-
cargo build --release
-
set
strip = true
to profile.release in cargo.toml -
set
opt-level = "s"
to optimize for size instead of runtime speed -
set
lto = true
to enable link-time optimization -
set
codegen-units = 1
to maximize size reduction but it results in slow builds -
use cargo bloat to understand which dependencies are taking up space
A hacky way to store data as video. The proof of concept uses youtube.
a new #project #idea is to build the whole thing, but not only for youtube, but for any possible API endpoint :D
Here is separation of the responsibility of such a program:
- the image processing layer takes the file and converts it into an image or video
- the compression layer reduces the size of the data as much as possible
- the encryption layer guarantees the confidentiality of the data until the algorithm is broken, since the data is shared with a third party anyway. It can intervene
- the network layer allows data to be uploaded and downloaded to different providers (why limit it to youtube?)
- the data layer: it is responsible for storing information about the videos/images (video metadata: at provider X, accessible from such and such a URL, last modification date, etc.). It inserts metadata into the files through the image processing layer in order to decode them correctly (type of compression, etc.).
- the application layer: it allows to use everything with a UI/CLI. In short, an interface for the user.
Kafka has a good throughput with sequential Writes and Reads.
Kafka can move a lot of data because of the zero copy read principle:
Before:
- Disc to OS buffer
- Write the content of the OS buffer to the RAM
- Copy the data to the application Buffer
- Copy the data back to the socket buffer
- Copy the data from the socket to the Network Interface Chip buffer and send it
With zero-copy read principle:
- Read from the disc and load it into the OS Buffer
- Directly copy to the NIC Buffer (the CPU is not involved)
Transition a CSS grid row from 0
to 1fr
to get an accordion working :D
An Advantage is the text is already rendered for screen readers.
Afin de favoriser la sobriété:
- Répondre à nos besoins fondamentaux
- l'autonomie
- la compétence
- l'appartenance sociale
- Développer les nudges (coups de pouce)
with material: https://web.stanford.edu/class/cs253/
A playlist of rust video about rust.
🎯
Tout abus d'@YLeBolloch peut nuire gravement à la santé d'une machine à propagande, même bien huilée.
On remarque les journalistes qui veulent tous le temps pousser à la phrase de trop au lieu d'avoir une clarté dans le discors. L'interviewé impose son rythme, c'est remarquable.
If you have some notes, you can have a melody
La prouesse technique d'avoir du Wifi à bord des TGV, et pourquoi c'est si compliqué !
Good points! I have use cases where a frontend database approach will work perfectly!
Numbers:
const formatter = Intl.NumberFormat(LOCALE, { notation: 'compact' })
Currency:
Intl.NumberFormat(LOCALE, {
notation: 'compact',
style: 'currency',
currency: 'ZWD'
})
- Extraction: make the nested code a function or its own piece of logic.
- Inversion: use guards to remove nested code. These guards will stop the function before the end of the function body. They act as gates.
La vidéo appuyé par Elon Musk n'a donc pas été tournée proche d'un aéroport, comme il le prétend
TL;DR $attrs defines every attribute that is not listed in the props.
Use cases:
- apply $attrs on a specific node
- tell Vue to apply $attrs on a node of a multiple root-node components
- using prop validations (in dev only for example)
Ils sont magnifiques et c'est si beau
Ducks and drum kit
Macron disant que c'est hypocrite de décaler l'âge de la retrait. C'était il y a peu années.