Coding Horrors: You can gather all your technical debt into one easy framework!

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Cavern of COBOL > Coding Horrors: You can gather all your technical debt into one easy framework!

«‹›1503 »

the littlest prince: Sep 23, 2006

SupSuper posted:

Did someone say use cases?

Somebody please upload that last one to imgur and post it here; it is too good to miss. I would do it but my work proxy keeps uploads from working.

# ? Nov 6, 2013 16:48

Adbot: ADBOT LOVES YOU

# ? May 16, 2024 05:22

DaTroof: Nov 16, 2000; CC LIMERICK CONTEST GRAND CHAMPION
There once was a poster named Troof
Who was getting quite long in the toof

the littlest prince posted:

Somebody please upload that last one to imgur and post it here; it is too good to miss. I would do it but my work proxy keeps uploads from working.

If the user gets all the way to View Travel Alerts, he turns into a king.

# ? Nov 6, 2013 17:02

JawnV6: Jul 4, 2004; So hot ...

Thank god for the little bumps in the lines. Can't imagine how you'd trace them if they just went point to point.

# ? Nov 6, 2013 17:06

1337JiveTurkey: Feb 17, 2005

Vasja posted:

Fear me, for I am Lord Use Case

Seriously though, what the hell. I think at a certain point you have to go 'these guys just look like they are firing lasers at each other, maybe this isn't a useful diagram any more.'

First thing that came to mind when seeing that.

Only registered members can see post attachments!

# ? Nov 6, 2013 18:29

Ornedan: Nov 4, 2009; Cybernetic Crumb

What's up with the faint (mirrored?) diagrams in the background of those use case images?

# ? Nov 6, 2013 18:47

pokeyman: Nov 26, 2006; That elephant ate my entire platoon.

Ornedan posted:

What's up with the faint (mirrored?) diagrams in the background of those use case images?

Scan of a double-sided printout.

# ? Nov 6, 2013 18:51

Suspicious Dish: Sep 24, 2011; 2020 is the year of linux on the desktop, bro; Fun Shoe

1337JiveTurkey posted:

First thing that came to mind when seeing that.

There's an iOS remake of FF6 now?

# ? Nov 6, 2013 18:55

shodanjr_gr: Nov 20, 2007

From some C++ code I got from a guy at the lab today:

code:

// Destructors
~ClassDefinedAndImplementedInFourHundredLineHeaderFile();
// End of Destructors

# ? Nov 6, 2013 23:44

shrughes: Oct 11, 2008; (call/cc call/cc)

He'll get the last laugh when C++23 allows passing parameters to destructors!

# ? Nov 6, 2013 23:52

Drastic Actions: Apr 7, 2009; FUCK YOU!
GET PUMPED!; Nap Ghost

code:

if isNumeric(objConfig.getItem(30)) then bSumRequired = (objConfig.getItem(30) = 1) end if
		if isNumeric(objConfig.getItem(31)) then bStartStopReq = (objConfig.getItem(31) = 1) end if
		if isNumeric(objConfig.getItem(34)) then bShowServiceDesk = cbool(objConfig.getItem(34)) end if
		if isNumeric(objConfig.getItem(44)) then bAllowFlexTime = (objConfig.getItem(44) = 1) end if
		if isNumeric(objConfig.getItem(55)) then bProhibitTimeEntry = (objConfig.getItem(55) = 1) end if
		if isNumeric(objConfig.getItem(80)) then bReduceResouceBillableHoursGoal = (objConfig.getItem(80) = 1) end if
		if isNumeric(objConfig.getItem(33)) then iRoundingMins = (objConfig.getItem(33)) end if
		if isNumeric(objConfig.getItem(61)) then DontRoundWorked = (objConfig.getItem(61)) end if
		if len(DontRoundWorked) < 1 then DontRoundWorked = 0 end if
		if CStr(DontRoundWorked) = "1" then DontRoundWorked = "true" else DontRoundWorked = "false"

This is ASP code used on a timesheet page.
What's getItem(30)? No idea. But I do know that bSumRequired is actually checking if the "summary" is required for a time entry. Why shorten "Summary" to a word that already exists that has nothing to do with it?

# ? Nov 7, 2013 15:04

Polio Vax Scene: Apr 5, 2009

Obviously it's getting the 30th input field on the form. So uh...don't add/remove any before that.

# ? Nov 7, 2013 18:25

raminasi: Jan 25, 2005; a last drink with no ice

"Remote to Local caster function. This stuff is complex, don't concern yourself with casting logic." is not acceptable documentation for a function :mad:

# ? Nov 8, 2013 19:40

KaneTW: Dec 2, 2011

GrumpyDoctor posted:

"Remote to Local caster function. This stuff is complex, don't concern yourself with casting logic." is not acceptable documentation for a function

You're developing teleportation spells?!

# ? Nov 8, 2013 20:24

Blotto Skorzany: Nov 7, 2008; He's a PSoC, loose and runnin'
came the whisper from each lip
And he's here to do some business with
the bad ADC on his chip
bad ADC on his chiiiiip

KaneTW posted:

You're developing teleportation spells?!

Arcanum sequel confirmed

# ? Nov 8, 2013 20:31

Opinion Haver: Apr 9, 2007

We've all seen PHP advocates defend PHP before. But have you seen them in the comments section of MLP fanart?

# ? Nov 10, 2013 01:58

nuvan: Mar 29, 2008; And the gentle call of the feral 3am "Everything is going so well you can't help but panic."

Opinion Haver posted:

We've all seen PHP advocates defend PHP before. But have you seen them in the comments section of MLP fanart?

Never before have I seen a link I was so scared to click.

# ? Nov 10, 2013 02:14

kitten smoothie: Dec 29, 2001

Opinion Haver posted:

We've all seen PHP advocates defend PHP before. But have you seen them in the comments section of MLP fanart?

The other comments to that piece are good, too. Like the poster who talked about programming Cobalt and Pearl.

# ? Nov 10, 2013 02:21

Flobbster: Feb 17, 2005; "Cadet Kirk, after the way you cheated on the Kobayashi Maru test I oughta punch you in tha face!"

Just moved to California and had to redo my auto insurance policy, and then print out a document to send to the DMV. So I logged on to their website and went to the online documents area. In Safari, the document downloaded as a 3-page PDF. No prob.

But since I needed to print it off at work, I logged on there where I primarily use Chrome. The document downloaded as... an HTML page with three <img> tags, one for each page, with poorly aliased text, encoded as a data: URI. :stare:

How does that even happen?

# ? Nov 10, 2013 03:30

Volmarias: Dec 31, 2002; EMAIL... THE INTERNET... SEARCH ENGINES...

Flobbster posted:

Just moved to California and had to redo my auto insurance policy, and then print out a document to send to the DMV. So I logged on to their website and went to the online documents area. In Safari, the document downloaded as a 3-page PDF. No prob.

But since I needed to print it off at work, I logged on there where I primarily use Chrome. The document downloaded as... an HTML page with three <img> tags, one for each page, with poorly aliased text, encoded as a data: URI.

How does that even happen?

Not Safari, Firefox or MSIE? Must be netscape or something from beyond the sands of time, better go to compatibility mode. PDFs may not yet be invented.

# ? Nov 10, 2013 04:57

pseudorandom name: May 6, 2007

Except the data URL protocol is new.

# ? Nov 10, 2013 05:46

Jewel: May 2, 2009

pseudorandom name posted:

Except the data URL protocol is new.

http://tools.ietf.org/html/rfc2397

"Proposed August 1998"

:confused:

# ? Nov 10, 2013 05:51

pseudorandom name: May 6, 2007

When was it actually implemented?

# ? Nov 10, 2013 07:38

Strong Sauce: Jul 2, 2003; You know I am not really your father.

All browsers support it at the moment except IE, with partial support.

http://caniuse.com/#feat=datauri

# ? Nov 10, 2013 07:53

NtotheTC: Dec 31, 2007

Opinion Haver posted:

We've all seen PHP advocates defend PHP before. But have you seen them in the comments section of MLP fanart?

"PHP is way more secure than Java or C"

:v:

"Uh, I'm not sure that's true"

:smug:

"Well unless you can provide me with hundreds of specific sources proving your point then your argument is indefensible"

Outstanding.

# ? Nov 11, 2013 23:58

ixnay: Jun 11, 2002; rainbow dash why are you making such a cool face?!

A customer is using "strings.php" to define all of the product sections on their website, which is comprised of this amazing set of arrays:

PHP code:

//Do NOT touch this section
$tab = array();
$category = array(array());
$image = array(array());
$family = array(array());
$prodName = array(array(array()));
$pcode = array(array(array()));
$promoIMG = array();
$promoLink = array();
$promoAlt = array();

//Product Section
$selectText             = "Please select your product...";
$moreText               = "More products...";
$i=1;           //Do NOT change
$k=1;           //Do NOT change
$j=1;           //Do NOT change
$tab[$i]                                = "Our Company";
$more[$i]                               = 0;
$category[$i][$k]               = "Network Storage";
$image[$i][$k]                  = "/zzz/qqq/images/enduser/feature_zzz.jpg";
$family[$i][$k]                 = "1188";
$prodName[$i][$k][$j]   = "Our Company zzz-450r";
$pcode[$i][$k][$j]              = "1195";
$j++;           //Do NOT change
$prodName[$i][$k][$j]   = "Our Company zzz-400r";
$pcode[$i][$k][$j]              = "1196";
$j++;           //Do NOT change
$prodName[$i][$k][$j]   = "Our Company zzz-300r";
$pcode[$i][$k][$j]              = "1198";
$k++;           //Do NOT change
$j=1;           //Do NOT change
$category[$i][$k]               = "Desktop Network Storage";
$image[$i][$k]                  = "/zzz/qqq/images/enduser/feature_zzz_desktop.jpg";
$family[$i][$k]                 = "1189";
$prodName[$i][$k][$j]   = "Our Company zzz-300d";
$pcode[$i][$k][$j]              = "1190";
$j++;           //Do NOT change
$prodName[$i][$k][$j]   = "Our Company xxx-300d";
$pcode[$i][$k][$j]              = "1191";
$j++;           //Do NOT change
$prodName[$i][$k][$j]   = "Our Company yyy-300d";
$pcode[$i][$k][$j]              = "1192";
$k++;           //Do NOT change
$j=1;           //Do NOT change

This repeats for a couple hundred lines.

# ? Nov 12, 2013 00:41

Volmarias: Dec 31, 2002; EMAIL... THE INTERNET... SEARCH ENGINES...

What I love most is the "Do NOT change" comments. I suspect that it was laid out this way because someone was getting sick of having to make code changes for the business side making minor text changes, so he decided to make this file to let them do it instead. Except that the business folks decided that they didn't need those i's or j's and just chucked them.

# ? Nov 12, 2013 00:58

EAT THE EGGS RICOLA: May 29, 2008

Someone blew up all the Jenkins plugins repositories because they had forced push permissions without realizing it.

quote:

AW - *PLEASE READ* Re: strange pushes on GitHub

Hi all,
I have triggered an involuntary "forced push" last night on the list of Jenkins-CI plugins indicated below in this e-mail.

My apology

I did not realise that I actually had forced push permissions and I do apologise for the inconvenience caused.
The operations pushed back the all the branches to around 1 month. The history is not lost and is still on the GitHub server but on detached branches.

The solution

I can raise a request to GitHub to provide the "reflog" of those repositories and restore the branches to the point before my forced push.
Alternatively the owners of those repositories can still perform a "forced push" to restore the correct position of the branches.
(if you would like to do so, please write to the mailing list so that we do not overlap the recovery operations)

The full list

See below the full list of repositories impacted:

[snip]

The prevention

Can we prevent this to happen again ?
I personally do not work on any of those repositories but still have "forced push" permissions ... and so many other people have.
I don't see the value of having such power of "potential disruption" associated to my account :-( ... can we remove the forced push by default and enable on a case-by-case basis ?

--- * ---

If you would like to propose an alternative approach to resolve the problem, feel free to follow-up !

... and again ... accept my sincere apologies :-(

Luca.

# ? Nov 12, 2013 01:57

prefect: Sep 11, 2001; No one, Woodhouse.
No one.; Dead Man’s Band

EAT THE EGGS RICOLA posted:

Someone blew up all the Jenkins plugins repositories because they had forced push permissions without realizing it.

Holy gently caress. Not updating any plugins this week.

# ? Nov 12, 2013 02:16

Pythagoras a trois: Feb 19, 2004; I have a lot of points to make and I will make them later.

quote:

Can we prevent this to happen again ?
I do not work on any of these projects, why do you let me force push to them? Shame on you. If you control user permissions please write to the mailing list to revoke force push on my account after recovery efforts have been completed.

# ? Nov 12, 2013 03:02

Strong Sauce: Jul 2, 2003; You know I am not really your father.

Git is pretty hard, but one of the stupid things is for you to just do a 'git push --force' without specifying the remote location/repo. It is annoying to me that he is blaming GitHub and others for giving him access when he was doing something as dangerous as that. Why would you even do that? If you want to overwrite history in one of your repos, you should be doing a git push specifically to that repo.

# ? Nov 12, 2013 03:25

Jabor: Jul 16, 2010; #1 Loser at SpaceChem

Strong Sauce posted:

Git is pretty hard, but one of the stupid things is for you to just do a 'git push --force' without specifying the remote location/repo. It is annoying to me that he is blaming GitHub and others for giving him access when he was doing something as dangerous as that. Why would you even do that? If you want to overwrite history in one of your repos, you should be doing a git push specifically to that repo.

I think if the default for a "dangerous" command like that is to just assume the user wants to overwrite history of every single repo they have access to, it's perfectly fair to blame git for having a horrible interface.

Additionally, it's also perfectly fair to say "hey maybe giving people who don't even work on this project the ability to horribly mess up your repository is a bad idea".

# ? Nov 12, 2013 04:21

OddObserver: Apr 3, 2009

Didn't git recently change to have push only affect one branch by default?

# ? Nov 12, 2013 04:27

necrotic: Aug 2, 2005; I owe my brother big time for this!

OddObserver posted:

Didn't git recently change to have push only affect one branch by default?

Yup. http://blog.nicoschuele.com/?p=217

# ? Nov 12, 2013 04:46

Plorkyeran: Mar 22, 2007; To Escape The Shackles Of The Old Forums, We Must Reject The Tribal Negativity He Endorsed

Jabor posted:

I think if the default for a "dangerous" command like that is to just assume the user wants to overwrite history of every single repo they have access to, it's perfectly fair to blame git for having a horrible interface.

Fortunately not only is that not the default, but it's not even a thing git can do at all (you can push a single repository to multiple remotes at once, but you can't operate on multiple repositories in a single git command). He did the equivalent of writing a script to loop over every file on his hard drive and call rm on each one under the assumption that it'd fail to delete most of the files.

# ? Nov 12, 2013 05:04

Pollyanna: Mar 5, 2005; Milk's on them.

I like to imagine that this is the same Luca as the one on Masterchef.

# ? Nov 12, 2013 06:49

Qwertycoatl: Dec 31, 2008

Why does this guy have permission to push to 180 repositories he's not involved with, anyway?

# ? Nov 12, 2013 09:22

Strong Sauce: Jul 2, 2003; You know I am not really your father.

Jabor posted:

I think if the default for a "dangerous" command like that is to just assume the user wants to overwrite history of every single repo they have access to, it's perfectly fair to blame git for having a horrible interface.

Additionally, it's also perfectly fair to say "hey maybe giving people who don't even work on this project the ability to horribly mess up your repository is a bad idea".

`git push --force` is never a good idea. It is only dangerous if you add the --force flag, in which case you should be well aware of the dangers of using that flag. If he wanted to be on the safe side he should have just done a git merge/git push of his changes. You should never run that git command unless you're absolutely sure that is what you want to do.

I am really curious of the following though:

1. He did a `git push --force` with repos that were months old. That is pretty irresponsible. Why would you even do a `git push --force` without updating all your repos? I assume in this case he just wanted to force push to one repo, but then why would you not specify the repo in your command as a preventative measure, especially when you are using the --force flag?

2. He has been on Github for almost 4 years and just _assumed_ Github has something to prevent force pushes? According to someone in that thread it may be in the Enterprise edition? Regardless it seems like later on in the thread he admits that the way they use GitHub probably isn't normal so, again I don't know why he just assumed GitHub would protect him from this. Which again begs the question of why he didn't specify a repo instead of doing this.

This guy has over 20 years of experience and has been on GitHub ~4 years and he used git push --force? I mean I agree that the people who just added him to all the repos are partially to blame as well and I am sure (maybe) that they will now actually restrict access to repos but he still did something stupid.

I mean people make mistakes but it seems like he is trying to deflect a lot of the blame for something that should never happen. Maybe your fingers accidentally hit the Enter key when you were rewriting the command while changing repos or something. Just fess up to that instead of doing this.

Edit: They have 562 members in the organization and apparently 1K+ repos. Uhm, yeah maybe the admins of the org should have done a better job with access controls.

Strong Sauce fucked around with this message at 09:55 on Nov 12, 2013

# ? Nov 12, 2013 09:51

Vanadium: Jan 8, 2005

Strong Sauce posted:

`git push --force` is never a good idea.

On github we keep doing that thing where we send a pull request, someone reviews+comments on the commits and suggests changes, and then we amend our commits and push -f them to the pull-requested branch until the pull request gets accepted. I don't think that's that weird a workflow?

# ? Nov 12, 2013 10:05

EssOEss: Oct 23, 2006; 128-bit approved

For those of us not englightened in the ways of git, what exactly is a forced push?

# ? Nov 12, 2013 10:40

Adbot: ADBOT LOVES YOU

# ? May 16, 2024 05:22

Vanadium: Jan 8, 2005

It's basically a push that doesn't just add commits to the top of the pushed-to branch in the remote repository, but can arbitrarily rewrite the branch to point to a completely different commit.

In the case of this story, the push reset 150 repositories' branches to point to a month-old commit, basically just rewinding them and effectively dropping all the commits since. The commits were left floating around in git's object space and could easily be reached by their hash ids, it's just that they weren't reachable by looking at the commit at the top of the branch and its ancestors, so they wouldn't show up in that branch's history. It's fairly easy to fix but I guess it's a pain to do it 150 times.

# ? Nov 12, 2013 10:49

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Cavern of COBOL > Coding Horrors: You can gather all your technical debt into one easy framework!

«‹›1503 »