aayushprime

Where are my bits getting lost?

Today I was trying to implement the Snowflake Id generator for Fly.io Gossip Glomers second challenge, Unique Id generation. Here’s what I came up with. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 package main import ( "encoding/json" "log" "strconv" "time" maelstrom "github.com/jepsen-io/maelstrom/demo/go" ) type Snowflake struct { sequenceNumber uint64 selfId int lastTs int64 } func nextId(s *Snowflake) uint64 { ts := time.Now().UnixMilli() // this is 41 bits if s.lastTs == ts { s.sequenceNumber++ } else { s.sequenceNumber = 0 } s.lastTs = ts var id uint64 = uint64(ts << 22) // (64 - 1 sign bit - 41 unix timestamp bits) log.Printf("shifted %d", id) id = id | ((uint64(s.selfId) & 0x3f) << 16) // machine id is 10 bits log.Printf("after self Id %d", id) id = id | ((s.sequenceNumber & 0xffff) ) return id } func main() { n := maelstrom.NewNode() sf := &Snowflake{} n.Handle("generate", func(msg maelstrom.Message) error { var body map[string]any if err := json.Unmarshal(msg.Body, &body); err != nil { return err } selfId, err := strconv.Atoi(msg.Dest[1:]) if err != nil { log.Fatal("Cannot parse destination as a number", msg.Dest) } sf.selfId = selfId body["type"] = "generate_ok" body["id"] = nextId(sf) return n.Reply(msg, body) }) if err := n.Run(); err != nil { log.Fatal(err) } } It wasn’t too bad for how simple its supposed to be. My solution wasn’t getting accepted by maelstrom because there were duplicate Ids generated by my program. After looking at my code for very long and not finding any idea what went wrong, I asked ChatGPT, it suggested the sequenceNumber bits were too few and must probably be overflowing. After looking at the maelstrom logs I found the lines which had duplicate Ids. ...

Python: Solving slow eval

Welcome to another one of my interesting findings, where I explore some stuff for fun and profit just fun A while back I was solving AOC and for a problem (day 13), I had this sort of clever idea of using eval to skip writing a if else ladder. The input file had the following structure Starting items: 91, 65 Operation: new = old * 13 Test: divisible by 5 If true: throw to monkey 7 If false: throw to monkey 4 I had to parse the file to get all values, but when parsing the second line I thought why not use the python interpretor itself. And so, I created a variable called old and whenever it needed updating I would use the following line ...

Why does my drive mount as read only?

Today I tried to write an fstab entry so that my drive would automount on boot. It went as smoothly as I had hoped (at first). So, I ran sudo blkid to find the UUID of the drive I wanted to mount and added the entry UUID=288C1E6C8C1E3532 /hdrive auto defaults 0 0 following the steps in this wonderful article. (Some options are different after I realized what I wanted.) Article Then I ran sudo mount -a to check if the mounts were working properly. But here the problem began. The drive was mounted as readonly. There is no mention of readonly mount in the options. In fact I learnt that defaults option is shorthand for rw (read write) and bunch of other options. Then I tried various mount options and got sick of rebooting. I tried directly mounting as ntfs instead of auto. But it didn’t work. After lot of struggle I figured out the problem. It was because of my Windows Installation! I have dual boot setup. And turns out windows stores some files for faster booting (fast boot) in the harddisk thus linux cannot mount the drive cleanly and can only mount it as a read only drive. So to fix this I had to boot to windows and change the fast boot setting. Here’s how, ...

Google Sheets as Database?

I recently learned that Google Sheets can be used as a database (although with caveats I hear). And I love it. It is so much easier to see what is in the spreadsheet than on a sqlite file. Plus, google sheets works everywhere! So, to get started, the official way to do this is from the google developers website. https://developers.google.com/sheets/api/guides/concepts It is not so easy to understand for a beginner(which I am). So, upon further looking around I came across this wonderful python library that makes things so much easier. It’s gspread. ...

String Bug

Today the coding calendar Advent of Code started. You should try it too, Advent of Code. I tried to solve the first challenge. It was pretty straight forward. Except, I was stumped by a simple problem. Here is my first attempt, 1 2 3 4 5 6 7 8 9 10 11 12 def part1(lines): prev = lines[0] more = 0 for i in range(len(lines) - 1): if lines[i+1] > prev: more += 1 prev = lines[i+1] print(more) if __name__ =='__main__': with open('./input/day1.txt', 'r') as f: lines = [a.strip() for a in f.readlines()] part1(lines) The thing is it gives correct result for the test data. So, I assumed it would work for the input data. But, the program always gave me 1 less than the actual answer. I couldn’t figure out why. Notice, I use string comparison in the if condition on line 5. I knew using string for comparison of numbers was a bad idea but, the following IDLE experiments convinced me otherwise. ...

First

Welcome! This is the first post of my blog! What is this about? In an attempt to make myself “productive”, I try to write about anything I find interesting. Tell me something interesting at [email protected]