标签：#apt

Background Take Lombok as an example: at compile time, it reads annotations like @Getter and @Setter through an Annotation Processor (APT, Annotation Processing Tool), then directly modifies the AST (Abstract Syntax Tree) in memory, injecting getter/setter methods for fields. Developers only need to annotate fields, and the compiled .class files automatically include these methods. Advantages of APT APT is an extension mechanism natively supported by the Java compiler. Annotation processors run at compile time and can read annotation information from source code to generate new source files. Its advantages are: Standardized — Simply implement the Processor interface; no need to hack the compiler Composable — Multiple annotation processors can work together in the same compilation process Generated code is visible — Source files generated through the Filer API are located in the build/generated/ directory, and IDEs can navigate to them after configuring the source path Disadvantages of APT APT has a fundamental limitation: it cannot generate a class with the same fully qualified name as the source code. The Java compiler explicitly specifies that two classes with the same fully qualified name are not allowed in the same compilation unit. Source code generated by APT participates in the same round of compilation as the original source code, so it can never generate a class with the same fully qualified name as the original class. This means APT cannot truly "modify" a class; it can only generate new classes (such as Builder, Factory, etc.). Advantages of AST Tools like Lombok bypass the APT limitation by directly modifying the AST in the compiler's memory — injecting methods for fields, injecting constructors for classes, etc. The advantages of AST modification are: Can modify existing classes — Methods are added directly to the AST in memory, breaking through the limitation that APT cannot generate Same-Name Classes Zero intrusion — Classes written by developers rema

Last year, Microsoft's 13-inch Surface Laptop quickly became one of my favorite thin-and-light Windows notebooks. At $900, it was easy to recommend to anyone wanting MacBook Air-like build quality and battery life on Windows - I even convinced my sister to buy one on sale. But that was last year. This year, thanks to RAMageddon, […]

With storage and memory prices still at an all-time high, we’re happy to tell you about a solid deal on a good laptop when we find one, rare as they are. Best Buy is selling the HP OmniBook X Flip 2 in 1 for $999.99 (usually $1,649.99) with an Intel Core 7 Ultra CPU, 16GB […]

The macOS 27 Golden Gate public beta is here, and anyone with an M-series Mac now has easier access to test-drive Apple's latest changes - including a more subdued Liquid Glass aesthetic. That's reason enough to be at least a little excited for macOS 27 (particularly if you're on Tahoe and disliking all the transparency). […]

You set up Puppeteer, navigate to a page, call page.screenshot() , and the bottom half of your image is blank placeholder boxes. Welcome to lazy loading. Most modern sites defer images and heavy content until the user scrolls. Your headless browser never scrolls. So those elements never load. Here's how to deal with it. The scroll trick The most common fix is to programmatically scroll down the page before taking the screenshot: async function scrollToBottom ( page ) { await page . evaluate ( async () => { const delay = ms => new Promise ( r => setTimeout ( r , ms )); const distance = 300 ; while ( window . scrollY + window . innerHeight < document . body . scrollHeight ) { window . scrollBy ( 0 , distance ); await delay ( 150 ); } window . scrollTo ( 0 , 0 ); }); } await page . goto ( " https://example.com " , { waitUntil : " networkidle2 " }); await scrollToBottom ( page ); await page . waitForTimeout ( 1000 ); await page . screenshot ({ fullPage : true }); The 150ms delay between scrolls gives IntersectionObserver -based lazy loaders time to trigger. Too fast and you'll scroll past elements before they start loading. That final waitForTimeout after scrolling back to top lets any remaining images finish rendering. Not elegant, but necessary. Why networkidle2 isn't enough You'd think waitUntil: "networkidle2" would handle this. It waits until there are no more than 2 network connections for 500ms. But lazy-loaded images haven't even been requested yet at that point — they're waiting for a scroll event that never happens. networkidle2 only helps with content that loads on page init. For scroll-triggered content, you need the scroll. The loading="eager" override Some sites use the native loading="lazy" attribute. You can override it before images load: await page . evaluateOnNewDocument (() => { Object . defineProperty ( HTMLImageElement . prototype , " loading " , { set : function ( val ) { this . setAttribute ( " loading " , " eager " ); }, get : function () { retu

While fully loaded models are quite pricey, the HP OmniBook Ultra is hard to beat for those in the market for a premium Windows ultraportable.

The Dell 14S represents the new normal of laptop pricing, but it has the quality to back up its cost.

CAP: khi network partition xảy ra, chỉ được chọn C hoặc A — không có "cả ba" CAP theorem là kết quả của Gilbert và Lynch (formal proof năm 2002 cho conjecture Brewer đưa ra ở PODC keynote 2000): một hệ phân tán có shared state không thể đồng thời cung cấp cả linearizable Consistency , Availability (mọi request tới non-failing node đều trả lời không lỗi), và Partition tolerance khi có network partition. Trong thực tế, partition là thứ sẽ xảy ra — TCP retransmit, GC pause dài, switch chết, cross-region link flap — nên P là ràng buộc bắt buộc, không phải lựa chọn. Câu hỏi thật là: khi partition xảy ra, hệ thống hy sinh C hay A? Chọn sai gây ra hai loại incident khác nhau: chọn AP mà dữ liệu cần linearizable dẫn tới double-charge, oversell inventory, split-brain; chọn CP mà dữ liệu chỉ cần eventually consistent dẫn tới downtime không cần thiết, user không đọc được profile của chính mình. Cơ chế hoạt động Định nghĩa formal theo Gilbert và Lynch: Consistency ở đây là linearizability : mọi read sau một write hoàn tất phải thấy giá trị mới (hoặc mới hơn); tồn tại một total order các operation phù hợp với real-time. Availability : mọi request tới một non-failing node phải nhận response (không timeout, không error). Partition tolerance : hệ thống tiếp tục hoạt động dù network drop tuỳ ý message giữa các node. Proof intuition: giả sử có 2 node N1, N2 giữ cùng key x=0 . Client ghi x=1 vào N1. Link N1 và N2 đứt. Một client khác đọc x từ N2. Nếu N2 trả về 0 thì không linearizable (mất C). Nếu N2 chờ đến khi thấy được N1 thì mất A. Nếu N2 từ chối phục vụ thì cũng mất A. Không có cách thứ ba. Trong hệ CP, mỗi write phải qua quorum (Raft, Paxos, ZAB); khi node bị isolate khỏi quorum, nó từ chối phục vụ để giữ linearizability: // etcd/Raft-style: khi mất quorum, leader step down và write fail resp , err := kv . Put ( ctx , "order/42" , "paid" ) if err != nil { // err là ErrLeaderChanged hoặc context.DeadlineExceeded khi ở minority side // client thấy unavailable — đúng contract CP re

Laptop chargers are much easier to travel with these days, and there's one specific technology that is largely to thank.

Next spring could be a fiesta of new iPads, MacBooks and iPhones.

Apple is working on a "revamped" version of its entry-level MacBook Pro that it could launch as soon as the first half of 2027, Bloomberg reports. The company is also testing four new iPad Pros that are set to launch in the spring with a focus on "internal improvements." The updated MacBook Pro, which will […]

Why your Cloudflare Turnstile token works in the browser but 403s from requests You solved the Turnstile widget. You can see the token in the page. You copy it into your script, POST the form from requests, and the server hands you back a 403 — or a JSON body with "success": false. The token clearly worked a second ago in the browser, so what changed? Short answer: a Turnstile token is not a password you can carry around. It's a one-time, short-lived proof bound to a very specific context, and replaying it from a different context is exactly what it's designed to reject. Below is what that context is, how to tell which constraint you're hitting, and the fix for each. The real scenario You're automating a flow on a Cloudflare-protected site. There's a cf-turnstile widget on the form. You get a token one of two ways: you render the page in a real browser (Playwright/Selenium) and read cf-turnstile-response, or you hand the sitekey + page URL to a solving service and get a token back. Either way, you then submit the form with a plain HTTP client requests, httpx, axios) and it fails. The frustrating part: it's intermittent-looking. The reason it feels random is that there are four separate constraints, and you're usually tripping a different one each time. The four things a Turnstile token is bound to 1. It's single-use Once Cloudflare validates a token server-side (the siteverify call your target makes), that token is spent. Submit twice, retry, or test it once by hand, and the second use returns false. You get a fresh one per submission. 2. It has a short TTL Turnstile tokens expire fast — a few minutes. Solve early, do other work, submit later, and the token can be dead on arrival. The widget auto-refreshes in the browser precisely because tokens go stale; a script that grabs the token and sits on it loses that refresh. 3. It's bound to the sitekey and the page URL Multiple widgets. Some pages embed more than one Turnstile (login + newsletter). Solving the wrong site

Thanks to the component crisis, it's a bad time to want a new computer. But if you are waiting on a preorder for the Framework Laptop 13 Pro - which Framework's CEO has called the "MacBook Pro for Linux users" - the company shared good news on Thursday that might mean yours will cost less […]

RAMageddon has come for computers. The price of memory chips, hard drives, and solid state storage has skyrocketed. That’s led to price increases on desktop and laptop RAM, SSDs, spinning hard drives, and pretty much everything that uses any of those things. Consoles are more expensive. Desktops are more expensive. Laptops are more expensive. Tablets […]

Apple just raised the prices on Macs and iPads in response to the rising costs of memory chips, right in the middle of Amazon Prime Day. That means existing discounts (even small ones) on Apple laptops like the MacBook Neo, MacBook Air, and MacBook Pro just became significantly better deals. As an example, the 13-inch […]

Microsoft just added a cheaper 12-inch Surface Pro and 13-inch Surface Laptop to its lineup. Both models come equipped with 8GB of RAM instead of 16GB, costing $849 for the specced-down Surface Pro and $949 for the Surface Laptop, as spotted earlier by Windows Central. When the 12-inch Surface Pro and 13-inch Surface Laptop launched […]

The Surface Laptop is down to $835 for Prime Day—a killer discount on one of my favorite laptops.

The RAM, storage, and component crunch has come for new laptop prices. Plenty of models have gone up in price after being on the market for months, and so have those that’ve been on shelves for over a year. Plus, many new models are releasing at higher prices than the previous generation. So, it really […]

HP’s OmniBook Ultra 14 is one of the best alternatives to the MacBook Air you can buy right now, and it’s more affordable than many of its Windows competitors.

I’ve been reviewing laptops for over a decade, and this is my advice on how to find the right laptop for you.