Extra softwares in Sailfish using CLI, repositories, etc - Page 11 - maemo.org

levone1 | # 101 | 2019-04-07, 17:29 | Report

Originally Posted by Halftux View Post

No problem and no need to say sorry

Maybe I need to say sorry. What ever.

Could mean everything maybe not enough ram. Or it is some problem with qemu. Or some configure option or compiler flag when you compile the needed packages on arm. Maybe it is better to try it on a real pc first, to make sure that the approach has success. And when everything is working like you want, then try to get it running on the mobile.

Try also to make a apt-get dist-upgrade. So the right procedure would be.

Code:

sudo apt-get update
sudo apt-get upgrade
sudo apt-get dist-upgrade
reboot
sudo apt-get install xfce4
reboot

So, turns out that I never had a network connection. I misunderstood some terminal output and thought I did ... I found this - https://ownyourbits.com/2017/02/06/r...network-access - which offered some hope, and followed all steps, including modding script to say NO_NETWORK=0 and IFACE=data0, as well as editing sudoers file... But, no luck.

edit - actually, now I seem to have connection - mu bin files referred to in sudoers were in sbin... I pinged the ip that showed up in

Code:

ip route ls

and got a normal response, so I guess connection is ok(?) Anyway, still segmentation fault error on apt-get update, even though connection is definitely there. I was able to download html of webpage with wget, but when I try to download any file, segmentation fault...

Last edited by levone1; 2019-04-08 at 02:16.

The Following 3 Users Say Thank You to levone1 For This Useful Post:
Amboss, hhbbap, juiceme

Maemish | # 102 | 2019-04-08, 09:07 | Report

Did you try to install on an sdcard or to phione itself? I got segmentation fault after an interrupted fetching of packages and only way to resolve the issue was to reformat the sdcard and start from the beginning what I was doing. Have got the same thing to happen couple of times (either with interruption likie a received call or with some unsuitable packages with dpkg error.

The Following 4 Users Say Thank You to Maemish For This Useful Post:
Amboss, Halftux, juiceme, levone1

mrsellout | # 103 | 2019-04-08, 16:13 | Report

@levone1. if I understand correctly, you're trying to run a raspbian image under qemu on your Xperia X compact?

Try having a read through this thread:
https://talk.maemo.org/showthread.php?t=98882

Here people managed to go down the chroot route with other distributions and get a VNC display up for a GUI.

You should be able to get Raspbian running in the same fashion.

The Following 3 Users Say Thank You to mrsellout For This Useful Post:
Amboss, juiceme, levone1

levone1 | # 104 | 2021-01-24, 04:31 | Report

edit - never mind, I found it - mapplauncherd-devel...

Anybody know how I can get qt5-boostable package? Trying to rpmbuild, and it's the last dependency I need, and I can't figure out which package it's in...

Last edited by levone1; 2021-01-24 at 05:03.

The Following User Says Thank You to levone1 For This Useful Post:
juiceme

levone1 | # 105 | 2021-07-07, 02:36 | Report

ok - how about this ...
Trying to rpm build from source. Hit several bumps and found solutions, but I'm stumped on this part.

Getting an error on code

Code:

= qMin(qMin(s.width() / 2, s.height() / 2), f)

.
The error is

Code:

graphicswidget/rect.cpp: In member function 'vir
tual void karin::rect::setRadius(float)':
graphicswidget/rect.cpp:293:55: error: no matchi
ng function for call to 'qMin(const double&, flo
at&)'
= qMin(qMin(s.width() / 2, s.height() / 2), f);
                                             ^

I'm totally unfamiliar with this type of code, but my searching hints that the key to the problem is in the fact that something tjere represents a 'double', so that 'f)' part needs to say something else, but can't figure it out ...

Last edited by levone1; 2021-07-07 at 02:48.

The Following User Says Thank You to levone1 For This Useful Post:
juiceme

pichlo | # 106 | 2021-07-07, 11:10 | Report

You pretty much figured it out yourself. qMin returns a double, so

Code:

qMin(qMin(s.width() / 2, s.height() / 2), f)

is seen by the compiler as

Code:

qMin(double, whatever_type_f_is)

If f is float, convert it to double. The best solution would be to make it double to start with, but if that is not practical, simply cast it:

Code:

qMin(qMin(s.width() / 2, s.height() / 2), double(f))

C++ is funny like that. In C, the conversion from float to double is automatic. C++ thinks it's too cool for that.

EDIT:
In this case, C++ cannot convert automatically because qMin is a template and requires the same type for both parameters. The compiler does not know whether to convert them both to float or to double, so it gives up and throws a wobbly. I still think it should be smart enough to convert them both to the wider type, but as I say, C++ can be funny sometimes.

Last edited by pichlo; 2021-07-07 at 11:17.

The Following 3 Users Say Thank You to pichlo For This Useful Post:
juiceme, levone1, peterleinchen

levone1 | # 107 | 2021-07-07, 16:33 | Report

Originally Posted by pichlo View Post

You pretty much figured it out yourself. qMin returns a double, so

Code:

qMin(qMin(s.width() / 2, s.height() / 2), f)

is seen by the compiler as

Code:

qMin(double, whatever_type_f_is)

If f is float, convert it to double. The best solution would be to make it double to start with, but if that is not practical, simply cast it:

Code:

qMin(qMin(s.width() / 2, s.height() / 2), double(f))

C++ is funny like that. In C, the conversion from float to double is automatic. C++ thinks it's too cool for that.

EDIT:
In this case, C++ cannot convert automatically because qMin is a template and requires the same type for both parameters. The compiler does not know whether to convert them both to float or to double, so it gives up and throws a wobbly. I still think it should be smart enough to convert them both to the wider type, but as I say, C++ can be funny sometimes.

Thanks - double(f)) did the trick. I still couldn't get rpm to build, since the build was creating a /usr/lib.. directory in BUILDROOT iwhile the app needed a /usr/lib64... directory, but all the proper files were created, so I could easily copy manually. Any insight there would be appreciated, jist out of curiosity ...

The Following User Says Thank You to levone1 For This Useful Post:
juiceme

pichlo | # 108 | 2021-07-08, 06:55 | Report

Originally Posted by levone1 View Post

I still couldn't get rpm to build, since the build was creating a /usr/lib.. directory in BUILDROOT iwhile the app needed a /usr/lib64... directory, but all the proper files were created, so I could easily copy manually. Any insight there would be appreciated, jist out of curiosity ...

I haven't got the first idea. Have you tried asking Mr DuckDuckGo?

My naive guess would be that building it on a 32-bit architecture will always create a 32-bit RPM. On a 64-bit architecture you may have a choice. This looks promising, but I have not tried it myself.

The Following 3 Users Say Thank You to pichlo For This Useful Post:
hhbbap, juiceme, levone1

levone1 | # 109 | 2021-07-08, 12:46 | Report

Originally Posted by pichlo View Post

I haven't got the first idea. Have you tried asking Mr DuckDuckGo?

My naive guess would be that building it on a 32-bit architecture will always create a 32-bit RPM. On a 64-bit architecture you may have a choice. This looks promising, but I have not tried it myself.

Thanks, I'll check it out... The error was file not found'. It was lookifor .../usr/lib64/..., but it created .../usr/lib..., which contained a 64-bit lib file.

The Following 3 Users Say Thank You to levone1 For This Useful Post:
hhbbap, juiceme, pichlo

levone1 | # 110 | 2021-07-30, 14:50 | Report

Originally Posted by pichlo View Post

I haven't got the first idea. Have you tried asking Mr DuckDuckGo?

My naive guess would be that building it on a 32-bit architecture will always create a 32-bit RPM. On a 64-bit architecture you may have a choice. This looks promising, but I have not tried it myself.

How about this one ...

A cpp file has a section like this

Code:

void IpHeartBeatWatcher::run()
{
#ifdef SFOS
    handle_ = iphb_open(NULL);
    if (handle_ != NULL)
    {
        while ( doWatch_ )
        {
            time_t time = iphb_wait(handle_, 0, (2 * IPHB_GS_WAIT_10_MINS), 1);
            printf("%ld seconds since the epoch began\n", (long)time);
            printf("%s", asctime(gmtime(&time)));

            emit triggered();

And the make error says that

Code:

2 * IPHB_GS_WAIT_10_MINS

is 'not declared in this scope'.

I'm not even sure where the variable or the scope is... Seems like the IPBH_GS... line needs to be either declared before, or included in some other declaration, or...?

Thanks

(tried to get past it with a hack :

Code:

time_t seconds;
            seconds = time(NULL);
            printf("%ld seconds since the epoch began\n", (long)time);

but then it complained about ...gmtime(&time) in the next line, that it 'cannot convert' )

Last edited by levone1; 2021-07-30 at 20:29.

The Following User Says Thank You to levone1 For This Useful Post:
juiceme