svn - What's the meaning of git's snapshot of a file? -

i'm reading git basics

git thinks of data more set of snapshots of miniature filesystem

i not understanding meaning of snapshot of git. git store entire file content in each snapshot/version? example, version 1

#include <stdio.h> int main() {         printf("hello, world");         return 0; } 

in version 2 added line file.

#include <stdio.h> int main() {         printf("hello, world");         printf("hello, git");         return 0; } 

will git store entire content rather store difference(printf("hello, git")) between these 2 versions svn etc?

if is, what's point?

will git store entire content rather store difference(printf("hello, git")) between these 2 versions svn etc?

git stores the entire contents of file. takes no space when file didn't change.

read brilliant answer git pack file format: are git's pack files deltas rather snapshots?

about sha1

files (and other stuff) stored in form of "blob". each sequence of bytes has own sha1-code, pretty unique it.

the following true sha1:

1. sha1 calculation file gives same result @ time, os, git version or implementation.
2. files different names or paths equal contents have equal sha1-s.
3. if 2 files have different sha1-s, not equal probability of 1.
4. if 2 files have equal sha1-s, equal probability of around 1 - 1 / 2⁴⁰⁰ (as remember) pretty 1.

what benefits system gives

1. revisions can compared equality quick. no file contents checked, sha1-s.
   • when push/pull, changed files transmitted.
   • checking status of current changes done in moment.
   • lets track n files equal contents, taking place of single file in git.
2. changing revision in working tree quick.
   • without applying consecutive patches
   • you can exclude commits branch, pull them branch, change order.

about diff (and git diff):

you may have noticed git indeed shows diff of text files, pointing out added , removed lines. done diff utility convenience. helps collect contribution statistics. , used resolving merge conflicts. nevertheless git treats , stores text (and binary) files single blobs.

exclusion git add --patch

there way force git break text files chunks when staging changes. may useful large files, pretty useless small ones.

git add --patch  

interactively choose hunks of patch between index , work tree , add them index. gives user chance review difference before adding modified contents index.

these favourite illustrations git pro git: