I am using UV as my package manager. I want the structure of the repo to be as shown below:
./ (backend)
./src/ (frontend)
However, whenever I initialize the repo using UV, it automatically creates a ./src/project_name/ folder. How can I avoid that?
About a month ago, I made a post asking for resources on building a very small compiler/interpreter before jumping into something larger like Crafting Interpreters.
I decided to stop looking for the perfect resource and just start building the smallest thing I could understand end-to-end.
Today I got the first version of a simple arithmetic interpreter working in Python.
Right now it supports:
For example:
calc> 2 + 3 * 4
14
calc> (2 + 3) * 4
20
calc> -10 + 5
-5
The structure is currently:
Source text
↓
Lexer
↓
Tokens
↓
Recursive-descent parser
↓
Evaluation
↓
Result
The lexer converts something like:
2 + 3 * 4
into tokens roughly equivalent to:
NUMBER(2)
PLUS
NUMBER(3)
MUL
NUMBER(4)
The parser implements a small grammar along these lines:
expr → term (("+" | "-") term)*
term → factor (("*" | "/") factor)*
factor → NUMBER | "(" expr ")" | "-" factor
One of the most useful things I learned today was how operator precedence can naturally come from the structure of the grammar. I initially assumed I would need to assign explicit precedence values to operators, but with recursive descent, expr, term, and factor already encode that hierarchy.
The parser currently evaluates expressions directly rather than producing an AST, so it is deliberately still very small. My next major step will probably be separating parsing from evaluation by building an AST.
I also spent some time turning it into a proper little Python project instead of keeping everything in one file. It now has separate lexer, parser, interpreter, and CLI modules, a src package layout, pyproject.toml, a command-line entry point, and Ruff for linting/formatting.
So this is obviously nowhere near a real compiler yet, but that was exactly the point of my original post. I wanted something small enough that I could understand every stage instead of immediately disappearing into a much larger implementation.
Building even this tiny version made concepts like tokenization, grammars, recursive descent, precedence, and parsing much less abstract than they were a month ago.
The plan from here is to keep extending it incrementally, probably with an AST, variables, and a few statements before eventually moving toward bytecode or compilation.
Are there any FOSS CRM that goes beyond a simple Sales Pipeline?
I have been a programmer for a while now, and I have decent credentials. My only problem is that I suffer from Imposter Syndrome. I feel like I don't know anything at all, that I will fumble during interviews and will end up embarrassing myself.
Senior Developers, do you suffer from something similar? If yes, then how do you deal with it?
Dear Devs, as you can see I have successfully completed 2 tasks from the COD Point Rush event, however, the CP that I earned from those 2 tasks are not showing up on my profile. I cleared cache and delete the data but still no go. Please help.
I've started a side project called SearchCraft, where I'm building a search engine from scratch using Python.
The goal isn't to compete with Elasticsearch, Lucene, or any existing search engine. Those projects are incredible, but they're also so mature that it's easy to use them without ever understanding what's happening underneath.
So I decided to build one from first principles.
I'm implementing each component myself, starting with the basics: loading documents, tokenizing text, building an inverted index, and searching through it. As the project grows, I'll be adding things like posting lists, phrase search, ranking algorithms (TF-IDF/BM25), snippets, fuzzy search, and whatever else I can reasonably build along the way.
The primary reason for this project is to learn. I find that the best way to understand how a system works is to build a simplified version of it yourself.
I'm documenting the journey as I go, both for myself and in the hope that it might help someone else who's curious about how search engines work under the hood. There are plenty of tutorials on using search engines, but far fewer resources that walk through building one piece by piece.
It's still in its early stages, but I'm excited to see where it goes. Even if it never becomes production-ready, I'll come away with a much deeper understanding of one of the most fundamental pieces of modern software. After all, humans spend a good chunk of their lives typing words into little boxes and expecting magic to happen. Figuring out how that "magic" works seemed like a worthwhile weekend habit.
Here's the link to my project: https://github.com/rajtilakjee/searchcraft
I would be writing about it in my blog here: https://rajtilakjee.github.io/
I've started a side project called SearchCraft, where I'm building a search engine from scratch using Python.
The goal isn't to compete with Elasticsearch, Lucene, or any existing search engine. Those projects are incredible, but they're also so mature that it's easy to use them without ever understanding what's happening underneath.
So I decided to build one from first principles.
I'm implementing each component myself, starting with the basics: loading documents, tokenizing text, building an inverted index, and searching through it. As the project grows, I'll be adding things like posting lists, phrase search, ranking algorithms (TF-IDF/BM25), snippets, fuzzy search, and whatever else I can reasonably build along the way.
The primary reason for this project is to learn. I find that the best way to understand how a system works is to build a simplified version of it yourself.
I'm documenting the journey as I go, both for myself and in the hope that it might help someone else who's curious about how search engines work under the hood. There are plenty of tutorials on using search engines, but far fewer resources that walk through building one piece by piece.
It's still in its early stages, but I'm excited to see where it goes. Even if it never becomes production-ready, I'll come away with a much deeper understanding of one of the most fundamental pieces of modern software. After all, humans spend a good chunk of their lives typing words into little boxes and expecting magic to happen. Figuring out how that "magic" works seemed like a worthwhile weekend habit.
Here's the link to my project: https://github.com/rajtilakjee/searchcraft
I would be writing about it in my blog here: https://rajtilakjee.github.io/
Pretty sure you may have heard this question previously on this sub, however, I would urge you to read my complete question before brushing it off.
I want to create a simple compiler and by "simple compiler" I mean a single-pass compiler. I know about https://craftinginterpreters.com/ which is a wonderful resource. But I would like to start by creating something much smaller and simpler, and only then would I like to move on to something more complex like what Robert Nystrom created on his website.
Are there any similar resource that would teach me about single-pass compilers along with showing me how to create one? Any help in the right direction would be highly appreciated.
Pretty sure you may have heard this question previously on this sub, however, I would urge you to read my complete question before brushing it off.
I want to create a simple compiler and by "simple compiler" I mean a single-pass compiler. I know about https://craftinginterpreters.com/ which is a wonderful resource. But I would like to start by creating something much smaller and simpler, and only then would I like to move on to something more complex like what Robert Nystrom created on his website.
Are there any similar resource that would teach me about single-pass compilers along with showing me how to create one? Any help in the right direction would be highly appreciated.
Pretty sure you may have heard this question previously on this sub, however, I would urge you to read my complete question before brushing it off.
I want to create a simple compiler and by "simple compiler" I mean a single-pass compiler. I know about https://craftinginterpreters.com/ which is a wonderful resource. But I would like to start by creating something much smaller and simpler, and only then would I like to move on to something more complex like what Robert Nystrom created on his website.
Are there any similar resource that would teach me about single-pass compilers along with showing me how to create one? Any help in the right direction would be highly appreciated.
If you've ever heard "Famous Blue Raincoat" by Cohen, you'll know it has a certain atmospheric quality. Could you suggest some books that capture a similar mood?
During video interviews and presentations, I always had the same problem:
If I looked at my notes, it was obvious I wasn't looking at the camera.
So I built Kivo, a lightweight desktop teleprompter that sits just below your webcam, making it much easier to glance at your script while still appearing to maintain eye contact.
https://reddit.com/link/1uh1aj4/video/nacing4xht9h1/player
Current features:
It's still an MVP, but it's already been useful for:
The project is open source, and I'd love feedback or feature suggestions.
https://reddit.com/link/1uh18wl/video/qpiuj75kht9h1/player
During video interviews and presentations, I always had the same problem:
If I looked at my notes, it was obvious I wasn't looking at the camera.
So I built Kivo, a lightweight desktop teleprompter that sits just below your webcam, making it much easier to glance at your script while still appearing to maintain eye contact.
Current features:
It's still an MVP, but it's already been useful for:
The project is open source, and I'd love feedback or feature suggestions.
During video interviews and presentations, I always had the same problem:
If I looked at my notes, it was obvious I wasn't looking at the camera.
So I built Kivo, a lightweight desktop teleprompter that sits just below your webcam, making it much easier to glance at your script while still appearing to maintain eye contact.
Current features:
It's still an MVP, but it's already been useful for:
The project is open source, and I'd love feedback or feature suggestions.
If anyone needs a replacement Jamai, I will be happy to be of service.
I have seen on Instagram people annotating the books they are reading. I often wonder what do people annotate in the books they read.
