|
Compiler warnings fail our build. If we really need to ignore a specific warning, we disable it.
/ravi
|
|
|
|
|
set warnings as errors. 
#SupportHeForShe
Government can give you nothing but what it takes from somebody else. A government big enough to give you everything you want is big enough to take everything you've got, including your freedom.-Ezra Taft Benson
You must accept 1 of 2 basic premises: Either we are alone in the universe or we are not alone. Either way, the implications are staggering!-Wernher von Braun
|
|
|
|
|
Warnings have the tendency to hide errors.
John Carmack wrote a pretty good article a few years ago about the best way to think of and treat warnings.
My plan is to live forever ... so far so good
|
|
|
|
|
While prototyping, who cares. When making the code production-ready, same as errors.
|
|
|
|
|
We ignore all warnings and catch bugs in user alpha testing.
Of course.
|
|
|
|
|
markrlondon wrote: catch bugs in user alpha testing.
 Finally, a real-world answer!
If it weren't for users, I'd have no bugs at all. 
"Go forth into the source" - Neal Morse
|
|
|
|
|
I don't take threats!
| Ravings en masse^ |
|---|
| "The difference between genius and stupidity is that genius has its limits." - Albert Einstein | | "If you are searching for perfection in others, then you seek disappointment. If you are seek perfection in yourself, then you will find failure." - Balboos HaGadol Mar 2010 |
|
|
|
|
|
We manage them. Warnings which disturb us (coding-style variances those warn us of, mostly) get disabled by the ruleset, merging anything is only possible with zero warnings. Too often have warnings led to bugs, and managing them to get only those which add value for our team has suited us well.
I only have a signature in order to let @DalekDave follow my posts.
|
|
|
|
|
I consider them Errors and fail the compile.
|
|
|
|
|
I prefer to call them suggestions. and if I wanted suggestions for my code, I'd just out-source it.
CQ de W5ALT
Walt Fair, Jr., P. E.
Comport Computing
Specializing in Technical Engineering Software
|
|
|
|
|
To the knife.
Software Zen: delete this;
|
|
|
|
|
Compiler warnings should never be ignored. If you ignore them, they build up and eventually obscure things like: "Unable to resolve reference to x as it was built with a higher framework version". This is just a warning, but try and publish such an application and you will quickly find it does not work in production. The number of times I am asked to help someone solve an issue which they could have solved themselves if they just read the warnings...
|
|
|
|
|
The only warnings ones I (usually) have are "returns" used to "disable" code meant for (eventual) deletion. So, I guess I do take notice when it's something else. They eventually get cleaned up; mostly unreferenced variables. Probably an OCD thing.
(Someone once noticed I had left an unused namespace in my XAML ... while I was developing; and felt they needed to bring it up).
The Master said, 'Am I indeed possessed of knowledge? I am not knowing. But if a mean person, who appears quite empty-like, ask anything of me, I set it forth from one end to the other, and exhaust it.'
― Confucian Analects
|
|
|
|
|
We are in the process of introducing the Coverity static code analyzer. It is very good at telling you why it warns you ("If [this] occurs, and then [this] and then [this], then you follow a null reference at [that] line" - the series of inferences may jump from source file to source file, and may go in five or six or more steps).
For this discussion it is more important that it maintains a history of all the "defects": If you have once reported that a given defect is in intentional, at the next Coverity run it will not be reported again. Same if you have flagged it as a false warning (which may occur if you set the agressiveness level to "high"). So you won't have the same warnings again and again. That makes it much more realistic to handle even moderate risk defects, because you do it once only. And if you give it a verdict of "intentional" or "false warning", you can leave it in your source code as it is.
Actually I am a little bit in love with Coverity at the moment. I never seen any compiler or other code analyzer that comes close to it neither in its ability to detect defects, nor its flexibility in handling them. The big disadvantage is that your billfold will complain loudly ... And it takes some heavy iron (or lot of patience). But when you employer can afford both the software and the iron, then it is great.
|
|
|
|
|
So I sat together with a client yesterday, they wanted some web application replaced.
He logs in with an admin account and gets on a page of all users that he can impersonate.
But why impersonate because... All passwords are stored AND SHOWN as plain text!
Forgot your password? Give us your email address and we'll send you your password, easy.
Oh yeah, and if you're not in the database we'll let you know so you can check if your competition is using this.
At least it's not the user's own password as the only way to get an account is... Actually, we couldn't find out, but some admin should create it (and the password I guess).
We did find how to reset a password... Change it directly in the database.
All this on HTTP without even the option for HTTPS.
As you can imagine, this wasn't the only thing that's wrong with it (don't even start on usability)...
I kind of assumed it was a quick and dirty intranet application, but it's on the public internet and apparently (business) customers are using it.
So... Should I keep these features when I rewrite it? 
Makes you wonder exactly how unqualified some people are for their job (or maybe this programmer wrote it exactly according to specs?) and that stuff like this happens everywhere.
|
|
|
|
|
Sander Rossel wrote: (or maybe this programmer wrote it exactly according to specs?)
The project spec sheet:
- Ensure that even non-hackers can trivially access your system in unexpected ways.
- Ignore best practices as often as possible.
- Confuse actual hackers by making them think it can't possibly be this easy.
- Just make it really, really dumb.
"Never attribute to malice that which can be explained by stupidity."
- Hanlon's Razor
|
|
|
|
|
Or:
1. We want full control over the system, including passwords because we know better.
2. We want to be able to log in as any user for testing purposes.
3. Changing passwords is a hassle and not user friendly, so just mail it to them.
You know how managers pointy haired bosses think (if at all) 
|
|
|
|
|
Doesn't cover the full range of findings.
I'm going to point back to what Uncle Bob said back when the VW testing scandal occurred: it doesn't matter what management wanted, the issue is that programmers did something knowing it was unethical.
Now, this of course brings up the question if ignoring security is unethical. I'm going to say yes, but I'm a bit biased on this one, since I multi-hat as a dev, security analyst, and backup SA.
Maybe @chris-maunder could make a poll out of that one 
"Never attribute to malice that which can be explained by stupidity."
- Hanlon's Razor
|
|
|
|
|
Nathan Minier wrote: if ignoring security is unethical
Maybe this programmer didn't ignore it, but simply didn't know about proper user and password management.
Or he knew, but didn't have the skills to implement it, and was afraid to ask for help because it would cost him his job.
Or he thought he knew, but obviously didn't.
I'm not making excuses, this guy should find another career asap.
I'm just saying we don't know the full story.
The only thing we know is that one or more people were not ready to take on such a project 
|
|
|
|
|
Sander Rossel wrote:
Why, as I saID TO ONE COMNPANY TRHAT WANTED TO HIRE ME:"i'VE MADE A PRETTY NICE CARREER OUT OF CLEANING UP AFTER THE MESSES YOUR COMPANY LEFT.i THINK i'LL JUST STAY HERE AND CONTINUE UNTIL YOU OFFER ME A POSITION IN CHARGE OF NOT LEAVING THE DISASTERS IN YOUYR WAKE.
Surprisingly, they agreed with me, unsurprisingly, they never called back again.
CQ de W5ALT
Walt Fair, Jr., P. E.
Comport Computing
Specializing in Technical Engineering Software
|
|
|
|
|
i need user managemnent coez, snd urgend plz.
GCS d--(d+) s-/++ a C++++ U+++ P- L+@ E-- W++ N+ o+ K- w+++ O? M-- V? PS+ PE- Y+ PGP t+ 5? X R+++ tv-- b+(+++) DI+++ D++ G e++ h--- r+++ y+++* Weapons extension: ma- k++ F+2 X
|
|
|
|
|
Nathan Minier wrote: Maybe @chris-maunder could make a poll out of that one
Sure! Send me some thoughts and I'll whip one up.
cheers
Chris Maunder
|
|
|
|
|
Problem is: if a developer ignores security it is unethical. If a developer knows Jack and sunshine about security, and Jack left town...
GCS d--(d+) s-/++ a C++++ U+++ P- L+@ E-- W++ N+ o+ K- w+++ O? M-- V? PS+ PE- Y+ PGP t+ 5? X R+++ tv-- b+(+++) DI+++ D++ G e++ h--- r+++ y+++* Weapons extension: ma- k++ F+2 X
|
|
|
|
|
Then that's not a developer; because a developer would do some research while assessing requirements. That's a random dude doing C&P from SO, and has behaved unethically by misrepresenting themselves.
"Never attribute to malice that which can be explained by stupidity."
- Hanlon's Razor
|
|
|
|
|
And you're right, except that a lot of "developers", either with engineering degrees or those certificates "become a Web Developer in 1 month and find a job" are actually certified developers.
Yet they do know less than 0 about developing, security, architecture and whatsnot.
Also, managers will invariably check out the prices of professiona, deem them too high and then have the work be done by their nephew who "knows computers".
GCS d--(d+) s-/++ a C++++ U+++ P- L+@ E-- W++ N+ o+ K- w+++ O? M-- V? PS+ PE- Y+ PGP t+ 5? X R+++ tv-- b+(+++) DI+++ D++ G e++ h--- r+++ y+++* Weapons extension: ma- k++ F+2 X
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
