开云体育

  1. BITX20
  2. Topics

compiling ubitx ver 4.3

 

when i compile ver 4.3 i get a number of warnings although the compile appears to work it says compile done

the warnings seem to be about conversions fro char literal to char

is thsi a problem

terry gm4dso
Gordon Gibby
 

开云体育

In general no. ? Compilers generate warnings when you haven’t use the brightest code. ??


On Aug 29, 2018, at 13:43, terryhugheskirkcudbright via Groups.Io <terryhugheskirkcudbright@...> wrote:

when i compile ver 4.3 i get a number of warnings although the compile appears to work it says compile done

the warnings seem to be about conversions fro char literal to char

is thsi a problem

terry gm4dso

 

As a professional software developer I'm required to deliver code that has no errors and no warnings.? I don't understand why people don't hold themselves to the same standard when releasing software for hobbyist use.? And please don't tell me that it's not worth the time, my response to that is that is indeed worth the time because of the questions it raises when people try to use your code.

73
K5KDT
Kevin
 

开云体育

No disrespect intended, but you do get paid for what you do.? This is a hobby, so in the words of Clint E. “Go ahead, fix my code…”.?

?

?

From: [email protected] [mailto:[email protected]] On Behalf Of Kevin Timm
Sent: Wednesday, August 29, 2018 1:31 PM
To: [email protected]
Subject: Re: [BITX20] compiling ubitx ver 4.3

?

As a professional software developer I'm required to deliver code that has no errors and no warnings.? I don't understand why people don't hold themselves to the same standard when releasing software for hobbyist use.? And please don't tell me that it's not worth the time, my response to that is that is indeed worth the time because of the questions it raises when people try to use your code.

73
K5KDT
Kevin


Virus-free.

--

…_. _._
 

Kevin and MadRadioModder

The Raduino seems to have brought some of us old geezers out of hiding after 10 to 30
years of non-coding.? Since some of us learned our C-language from Dennis Ritchie and
the K&R manual we may not be up to speed with all the changes in C-code standards nor
be using the latest compilers.? This does not fix the warning messages situation but may
explain how some of the present day code is not up to latest standards.?

Part of the problem is that there seem to be several competing standards and slightly
different compilers depending on whether one is using Microsoft, Linux, BSD, Oracle,
IRIX, Solaris, or OS-X as the programming platform.?

So we depend rightly or wrongly on you professional programmers to help us get up to
speed with the latest standards and techniques.? While most will be using some variation
of the GCC compiler, it may not be up to date with latest standards if a new version has
not been installed for a couple of years.? Others using a Microsoft compiler will find a
few places where GCC C++ code might throw an error, and vice-versa.?

On Wed, Aug 29, 2018 at 1:08 PM MadRadioModder <madradiomodder@...> wrote:

No disrespect intended, but you do get paid for what you do.? This is a hobby, so in the words of Clint E. “Go ahead, fix my code…”.?

?

?

From: [email protected] [mailto:[email protected]] On Behalf Of Kevin Timm
Sent: Wednesday, August 29, 2018 1:31 PM
To: [email protected]
Subject: Re: [BITX20] compiling ubitx ver 4.3

?

As a professional software developer I'm required to deliver code that has no errors and no warnings.? I don't understand why people don't hold themselves to the same standard when releasing software for hobbyist use.? And please don't tell me that it's not worth the time, my response to that is that is indeed worth the time because of the questions it raises when people try to use your code.

73
K5KDT
Kevin


Virus-free.

--

…_. _._

 

I agree that warnings and errors in released code detract from the apparent product quality. In one of my jobs I had to deal with 90,000 warnings in one release. Everyone told me that they were insignificant and fixing the warnings would require massive effort, so they simply postponed dealing with them until the final release. I could not stand the situation and started pruning through the code. After two weeks, I uncovered about 100 serious issues and with the help of others (who started paying attention to my work) managed to bring the number of warnings to zero. Only after the value of my work was recognized, the rule became "Zero Warnings".

The above demonstrates that warnings should not be ignored. However, the advantage of open source software (or for that matter hardware), is its transparency. Greater progress can be achieved when hundreds of eyeballs are scanning the code. This is a collaborative effort that requires group collaboration. Maybe a strange concept to people who are used to only consume... If you find an issue in the code, fix it and contribute your fix to the code repository. At least contact the author and request them to fix it. By complaining, you will achieve not much.

--Ron? ?N7FTZ

On Wed, Aug 29, 2018 at 12:08 PM MadRadioModder <madradiomodder@...> wrote:

No disrespect intended, but you do get paid for what you do.? This is a hobby, so in the words of Clint E. “Go ahead, fix my code…”.?

?

?

From: [email protected] [mailto:[email protected]] On Behalf Of Kevin Timm
Sent: Wednesday, August 29, 2018 1:31 PM
To: [email protected]
Subject: Re: [BITX20] compiling ubitx ver 4.3

?

As a professional software developer I'm required to deliver code that has no errors and no warnings.? I don't understand why people don't hold themselves to the same standard when releasing software for hobbyist use.? And please don't tell me that it's not worth the time, my response to that is that is indeed worth the time because of the questions it raises when people try to use your code.

73
K5KDT
Kevin


Virus-free.

--

…_. _._

 

In one case, I've worked on some of the Google TensorFlow 开云体育 code. Was the Google code free of warnings, no, not at all.?

Often the C++ complier introduces new warnings that are due to improvements in what the compiler can detect or pointing the direction to what is coming in the next versions of the language that will be?deprecated in the future. So, fixing warnings is a matter of judgment and priorities.

A great case example of overzealous commitment to no warnings allowed was a case where someone made no warnings allowed their mission and set the -Werror flag on release code. There were many people working on projects that relied on that code. So along comes a new release of the C++ complier and hundreds of projects are halted and much time wasted (days) until someone could track down the issue not being a real problem, just a ill-advised complier flag setting. The person defended their position to set that flag in spite of the hundreds of person days lost by users.?

In the development build chain, warning free code is a good aspiration. But even companies will weigh the costs of refactoring code against the benefits. Often warnings are generated by included libraries, so unless you have the time to rewrite those yourself it is necessary to wait until the maintainers of the library can prioritize fixing and releasing a new version.?

Tom, wb6b

?
 

Perhaps I'm just old and in the way, but this "char literal to char" warning seems a bit much,
If the compiler saw where I was later assigning a different value to that char, then it can complain
And I would not.

If you want zero warnings when especially fastidious checking is turned on,
you might be re-writing a bunch of libraries.

Jerry


On Wed, Aug 29, 2018 at 10:43 AM, <terryhugheskirkcudbright@...> wrote:
the warnings seem to be about conversions fro char literal to char
is thsi a problem
 

I extremely disagree!? As a hobbyist we have NO schedules and NO budget constraints!


There is no reason not to produce clean code.? It's called being LAZY.


There are a lot of people that look at our code and try to learn from it.? Jack has written many posts

about proper coding which most people just ignore.


rOn


On August 29, 2018 at 3:08 PM MadRadioModder <madradiomodder@...> wrote:

No disrespect intended, but you do get paid for what you do.? This is a hobby, so in the words of Clint E. “Go ahead, fix my code…”.?

?

?

From: [email protected] [mailto:[email protected]] On Behalf Of Kevin Timm
Sent: Wednesday, August 29, 2018 1:31 PM
To: [email protected]
Subject: Re: [BITX20] compiling ubitx ver 4.3

?

As a professional software developer I'm required to deliver code that has no errors and no warnings.? I don't understand why people don't hold themselves to the same standard when releasing software for hobbyist use.? And please don't tell me that it's not worth the time, my response to that is that is indeed worth the time because of the questions it raises when people try to use your code.

73
K5KDT
Kevin

?




Virus-free.

?

--

…_. _._

?
 

开云体育

Right.? And this from the hobby group that routinely builds test gear that is essentially a ball of electrical tape…? Jack is a professional and acts that way.? I just barely produce code that works… if at all.

?

?

From: [email protected] [mailto:[email protected]] On Behalf Of W2CTX
Sent: Wednesday, August 29, 2018 3:40 PM
To: [email protected]
Subject: Re: [BITX20] compiling ubitx ver 4.3

?

I extremely disagree!? As a hobbyist we have NO schedules and NO budget constraints!

?

There is no reason not to produce clean code.? It's called being LAZY.

?

There are a lot of people that look at our code and try to learn from it.? Jack has written many posts

about proper coding which most people just ignore.

?

rOn



On August 29, 2018 at 3:08 PM MadRadioModder <madradiomodder@...> wrote:

No disrespect intended, but you do get paid for what you do.? This is a hobby, so in the words of Clint E. “Go ahead, fix my code…”.?

?

?

From: [email protected] [mailto:[email protected]] On Behalf Of Kevin Timm
Sent: Wednesday, August 29, 2018 1:31 PM
To: [email protected]
Subject: Re: [BITX20] compiling ubitx ver 4.3

?

As a professional software developer I'm required to deliver code that has no errors and no warnings.? I don't understand why people don't hold themselves to the same standard when releasing software for hobbyist use.? And please don't tell me that it's not worth the time, my response to that is that is indeed worth the time because of the questions it raises when people try to use your code.

73
K5KDT
Kevin

?

?

?

Virus-free.


?


--

…_. _._


?


--

…_. _._

1 - 10 of 10