Port to Qt 6 by alexispurslane · Pull Request #21 · realnc/qtads

alexispurslane · 2023-10-28T15:12:10Z

In the interests of future maintainability and compatibility, I ported QTADS to Qt6. Everything should work identically, I just went through and replaced uses of constructs that were removed in Qt6. I also removed the uses of globalStrut entirely because according to the Qt developers it was useless and usually set to zero anyways.

I also add the capacity to accept command line flags and two flags (--help and --embed), because I need the ability to tell qtads to print out its window id for something else I'm working on. I hope you don't mind.

alexispurslane · 2023-10-28T20:13:08Z

I'm not familiar at all with this codebase, and not even that experienced with C++, and I'm very foggy today, so please feel free to comment/critique and ill change it to get this PR merged! :D

a-detiste · 2023-11-15T11:11:32Z

Nice work. I would upload this to debian/experimental as a patch if I'm really bored or anyone ask for ;-)

fos1 · 2023-11-15T11:35:15Z

Any new uploads to Debian regarding TADS would be very nice. Maybe they wouldn't think TADS is obsolete. Thanks, Jeff ( Fos1)

…

On Wed, Nov 15, 2023 at 5:11 AM Alexandre Detiste ***@***.***> wrote: Nice work. I would upload this to debian/experimental as a patch if I'm really bored or anyone ask for ;-) — Reply to this email directly, view it on GitHub <#21 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADDAWSLR6C4RIJH4UU2L6HLYESPPDAVCNFSM6AAAAAA6UEY4S2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMJSGMYTCMZWG4> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

a-detiste · 2023-11-15T11:43:19Z

it is very well alive, I enabled CI right now.

one can have a look a the build warnings if interrested

https://salsa.debian.org/games-team/qtads/-/commit/a8cf846ac80918031a37c3a33da8981cf506cec7

realnc · 2023-11-15T17:07:20Z

Thank you very much for doing the work! I'll need to find some time go over it and merge it. Right now I'm kind of swamped with other things.

alexispurslane · 2023-11-15T21:12:04Z

Thank you very much for doing the work! I'll need to find some time go over it and merge it. Right now I'm kind of swamped with other things.

Thank you! And yeah that's perfectly understandable.

a17r · 2025-05-02T23:29:54Z

Any status update? At least Gentoo and Ubuntu are now pursuing Qt5 removal trackers.

alexispurslane · 2025-05-05T11:56:20Z

Any status update? At least Gentoo and Ubuntu are now pursuing Qt5 removal trackers.

I'm still around to work on any changes the author needs me to make (although I've been flaking on a number of my contributions to open source projects lately because work has been absolutely hammering me so there may be some delay, but my attention is still available at least). The bottleneck is just whenever he has time.

a17r · 2025-08-27T16:20:26Z

Qt5 is now declared deprecated in Gentoo

At this point in time, Linux distributions are already tracking which packages are blocking Qt5 cleanup. Real Qt Company OSS support stopped back in 2020, when public availability of LTS releases started to be delayed for 1 year, stable branch repository access got closed down (with 5.15 branch repository receiving any new LTS tag one year after commercial release) and only Qt6-relevant bugs are being fixed (then backported) since then. The final release happened on 2025-05-19, which means a) we will not get it until 2026-06-19, b) any bugs discovered since last May will not be fixed and c) no bugs since 2020-12-08 that haven't also been affecting Qt6 were fixed at all.

At least Gentoo, Alpine and Ubuntu are entertaining Qt5 porting trackers by now, for Gentoo, Qt5 revdep cleanups are in full swing already.
https://bugs.gentoo.org/show_bug.cgi?id=qt5-removal
https://gitlab.alpinelinux.org/alpine/aports/-/issues/17114
https://ubuntu-archive-team.ubuntu.com/transitions/html/qt5-removal.html

ferdnyc

Thanks for doing this @alexispurslane!

ferdnyc · 2025-09-06T08:21:50Z

+    const QStringConverter::Encoding codecs[9]{
+        QStringConverter::Encoding::System,  QStringConverter::Encoding::Latin1,
+        QStringConverter::Encoding::Utf16,   QStringConverter::Encoding::Utf16BE,
+        QStringConverter::Encoding::Utf16LE, QStringConverter::Encoding::Utf32,
+        QStringConverter::Encoding::Utf32BE, QStringConverter::Encoding::Utf32LE,
+        QStringConverter::Encoding::Utf8,
+    };


I don't think this list actually makes much sense. As the comment in the removed code said, in the original set only UTF-8 was included (despite even that not being common in practice) because at least TADS 2 has a one-byte-per-character format. So UTF16 is unlikely-to-impossible, and I don't think I've ever seen UTF32-encoded... anything. Anywhere. Ever.

On my (Fedora Linux) system, presumably thanks to the ICU library, QStringConverter::availableCodecs() produces a list of 233 codecs, pretty much a drop-in replacement for QTextCodec::availableCodecs(). Filtering that list by the same criteria as the old code leaves these 78:

['UTF-8', 'ISO-8859-1', 'ISO-8859-2', 'ISO-8859-3', 'ISO-8859-4', 'ISO-8859-5', 'ISO-8859-6', 'ISO-8859-7', 'ISO-8859-8', 'ISO-8859-9', 'ISO-8859-10', 'ISO-8859-13', 'ISO-8859-14', 'ISO-8859-15', 'EUC-JP', 'EUC-KR', 'windows-874-2000', 'IBM437', 'IBM775', 'IBM850', 'IBM852', 'IBM855', 'IBM857', 'IBM00858', 'IBM860', 'IBM861', 'IBM862', 'IBM863', 'IBM864', 'IBM865', 'IBM866', 'IBM868', 'IBM869', 'KOI8-R', 'KOI8-U', 'windows-1250', 'windows-1251', 'windows-1252', 'windows-1253', 'windows-1254', 'windows-1255', 'windows-1256', 'windows-1257', 'windows-1258', 'ISO-2022-JP', 'ISO-2022-JP-1', 'ISO-2022-JP-2', 'ISO-2022-KR', 'ISO-2022-CN', 'ISO-2022-CN-EXT', 'IBM037', 'IBM273', 'IBM277', 'IBM278', 'IBM280', 'IBM284', 'IBM285', 'IBM290', 'IBM297', 'IBM420', 'IBM424', 'IBM500', 'IBM-Thai', 'IBM870', 'IBM871', 'IBM918', 'IBM1026', 'IBM1047', 'IBM01140', 'IBM01141', 'IBM01142', 'IBM01143', 'IBM01144', 'IBM01145', 'IBM01146', 'IBM01147', 'IBM01148', 'IBM01149']

That's probably excessive, but OTOH some legacy codecs like windows-1250, windows-1252, ISO-8859-15, etc. were extremely common in older files, and still come up fairly often even today.

ferdnyc · 2025-09-06T08:33:07Z

-        if (ui->encodingComboBox->findText(QString::fromLatin1(codec)) == -1) {
-            ui->encodingComboBox->addItem(QString::fromLatin1(codec));
+        if (ui->encodingComboBox->findText(
+                QString::fromLatin1(QStringConverter::nameForEncoding(codec)))


According to QStringConverter's docs, nameForEncoding() outputs a const char * string that's UTF-8 encoded, so fromLatin1() sounds like the wrong interpretation. It's probably best to just drop it, since QString will default to interpreting char * as UTF-8 anyway.

(fromLatin1() is used to override that default when dealing with strings that can be processed faster as 8-bit Latin1. But the names returned by nameForEncoding() aren't guaranteed to be 8-bit fixed-width.)

ferdnyc · 2025-09-06T10:09:16Z

-            gameFileName = args.at(1) + ".t3";
-        } else {
-            qWarning() << "File" << args.at(1) << "not found.";
+    if (args.size() >= 1) {


Suggestion: Replace all of this tedious manual command-line parsing with,

#include <QCommandLineParser> // ... auto main(int argc, char** argv) -> int { QString gameFileName; bool embedOption = false; // ... QCommandLineParser parser; parser.setApplicationDescription("QTads: Interpreter for TADS 2 and 3 game files"); parser.addHelpOption(); parser.addVersionOption(); parser.addPositionalArgument("FILE", "Game file to load"); parser.addOptions({ {{"e", "embed"}, "Print the QTads window ID to stdout on startup (for embedding)"}, }); parser.process(app); const QStringList args = parser.positionalArguments(); embedOption = parser.isSet("embed"); if (args.length() > 0) { gameFileName = checkFileName(args.at(0)); if (gameFileName.isEmpty()) { qWarning() << "File" << args.at(0) << "not found."; } }

(And a helper checkFileName, something like:)

auto checkFileName(const QString& input) -> QString { QStringList fileExtensions; fileExtensions << QLatin1String("") << QStringLiteral(".gam") << QStringLiteral(".t3"); for (auto& ext : fileExtensions) { QString name = input + ext; if (QFile::exists(name)) { return name; } } return nullptr; }

Qt will handle processing of -h / --help, -v / --version, failing with a message when given unrecognized arguments, and all of the other tedium involved in command-line parsing. When called with --help / --h it'll output something like this:

$ qtads --help Usage: qtads [options] FILE QTads: Interpreter for TADS 2 and 3 game files Options: -h, --help Displays help on commandline options. --help-all Displays help including Qt specific options. -v, --version Displays version information. -e, --embed Print the QTads window ID to stdout on startup (for embedding) Arguments: FILE Game file to load

The Windows version will even show the help in a window, if QTads is launched from a script or desktop shortcut with a bad argument, instead of in a shell.

alexispurslane force-pushed the master branch 2 times, most recently from c4a4322 to c422228 Compare October 28, 2023 16:27

Port to Qt 6 and add command line arguments for embedding

2fef136

alexispurslane force-pushed the master branch from c422228 to 2fef136 Compare October 28, 2023 16:38

switch to Qt 6 text encoding/decoding handling to get rid of qt5compat

7154efc

This was referenced Jan 17, 2024

QTads doesn't build on QT 6 #23

Closed

macOS: QTads crashes on launch when I build from source on Apple Silicon #25

Open

cho-m mentioned this pull request Aug 28, 2025

qt@5: deprecate on 2026-05-19 Homebrew/homebrew-core#235319

Merged

18 tasks

ferdnyc reviewed Sep 6, 2025

View reviewed changes

Navigation Menu

Port to Qt 6 - #21

Port to Qt 6#21
alexispurslane wants to merge 2 commits into
realnc:masterfrom
tads-working-group:master

alexispurslane commented Oct 28, 2023

Uh oh!

alexispurslane commented Oct 28, 2023 •

edited

Loading

Uh oh!

a-detiste commented Nov 15, 2023

Uh oh!

fos1 commented Nov 15, 2023 via email

Uh oh!

a-detiste commented Nov 15, 2023

Uh oh!

realnc commented Nov 15, 2023

Uh oh!

alexispurslane commented Nov 15, 2023

Uh oh!

a17r commented May 2, 2025

Uh oh!

alexispurslane commented May 5, 2025

Uh oh!

a17r commented Aug 27, 2025

Uh oh!

ferdnyc left a comment

Uh oh!

ferdnyc Sep 6, 2025

Uh oh!

ferdnyc Sep 6, 2025

Uh oh!

ferdnyc Sep 6, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

Conversation

alexispurslane commented Oct 28, 2023

Uh oh!

alexispurslane commented Oct 28, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

a-detiste commented Nov 15, 2023

Uh oh!

fos1 commented Nov 15, 2023 via email

Uh oh!

a-detiste commented Nov 15, 2023

Uh oh!

realnc commented Nov 15, 2023

Uh oh!

alexispurslane commented Nov 15, 2023

Uh oh!

a17r commented May 2, 2025

Uh oh!

alexispurslane commented May 5, 2025

Uh oh!

a17r commented Aug 27, 2025

Qt5 is now declared deprecated in Gentoo

Uh oh!

ferdnyc left a comment

Choose a reason for hiding this comment

Uh oh!

ferdnyc Sep 6, 2025

Choose a reason for hiding this comment

Uh oh!

ferdnyc Sep 6, 2025

Choose a reason for hiding this comment

Uh oh!

ferdnyc Sep 6, 2025

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

alexispurslane commented Oct 28, 2023 •

edited

Loading