Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

fix(parse): Hand-roll attribute parsing #503

Merged
merged 1 commit into from
May 28, 2021
Merged

fix(parse): Hand-roll attribute parsing #503

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 17 additions & 0 deletions src/__fixtures__/tests.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -801,6 +801,23 @@ export const tests: [
"XML mode",
{ xmlMode: true },
],
[
"#myID",
[
[
{
action: "equals",
name: "id",
type: "attribute",
namespace: null,
value: "myID",
ignoreCase: null,
},
],
],
"IDs in XML mode",
{ xmlMode: true },
],
[
"fOo[baR]",
[
Expand Down
22 changes: 13 additions & 9 deletions src/parse.spec.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,15 +10,15 @@ describe("parse own tests", () => {
}
});

describe("Collected selectors", () => {
test("(qwery, sizzle, nwmatcher)", () => {
const out = JSON.parse(
readFileSync(`${__dirname}/__fixtures__/out.json`, "utf8")
);
for (const s of Object.keys(out)) {
describe("Collected selectors (qwery, sizzle, nwmatcher)", () => {
const out = JSON.parse(
readFileSync(`${__dirname}/__fixtures__/out.json`, "utf8")
);
for (const s of Object.keys(out)) {
test(s, () => {
expect(parse(s)).toStrictEqual(out[s]);
}
});
});
}
});

const broken = [
Expand All @@ -32,13 +32,17 @@ const broken = [
",a",
"a,",
"[id=012345678901234567890123456789",
"input[name=foo.baz]",
"input[name=foo b]",
"input[name!foo]",
"input[name|]",
"input[name=']",
"input[name=foo[baz]]",
':has("p")',
":has(p",
":foo(p()",
"#",
"##foo",
"/*",
];

describe("Broken selectors", () => {
Expand Down
179 changes: 127 additions & 52 deletions src/parse.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -81,29 +81,24 @@ export type TraversalType =

const reName = /^[^\\#]?(?:\\(?:[\da-f]{1,6}\s?|.)|[\w\-\u00b0-\uFFFF])+/;
const reEscape = /\\([\da-f]{1,6}\s?|(\s)|.)/gi;
// Modified version of https://github.com/jquery/sizzle/blob/master/src/sizzle.js#L87
const reAttr =
/^\s*(?:(\*|[-\w]*)\|)?((?:\\.|[\w\u00b0-\uFFFF-])+)\s*(?:(\S?)=\s*(?:(['"])((?:[^\\]|\\[^])*?)\4|(#?(?:\\.|[\w\u00b0-\uFFFF-])*)|)|)\s*([iIsS])?\s*\]/;

const actionTypes: { [key: string]: AttributeAction } = {
undefined: "exists",
"": "equals",
"~": "element",
"^": "start",
$: "end",
"*": "any",
"!": "not",
"|": "hyphen",
};

const Traversals: { [key: string]: TraversalType } = {
const actionTypes = new Map<string, AttributeAction>([
["~", "element"],
["^", "start"],
["$", "end"],
["*", "any"],
["!", "not"],
["|", "hyphen"],
]);

const Traversals: Record<string, TraversalType> = {
">": "child",
"<": "parent",
"~": "sibling",
"+": "adjacent",
};

const attribSelectors: { [key: string]: [string, AttributeAction] } = {
const attribSelectors: Record<string, [string, AttributeAction]> = {
"#": ["id", "equals"],
".": ["class", "element"],
};
Expand Down Expand Up @@ -302,10 +297,7 @@ function parseSelector(
tokens = [];
sawWS = false;
stripWhitespace(1);
} else if (
firstChar === "/" &&
selector.charAt(selectorIndex + 1) === "*"
) {
} else if (selector.startsWith("/*", selectorIndex)) {
const endIndex = selector.indexOf("*/", selectorIndex + 2);

if (endIndex < 0) {
Expand All @@ -332,51 +324,134 @@ function parseSelector(
ignoreCase: options.xmlMode ? null : false,
});
} else if (firstChar === "[") {
const attributeMatch = selector
.slice(selectorIndex + 1)
.match(reAttr);

if (!attributeMatch) {
throw new Error(
`Malformed attribute selector: ${selector.slice(
selectorIndex
)}`
);
stripWhitespace(1);

// Determine attribute name and namespace

let name;
let namespace: string | null = null;

if (selector.charAt(selectorIndex) === "|") {
namespace = "";
selectorIndex += 1;
}

if (selector.startsWith("*|", selectorIndex)) {
namespace = "*";
selectorIndex += 2;
}

const [
completeSelector,
namespace = null,
baseName,
actionType,
,
quotedValue = "",
value = quotedValue,
forceIgnore,
] = attributeMatch;
name = getName(0);

selectorIndex += completeSelector.length + 1;
let name = unescapeCSS(baseName);
if (
namespace === null &&
selector.charAt(selectorIndex) === "|" &&
selector.charAt(selectorIndex + 1) !== "="
) {
namespace = name;
name = getName(1);
}

if (options.lowerCaseAttributeNames ?? !options.xmlMode) {
name = name.toLowerCase();
}

const ignoreCase =
stripWhitespace(0);

// Determine comparison operation

let action: AttributeAction = "exists";
const possibleAction = actionTypes.get(
selector.charAt(selectorIndex)
);

if (possibleAction) {
action = possibleAction;

if (selector.charAt(selectorIndex + 1) !== "=") {
throw new Error("Expected `=`");
}

stripWhitespace(2);
} else if (selector.charAt(selectorIndex) === "=") {
action = "equals";
stripWhitespace(1);
}

// Determine value

let value = "";
let ignoreCase: boolean | null = null;

if (action !== "exists") {
if (quotes.has(selector.charAt(selectorIndex))) {
const quote = selector.charAt(selectorIndex);
let sectionEnd = selectorIndex + 1;
while (
sectionEnd < selector.length &&
(selector.charAt(sectionEnd) !== quote ||
isEscaped(sectionEnd))
) {
sectionEnd += 1;
}

if (selector.charAt(sectionEnd) !== quote) {
throw new Error("Attribute value didn't end");
}

value = unescapeCSS(
selector.slice(selectorIndex + 1, sectionEnd)
);
selectorIndex = sectionEnd + 1;
} else {
const valueStart = selectorIndex;

while (
selectorIndex < selector.length &&
((!isWhitespace(selector.charAt(selectorIndex)) &&
selector.charAt(selectorIndex) !== "]") ||
isEscaped(selectorIndex))
) {
selectorIndex += 1;
}

value = unescapeCSS(
selector.slice(valueStart, selectorIndex)
);
}

stripWhitespace(0);

// See if we have a force ignore flag

const forceIgnore = selector.charAt(selectorIndex);
// If the forceIgnore flag is set (either `i` or `s`), use that value
forceIgnore
? forceIgnore.toLowerCase() === "i"
: // If `xmlMode` is set, there are no rules; return `null`.
options.xmlMode
? null
: // Otherwise, use the `caseInsensitiveAttributes` list.
caseInsensitiveAttributes.has(name);
if (forceIgnore === "s" || forceIgnore === "S") {
ignoreCase = false;
stripWhitespace(1);
} else if (forceIgnore === "i" || forceIgnore === "I") {
ignoreCase = true;
stripWhitespace(1);
}
}

// If `xmlMode` is set, there are no rules; otherwise, use the `caseInsensitiveAttributes` list.
if (!options.xmlMode) {
// TODO: Skip this for `exists`, as there is no value to compare to.
ignoreCase ??= caseInsensitiveAttributes.has(name);
}

if (selector.charAt(selectorIndex) !== "]") {
throw new Error("Attribute selector didn't terminate");
}

selectorIndex += 1;

const attributeSelector: AttributeSelector = {
type: "attribute",
name,
action: actionTypes[actionType],
value: unescapeCSS(value),
action,
value,
namespace,
ignoreCase,
};
Expand Down
3 changes: 2 additions & 1 deletion src/stringify.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
import { Selector } from "./parse";

const actionTypes: { [key: string]: string } = {
const actionTypes: Record<string, string> = {
equals: "",
element: "~",
start: "^",
Expand All @@ -21,6 +21,7 @@ const charsToEscape = new Set([
"\\",
"(",
")",
"'",
]);

/**
Expand Down