|
Automatic whitespace formatting step on commit.
|
#
?
Aug 22, 2017 17:42
|
|
|
|
| # ? May 10, 2024 01:59 |
|
|
Doom Mathematic posted:Automatic whitespace formatting step on commit. Counterpoint: Diffs from hell.
|
|
#
?
Aug 22, 2017 17:43
|
|
|
Pollyanna posted:Counterpoint: Diffs from hell. git diff -w hth
|
|
#
?
Aug 22, 2017 18:00
|
|
|
leper khan posted:git diff -w It doesn't get everything! Especially if you make some smaller changes alongside the whitespace that show up anyway.
|
|
#
?
Aug 22, 2017 18:03
|
|
|
|
|
#
?
Aug 22, 2017 18:05
|
|
|
aggressively LF and space every single file that touches your computer
|
|
#
?
Aug 22, 2017 18:13
|
|
|
Seriously, whitespace is a solved problem in diffing. Your tool should show a light background for the whitespace and a strong background for content changes. If this is an issue, fix your tools or get better ones.
|
|
#
?
Aug 22, 2017 19:18
|
|
|
Spaces because it's the goddamn Visual Studio default and setting up pre/postprocessing filters in Git/Bitbucket is annoying.
|
|
#
?
Aug 22, 2017 19:51
|
|
|
Tbh your editor/gofmt/etc should just do all of this transparently. I had to double check because I wasn't entirely certain whether or not I had actually been using tabs for the past year
|
|
#
?
Aug 22, 2017 20:32
|
|
|
Ok but 3 or 4 or 5 spaces?
|
|
#
?
Aug 22, 2017 21:54
|
|
|
Keetron posted:Ok but 3 or 4 or 5 spaces?
|
|
#
?
Aug 22, 2017 22:00
|
|
|
Keetron posted:Ok but 3 or 4 or 5 spaces? 1 HTML nbsp character and 1 ASCII space separated by a poop emoji. Finding an indent is just about counting the poops.
|
|
#
?
Aug 22, 2017 22:00
|
|
|
Keetron posted:Ok but 3 or 4 or 5 spaces? 2 spaces
|
|
#
?
Aug 23, 2017 00:31
|
|
|
leper khan posted:2 spaces maybe for yamls you heathen
|
|
#
?
Aug 23, 2017 00:34
|
|
|
Doom Mathematic posted:Automatic whitespace formatting step on commit. Python
|
|
#
?
Aug 23, 2017 01:18
|
|
|
Smugworth posted:maybe for yamls you heathen Or Scala, the best language
|
|
#
?
Aug 23, 2017 02:25
|
|
|
Clanpot Shake posted:Or Scala, the best language  maybe if you've only ever seen java
|
|
#
?
Aug 23, 2017 02:31
|
|
|
JewKiller 3000 posted:
https://vimeo.com/216330850 Set vim shiftwidth to 3 and softtabstop to 5 and call it a compromise
|
|
#
?
Aug 23, 2017 03:15
|
|
|
MisterZimbu posted:Spaces because it's the goddamn Visual Studio default and setting up pre/postprocessing filters in Git/Bitbucket is annoying. I set up a script a year or so ago that fails the build if anyone pushes code with tabs or mismatched line endings, and sends the submitter an email accusing them of tabcrimes. I wanted to accuse them of <i><b><blink>TABCRIMES!</blink></b></i> but apparently blink tags don't work in modern email readers and my knowledge of HTML ends somewhere around 1998. Sad. Anyhow, automated yelling was a lot easier to set up than mandatory pre-commit hooks for clang-format and about as effective (we do have .editorconfigs and .clang-formats in our repos so it's not like conformance is that hard). It's possibly also something for the horror thread, but it means I never have to see a dirty stinking tab fouling up my code so I'm happy.
|
|
#
?
Aug 23, 2017 08:17
|
|
|
Xerophyte posted:I set up a script a year or so ago that fails the build if anyone pushes code with tabs or mismatched line endings, and sends the submitter an email accusing them of tabcrimes. I wanted to accuse them of <i><b><blink>TABCRIMES!</blink></b></i> but apparently blink tags don't work in modern email readers and my knowledge of HTML ends somewhere around 1998. Sad. doing god's work, soldier 
|
|
#
?
Aug 23, 2017 13:12
|
|
|
Xerophyte posted:I set up a script a year or so ago that fails the build if anyone pushes code with tabs or mismatched line endings, and sends the submitter an email accusing them of tabcrimes. I wanted to accuse them of <i><b><blink>TABCRIMES!</blink></b></i> but apparently blink tags don't work in modern email readers and my knowledge of HTML ends somewhere around 1998. Sad. I think you can do gifs now so you know what to do 
|
|
#
?
Aug 23, 2017 15:12
|
|
|
love2get a ticket to copy work from a half-finished branch that bugs the gently caress out on a new piece of tech and have to wait 3~4 days to pull lovely vague replication knowledge from my team lead, love love love it
|
|
#
?
Aug 24, 2017 14:49
|
|
|
Pollyanna posted:love2get a ticket to copy work from a half-finished branch that bugs the gently caress out on a new piece of tech and have to wait 3~4 days to pull lovely vague replication knowledge from my team lead, love love love it Your team lead sounds bad and you should make a list of all the ways they are for your exit interview
|
|
#
?
Aug 24, 2017 14:55
|
|
|
Munkeymon posted:Your team lead sounds bad and you should make a list of all the ways they are for your exit interview The grand majority of my time is spent trying to implement his stuff, bashing my head on my desk when it inevitably fails somewhere, then waiting on a response from him for a long time to explain to me WTF. He never automates, simplifies, or documents what he does unless specifically asked to do it, and it really slows us all down. He goes cowboy on everything and really isn't much of a technical player at all. Our new team member is now realizing how poo poo he is at his job and is very unhappy. I'll definitely be bringing it up in the exit interview.
|
|
#
?
Aug 24, 2017 15:23
|
|
|
The early return argument is one that most people don't have the full context on. There's two original arguments. The first is that code should have one entry and one exit destination (except for error handling) and it's relatively uncontroversial. You enter at the top of a function, and you exit to the location that called you. Most modern languages only provide exceptions as an alternative to this (with an exception your exit destination is the first catch that can handle the exception). The other argument was for 100% structured languages like BASIC (IIRC). This style considers stuff like breaks to be the same as early returns. People stopped doing this as well because it's pretty easy to prove that it's objectively bad but people still cling to the early return (but not break/continue) part of it for some reason.
|
|
#
?
Aug 24, 2017 17:12
|
|
|
Pollyanna posted:I think you can do gifs now so you know what to do You can serve a video on an iPhone. Give 'em hell soldier.
|
|
#
?
Aug 25, 2017 20:59
|
|
|
Alright, question, because all my searching has found nothing. Is there any tool out there to automatically extract diffs from a git commit and allow you to comment on them? Right now one guy on my team is basically doing this by hand, and I could have sworn I'd seen an utility like this before.
|
|
#
?
Aug 31, 2017 19:45
|
|
|
TheCog posted:Alright, question, because all my searching has found nothing. Is there any tool out there to automatically extract diffs from a git commit and allow you to comment on them? Right now one guy on my team is basically doing this by hand, and I could have sworn I'd seen an utility like this before. We use BitBucket (formally known as Stash) and it has this feature where we can comment line by line on a PR. Not sure about commits but who cares about commits until the PR is made (when it needs to be squashed anyway).
|
|
#
?
Aug 31, 2017 20:13
|
|
|
TheCog posted:Alright, question, because all my searching has found nothing. Is there any tool out there to automatically extract diffs from a git commit and allow you to comment on them? Right now one guy on my team is basically doing this by hand, and I could have sworn I'd seen an utility like this before.
|
|
#
?
Aug 31, 2017 20:19
|
|
|
Vulture Culture posted:Are you asking about code review tools? GitHub/GitLab/BitBucket pull requests, Gerrit, and Crucible are all popular, but there's a bunch of other niche tools too if you go hunting for them. Yeah, these are pretty standard nowadays. I understand not everyone can use the GitHub service, but they do have https://enterprise.github.com/home
|
|
#
?
Aug 31, 2017 23:52
|
|
|
We use Phabricator's differential tool, and I really like it.
|
|
#
?
Sep 1, 2017 00:21
|
|
|
We use gitlab internally, it does a pretty good job of trying to be github.
|
|
#
?
Sep 1, 2017 14:15
|
|
|
captkirk posted:We use gitlab internally, it does a pretty good job of trying to be github. It's better imo. The CI support is really nice, and the doxygen plugins are pgood as well.
|
|
#
?
Sep 1, 2017 14:22
|
|
|
Overslept, rushed to the office, got in at 9:45AM, am somehow still the first person on my team to be in the office today. v  v
|
|
#
?
Sep 5, 2017 14:50
|
|
|
Pollyanna posted:Overslept, rushed to the office, got in at 9:45AM, am somehow still the first person on my team to be in the office today. You can leave early without anyone knowing!
|
|
#
?
Sep 5, 2017 15:02
|
|
|
Pollyanna posted:Overslept, rushed to the office, got in at 9:45AM, am somehow still the first person on my team to be in the office today. Why are you trying to ruin it for your team??
|
|
#
?
Sep 5, 2017 15:05
|
|
|
Keetron posted:You can leave early without anyone knowing! "Oh yeah, I got in at 7:00 this morning, so I'll be taking off at 3:15."
|
|
#
?
Sep 5, 2017 15:06
|
|
|
lmbo, I get in at 5 and leave at 1. I only have a 3-hour window when people can schedule meetings with me, and I am the lead architect of our main project right now at work.
|
|
#
?
Sep 5, 2017 15:08
|
|
|
Anticipating the follow-up post where Pollyana's entire team was laid off...
|
|
#
?
Sep 5, 2017 15:08
|
|
|
|
| # ? May 10, 2024 01:59 |
|
|
ratbert90 posted:I only have a 3-hour window when people can schedule meetings with me, Living the dream.
|
|
#
?
Sep 5, 2017 15:10
|
|