-identifier. "
|
|
"Example: when set, will not highlight \"foo.open\" when \"open\" is a keywords2 item.");
|
|
|
|
DefineProperty("fold", &OptionsPython::fold);
|
|
|
|
DefineProperty("fold.quotes.python", &OptionsPython::foldQuotes,
|
|
"This option enables folding multi-line quoted strings when using the Python lexer.");
|
|
|
|
DefineProperty("fold.compact", &OptionsPython::foldCompact);
|
|
|
|
DefineWordListSets(pythonWordListDesc);
|
|
}
|
|
};
|
|
|
|
const char styleSubable[] = { SCE_P_IDENTIFIER, 0 };
|
|
|
|
}
|
|
|
|
class LexerPython : public ILexerWithSubStyles {
|
|
WordList keywords;
|
|
WordList keywords2;
|
|
OptionsPython options;
|
|
OptionSetPython osPython;
|
|
enum { ssIdentifier };
|
|
SubStyles subStyles;
|
|
public:
|
|
explicit LexerPython() :
|
|
subStyles(styleSubable, 0x80, 0x40, 0) {
|
|
}
|
|
virtual ~LexerPython() {
|
|
}
|
|
void SCI_METHOD Release() {
|
|
delete this;
|
|
}
|
|
int SCI_METHOD Version() const {
|
|
return lvSubStyles;
|
|
}
|
|
const char * SCI_METHOD PropertyNames() {
|
|
return osPython.PropertyNames();
|
|
}
|
|
int SCI_METHOD PropertyType(const char *name) {
|
|
return osPython.PropertyType(name);
|
|
}
|
|
const char * SCI_METHOD DescribeProperty(const char *name) {
|
|
return osPython.DescribeProperty(name);
|
|
}
|
|
Sci_Position SCI_METHOD PropertySet(const char *key, const char *val);
|
|
const char * SCI_METHOD DescribeWordListSets() {
|
|
return osPython.DescribeWordListSets();
|
|
}
|
|
Sci_Position SCI_METHOD WordListSet(int n, const char *wl);
|
|
void SCI_METHOD Lex(Sci_PositionU startPos, Sci_Position length, int initStyle, IDocument *pAccess);
|
|
void SCI_METHOD Fold(Sci_PositionU startPos, Sci_Position length, int initStyle, IDocument *pAccess);
|
|
|
|
void * SCI_METHOD PrivateCall(int, void *) {
|
|
return 0;
|
|
}
|
|
|
|
int SCI_METHOD LineEndTypesSupported() {
|
|
return SC_LINE_END_TYPE_UNICODE;
|
|
}
|
|
|
|
int SCI_METHOD AllocateSubStyles(int styleBase, int numberStyles) {
|
|
return subStyles.Allocate(styleBase, numberStyles);
|
|
}
|
|
int SCI_METHOD SubStylesStart(int styleBase) {
|
|
return subStyles.Start(styleBase);
|
|
}
|
|
int SCI_METHOD SubStylesLength(int styleBase) {
|
|
return subStyles.Length(styleBase);
|
|
}
|
|
int SCI_METHOD StyleFromSubStyle(int subStyle) {
|
|
int styleBase = subStyles.BaseStyle(subStyle);
|
|
return styleBase;
|
|
}
|
|
int SCI_METHOD PrimaryStyleFromStyle(int style) {
|
|
return style;
|
|
}
|
|
void SCI_METHOD FreeSubStyles() {
|
|
subStyles.Free();
|
|
}
|
|
void SCI_METHOD SetIdentifiers(int style, const char *identifiers) {
|
|
subStyles.SetIdentifiers(style, identifiers);
|
|
}
|
|
int SCI_METHOD DistanceToSecondaryStyles() {
|
|
return 0;
|
|
}
|
|
const char * SCI_METHOD GetSubStyleBases() {
|
|
return styleSubable;
|
|
}
|
|
|
|
static ILexer *LexerFactoryPython() {
|
|
return new LexerPython();
|
|
}
|
|
};
|
|
|
|
Sci_Position SCI_METHOD LexerPython::PropertySet(const char *key, const char *val) {
|
|
if (osPython.PropertySet(&options, key, val)) {
|
|
return 0;
|
|
}
|
|
return -1;
|
|
}
|
|
|
|
Sci_Position SCI_METHOD LexerPython::WordListSet(int n, const char *wl) {
|
|
WordList *wordListN = 0;
|
|
switch (n) {
|
|
case 0:
|
|
wordListN = &keywords;
|
|
break;
|
|
case 1:
|
|
wordListN = &keywords2;
|
|
break;
|
|
}
|
|
Sci_Position firstModification = -1;
|
|
if (wordListN) {
|
|
WordList wlNew;
|
|
wlNew.Set(wl);
|
|
if (*wordListN != wlNew) {
|
|
wordListN->Set(wl);
|
|
firstModification = 0;
|
|
}
|
|
}
|
|
return firstModification;
|
|
}
|
|
|
|
void SCI_METHOD LexerPython::Lex(Sci_PositionU startPos, Sci_Position length, int initStyle, IDocument *pAccess) {
|
|
Accessor styler(pAccess, NULL);
|
|
|
|
const Sci_Position endPos = startPos + length;
|
|
|
|
// Backtrack to previous line in case need to fix its tab whinging
|
|
Sci_Position lineCurrent = styler.GetLine(startPos);
|
|
if (startPos > 0) {
|
|
if (lineCurrent > 0) {
|
|
lineCurrent--;
|
|
// Look for backslash-continued lines
|
|
while (lineCurrent > 0) {
|
|
Sci_Position eolPos = styler.LineStart(lineCurrent) - 1;
|
|
int eolStyle = styler.StyleAt(eolPos);
|
|
if (eolStyle == SCE_P_STRING
|
|
|| eolStyle == SCE_P_CHARACTER
|
|
|| eolStyle == SCE_P_STRINGEOL) {
|
|
lineCurrent -= 1;
|
|
} else {
|
|
break;
|
|
}
|
|
}
|
|
startPos = styler.LineStart(lineCurrent);
|
|
}
|
|
initStyle = startPos == 0 ? SCE_P_DEFAULT : styler.StyleAt(startPos - 1);
|
|
}
|
|
|
|
const literalsAllowed allowedLiterals = options.AllowedLiterals();
|
|
|
|
initStyle = initStyle & 31;
|
|
if (initStyle == SCE_P_STRINGEOL) {
|
|
initStyle = SCE_P_DEFAULT;
|
|
}
|
|
|
|
kwType kwLast = kwOther;
|
|
int spaceFlags = 0;
|
|
styler.IndentAmount(lineCurrent, &spaceFlags, IsPyComment);
|
|
bool base_n_number = false;
|
|
|
|
const WordClassifier &classifierIdentifiers = subStyles.Classifier(SCE_P_IDENTIFIER);
|
|
|
|
StyleContext sc(startPos, endPos - startPos, initStyle, styler);
|
|
|
|
bool indentGood = true;
|
|
Sci_Position startIndicator = sc.currentPos;
|
|
bool inContinuedString = false;
|
|
|
|
for (; sc.More(); sc.Forward()) {
|
|
|
|
if (sc.atLineStart) {
|
|
styler.IndentAmount(lineCurrent, &spaceFlags, IsPyComment);
|
|
indentGood = true;
|
|
if (options.whingeLevel == 1) {
|
|
indentGood = (spaceFlags & wsInconsistent) == 0;
|
|
} else if (options.whingeLevel == 2) {
|
|
indentGood = (spaceFlags & wsSpaceTab) == 0;
|
|
} else if (options.whingeLevel == 3) {
|
|
indentGood = (spaceFlags & wsSpace) == 0;
|
|
} else if (options.whingeLevel == 4) {
|
|
indentGood = (spaceFlags & wsTab) == 0;
|
|
}
|
|
if (!indentGood) {
|
|
styler.IndicatorFill(startIndicator, sc.currentPos, indicatorWhitespace, 0);
|
|
startIndicator = sc.currentPos;
|
|
}
|
|
}
|
|
|
|
if (sc.atLineEnd) {
|
|
if ((sc.state == SCE_P_DEFAULT) ||
|
|
(sc.state == SCE_P_TRIPLE) ||
|
|
(sc.state == SCE_P_TRIPLEDOUBLE)) {
|
|
// Perform colourisation of white space and triple quoted strings at end of each line to allow
|
|
// tab marking to work inside white space and triple quoted strings
|
|
sc.SetState(sc.state);
|
|
}
|
|
lineCurrent++;
|
|
if ((sc.state == SCE_P_STRING) || (sc.state == SCE_P_CHARACTER)) {
|
|
if (inContinuedString || options.stringsOverNewline) {
|
|
inContinuedString = false;
|
|
} else {
|
|
sc.ChangeState(SCE_P_STRINGEOL);
|
|
sc.ForwardSetState(SCE_P_DEFAULT);
|
|
}
|
|
}
|
|
if (!sc.More())
|
|
break;
|
|
}
|
|
|
|
bool needEOLCheck = false;
|
|
|
|
// Check for a state end
|
|
if (sc.state == SCE_P_OPERATOR) {
|
|
kwLast = kwOther;
|
|
sc.SetState(SCE_P_DEFAULT);
|
|
} else if (sc.state == SCE_P_NUMBER) {
|
|
if (!IsAWordChar(sc.ch) &&
|
|
!(!base_n_number && ((sc.ch == '+' || sc.ch == '-') && (sc.chPrev == 'e' || sc.chPrev == 'E')))) {
|
|
sc.SetState(SCE_P_DEFAULT);
|
|
}
|
|
} else if (sc.state == SCE_P_IDENTIFIER) {
|
|
if ((sc.ch == '.') || (!IsAWordChar(sc.ch))) {
|
|
char s[100];
|
|
sc.GetCurrent(s, sizeof(s));
|
|
int style = SCE_P_IDENTIFIER;
|
|
if ((kwLast == kwImport) && (strcmp(s, "as") == 0)) {
|
|
style = SCE_P_WORD;
|
|
} else if (keywords.InList(s)) {
|
|
style = SCE_P_WORD;
|
|
} else if (kwLast == kwClass) {
|
|
style = SCE_P_CLASSNAME;
|
|
} else if (kwLast == kwDef) {
|
|
style = SCE_P_DEFNAME;
|
|
} else if (kwLast == kwCDef || kwLast == kwCPDef) {
|
|
Sci_Position pos = sc.currentPos;
|
|
unsigned char ch = styler.SafeGetCharAt(pos, '\0');
|
|
while (ch != '\0') {
|
|
if (ch == '(') {
|
|
style = SCE_P_DEFNAME;
|
|
break;
|
|
} else if (ch == ':') {
|
|
style = SCE_P_CLASSNAME;
|
|
break;
|
|
} else if (ch == ' ' || ch == '\t' || ch == '\n' || ch == '\r') {
|
|
pos++;
|
|
ch = styler.SafeGetCharAt(pos, '\0');
|
|
} else {
|
|
break;
|
|
}
|
|
}
|
|
} else if (keywords2.InList(s)) {
|
|
if (options.keywords2NoSubIdentifiers) {
|
|
// We don't want to highlight keywords2
|
|
// that are used as a sub-identifier,
|
|
// i.e. not open in "foo.open".
|
|
Sci_Position pos = styler.GetStartSegment() - 1;
|
|
if (pos < 0 || (styler.SafeGetCharAt(pos, '\0') != '.'))
|
|
style = SCE_P_WORD2;
|
|
} else {
|
|
style = SCE_P_WORD2;
|
|
}
|
|
} else {
|
|
int subStyle = classifierIdentifiers.ValueFor(s);
|
|
if (subStyle >= 0) {
|
|
style = subStyle;
|
|
}
|
|
}
|
|
sc.ChangeState(style);
|
|
sc.SetState(SCE_P_DEFAULT);
|
|
if (style == SCE_P_WORD) {
|
|
if (0 == strcmp(s, "class"))
|
|
kwLast = kwClass;
|
|
else if (0 == strcmp(s, "def"))
|
|
kwLast = kwDef;
|
|
else if (0 == strcmp(s, "import"))
|
|
kwLast = kwImport;
|
|
else if (0 == strcmp(s, "cdef"))
|
|
kwLast = kwCDef;
|
|
else if (0 == strcmp(s, "cpdef"))
|
|
kwLast = kwCPDef;
|
|
else if (0 == strcmp(s, "cimport"))
|
|
kwLast = kwImport;
|
|
else if (kwLast != kwCDef && kwLast != kwCPDef)
|
|
kwLast = kwOther;
|
|
} else if (kwLast != kwCDef && kwLast != kwCPDef) {
|
|
kwLast = kwOther;
|
|
}
|
|
}
|
|
} else if ((sc.state == SCE_P_COMMENTLINE) || (sc.state == SCE_P_COMMENTBLOCK)) {
|
|
if (sc.ch == '\r' || sc.ch == '\n') {
|
|
sc.SetState(SCE_P_DEFAULT);
|
|
}
|
|
} else if (sc.state == SCE_P_DECORATOR) {
|
|
if (!IsAWordChar(sc.ch)) {
|
|
sc.SetState(SCE_P_DEFAULT);
|
|
}
|
|
} else if ((sc.state == SCE_P_STRING) || (sc.state == SCE_P_CHARACTER)) {
|
|
if (sc.ch == '\\') {
|
|
if ((sc.chNext == '\r') && (sc.GetRelative(2) == '\n')) {
|
|
sc.Forward();
|
|
}
|
|
if (sc.chNext == '\n' || sc.chNext == '\r') {
|
|
inContinuedString = true;
|
|
} else {
|
|
// Don't roll over the newline.
|
|
sc.Forward();
|
|
}
|
|
} else if ((sc.state == SCE_P_STRING) && (sc.ch == '\"')) {
|
|
sc.ForwardSetState(SCE_P_DEFAULT);
|
|
needEOLCheck = true;
|
|
} else if ((sc.state == SCE_P_CHARACTER) && (sc.ch == '\'')) {
|
|
sc.ForwardSetState(SCE_P_DEFAULT);
|
|
needEOLCheck = true;
|
|
}
|
|
} else if (sc.state == SCE_P_TRIPLE) {
|
|
if (sc.ch == '\\') {
|
|
sc.Forward();
|
|
} else if (sc.Match("\'\'\'")) {
|
|
sc.Forward();
|
|
sc.Forward();
|
|
sc.ForwardSetState(SCE_P_DEFAULT);
|
|
needEOLCheck = true;
|
|
}
|
|
} else if (sc.state == SCE_P_TRIPLEDOUBLE) {
|
|
if (sc.ch == '\\') {
|
|
sc.Forward();
|
|
} else if (sc.Match("\"\"\"")) {
|
|
sc.Forward();
|
|
sc.Forward();
|
|
sc.ForwardSetState(SCE_P_DEFAULT);
|
|
needEOLCheck = true;
|
|
}
|
|
}
|
|
|
|
if (!indentGood && !IsASpaceOrTab(sc.ch)) {
|
|
styler.IndicatorFill(startIndicator, sc.currentPos, indicatorWhitespace, 1);
|
|
startIndicator = sc.currentPos;
|
|
indentGood = true;
|
|
}
|
|
|
|
// One cdef or cpdef line, clear kwLast only at end of line
|
|
if ((kwLast == kwCDef || kwLast == kwCPDef) && sc.atLineEnd) {
|
|
kwLast = kwOther;
|
|
}
|
|
|
|
// State exit code may have moved on to end of line
|
|
if (needEOLCheck && sc.atLineEnd) {
|
|
lineCurrent++;
|
|
styler.IndentAmount(lineCurrent, &spaceFlags, IsPyComment);
|
|
if (!sc.More())
|
|
break;
|
|
}
|
|
|
|
// Check for a new state starting character
|
|
if (sc.state == SCE_P_DEFAULT) {
|
|
if (IsADigit(sc.ch) || (sc.ch == '.' && IsADigit(sc.chNext))) {
|
|
if (sc.ch == '0' && (sc.chNext == 'x' || sc.chNext == 'X')) {
|
|
base_n_number = true;
|
|
sc.SetState(SCE_P_NUMBER);
|
|
} else if (sc.ch == '0' &&
|
|
(sc.chNext == 'o' || sc.chNext == 'O' || sc.chNext == 'b' || sc.chNext == 'B')) {
|
|
if (options.base2or8Literals) {
|
|
base_n_number = true;
|
|
sc.SetState(SCE_P_NUMBER);
|
|
} else {
|
|
sc.SetState(SCE_P_NUMBER);
|
|
sc.ForwardSetState(SCE_P_IDENTIFIER);
|
|
}
|
|
} else {
|
|
base_n_number = false;
|
|
sc.SetState(SCE_P_NUMBER);
|
|
}
|
|
} else if ((IsASCII(sc.ch) && isoperator(static_cast<char>(sc.ch))) || sc.ch == '`') {
|
|
sc.SetState(SCE_P_OPERATOR);
|
|
} else if (sc.ch == '#') {
|
|
sc.SetState(sc.chNext == '#' ? SCE_P_COMMENTBLOCK : SCE_P_COMMENTLINE);
|
|
} else if (sc.ch == '@') {
|
|
if (IsFirstNonWhitespace(sc.currentPos, styler))
|
|
sc.SetState(SCE_P_DECORATOR);
|
|
else
|
|
sc.SetState(SCE_P_OPERATOR);
|
|
} else if (IsPyStringStart(sc.ch, sc.chNext, sc.GetRelative(2), allowedLiterals)) {
|
|
Sci_PositionU nextIndex = 0;
|
|
sc.SetState(GetPyStringState(styler, sc.currentPos, &nextIndex, allowedLiterals));
|
|
while (nextIndex > (sc.currentPos + 1) && sc.More()) {
|
|
sc.Forward();
|
|
}
|
|
} else if (IsAWordStart(sc.ch)) {
|
|
sc.SetState(SCE_P_IDENTIFIER);
|
|
}
|
|
}
|
|
}
|
|
styler.IndicatorFill(startIndicator, sc.currentPos, indicatorWhitespace, 0);
|
|
sc.Complete();
|
|
}
|
|
|
|
static bool IsCommentLine(Sci_Position line, Accessor &styler) {
|
|
Sci_Position pos = styler.LineStart(line);
|
|
Sci_Position eol_pos = styler.LineStart(line + 1) - 1;
|
|
for (Sci_Position i = pos; i < eol_pos; i++) {
|
|
char ch = styler[i];
|
|
if (ch == '#')
|
|
return true;
|
|
else if (ch != ' ' && ch != '\t')
|
|
return false;
|
|
}
|
|
return false;
|
|
}
|
|
|
|
static bool IsQuoteLine(Sci_Position line, Accessor &styler) {
|
|
int style = styler.StyleAt(styler.LineStart(line)) & 31;
|
|
return ((style == SCE_P_TRIPLE) || (style == SCE_P_TRIPLEDOUBLE));
|
|
}
|
|
|
|
|
|
void SCI_METHOD LexerPython::Fold(Sci_PositionU startPos, Sci_Position length, int /*initStyle - unused*/, IDocument *pAccess) {
|
|
if (!options.fold)
|
|
return;
|
|
|
|
Accessor styler(pAccess, NULL);
|
|
|
|
const Sci_Position maxPos = startPos + length;
|
|
const Sci_Position maxLines = (maxPos == styler.Length()) ? styler.GetLine(maxPos) : styler.GetLine(maxPos - 1); // Requested last line
|
|
const Sci_Position docLines = styler.GetLine(styler.Length()); // Available last line
|
|
|
|
// Backtrack to previous non-blank line so we can determine indent level
|
|
// for any white space lines (needed esp. within triple quoted strings)
|
|
// and so we can fix any preceding fold level (which is why we go back
|
|
// at least one line in all cases)
|
|
int spaceFlags = 0;
|
|
Sci_Position lineCurrent = styler.GetLine(startPos);
|
|
int indentCurrent = styler.IndentAmount(lineCurrent, &spaceFlags, NULL);
|
|
while (lineCurrent > 0) {
|
|
lineCurrent--;
|
|
indentCurrent = styler.IndentAmount(lineCurrent, &spaceFlags, NULL);
|
|
if (!(indentCurrent & SC_FOLDLEVELWHITEFLAG) &&
|
|
(!IsCommentLine(lineCurrent, styler)) &&
|
|
(!IsQuoteLine(lineCurrent, styler)))
|
|
break;
|
|
}
|
|
int indentCurrentLevel = indentCurrent & SC_FOLDLEVELNUMBERMASK;
|
|
|
|
// Set up initial loop state
|
|
startPos = styler.LineStart(lineCurrent);
|
|
int prev_state = SCE_P_DEFAULT & 31;
|
|
if (lineCurrent >= 1)
|
|
prev_state = styler.StyleAt(startPos - 1) & 31;
|
|
int prevQuote = options.foldQuotes && ((prev_state == SCE_P_TRIPLE) || (prev_state == SCE_P_TRIPLEDOUBLE));
|
|
|
|
// Process all characters to end of requested range or end of any triple quote
|
|
//that hangs over the end of the range. Cap processing in all cases
|
|
// to end of document (in case of unclosed quote at end).
|
|
while ((lineCurrent <= docLines) && ((lineCurrent <= maxLines) || prevQuote)) {
|
|
|
|
// Gather info
|
|
int lev = indentCurrent;
|
|
Sci_Position lineNext = lineCurrent + 1;
|
|
int indentNext = indentCurrent;
|
|
int quote = false;
|
|
if (lineNext <= docLines) {
|
|
// Information about next line is only available if not at end of document
|
|
indentNext = styler.IndentAmount(lineNext, &spaceFlags, NULL);
|
|
Sci_Position lookAtPos = (styler.LineStart(lineNext) == styler.Length()) ? styler.Length() - 1 : styler.LineStart(lineNext);
|
|
int style = styler.StyleAt(lookAtPos) & 31;
|
|
quote = options.foldQuotes && ((style == SCE_P_TRIPLE) || (style == SCE_P_TRIPLEDOUBLE));
|
|
}
|
|
const int quote_start = (quote && !prevQuote);
|
|
const int quote_continue = (quote && prevQuote);
|
|
if (!quote || !prevQuote)
|
|
indentCurrentLevel = indentCurrent & SC_FOLDLEVELNUMBERMASK;
|
|
if (quote)
|
|
indentNext = indentCurrentLevel;
|
|
if (indentNext & SC_FOLDLEVELWHITEFLAG)
|
|
indentNext = SC_FOLDLEVELWHITEFLAG | indentCurrentLevel;
|
|
|
|
if (quote_start) {
|
|
// Place fold point at start of triple quoted string
|
|
lev |= SC_FOLDLEVELHEADERFLAG;
|
|
} else if (quote_continue || prevQuote) {
|
|
// Add level to rest of lines in the string
|
|
lev = lev + 1;
|
|
}
|
|
|
|
// Skip past any blank lines for next indent level info; we skip also
|
|
// comments (all comments, not just those starting in column 0)
|
|
// which effectively folds them into surrounding code rather
|
|
// than screwing up folding.
|
|
|
|
while (!quote &&
|
|
(lineNext < docLines) &&
|
|
((indentNext & SC_FOLDLEVELWHITEFLAG) ||
|
|
(lineNext <= docLines && IsCommentLine(lineNext, styler)))) {
|
|
|
|
lineNext++;
|
|
indentNext = styler.IndentAmount(lineNext, &spaceFlags, NULL);
|
|
}
|
|
|
|
const int levelAfterComments = indentNext & SC_FOLDLEVELNUMBERMASK;
|
|
const int levelBeforeComments = Maximum(indentCurrentLevel,levelAfterComments);
|
|
|
|
// Now set all the indent levels on the lines we skipped
|
|
// Do this from end to start. Once we encounter one line
|
|
// which is indented more than the line after the end of
|
|
// the comment-block, use the level of the block before
|
|
|
|
Sci_Position skipLine = lineNext;
|
|
int skipLevel = levelAfterComments;
|
|
|
|
while (--skipLine > lineCurrent) {
|
|
int skipLineIndent = styler.IndentAmount(skipLine, &spaceFlags, NULL);
|
|
|
|
if (options.foldCompact) {
|
|
if ((skipLineIndent & SC_FOLDLEVELNUMBERMASK) > levelAfterComments)
|
|
skipLevel = levelBeforeComments;
|
|
|
|
int whiteFlag = skipLineIndent & SC_FOLDLEVELWHITEFLAG;
|
|
|
|
styler.SetLevel(skipLine, skipLevel | whiteFlag);
|
|
} else {
|
|
if ((skipLineIndent & SC_FOLDLEVELNUMBERMASK) > levelAfterComments &&
|
|
!(skipLineIndent & SC_FOLDLEVELWHITEFLAG) &&
|
|
!IsCommentLine(skipLine, styler))
|
|
skipLevel = levelBeforeComments;
|
|
|
|
styler.SetLevel(skipLine, skipLevel);
|
|
}
|
|
}
|
|
|
|
// Set fold header on non-quote line
|
|
if (!quote && !(indentCurrent & SC_FOLDLEVELWHITEFLAG)) {
|
|
if ((indentCurrent & SC_FOLDLEVELNUMBERMASK) < (indentNext & SC_FOLDLEVELNUMBERMASK))
|
|
lev |= SC_FOLDLEVELHEADERFLAG;
|
|
}
|
|
|
|
// Keep track of triple quote state of previous line
|
|
prevQuote = quote;
|
|
|
|
// Set fold level for this line and move to next line
|
|
styler.SetLevel(lineCurrent, options.foldCompact ? lev : lev & ~SC_FOLDLEVELWHITEFLAG);
|
|
indentCurrent = indentNext;
|
|
lineCurrent = lineNext;
|
|
}
|
|
|
|
// NOTE: Cannot set level of last line here because indentCurrent doesn't have
|
|
// header flag set; the loop above is crafted to take care of this case!
|
|
//styler.SetLevel(lineCurrent, indentCurrent);
|
|
}
|
|
|
|
LexerModule lmPython(SCLEX_PYTHON, LexerPython::LexerFactoryPython, "python",
|
|
pythonWordListDesc);
|