feat: migrate SlowBuffer to Buffer.allocUnsafeSlow() #193

lluisemper · 2025-08-28T20:37:37Z

Add codemod to handle deprecated SlowBuffer API migration covering CommonJS, ES6 imports, and usage patterns

AugustinMauroy

WOW, thank you and congratulations on your first contribution.
Here is the first wave of reviews. Overall, it's good. You've covered a lot of cases except for dynamic imports. For that, you have the getNodeImportCalls function, which allows you to query all nodes that are dynamic imports.
As mentioned in one of the comments, you're using a lot of regex. It's better to base it on the AST for greater stability. Also, as mentioned in a message, you have tools to visualise things, and for query examples, you have utility functions that are quite complex.

AugustinMauroy · 2025-08-29T10:05:50Z

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/README.md

+2. Direct `SlowBuffer` calls to `Buffer.allocUnsafeSlow()`
+3. Import/require statements to use `Buffer` instead of `SlowBuffer`
+
+## Examples


example on rest of codemods use this structure :

## Examples **Before:**  **After:** 

I think your aproach is better in terms of semantics.

@nodejs/userland-migrations what should we do modify this pr or update other codemod ?

I think it’s fine to just update this PR for now. In the future, it might be a good idea to add a script or system at the project root that, when run with
npm run new:codemon --name='slow-buffer-to-buffer-alloc-unsafe-slow',
generates a template project for implementing a new codemon. New contributors and existing ones would benefit from it to have all the basic down.

If this template project would have a different structure, it would be then a good opportunity to correct existing codemons.

Let me know if you find this idea interesting and I can expand on it, or we can discuss how it could be implemented.

oh interesting point.

Codemod CLI ofter a codemod init but it's don't really cover our code style/architecture.

We can have a script that generate the basic of a codemod.

Simple way copy paste a codemod (what I do now)

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/README.md

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/src/workflow.ts

AugustinMauroy · 2025-08-29T10:17:50Z

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/src/workflow.ts

+ * For ESM imports: { SlowBuffer as SB } -> { Buffer as SB }
+ * Preserves original formatting as much as possible
+ */
+function transformDestructuringPattern(originalText: string): string {


Here it's use too much regex, we need to relie on AST for reliability. If you have difficult to visualize the ast you can use Codemod Studio, ast-grep playground or any Tree-sitter sitter visualization tools.

AugustinMauroy · 2025-08-29T10:19:34Z

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/src/workflow.ts

+/**
+ * Process comments mentioning SlowBuffer and update them
+ */
+function processSlowBufferComments(rootNode: SgNode, edits: Edit[]): boolean {


Good idea but in practice we doesn't want to update the user comment. But you can console.warn + file://<path>:<line> where a comment mention this api.

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/workflow.yaml

AugustinMauroy · 2025-08-29T10:27:19Z

for the red ci just run npm I to add your new recipes to the package-lock.json

lluisemper · 2025-08-29T13:53:34Z

@AugustinMauroy thank you very much for the review 💪. I have commited the suggestions and marked the addressed comments with a thums up emoji.

The comments that have a requested change, do not have any emoji. I will aim to commit this over the weekend.

AugustinMauroy · 2025-08-29T14:00:19Z

@AugustinMauroy thank you very much for the review 💪. I have commited the suggestions and marked the addressed comments with a thums up emoji.

The comments that have a requested change, do not have any emoji. I will aim to commit this over the weekend.

Wow super! also any pressure we are open source.

Also for your information you have a button with "resolve conversation" which is collapsing the review 😁

brunocroh · 2025-08-31T19:00:48Z

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/src/workflow.ts

+	if (
+		originalText.includes("SlowBuffer") &&
+		!originalText.includes(",") &&
+		!originalText.includes(":") &&
+		!originalText.includes(" as ")
+	) {
+		return originalText.replace(/SlowBuffer/g, "Buffer");
+	}
+
+	let newText = originalText;
+
+	// Handle aliased patterns (both CommonJS : and ESM as syntax)
+	// { SlowBuffer: SB } -> { Buffer: SB }
+	newText = newText.replace(/SlowBuffer(\s*:\s*\w+)/g, "Buffer$1");
+	// { SlowBuffer as SB } -> { Buffer as SB }
+	newText = newText.replace(/SlowBuffer(\s+as\s+\w+)/g, "Buffer$1");
+
+	// If Buffer is already present in this specific pattern, just remove SlowBuffer
+	if (originalText.includes("Buffer") && originalText.includes("SlowBuffer")) {
+		// Remove non-aliased SlowBuffer references very carefully to preserve spacing
+		newText = newText
+			.replace(/,\s*SlowBuffer(?!\s*[:as])/g, "") // Remove SlowBuffer with leading comma
+			.replace(/SlowBuffer\s*,\s*/g, "") // Remove SlowBuffer with trailing comma and space
+			.replace(/SlowBuffer(?!\s*[:as])/g, ""); // Remove standalone SlowBuffer (not followed by : or as)
+
+		// Clean up any double spaces after opening brace
+		newText = newText.replace(/{\s{2,}/g, "{ ");
+		// Clean up any double commas but preserve spacing
+		newText = newText.replace(/,\s*,/g, ",");
+	}
+	// If Buffer is not present, replace first SlowBuffer with Buffer, remove others
+	else if (originalText.includes("SlowBuffer")) {
+		// Replace the first non-aliased SlowBuffer with Buffer
+		newText = newText.replace(/SlowBuffer(?!\s*[:as])/, "Buffer");
+		// Remove any remaining non-aliased SlowBuffer references
+		newText = newText
+			.replace(/,\s*SlowBuffer\s*(?![\s:as])/g, "")
+			.replace(/SlowBuffer\s*,\s*/g, "")
+			.replace(/SlowBuffer\s*(?![\s:as])/g, "");
+		// Clean up commas
+		newText = newText.replace(/,\s*,/g, ",");
+	}
+
+	return newText;
+}


We have a utility function that resolves these scenarios

please check utils/ast-grep/resolve-binding-path.ts

brunocroh · 2025-08-31T19:31:04Z

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/src/workflow.ts

+	const objectPatterns = rootNode.findAll({ rule: { kind: "object_pattern" } });
+	for (const pattern of objectPatterns) {
+		const text = pattern.text();
+		// Match patterns like { SlowBuffer: SB } to extract alias "SB"
+		const aliasMatches = text.matchAll(/SlowBuffer\s*:\s*(\w+)/g);
+		for (const match of aliasMatches) {
+			slowBufferAliases.add(match[1]);
+		}
+	}
+
+	// Find all named_imports that include SlowBuffer aliases (ESM imports)
+	const namedImports = rootNode.findAll({ rule: { kind: "named_imports" } });
+	for (const importNode of namedImports) {
+		const text = importNode.text();
+		// Match patterns like { SlowBuffer as SB } to extract alias "SB"
+		const aliasMatches = text.matchAll(/SlowBuffer\s+as\s+(\w+)/g);
+		for (const match of aliasMatches) {
+			slowBufferAliases.add(match[1]);
+		}
+	}
+
+	// Handle constructor patterns: new SlowBuffer(size) and new SB(size) for aliases
+	const constructorCalls = rootNode.findAll({
+		rule: {
+			kind: "new_expression",
+			has: {
+				field: "constructor",
+				kind: "identifier",
+				regex: "^SlowBuffer$",
+			},
+		},
+	});
+
+	for (const constructorCall of constructorCalls) {
+		const args = constructorCall.field("arguments");
+		if (args) {
+			// Extract the arguments text (without parentheses)
+			const argsText = args.text().slice(1, -1); // Remove ( and )
+			edits.push(constructorCall.replace(`Buffer.allocUnsafeSlow(${argsText})`));
+			hasChanges = true;
+		}
+	}
+
+	// Handle constructor calls with aliases
+	for (const alias of slowBufferAliases) {
+		const aliasConstructorCalls = rootNode.findAll({
+			rule: {
+				kind: "new_expression",
+				has: {
+					field: "constructor",
+					kind: "identifier",
+					regex: `^${alias}$`,
+				},
+			},
+		});
+
+		for (const constructorCall of aliasConstructorCalls) {
+			const args = constructorCall.field("arguments");
+			if (args) {
+				const argsText = args.text().slice(1, -1);
+				edits.push(constructorCall.replace(`${alias}.allocUnsafeSlow(${argsText})`));
+				hasChanges = true;
+			}
+		}
+	}
+
+	// Handle direct function calls: SlowBuffer(size)
+	const directCalls = rootNode.findAll({
+		rule: {
+			kind: "call_expression",
+			has: {
+				field: "function",
+				kind: "identifier",
+				regex: "^SlowBuffer$",
+			},
+		},
+	});
+
+	for (const directCall of directCalls) {
+		const args = directCall.field("arguments");
+		if (args) {
+			// Extract the arguments text (without parentheses)
+			const argsText = args.text().slice(1, -1); // Remove ( and )
+			edits.push(directCall.replace(`Buffer.allocUnsafeSlow(${argsText})`));
+			hasChanges = true;
+		}
+	}
+
+	// Handle direct function calls with aliases
+	for (const alias of slowBufferAliases) {
+		const aliasDirectCalls = rootNode.findAll({
+			rule: {
+				kind: "call_expression",
+				has: {
+					field: "function",
+					kind: "identifier",
+					regex: `^${alias}$`,
+				},
+			},
+		});
+
+		for (const directCall of aliasDirectCalls) {
+			const args = directCall.field("arguments");
+			if (args) {
+				const argsText = args.text().slice(1, -1);
+				edits.push(directCall.replace(`${alias}.allocUnsafeSlow(${argsText})`));
+				hasChanges = true;
+			}
+		}
+	}


This scenario can be simplified using utils/resolve-binding-path, and improving findAll from ast-grep with any and multiple matches.

Suggested change

const objectPatterns = rootNode.findAll({ rule: { kind: "object_pattern" } });

for (const pattern of objectPatterns) {

const text = pattern.text();

// Match patterns like { SlowBuffer: SB } to extract alias "SB"

const aliasMatches = text.matchAll(/SlowBuffer\s*:\s*(\w+)/g);

for (const match of aliasMatches) {

slowBufferAliases.add(match[1]);

}

}

// Find all named_imports that include SlowBuffer aliases (ESM imports)

const namedImports = rootNode.findAll({ rule: { kind: "named_imports" } });

for (const importNode of namedImports) {

const text = importNode.text();

// Match patterns like { SlowBuffer as SB } to extract alias "SB"

const aliasMatches = text.matchAll(/SlowBuffer\s+as\s+(\w+)/g);

for (const match of aliasMatches) {

slowBufferAliases.add(match[1]);

}

}

// Handle constructor patterns: new SlowBuffer(size) and new SB(size) for aliases

const constructorCalls = rootNode.findAll({

rule: {

kind: "new_expression",

has: {

field: "constructor",

kind: "identifier",

regex: "^SlowBuffer$",

},

},

});

for (const constructorCall of constructorCalls) {

const args = constructorCall.field("arguments");

if (args) {

// Extract the arguments text (without parentheses)

const argsText = args.text().slice(1, -1); // Remove ( and )

edits.push(constructorCall.replace(`Buffer.allocUnsafeSlow(${argsText})`));

hasChanges = true;

}

}

// Handle constructor calls with aliases

for (const alias of slowBufferAliases) {

const aliasConstructorCalls = rootNode.findAll({

rule: {

kind: "new_expression",

has: {

field: "constructor",

kind: "identifier",

regex: `^${alias}$`,

},

},

});

for (const constructorCall of aliasConstructorCalls) {

const args = constructorCall.field("arguments");

if (args) {

const argsText = args.text().slice(1, -1);

edits.push(constructorCall.replace(`${alias}.allocUnsafeSlow(${argsText})`));

hasChanges = true;

}

}

}

// Handle direct function calls: SlowBuffer(size)

const directCalls = rootNode.findAll({

rule: {

kind: "call_expression",

has: {

field: "function",

kind: "identifier",

regex: "^SlowBuffer$",

},

},

});

for (const directCall of directCalls) {

const args = directCall.field("arguments");

if (args) {

// Extract the arguments text (without parentheses)

const argsText = args.text().slice(1, -1); // Remove ( and )

edits.push(directCall.replace(`Buffer.allocUnsafeSlow(${argsText})`));

hasChanges = true;

}

}

// Handle direct function calls with aliases

for (const alias of slowBufferAliases) {

const aliasDirectCalls = rootNode.findAll({

rule: {

kind: "call_expression",

has: {

field: "function",

kind: "identifier",

regex: `^${alias}$`,

},

},

});

for (const directCall of aliasDirectCalls) {

const args = directCall.field("arguments");

if (args) {

const argsText = args.text().slice(1, -1);

edits.push(directCall.replace(`${alias}.allocUnsafeSlow(${argsText})`));

hasChanges = true;

}

}

}

var ImportNode // importNode can be fetch with getNodeImportStatement or getNodeRequireCalls

const binding = resolveBindPath(importN, "$.SlowBuffer") //it will resolve the name of the SlowBuffer in this file, independent of the way it was imported.

const aliasConstructorCalls = rootNode.findAll({

rule: {

any: [

{

kind: "new_expression",

pattern: `new ${binding}($$$ARGS)`,

},

{

kind: "call_expression",

pattern: `${binding}($$$ARGS)`,

},

],

},

});

for(const match of matches) {

if(match.kind() === 'new_expression') { // each scenario can be handled by kind here if needed

const args = match.getMultipleMatches() //get args

edits.push(match.replace(/* handle replace */)

}

}

brunocroh · 2025-08-31T19:44:10Z

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/src/workflow.ts

+	if (
+		originalText.includes("SlowBuffer") &&
+		!originalText.includes(",") &&
+		!originalText.includes(":") &&
+		!originalText.includes(" as ")
+	) {
+		return originalText.replace(/SlowBuffer/g, "Buffer");
+	}
+
+	let newText = originalText;
+
+	// Handle aliased patterns (both CommonJS : and ESM as syntax)
+	// { SlowBuffer: SB } -> { Buffer: SB }
+	newText = newText.replace(/SlowBuffer(\s*:\s*\w+)/g, "Buffer$1");
+	// { SlowBuffer as SB } -> { Buffer as SB }
+	newText = newText.replace(/SlowBuffer(\s+as\s+\w+)/g, "Buffer$1");
+
+	// If Buffer is already present in this specific pattern, just remove SlowBuffer
+	if (originalText.includes("Buffer") && originalText.includes("SlowBuffer")) {
+		// Remove non-aliased SlowBuffer references very carefully to preserve spacing
+		newText = newText
+			.replace(/,\s*SlowBuffer(?!\s*[:as])/g, "") // Remove SlowBuffer with leading comma
+			.replace(/SlowBuffer\s*,\s*/g, "") // Remove SlowBuffer with trailing comma and space
+			.replace(/SlowBuffer(?!\s*[:as])/g, ""); // Remove standalone SlowBuffer (not followed by : or as)
+
+		// Clean up any double spaces after opening brace
+		newText = newText.replace(/{\s{2,}/g, "{ ");
+		// Clean up any double commas but preserve spacing
+		newText = newText.replace(/,\s*,/g, ",");
+	}
+	// If Buffer is not present, replace first SlowBuffer with Buffer, remove others
+	else if (originalText.includes("SlowBuffer")) {
+		// Replace the first non-aliased SlowBuffer with Buffer
+		newText = newText.replace(/SlowBuffer(?!\s*[:as])/, "Buffer");
+		// Remove any remaining non-aliased SlowBuffer references
+		newText = newText
+			.replace(/,\s*SlowBuffer\s*(?![\s:as])/g, "")
+			.replace(/SlowBuffer\s*,\s*/g, "")
+			.replace(/SlowBuffer\s*(?![\s:as])/g, "");
+		// Clean up commas
+		newText = newText.replace(/,\s*,/g, ",");
+	}
+
+	return newText;


I think it can be handled with updateBinding utility function #182 , but it hasn’t been merged yet.

@AugustinMauroy Could we check with someone to follow up on the #182 merge?

@AugustinMauroy Could we check with someone to follow up on the #182 merge? yess

lluisemper · 2025-09-13T16:42:15Z

Hey guys! Based on your and suggestiongs I have pushed some changes. I have included some test for dynamic imports and I decided to remove the warning for the comments, with all the suggestions it was increasing the complexity and I was loosing track of what was what.

In addition to this I am not sure if updating this branch with a commit merge from main is fine or if you would prefer to rebase it. I will leave it down to you

AugustinMauroy · 2025-09-13T21:05:09Z

Hey guys! Based on your and suggestiongs I have pushed some changes. I have included some test for dynamic imports and I decided to remove the warning for the comments, with all the suggestions it was increasing the complexity and I was loosing track of what was what.

In addition to this I am not sure if updating this branch with a commit merge from main is fine or if you would prefer to rebase it. I will leave it down to you

It's look good I am still on holiday I am going to review it when I'll be back !

AugustinMauroy

LGMT !

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/src/workflow.ts

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/codemod.yaml

lluisemper · 2025-09-17T06:01:18Z

@AugustinMauroy @brunocroh Thank you very much for the reviews! I managed to learn a lot.
By the way, I have commited the change in which I was removing the hasChanged variables which were usseles. This has caused that approval is needed once again. However I will leave this in your hands.

AugustinMauroy · 2025-09-21T21:03:35Z

@AugustinMauroy @brunocroh Thank you very much for the reviews! I managed to learn a lot. By the way, I have commited the change in which I was removing the hasChanged variables which were usseles. This has caused that approval is needed once again. However I will leave this in your hands.

It's good now ! I'm just waiting the @JakobJingleheimer (actual formal maintainer) approval to merge and publish

JakobJingleheimer

This looks pretty good! A couple small things, and then I think this is good to go.

PS Sorry for the delayed review.

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/src/workflow.ts

JakobJingleheimer · 2025-09-22T07:06:32Z

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/src/workflow.ts

+		if (!nameField) continue;
+
+		// Handle direct assignment: const SlowBuffer = require('buffer').SlowBuffer
+		if (nameField.kind() === "identifier" && nameField.text() === "SlowBuffer") {


I would use a switch statement:

Suggested change

if (nameField.kind() === "identifier" && nameField.text() === "SlowBuffer") {

switch (nameField.kind()) {

case "identifier":

if (nameField.text() === "SlowBuffer") {

// …

case "object_pattern":

JakobJingleheimer · 2025-09-22T07:35:20Z

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/src/workflow.ts

+ * @param root - The AST root to search for require statements
+ * @param edits - Array to collect edit operations
+ */
+function processRequireStatements(root: SgRoot, edits: Edit[]): void {


This function and the other 2 are almost exactly the same. Could they be DRY'd up?

type StatementType = 'import-dynamic' | 'import-static' | 'require'; const nodeGetterMap = { 'import-dynamic': …, 'import-static': … 'require': getNodeRequireCalls, } as const satisfies Record<StatementType, Function>; function processStatements( root: …, edits: …, type: StatementType, ) { const statements = nodeGetterMap[type](root, 'buffer'); // I think you can keep the rest the same }

I think it makes the code less readable by adding this abstraction and incresses the complexity of the code base.
It is not like this is a system that will need to be scalable, it is just like a script that will run once.

I think that this change is not needed. Not sure why would you want to do this when it is as simple as it can get.

I can see arguments for both ways. My rationale is:

When consolidated, you're certain it's doing the same thing and there are no subtle differences (which a human may not even notice).

When duplicated, they can drift over time, very often in unimportant ways, like optional chaining gets added to one but not the other and then you're not sure which is correct or whether they're truly supposed to be different. That "not-sure-if" uncertainty makes the code untouchable.

Consolidated, there's no opportunity for that. So I tend to lean towards a slight increase in complexity to preclude code becoming untouchable.

both is right but I tend to follow Jacob POV

I have implement this, there were 2 functions that were nearly the same. Third one being 'import-static' was quite different.

JakobJingleheimer · 2025-09-22T07:43:52Z

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/tests/input/file-03.mjs

Could we get a case for aliasing the destructured require too? I believe under the hood it uses the same utility as dynamic import, which has a case, but good to have one for require too to ensure it doesn't get broken.

const { SlowBuffer: SB } = require('buffer'); const buf3 = new SB(256);

Thank you for spotting this. This must be covered. I will add this when I get a chance

While working on it I realized that this test was already covered here: recipes/slow-buffer-to-buffer-alloc-unsafe-slow/tests/expected/file-06.js

Add codemod to handle deprecated SlowBuffer API migration covering CommonJS, ES6 imports, and usage patterns Closes: nodejs#125 (comment)

lluisemper · 2025-09-25T11:36:17Z

@JakobJingleheimer I have implemented your small requested changes, let me know what do you think of them

JakobJingleheimer · 2025-09-25T20:00:14Z

I'm not sure how you were able to bypass the repo's protections and force-push, but that has broken GitHub's review system so I can't review your recent change in isolation—I'll have to do the whole thing over again, which will take a huge amount of time. I probably won't get to it before I'm away next week.

JakobJingleheimer · 2025-09-25T20:01:24Z

recipes/slow-buffer-to-buffer-alloc-unsafe-slow/src/workflow.ts

+	if (type === "import-static") {
+		// ESM imports have a different structure - handle them separately


@brunocroh don't we have a utility for this?

Yes, we have, but I can see one case here that is not covered by our utility. The new one added in #182 will probably simplify this code a lot. I’ll work on that now so we can use it here.

lluisemper · 2025-09-25T20:37:56Z

I'm not sure how you were able to bypass the repo's protections and force-push, but that has broken GitHub's review system so I can't review your recent change in isolation—I'll have to do the whole thing over again, which will take a huge amount of time. I probably won't get to it before I'm away next week.

Oh no, I am so sorry. The bypass of the repo's protection will be because I am in a fork, therefore the original repo setting do not affect it. Could it be a good idea to document this?

I was just trying to avoid a bloated history with many commits.

JakobJingleheimer · 2025-09-25T20:59:48Z

Hmm. The docs had previously said it would apply to fork branches (and we had previously tested it). I didn't document it because the error GitHub threw made it fairly obvious. But if the protection is now broken by design, I guess yeah good idea to document it 🫠

I was just trying to avoid a bloated history with many commits.

PRs are squash-merged, so only 1 commit per PR is added to the history.

AugustinMauroy · 2025-10-09T21:14:09Z

@JakobJingleheimer @brunocroh still blocked ?

JakobJingleheimer · 2025-10-09T21:38:32Z

still blocked ?

No, the upstream blocker is resolved, so this PR can now consume it per Bruno's #193 (comment)

lluisemper force-pushed the main branch 4 times, most recently from b04a60b to d2d54d6 Compare August 28, 2025 20:56

AugustinMauroy reviewed Aug 29, 2025

View reviewed changes

lluisemper force-pushed the main branch 3 times, most recently from a68a702 to 48798a1 Compare August 29, 2025 13:51

brunocroh reviewed Aug 31, 2025

View reviewed changes

lluisemper force-pushed the main branch from 48798a1 to 7c22e79 Compare September 13, 2025 16:35

lluisemper requested review from AugustinMauroy and brunocroh September 13, 2025 16:42

brunocroh approved these changes Sep 15, 2025

View reviewed changes

AugustinMauroy approved these changes Sep 16, 2025

View reviewed changes

lluisemper force-pushed the main branch from 7c22e79 to 944cbcd Compare September 17, 2025 05:58

avivkeller approved these changes Sep 21, 2025

View reviewed changes

JakobJingleheimer reviewed Sep 22, 2025

View reviewed changes

JakobJingleheimer added the awaiting author Reviewer has requested something from the author label Sep 23, 2025

feat: migrate SlowBuffer to Buffer.allocUnsafeSlow()

5034ce6

Add codemod to handle deprecated SlowBuffer API migration covering CommonJS, ES6 imports, and usage patterns Closes: nodejs#125 (comment)

lluisemper force-pushed the main branch from 944cbcd to 5034ce6 Compare September 25, 2025 11:33

lluisemper requested a review from JakobJingleheimer September 25, 2025 11:36

JakobJingleheimer removed the awaiting author Reviewer has requested something from the author label Sep 25, 2025

JakobJingleheimer added the awaiting reviewer Author has responded and needs action from the reviewer label Sep 25, 2025

JakobJingleheimer reviewed Sep 25, 2025

View reviewed changes

JakobJingleheimer added blocked:upstream Depends on another PR or external change/fix and removed awaiting reviewer Author has responded and needs action from the reviewer labels Sep 26, 2025

brunocroh mentioned this pull request Sep 28, 2025

feat(util/update-binding): add new utility to update binding in the file #182

Merged

JakobJingleheimer added awaiting author Reviewer has requested something from the author and removed blocked:upstream Depends on another PR or external change/fix labels Oct 9, 2025

-		if (nameField.kind() === "identifier" && nameField.text() === "SlowBuffer") {
+	switch (nameField.kind()) {
+		case "identifier":
+			if (nameField.text() === "SlowBuffer") {
+			// …
+		case "object_pattern":

		if (type === "import-static") {
		// ESM imports have a different structure - handle them separately

Uh oh!

feat: migrate SlowBuffer to Buffer.allocUnsafeSlow() #193

Are you sure you want to change the base?

feat: migrate SlowBuffer to Buffer.allocUnsafeSlow() #193

Uh oh!

Conversation

lluisemper commented Aug 28, 2025

Uh oh!

AugustinMauroy left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

AugustinMauroy commented Aug 29, 2025

Uh oh!

lluisemper commented Aug 29, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

AugustinMauroy commented Aug 29, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

lluisemper commented Sep 13, 2025

Uh oh!

AugustinMauroy commented Sep 13, 2025

Uh oh!

AugustinMauroy left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

lluisemper commented Sep 17, 2025

Uh oh!

AugustinMauroy commented Sep 21, 2025

Uh oh!

JakobJingleheimer left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

JakobJingleheimer Sep 22, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

lluisemper commented Aug 29, 2025 •

edited

Loading

JakobJingleheimer Sep 22, 2025 •

edited

Loading

JakobJingleheimer commented Sep 25, 2025 •

edited

Loading