Identical consecutive commit messages from the same user

I was fortunate enough to prepare some SCM recommendations for our Subversion users and came across a point of disagreement with the team. Is there ever a valid use case for someone committing consecutive commits with the same message?

If you take an approach in which the commit message should describe the “what” and “why” of the code, it is hard to see the actual case for this. Looking at our story, the instances when this happened are apparently more convenient than anything else and really don't tell the story of what the code does.

Does anyone have any views on the legitimacy of this? Can recommendations (or even intercept hooks) be overly diligent, or is that a reasonable expectation?

Edit: allows you to work with the assumption that people are already leaving good commit messages. IMHO, a single word, such as "updated" or "typo", is not a satisfactory commit message. I expected to see something more, like "Updated color of the submit button on green" or "Typo in the training copy fixed." It is very difficult to simply look at the repository log and understand what is going on in the project without drilling individual messages if this message is a word or two.

+3
source share
6

. - . , , , . , , , , .

+1

, , . , , , .

, , :

typo

. . .

+6

, , diff. .

+3

... .

, : , , , " "

+2

, ( ). , , , .

+1
source

I don’t understand why this is bad ... you cannot describe what you did every time ... or else you will spend most of your time on a lot of words that are practically not.

Examples like:

commented here and there
typo
tidyed something ...

(examples from the top of my head ...)

So, for example, "typo" means that if I want to write good commit messages, I have to spend 10 characters describing the change in one letter. Bureocracy at its best ...

0
source

Source: https://habr.com/ru/post/1720892/


All Articles