I'm too sick for this nonsense
-
Update: Not a compiler bug. User error. GCC is more forgiving and the error message was weird. That was my first suspicion but after banging on it for an hour I decided it wasn't me. Then I posted here, which invoked Murphy's Law by making me look like an idiot, but at least my code works now, so it did its job. :laugh:
template class has_channel_names_impl;
template
class has_channel_names_impl {
using chidx = typename PixelType::template channel_index_by_name;
public:
constexpr static const bool value = (-1!= chidx::value) &&
has_channel_names_impl::value;
};
template
class has_channel_names_impl {
public:
constexpr static const bool value = true;
};This isn't pretty, but what it does is it looks through a pixel definition to see if the pixel has a channel or channels with particular names, at compile time. So Microsoft, explain to me the situation in which
valuedoes not resolve to a compile time constant, because that's what your compiler is telling me sometimes. Apparently it depends on the context in which it's being used/instantiated, or maybe the day of the week. Who knows? I'm not well. I don't need compiler bugs right now.Check out my IoT graphics library here: https://honeythecodewitch.com/gfx And my IoT UI/User Experience library here: https://honeythecodewitch.com/uix
-
Update: Not a compiler bug. User error. GCC is more forgiving and the error message was weird. That was my first suspicion but after banging on it for an hour I decided it wasn't me. Then I posted here, which invoked Murphy's Law by making me look like an idiot, but at least my code works now, so it did its job. :laugh:
template class has_channel_names_impl;
template
class has_channel_names_impl {
using chidx = typename PixelType::template channel_index_by_name;
public:
constexpr static const bool value = (-1!= chidx::value) &&
has_channel_names_impl::value;
};
template
class has_channel_names_impl {
public:
constexpr static const bool value = true;
};This isn't pretty, but what it does is it looks through a pixel definition to see if the pixel has a channel or channels with particular names, at compile time. So Microsoft, explain to me the situation in which
valuedoes not resolve to a compile time constant, because that's what your compiler is telling me sometimes. Apparently it depends on the context in which it's being used/instantiated, or maybe the day of the week. Who knows? I'm not well. I don't need compiler bugs right now.Check out my IoT graphics library here: https://honeythecodewitch.com/gfx And my IoT UI/User Experience library here: https://honeythecodewitch.com/uix
nope, sorry. You beat your head on this for an hour. You had three choices: 1) go take a dump. bring notebook; 2) take a shower, while washing hair answer will come to you; 3) Post on CP, answer will become immediately obvious - where's the delete button. There is an option 4, post on stackoverflow (worst site ever imho) and get reamed for being stupid, reposting, duplicating, yada yada.
Charlie Gilley “They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759 Has never been more appropriate.
-
nope, sorry. You beat your head on this for an hour. You had three choices: 1) go take a dump. bring notebook; 2) take a shower, while washing hair answer will come to you; 3) Post on CP, answer will become immediately obvious - where's the delete button. There is an option 4, post on stackoverflow (worst site ever imho) and get reamed for being stupid, reposting, duplicating, yada yada.
Charlie Gilley “They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759 Has never been more appropriate.
Well, it worked. Y'all got to be my rubber duck. As an aside, MSVC++ gives the WORST error messages.
static inline gfx_result image(Destination& destination, const rect16& destination_rect, const image& source_image, const rect16& source_rect = rect16(0, 0, 65535, 65535), const srect16* clip = nullptr);
It said this function did not take 5 arguments. I was pulling my hair out for an hour on this too.
static inline gfx_result image(Destination& destination, const rect16& destination_rect, const ::gfx::image& source_image, const rect16& source_rect = rect16(0, 0, 65535, 65535), const srect16* clip = nullptr);
That fixed it. ETA: It may look strange to call the method image that itself takes an image. It's part of my draw facilities and ends up getting used like this:
png_image img(VisualFA128);
gfx_result r = img.initialize();
if(r!=gfx_result::success) {
printf("img init failure %d\n",(int)r);
while(1);
}
r=draw::image(panel_fb,img.bounds().center(panel_fb.bounds()),img);
if(r!=gfx_result::success) {
printf("draw error: %d\n",(int)r);
}Check out my IoT graphics library here: https://honeythecodewitch.com/gfx And my IoT UI/User Experience library here: https://honeythecodewitch.com/uix
-
Well, it worked. Y'all got to be my rubber duck. As an aside, MSVC++ gives the WORST error messages.
static inline gfx_result image(Destination& destination, const rect16& destination_rect, const image& source_image, const rect16& source_rect = rect16(0, 0, 65535, 65535), const srect16* clip = nullptr);
It said this function did not take 5 arguments. I was pulling my hair out for an hour on this too.
static inline gfx_result image(Destination& destination, const rect16& destination_rect, const ::gfx::image& source_image, const rect16& source_rect = rect16(0, 0, 65535, 65535), const srect16* clip = nullptr);
That fixed it. ETA: It may look strange to call the method image that itself takes an image. It's part of my draw facilities and ends up getting used like this:
png_image img(VisualFA128);
gfx_result r = img.initialize();
if(r!=gfx_result::success) {
printf("img init failure %d\n",(int)r);
while(1);
}
r=draw::image(panel_fb,img.bounds().center(panel_fb.bounds()),img);
if(r!=gfx_result::success) {
printf("draw error: %d\n",(int)r);
}Check out my IoT graphics library here: https://honeythecodewitch.com/gfx And my IoT UI/User Experience library here: https://honeythecodewitch.com/uix
-
What did you change ? Edit just spotted it ::gfx:: No wonder it had you flummoxed
In a closed society where everybody's guilty, the only crime is getting caught. In a world of thieves, the only final sin is stupidity. - Hunter S Thompson - RIP
1st vs 2nd codeblock in the previous comment, 3rd parameter to the function from const image& to const ::gfx::image&
Check out my IoT graphics library here: https://honeythecodewitch.com/gfx And my IoT UI/User Experience library here: https://honeythecodewitch.com/uix
-
1st vs 2nd codeblock in the previous comment, 3rd parameter to the function from const image& to const ::gfx::image&
Check out my IoT graphics library here: https://honeythecodewitch.com/gfx And my IoT UI/User Experience library here: https://honeythecodewitch.com/uix
-
Update: Not a compiler bug. User error. GCC is more forgiving and the error message was weird. That was my first suspicion but after banging on it for an hour I decided it wasn't me. Then I posted here, which invoked Murphy's Law by making me look like an idiot, but at least my code works now, so it did its job. :laugh:
template class has_channel_names_impl;
template
class has_channel_names_impl {
using chidx = typename PixelType::template channel_index_by_name;
public:
constexpr static const bool value = (-1!= chidx::value) &&
has_channel_names_impl::value;
};
template
class has_channel_names_impl {
public:
constexpr static const bool value = true;
};This isn't pretty, but what it does is it looks through a pixel definition to see if the pixel has a channel or channels with particular names, at compile time. So Microsoft, explain to me the situation in which
valuedoes not resolve to a compile time constant, because that's what your compiler is telling me sometimes. Apparently it depends on the context in which it's being used/instantiated, or maybe the day of the week. Who knows? I'm not well. I don't need compiler bugs right now.Check out my IoT graphics library here: https://honeythecodewitch.com/gfx And my IoT UI/User Experience library here: https://honeythecodewitch.com/uix
I've spent all day beating on a problem with no success. Then: 1) I put it away, the next morning I spot the problem in 3 seconds. 2) Or, I call someone over; they spot the problem in 3 seconds. 3) Or, I call someone over; then *I* spot the problem in 3 seconds. Of course, I've had co-workers call me to help, and I spot the problem in 3 seconds. Typically, the person feels stupid, but the reality is that after staring at something too long, we don't see what's there -- we see what we expect to be there. The best course of action is to walk away for an hour -- do something else, then come back. To be fair, not all problems are solved so easily. My freshman year in college, working on a DEC/VAX, the program I wrote was failing to compile on a specific line, and the error made no sense. I took it to the professor, and he acted disgusted, expecting a newbie programmer problem. He read through that section twice. Frowned, and said, "That is written correctly." He thought for another moment and said, "Delete that line plus the one above and below. Retype them." An unprintable character had been inserted into the file and the compiler choaked on it. Retyping those 3 lines solved it.
