Created
August 2, 2020 00:06
-
-
Save pconerly/60d26db96414422c00b6f95107e1e537 to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#pragma once | |
#include "defs.h" | |
struct StringParser { | |
const char *pos, *end; | |
char peek() { return *pos; } | |
bool atEnd() { return pos >= end; } | |
void skip(int skip = 1) { pos += skip; } | |
void rewind(int skip = 1) { pos -= skip; } | |
bool accept(char c) { | |
if (*pos == c) { | |
++pos; | |
return true; | |
} | |
return false; | |
} | |
bool accept(const char* str) { | |
auto start = pos; | |
while (*str) { | |
if (!accept(*str)) { | |
pos = start; | |
return false; | |
} | |
++str; | |
} | |
return true; | |
} | |
bool acceptAnyOf(const char *str, char *cout = nullptr) { | |
while (*str) { | |
if (accept(*str)) { | |
if (cout) { | |
*cout = *str; | |
} | |
return true; | |
} | |
++str; | |
} | |
return false; | |
} | |
bool acceptInRanges(std::initializer_list<std::initializer_list<char>> ranges, char *cout = nullptr) { | |
auto r = ranges.begin(); | |
for (auto&&range : ranges) { | |
if (range.size() == 1) { | |
if (*pos == *range.begin()) { | |
if (cout) { *cout = *pos; } | |
++pos; | |
return true; | |
} | |
} | |
else if (range.size() == 2) { | |
if (*pos >= *range.begin() && *pos <= *(range.begin() + 1)) { | |
if (cout) { *cout = *pos; } | |
++pos; | |
return true; | |
} | |
} | |
} | |
return false; | |
} | |
}; |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
static bool _acceptSigil(StringParser& p) { | |
while (!p.atEnd()) { | |
if (p.accept('@')) return true; | |
p.skip(); | |
} | |
return false; | |
} | |
static bool _acceptTagCharacter(StringParser& p) { | |
return p.acceptInRanges({ { 'a', 'z' },{ 'A', 'Z' },{ '0', '9' },{ '_' },{'.'} }); | |
} | |
static bool _acceptTag(StringParser& p) { | |
auto start = p.pos; | |
while (!p.atEnd() && _acceptTagCharacter(p)); | |
return p.pos != start; | |
} | |
static void _parseSpan(VexSpan span, VexNode* parent); | |
static void _parseNode(VexNode* node, StringParser &p) { | |
node->tag.begin = p.pos; | |
_acceptTag(p); | |
node->tag.end = p.pos; | |
bool parseBody = false; | |
if (p.atEnd()) { | |
node->body = { p.pos, p.pos }; | |
} | |
else { | |
if (p.accept(':')) { // body is until line ending | |
node->body.begin = p.pos; | |
while (!p.atEnd() && p.peek() != '\n') p.skip(); | |
node->body.end = p.pos; | |
parseBody = true; | |
} | |
else if (p.accept('=')) { // body is next tag | |
node->body.begin = p.pos; | |
_acceptTag(p); | |
node->body.end = p.pos; | |
} | |
else if (p.accept('{')) { // start scope | |
int bracketCount = 1; | |
while (p.accept('{')) { ++bracketCount; } | |
node->body.begin = p.pos; | |
// find end | |
int scopeStack = 1; | |
while (true) { | |
if (p.atEnd()) { | |
node->body.end = p.pos; | |
break; | |
} | |
if (p.accept('}')) { | |
int newbracketCount = 1; | |
while (newbracketCount < bracketCount && p.accept('}')) { ++newbracketCount; } | |
if (newbracketCount == bracketCount) { | |
--scopeStack; | |
} | |
if (!scopeStack) { | |
node->body.end = p.pos - bracketCount; | |
break; | |
} | |
} | |
else if (p.accept('{')) { | |
int newbracketCount = 1; | |
while (newbracketCount < bracketCount && p.accept('{')) { ++newbracketCount; } | |
if (newbracketCount == bracketCount) { | |
++scopeStack; | |
} | |
} | |
else { | |
p.skip(); | |
} | |
} | |
parseBody = true; | |
} | |
else if (p.accept('(')) { // everythings a node | |
int bracketCount = 1; | |
while (p.accept('(')) { ++bracketCount; } | |
node->body.begin = p.pos; | |
int scopeStack = 1; | |
// find end parens | |
while (true) { | |
if (p.atEnd()) { | |
node->body.end = p.pos; | |
break; | |
} | |
if (p.accept(')')) { | |
int newbracketCount = 1; | |
while (newbracketCount < bracketCount && p.accept(')')) { ++newbracketCount; } | |
if (newbracketCount == bracketCount) { | |
--scopeStack; | |
} | |
if (!scopeStack) { | |
node->body.end = p.pos - bracketCount; | |
break; | |
} | |
} | |
else if (p.accept('(')) { | |
int newbracketCount = 1; | |
while (newbracketCount < bracketCount && p.accept('(')) { ++newbracketCount; } | |
if (newbracketCount == bracketCount) { | |
++scopeStack; | |
} | |
} | |
else { | |
p.skip(); | |
} | |
} | |
// now parse the inner span for nodes | |
StringParser subp = { node->body.begin, node->body.end }; | |
VexNode* lastChild = nullptr; | |
while (!subp.atEnd()) { | |
VexNode child; | |
child.span.begin = subp.pos; | |
_parseNode(&child, subp); | |
// if the parse found a valid node, alloc it and link it up | |
if (child.span.end > child.span.begin) { | |
auto childPtr = new VexNode(child); | |
if (!lastChild) { node->children = childPtr; } | |
else { lastChild->next = childPtr; } | |
lastChild = childPtr; | |
} | |
// skip to next tag | |
while (!subp.atEnd()) { // the sad vex mustache man | |
if (_acceptTagCharacter(subp) || subp.acceptAnyOf(":={(")) { | |
subp.rewind(); | |
break; | |
} | |
subp.skip(); | |
} | |
} | |
} | |
else { // this node is done | |
node->body = { p.pos, p.pos }; | |
} | |
} | |
node->span.end = p.pos; | |
if (parseBody) { | |
_parseSpan(node->body, node); | |
} | |
} | |
static void _parseSpan(VexSpan span, VexNode* parent) { | |
VexNode* lastChild = nullptr; | |
StringParser p = { span.begin, span.end }; | |
while (_acceptSigil(p)) { | |
auto newNode = new VexNode(); | |
if (!lastChild) {// first child | |
parent->children = newNode; | |
} | |
else { | |
lastChild->next = newNode; | |
} | |
lastChild = newNode; | |
newNode->span.begin = p.pos - 1; //include sigil into span | |
_parseNode(newNode, p); | |
} | |
} | |
VexNode* vexCreate(const char* docStart, const char* docEnd) { | |
auto out = new VexNode(); | |
out->span = out->body = { docStart, docEnd }; | |
out->children = out->next = nullptr; | |
out->tag = { docStart, docStart }; | |
_parseSpan(out->body, out); | |
return out; | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment