Skip to content

Update French & Turkish translations - #691

Merged
mrjimenez merged 1 commit into
amule-project:masterfrom
cardpuncher:master
May 23, 2026
Merged

Update French & Turkish translations#691
mrjimenez merged 1 commit into
amule-project:masterfrom
cardpuncher:master

Conversation

@cardpuncher

Copy link
Copy Markdown
Contributor

Hi there,

Firstly, thanks a ton for working on aMule. I was astonished to see that there was new activity, so here are updated French and Turkish translations, both completed. In French I mostly removed the "AI translation please review" comments because those translations were almost all good. Turkish took a bit more time, but here it is. Again, thanks a lot and keep up the good work :)

@mrjimenez

Copy link
Copy Markdown
Contributor

Hi @cardpuncher ,

Welcome and thank you very much for your effort!

Hopefully we will have a new release real soon.

Best regards!

@mrjimenez mrjimenez added the translation Updates to translation files label May 23, 2026
@mrjimenez
mrjimenez merged commit f7005ec into amule-project:master May 23, 2026
12 checks passed
@got3nks

got3nks commented May 23, 2026

Copy link
Copy Markdown
Contributor

Hi there,

Firstly, thanks a ton for working on aMule. I was astonished to see that there was new activity, so here are updated French and Turkish translations, both completed. In French I mostly removed the "AI translation please review" comments because those translations were almost all good. Turkish took a bit more time, but here it is. Again, thanks a lot and keep up the good work :)

Thanks for your work! 👌

ngosang pushed a commit to ngosang/amule that referenced this pull request Jul 29, 2026
…#691)

Since amule-project#440 the core resolves each ed2k server's host country and hands it
to the GUI, but the two lists rendered the same piece of data two
different ways: the peer list drew a flag bitmap, while the server list
prefixed the Server Name column with the ISO code ("de - ServerName", or
"? - " when unresolved). The prefix ate horizontal space in a column that
is often too narrow already and broke up the name itself.

The server list now draws the same flag the peer list does, with the same
rule for the unresolved case: no icon rather than a placeholder.

Getting an icon into that column meant finishing a migration that was
started when the virtual lists landed and deliberately left out: the
server list was the last report-mode CMuleListCtrl of the pair, and only a
virtual list can answer OnGetItemColumnImage per row. CServerListCtrl now
derives from CMuleVirtualListCtrl and renders its cells on demand from the
model via GetItemColumnText, which also buys what the other virtual lists
already have -- O(1) row lookup instead of the linear FindItem that
RefreshServer, RemoveServer and HighlightServer each ran, a std::sort over
the model instead of moving native rows, and one coalesced re-sort per
update batch instead of an inline SortList per refreshed server. Ping,
Users and Files are declared live-sort columns, so sorting by one of them
now follows the data.

Three things had to be rebuilt against the model rather than the native
items:

  - The connected server's bold font moves from a per-item wxListItem to
    OnGetItemAttr; HighlightServer just moves m_connected and repaints.
  - FitColumnsToContent measures the model itself. wxLIST_AUTOSIZE is a
    no-op on a virtual control -- it has no native items to measure and
    returns a default width -- so only the header half can still be left
    to wxLIST_AUTOSIZE_USEHEADER.
  - RemoveAllServers collects its victims before deleting any, instead of
    deleting while walking row indices. The confirmations it raises run a
    nested event loop, and a row index does not survive that; a server
    pointer does.

The flags go in a small image list owned by this one control, seeded with
the header sort arrows so they keep their indices, rather than in the list
every CMuleListCtrl shares. Each flag is padded onto a transparent 16x16
cell -- wxImageList has one fixed size, and the bundled flags are 16x11,
so adding them as-is would stretch them.

The gate deciding which country code to show was duplicated between the
two lists, and the server list's copy had an extra
thePrefs::IsGeoIPEnabled() check that the peer list deliberately does not
have. It is redundant on both paths: the core only emits its country tag
when its resolver is on, so a tag received over EC already implies it, and
the monolithic local lookup is guarded by the resolver's own IsEnabled().
On amulegui that pref is merely a mirror of the core's, refreshed at
prefs-sync time, so consulting it could only ever disagree with the tag we
were handed. Both lists now share GetDisplayCountryCode().

That helper is a header, not a .cpp, because the answer depends on
ENABLE_IP2COUNTRY / CLIENT_GUI and those are per-target compile
definitions. CountryFlags.cpp -- the obvious neighbour -- is built into
muleappgui, one static library linked into both binaries with neither
define set, where it could only ever produce the amulegui answer.

Two base-class methods assumed native items and were dead against a
virtual model: GetSelectedItems returned nothing, and DeleteAllItems left
the model holding pointers to objects the caller was about to free. Both
are now handled in CMuleVirtualListCtrl.

No new translatable strings.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

translation Updates to translation files

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants