Problem with configuration

[Hamousher]

On my desktop language is selected English but most of the boxes containg numerals appear as foreign and can't be changed. On my laptop is no problem

Expected Behaviour

Actual Behaviour

Describe or maybe attach a screenshot?

Steps to reproduce

System

• Stellarium version: <Name of downloaded installable file?>
• Operating system: <Name, version number>
• Graphics Card: <Manufacturer (likely Intel, NVidia, AMD?), Model (HD, Geforce, Radeon..., with model number), driver version?>
• Screen type (if applicable): Resolution, HighDPI, scaling

Logfile

If possible, attach the logfile log.txt from your user data directory. Look into the Guide for its location.

[10110111]

Stellarium version: <Name of downloaded installable file?> Operating system: <Name, version number>

These should have been filled in.

[Hamousher]

Hi I have tried ver 24.2 -qt6-win64 and an earlier version my wife has this version which is ok so I copied to my pc which still has the same fault and my desktop is running windows10 pro ver 22H2 OS build 19045.4598 PC is a lenovo m900 tower Intel i7-6700 CPU 3.4ghz with 32 gb memory and Radeon RX6400 gpu 4gb memory 64 bit.

Regards David

On Sun, 30 Jun 2024 at 08:58, Ruslan Kabatsayev @.***> wrote:

Stellarium version: <Name of downloaded installable file?> Operating system: <Name, version number>

These should have been filled in.

— Reply to this email directly, view it on GitHub https://github.com/Stellarium/stellarium/issues/3791#issuecomment-2198445064, or unsubscribe https://github.com/notifications/unsubscribe-auth/BAXT2B36PB2WBIBUIXJNCWTZJ6NCFAVCNFSM6AAAAABKDZHYVKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCOJYGQ2DKMBWGQ . You are receiving this because you authored the thread.Message ID: @.***>

[gzotti]

What is your operating system language set to? Can you think of any other reason that a system-provided date dialog shows Arab numbers? If not,

attach the logfile log.txt from your user data directory. Look into the Guide for its location.

[Hamousher]

Hi Georg, OS language is English UK and options English US and Bulgarian. The characters appearing in Stellarium are Persian numbers only and there would be no reason for me to use them. Maybe some code in the software is corrupted in relation to the language to use in the boxes that require a numerical code. With regard to the log, I have searched the complete computer and there is no log.txt file. I have even checked all the files in Users \David manually. In the Stellarium program there is a log. The only way I could send this is to do multiple screen shots, unless there is a way to copy it from the app.

Regards David

On Sun, 30 Jun 2024 at 23:08, Georg Zotti @.***> wrote:

What is your operating system language set to? Can you think of any other reason that a system-provided date dialog shows Arab numbers? If not,

attach the logfile log.txt from your user data directory. Look into the Guide for its location.

— Reply to this email directly, view it on GitHub https://github.com/Stellarium/stellarium/issues/3791#issuecomment-2198695330, or unsubscribe https://github.com/notifications/unsubscribe-auth/BAXT2B3OJSZBDQIMOTH2XPLZKBQTBAVCNFSM6AAAAABKDZHYVKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCOJYGY4TKMZTGA . You are receiving this because you authored the thread.Message ID: @.***>

[Hamousher]

Hi George, I have managed to get setup in the observatory and have a log file attached regards David

On Mon, 1 Jul 2024 at 05:42, David Hamilton @.***> wrote:

Hi Georg, OS language is English UK and options English US and Bulgarian. The characters appearing in Stellarium are Persian numbers only and there would be no reason for me to use them. Maybe some code in the software is corrupted in relation to the language to use in the boxes that require a numerical code. With regard to the log, I have searched the complete computer and there is no log.txt file. I have even checked all the files in Users \David manually. In the Stellarium program there is a log. The only way I could send this is to do multiple screen shots, unless there is a way to copy it from the app.

Regards David

On Sun, 30 Jun 2024 at 23:08, Georg Zotti @.***> wrote:

What is your operating system language set to? Can you think of any other reason that a system-provided date dialog shows Arab numbers? If not,

attach the logfile log.txt from your user data directory. Look into the Guide for its location.

— Reply to this email directly, view it on GitHub https://github.com/Stellarium/stellarium/issues/3791#issuecomment-2198695330, or unsubscribe https://github.com/notifications/unsubscribe-auth/BAXT2B3OJSZBDQIMOTH2XPLZKBQTBAVCNFSM6AAAAABKDZHYVKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCOJYGY4TKMZTGA . You are receiving this because you authored the thread.Message ID: @.***>

2024-07-01T09:41:16 Operating System: Windows 10 Version 22H2 Compiled using MSVC 19.29.30154.0 Qt runtime version: 6.5.3 Qt compilation version: 6.5.3 Build ABI: x86_64-little_endian-llp64 Addressing mode: 64-bit Processor architecture: x86_64 Processor name: Intel(R) Core(TM) i7-6700 CPU @ 3.40GHz Processor maximum speed: 3408 MHz Processor logical cores: 8 Total physical memory: 32768 MB Video controller name: AMD Radeon RX 6400 Video controller RAM: 4080 MB Current resolution: 1920x1080 C:\Program Files\Stellarium\stellarium.exe

[ This is Stellarium 24.2 (v24.2.0) - https://stellarium.org/ ] [ Copyright (C) 2000-2024 Stellarium Developers ]

Writing log file to: C:\Users\David Hamilton\AppData\Roaming\Stellarium\log.txt File search paths: [0]: C:\Users\David Hamilton\AppData\Roaming\Stellarium

Config file: C:\Users\David Hamilton\AppData\Roaming\Stellarium\config.ini Default surface format: QSurfaceFormat(version 2.0, options QFlags(), depthBufferSize -1, redBufferSize -1, greenBufferSize -1, blueBufferSize -1, alphaBufferSize -1, stencilBufferSize -1, samples -1, swapBehavior QSurfaceFormat::DefaultSwapBehavior, swapInterval 1, colorSpace QColorSpace(), profile QSurfaceFormat::NoProfile) OpenGL module type: desktop OpenGL Warning: Setting a new default format with a different version or profile after the global shared context is created may cause issues with context sharing. Default surface format: QSurfaceFormat(version 3.3, options QFlags(), depthBufferSize 24, redBufferSize 8, greenBufferSize 8, blueBufferSize 8, alphaBufferSize -1, stencilBufferSize 8, samples -1, swapBehavior QSurfaceFormat::DefaultSwapBehavior, swapInterval 1, colorSpace QColorSpace(), profile QSurfaceFormat::CoreProfile) OpenGL module type: desktop OpenGL StelGLWidget constructor StelGraphicsScene constructor initializeGL(windowWidth = 1474, windowHeight = 818) OpenGL supported version: "3.3.0 Core Profile Context 24.5.1.240502" Current Format: QSurfaceFormat(version 3.3, options QFlags(), depthBufferSize 24, redBufferSize 8, greenBufferSize 8, blueBufferSize 8, alphaBufferSize 8, stencilBufferSize 8, samples 0, swapBehavior QSurfaceFormat::DoubleBuffer, swapInterval 1, colorSpace QColorSpace(), profile QSurfaceFormat::CoreProfile) Initialization StelMainView Luminance textures are not supported Running in High Graphics Mode Maximum texture anisotropy: 16 Maximum 2D texture size: 16384 Detected: OpenGL 3.3 Driver version string: 3.3.0 Core Profile Context 24.5.1.240502 GL vendor: ATI Technologies Inc. GL renderer: AMD Radeon(TM) RX 6400 GL Shading Language version: 4.60 GLSL Version Number detected: 4.6 GLSL version is fine, we should not see a graphics problem. stel.OpenGLArray: Vertex Array Objects are supported Sky language: en_GB Empty translation file for language "en_GB" in section "stellarium-planetary-features" Planetary features language: en_GB Application language: en_GB Scripts language: en_GB Cache directory: C:\Users\David Hamilton\AppData\Local\stellarium\stellarium\cache Loaded 252 countries Loaded 193 regions Loading Solar System data (1: planets and moons) ... Loading from: ./data/ssystem_major.ini Solar System has 83 entries. Loaded 83 Solar System bodies Loading Solar System data (2: minor bodies) ... Loading from: C:\Users\David Hamilton\AppData\Roaming\Stellarium/data/ssystem_minor.ini qt.gui.imageio: libpng warning: iCCP: profile 'icc': 'GRAY': Gray color space not permitted on RGB PNG Solar System has 403 entries. Loaded 320 Solar System bodies File ssystem_minor.ini is loaded successfully... qt.gui.imageio: libpng warning: iCCP: profile 'ICC profile': 'RGB ': RGB color space not permitted on grayscale PNG Loading nomenclature for Solar system bodies ... Loaded 15959 / 15972 items of planetary surface nomenclature INFO: Cannot find these planetary objects to assign nomenclature items: "Dactyl, Dimorphos" Loading star data ... Loading star catalog: .\stars\default\stars_0_0v0_8.cat - 0_0v0_8; 4979 entries Loading star catalog: .\stars\default\stars_1_0v0_8.cat - 1_0v0_8; 21806 entries Loading star catalog: .\stars\default\stars_2_0v0_8.cat - 2_0v0_8; 150826 entries Loading star catalog: .\stars\default\stars_3_1v0_4.cat - 3_1v0_4; 425807 entries Loading star catalog: C:\Users\David Hamilton\AppData\Roaming\Stellarium\stars\default\stars_4_1v0_2.cat - 4_1v0_4; 1692779 entries Finished loading star catalogue data, max_geodesic_level: 4 Loading scientific star names from .\stars\default\name.fab Loaded 4942 / 4942 scientific star names Loading scientific star extra names from .\stars\default\extra_name.fab Loaded 26200 / 26200 scientific star extra names Loading variable stars from .\stars\default\gcvs_hip_part.dat Loaded 6862 / 6862 variable stars Loading double stars from .\stars\default\wds_hip_part.dat Loaded 22992 / 22992 double stars Loading cross-identification data from .\stars\default\cross-id.dat Loaded 108378 / 108378 cross-identification data records for stars Loading parallax errors data from .\stars\default\hip_plx_err.dat Loaded 117703 / 117703 parallax error data records for stars Loading proper motion data from .\stars\default\hip_pm.dat Loaded 117955 / 117955 proper motion data records for stars navigation/preset_sky_time is a double - treating as jday: 2460492.31248 Loading DSO data ... [...] Stellarium DSO Catalog, version 3.20 (standard edition) Loaded 94660 DSO records Loading DSO outline data ... Loaded 98 DSO outline records successfully Loading DSO discovery data ... Loaded 183 / 183 DSO discovery records successfully LandscapeMgr: initialized Cache for 100 MB. Loading star names from .\skycultures\modern\star_names.fab Loaded 1104 / 1104 common star names Loading DSO name data ... Loaded 1338 / 1363 DSO name records successfully WARNING - No position data for 25 objects: PGC 2798, PGC 2907, PGC 23521, PGC 28759, PGC 29167, PGC 34658, PGC 42102, PGC 44750, PGC 54559, PGC 59117, PGC 59858, PGC 59953, PGC 69018, PGC 69877, PGC 73957, PGC 95597, PGC 100170, PGC 100170, PGC 119230, PGC 1000714, PGC 1436754, PGC 1694462, PGC 2822840, PGC 2826829, PGC 3098124 Loaded 88 / 88 constellation records successfully for culture "modern" Loaded 85 / 85 constellation art records successfully for culture "modern" Loaded 88 / 88 constellation names Loading constellation boundary data ... Loaded 782 constellation boundary segments Loaded 84 / 84 asterism records successfully for culture "modern" Loaded 74 / 74 asterism names Initializing basic GL shaders... Creating GUI ... Loading style file: :/graphicGui/normalStyle.css Loaded plugin "Exoplanets" [Exoplanets] Version of the format of the catalog: 1 [Exoplanets] loading catalog file: C:\Users\David Hamilton\AppData\Roaming\Stellarium\modules\Exoplanets\exoplanets.json Loaded plugin "MeteorShowers" [MeteorShowersMgr] Loading catalog file: C:\Users\David Hamilton\AppData\Roaming\Stellarium\modules\MeteorShowers\MeteorShowers.json [MeteorShowersMgr] Version of the format of the catalog: 2 Loaded plugin "MissingStars" [MissingStars] Loaded 61 extra stars (missing in main catalogs) Loaded plugin "Novae" [Novae] version of the catalog: 1 [Novae] Loading catalog file: C:\Users\David Hamilton\AppData\Roaming\Stellarium\modules\Novae\novae.json Loaded plugin "Oculars" Oculars::validateAndLoadIniFile() ocular.ini exists at: C:\Users\David Hamilton\AppData\Roaming\Stellarium\modules\Oculars\ocular.ini . Checking version... Oculars::validateAndLoadIniFile() found existing ini file version 3.1 Loaded plugin "Satellites" [Satellites] loading catalogue file: "C:\Users\David Hamilton\AppData\Roaming\Stellarium\modules\Satellites\satellites.json" Satellite has invalid orbit: "AIS & APRS & ARIS & R/B" "44104" Loaded plugin "SolarSystemEditor" Using the ssystem_minor.ini file that already exists in the user directory... SSE: Comet cross-index data: Overwriting entry for "453P" Loaded plugin "TelescopeControl" [TelescopeControl] Only embedded telescope servers are available. [TelescopeControl] Adding device model: "Meade AutoStar compatible" "Any telescope or telescope mount compatible with Meade's AutoStar controller." "TelescopeServerLx200" 500000 [TelescopeControl] Adding device model: "Meade LX200 (compatible)" "Any telescope or telescope mount compatible with Meade LX200." "TelescopeServerLx200" 500000 [TelescopeControl] Adding device model: "Meade ETX70 (#494 Autostar, #506 CCS)" "Meade's ETX70 with the #494 Autostar controller and the #506 Connector Cable Set." "TelescopeServerLx200" 1500000 [TelescopeControl] Adding device model: "Losmandy G-11" "Losmandy's G-11 telescope mount." "TelescopeServerLx200" 500000 [TelescopeControl] Adding device model: "Wildcard Innovations Argo Navis (Meade mode)" "Wildcard Innovations' Argo Navis DTC in Meade LX200 emulation mode." "TelescopeServerLx200" 500000 [TelescopeControl] Adding device model: "Celestron NexStar (compatible)" "Any telescope or telescope mount compatible with Celestron NexStar." "TelescopeServerNexStar" 500000 [TelescopeControl] Adding device model: "Sky-Watcher SynScan (version 3 or later)" "Any Sky-Watcher mount that uses version 3 or later of the SynScan hand controller." "TelescopeServerNexStar" 500000 [TelescopeControl] Adding device model: "Sky-Watcher SynScan AZ GOTO" "The Sky-Watcher SynScan AZ GOTO mount used in a number of telescope models." "TelescopeServerNexStar" 500000 connectionType: TelescopeControl::ConnectionInternal initString: "New Telescope 1:TelescopeServerNexStar:J2000:COM1:500000" Creating telescope "New Telescope 1:TelescopeServerNexStar:J2000:COM1:500000" ; name/type/equinox/params: "New Telescope 1" "TelescopeServerNexStar" J2000 "COM1:500000" TelescopeClientDirectNexStar parameters: port, time_delay: "COM1" 500000 [TelescopeControl] Loaded successfully 1 telescopes. Creating scene FBO with size 1474x818 Satellite has invalid orbit: "IRIDIUM 2" "25527"

[alex-w]

Hmm... Qt ignore language settings for "inner" Qt UI elements like calendars

[github-actions[bot]]

Hello @Hamousher!

OK, developers can reproduce the issue. Thanks for the report!

[Hamousher]

so, its not just the calender that is the problem all the boxes that need a numerical value are also showing Persian or Farsi numericals and the telescope mount does not operate from the Stellarium app. On disconnecting the cable to the synscan hand control from the pc and connecting to an adjacent laptop it all works just fine

[gzotti]

We know there are issues with the telescope plugin on Qt6-based builds. (See release notes and FAQ.) Do you say that the Arab numerals show up only when your telescope is actually attached (at least cable-connected)? Does it work OK (show European numbers) when you have the Telescope plugin active, but no telescope plugged in?

[Hamousher]

Good morning Georg, the arab numerals are on the app when the telescope is attached or not, so all the time. When the cable is connected the telescope does not work and it does not matter the arab numerals are there all the time. Yesterday I took the pc and laptop to the observatory, using the same telescope and cable the laptop worked fine, found the moon on the app and selected it then slew the telescope which then tracked the moon. Disconnected the cable from the laptop and plugged it into the pc, found the moon and selected it but no bullseye for the telescope though the app says it was connected and the telescope remained stationary. I have used the pc with the same telescope and cables previously with an earlier version of stellarium and it worked ok!! I have now deleted the app several times and re installed and the same results. I have two telescopes both with RS232 and cables and have tried both cables and RS232 But cant try the other telescope as it has ZWO system with guide and main camera and EAF on the focuser Please note I will be unable to carry out any more tests for 1 week as we are going away for a few days Regards David

http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail Virus-free.www.avg.com http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

On Tue, 2 Jul 2024 at 01:19, Georg Zotti @.***> wrote:

We know there are issues with the telescope plugin on Qt6-based builds. (See release notes and FAQ.) Do you say that the Arab numerals show up only when your telescope is actually attached (at least cable-connected)? Does it work OK (show European numbers) when you have the Telescope plugin active, but no telescope plugged in?

— Reply to this email directly, view it on GitHub https://github.com/Stellarium/stellarium/issues/3791#issuecomment-2201177761, or unsubscribe https://github.com/notifications/unsubscribe-auth/BAXT2BY2KLTCVZEOZCHFOZ3ZKHIVXAVCNFSM6AAAAABKDZHYVKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEMBRGE3TONZWGE . You are receiving this because you were mentioned.Message ID: @.***>

[gzotti]

OK, then these two are at least unrelated. This would have been really weird voodoo...

• We know about issues around the Telescope plugin. The "solution" (circumvention) for users of version 24.2 is using a Qt5-based build of 24.2.
• The Arab numerals are still unexpected and unexplained. It seems one in about 200.000 users report various issues which cannot easily be reproduced or even explained.

[10110111]

cannot easily be reproduced or even explained

Hasn't @alex-w reproduced this one?

[gzotti]

The tag looks like it, but I don't know how.

[alex-w]

Hasn't @alex-w reproduced this one?

Not complete reproducible - I can reproduce the issue of ignoring language settings in Qt's dialogs (this is missing feature IMHO) and few years ago we're got the report with very similar issue for arabic/persian/etc. numbers in dialogs.

@Hamousher Could you attach screenshot for locale settings in operating system?

[gzotti]

AFAIK the program is initialized in the "C" locale, right? Is there anything missing then? How does that affect the Qt modules? Are they used uninitialized?

[alex-w]

AFAIK the program is initialized in the "C" locale, right? Is there anything missing then? How does that affect the Qt modules? Are they used uninitialized?

This is a good and difficult question - please see log from current master

[10110111]

AFAIK the program is initialized in the "C" locale, right?

Only LC_NUMERIC is set to "C", but QApplication uses what it got from the system before this (and calls setlocale accordingly, which is then overridden in LC_NUMERIC by Stellarium).

Now, setting LC_TIME=ar.utf8 (in environment variable on Linux) leads to days of week inside the calendar widget being in Arabic. Setting LC_NUMERIC=ar.utf8 results in the numbers in the calendar being Arabic. This reflects the fact that QApplication reads these variables before we reset any of them in Stellarium.

P.S. I'm not sure how this works on Windows, maybe QApplication reads the registry or queries some locale-related Win32 API.

[gzotti]

Good find. I did not check the code so far. Just that I assumed some trace of an Arab setting may have remained from some previous system setting. But en_UK, enUS and bg system language settings should not cause that. Again, just suggesting, can we read system language, initialize LC* with whatever is appropriate for current system language, then overwrite LC_NUMERIC=C. Or is it enough to set LC_NUMERIC earlier, before any Qt module has a chance to initialize?

[10110111]

Again, just suggesting, can we read system language, initialize LC_* with whatever is appropriate for current system language

Just checked, on Windows Qt calls GetLocaleInfo. So no, you can't easily force it this way. But you can call QLocale::setDefault(desiredLocale) before creating any widget (or the instance of QApplication?). This won't work when you change locale on the fly though.

I guess we still need to know the defaults when there's no config yet.

[gzotti]

I see in the helpfile:

Warning: In a multithreaded application, the default locale should be set at application startup, before any non-GUI threads are created.

Probably adding in line 189 in main.cpp helps: QLocale::setDefault(QLocale::c());

What other critical settings would change by this?

[10110111]

QLocale::setDefault(QLocale::c());

You should use the actually desired locale, rather than "C". Otherwise you'll simply get English in the calendar widgets instead of the language chosen by the user.

Actually, I've now realised that we could set the locale for the calendars explicitly by using QWidget::setLocale. The need for this is strange though. How do we localize other stock widgets, like, say, file open/save dialogs?

[gzotti]

We have our own language settings which should be used for the actual widgets. The "C" would just act as global initializer before reading config.ini. We could recommend restarting the program after a language change if dynamic re-setting is too cumbersome. OTOH there is the "languageChanged" (or similar) signal. Here GUI elements can be nudged at runtime.

[10110111]

The "C" would just act as global initializer before reading config.ini.

You still need to get the default for the case when there's nothing in config.ini yet.

OTOH there is the "languageChanged" (or similar) signal

Apparently, the uic-generated retranslateUi() method doesn't touch the QDateTimeEdit widgets. So we need to write the code to call setLocale on them in the languageChanged handlers everywhere where we have these widgets.

[gzotti]

Prepend a qDebug() << QLocale::system().bcp47Name(); for diagnostic purposes? With this in a future beta @Hamousher could possibly find what's wrong with the system and why these wrong arab numbers appear on just this one computer or under such rare circumstances. Then reconstruct a meaningful name and set this instead of C for global initialisation. Sure, if a locale is set in config.ini, we could use that instead.

Yes, our *Dialog::retranslate() methods are the places to fix the widget locales.

[10110111]

Prepend a qDebug() << QLocale::system().bcp47Name(); for diagnostic purposes?

It's not only needed for diagnostics. You also need a sensible default for a first-time launch.

why these wrong arab numbers appear on just this one computer or under such rare circumstances

Similarly wrong Russian weekday names I have on my machine with English being the Stellarium locale, because I have LANG=en_US.UTF-8 and LC_TIME=ru_RU.UTF-8 in my system. I just never paid enough attention to such details, but this has always been present in Stellarium's calendars.

I still wonder how the calls to retranslateUi etc. appear to affect the file dialogs, because they are stock dialogs, just as the calendar widget, and we don't call QWidget::setLocale on anything.

[Hamousher]

Hi guys sorry I just read your latest and realised I had not given full info. The laptop I had been using along with the desktop is running Windows 10 Home not Windows 10 Pro like the desktop maybe this has something to do with the problem

[alex-w]

Could you attach screenshot for locale settings in operating system?

Could you attach screenshot for region settings from operating system?

[Hamousher]

I can't send any screenshots at present as I am at Sofia airport waiting for a flight. Won't be home until tuesday

[Hamousher]

Hi guys, so I have checked the pc under advice from lenovo and found the problem. In settings/time & Language/Language/Preferred Languages. Next to English US were 5 icons but next to English UK there was only 1. So I presumed the 4 were missing and I found a download for the missing ones, installed them and now all is ok, see screen shot many thanks for you help.

[alex-w]

@Hamousher so, is the issue resolved by installing missing packages in operating system?

[Hamousher]

Yes, The issue is fully resolved, thank you all