Skip to content

Commit

Permalink
extract rewriting from main module
Browse files Browse the repository at this point in the history
  • Loading branch information
@davidchambers
davidchambers committed Jan 4, 2024
1 parent 51ae7a4 commit 6aa7f6d
Show file tree
Hide file tree
Showing 5 changed files with 257 additions and 328 deletions.
324 changes: 4 additions & 320 deletions lib/doctest.js
View file
Original file line number Diff line number Diff line change
Expand Up @@ -13,332 +13,17 @@ import {dirname, resolve} from 'node:path';
import url from 'node:url';
import vm from 'node:vm';

import * as acorn from 'acorn';
import CoffeeScript from 'coffeescript';
import show from 'sanctuary-show';
import Z from 'sanctuary-type-classes';

import {Incorrect, Correct} from './Comparison.js';
import {Success, effect, encase} from './Effect.js';
import {Line} from './Line.js';
import {Channel} from './Output.js';
import require from './require.js';
import {rewriteCoffee} from './rewriters/coffeescript.js';
import {rewriteEsm, rewriteJs} from './rewriters/javascript.js';


const wrapJs = sourceType => logFunctions => ({input, outputs}) => {
let source = '';
for (const {text} of input.lines) {
source += text.replace (/^\s*([>]|[.]+)[ ]?/, '') + '\n';
}

if (
acorn.parse (
source.replace (/^[{].*[}]\n$/s, '($&)'),
{ecmaVersion: 2023, sourceType}
)
.body[0].type !== 'ExpressionStatement'
) return source;

source = '';
source += '\n';
source += '__doctest.enqueue({\n';
source += ' input: {\n';
source += ' lines: [\n';
for (const {number, text} of input.lines) {
source += ` {number: ${show (number)}, text: ${show (text)}},\n`;
}
source += ' ],\n';
source += ` thunk: ([${logFunctions.join (', ')}]) => {\n`;
source += ' return (\n';
for (const {text} of input.lines) {
source += ` ${text.replace (/^\s*([>]|[.]+)[ ]?/, '')}\n`;
}
source += ' );\n';
source += ' },\n';
source += ' },\n';
source += ' outputs: [\n';
for (const {lines, channel} of outputs) {
source += ' {\n';
source += ' lines: [\n';
for (const {number, text} of lines) {
source += ` {number: ${show (number)}, text: ${show (text)}},\n`;
}
source += ' ],\n';
source += ` channel: ${show (channel)},\n`;
source += ` thunk: ([${logFunctions.join (', ')}]) => {\n`;
{
const parts = lines.map (line =>
line.text.replace (/^\s*([.]+[ ]?)?/, '')
);
const options = {
ecmaVersion: 2023,
allowReturnOutsideFunction: true,
};
let isExpression = false;
try {
isExpression = (
(acorn.parse (parts.join ('\n'), options)).body[0].type
!== 'ExpressionStatement'
);
} catch {}
if (isExpression) {
for (const part of parts) {
source += ` ${part}\n`;
}
} else {
source += ' return (\n';
for (const part of parts) {
source += ` ${part}\n`;
}
source += ' );\n';
}
}
source += ' },\n';
source += ' },\n';
}
source += ' ],\n';
source += '});\n';
return source;
};

const wrapCoffee = logFunctions => ({input, outputs, indent}) => {
let source = '';
source += `${indent}__doctest.enqueue {\n`;
source += `${indent} input: {\n`;
source += `${indent} lines: [\n`;
for (let {number, text} of input.lines) {
number = show (number);
text = show (text);
source += `${indent} {number: ${number}, text: ${text}}\n`;
}
source += `${indent} ]\n`;
source += `${indent} thunk: ([${logFunctions.join (', ')}]) ->\n`;
for (const {text} of input.lines) {
source += `${indent} ${text.replace (/^\s*([>]|[.]+)[ ]?/, '')}\n`;
}
source += `${indent} }\n`;
source += `${indent} outputs: [\n`;
for (const {lines, channel} of outputs) {
source += `${indent} {\n`;
source += `${indent} lines: [\n`;
for (let {number, text} of lines) {
number = show (number);
text = show (text);
source += `${indent} {number: ${number}, text: ${text}}\n`;
}
source += `${indent} ]\n`;
source += `${indent} channel: ${show (channel)}\n`;
source += `${indent} thunk: ([${logFunctions.join (', ')}]) ->\n`;
for (const {text} of lines) {
source += `${indent} ${text.replace (/^\s*([.]+[ ]?)?/, '')}\n`;
}
source += `${indent} }\n`;
}
source += `${indent} ]\n`;
source += `${indent}}\n`;
return source;
};

// contiguous :: Line -> NonEmpty (Array Line) -> Boolean
const contiguous = line => lines => (
line.number === lines[lines.length - 1].number + 1
);

const rewriteJs = sourceType => ({
prefix,
openingDelimiter,
closingDelimiter,
logFunctions,
}) => input => {
const comments = [];
acorn.parse (input, {
ecmaVersion: 2023,
sourceType,
locations: true,
onComment: comments,
});

// literalChunks :: Array (Array2 NonNegativeInteger String)
const literalChunks =
comments
.concat ([{start: Infinity, end: Infinity}])
.reduce (
(accum, {start, end}) => {
accum.chunks.push ([
accum.offset,
input.slice (accum.offset, start),
]);
accum.offset = end;
return accum;
},
{chunks: [], offset: 0}
)
.chunks;

// commentChunks :: Array (Array2 NonNegativeInteger String)
const commentChunks =
comments
.flatMap (
({type, value, start, end, loc}) =>
type === 'Line'
? [{value, start, end, number: loc.start.line}]
: value
.split ('\n')
.reduce (
(accum, value) => {
accum.lines.push ({
value: value.replace (/^\s*[*]/, ''),
start: accum.offset,
end: accum.offset += value.length,
number: accum.number,
});
accum.offset += '\n'.length;
accum.number += 1;
return accum;
},
{offset: start, number: loc.start.line, lines: []}
)
.lines
)
.reduce (
(accum, {value, start, number}) => {
if (value.startsWith (prefix)) {
const text = value
.slice (prefix.length)
.replace (/^\s*/, '');
const line = Line (number) (text);
if (accum.state === 'closed') {
if (text === openingDelimiter) accum.state = 'open';
} else if (text === closingDelimiter) {
accum.state = 'closed';
} else if (text.startsWith ('>')) {
accum.state = 'input';
accum.tests.push ({
input: {channel: null, lines: [line], start},
outputs: [],
});
} else if (text.startsWith ('.')) {
const {input, outputs} = accum.tests[accum.tests.length - 1];
if (accum.state === 'input') {
input.lines.push (line);
} else {
outputs[outputs.length - 1].lines.push (line);
}
} else if (accum.state === 'input' || accum.state === 'outputs') {
const {input, outputs} = accum.tests[accum.tests.length - 1];
if (
// A comment immediately following an input line is an output
// line if and only if it contains non-whitespace characters.
contiguous (line)
(input.lines.concat (outputs.flatMap (o => o.lines))) &&
text !== ''
) {
accum.state = 'outputs';
outputs.push ({channel: null, lines: [line], start});
} else {
accum.state = 'open';
}
}
}
return accum;
},
{tests: [], state: openingDelimiter == null ? 'open' : 'closed'}
)
.tests
.map (test => [test.input.start, wrapJs (sourceType) (logFunctions) (test)]);

return Z.sort (Z.concat (literalChunks, commentChunks))
.map (([, text]) => text)
.join ('');
};

const rewriteCoffee = ({
prefix,
openingDelimiter,
closingDelimiter,
logFunctions,
}) => input => {
const lines = input.match (/^.*(?=\n)/gm);
const chunks = lines.reduce ((accum, text, idx) => {
const isComment = /^[ \t]*#(?!##)/.test (text);
const current = isComment ? accum.commentChunks : accum.literalChunks;
const line = Line (idx + 1) (text);
if (isComment === accum.isComment) {
current[current.length - 1].push (line);
} else {
current.push ([line]);
}
accum.isComment = isComment;
return accum;
}, {
literalChunks: [[]],
commentChunks: [],
isComment: false,
});

const testChunks = chunks.commentChunks.map (commentChunk =>
commentChunk.reduce ((accum, {number, text}) => {
const [, indent, uncommented] = text.match (/^([ \t]*)#(.*)$/);
if (uncommented.startsWith (prefix)) {
const unprefixed = uncommented
.slice (prefix.length)
.replace (/^\s*/, '');
const line = Line (number) (unprefixed);
if (accum.state === 'closed') {
if (unprefixed === openingDelimiter) accum.state = 'open';
} else if (unprefixed === closingDelimiter) {
accum.state = 'closed';
} else if (unprefixed.startsWith ('>')) {
accum.state = 'input';
accum.tests.push ({
indent,
input: {
lines: [line],
},
outputs: [],
});
} else if (unprefixed.startsWith ('.')) {
const {input, outputs} = accum.tests[accum.tests.length - 1];
if (accum.state === 'input') {
input.lines.push (line);
} else {
outputs[outputs.length - 1].lines.push (line);
}
} else if (accum.state === 'input' || accum.state === 'outputs') {
const {input, outputs} = accum.tests[accum.tests.length - 1];
if (
// A comment immediately following an input line is an output
// line if and only if it contains non-whitespace characters.
contiguous (line)
(input.lines
.concat (outputs.flatMap (o => o.lines))) &&
unprefixed !== ''
) {
accum.state = 'outputs';
outputs.push ({channel: null, lines: [line]});
} else {
accum.state = 'open';
}
}
}
return accum;
}, {state: openingDelimiter == null ? 'open' : 'closed', tests: []})
.tests
.map (wrapCoffee (logFunctions))
);

return CoffeeScript.compile (
chunks.literalChunks.reduce (
(s, chunk, idx) => Z.reduce (
(s, line) => `${s}${line}\n`,
chunk.reduce ((s, line) => `${s}${line.text}\n`, s),
idx < testChunks.length ? testChunks[idx] : []
),
''
)
);
};

const sentinel = {};

const readSourceFile = async path => (
Expand Down Expand Up @@ -465,7 +150,7 @@ export default ({
const options = {prefix, openingDelimiter, closingDelimiter, logFunctions};
switch (module) {
case 'esm': {
const rewrite = rewriteJs ('module');
const rewrite = rewriteEsm;
const source = rewrite (options) (await readSourceFile (path));
if (print) return source;
const moduleUrl = url.pathToFileURL (__filename);
Expand All @@ -479,7 +164,7 @@ export default ({
context = {process, exports, module, require, __dirname, __filename};
} // fall through
case undefined: {
const rewrite = coffee ? rewriteCoffee : rewriteJs ('script');
const rewrite = coffee ? rewriteCoffee : rewriteJs;
const source = rewrite (options) (await readSourceFile (path));
if (print) return source;
const queue = await evaluateScript (context) (source);
Expand All @@ -489,5 +174,4 @@ export default ({
throw new Error (`Invalid module ${show (module)}`);
}
}
/* c8 ignore next */
};
Loading

0 comments on commit 6aa7f6d

Please sign in to comment.