Fixes#4848.
This error arose because the form was destroyed after kmshell.dpr
completes, which meant that sentry had already been destroyed. The
sentry call was simply to record a breadcrumb of the form being closed.
The fix is to move the lifecycle of the form into the `OpenTextEditor`
function. `OpenTextEditor` is used only by `initprog.RunKMCOM` and so it
is safe to remove the ApplicationRunning path of the code, which
simplifies the lifecycle of the form.
May fix#4889.
In some situations, Keyman is unable to precisely match the user's
default language with its own keyboard settings. In this situation,
Keyman may crash when attempting to install the TIP, or may give a
spurious error such as "Too many transient languages" or similar.
I have been unable to reproduce the crash described in #4889 on my
machines so this is an attempted fix.
This does fix a bug with default languages in any case so is a good fix
to include: if the user's default language is not found in Keyman's LCID
table, then Keyman would fail to install the TIP when the keyboard is
installed and would return an error message (but not crash).
Fixes#4886.
This patch includes 5 separate fixes for the various scenarios we
encountered during package installation.
1. RunTools.pas: don't attempt to install TIPs if package is not also
being installed.
2. Keyman.Setup.System.OnlineResourceCheck.pas, OnlineUpdateCheck.pas:
handle online update check for packages with spaces or periods in the
id (legacy packages only).
3. Keyman.Setup.System.InstallInfo.pas: Handle browsers renaming
downloaded files with `(n)` and `[n]` patterns as well as the current
` (n)` pattern.
4. Keyman.Setup.System.InstallInfo.pas: Handle browsers renaming
downloaded files with `_n` pattern. See code comments for details.
5. Keyman.Configuration.System.TIPMaintenance.pas: Avoid crashing if
`kmshell -install-tips-for-packages` is passed an invalid package
name.
The crash is actually fixed in point 5 above, but the root causes are
addressed in the other points.
Fixes#4830.
There remain some additional buffer size checks we should be doing in
`GetXString` but this will address the current issue by using the
standard maximum buffer size.
Fixes#4751.
Fixes KEYMAN-WINDOWS-76.
Fixes KEYMAN-WINDOWS-79.
Fixes KEYMAN-WINDOWS-7B.
Fixes KEYMAN-WINDOWS-7E.
Fixes KEYMAN-WINDOWS-7F.
Fixes KEYMAN-WINDOWS-7K.
Fixes KEYMAN-WINDOWS-7N.
Fixes KEYMAN-WINDOWS-80.
This resolves an issue where the TLangSwitchRefreshWatcher thread would
terminate early on Win7 (as it had no work to do), and free itself, but
then the main form would reference it post-free. Responsibility for free
has been moved to the main form.
This would result in various exceptions as state is corrupted at this
point.
This does not trigger a build failure; validation warnings currently always pass.
Info messages are currently suppressed with -s which also suppresses the banner,
which is more plumbing than we should change just now.
Some of our Sentry reports are difficult to track down. Adding user
interaction breadcrumbs to Keyman Configuration (to start with) may
simplify this. I may add more breadcrumbs to other projects now that
this is available.
Fixes#4520.
Start Keyman after Setup finishes only if Keyman is not already running.
The pattern in this code change is found in a few places. It would be
nice to centralise that in a unit somewhere but I don't want to do
that much refactoring at this point in beta.
Fixes#4447.
Fixes#4222.
The symptoms for these two issues are related: the language associations
change and keyboards stop functioning correctly. The issue is described
in detail in #4447.
Although this fix should be considered 'experimental', we should
probably include it in the release of 14.0, because it fixes a
longstanding issue with Keyman and Windows languages.
